BELGIAN ELECTRICITY SYSTEM BLUEPRINT FOR 2035-2050 PDF Free Download
1 / 140
/140
100%
BELGIAN
ELECTRICITY
SYSTEM
BLUEPRINT
FOR 2035-2050
BLUEPRINT
| SEPT. 2024 |
1
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword
TABLE OF CONTENTS
EXECUTIVE SUMMARY 11
1. INTRODUCTION 32
1.1. Objectives 33
1.2. Context 34
1.3. Stakeholder interactions 40
2. METHODOLOGY 42
2.1. Methodology: in a nutshell 44
2.2. Time horizons and simulation perimeter 45
2.3. The multi-energy model 47
2.3.1. Inputs, outputs and optimisation 49
2.3.2. The KARI model (electricity zonal model) 51
2.4. Total system costs calculation 52
3. SCENARIOS 53
3.1. European scenarios and boundaries 56
3.1.1. Introduction 56
3.1.2. Energy demand 58
3.1.3. Electricity supply 64
3.1.4. Molecule supply 70
3.1.5. Greenhouse gases 72
3.1.6. Transformation processes 76
3.1.7. Grid 77
3.1.8. Climate years 78
3.2. Belgian electricity scenarios 80
3.2.1. Introduction 80
3.2.2. Energy demand 81
3.2.3. Energy supply 86
3.2.4. Demand flexibility and storage 98
3.2.5. Belgian electricity grid 101
3.3. Financial assumptions 103
3.3.1. Approach to total cost quantification 103
3.3.2. Investment costs 104
3.3.3. Cost of capital 105
3.3.4. Treatment of non-domestic offshore in the costs 105
3.3.5. Other costs components 105
3.3.6. Costs of imports outside of
Europe & domestic fuels 106
4. MULTIENERGY EUROPEAN RESULTS 108
4.1. Supply and demand per vector 110
4.1.1. Methane balances 110
4.1.2. Hydrogen balances 111
4.1.3. Liquids 112
4.1.4. Imports from outside of Europe 113
4.1.5. Total primary energy supply 114
4.1.6. Summary of insights 117
4.2. Supply and demand of electricity 119
4.2.1. European electricity supply and demand 119
4.2.2. Optimal amount of offshore 120
4.2.3. Zonal energy mix 122
4.2.4. Electricity flows 122
4.3. Interactions between energy vectors 126
4.3.1. Sankey diagrams for Europe 126
4.3.2. Interactions between the
electricity system and the other vectors 127
4.4. Management of emissions 129
4.4.1. Changes in GHG emissions 129
4.4.2. Carbon capture, usage and storage 130
4.5. European electricity grid 133
4.5.1. Important elements for the
interpretation of results 133
4.5.2. Optimal European grid found 134
4.5.3. Results of the European
optimisation around Belgium 135
4.5.4. Main takeaways regarding the
development of the high-voltage grid 138
4.6. Adequacy and flexibility 139
4.6.1. Required thermal generation 139
4.6.2. Generation characteristics of thermal units 140
4.7. System costs across the different scenarios 143
4.7.1. Total energy system costs including end uses 143
4.7.2. Energy system costs only 144
4.7.3. Zooming into the power system costs 145
4.8. Key takeaways 146
5. RESULTS FOR BELGIUM 147
5.1. Multi-energy results 149
5.1.1. Yearly methane balances 149
5.1.2. Yearly hydrogen balances 150
5.1.3. Yearly liquid balances 151
5.1.4. Imports 152
5.1.5. Primary energy supply 153
5.1.6. GHG emissions and their management 154
5.1.7. Link between the molecule and electricity systems 156
5.2. Current policies and levers 158
5.2.1. Results in the ‘Current Policies’ scenario 158
5.2.2. Domestic low-carbon supply and
expected demand 161
5.2.3. Overview of the different options
to complement Belgium’s supply 162
5.3. Electricity demand 164
5.3.1. Sufficiency as a lever 166
5.3.2. Peak demand and flexibility 168
5.4. Electricity long-term supply options 171
5.4.1. Electricity mix dashboard for 2050 171
5.4.2. Imports, exports and thermal generation 172
5.4.3. Electricity system costs of the different options 174
5.4.5. Additional indicators 186
5.4.6. Total system costs (all vectors) 188
5.5. Transition period (the road to 2050) 189
5.5.1. Electricity mix dashboard for 2036 and 2040 189
5.5.2. Electricity system costs 190
5.5.3. Nuclear extensions 191
5.5.4. Adequacy 192
5.6. Summary of the different levers 194
5.7. Electricity grid 197
5.7.1. Development and integration of
the offshore network 198
5.7.2. The further development of
onshore interconnectors 203
5.7.3. The creation of hosting capacity 206
5.7.4. The development of a strong and
robust internal backbone grid 209
5.7.5. An overview of no-regret,
minimum-regret and policy
dependent evolutions to be
envisaged for the Belgian off- and onshore grid 212
5.8. Other key insights 213
5.8.1. Material needs and other environmental aspects 213
5.8.2. Long-duration energy storage 216
5.8.3. Marginal costs and production costs 217
5.9. Key takeaways 218
6. APPENDIX 220
Appendix A - KARI dispatch and investment
electricity model 221
Appendix B - Molecules and liquids model 225
Appendix C - Carbon capture, utilisation and
storage model 227
Appendix D - Adequacy electricity model 228
Appendix E – Marginal Abatement Cost Curve
methodology 230
Appendix F – Total cost methodology 235
F.1. General introduction 235
F.2. Structure of the Cost tool 235
F.3. Energy production system costs of
different energy carriers 238
F.4. Energy consumption system cost
of End-use sectors 240
F.5. Material needs 241
F.6. Comparing the results to previous studies 242
Appendix G – Schematic view of the model 243
Appendix H – Non-CO2 emissions methodology 244
Appendix I – Details on energy demand 246
Most commonly used abbreviations 247
References 248
3
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword2
DECISIONS ABOUT THE PERIOD
2035-2050 MUST BE TAKEN SOON,
GIVEN HOW CRITICAL IT WILL BE FOR
BELGIUM'S FUTURE ENERGY SUPPLY
Dear reader,
In late 2019, when the European Commission presented its Green Deal and its goal of achieving
climate neutrality by 2050, net zero appeared to be a far-off ambition. Barely two years later, the gas
crisis and war in Ukraine have brought energy to the forefront of the political agenda, underlining
its strategic importance. Consequently, renewable energy targets have been raised and policies
aimed at the swift phasing out of fossil fuels have been implemented. This shift has transformed
Europe’s climate agenda into an investment strategy to strengthen energy security and anchor
industry to the Union.
The phasing out of fossil fuels carries significant implications for Belgium’s energy policy. Major
transformations are expected to occur between 2035 and 2050. Key developments will include
a substantial decrease in energy demand (a reduction of 25-45%) coupled with an unprece-
dented rise in electrification (an increase of 95-130%). This will place Belgium in a new position
regarding its energy supply.
This significant increase in electricity demand can be mitigated via changes in consumer
behaviour. However, ensuring access to a sufficient supply of carbon-neutral energy sources in
the long term is crucial.
Defining Belgium's future energy mix in order to ensure its security of supply will be a complex
and critical process which the next government will be responsible for. The lead times associated
with the development of the electricity grid and capital-intensive electrical power projects, such
as offshore or nuclear generation projects, exceed 10 years. With this in mind, 2035 and 2050 are
much closer than they might appear.
The development of long-term options will require the government to explore and weigh up
numerous considerations. None of the options that Belgium is facing – from offshore wind
farms located far off the coast and new nuclear plants through to sufficiency measures and the
country’s dependence on electricity imports – will be easy to implement. Each option entails
crucial development questions. Guidelines are needed to mobilise different private and public
actors to work towards Belgium’s desired energy mix.
This publication provides readers with valuable insights into the country’s options regarding its
future energy mix and evaluates their technological and economic impacts. Its goal is to assist
policymakers as they take decisions about Belgium's future energy mix and the path it will follow
in the lead-up to 2050.
We hope you find this study both engaging and insightful.
Frédéric Dunon
CEO of Elia Transmission Belgium
FOREWORD
Deciding what energy sources Belgium
will rely on in the future is crucial for
the timely development of low-carbon
technologies and grid infrastructure.
Though 2040-2050 may seem distant,
when it comes to infrastructure, we must
start planning for it soon.
Belgium’s Electricity System Blueprint
for 2035-2050 provides insights into
the country’s options regarding its
future energy mix and evaluates their
technological and economic impacts.
Its goal is to assist policymakers as they
take decisions about Belgium's future
energy mix and the path it will follow in
the lead-up to 2050.
IN SHORT
NOTE: This study has been undertaken by the Belgian system operator Elia
Transmission Belgium (ETB), which is referred to as ‘Elia’ throughout.
5
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword4
MESSAGE 3
The source of half of Belgium’s
electricity supply in the lead-up to 2050
still needs to be defined. Without a
clear policy regarding electricity supply
towards 2050, Belgium will likely end up
in the most costly scenario. Large-scale
options, like new nuclear units and non-
domestic offshore wind farms, require
clear signals to be provided in the years
to come.
MESSAGE 4
Alongside long-term preparations,
managing the transition period will
require some attention. Cost-effective
options include maximising Belgium’s
domestic renewable energy sources
(RES), applying sufficiency measures,
prolonging the lifespan of existing
generation units and developing the
country’s access to non-domestic
offshore wind. Each of these is subject
to their own specific constraints.
MESSAGE 5
The future energy mix and the location
of future power projects will play a
crucial role in the development of the
electricity grid. In all scenarios, the
reinforced and completed 380 kV grid
(backbone) is the basis for further
developments.
5 KEY INSIGHTS ABOUT
BELGIUM’S ENERGY SYSTEM
IN THE LEAD-UP TO 2050
MESSAGE 1
By 2050, Belgium’s final energy demand
will decrease by 25-45%, meaning its
energy dependency will reduce by a
factor 2. Both electrons and molecules
will play a role in the country’s future
energy supply.
MESSAGE 2
By 2050, Belgium’s final electricity
consumption is expected to rise by
95-130%. Without new policies to shape
its future energy mix, domestic supplies
are likely to cover only half of this
demand.
2050
7
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword6
A MULTI-ENERGY MODEL WAS DEVELOPED
AS PART OF THIS STUDY
Following requests from our stakeholders, this study encompasses the entire energy
system as part of a quantified modelling approach – the first time such an approach has
been taken for an Elia study. This study therefore covers both electricity and molecules
(hydrogen & derivatives, CO
2
, methane, etc.). The aim of the model is to find the
European cost optimum across all energy vectors for a given carbon target.
The different energy carriers are modelled in the study according to their specificities:
◆ the electricity system is modelled with an hourly time gran-
ularity and small sub-country geographical zones to realisti-
cally capture its behaviour;
◆ hydrogen, methane, ammonia and liquids are modelled on
a daily basis with adapted levels of geographical granularity.
Electricity model H
2
model CH
4
model Liquids model
> 100 onshore zones
> Over 400 offshore zones
> 25 000 interconnectors assessed
H
2
H
2
H
2
H
2
H
2
H
2
H
2
> 20 Zones
1 direct RES offshore zone
Onshore and offshore pipelines
assessed
Imports possible via sea terminals
for coastal areas
Imports possible via pipelines from
Ukraine and Africa
10 Zones
Existing grid kept constant
Imports possible via LNG terminals
for coastal areas
Imports possible via pipelines from
Russia/Ukraine and Africa
Domestic production of CH
4
(bio
and/or fossil) modelled in all zones
Liquid supply modelled on an EU
level
Import of liquids possible via
terminals
Domestic production of liquids
(bio and/or fossil) modelled
CH
4
CH
4
CH
4
CH
4
CH
4
CH
4
CH
4
THE MOST TRANSFORMATIVE CHANGES
ARE YET TO COME
Over the next few decades, Belgium’s economy will undergo a shift from running
on fossil fuels to optimising carbon-neutral resources and aiming for maximum
electrification. The most transformative changes to make this happen still lie ahead,
requiring the establishment of a long-term strategy for reaching net zero.
Given our expertise and techno-economic knowledge about
energy and electricity systems, this study offers up expert insights
for use by Belgian policymakers. They will shape the country’s
energy future in the lead-up to 2050 and need to take a multitude
of factors into consideration as they do so.
From extending the lifespan of current nuclear power plants,
constructing new nuclear plants, and significantly ramping up
renewable energy production to building more interconnectors,
each decision about Belgium’s future energy mix will have direct
impacts on the development and management of the electricity
grid over the next few decades.
This study will support the preparation of the upcoming Federal Development Plan 2028-2038. It will help
to evaluate long-term transmission grid infrastructure needs and their alignment with Belgium’s future
electricity mix choices. Defining future energy policy will be a crucial prerequisite for elaborating this plan.
THIS STUDY FOCUSES ON THE PERIOD
BETWEEN 2035 AND 2050
The time frame that this study focuses on extends beyond Elia’s
last Adequacy & Flexibility Study for Belgium (2024-2034) and the
Federal Development Plan 2024-2034 that included projections
for Belgium up to the year 2034.
The political decisions listed below have been taken regarding
Belgium’s energy mix and security of supply over the next 10 years,
and are taken as a starting point for this study.
ADEQUACY GRID
DEVELOPMENT
ELECTRICITY
MIX
MOLECULE
SYSTEM
Extension of the lifespan
of two nuclear reactors to
2035 and the establishment
of Belgium’s Capacity
Remuneration Mechanism
(CRM).
Approval of the FDP 2024-
2034, and the exploration
of / preparation for future
hybrid interconnectors
which can supply non-
domestic offshore wind to
Belgium.
Development of 3.5 GW of
extra offshore wind through
the Princess Elisabeth
Island and the further
development of onshore
wind and photovoltaic
(PV) capacity as well as the
latest regional ambitions
regarding domestic RES (PV
and onshore wind).
Development of a Federal
Hydrogen Strategy and the
appointment of a hydrogen
system operator.
These steps are currently being implemented and will take several years to complete. However, to help
Belgium move towards climate neutrality, a long-term vision is needed that extends beyond the next 10
years. Our Electricity System Blueprint aims to offer up our expert insights regarding the period 2035-2050.
9
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword8
BELGIUM
For Belgium, the scenario framework consists of a combination
of several elements, leading to over 300 sensitivities, which are
obtained by combining the different options for demand and
supply. Several future trajectories are defined for each component
allowing to assess the impact of the different options.
The starting scenarios for the demand in Belgium and all other
assumptions abroad are based on the European scenario frame-
work. The Belgian scenarios aim to assess the impact of a change
in Belgium only (while keeping all other things equal).
Demand scenarios (
DE
,
GA
and
ELEC
)
- A sufficiency sensitivity is assessed to quantify the impact
of behavioural changes on the consumption and hence its
impact on the costs and other parameters.
-
A district heating sensitivity is explored where more heating
networks using waste/direct heat would decrease the need
for other sources.
◆ Different supply options (on top of the current policies sce-
nario)
- More domestic RES (High RES, Very High PV).
-
Non-domestic offshore wind directly connected to Belgium
(or via hybrid interconnections).
-
New nuclear plants that could be built in Belgium (large
scale units or small modular reactors).
- Access to far-out baseload RES.
- Extension of existing nuclear plants for the duration of the
transition period.
◆ Different investment cost combinations for the options
above.
- Low/Medium/High assumptions for investment costs of the
different technologies
- Different WACC (Weighted Average Costs of Capital) for the
different technologies
THIS STUDY QUANTIFIES A DIVERSE SET OF
POSSIBLE FUTURES OVER 3 TIME HORIZONS
This study quantifies the impact and consequences of a very broad set of possible
choices for Belgium’s future energy system. A huge number of both European and
Belgian scenarios and sensitivities are taken into consideration. To present a realistic
overview of the changes to Belgium’s energy system, three sequential time horizons are
modelled: 2036, 2040 and 2050.
EUROPEAN DEMAND SCENARIOS
This study’s goal is to assess the impact of different pathways
Belgium could take in the lead-up to net zero. These use the
Ten-Year Network Development Plan 2024 (TYNDP2024) demand
scenarios as starting points.
Each of the demand scenarios mentioned below includes a dif-
ferent combination of technologies and strategies, highlighting
the diverse range of options available for reducing Belgium’s
emissions. Choosing between these scenarios will depend on a
variety of factors, including their technological feasibility, cost,
level of political support, and level of public acceptance.
1.
GA
= Global Ambition demand scenario
This scenario considers a lower electrification rate. It
includes the use of gas boilers and hydrogen-based
heat in addition to heat pumps. With regard to
transport, it assumes that up to 30% of cars and
50% of trucks are fuelled by hydrogen. This scenario
suggests that a mix of technologies, including those
based on molecules like hydrogen, can contribute to
efforts towards carbon neutrality.
2.
DE
= Distributed Energy demand scenario
This scenario is more focused on electrification but
still includes the use of molecules in the transport,
heat, and industrial sectors. It suggests that while
electrification can play a significant role in reducing
emissions, other energy vectors may still be neces-
sary for certain applications.
3.
ELEC
= Increased Electrification demand scenario
This scenario assumes a high degree of electrifica-
tion across the transport, buildings, and industrial
sectors, with other energy vectors still used for
specific applications. It suggests that widespread
electrification could be a key way to achieve carbon
neutrality.
The above demand scenarios are combined with several supply
options for onshore renewables at a European level (acceleration,
more PV), different carbon targets for intermediary years, more
flexible devices, different ways of connecting offshore wind and
different prices for the import of molecules from outside of Europe.
In total, 15 European scenarios are assessed.
EUROPEAN SCENARIOS AND SENSITIVITIES
Demand
CENTRAL
GA DE
Supply & import
Global
ambition Distributed
energy
GHG sensitivities
European scenarios and sensitivities
Onshore supply options
Offshore and grid sensitivities
Molecule import options
CO2
80%
non
CO2+
NUC
PV+
RES+
NIM
MOL
EUR+
MOL
EUR-
RAD
RAD+
400
GW
150 GW nuclear in EU in 2050
(instead of 75 GW)
+100 GW/year for PV. 2,700 GW in 2050
Acceleration for domestic PV and
onshore wind (+25 GW/y for wind and
+75 GW/y for PV)
Nimby scenario (lower onshore wind
+10 GW/y & more expensive onshore
grids)
Higher costs for imported
molecules
Lower costs for imported molecules
Only national radial
Only radial, international allowed
Connect 400 GW offshore
80% at EU level
instead of 90% in
2040
Lower ambition
on non-CO2 &
LULUCF
Demand & flex options
Over 15
scenarios
ELEC
FLEX
Increased electrification
x2 the installed demand
flex/storage in Europe
SCENARIOS OF TYNDP 2024
Multi-energy capacity
expansion European scenarios and sensitivities
SCENARIOS OF TYNDP 2024 AS STARTING POINT
+
BELGIAN SCENARIOS AND SENSITIVITIES
Over 300
sensitivities
for the
electricity
demand/
supply
Over 9 cost
combina-
tions for
each
technology
Electricity Belgian electricity sensitivities
GA DE
Global
ambition Distributed
energy
Demand
Supply
Investment
costs
ELEC, SUFF and HEAT scenario
CENTRAL, PV+ and RES+
0 to 16 GW offshore in 2050 or 'far-out baseload' RES
Extension existing capacity and 0 up to 8 GW new in 2050
Ex ante defined linked to demand and amount of PV
Enough to comply with SoS criteria
Low – medium - high
Results of the choices above
4% - 7% - 10%
Domestic RES
Non-domestic RES
Nuclear
Flexibility
Adequacy
CAPEX & FOM
Import
WACC
11
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Foreword10
EXECUTIVE
SUMMARY
EXECUTIVE SUMMARY
THIS STUDY TAKES INTO ACCOUNT THE
INPUT OF MANY OF ELIA’S STAKEHOLDERS
Given the broad scope of this study and its goal of being as comprehensive as possible,
many external parties were engaged with throughout its development.
Elia would like to extend its sincere gratitude to the following
partners for their valuable contributions to this study:
research
Academic
& advisory board
The Elia Academic Board, which was
tasked with challenging the study’s
methodology;
Specialised consulting partners,
which contributed to developing cost
assumptions and some parts of the
analysis, and benchmarking these
with other studies in the field;
The European network transmission
system operators for electricity and
gas (ENTSO-E and ENTSO-G), which
carry the legal responsibility of
providing a consistent set of cross-
sector demand scenarios for the
European system in the framework
of the upcoming ten-year network
development plan (TYNDP 2024);
Fluxys, the Belgian system operator for
gas and hydrogen, which was involved
in the alignment of some input
parameters and scenarios, challenging
the methods used to model the
molecule systems and testing some of
the resulting outcomes;
Belgium’s electricity distribution
system operators (DSOs), which
were contacted to challenge cost
assumptions regarding infrastructure
developments across the DSO grids.
Energyville, a partnership that
includes Belgian research partners
KU Leuven, VITO, Imec and UHasselt,
which was involved in cross-checking
the study’s inputs, methodology and
results;
The Elia Think Tank, which comprises
relevant Belgian stakeholders from
across the energy sector, was con-
sulted on during several workshops
that focused on the methods applied
and assumptions adopted for this
study;
13
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary12
2050
THE ELECTRIFICATION OF SOCIETY IS A NOREGRET MEASURE
Although significant transformative changes are still needed
to reach net zero, the outlook for 2050 is optimistic. Thanks to
efficiency gains, mainly linked to electrification, Europe’s final
demand for energy can be reduced by 40%, which will be key for
ensuring that the energy transition is affordable. The numbers for
Belgium are similar, although larger differences emerge between
the molecule-focused scenario (GA: 25%) and the more electrified
scenario (ELEC: 40%).
1. 2.
A
B
C3.
Renovating buildings Improving the efficiency
of user devices
Increasing electrification
Through direct electrification, the primary energy consumption of buildings, heating and transportation can be
reduced by a factor of two to three.
The electrification of society is a no-regret approach for Belgium as it can contribute to a 40% reduction in
the country’s energy demand by 2050 (compared with today). Although the pace of electrification remains
uncertain, taking action to enable it is crucial in order to avoid delaying or obstructing the energy transition.
FINAL ENERGY DEMAND IN BELGIUM
The figure below depicts how Belgium's total energy demand (in TWh, excluding international aviation & shipping and non-energetic
feedstock) is assumed to change over time. The different colours represent the different energy carriers. Belgium’s historical energy
demand is on the left-hand side of the diagram, whilst the different scenarios are simulated over three time horizons on the right-
hand side.
Final energy demand reduction
(electrification, sufficiency and
energy efficiency)
-25% to -45% reduction of final
energy demand
Electricity
Increasing from 24% in 2019
towards
55% to 80% share of final energy
demand
Gaseous Molecules
12% to 38% share of final energy
demand
Coal**
Liquids*
Hydrogen
Methane
Heat
Biomass &
waste
Electricity
Other
GA
DE
ELEC
GA
DE
ELEC
GA
DE
ELEC
360 360 -25% to -45%
+95% to +130%
300
320 310
250
310
280
240
2036 2040 2050
2010 2015 2020
Historical
Excluding international aviation & shipping and non-energetic feedstock, including grid losses.
Note that energy demand for transformations such as power-to-hydrogen and carbon capture are not included. Values are normalised for historical climate while in
the simulations, a forward-looking climate database is used, therefore the simulated demand can differ from these input values.
* Methane & liquids could be fossil, bio or synthetically sourced, which is defined in the model.
** Coal as defined as final energy demand per EUROSTAT (i.e. excluding coal consumed in blast furnaces)
Historical values based on EUROSTAT
450
400
350
300
250
200
150
100
50
0
Final energy demand [TWh]
Not yet included are electricity for
electrolysis and CCS/U (see later)
370
400
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ Belgium's final energy demand is seen to decrease by
25% to 45% by 2050.
◆ The composition of the energy mix changes drastically
over the years: the phase-out of fossil-based molecules is
accelerated and the use of electricity increases sharply.
◆ The use of electricity (in orange) rises by 95-130% to make
up 55-80% of Belgium’s final energy demand in 2050.
Gaseous molecules make up 12-38% of the country’s
domestic final energy demand in 2050 (feedstock and
international transport excluded).
WHAT DOES THIS TELL US?
Both electrons and molecules will remain
necessary, although the share they
occupy in the country’s final demand will
differ compared with today.
In the lead-up to 2050, Belgium’s energy system is expected to
undergo significant changes. Efficiency gains will lead to a notable
25-45% reduction in the country’s final energy demand, while its
total electricity consumption is anticipated to rise by 95-130% due to
electrification. These changes will place Belgium in a new situation
regarding its energy supply, in which both electrons and molecules
will play a crucial role.
MESSAGE 1
By 2050, Belgium’s final
energy demand will decrease
by 25-45%, meaning its
energy dependency will
reduce by a factor 2. Both
electrons and molecules will
play a role in the country’s
future energy supply.
15
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary14
IN BELGIUM, THE INFRASTRUCTURE DESIGN OF THE ELECTRON AND
MOLECULE SYSTEMS CAN BE DECOUPLED
INTERACTION BETWEEN THE ELECTRON AND MOLECULE
SYSTEMS IN BELGIUM IN 2050
The figure below depicts the interactions between electrons and molecules in the Belgian energy system in 2050 for two different
demand scenarios. The inputs – how electrons and molecules are supplied – are depicted on the left-hand side of each diagram.
The outputs – what form the energy is in and which sectors use it – are depicted on the right-hand side of each diagram.
GA
Ammonia
Ammonia
Imports
Ammonia-to-Hydrogen
Bioliquids
Biomass
Biomethane
Electricity
Electrolyser
Generation
Hydro
Hydrogen
Liquids
Liquids
Imports
Methane
Methane
Imports
Solar
Solids
Fertilisers
Feedstock
Households
Industry
Tertiary
Curtailment
Grid Losses
Storage Losses
Agriculture
Aviation
Other
Shipping
Ammonia
Ammonia-to-Hydrogen
Bioliquids
Biomass
Biomethane
Electricity
Generation
Hydro
Hydrogen
Liquids
Methane
Solar
Solids
Fertilisers
Feedstock
Households
Industry
Tertiary
Grid Losses
Storage Losses
Agriculture
Aviation
Other
Shipping
No or very few electrolysers are
economically viable in Belgium
How is final energy demand dis-
tributed over different end usages
and sectors, and in which ratio?
How are electrons and
molecules supplied in
Belgium?
Which conversions
between carriers take
place in the system?
Marginal contribution of electricity
generated from molecules
2050 – European scenario “Distributed Energy”
Medium electrification level of demand 2050 – European scenario “Global Ambition”
Limited electrification level of demand
DE
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ Depending on the final energy demand scenario, the
manner in which different end usages are assigned to
energy carriers can look very different.
◆ Electrons and molecules are generally consumed in their
original form, meaning conversion losses can be avoided
as much as possible.
◆ The oval shapes demonstrate how little conversion
between molecules and electrons occurs in Belgium.
◆ The volume of electricity generated through molecules
is quite limited.
WHAT DOES THIS TELL US?
While developing integrated views of
supply and consumption scenarios is
key, the low level of interaction between
electrons and molecules means the
infrastructure design of these two systems
can be decoupled in Belgium. This is not
the case in all European countries, since
it is a characteristic of countries which do
not carry sufficient amounts of domestic
renewable potential.
The primary focus of this study is the electricity system.
Questions about the design of the future molecule system
are also pertinent, and are being analysed by Fluxys, the
operator of the Belgian gas transmission network.
2050
To cover the growing demand for electricity, having access to a
sufficient supply of carbon-neutral energy sources is crucial.
MESSAGE 2
By 2050, Belgium’s final
electricity consumption is
expected to rise by 95-130%.
Without new policies to
shape its future energy mix,
domestic supplies are likely
to cover only half of this
demand.
17
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary16
CHANGES IN BELGIUM’S FINAL ENERGY DEMAND AND
DOMESTIC SUPPLY IN THE LEADUP TO 2050
The figure below depicts the changes in Belgium's final energy demand (TWh) and domestic supply in the lead-up to 2050. Each
of the different colours represents a different energy carrier.
500
400
300
200
100
0
Historical Simulated
1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
Illustration showing DE demand, current nuclear policy
and Central domestic RES
Biomass & Heat
H
2
&
derivatives
Molecules
(fossil fuels) Molecules
150
Electricity
Domestic RES and nuclear production
≈400
≈35
≈70
/2
x2
2050
Excluding international aviation & shipping and non-energetic feedstock, including grid losses.
Today
WHAT DOES THIS FIGURE DEMONSTRATE?
TODAY
◆ Belgium is shown to import around 80% of its energy, of
which the biggest part is imported as molecules (both in
liquid and gaseous forms). Electricity makes up less than
20% of the country’s final energy demand.
◆ Belgium does not generate enough electricity through
domestic RES and nuclear units to meet its electricity
demand (see pale orange area).
IN 2050
◆ The final energy demand decreases significantly (it is
halved compared with 2020).
◆ Without additional measures, domestic generation
increases by 2050 but remains insufficient to meet the
country’s electricity demand. As a result, electricity
imports will double compared with 2020.
WHAT DOES THIS TELL US?
The composition of the energy mix in
terms of electricity and molecules changes
drastically. If no new policies are adopted
regarding the country’s future electricity
supply, its domestic generation won’t be
enough to meet its rising demand. As
a result, the volume of electricity that
needs to be imported might rise steeply.
BY 2050, BELGIUM WILL NEED TWICE AS MUCH ELECTRICITY
AS IT DOES TODAY
EVOLUTION OF BELGIUM’S ELECTRICITY DEMAND AND
SUPPLY IN THE LEADUP TO 2050
The figure below demonstrates the growing gap between the increasing demand for electricity (pink lines) and Belgium’s domestic
low-carbon supply. It shows how much electricity will be needed to cover the increasing electricity demand. This is separate from
the adequacy requirement, which relates to maintaining the country’s security of supply during peak moments.
* For year X, the 5-year average in the range [X-2,X+2] is shown instead
** Approved policies: Extension of offshore wind in Belgium to 5,8 GW, extension of D4/T3 for 10 years, National/Regional energy climate plans (domestic RES,
electrification, energy efficiency...), CRM
’00* ’05* ’10* ’15* ’20* ’25 ’30 ’36 ’40 ’45 ’50
Focus of the studyHistorical & approved** policies
Historical 5-year averages Result of approved
policies Accounting for the Central scenario
of domestic RES
220
200
180
160
140
120
100
80
60
40
20
0
[TWh]
Thermal
Solar
Wind onshore
Domestic offshore wind
Biomass & Hydro
Existing nuclear
47 47 46 38 39
15 15
6
6 6
6 6 6 6
6
8
4
4
32 34 33 25
25
27 15
23
29
35
41
30
917
19 27 29 29
10 12
16
19 22 25
70-80
70-90
ELEC
DE
GA
50-60
70-80
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ The result of approved policies and the central scenario
of domestic RES leads to a doubling of the domestic
low-carbon electricity supply between 2025 and 2050.
◆ By 2036, Belgium is shown to be 50-60 TWh short in
domestic electricity supply.
◆ By 2050, its additional electricity supply needs grow to
70-90 TWh.
WHAT DOES THIS TELL US?
The domestic production of electricity
will not suffice to meet the country’s
increasing electricity demand. Without
the development of new policies related
to the country’s long-term electricity mix,
Belgium will increase its dependence
on imports. If network assets are to be
developed in time, energy policies will
have to be drawn up regarding both the
long-term structural choices for Belgium’s
electricity supply and the way in which
the transition will be organised.
19
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary18
MULTIPLE OPTIONS EXIST FOR COMPLEMENTING BELGIUM’S
DOMESTIC BASECASE LOWCARBON SUPPLY
The figure below outlines multiple levers which can be used to complement Belgium’s domestic low-carbon supply. They serve
as the building blocks of the country’s 2050 energy mix. A strategic combination of these levers will be essential for bridging the
70-90 TWh gap that will emerge between Belgium’s increasing electricity demand and its base-case domestic low-carbon supply.
Various considerations and diversification strategies must be taken into account when selecting the most effective levers.
~ 70-90 TWh
supply need 2050
On top of central
domestic RES supply
Range over the demand
scenarios (DE, GA and ELEC)
Maximum for each lever on each time horizon [TWh]
Onshore wind
x2 installation rate
Nuclear
Extend existing
units
Nuclear
Build new units
Solar PV2
x2-x4 installation rate
Molecule-fired
generation
Sufficiency levers
Lower
consumption
Outcome of other
choices
Analysed on ad-hoc
basis
Outcome of the
European dispatch
and type of mol-fired
generation installed
Range observed in the
simulations
Imports
Non-domestic
baseload RES
Build interco & RES
Non-domestic offshore
1
Build interconnectors
& offshore wind
’36 ’40 ’45 ’50 ’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45
’36 ’40 ’45 ’50
3
Range
5912
7 7 7
7 7 7
15
15 15
15
15
15
15
15 15 612 18 24
15
50
70 70 70
10 85
6
13
15
17
15
10
60
TWh
80
TWh 80
TWh
90
TWh
14 GW
10 GW
16 GW
12 GW
8 GW
4 GW
8 GW
6 GW
4 GW
2 GW
98 GW
66 GW
42 GW
4 GW
0,5 GW
4 GW
3 GW
2 GW
+4 GW
+1,5 GW
+3 GW
+3 GW
+4 GW
+4 GW
10 10 20
20 23 25
15 15
15 15
15 15
15 15 18 20 22 22
1. Non-domestic offshore refers to offshore wind capacity installed outside of Belgium’s Exclusive Economic Zone (EEZ) which still counts towards Belgium’s
domestic supply given Belgian financing/support of the wind generation itself.
2. Note that the capacity factor in the highest solar PV scenario is lower, because some PV capacity is curtailed when generation exceeds the limits of what the distribution
network can handle.
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ The bar charts depict the additional electricity (in TWh)
that can be generated (from 2036 onwards) if the
production capacity or sufficiency levers are used to their
maximum. Significant and timely efforts will be needed
to achieve these values.
◆ For example, the installation rate of onshore wind could
be doubled (compared to the Central-RES scenario) and
the installation rate of solar PV could even be quadrupled
(very high-RES scenario). An additional 4 GW of non-
domestic offshore wind could be added every 5 years.
Options related to the maximum lifetime extension of
current nuclear plants and development of new nuclear
plants are also included.
WHAT DOES THIS TELL US?
Belgium’s domestic RES generation
(PV, onshore wind and Belgian offshore
wind) can contribute significantly to
the electricity supply mix. However, this
alone will not suffice. Several options
are available for meeting the country’s
increasing electricity demand. By
leveraging their combined potential,
Belgium has the means to cover its
required electricity supply towards 2050.
2050
Compared with today, Belgium will need an additional 70 to 90
TWh of electricity to cover its electricity demand in 2050. Defining
Belgium's future energy mix will therefore be one of the critical
actions that the country’s next governments will have to take.
Multiple options exist for the country’s 2050 energy mix. In order to
shape the desired mix, different considerations and diversification
strategies need to be considered. These different options are
related to different decisional levels (federal, regional), which makes
cooperation essential.
MESSAGE 3
The source of half of Belgium’s elec-
tricity supply in the lead-up to 2050
still needs to be defined. Without a
clear policy regarding electricity sup-
ply towards 2050, Belgium will likely
end up in the most costly scenario.
Large-scale options, like new
nuclear units and non-domestic
offshore wind farms, require
clear signals to be provided
in the years to come.
21
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary20
KEY QUESTIONS TO BE ADDRESSED TO DEVELOP NONDOMESTIC OFF
SHORE WIND AND NEW NUCLEAR PLANTS
The development of large-scale energy sources like non-domestic offshore wind and new nuclear plants won’t be a walk in the park.
Below is a list of factors that need to be taken into consideration for the further development of both technologies.
NON-DOMESTIC OFFSHORE WIND NEW NUCLEAR PLANTS
INTERNATIONAL COORDINATION & JOINT
PLANNING
Planning related to offshore generation and offshore
grid development is currently organised in a decen-
tralised way, with each country identifying and tak-
ing decisions about their investments from a mostly
national perspective. In this context, the need for
non-domestic offshore is likely to be under-devel-
oped. The solution to this lies in adopting a genuinely
regional and joint approach to planning. At the same
time, efforts must be made to improve the regulatory
framework surrounding these processes.
JOINT FUNDING
The funding of offshore infrastructure and, where
required, the de-risking of wind farms through support
mechanisms currently involve simplistic approaches
which are essentially based on a territorial principle
or a fifty-fifty approach. Such approaches are not fit
for increasingly complex and meshed offshore infra-
structure to be built. Instead, countries based around
a particular sea basin need to team up to develop
sustainable cost and benefit sharing mechanisms to
ensure that proper incentives are created for all par-
ties to commit to the buildout of offshore infrastruc-
ture and offshore wind. This should also attract private
investors to help cope with the financing challenges
that the TSOs will face.
A POLITICAL DECISION
The policy makers will need to define a quantified
ambition (number of MW) and a timeline for the devel-
opment of new nuclear power plants. This will also
require a revision of the 2003 law on nuclear phase-
out, an environmental impact assessment, the consul-
tation of neighboring countries and an EU State Aid
notification.
DEFINE LOCATIONS AND SECURE PERMITS
The determination of locations for new nuclear units,
together with the securing of permits, is a prerequisite
for commissioning of a new plant. It is a complex pro-
cess, involving many stakeholders.
SETTING-UP THE FRAMEWORK
It will be imperative to define a framework for attract-
ing investment (tender, private initiative, joint ven-
ture).
COST AND TIME OF THE CONSTRUCTION
Europe carries limited recent development experience
in the field and the most recent projects took more
than 15 years to finish, while their costs severely over-
ran the original budgets. For new technologies such as
SMRs or next generation reactors, there will be limited
information available when an investment decision is
taken.
TOTAL SYSTEM COST COMPARISON WHEN COMBINING NEW
NUCLEAR WITH NONDOMESTIC OFFSHORE WIND
The figure below depicts the changes in Belgium’s total system
cost (€/MWh) brought about by different combinations of large-
scale electricity sources. The total system cost encompasses all
investment and operating expenses associated with the elec-
tricity supply under a particular scenario. As security of supply is
enforced in all scenarios, the cost of backup capacity required is
accounted for as well, and varies across the different scenarios.
In the chart in the middle of the figure, the total system costs
associated with different volumes of non-domestic offshore wind
and new nuclear plants are compared with the reference cost
assumptions.
The outer charts on the right- and left-hand sides of the figure
demonstrate the impact on the total system cost when more
conservative cost assumptions (that reflect various risks) are taken
into account. For offshore wind (on the left-hand side), these
assumptions could be linked to supply chain issues and material
costs. The cost risks for new nuclear plants are associated with
increased design complexity, safety-specific requirements, and
the immature state of small modular reactors (SMRs) (generation
III & IV).
2050 DISTRIBUTED ENERGY CENTRAL DRES
TOTAL ELECTRICITY SYSTEM COST FOR BELGIUM IN €MWH
145 144 143
137 136
133 127 124
124 121
123 115 110
119 117 117
118 117
120 114 112
118 114
123 115 110
120 124 129
122 126
122 121 124
122 124
124 122 123
Nuclear WACC 7%; overnight CAPEX 7 500 €/kW
Offshore wind WACC 7%; overnight CAPEX 1 600 €/kW
Reference grid costs
Nuclear WACC 7%; overnight CAPEX 7 500 €/kW
Offshore wind WACC 7%; overnight CAPEX 2 200 €/kW
High grid costs
Cost increase
for new nuclear
Cost increase for
non-domestic
offshore wind
Nuclear WACC 10%; overnight CAPEX 10 000 €/kW
Offshore wind WACC 7%; overnight CAPEX 1 600 €/kW
Reference grid costs
NEW
NEW NEW
NEW
8 GW
8 GW 8 GW
4 GW
4 GW 4 GW
6 GW
6 GW 6 GW
2 GW
2 GW 2 GW
0 GW
0 GW 0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
0 GW 0 GW4 GW 4 GW8 GW 8 GW12 GW 12 GW16 GW 16 GW
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ According to the reference cost assumptions (middle
chart), taking no action (0 GW along both axes) emerges
as the most costly option. Non-domestic offshore wind
is the most cost-efficient solution when compared with
the development of new nuclear generation.
◆ However, depending on how price risks are assessed, the
point above may change. This highlights the importance
of accounting for all relevant risks when taking decisions
about the desired electricity supply mix for Belgium.
WHAT DOES THIS TELL US?
As a large-scale energy source, non-
domestic offshore wind appears to be
more cost effective than the development
of new nuclear generation. Nonetheless,
the scaling up of offshore development
requires a step change in international
coordination, joint planning, and funding.
While new nuclear plants are a viable
solution, this option carries its own
challenges related to areas including
safety, complexity, and financing.
Depending on the actual evolution of Europe's energy landscape, advantages could also potentially come from linking Belgium to
a European region with extensive deployment of low-carbon energy sources, f.i. solar energy. As long as appropriate international
agreements are put in place, this large-scale local deployment could prove its benefits due to regional attributes such as space
availability, public approval, and load factor.
23
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary22
CROSSBORDER EXCHANGES BETWEEN EUROPEAN COUNTRIES WILL
REMAIN VITAL FOR A WELLFUNCTIONING INTEGRATED EUROPEAN
POWER MARKET
The future electricity system will be much more volatile. Both weather-dependent electricity generation as well as electrified appliances
such as electric vehicles, heat pumps and industrial processes are contributing to this volatility (if not carefully managed). Whereas
unlocking end-user flexibility is key for managing this volatility over time, the interconnected electricity transmission system will be
crucial for accommodating geographical fluctuations in supply and demand.
ELECTRICITY IMPORTSEXPORTS OF BELGIUM IN 2050
ACROSS DIFFERENT SCENARIOS
Non-domestic
offshore wind
Imports
Exports
New nuclear
units
0 GW 16 GW 8 GW 0 GW 16 GW
2020 - 23
0 GW 0 GW 4 GW 8 GW 8 GW
H
I
G
H
H
I
G
H
H
I
G
H
H
I
G
H
H
I
G
H
C
E
N
C
E
N
C
E
N
C
E
N
C
E
N
100
50
0
-50
-100
[TWh]
Domestic
RES
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ The bar chart shows the electricity that will be exchanged
between Belgium and its neighbouring countries in 2050
(imports are separated from exports).
◆ Different supply scenarios for Belgium are depicted,
ranging from no new large-scale electricity sources to
the installation of maximum amounts of non-domestic
offshore wind (16 GW) and new nuclear generation (8 GW),
and from the Central domestic RES scenario to the High
domestic RES scenario.
◆ For comparison purposes, the electricity Belgium
exchanged with its neighbours during the period 2020-
2023 is also included.
WHAT DOES THIS TELL US?
In all supply scenarios for Belgium in the
lead-up to 2050, the volume of cross-border
exchanges is several times higher than it is
today. The integrated European electricity
market remains a cornerstone of the future
energy system. It optimises dispatch and
mitigates geographical fluctuations and
volatility, contributing to an efficient and
affordable electricity system. Irrespective of
electricity supply decisions that are taken
for Belgium, further extending the cross-
border transmission system is therefore a
no-regret measure.
TOTAL SYSTEM COST COMPARISON WHEN MAXIMISING THE
CONTRIBUTION OF DOMESTIC RENEWABLES
The figure below demonstrates how boosting domestic RES (wind and solar PV) is very cost effective in all scenarios.
8 GW
4 GW
6 GW
2 GW
0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
119
118
118
120
117
117
114
123
117
112114
115 110
8 GW
4 GW
6 GW
2 GW
0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
114
113
112
114
114
113
110
114
115
109110
109 107
CENTRAL DOMESTIC RES
Onshore wind
x2 installation rate
With respect to offshore wind
in the Belgian EEZ, a capacity of
8 GW by 2050 is cost efficient
and considered in all simulated
scenarios
-9 to -2€/MWh
Solar PV
x2 installation rate
HIGH DOMECTIC RES
2050 DISTRIBUTED ENERGY TOTAL ELECTRICITY SYSTEM COSTS FOR BELGIUM IN €MWH
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ Accelerating the deployment of new domestic RES
proves to be cost effective independent of the capacity
assumptions for new large-scale energy sources in
Belgium.
◆ The downwards effect on system costs of the integration
of more domestic renewables is higher in scenarios
where less additional large-scale energy sources are
assumed.
WHAT DOES THIS TELL US?
Alongside the development of large-scale
energy sources, maximising Belgium’s
domestic RES (including offshore wind
in the Belgian EEZ), proves to be a very
cost-effective option in all scenarios.
Nevertheless, maximising the country’s
PV capacity will require tailored strategies
to manage potential overgeneration
challenges during certain periods. Spatial
limitations (notably related to offshore
wind) and public acceptance issues
(particularly related to onshore wind farms)
will play a crucial role in determining the
final potential of these technologies.
25
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary24
THE DEVELOPMENT OF FAR
OFFSHORE SOLUTIONS, COMPARED
WITH NEW NUCLEAR POWER
PLANTS, APPEARS TO BE MORE
ECONOMICAL IN MOST SCENARIOS.
Further harnessing the offshore renewable
potential of the North Sea appears to
be beneficial for Belgium. However,
the benefits of this option need to be
weighed against other supply options,
such as the development of new nuclear
generation units or connecting far-out
baseload RES. Important elements linked
to these options are the cost assumptions,
the time to market and the risk profile
(technological, financial, environmental,
etc.) of each technology.
SEE PAGE 174.
MANAGING THE SYSTEM’S
ADEQUACY WILL REQUIRE THE
DEVELOPMENT OF NEW THERMAL
CAPACITIES BY 2050. THE RUNNING
HOURS OF THESE PLANTS WILL BE
LIMITED 7002000 HOURS A YEAR.
The tools for managing adequacy were
put in place by the outgoing government.
The need for similar tools will be felt
throughout the analysed horizon.
However, the contribution that adequacy
measures make to the overall cost of the
future energy system is rather limited,
and technical solutions can be deployed
within a relatively short time frame (1 to
5 years).
SEE PAGE 192.
UNLOCKING AS MUCH FLEXIBILITY AS
POSSIBLE ACROSS THE SYSTEM TO MANAGE
ITS INCREASED VOLATILITY IS OF PARAMOUNT
IMPORTANCE. EFFICIENT MARKET ACCESS IS
CRUCIAL.
The energy system will become increasingly volatile.
The development of and access to different modes of
flexibility as well as the European integrated electricity
market will be key for coping with this volatility. This
will be essential for managing the energy system in the
most cost-efficient way and for limiting the economic
curtailment of RES.
SEE PAGE 186.
4 6
5
POLICYMAKERS CAN USE THE
FOLLOWING KEY INSIGHTS WHEN
TAKING DECISIONS RELATED TO
BELGIUM’S 2050 ENERGY MIX
SUFFICIENCY MEASURES HAVE THE
POTENTIAL TO REDUCE THE TOTAL
SYSTEM COSTS BY 15%.
The moderation of energy consumption
(sufficiency) holds a great deal of
potential in terms of keeping system costs
under control. Since this is predominantly
related to changes in human behaviour,
the main challenge of this measure lies
in citizen acceptance, particularly when
individuals believe that changes in their
behaviour will lead to a loss of comfort.
SEE PAGE 166.
MAXIMISING THE DEVELOPMENT OF
DOMESTIC RENEWABLES IS A
COSTOPTIMAL SOLUTION.
Even when accounting for the entire
system cost, maximising the development
of domestic renewable energy (onshore
wind, PV panels and offshore wind in the
Belgian EEZ) is demonstrated to be part
of a cost-optimal solution for Belgium in
all scenarios.
SEE PAGE 174.
THE MOST EXPENSIVE SCENARIO IS THE ONE IN
WHICH NO LARGESCALE SUPPLY SOLUTIONS
ARE DEVELOPED BY BELGIUM.
One key policy choice that should be made relates
to finding the right balance (over time) between
relying on electricity imports and undertaking
domestic investments in electricity supply. Numerous
considerations have to be made. These include:
affordability considerations, opportunities for the
redistribution of costs and benefits, agility in the face
of uncertainties, resilience against supply shocks,
international cooperation to ensure a coordinated
approach to offshore development, risks related
to budget and timing overruns, private – public
partnerships for financing, funding, etc.
SEE PAGE 174.
1
3
2
27
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary26
SHORTTERM ACTIONS THAT CAN CONTRIBUTE TO MEETING
SUPPLY NEEDS IN 2036
Promoting sufficiency and maximising the country’s domestic RES are part of a cost-optimal solution in the long run, and can play
an important role during the transition period as well. Other options that should be explored are developing access to non-do-
mestic offshore electricity and prolonging the lifespan of existing generation units (both thermal and nuclear). Since volatility
across the system will significantly increase, unlocking flexibility across the system is key for reducing system costs.
The figure below provides an overview of ongoing and short-term actions and their contribution (TWh) to meeting
Belgium’s supply needs in 2036 and 2050.
CONTRIBUTION TO
THE 50-60 TWh
SUPPLY NEED IN
2036
CONTRIBUTION
IN 2050
ONGOING ACTIONS: IMPLEMENTING CURRENT POLICIES
− Prolonging the lifespan of the Tihange 3 and Doel 4 nuclear units
(by 10 years). Already included
in basis
− Extending offshore wind in the Belgian EEZ to reach a capacity of
5.8 GW through the Princess Elisabeth Island.
− Further developing the transmission grid and interconnectors, and a
first batch of non-domestic offshore wind hybrid interconnectors.
7.5-15 TWh
15-30%
CONTRIBUTION IN THE SHORT TERM
Additional domestic RES + sufficiency
− Measures to speed up the deployment of domestic RES, as well as
actions to ensure their efficient integration into the power system.
9-15 TWh
15-30%
− Consumer moderation, synonymous with behavioural adaptations
and also known as sufficiency, is an opportunity for further reduc-
ing the final energy demand. This approach predominantly relates
to changes in human behaviour. Its implementation is hindered by
challenges related to its acceptance, particularly when individuals
believe that changes in their behaviour will lead to a loss of comfort.
Sufficiently long implementation lead times are required for encour-
aging it.
Up to 18 TWh
Up to 30%
Prolonging the lifespan of existing generation
− Further extending the operational life of the nuclear fleet beyond 2035
(subject to technical, safety and regulatory constraints) is a cost-effective
transitory solution. Whilst the prolongation of existing nuclear units
beyond 2035 seems to be cost efficient, it is only a transitory solution
and won’t to fill the supply gap in its entirety.
15-29 TWh
30-50%
− Next to contributing to adequacy, prolonging the lifetime of existing
thermal (gas) generation will contribute to the supply mix of Belgium.
The actual contribution to supply as well as the mix of (green and/or
fossil) molecules used in this type of generation is strongly dependent
on the energy landscape that will materialise in Belgium and abroad.
15-30 TWh
30-50%*
~
More imports
− An increased reliance on imports/foreign supplies could in any case
contribute to a (transitory) solution.
10-43 TWh
30-100%
*not necessarily carbon neutral
2050
As we navigate Belgium's energy landscape, it is essential to continue
implementing current policies and prioritising short-term actions that
accommodate the increasing electricity demand.
MESSAGE 4
Alongside long-term prepara-
tions, managing the transition
period will require some attention.
Cost-effective options include
maximising Belgium’s domestic
renewable energy sources (RES),
applying sufficiency measures,
prolonging the lifespan of existing
generation units and developing
the country’s access to non-
domestic offshore wind. Each
of these is subject to their
own specific constraints.
29
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary28
A FOCUS ON SHORTTERM ACTIONS SHOULD NOT REDUCE THE
URGENT NEED TO UNDERTAKE LONGTERM PREPARATIONS
◆ NEW NUCLEAR
If new nuclear capacity is envisioned, preparatory actions
such as the identification of potential sites, an analysis of
potential investment vehicles and the preparation of the grid
infrastructure should be initiated.
◆ NONDOMESTIC OFFSHORE WIND
In terms of the further development of non-domestic
offshore wind, existing and new international partner-
ships should be established, feasibility studies should be
performed, and existing barriers (e.g. financing) should be
addressed.
COST EFFECTIVENESS OF PROLONGING EXISTING NUCLEAR
GENERATION
2036 2040
Total electricity system costs [€/MWh] Total electricity system costs [€/MWh]
1000 EUR/kW - 7% 1000 EUR/kW - 7%1200 EUR/kW - 10% 1200 EUR/kW - 10%
99 10899 10898 10598 105
4GW 4GW
3 GW 3 GW
2GW
-2 to -5
-2 to -5
-2 to -4
-9 to -12
-7 to -10
-5 to -7
-4 to -7
-4 to -7
-3 to -5
-11 to -14
-8 to -11
-6 to -82GW
0 GW 0 GW
0 GW 0 GW0 GW 0 GW4 GW 8 GW4 GW 8 GW
€ €
extension of
impact when
extending…
impact when
extending…
extension of
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ The figure shows the total electricity system cost in €/
MWh when no existing nuclear generation is prolonged,
and when 2-3-4 GW of existing nuclear generation
capacity is prolonged:
-
For two time horizons: 2036 on the left-hand side and
2040 on the right-hand side;
- For two cost assumptions: reference cost on the left and
increased cost on the right;
-
For two levels of connected non-domestic offshore wind:
0 GW on the left and 4/8 GW on the right.
◆ Prolonging the lifespan of nuclear plants reduces
Belgium’s electricity system cost. While in 2036
prolonging more than 2 GW might not be financially
beneficial (especially in cases where the cost of extending
a plant is high), this possibility does become financially
beneficial in 2040 due to the increased demand for
electricity.
WHAT DOES THIS TELL US?
Extending the lifespan of 2 GW of nuclear
power plants is found to be economically
beneficial given the assumptions. The
benefit of prolonging additional reactors
depends on the Belgian electricity
demand. In any case, nuclear extensions
should be considered from a broader
perspective that encompasses more
than the costs involved: aspects such
as feasibility, safety regulations, grid
availability, socioeconomics, etc. should
all be accounted for.
2050
While some grid reinforcement projects are shown to be necessary
in all scenarios outlined in this study, the utility of other projects
strongly depends on the location of new or extended onshore power
plants and the level of offshore wind integration. In all situations, the
reinforced and completed 380 kV grid (backbone) is the basis for any
further developments.
MESSAGE 5
The future energy mix and
the location of future power
projects will play a crucial role
in the development of the
electricity grid. In all scenarios,
the reinforced and completed
380 kV grid (backbone) is the
basis for further developments.
31
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Executive Summary30
CERTAIN GRID INFRASTRUCTURE INVESTMENTS DEPEND ON POLICY
DECISIONS
Important grid investments, particularly in the Belgian backbone and offshore grid, depend significantly on policy decisions
regarding Belgium’s electricity mix.
While the Federal Development Plan 2024-2034 outlines investments in the backbone which are required up to 2035, decisions need
to be taken for the period beyond this. Given long project lead times, it is crucial to adopt specific actions that facilitate or initiate
preparations to activate relevant options.
The strong AC backbone network with its new links and high-temperature low-sag (HTLS) upgrades including Ventilus and Boucle
du Hainaut will remain instrumental for creating the possibility of connecting additional centralised generation or HVDC links.
REINFORCEMENTS NEEDED TO CONNECT DOMESTIC,
CENTRALISED GENERATION
On prolonging existing nuclear generation plants
If extending the lifespan of over 2 GW of existing nuclear units
is selected, the electrical infrastructure around current nuclear
sites needs to be prepared. Belgian nuclear phase-out plans
since 2003, the arrival of additional grid users nearby, and
changes to European legislation have reduced the grid host-
ing capacity for such extensions.
On new nuclear plants
Identifying potential future new nuclear sites is an essential
step. This involves preparing the most probable locations of
these sites and integrating them into the overall Belgian back-
bone.
REINFORCEMENTS NEEDED TO CONNECT ADDITIONAL
NONDOMESTIC OFFSHORE WIND
Hybrid offshore solutions and offshore hubs prove to be the
most cost-efficient approach for incorporating non-domestic
offshore wind into the Belgian electricity mix.
Collaboration with international partners, such as TritonLink
(BE-DK), BE-NO, BE-NL, BE-UK-IR, BE-FR, and BE-DE, is essen
-
tial for identifying promising options and establishing the ne-
cessary organisational structures and agreements to success-
fully implement chosen projects.
Addressing existing barriers related to potential hybrid systems
remains equally important (see the Elia-Orsted [ELI-9] and OTC
papers [OTC-1]).
The concrete developments related to connecting a first batch
of non-domestic offshore wind hybrid interconnectors will have
to be approved in the next federal development plan if the
aim is to have them commissioned before 2040. To connect
additional offshore wind beyond these, the east-west axis of the
internal backbone will have to be further reinforced.
NOREGRET GRID INFRASTRUCTURE INVESTMENTS
Certain grid infrastructure investments are no-regret measures and are resilient to changes in the choice of energy sources for
the electricity supply mix. Their main drivers are typically the electrification of demand and the development of domestic RES.
These investments should be prioritised and implemented without delay to avoid any potential setbacks in the energy transition.
STRENGTHENING DISTRIBUTION
AND LOCAL TRANSMISSION GRIDS
Following the electrification of resi-
dential end uses (such as electric vehi-
cles and heat pumps) and small and
medium enterprises, reinforcing both
the distribution grids and the local
transmission grid is essential. This rein-
forcement is particularly crucial in the
short term, as the electrification of local
end uses is expected to unfold primarily
over the next 10 to 15 years.
EXPANDING GRIDS FOR INDUSTRIAL
CLUSTERS
Similarly, the electrification of large indus-
trial clusters requires local extra-high-volt-
age grids to be significantly extended.
Despite uncertainties regarding some
industrial plans, there are enough existing
and potential grid users in these clusters
to justify the reinforcement of the grid. The
timely development of infrastructure, or
even its timely anticipation, is especially
vital in this context. Failure in this could
potentially lead to industries relocating
outside of the country due to a lack of
infrastructure.
REINFORCING ONSHORE INTERCON
NECTION
Onshore interconnectors continue to
play a crucial role in ensuring the effi-
cient dispatch of electricity, regardless
of Belgium’s dependence on imports.
Strengthening Belgium’s interconnec-
tion with its neighbours is highly cost
effective and yields significant benefits.
FUTURE ONSHORE CROSSBORDER INVESTMENTS
The figure below shows that cross-border investments partially depend on the energy visions that are selected both for Belgium
and its neighbours, particularly in terms of their mutual priorities.
Onshore cross-border investments: these are a no-regret, but a
link should be made with the vision for Belgium's energy future to
determine priorities
Reinforcing onshore interconnectors
4GW
2GW
0GW
8GW
16GW
12GW8GW0GW 4GW
6GW
8GW
6GW
4GW
2GW
0GW
16GW
12GW8GW0GW 4GW
More interesting
in case of less
offshore generation
directly connected
to BE
Invariant of
the offshore/
domestic
generation
More interesting in case of
more generation in Belgium
Note: this is the need for XB reinforcement calculated as the marginal benefit reducing European electricity costs
in a zonal setting. The impact on the Belgian costs can be different and should be further investigated.
NEW
8
6
4
2
0
GW
0 4 8 12 16
GW
Legend
Lower
need
Higher
need
WHAT DOES THIS FIGURE DEMONSTRATE?
◆ Further reinforcing Belgium’s interconnection with its
neighbours is a no-regret measure. However, the borders
that should be prioritised appear to depend on the
electricity mixes chosen for Belgium and its neighbours.
◆ For example, when large volumes of offshore wind are
harvested from the North Sea, the development of
further east-west interconnectors between Belgium and
Germany is crucial; if this is not the case in the medium
term, an increased access to countries in the north
should be prioritised.
33
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction32
1.
INTRODUCTION
1.1. OBJECTIVES
AN ENERGY COMPASS FOR BELGIUM
Achieving the commitment to reach a net-zero energy mix by
2050 requires a clear vision that should be implemented via deci
-
sive measures. Given the recent European, federal and regional
elections in Belgium, and clear signs that additional measures
are required at different political levels, the time is ripe to take
important decisions about the future of our energy system. As
Belgium’s electricity transmission system operator (TSO) Elia
has applied its modelling expertise in this study by quantifying
different possible energy pathways that Belgium could adopt
in the lead-up to 2050 and assessing the challenges associated
with each of these.
This study aims to illustrate the different electricity supply options
that are still open to Belgium whilst considering the influence of
other energy vectors on its power system. The study evaluates
a wide array of scenarios for Belgium and Europe, reflecting the
broad spectrum of potential futures that both face.
Instead of prescribing one single solution or setting out one clear
direction for Belgium and Europe to follow, this study outlines
the choices that policymakers face regarding our energy mix,
the effects of these choices on several crucial indicators such as
costs or imports, and the time frames related to these choices, to
ensure that sufficient time can be allocated to considering each of
them. It should be noted that the impacts of some choices cannot
be quantified, and certain pathways entail more uncertainties
than others. Policymakers should take this into consideration
when making decisions about the future of our energy mix and
what it will resemble in (the transition to) 2050.
Decisions about the period 2035-2050 must be taken soon, given
how critical it will be for Belgium's future energy supply.
◆ 15 European scenarios and
sensitivities
◆ 300 Belgian sensitivities
◆ A large set of quantified and
qualitative indicators calculated
PREPARING THE ELECTRICITY
INFRASTRUCTURE OF THE FUTURE
Elia is required to evaluate and identify future electricity grid
requirements to ensure that these can be met in an efficient man-
ner that is aligned with the interests of society. This is crucial, since
infrastructure projects often take several years to complete, and
decisions taken today influence how the grid will be developed
years down the line. With Elia’s next federal development plan as
a reference point in mind (see BOX 1-1 on Elia’s other studies for
more information), Elia is keen to outline the potential trajectories
that Belgium could adopt in the lead-up to 2050.
• Further inform the general
public and policymakers about
the impact of different visions
relating to Belgium's electricity
landscape
• First step for future federal
network development plan
post 2035
• Divergent scenarios BE/EU based on different
visions
• Focus on power system
• Specific strengths/characteristics: hourly
granularity, EU scope, physical grid constraints, …
• Grid infrastructure projects >10 years to build
• Need to define grid infrastructure corridors
• Highlight necessary steps and decisions in the
forthcoming legislation period
…carry expertise and
tools for scenario
building
…need sufficient time
to prepare an electricity
grid which is ‘fit for
purpose’
BELGIAN ELECTRICITY SYSTEM BLUEPRINT FOR 20352050
PROVIDING A COMPASS FOR POLICYMAKERS WHEN TAKING DECISIONS ABOUT THE ENERGY MIX
As an electricity TSO we…
1.1. Objectives 33
1.2. Context 34
1.3. Stakeholder interactions 40
35
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction34
1.2. CONTEXT
25 YEARS LEFT TO ACHIEVE NET ZERO:
AFFORDABILITY AND SECURITY OF SUPPLY
SHOULD BE FOCUSED ON WHILST WORKING
TOWARDS NET ZERO
As we shift away from fossil fuels, the electrification of our society
is happening at an unprecedented pace. This entails increasingly
ambitious goals for renewable energy. In addition, geopolitical
instability is straining our energy security and affordability (linked
both to the Russian invasion of Ukraine and the energy crisis).
Three facets of the energy system (also called the ‘energy tri-
lemma’) lie more than ever at the forefront of public debate:
security of supply, affordability and sustainability (see Figure 1-1).
Security of supply and affordability are essential for socioeco-
nomic prosperity. The recent energy crisis underscored the crucial
role of energy security; REPowerEU listed it as a key area of focus
- the first time in several years that it was highlighted. Depend-
ing on a single energy source or supplier increases a country’s
vulnerability and increases the risk of it being exposed to supply
interruptions. In addition to diversifying their energy sources, the
energy crisis demonstrated that countries should invest in solid
energy infrastructure to ensure a stable, affordable, and secure
energy supply for the future.
The fight against climate change is one of the most pressing
challenges facing the world today. The EU's commitment to cli-
mate action and the implementation of the European Green Deal
form important contributions towards limiting global warming to
well below 2°C, as outlined in the 2015 Paris Agreement.
THE ENERGY TRILEMMA FIGURE 11
Affordable and
available
Green and
clean
Secure and
reliable
The global focus on sustainable development and combating
climate change has led to a significant emphasis on the energy
transition and electrification. This shift is characterised by the
phasing out of fossil fuels, the adoption of renewable energy
sources (RES), advancements in energy storage technologies,
and the electrification of transportation and industrial sectors.
Governments, all sectors of the economy and communities are
prioritising energy transition initiatives to reduce greenhouse
gas emissions, enhance energy security, and foster a more sus-
tainable future.
CURRENT AMBITIONS AND TARGETS
European plans to mitigate climate change consist of a range
of measures adopted by European Union (EU) Member States.
For example, the EU has set emission reduction targets for the
next few decades.
The Union’s historical GHG emissions along with its targets are
illustrated in Figure 1-2.
2020
TARGETS 2040
TARGETS
2030
TARGETS 2050
TARGETS
By 2020, reduce greenhouse gas (GHG)
emissions by 20% compared with 1990 levels;
increase energy efficiency in the EU by 20%;
and ensure that 20% of total final energy
consumption in the EU is met by renewables
[EUC-1]. The EU successfully reduced its
emissions by 24% in 2019 and 31% in 2020
(due, in part, to the COVID-19 pandemic).
The EC recommended in Feb. 2024 to aim
for a 90% net reduction in GHG emissions
compared with 1990 levels by 2040 [EUC-
3]. The legislative proposal needs to be
submitted after the European elections and
then needs to be agreed on by the European
parliament and Member States.
The EU has adopted a set of proposals
to make its climate, energy, transport
and taxation policies fit for reducing
net GHG emissions by at least 55%
by 2030 compared with 1990 levels
[EUC-2].
The EU’s goal is to reach net-zero
emissions by 2050, meaning that
any remaining emissions are
counterbalanced by measures that
remove GHG from the atmosphere.
[EUC-4].
HISTORICAL EUROPEAN GHG EMISSIONS AND NETZERO TARGETS IN THE LEADUP TO 2050 FIGURE 12
1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
Net GHG emissions for Europe including UK, NO, CH.
‘Energy’ category: includes international aviation and 50% of international shipping.
‘Other’ category includes agriculture, waste management and other sectors.
‘LULUCF’ category includes Land Use, Land-Use Change and Forestry.
Source: European Environment Agency.
GHG emissions [MtCO
2
eq]
7000
6000
5000
4000
3000
2000
1000
0
-1000
-20 %
-55 %
Net zero
-90 %
(EC recommendation -
Feb´24)
-80 % (upper range
considered in this study)
Reduction target
compared to 1990
levels
Other
Industrial process
Energy
LULUCF
Net GHG emissions
At the European level, multiple policy measures, commitments and communications have therefore been released by the EU in
relation to the above targets. Some of these are included below (in chronological order).
12 DECEMBER 2015
Adoption of Paris Agreement, the legally binding international treaty on climate change at COP21
with the goal to hold “the increase in the global average temperature to well below 2°C above pre-
industrial levels” and pursue efforts “to limit the temperature increase to 1.5°C above pre-industrial
levels.” [UNF-1]
11 DECEMBER 2019
Presentation of the European Green Deal and adoption of the ‘Clean energy for all Europeans
package’
17 SEPTEMBER 2020
Presentation of the 2030 Climate Target Plan communication
9 JULY 2021
Adoption of the EU Climate Law, setting a binding objective for the Union to reach climate neutrality
by 2050 and a binding Union target for reducing net greenhouse gas emissions by at least 55% by
2030, compared with 1990 levels (in line with the European Green Deal communication of 11 December
2019)
14 JULY 2021
Presentation of the 'Fit for 55' legislative package, consisting of a major review of relevant EU
legislation on climate and energy which aims to enable the EU to reduce its net GHG emissions by at
least 55% by 2030 compared with 1990 levels and achieve climate neutrality by 2050
8 MARCH 2022 AND
18 MAY 2022
Proposal and publication of the REPowerEU plan, a joint European action plan which aims to ensure
Europe has access to more affordable, secure and sustainable energy
9 OCTOBER 2023
Adoption of two final pieces of legislation of the 'Fit for 55' legislative package, which includes relevant
updates to European legislation such as the revised Energy Efficiency Directive (EED) and revised
Renewable Energy Directive (RED), the updated EU Emission Trading System (ETS) and Effort Sharing
Regulation (ESR), and the new Carbon Border Adjustment Mechanism (CBAM)
14 DECEMBER 2023
Political agreement regarding a reform of the EU's electricity market design achieved
6 FEBRUARY 2024
Recommendation for the introduction of a 2040 emissions reduction target to set the path to climate
neutrality in 2050; political agreement regarding the Net-Zero Industry Act achieved
37
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction36
FUNDAMENTAL SHIFTS IN THE ENERGY SYSTEM
Figure 1-3 illustrates some of the major changes in the energy
system that are required to reach net zero by 2050:
◆ a reduction in total energy needs through sufficiency and
energy efficiency (with electrification being the most impor-
tant lever);
◆ an increase in the volume of RES integrated into the system
which will be complemented by other low-carbon sources;
◆ the massive electrification of final energy consumption, so
increasing the share occupied by electricity in the final energy
demand as well as increasing the consumption of electricity
in absolute terms.
The pace at which these changes are occurring has recently
accelerated significantly. Numerous countries have revised their
offshore wind ambitions, while the installation rate of solar photo-
voltaic systems in Europe continues to rise. For example, Germany
raised its 2030 offshore wind capacity target from 20 GW in 2020
[REU-1] to 30 GW in 2022 via the Easter Package [BMW-1]; the
Dutch Government raised in 2022 the target for offshore wind
capacity from 11 to 21 GW by 2030 [RVO-1]; while the total EU
solar photovoltaic capacity increased by 21% in 2021 compared to
2022 [SOL-1] and by 27% in 2023 compared to 2022 [SOL-2] with
Germany and Spain in the lead. Furthermore, the ban on the sale
of new light-duty fossil fuel vehicles from 2035 onwards will expe-
dite the adoption of electric vehicles, in turn leading to increased
electrification. Likewise, there has been a marked increase in the
installation of heat pumps across multiple European countries.
These changes will affect the energy supply mix as well as energy
consumption patterns across Europe.
CHANGES UNDERGONE BY THE ENERGY SECTOR IN THE LEADUP TO NET ZERO FIGURE 13
FROM TO
Electricity
Energy efficiency, Sufficiency and Electrification
Electrification
>> Share of RES
Total Energy Non RES RES
The total amount of energy
consumed will be reduced
through the use of additional
energy efficiency and sufficiency
measures but also through
additional electrification as it
mostly uses less energy to deliver
the same energy use
The share occupied by electricity
in final/end energy consumption
will increase with additional
electrification
Renewable generation will
increase both in the overall
energy mix and in the electricity
mix
Source: Inspired from ‘Increasing the EU’s 2030 emissions reduction target’ report from European Climate Foundation
and Climact.
Figure 1-4 illustrates the historical RES share (in the energy mix)
and RES-E share (in electricity consumption) in Europe and Bel-
gium. The EU’s target for 2020 was for RES to make up 20% of its
energy consumption - a goal that the EU achieved. In November
2023, a new binding RES share target of 42.5% was set for 2030
[EUC-5], with the EC estimating that the RES-E share would reach
69% in order to support the 'REPowerEU' plan [EUC-6].
Also in May 2024, Belgium planned its draft (updated) National
Energy and Climate Plan (NECP) to be submitted to the European
Commission by end of June 2024. The RES share calculated in this
plan would be 21.7% by 2030, based on the measures outlined in
the plan’s WAM scenario [BEL-1]. However, additional measures
should be taken by Belgium in case the country is required to
raise its RES share to 33% [BEL-1], as outlined in EU legislation
[EUC-7].
RES AND RESE SHARES IN EUROPE AND IN BELGIUM FIGURE 14
Share[%]Share[%]
80%
70%
60%
50%
40%
30%
20%
10%
0%
60%
50%
40%
30%
20%
10%
0%
20042004
2012
2012
2027
2027
2008
2008
2023
2023
2016
2016
2006
2006
2021
2021
2014
2014
2029
2029
2010
2010
2025
2025
2018
2018
2005
2005
2013
2013
2028
2028
2009
2009
2024
2024
2017
2017
2007
2007
2022
2022
2015
2015
2030
2030
2011
2011
2026
2026
2019
2020
2019
2020
Sources: Eurostat for historical values (EU-27 and Belgium).
(*) European Commission - REPowerEU Plan [EUC-5]
(**) European Commission - Renewable energy targets [EUC-6]
(***) Draft updated NECP for Belgium (‘Projet de mise à jour du Plan National Energie et Climat belge 2021-2030’) - Nov. 2023.[BEL-1]
(****) European Commission - Highlights of the Commission’s assessment - NECP Belgium. ‘33% - according to the formula set out in Annex II of the
Regulation (EU) 2018/1999 on the Governance Regulation of the Energy Union and Climate Action. ‘ [EUC-7]
Historical values
Historical values
Estimated target by EU to support the
REPowerEU plan (*)
Estimated target in Belgium’s draft
updated NECP of Nov. 2023 based on
WAM scenario (***)
Revised RES share target in Renewable
Energy Directive (RED) of Nov. 2023.(**)
Estimated target for Belgium according
to EU legislation(****)
Estimated target in Belgium’s draft
updated NECP of Nov. 2023 based on
WAM scenario (***)
2030 target
2030 target
41%
29%
RES-E share
RES-E share
RES share
RES share
69%
48.5%
23%
14%
42.5%
33%
21.7%
RES share of energy and electricity consumption in Belgium
RES share of energy and electricity consumption in Europe
39
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction38
PUTTING BELGIUM CENTRE STAGE
The challenge for Belgium lies in carving out an energy future that
ensures that it has access to a sustainable and affordable energy
supply. Belgium's strong industrial foundation and its well-con-
nected and interconnected methane and electricity systems are
key assets that have helped to achieve the welfare the country has
today. Additionally, the country is home to major ports, logistics
centres and industry hubs.
In addition to occupying a central position in terms of its geog-
raphy, Belgium occupies a central position in terms of Europe's
policy changes. It actively participates in and spearheads con-
versations at both industrial and political levels. This is evident in
its hosting of key events such as the second North Sea Summit
in Ostend in 2023 and coordination of the 'Antwerp Declaration
for a European Industrial Deal' in February 2024. Belgium is also
showing its innovation skills in energy infrastructure with the
plan to build the first energy island (Princess Elisabeth island) to
integrate more offshore wind energy and further interconnectors
into the system.
Given the fact that it is a highly densely populated country with
little to no economic potential for primary fossil fuel resources,
Belgium's primary energy is derived from renewable sources
(which also have a limited potential). Having phased out its coal
production during the twentieth century and with no indige-
nous sources of methane or oil on its territory, Belgium relies on
imports to meet over 90% of its primary energy supply as all fossil
fuels are sourced from abroad. Over 75% of the primary energy
consumed in the country still comes from fossil fuels like oil,
gas and coal. Similar figures are reflected in the country’s final
energy consumption.
Belgium’s primary energy supply was 52.3 Mtoe (which corre
-
sponds to 608 TWh) in 2022 [FPS-1] (including energy carriers
used for non-energy purposes such as petroleum used for pro-
ducing plastics and excluding international transport). Minus
transformation and losses, its final energy consumption was 36.9
Mtoe in 2022 (429 TWh). Figure 1-5 covers the shares occupied
by different fuels in the country’s primary and final energy con-
sumption in 2022.
Belgium’s final energy consumption has remained stable over
the past 10 years, as has the share of energy carriers:
◆ around 50%: oil;
◆ around 25%: methane;
◆ less than 20%: electricity;
◆ approximately 5%: direct heat, renewables and solid fossil
fuels.
PRIMARY ENERGY SUPPLY AND FINAL ENERGY CONSUMPTION IN BELGIUM IN 2022 FIGURE 15
52.3 Mtoe
(608 TWh)
Energy transformation
and other losses
Data for 2022
Source: FPS Economy - Belgian Energy Data Overview
Primary energy supply includes the energy consumed for non-energy purpose (feedstock) and excludes the energy used for international shipping. Note that
there is one missing category ‘Other’, which accounts for a negative energy supply of -0.7% in 2022 and which represents net imports of electricity and heat, as
well as heat recovery from chemical processes.
Finale energy consumption includes the energy consumed for non-energy purpose (feedstock) and excludes the energy used for international aviation.
Almost 70% of Belgium’s primary
energy supply was made up of fossil
fuels in 2022
Primary energy supply
18% of Belgium’s final energy
consumption was met by electricity
in 2022
Final energy consumption
36.9 Mtoe
(429 TWh)
Oil Gas Solid fossil Nuclear Electricity Heat RES
Interestingly, fuels for international aviation and shipping used
in Belgium amount to about 17 and 80 TWh per year respectively
(calculated as an average over the last ten years) [EUS-1]. This is
mainly due to ship refuelling (bunkering), with the Port of Ant-
werp being a significant hub for this in Europe.
Belgium has the fifth largest non-energy feedstock demand in
the EU, mainly explained by the presence of the petrochemical
cluster in the port of Antwerp with a heavy concentration of refin-
eries, chemical producers, and related industries that transform
crude oil and natural gas into a multitude of chemical interme-
diary and final products.
HOW THIS BLUEPRINT RELATES TO ELIA TRANSMISSION BELGIUM’S OTHER STUDIES
Elia has gained experience in energy modelling and sce-
nario development by producing prospective studies,
which require modelling methodologies to be constantly
improved and require performant data quality and data
management processes to be in place.
Studies performed by Elia linked to legal require-
ments
As required by the Electricity Act 1999, Elia publishes ten-
year adequacy and flexibility studies (AdeqFlex) on a
biennial basis. These explore the electricity system’s pro-
jected adequacy and flexibility needs for the following
ten-year period. Assessments of the system’s ‘adequacy’
explore whether the sum of expected available capac-
ities, including electricity imports, is sufficient to meet
Belgium’s reliability standard - or the necessary level of
adequacy. It should be noted that the study also assesses
the economic viability of the needed capacities. Assess-
ments of the system’s ‘flexibility’ investigate the extent to
which this capacity carries the right technical characteris-
tics to cope with future (un)expected variations in power
generation (in particular, power produced from RES) and
demand. The most recent ten-year Adequacy and Flexibil-
ity study was published in June 2023 (AdeqFlex’23) [ELI-1].
Elia has been mandated by law to publish Capacity Remu-
neration Mechanism (CRM) calibration reports which
contain information that is required to determine the vol-
ume of capacity to be contracted and proposed parame-
ters for each CRM auction. These calibration reports are
published every year in November in line with the Royal
Decree that sets out the method for calculating the vol-
ume of required capacity and the necessary parameters
for the organisation of auctions within the framework of
the CRM (‘Royal Decree on Methodology’) [ELI-2].
Elia is also responsible for writing and publishing quad-
rennial federal development plans and regional plans.
Each federal development plan covers a period of ten
years and includes a detailed estimate of onshore and
offshore 150-380 KV transmission capacity needs, along-
side an explanation of the assumptions and methods
used to calculate them. It also includes the investment
programme that Elia will need to implement to meet the
identified needs. The federal plan must be approved by
the Minister of Energy before being officially adopted.
The latest plan, which covers the period 2024-34, was
approved in May 2023 [ELI-3]. Given that Elia also owns
and operates the 30kV to 70 kV high-voltage sections of
the power grid which fall under the competence of the
different regions, a similar (but slightly different) process
of developing regional investment plans exists for Flan-
ders, Wallonia and the Brussels region.
Long-term prospective studies
Elia also produces ad-hoc system of the future studies
which cover longer periods of time (for example, up to
2050). In November 2017, Elia published its ‘Electricity
scenarios for Belgium towards 2050 – Elia’s quantified
study on the energy transition in 2030 and 2040’ [ELI-4].
The current Blueprint study is another example of one of
these long-term prospective studies.
Such studies are designed to complement existing stud-
ies that explore the lead-up to 2050 whilst focusing spe-
cifically on the Belgian electricity sector within Europe.
Additionally, Elia Group publishes specific viewpoint stud-
ies pertaining to a specific topic of the electricity value
chain. The viewpoint study of 2023 (‘The Power of Flex’)
focused on the barriers to the development of decen-
tralised flexibility whilst the View Point 2022 (‘Powering
Industry towards Net Zero’) offered a deep dive into the
electrification needs of industry as a result of their net-
zero ambitions. This year’s viewpoint will be focused on
European offshore development.
OVERVIEW OF ELIA TRANSMISSION BELGIUM’S PUBLICATIONS FIGURE 16
2010 2015 2020 2025 2030 2036 2040 2045 2050
[TWh]
200
160
120
80
Electricity demand
A lot of uncertainties
and choices to be
made
?
… to long term
ad-hoc vision studies
Y+10 - Y+25
Y+10
From regular legally
binding studies …
Studied horizons
in the Blueprint
Federal
Development Plan
Quadriennal
study. Last
published in May
2023.
Energy System
BluePrint
for Belgium
Published in
September 2024.
AdeqFlex study
Biannual study.
Last published in
June 2023.
Electricity Scenarios for
Belgium towards 2050
Last published in Nov.
2017
BOX 1-1
41
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Introduction40
1.3. STAKEHOLDER INTERACTIONS
1 The Elia academic board is collaborative platform launched in 2022 where Elia and the academic community can discuss the challenges and issues crucial to the future of
the Belgian energy sector [ELI-6].
INVOLVEMENT OF THE HORIZONTAL ELECTRICITY SYSTEM THINK TANK
As part of the stakeholder engagement process for this study,
Elia asked for feedback from partners in order to define its scope,
assumptions and methodology. Each of these was discussed
during sessions organised by the Horizonal Electricity System
Think Tank [ELI-5], which is made up of a wide range of energy
stakeholders in Belgium.
In addition to the plenary sessions held by the Think Tank in
September 2023, December 2023 and March 2024, Elia also organ-
ised 3 dedicated workshops that covered specific aspects of
the study:
◆ On 24 October 2023, the main methodological elements were
covered, alongside an overview of assumptions and scenarios.
◆ On 13 November 2023, the cost components were presented
by Compass Lexecon.
◆ On 13 December 2023, an update on the scenarios and meth-
odology was provided.
The workshops were complemented by a consultation of the
members of the Think Tank, which produced 9 replies and more
than 50 comments (see BOX 1-2 for further details).
A plenary Think Tank session was held on 19 December 2023, dur-
ing which the consultation comments were presented alongside
several adaptations to the models.
In addition to the above, Elia took part in a number of exchanges
with interested parties in order to discuss the study’s methodol-
ogy and assumptions. In particular, Elia worked closely with Fluxys
(the Belgian gas TSO) to define the main European scenarios
that should be used in the study. Several bilateral meeting were
held during the study's development to debate and discuss the
scenarios, methodology and results. Elia and Fluxys agreed on the
Ten-Year Network Development Plan 2024 (TYNDP 2024) scenario
framework from ENTSO-E/ENTSO-G (European Network of Trans-
mission System Operators for Electricity/Gas) as a starting point
for the study’s scenarios with relevant adaptations (see Chapter
3). In addition, other costs and parameters were further aligned
with Fluxys after the consultation.
The Elia Academic Board1 met in October 2023; during this, the
methodology and scope of the study were discussed.
The interactions with stakeholders lead to several key modelling
changes and improvements, as well as additional sensitivities to
be investigated. The main criticisms received during the consul-
tation phase were linked to the modelling choice to only model
electricity. Therefore, Elia expanded its model to all energy vectors,
not only electricity. This major change required the creation of
modules for hydrogen, methane, ammonia, liquids and carbon
emissions.
TIMELINE OF STAKEHOLDER INVOLVEMENT DURING THE DEVELOPMENT OF THE STUDY FIGURE 17
13/09
Think
Tank #1 Think
Tank #2 Think
Tank #3
Consultation
on data and
methodology
WS n°1
Methodology
& assumptions
& scenarios
WS n°2
Cost components
(Compass-
Lexecon)
Publication
and
presentation to
stakeholders
24/10 13/11 18/11 18/12 19/12 01/03 23/09
WS n°3
Improvements on
methodology,
assumptions &
scenarios
13/12
Performing the study
Fluxys Bilateral alignment
2023 2024
SUPPORT FROM CONSULTING FIRMS
Elia hired Compass Lexecon in order to estimate the total system
costs (and its components) and material usage associated with
future energy system scenarios. These were discussed during
the second workshop mentioned above and were included in
the consultation.
In addition, Sia Partners was hired to challenge and provide input
to Elia with regard to its scenarios (aviation & shipping, industry…)
and methodology. Sia Partners also carried out the work under-
lying and related to the Marginal Abatement Cost Curve (MACC)
(see Appendix E for more details).
Elia also asked VITO/EnergyVille to verify and provide feedback
on the assumptions and results used in the study. Their reasoned
opinion is also included in this report as appendix.
?
Stakeholders feedbacks
• 10 replies received
• More than 50 comments
Main points raised during the consultation
• questions related to the modelling of other vectors (hydrogen, heat, methane, liquids...)
• costs assumptions for certain technologies, WACC
• costs for non explicitly modelled vectors and scope of the cost assessment
• flexibility that can be harvested in heating networks, cogeneration...
• optimisation of investments in other generation assets than offshore and thermal
• simplifications that could be introduced in the geographical granularity
• clarifications regarding methodology, models, assumptions, scope of the assessment
• proposal for sensitivities
...
Main improvements applied after consultation
• Update of the costs based on the comments and more recent sources
• Alignment of the main scenarios to be used with Fluxys (based on the TYNDP2024)
• Addition of several sensitivities at Belgian and European level based on the comments
• Expansion of the modelling towards all energy vectors (initially only electricity was to be considered)
Documents submitted to consultation from 18/11/2023 to 18/12/2023
• Document providing explanations on the methodology, scenarios and input data
• Excel file with detailed input data for the costs prepared by Compass-Lexecon
Belgian offshore platform
EnergyVille
Engie
EDF Luminus
Essencia
Edora
Febeliec
Fluxys
FEBEG
GE Vernova
Think Tank
Academic Board
Input data / Generation
Input data / Total electricity demand
Input data / Demand Side Response
Input data / Economic Costs
assumptionsfor certain technologies,
WACC assumption
Input data / Scenariosfor Belgium
Input data / Scenariosfor Europe
Input data / Other topics
Methodology / General
Methodoolgy / Cross-border exchanges
Methodology / modelling ofother vectors
than electricity (hydrogen, methane, heat,
liquids,..)
Methodology/CO
2
computation
Questions on the modelling/clarifications
General comments
CONSULTATION FOR THE ELECTRICITY SYSTEM BLUEPRINT FOR 2035-2050
Three dedicated workshops
◆ 24/10/2023: scenarios and methodology
◆ 13/11/2023: costs
◆ 13/12/2023: update on scenarios and methodology
BOX 1-2
43
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology42
2.
METHODOLOGY
The methodology used in this study was developed based on the expertise Elia has
gained over the past decade via the publication of numerous studies as outlined in the
Chapter 1. As described in the previous chapter, the methodology was presented to and
discussed with stakeholders during several workshops and meetings and was subject
to a consultation. As a consequence, several novel approaches were then used as part of
this study.
The main changes compared with previous Elia studies lie in:
◆ the expansion of European multi-energy scenarios to feedstock, international aviation and shipping;
◆ the use of a capacity expansion model which optimises the location and the amount of selected technologies;
◆ the adoption of hourly/daily multi-energy modelling across the whole of Europe;
◆ the use of a flow-based zonal modelling for the electricity system to also reflect electric bottlenecks within countries;
◆ the consideration of all carbon emissions (processes, non-CO2, LULUCF, energy…) and options for capturing or using it.
Indeed, while Elia’s past studies have focused mainly on the elec-
tricity sector, the current study is the first multi-energy study that
Elia has performed for the whole of Europe. It should be noted
that ‘multi-energy’ refers to the explicit modelling of multiple
energy vectors (electricity, liquids, gases…). In addition, in order to
grasp the physical reality of the electricity network, this study uses
a flow-based model for the entire European perimeter (through
an equivalent zonal grid model).
This chapter includes a summary of the methodology used. Next,
an overview of the multi-energy framework that was developed
and is used in this study is covered. This chapter then ends with a
discussion about the creation of the marginal cost of abatement
curves. Several of the study’s appendices act as a complement
to this chapter: they include more information about specific
aspects of the methodology.
FEATURES OF THE MODELS USED IN THIS STUDY FIGURE 21
Dispatch includes storage, hydro and flexibility optimisation
Legacy thermal accounted for (existing units)
Zonal flow-based European model (EU27 + NO, CH, UK and the Balkans with >100 onshore nodes,
>400 offshore nodes/wind farms). Over 25k candidates for grid investment.
Endogenously optimised given the geographical scope
Endogenous investments for whole Europe in the grid (onshore and offshore), supply (thermal)
and offshore wind (location).
Sensitivities on other supply options
Includes energetic emissions, process emissions per sector, non-CO
2
emissions, LULUCF,
CCS/U and emissions stored in materials.
Europe (EU27+NO, CH, UK)
Carbon capture for power generation and process emissions per sector, reconversion of
thermal units to hydrogen units, permanent underground storage, direct air capture.
275 MT per year in 2050 (based on EC impact assessment)
Energy vectors
modelled
Final energy
demand
Investment
Temporal
granularity
Geographical
scope
Climate years
Goal function
Electricity dispatch
Electricity physical grid
constraints
Imports/exports
Investments
Emission accounting
Geographic extent
Investment options
Carbon permanent
storage constraint
Multi-energy model with several explicitly modelled energy vectors (electricity, H
2
, CH
4
,
ammonia, liquids & CO
2
options)
Fixed ex ante for each carrier explicitly (not optimised). Includes feedstock, international
aviation and shipping.
Investment options: H
2
and electricity infrastructure, cross-vector options (e.g. electrolysers,
thermal units...)
Hourly economic dispatch for electricity supply and demand
Daily (or 2-daily) for molecules supply and demand
Europe (EU27 + NO, CH, UK)
Import merit order for outside of Europe
Forward-looking database that includes climate change
• 3 CY for capacity expansion and dispatch
• 200 CY for electricity adequacy related results
Minimisation of total system costs at European level for a given carbon target (includes
optimisation of the CO
2
price).
Cross-vector features
Electricity features
GHG emissions modelling
2.1. Methodology: in a nutshell 44
2.2. Time horizons and simulation perimeter 45
2.3. The multi-energy model 47
2.4. Total system costs calculation 52
45
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology44
2.1. METHODOLOGY: IN A NUTSHELL
The methodology adopted for this study is presented in Figure 2-2. It can be subdivided into the following steps:
1. EU scenario definition - combining ex ante defined parameters, potentials and candidates for investments
2. Multi-energy capacity expansion and dispatch model – deriving the optimal mix and dispatch to meet the key character-
istics of the scenario defined in the previous step
3. Adequacy study – full adequacy check on all climate years for the required thermal generation in the power system
4. Application of Belgian sensitivities – changing the installed capacities of certain technologies for Belgium and re run the
multi-energy dispatch for whole Europe
5. Indicators assessments
OVERVIEW OF THE METHODOLOGY FIGURE 22
2
3
4
5
DE
ELEC+
GA
1EU scenario
definition
Multi-E economic dispatch
and capacity expansion
EU optimised
Infrastructure & supply Adequacy study
BE sensitivities with full multi-E
economic dispatch application
Dispatch
• Multi-E energy mix
• Energy exchanges
• Security of supply
• Import dependence
Sustainability
• CO
2
emissions
• RES shares
• Bio and e-fuel share
• Material needs
Economic
• System costs
• Supply
• Infrastructure
• Import costs
Assessment of
indicators
Based on scenario storylines capacities are quantified for RES, demand per vector (electricity, hydrogen, liquids, methane, …) , DSR,
thermal generation, storage, interconnectors, costs for import and domestic production of fuels.
Combination of:
• Market dispatch optimisation: electricity supply, flexibility, molecule imports and
domestic supply
• Optimisation of infrastructure: onshore grid, offshore wind, offshore grid, hydrogen
grid, direct-RES hydrogen, carbon-capture and storage technologies, electrolysers,…
• All under the constraint of a given target.
Adequacy study for electrical supply with an increased
number of ‘Monte Carlo’ years with a reduced granularity
in electrical grid data and with optimisation of required
thermal infrastructure to meet SoS targets.
Application of sensitivities for
Belgium. Sensitivities comply
to SoS criteria.
Post-processing of results.
...
...
...
CCS
Potential additional sensitivities
Initial infrastructure Infrastructure
CH4H2
==
MOL
EUR-
non
CO
2
+NUC
RAD
H2
1.
Based on scenario storylines, multi-energy demand vec-
tors and supply capacities, costs and (import) potentials are
defined for Europe. These quantified scenarios are supple-
mented with additional sensitivities which aim to identify the
effects of changing certain input variables. The final demand
for each energy vector is therefore defined ex ante (there is
no optimisation between energy vectors for a given energy
usage). However the additional consumption from one vector
to another or storage losses is endogenously modelled (car-
bon capture and storage/utilisation or CCS/U, P2x, batteries…).
2. In a second step, the quantified scenario is subject to a com-
bined optimisation of the multi-energy system dispatch,
capacity expansion and location of key technologies such as
electrical and hydrogen transmission infrastructure, thermal
generation based on molecules, electrolysis, offshore wind
and carbon abatement technologies. This optimisation deter-
mines the energy system which can provide the total energy
load at the lowest cost subject to a European GHG emission
target. The optimisation is performed sequentially and starts
in 2036, meaning that the results of previous target years are
used to initialise subsequent ones.
3. New electricity generation capacity is added where needed
to reach the security of supply criteria (adequacy). For this
simulation step, a model with a reduced level of geographical
granularity is used to enable the timely calculation of the large
amount of climate years needed for an adequacy assessment.
4. Starting from the models obtained after step 3, sensitivities
are created to assess the effect of different choices for Bel-
gium’s domestic electric supply. These are further detailed
in the scenario chapter.
5. The results are then further processed to extract key indica-
tors related to dispatch, sustainability, and economics. These
also include an assessment of the infrastructure costs for
electricity for each scenario.
2.2. TIME HORIZONS AND SIMULATION
PERIMETER
TEMPORAL GRANULARITY 2036, 2040 AND 2050
Three time horizons are explicitly modelled in this study: 2036,
2040 and 2050. These time horizons were chosen for their rele-
vance for the Belgian energy system:
◆ 2036 was chosen to represent the year in which, according to
the current legal framework, the last nuclear reactor would
close in Belgium;
◆ The availability of input data from other studies and Euro-
pean (proposed) targets (2040, 2050) explain the choice for
the other two years.
The optimisation of the energy system is performed sequentially.
As such, investments from previous time horizons are considered
in the initialisation of the optimisation of subsequent years. This
process is schematically presented in Figure 2-3.
SEQUENTIAL OPTIMISATION OVER TIME FIGURE 23
Investments of previous time horizons are considered in the next optimisation
Start
Optimise for 2036
starting from
initial situation
Optimise for 2040
starting from
2036 optimum
Optimise for 2050
starting from
2040 optimum
Stop
Initial
situation 2036 2040 2050
Other aspects of the methodology can be found in the appendices with details about:
◆ the KARI electricity zonal model (Appendix A)
◆ the molecule and liquid model (Appendix B)
◆ the carbon capture, utilisation and storage model (Appendix C)
◆ the adequacy model (Appendix D)
◆ the marginal abatement cost curve (MACC) model (Appendix E)
◆ the total cost methodology (Appendix F)
◆ the schematic view of the model (Appendix G)
◆ the non-CO2 emissions methodology (Appendix H)
H2
47
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology46
GEOGRAPHICAL COVERAGE WHOLE OF EUROPE
The perimeter for this study includes most of Europe. Depending
on the energy vector being simulated, a different geographical
and temporal granularity is used. The geographical perimeter
and granularity for each of the vectors is shown in Figure 2-4.
In order to accurately capture the benefits of additional electrical
interconnector capacity, a well-chosen geographical subdivision
and sufficiently short time step is necessary. For this reason, the
finest granularity in terms of both geography and time is used
for the modelling of the electricity system. In total, over 500 zones
are individually modelled across Europe: over 1800 thermal units
(grouped into more than 600 clusters), more than 600 profiles
for wind and solar and over 25,000 possible grid reinforcements
are assessed using an hourly time step.
For hydrogen and methane, a longer time step can be justified
since the storage possibilities inherent to the system (line packing
for gases, storage tanks, etc.) are sufficient to cover a period of
several days. In addition, flows for these other vectors are highly
steerable so a coarser geographical granularity still allows for
accurate simulations of energy exchanges. Therefore, a daily or
twice daily time step (every two days) and a close-to-country level
granularity is used for these vectors. In total 32 zones and one
direct RES offshore zone are modelled, imports from pipelines (6
corridors for H2 and 7 for CH4), ammonia (4 geographical origins)
and LNG (4 geographical origins and choice between fossil and
synthetic) are taken into account on top of domestic bio and fossil
production. Steam methane reforming (hereafter called SMR-H
2
)
with CCS, direct-H2 offshore RES and over 50 potential hydrogen
pipeline interconnectors as well as several processes related to
the conversion of (molecule) energy vectors are assessed (Haber-
Bosch, ammonia cracking, methanation, hydrogen to liquids, …).
Those conversion processes are discussed in Section 3.1.6 and
illustrated on Figure 2-5.
Finally, for liquids, a similar line of reasoning as the one adopted
for hydrogen and methane can be followed regarding the tem-
poral time step. Given that the transportation of liquids is well
established through multiple modes (pipelines, trucks, tankers,
trains, etc.) the assumption taken is that there would be no con-
gestions for its transport within Europe, leading to the assump-
tion of a single node. On top of potential domestic fossil and bio
production, 17 import options from 4 geographical regions are
assessed including fossil liquids, synthetic liquids and jet fuel. The
conversion of hydrogen to liquids is explicitly assessed.
MULTIENERGY MODEL AND GEOGRAPHICAL SCOPES FIGURE 24
Electricity model H2 model CH4 model Liquids model
> 100 onshore zones
Over 400 offshore zones
>25 000 interconnectors assessed
H
2
H
2
H
2
H
2
H
2
H
2
H2
> 20 Zones
1 direct RES offshore zone
Onshore and offshore pipelines assessed
Imports possible via sea terminals for
coastal areas
Imports possible via pipelines from Ukraine
and Africa
10 Zones
Existing grid kept constant
Imports possible via LNG terminals for
coastal areas
Imports possible via pipelines from Russia/
Ukraine and Africa
Domestic production of CH
4
(bio and/or
fossil) modelled in all zones
Liquid supply modelled on an EU level
Import of liquids possible via terminals
Domestic production of liquids (bio and/or
fossil) modelled
CH
4
CH
4
CH
4
CH
4
CH
4
CH
4
CH
4
The entire energy requirement for Europe is taken into account,
which encompasses feedstock, international aviation and inter-
national shipping. These components are frequently overlooked
or excluded from future-oriented studies, as they are often not
included in domestic emission calculations.
◆ Feedstock: This term pertains to the starting materials for a
variety of chemical processes and are typically converted into
a wide range of products in the chemical industry, including
plastics, fertilisers, pharmaceuticals, and other chemical com-
pounds. Examples of chemical feedstocks include methane,
oil, coal, and biomass.
◆ International Aviation: This involves the energy utilized by
airplanes entering or leaving Europe. It's crucial to incorpo-
rate this into energy demand estimates, as the aviation sector
consumes a significant amount of fossil fuels and substan-
tially contributes to greenhouse gas emissions.
◆ International Shipping: This, much like aviation, relates to
the energy used by vessels for global transportation of goods
or passengers to and from Europe. Including the shipping
industry, another significant energy consumer, ensures a
more inclusive perspective on Europe's energy demand.
2.3. THE MULTI-ENERGY MODEL
The aim of the model is to find the European cost optimum across all energy vectors for a given carbon target by:
◆ optimally dispatching the necessary production assets imports on an hourly basis for electricity and daily for other vectors;
◆ optimising the needed infrastructure and the amount of certain production technologies.
One key novelty of the approach used in this study compared
to previous Elia studies is the explicit integration of a multi-en-
ergy dispatch modelling framework. This framework enables the
simulation of the exchange of energy vectors such as hydrogen,
methane, and liquids along with the electricity dispatch.
This multi-energy model is coupled with a carbon emission model
which carries the capability of enforcing a GHG emission tar-
get and deriving an associated carbon price. Starting from the
emissions resulting from the multi-energy dispatch, the carbon
emissions model invests cost optimally in carbon abatement
options (CCS, CCU, conversion of CH
4
to H
2
turbines, Direct-Air
Capture (DAC), ...) to reach the GHG emission target. Inherently
this causes the tool to define a shadow cost of carbon (the cost
of the marginal carbon abatement technology). This shadow
cost of carbon is then used in a series of subsequent iterations
of the molecule models (H2, CH4, liquids), enabling the import
of low-carbon molecules over fossil ones (which are expected to
remain cheaper when not accounting for a carbon price). The
cost optimal molecule dispatch and carbon abatement tech-
nologies selection is then used in the next electrical model runs.
The models and their interactions are represented in Figure 2-5.
SCHEMATIC REPRESENTATION OF THE MULTIENERGY MODEL AND ITS MODULES FIGURE 25
Methanation
Other
uses
Fischer-Tropsch
SMR-H2 + CCS
Methane
Pipeline imports
+ domestic fossile
LNG imports
FR
BE
DE
...
CO
2
CO
2
CO
2
CO
2
CO
2
CO
2
CO
2
Hydrogen
ElectricityOther fuels
Pipeline
imports
FR
BE
DE
...
Power
generation
Power generation
Power generation
Power generation
Power
generation
Power
generation
Ammonia
cracking
ElectrolysisHaber-Bosh
Liquids
Aviation
Shipping
Feedstock
Other
EU
EU
EU
EU
Imports
outside EU
Fossil liquids
from EU
Ammonia
EU
*Bio
*Bio
Bio/waste
Nuclear
Coal &
lignite
Imports from Middle-East
Imports from Africa
Imports from Australia
Imports from South-America
Transport via pipeline
Pipelines the model can invest in
Transport via shipping
Emissions from the subsystem/
transformation
Emissions absorbed by the
subsystem/transformation
Transformation
Transmission of electricity
CO
2
CO
2
FR1
BE1
BE2
FR2
...
...
BE3
NL1
OFW1
*Bio
CO
2
CO
2
49
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology48
The multi-energy model comprises three modules:
◆ The electricity dispatch and investment model (KARI) – see
Appendix A;
◆ The molecule model (H2, liquids, CH4 and ammonia) – see
Appendix B;
◆ The carbon capture and storage model – see Appendix C.
This study utilises a model of high complexity. For instance, the
electrical model considers approximately one million constraints
and over three million variables for each weekly optimisation
problem. Furthermore, the model evaluates more than 25,000
investment options for the electrical grid. Due to its intricacy -
which is ten times greater than the model used in the Adequacy
& Flexibility study - it's run iteratively until an optimal result is
achieved. This optimum is identified as the iteration where the
total system costs reach a minimum for a specified GHG emission
target. The interactions between the different models and the
iterative process are outlined in the Figure 2-6 below.
The iterative process (see Figure 2-6) consists of:
1. A run of the hourly electricity model for three climate years:
a set of electricity divestments/investments is identified and
applied.
2.
The results of the hourly electricity model run are aggre-
gated temporally and/or geographically and integrated in
the molecule and CO2 model. The exported results contain
the molecule consumption for electricity generation and
hydrogen production via electrolysis in each of the zones of
the molecule model as well as the unit profitability to deter-
mine the economic viability of the conversion of conventional
units to CCS or hydrogen. Finally, electricity marginal costs are
exported to estimate costs of the, where relevant, additional
electricity consumption of carbon abatement technologies.
3.
Several iterations (a daily temporal granularity for multiple
climate years) of the molecule model and carbon capture
model are carried out in order to find the optimal molecule
infrastructure, fuel prices, CO2 prices, CCS/U and imported
fuels.
4.
The outcome of the optimal molecule & carbon capture
model is introduced into KARI (prices, additional electricity
demand, etc... see the next subsection for details) and a new
iteration can be performed.
5.
The optimum reached for the combined system can then be
used for further sensitivities and/or analysis.
In order to speed up the iterations, the user can also choose to
simulate clustered weeks instead of full year hourly simulations.
ITERATIVE PROCESS USED IN THE MULTIENERGY MODEL FIGURE 26
Total system costs
(including molecule costs)
For each iteration of the electricity model
Iteration of the electricity model
1
5
2
34
Optimum for
the entire
system
CO2 model
Optimum of the
molecule model
Export
results of the
electricity
iteration to
the molecule
models
In each electrical iteration, the framework optimises dispatch,
invests in grid, in offshore wind farms and in electrolysers,
calculates adequacy benefits of new investments, ...
Optimal system
ready for
application
of sensitivities
and further
analysis
Iterations of the electricity
model
- Molecule demand for electricity
generation
- Generation of molecules (P2X)
- Electricity prices
- Unit profitabilities
- Daily molecule prices
- Carbon price
- Electricity consumption for carbon
abatement
- Conversion of thermal power plants
Results of optimal molecule
model are fed back to the
electricity model.
H2 model
CH4 model
Liquids model
For each electrical iteration, the model fully optimises for all
molecule models (H2, CH4, CO2, liquids and other fuels) the dispatch,
the conversion of energy vectors (Haber-Bosch, methanation...),
imports, hydrogen grid, carbon abatement, conversion of thermal
plants to CCS and/or hydrogen turbines, ...
...
...
Molecule system costs
Iterations of the molecule model
new offshore-offshore
new onshore-offshore
new onshore-onshore
2.3.1. INPUTS, OUTPUTS AND OPTIMISATION
Several parameters are fixed ex ante in the model. Other parameters are optimised by the model.
Ex ante defined parameters:
◆ Final demand for each energy carrier, including feedstock,
international aviation and shipping;
◆ Electricity storage capacities and demand-side flexibility;
◆ Installed PV, onshore wind, biomass, nuclear and starting leg-
acy thermal (existing gas and coal units) capacity;
◆ Potential for new offshore wind farms;
◆ Climate years used to create hourly/daily consumption pro-
files and production profiles;
◆ Potential (price and quantities) for each molecule carrier
(imports of liquids and molecules; domestic biomethane,
domestic/imported fossil fuels);
◆ Conversion efficiency, electricity requirements and costs
between energy vectors and other technical parameters
related to the conversion technology (for example, P2X,
Haber-Bosch, SMR-H2 but also CCGTs, nuclear powerplants…);
◆ Costs for each technology that can be invested in and variable
and operating costs;
◆ Reference (starting) grid, existing methane grid;
◆ Process emissions and related potential for CCS;
◆ Non-CO2 emissions and other non-modelled sector emis-
sions;
◆ GHG emission target that needs to be reached at EU level;
◆ Maximum CO2 permanent storage per year.
Optimised by the model:
◆ Electricity high-voltage grid (between countries and within
countries), including offshore grid - both AC and DC reinforce-
ments are accounted for;
◆ H2 grid between countries and to North Africa/Ukraine (no
intra-bidding zone assessment), storage of H2;
◆ Location and type of new thermal generation (H2, CH4…) for
electricity generation;
◆ Location and capacity of new offshore wind;
◆ Location and capacity of new electrolysis for each electrical
zone;
◆ Creation of synthetic fuels from hydrogen (ammonia, metha-
nol, e-kerosene, e-methane…);
◆ Usage of liquid biofuels and biomethane;
◆ Carbon price (shadow carbon price of the model);
◆ Carbon capture in industry (process), electricity (thermal gen-
eration), industry (energy), SMR-H2 ;
◆ Energy dispatch of each carrier;
◆ Amount of imported fuels of each type, marginal prices.
51
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology50
WHAT IS OPTIMISED IN THE PRESENT STUDY ? FIGURE 27
Localisation
optimisation
INPUTS MODEL
Demand
Supply
Grid
Emission
Other parameters
Database
Final energy demand for each energy carrier
Process emissions and related potential for CCS
Energy conversion & costs
Climate years
Costs for each technology
Non-CO2 emissions & other sectors emissions
EU GHG emission target
Max yearly CO2 permanent storage
Reference electricity grid and methane grid
Including feedstock and international aviation
and shipping
Location (>400 candidates) &
amount
Amount (for adequacy) & type
(existing, H2, CCS...)
Location (zonal level) & amount
Inter-zonal (AC and DC)
& offshore grid
Inter-country capacities
Location & amount (CCS, DAC...).
CCS in process and usage of CO2
in synthetic liquids, materials
Shadow cost calculation
Electricity: for each hour
(generation, storage, flexibility,
imports/exports). Molecules:
for each (2-)day
Amount per type & from where
Conversion efficiency and costs between energy
vectors and other technical parameters related
to the conversion technology
(P2X, Haber-Bosch,…)
Climate years used to create hourly/daily
consumption profiles and production profiles
Costs for each technology that can be invested
in and variable and operating costs
Storage & flexibilty
Offshore wind quantity
and location
Thermal quantity and type
Electrolysers
Electricity high voltage grid
H2 grid & storage
CCS/U
CO2 prices
Electricity & molecule dispatch
Amount of imported fuel per type
PV, onshore wind & biomass
Potential for imported fuel per type for each
molecule carrier
Nuclear and starting legacy thermal
(existing gas & coal)
Potential for new offshore wind
CAPEX
optimised
OPEX
optimised
OP
EX CAP
EX
2.3.2. THE KARI MODEL (ELECTRICITY ZONAL MODEL)
The KARI model, which is based on the open-source Antares-Sim-
ulator, is described in more detail in Appendix A.
In a nutshell, the model is a Unit Commitment model of the entire
European electricity system on a zonal basis. It applies flow-based
constraints based on a reduced equivalent grid model. It therefore
captures grid constraints and flow patterns more accurately inside
and between countries. The grid model is based on the data used
for the TYNDP 2022 due to the unavailability of the TYNDP 2024
grid model when performing this study. The model was further
expanded to account for more granular offshore zones.
In total there are:
◆ >100 onshore zones considered in Europe;
◆ >400 (potential) new offshore wind farms considered;
◆ >25,000 potential transmission candidates from/to all
onshore/offshore zones considered.
The model calculates the optimal dispatch for the electricity
system including the production of hydrogen and can make
investments in infrastructure such as onshore (AC and DC) inter-
connectors, offshore HVDC interconnectors, hybrid interconnec-
tors, offshore wind farms, wind farm platform extensions and
electrolysers. For a more elaborate explanation see Appendix
A - KARI dispatch and investment electricity model.
After stakeholder feedback, the KARI model was linked with other
models as described at the start of Section 2.3 (for a more detailed
overview see Appendix G – Schematic view of the model). As such,
the results of the KARI model dispatch are taken into account
for the consumption and production of fuels (see earlier in this
section), the system costs (see section 2.4), the calculation of
the GHG emissions and the calculation of the costs of carbon
abatement options (see section 3.1.5).
This linkage between the models means that their dispatches
are consistent. As such interactions between for example mole-
cule consumption to generate electricity and the resulting effect
on the prices of molecules is taken into account. Inversely, the
prices of molecules will also influence the profitability and as such
installed capacities of electrolysers in the electricity model. The
final interactions that are observed between the different energy
vectors are further described in chapters 4 and 5.
The main inputs and outputs of the KARI model are shown in
Figure 2-8, where the dots represent existing or potential future
offshore wind farms.
MAIN INPUTS AND OUTPUTS OF THE KARI MODEL FIGURE 28
INPUTS OUTPUTSEUSCALE ZONAL ELECTRICITY MODEL KARI
Hourly electricity
demand, PV, wind...
including flexibility
options
Electricity mix
System Costs
(CAPEX + OPEX)
CO
2
emissions
Onshore & offshore
transmission grid
Cross-vector capacities
and location (thermal,
P2X)
Marginal prices
…
Physical constraints on
the European
HV grid
Investment candidates
(offshore, grid,
electrolysers, thermal...)
Thermal capacities,
constraints, fuel and CO
2
prices
Hourly molecule prices
…
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Methodology52
2.4. TOTAL SYSTEM COSTS CALCULATION
Total energy system costs are quantified in this study. Depending on the type of analysis, the scope will differ:
◆ The geographical scope: Belgium only or Europe;
◆ The type of costs and sectors that are accounted for in the analysis.
Other types of benefits or costs (e.g. socio economic impact, employment) are excluded from the analysis.
The main assumptions behind the costs are further detailed in Section 3.3.
System costs are defined in four different parts:
◆ End use investments (e.g. investments in electro-mobility,
energy efficiency…):
-
Those are the investments made by the end users of the
energy, and include the cost for acquisition of new cars,
charging infrastructure, heating device or renovation in
buildings;
◆ Energy vector – molecules (oil, methane…):
- Costs related to molecule infrastructure (grid and transfor-
mation processes);
-
Costs related to the molecule supply (e.g. imports, domestic
production but excluding the fuel used for electricity gener-
ation and fuel generated from electricity);
◆ Energy vectors – electricity:
- Electricity grid costs (offshore, backbone (high-voltage grid
within a zone), regional grid (also called ‘vertical’ grid: 150-
70-36-30 kV), DSO grids);
- Electricity supply capital expenditure (CAPEX) costs (invest-
ments and fixed costs of production facilities);
-
Electricity supply operational expenditure (OPEX) costs (fuel
used to generate electricity).
In this study there are mainly three types of comparisons:
◆ When comparing divergent demand scenarios, the totality
of the system costs is accounted for. This includes all energy
vectors (CAPEX and OPEX) but also end use CAPEX. This will
be referred as ‘total system costs’;
◆ When comparing different scenarios within a similar demand
scenario, end use CAPEX costs are the same (defined ex ante
via the demand scenario) and only the costs related to the
CAPEX and OPEX of the energy system will be shown. This is
further called ‘energy system costs’;
◆ When only comparing electricity sector for Belgium costs,
the electricity costs and interfaces to the other vectors are
accounted for (power to X and X to power). In this case end
uses are the same across all scenarios and can be omitted for
the comparison. This is called ‘electricity system costs’.
Depending on the analysis, the costs can relate to either Europe
or Belgium. Figure 2-9 summarises the different approaches.
SYSTEM COSTS DEFINITION USED IN THIS STUDY FIGURE 29
End-uses CAPEX
Molecules grid &
transformation
CAPEX
Molecules
(excl. elec. supply)
OPEX
Electricity grid
CAPEX & OPEX
Electricity supply
CAPEX
Electricity supply
OPEX
Electricity Electricity
system costs
Energy system costs
Total System costs
Comparing supply
sensitivities for
electricity
Comparing
sensitivities within the
same final demand
scenario
Comparing final
demand scenarios
between each
other
Molecules
End uses
is kept fixed when
comparing supply
sensitivities for electricity
only changes if the
demand scenario
changes
• Offshore
• Backbone
• Regional
• DSO
55
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios54
OVERVIEW OF THE BLUEPRINT SCENARIO FRAMEWORK FIGURE 31
Demand
CENTRAL
GA DE
Supply & import
Global
ambition Distributed
energy
GHG sensitivities
European scenarios and sensitivities
Onshore supply option
Offshore and grid
sensitivities
Molecule import option
CO2
80%
non
CO2+
NUC
PV+
RES+
NIM
MOL
EUR+
MOL
EUR-
RAD
RAD+
400
GW
150 GW nuclear in EU
in 2050 (instead of 75 GW)
+100 GW/year for PV. 2,700 GW
in 2050
Acceleration for domestic PV
and onshore wind (+25 GW/y
for wind and +75 GW/y for PV)
Nimby scenario (lower
onshore +10 GW/y & more
expensive onshore grids)
Higher costs for imported
molecules
Lower costs for imported
molecules
Only national radial
Only radial, international allowed
Connect 400 GW offshore
80% at EU level
instead of 90%
in 2040
Lower ambition
on non-CO2 &
LULUCF
Demand & flex options
Over 15
scenarios
Over 300
sensitivities
for the
electricity
demand/
supply
Over 9 cost
combina-
tions for
each
technology
ELEC
FLEX
Increased electrification
x2 the installed demand
flex/storage in Europe
SCENARIOS OF TYNDP 2024
Multi-energy capacity
expansion
Electricity
European scenarios and sensitivities
Belgian electricity sensitivities
SCENARIOS OF TYNDP 2024
AS STARTING POINT
GA DE
Global
ambition Distributed
energy
Demand
Supply
Investment
costs
ELEC, SUFF and HEAT scenario
CENTRAL, PV+ and RES+
0 to 16 GW offshore in 2050 or 'far-out baseload' RES
Extension existing capacity and up to 0 to 8 GW new in 2050
Ex ante defined linked to demand and amount of PV
Enough to comply with SoS criteria
Low – medium - high
Results of the choices above
4% - 7% - 10%
Domestic RES
Non-domestic RES
Nuclear
Flexibility
Adequacy
CAPEX & FOM
Import
WACC
+
In order to lay the ground for establishing different scenarios, this section first outlines
storylines. Storylines are high-level overviews of what the energy landscape could
look like in the future (including the context, energy ambitions, etc.). These are then
translated into concrete datasets - or scenarios - for different components (supply,
demand, the grid, climate years, etc.) for the different countries that are simulated and
in the needed granularity. Scenarios are not used to predict the future; instead, they
cover a range of plausible futures so that the impact of these futures can be assessed.
As explained in Chapter 2, one part of the scenarios is fixed ex
ante, while another part is optimised throughout the simulations
(the constraints/ranges that are used for the optimised variables
are provided below). The optimised scenarios are the result of the
simulations and are provided in chapters 4 and 5.
The scenario framework is divided into 3 parts, as outlined below
and in Figure 3-1.
◆ The European framework (see also Section 3.1). This is
needed to assess changes in multi-energy consumption and
supply at European level and the impact this will have on Bel-
gium. In order to grasp different views relating to demand,
supply and ambitions, several sensitivities are simulated.
These cover:
- supply options;
- offshore supply and grid development options;
- the import costs of molecules from outside of Europe;
- the level of electrification and flexibility;
- carbon target options.
◆ Given that this study focuses primarily on Belgium and
future electricity supply options (see also Section 3.2), a
large amount of sensitivities (over 300 combinations) are
investigated. These consist of a combinations of:
- sufficiency measures related to consumption;
-
more domestic RES (PV, onshore and offshore wind installed
in Belgium);
-
the connection of foreign offshore wind to Belgium or far-out
RES (not in Europe);
- new nuclear units in Belgium as well as the extension of its
current fleet beyond 2036;
-
the impact of flexibility options (storage and demand
response) and the calculation of the need for additional
thermal capacity to comply with the adequacy criteria;
- imports that have resulted from the above.
◆ Financial assumptions are also covered (see also Section 3.3).
Indeed, these are used for the European optimisation and
are used when quantifying the costs of the different Belgian
sensitivities. Several cost sensitivities are also applied such as
high/low estimates for CAPEX or different WACC.
Note: the area investigated by this study comprises all 27 Member
States along with Norway, the United Kingdom and Switzerland.
References to Europe throughout this study therefore cover these
30 states. By contrast, specific references to the European Union
cover its 27 Member States only.
57
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios56
OVERVIEW OF THE EUROPEAN SCENARIO FRAMEWORK FIGURE 32
RAD RAD+ 400
GW
Base scenario Options References
GA DE
Global
ambition
PV: + 50GW/y
Wind onshore:
+15 GW/y
Nuclear: known
plans (new and
phase out)
52% 64%
Electrification level of
the final demand
Distributed
energy
Demand &
flexibility
Onshore supply
Offshore &
electricity grid
Molecule import
Greenhouse gases
CO
2
target
non-CO
2
& LULUCF
CENTRAL
All options for
offshore/onshore
CENTRAL
-90% in 2040 and
net zero 2050
S3 EC scenario
ELEC
FLEX
Higher electrification
(with 72% in the final demand)
Base scenario
• Based on TYNDP2024, RTE Futurs
Energétiques & FES UK
Options
• ELEC scenario: no hydrogen in buildings
and road transport, less hydrogen
in industry. Shifted to increased
electrification
• Flexibility taken from TYNDP 2024
associated to each demand scenario
(storage and demand flexibility)
• FLEX scenario: longer storage duration
assumed
Based on TYNDP2024 central
input from TSOs
Sensitivities based on the max
potential for each country
Grid: all options of offshore allowed
(radial, hybrids, multi-terminal, cross
country)
Molecule import: central scenario based
on literature review and CAPEX of
underlying technologies
Proposal from the EC for 2040. Lower
target assessed
S3 scenario from the EC impact
assessment (most ambitious). Lower
ambition assessed
x2 the installed demand flex/storage
in Europe
NUC
PV+
RES+
NIM
170 GW nuclear in Europe in 2050
(instead of 75 GW)
Different offshore
grid topologies
(radially, 400 GW
offshore)
2,700 GW vs.
1,600 GW +100 GW/y for PV
+25 GW/y for wind
onshore
+75 GW/y for PV
+10 GW/y for wind
onshore
High costs for the
electricity grid
Acceleration for
domestic PV and
onshore wind
Nimby scenario
(lower onshore &
more expensive
onshore grids)
MOL
EUR+ MOL
EUR-
Higher and lower cost
for molecule import
-80% at EU level instead of -90% for
2040
Lower ambition on non-CO2 & LULUCF
CO2
80%
non
CO2+
3.1. EUROPEAN SCENARIOS AND
BOUNDARIES
3.1.1. INTRODUCTION
The scenario framework for Europe is depicted in Figure 3-2.
Most of the data is based on the TYNDP 2024
Most of the data used in this study is based on the available
preliminary TYNDP 2024 scenarios. ENTSO-E and ENTSO-G pro-
duce these long-term scenarios with numerous stakeholders
for their ten-year network development plans every two years
[ENT-1]. These are so-called ‘top-down’ scenarios for 2040 and
2050, built from the national trends scenarios for 2030 (also called
‘bottom-up’ scenarios based on official targets such as NECP).
These two scenarios deviate from the national trends scenarios
in order to capture uncertainties, but still reach the EU targets.
The Distributed Energy (DE) and Global Ambition (GA) demand
scenarios used in this study originate from the preliminary TYNDP
2024 scenarios and are complemented by data for France based
on RTE’s Futurs Energétiques [RTE-1] and the latest future energy
scenarios published by National Grid [NGE-1].
Differences compared with the TYNDP 2024
Both DE and GA scenarios used by the TYNDP 2024 also assume
different supply options, carbon targets, prices, etc. In order to be
able to assess the impact of different supply options, this study
uses the same starting points for both scenarios in terms of RES
supply options and quantities, prices and carbon targets. Several
sensitivities are performed as explained in Figure 3-2 in order
to assess the impact of different assumptions. The DE and GA
scenarios in this study differ only in terms of the final demand
assumptions. Most of the sensitivities for Europe are applied on
the DE scenario as it was computationally not possible to sim-
ulate all options.
In addition, the present model takes certain energy vectors into
account that are not modelled by ENTSO-E or ENTSO-G. Indeed,
this study goes beyond the definition of ‘hydrogen demand &
derivatives’ and splits the demand of ‘hydrogen’ into end uses
such as ammonia and liquids. This is a major improvement that
is further detailed in the current study.
Onshore supply options
Not all options are optimised by the multi-energy capacity expan-
sion model. Indeed, certain technologies such as onshore wind
or PV are not solely driven by a pure wholesale market approach.
As part of the current study, the choice was made to work with
different onshore RES development scenarios. As a starting point,
the CENTRAL scenario assumes the ‘central’ growth trajectories
defined in the TYNDP 2024 and includes several sensitivities. The
same approach is used for nuclear capacities.
Electricity grid and several options for offshore grids
The granularity of the model (zonal model for electricity) allows
this study to go beyond the ‘country-based’ capacities and
assesses grid bottlenecks across Europe. In addition, in terms
of offshore wind supply, around 400 candidates amounting to 2
GW each are assessed which leads to several sensitivities applied
to offshore development. Indeed, options such as radial con-
nections, hybrids or meshed grids are tested and compared at
European level.
Carbon targets
While all the simulated scenarios involve Europe reaching carbon
neutrality in 2050, several options are considered for the interme-
diate period (the base being -90% by 2040 while the simulated
alternative considers -80% by 2040). Similarly, a less optimistic
assumption than the base scenario (S3 scenarios from the EC
impact assessment) is also considered for non-CO2 and LULUCF
as those emissions are not explicitly modelled but are accounted
for when calculating the total emissions for Europe.
Alignment with Fluxys regarding a large set of assumptions
The assumptions were also further aligned with Fluxys, the Bel-
gian gas TSO, including (non-exhaustive list):
◆ domestic production potentials;
◆ carbon content of the different supply options;
◆ the efficiency of conversion technologies;
◆ the cost of supply grid and transformation technologies;
◆ final energy demand scenarios for DE and GA for Belgium
and Europe;
◆ starting grids in each energy vector.
59
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios58
3.1.2.1. FINAL ENERGY DEMAND
In the three demand scenarios which are examined (DE, GA and
ELEC), the final energy demand decreases significantly, with this
decrease ranging between 38% and 42% by 2050 when compared
with 2021. This reduction can be attributed to a combination
of energy efficiency measures and behavioural changes such
as building renovations and a shift in means of transport. The
most important factor (as part of the energy efficiency meas-
ures) driving this reduction is electrification. The transition from
fossil-based heating systems in buildings (such as oil and gas
boilers) to heat pumps, and the replacement of internal com-
bustion engine (ICE) vehicles with battery-electric alternatives,
results in lower energy consumption for the same uses (due to
the inherent efficiency of the electrical alternatives). Similarly,
substituting fossil-based heating supplies in the industrial sector
with electrical alternatives contributes to this decline. Therefore,
it is clear that a higher level of electrification corresponds to a
lower final energy demand.
FINAL ENERGY DEMAND IN EUROPE FIGURE 33
DE DE DE
ELEC ELEC ELEC
GA GA GA
Final energy demand reduction
(electrification, sufficiency and
energy efficiency)
-38% to -42% reduction of final
energy demand
Electricity
52% to 72% share of final energy
demand
Gaseous Molecules
10% to 30% share of final energy
demand
Coal**
Liquids
Hydrogen
Methane*
Heat
Biomass &
waste
Electricity
Other
10,330
13,500
9,570
8,430
10,140
9,270
7,980
10,020
9,100
7,780
2010 2015 2020
2036 2040 2050
Historical
Final energy demand for Europe (incl. UK, NO, CH)
Excluding international aviation & shipping and non-energetic feedstock, including grid losses.
Energy demand for transformations such as power-to-hydrogen and carbon capture are not included. Values are normalised for historical climate while in the
simulations, a forward-looking climate database is used, therefore the simulated demand can differ from these input values.
* Methane & liquids could be fossil, bio or synthetically sourced, which is defined in the model.
** Coal as defined as final energy demand per EUROSTAT (i.e. excluding coal consumed in blast furnaces).
Historical values based on EUROSTAT
16,000
14,000
12,000
8,000
6,000
4,000
2,000
0
+
Final energy demand [TWh]
Figure 3-4 depicts the share occupied by electricity versus other
energy vectors in final energy demand for some key demand
sectors at European level in the year 2050. Electrification clearly
plays a key role in these sectors – a fact which is more pronounced
in the DE and ELEC scenarios.
SHARE OCCUPIED BY ELECTRICITY, HYDROGEN, METHANE AND OTHER ENERGY VECTORS
IN FINAL ENERGY DEMAND IN 2050 FOR EUROPE FIGURE 34
GA
ELEC
DE
GA
ELEC
DE
GA
ELEC
DE
GA
ELEC
DE
GA
Electricity
97%
80%
63%
86%
77%
74%
52%
34%
33%
63%
44%
59%
85%
71%
55%
52%
70%
56%
28%
87%
66%
72%
44%
34%
20%
37%
14%
23%
26%
37%
56%
48%
66%
67%
29%
45%
15%
30%
41%
48%
Hydrogen Methane Other
ELEC
ELEC
ELEC
DE
DE
DE
GA
GA
Cars
Trucks
Buildings
Food
Chemicals
(energetic)
Other
Industry
Steel
13%
3%
3.1.2. ENERGY DEMAND
Base scenario Options References
GA DE
Global
ambition
52% 64%
Electrification level of
the final demand
Distributed
energy
Demand &
flexibility
ELEC
FLEX
Higher electrification
(with 72% in the final demand)
Base scenario
• Based on TYNDP2024, RTE Futurs
Energétiques & FES UK
Options
• ELEC scenario: no hydrogen in buildings
and road transport, less hydrogen
in industry. Shifted to increased
electrification
• Flexibility taken from TYNDP 2024
associated to each demand scenario
(storage and demand flexibility)
• FLEX sensitivity: longer storage
duration assumed
x2 the installed demand flex/storage
in Europe
Scenarios
The final energy demand is used as an exogenous input in the
models. This means that it is fixed ex ante in the optimisation.
Two main reference scenarios are used: the DE and GA scenarios
from the TYNDP 2024 study. An additional European demand
scenario was created by Elia and is modelled in the ELEC scenario,
following a request from a number of stakeholders that were
consulted but also based on several literature references that
are outlined in BOX 3-1.
The FLEX sensitivity assumes the same demand as in DE but the
flexible assets (EV, HP, batteries, etc.) are considered with a longer
storage duration. This scenario is illustrated in Section 3.1.2.3.
Storylines
The general drivers related to energy demand are included in
Table 3-1. More details about the storylines for the final energy
demand for GA and DE can be found in the TYNDP 2024 scenario
report that was recently published [ENT-2]. In concrete terms, the
ELEC scenario is constructed from the DE scenario, with hydrogen
used in road transport and heating in buildings replaced by elec-
tric vehicles and heat pumps, respectively. Additionally, the use of
hydrogen as energy in industrial settings is limited, mainly based
on the policy paper by the Florence School of Regulation [FSR-1]
and other academic studies regarding the future of hydrogen
and the value of efficient and smart electrification [ROS-1]. BOX
3-1. provides more background to consider higher electrification
in those sectors.
OVERVIEW OF THE DEMAND SCENARIO TABLE 31
Scenario
GA DE
ELEC
Energy intensity
Energy demand reduction through
energy efficiency measures (but
slower than the other two scenarios)
Stronger focus on energy efficiency Stronger focus on energy efficiency
(same as DE)
Buildings
Wide range of heating technologies
such as (hybrid) heat pumps, gas
boilers, district heating and hydrogen-
based heat
Focus on electric heat pumps and
district heating, some gaseous
heating remains
Maximised focus on electric heat
pumps
Road Transport
Full range of energy carriers for
both light and heavy-duty transport
(electric, hydrogen, liquid fuels)
Electrification of light weight road
transport, some hydrogen and liquids
remaining for heavy-duty transport
Nearly full electrification of all road
transport
Industry
In industrial settings, only lower
temperature and (to a limited extent)
medium temperature heat is assumed
to be electrified, whereas (green)
gas-based heating remains important,
especially for high temperature
processes
Most low temperature heat as well as
an important share of medium to high
temperature heat is
assumed to be electrified
Full electrification of low and medium
temperature heat, new breakthrough
technologies would also allow the
application of electricity-based
processes in (very) high temperature
heat processes
61
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios60
ENERGY DEMAND FOR FEEDSTOCK IN EUROPE FIGURE 36
DE DE DE
ELEC ELEC ELEC
GA GA GA
Historical
2010 2015 2020
2036 2040 2050
1,400
1,200
1,000
800
600
400
200
0
Feedstock demand [TWh]
+
1,110 1,090 1,120 1,100 1,110 1,110
1,110
1,120
1,090
1,090
Coal
Oil ***
Methane**
Ammonia*
Biomass &
waste
Synthetic
Liquid***
Energy demand for Europe (incl. UK, NO, CH)
ELEC scenario is based on DE scenario for energy demand for feedstock (note that the final synthetic liquid demand might differ based on the model dispatch)
* All historical hydrogen consumption is expressed in fossil fuel origin (methane or naptha via reforming or byproduct) except for the hydrogen for fertilisers which is
expressed as ammonia.
** Methane could be fossil, bio or synthetically sourced, which is defined in the model. It excludes methane used in SMR-H
2
for the production of ammonia (shown
seperately as ammonia).
*** Oil and synhtetic liquids are a result of the model dispatch
Historical values based on EUROSTAT
International transport (aviation, maritime shipping)
International transport, which includes maritime shipping and
aviation, is heavily reliant on fossil fuels due to its requirement for
fuel with a high energy density. The reduction of fossil fuel use of
this sector is relatively complex as cost-effective alternatives are
not yet fully commercially available and the operating margins
of these sectors are relatively thin.
A general decrease in energy demand can be observed due to
energy and operational efficiencies. This can be achieved through
better engine technology, improved aerodynamics and lighter
materials. Adjusting operational practices can also lead to sig-
nificant reductions in demand. In aviation, this might involve
optimising flight paths for fuel efficiency purposes. In shipping,
slower speeds can greatly improve fuel efficiency.
However, in order to further lower emissions of this segment,
these sectors will need to switch to low-carbon or zero-carbon
fuels. The modelling results demonstrate that both bio and syn-
thetic liquids such as methanol in shipping and e-kerosene in
aviation have a role to play in combination with (low-carbon)
methane. Note that even with relatively high CO
2
prices, oil-based
fuels appear to remain economically viable, even in 2050. These
emissions will then need to be compensated by negative emis-
sions (such as CCS, or land use, land use change and forestry) in
other sectors.
ENERGY DEMAND FOR INTERNATIONAL TRANSPORT AVIATION & SHIPPING IN EUROPE FIGURE 37
Oil**
Methane*
Biofuels**
Synthetic
Liquid**
Historical
2010 2015 2020
1,080
1,400
1,200
1,000
800
600
400
200
0
Energy demand [TWh]
Energy demand for Europe (incl. UK, NO, CH)
*Methane could be fossil, bio or synthetically sourced, which is defined in the model.
**Biofuels, synthetic liquids and oil are a result of the model dispatch.
Historical values based on EUROSTAT
+
DE DE DE
ELEC ELEC ELEC
GA GA GA
2036 2040 2050
870
1,040 1,010
880
1,040 1,010
880
1,070 1,060
940
THIS STUDY EXPLICITLY MODELS THE HYDROGEN DERIVATIVES
Today, most hydrogen is used as a building block to cre-
ate other products such as ammonia (for fertilisers) and
methanol [EUH-1]. In Europe, 99% of the time, hydrogen is
made from fossil fuels via reforming and/or as a byprod-
uct in certain industrial processes [EUH-2]. For this rea-
son, historical energy balances do not include hydrogen
as a final energy carrier; instead , these are expressed in
primary sources such as methane. To increase transpar-
ency regarding the demand and supply of hydrogen,
EUROSTAT will start undertaking detailed surveys of the
area from calendar year 2024 onwards [EUS-2].
The current study draws a clear distinction between hydro-
gen used in its final gaseous form and hydrogen used as a
building block to create synthetic fuels such as methanol,
ammonia, e-methane, e-kerosene, etc. This consitutes
another key difference with the TYNDP demand scenar-
ios, where the demand for hydrogen for the creation of
synthetic fuels for fertilisers, international shipping and
feedstock is also expressed in the final demand for hydro-
gen. As such, all hydrogen demands, whether for energy
and non-energy purposes, is treated as a similar form of
final energy demand in which case the system is forced to
either produce or import this hydrogen, whereas it could
also be more economical to instead import these other
molecules and/or to use a bio/fossil energy vector to sup-
ply this demand.
Figure 3-5 provides an example of how liquid demand is
modelled within this study and how it compares with the
TYNDP modelling. The following step-wise approach is
used:
◆ energy demand is taken from the TYNDP 2024 sce-
nario, in which the demand for hydrogen derivatives for
international shipping and feedstock such as methanol
and ammonia is expressed in an aggregated ‘hydrogen’
demand category without the disctinction between
the different derivatives behind it;
◆ the TYNDP demand is translated into an ‘undefined’
demand vector as an input for the model used within
this study;
◆ the supply of liquids is optimised in the molecule model
dispatch. In this dispatch the model can tap different
sources (domestic/imported) and types of molecules
(fossil/bio/synthetic) each with their own supply poten-
tial and price. The options and their characteristics can
be varied to simulate different scenarios and/or sensi-
tivities.
MODELLING OF LIQUIDS FIGURE 35
Hydrogen
Imported
synfuel
Domestic
synfuel
Model
dispatch
Liquids (undefined)
Translate into
undefined
demand
vector
Bioliquids Bioliquids
Oil
TYND 24- fixed
demand before
optimisation after
optimisation
Imported
oil
Domestic
oil
This method is applied for international aviation and
shipping, chemical feedstock and fertilisers
By using this method, the impact of sensitivities that
exploit parameters such as the price of CO2, the price of
(green) molecule imports, the CAPEX/OPEX of domestic
electrolysis, the availability of biomass, etc. is more cor-
rectly taken into account as it will directly determine the
supply mix and source of molecules that are consumed
by these sectors.
BOX 3-1
International transport and feedstock are significant components
of the global energy system, but they entail substantial chal-
lenges to reach carbon neutrality. Both components account for
around 1,000 TWh each today across Europe. This study assumes
that both sectors would fall under the Emissions Trading System
(ETS) mechanism, meaning CO
2
prices would apply for the use of
fossil fuels. In the case of feedstock, this implies that the life cycle
emissions of fossil-derived end products are taken into account.
Feedstock (for non-energy purposes)
Feedstock refers to the raw materials that are used in industrial
processes because of their physical and chemical properties (not
for their energy content). These materials are transformed in the
process to create a final product. Examples include raw materials
in the petrochemical industry which are used to produce prod-
ucts such as plastics, fertilisers, synthetic fibres and other chem-
ical products. Today, all feedstock demand relates to fossil fuels
like methane, coal and crude oil. The use of non-fossil feedstock
(for non-energy uses) is challenging because it often requires
finding new materials or processes that can perform the same
functions as fossil-based feedstock (and, in any case, carbon is a
key building block for the creation of these products).
Figure 3-6 shows the changes in feedstock demand in this study.
Until 2036, oil products such as naphtha remain most in demand
for feedstock along with some biomass. In the lead-up to 2040
and especially 2050, synthetic liquids are due to become viable,
for example via the methanol to olefins process in which meth-
anol is converted into ethylene and propylene. The demand for
ammonia for the production of fertilizers is expected to remain
relatively stable compared with today. The origin of the ammonia
is explained in Section 5.1. The split between the different vectors
supplying the feedstock is the result of the model optimisation.
63
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios62
3.1.2.3. FLEXIBILITY
In any future power system, sufficient flexibility will be required
to cope with the high volatility of RES infeed. The development
of large-scale storage devices (batteries or pumped-storage), end
user flexibility (heat pumps, electric mobility or home batteries)
and demand side response will be an important enabler to deal
with future challenges in this regard.
The flexibility associated with batteries, electric mobility and heat
pumps at a European level in this study is depicted in Figure 3-9.
The figure summarises information about the historical volume
of flexibility and the volume assumed in both the DE and GA
scenarios. The data used stems from the TYNDP 2024 scenarios.
The ELEC scenario is derived from the DE scenarios with slightly
more flexibility due to the higher electrification compared to DE.
An additional FLEX scenario is also created in order to assess the
impact of more flexibility across Europe and assumes a doubling
in energy content (not in power) from the DE scenario.
The values in the figure for the flexibility in electric vehicles and
heat pumps should be considered as indicative as those vary
within the day/year depending on the demand for heat and vehi-
cles connected to chargers.
INSTALLED FLEXIBILITY IN THE ELECTRICITY SYSTEM IN EUROPE FIGURE 39
Historical
Capacity [GW]
1,000
900
800
700
600
500
400
300
200
`100
0
+
Flexibility
DE DE DE
FLEX
ELEC ELEC ELEC
GA GA GA
2036 2040 2050
The FLEX scenario assumes a higher
energy content than DE and ELEC,
while assuming the same capacity
Higher
energy
content
700 GW
530 GW
GA
ELEC
Heat pumps
Electric vehicles
Large-scale batteries
Residential batteries
Market Response
Pumped Storage
Capacity assumed in Europe (incl. UK, NO, CH)
Historical data based on BNEF data, IEA data, ENTSO-E data
The figures given represent installed capacities (power), but the flexibility of the system is modeled based on its energy
content and other constraints. Therefore, the values provided are an estimate of the flexible power at a certain point in
time, as it fluctuates depending on the availability of the different components.
Historical values based on AdeqFlex23, ENTSO-E, energytrend.com and energy-storage.news
DE
2010 2015 2020
3.1.2.2. ELECTRICITY DEMAND
Whereas final energy demand decreases in all 3 demand scenar-
ios, a strong increase (ranging between 33% and 70% compared
to 2021) in the demand for electricity can be observed.
The transport sector is expected to experience the largest para-
digm shift in terms of energy consumption. Whilst in 2022, more
than 95% of the energy demand was still met by fossil fuels, the
sector is due to undergo the strongest relative and absolute
increase in demand for electricity. In the ELEC scenario, which
assumes that road transport methods are almost fully electrified,
covering the defined electricity demand requires more than 1,100
TWh of additional electricity, equal to 1/3 of today’s European total
electricity demand.
Electricity demand in buildings also increases due to the rollout
of electric heat pumps (depending on the scenario). However,
this is partially compensated for by the high efficiency of heat
pumps and energy efficiency measures.
Today, the industrial demand for electricity mainly stems from
non-thermal workloads such as compressors, machinery, lighting
etc. Nearly all industrial heat is supplied by combustible fuels.
Electrification has a key role to play in order to decarbonise heat
in this sector. The range between the DE, GA and ELEC scenarios
can mainly be explained by the uncertainty linked to the cost
and technical feasibility of electrification of higher temperature
heat processes.
In the GA scenario, combustible fuels remain the key energy
driver, albeit in the form of decarbonised molecules such as biom-
ethane and hydrogen (derivatives). In the DE scenario, most of the
low and medium temperature heat is assumed to be electrified
using already existing technologies. This includes industrial heat
pumps in the food and paper industries, along with the recovery
of derived heat from other industrial processes and e-boilers in
the chemical sector. The direct reduction of iron with methane
(and, in later, years hydrogen) in combination with electric arc
furnaces is assumed to be applied for steelmaking. (Green) mol-
ecules such as biomethane and hydrogen still have a role to play
in some high-temperature heat processes.
The ELEC scenario assumes that all industrial heat is mostly elec-
trified in the form of industrial heat pumps, e-boilers, microwaves,
infrared heaters, induction and resistance heaters in the metal
sector, electric boilers and crackers in the chemical sector, electric
arc furnaces and electrolysis steel in the steel industry and electric
kilns in the cement industry; each of these are considered to be
commercially available and implemented at scale by 2050. In this
scenario, almost no hydrogen is used for process heat and it has
only a limited role to play in some industrial processes such as in
steelmaking as a reducing agent. (Bio-)methane still has a small
role to play for some high-temperature energy uses.
TOTAL ELECTRICITY DEMAND IN EUROPE FIGURE 38
Historical
GA GA GADE DE DE
ELEC ELEC ELEC
Transport
Industry
Electrolysis
CCS +
Synfuels
Household
Other
Losses
Tertiary
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
Indirect
electrification
~ model
dispatch
4,670
4,950
5,000
4,980
5,440
5,600
5,520
6,150
6,430
2010 2015 2020
2036 2040 2050
Electricity demand for Europe (incl. UK, NO, CH).
Values are normalised for historical climate while in the simulations, a forward looking climate database is used, therefore the simulated demand can differ from
these input values.Electrolysis, CCS/U is optimised within the model and depends therefore on each potential scenario and sensitivity.
Historical values based on EUROSTAT
Electricity demand [TWh]
+
ELEC
95%increase
DE
85%increase
GA
75%increase
65
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios64
Solar PV scenario
Regarding solar PV, the CENTRAL scenario assumes a total
capacity of 1,600 GW of solar capacity by 2050 in the simulated
European area. This is six times today’s existing capacity and
corresponds to a yearly installation rate of 50 GW. An installation
rate of 75 GW per year is assumed in the RES+ scenario, leading
to 2,100 GW by 2050 in Europe. A third scenario called PV+ is
also considered, with an annual installation rate of 100 GW and
an installed capacity of 2,700 GW by 2050.
ASSUMED IN INSTALLED SOLAR CAPACITY IN EUROPE FIGURE 310
Historical
1,600 GW
2,100 GW
2,700 GW 100 GW/y
50 GW/y
75 GW/y
2036 2040 2050
RES+
PV+
CEN
3,000
2,500
2,000
1,500
1, 000
500
0
Installed capacity [GW]
+
Solar
2010 2015
2020
Capacity assumed in Europe (incl. UK, NO, CH).
Historical data based on EMBER data.
20 GW/y
Onshore wind scenarios
A total installed capacity of 620 GW of onshore wind in Europe
by 2050 is assumed in the CENTRAL scenario (three times today’s
existing capacity), corresponding to a yearly installation rate of 15
GW. In the RES+ scenario, a rate of +25 GW per year is assumed,
with 850 GW installed by 2050. A third scenario, the NIMBY sce-
nario, assumes an additional 10 GW per year which is what has
been observed historically.
ASSUMED IN INSTALLED ONSHORE WIND CAPACITY IN EUROPE FIGURE 311
620 GW
490 GW
850 GW
15 GW/y
10 GW/y
25 GW/y
Historical
900
800
700
600
500
400
300
200
100
02036 2040 2050
Installed capacity [GW]
+
RES+
CEN
NIM
Onshore wind
2010 2015 2020
Capacity assumed in Europe (incl. UK, NO, CH).
Historical data based on EMBER data.
10 GW/y
Other continental RES
In addition to solar PV and onshore wind, European RES also
include hydroelectric production and biomass. The assump-
tions for those categories are fixed, based on the expected
capacity in 2030: about 5 GW of biomass (used for electricity
production) by 2030 and about 170 GW of hydro power plants
(i.e. run-of-river, reservoir and pondage, excluding pumped
storage). No increase or decrease in these capacities is
assumed. Note the amount of hydroelectric production is de-
pendent on the climate years that are used given the linkage
with precipitation.
3.1.3. ELECTRICITY SUPPLY
Several sensitivities are applied to the electricity supply. Indeed, there are still many uncertainties regarding the pace at which certain
technologies might spread. What is certain is that a higher amount of renewables will be integrated into the system. Certain types
of dispatchable power are expected to be phased out in the future.
RAD RAD+ 400
GW
Base scenario Options References
PV: + 50GW/y
Wind onshore:
+15 GW/y
Nuclear: known
plans (new and
phase out)
Onshore supply
Offshore &
electricity grid
CENTRAL
All options for
offshore/onshore
Based on TYNDP2024 central
input from TSOs
Sensitivities based on the max
potential for each country
Grid: all options of offshore allowed
(radial, hybrids, multi-terminal, cross
country)
NUC
PV+
RES+
NIM
170 GW nuclear in Europe in 2050 (ins-
tead of 75 GW)
Different offshore
grid topologies
(radially, 400 GW
offshore)
2,700 GW vs.
1,600 GW +100 GW/y for PV
+25 GW/y for wind
onshore
+75 GW/y for PV
+10 GW/y for wind
onshore
High costs for the
electricity grid
Acceleration for
domestic PV and
onshore wind
Nimby scenario
(lower onshore &
more expensive
onshore grids)
3.1.3.1. CONTINENTAL RES
The future role played by renewables in Europe is set to be signif-
icant as the continent strives to shift towards a more sustainable
and green economy. Europe is already a leader in renewable
energy, with many countries like Germany, Spain and Denmark
setting the pace with significant investments in wind and solar
energy. This section mainly focuses on solar photovoltaic and
onshore wind as they are assumed to be the two most important
sources of renewable energy for the European continent in the
future.
Solar photovoltaic (PV) panels are said to become the largest
source of renewable energy in terms of installed capacity across
Europe. As outlined in its EU Solar Energy Strategy [EUC-11] (which
is part of the ‘REPowerEU’ plan), the European Commission is aim-
ing to have 600 GW of solar PV installed by 2030 across the EU27.
In addition to policy that supports its development, technological
advancements and cost reductions are expected to make solar
PV more accessible and affordable, so accelerating its adoption.
However, challenges such as its integration into the power system
and the need for significant infrastructure investment (on both
the DSO and TSO side) and storage will need to be overcome.
Europe has also been a front-runner in harnessing onshore wind
energy and this trend is expected to continue in the coming
decades. In October 2023, the European Commission published
its Wind Power Action Plan, which aims to ensure the success
of Europe’s wind energy industry through measures such as
an improved auction design, the faster deployment of projects,
access to finance and the development of a skilled workforce.
Although a number of challenges will have to be addressed (land
use conflicts, public acceptance issues, and the need for exten-
sive grid upgrades), with continued policy support and societal
commitment to a green transition, onshore wind energy has a
promising future in Europe.
Four scenarios are studied for solar PV and onshore wind
development at European level:
◆ The CENTRAL scenario is based on the ‘central’ trajectories
that were brought forward during the public consultation of
the TYNDP2024 scenarios (with adaptations for the PV trajec-
tories reflecting recent growth trends):
- PV: +50 GW/year and onshore wind: +15 GW/year.
◆ The RES+ scenario assumes a higher installation rate and final
installed capacity for both PV and wind onshore. This scenario
reaches the maximum potential identified in the TYNDP 2024
consultation phase.
- PV +75 GW/year and onshore wind +25 GW/year
◆ The NIMBY scenario will further reduce the installation rate of
wind onshore along with higher costs for onshore grids.
- Onshore wind +10 GW/year (while PV follows the CENTRAL
scenario: +50 GW/year)
◆ The PV+ scenario will further increase the PV capacity for
2050 on top of the RES+ scenario.
- PV +100 GW/year (while onshore wind follows the RES+ sce-
nario: +25 GW/year)
67
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios66
Note that 6 offshore hubs are also assumed in the initial situa-
tion. These are meant to represent the known projects/ambitions
related to multi-terminal offshore hubs (e.g. the Princess Elisabeth
Island in Belgium, the North Sea Power Hub in the Netherlands,
…). The optimiser does not assume additional platform exten-
sion costs for the islands, as it is assumed that the necessary
infrastructure for hosting additional connections has been taken
into account from the start. If it brings the model closer to the
optimum, the optimiser is able to reinforce the connections to
these hubs to transform them into multi-terminal offshore hubs.
Offshore wind farms can be connected to the shore in different
ways: via radial connections that link the farm directly to the
shore; via two connections to two different coasts (hybrids); via a
hub, which includes different connections to different shores; and
via connections to other offshore wind farms, which are them-
selves connected to the shore (see also Appendix A for more
information).
In order to see the impact of allowing or limiting some types of
connections, different options are simulated:
◆ only radial connections to the shore of the ‘home country’ are
allowed (RAD);
◆ radial connections to other countries are allowed (RAD+);
◆ starting with already 400 GW of offshore wind in Europe that
is pre-connected.
OFFSHORE WIND POTENTIAL BEING CONSIDERED ACROSS EUROPE FIGURE 313
Up to 500 GW
Up to 850 GW
180 GW
OFFSHORE WIND CAPACITY POTENTIAL PER SEA BASIN
TYPE OF OFFSHORE WIND FARMSPOTENTIAL OFFSHORE WIND CAPACITY [GW]
1000
800
600
400
200
0
Potential offshore wind farms candidate that can be radially
connected to shore or with hybrid interconnections (part of the
optimisation)
Technical potential used as upper boundary
in 2050
Technical potential used as upper boundary
in 2036 and 2040
Assumed offshore wind farms in the
initial model for 2036 Existing offshore wind farms and future offshore wind farms
assumed radially connected to shore by 2036 (not part of the
optimisation)
Future offshore wind farms assumed radially connected to shore
by 2036 that can become hybrid interconnections (part of the
optimisation)
Future offshore ‘hub’ assumed radially connected to shore by
2036 that can become a multi-hub (part of the optimisation)
110 GW
170 GW
210 GW
360 GW
3.1.3.2. OFFSHORE RES
By 2050, offshore wind energy is expected to play a pivotal role in
Europe's energy mix, contributing substantially to the European
Union's ambitious target of achieving net-zero carbon emissions.
This will be facilitated by advancements in offshore wind tech-
nology, which will make turbines more efficient and cost effec-
tive. Offshore wind capacity is expected to be spread out across
different seas and oceans of Europe.
Important developments occurred in this respect last year. Europe
committed to transforming the North Sea into Europe’s green
power plant during the second North Sea Summit in Ostend in
April, and to strengthening regional cooperation during the Baltic
Offshore Wind Forum in Berlin in May.
The emergence of hybrid interconnectors and energy islands will
facilitate the exchange of electricity between countries with vary-
ing levels of RES potential, while also connecting them to offshore
wind farms. These hybrid interconnectors and energy islands will
constitute important steps on the journey to the establishment
of a European meshed offshore grid.
In this study, installed offshore capacity and its location is
optimised by the tool. Therefore, this section focuses on the ini-
tial level of wind considered in the system as well as the upper
boundaries allowed for the investment. The final invested off-
shore wind capacity is outlined in the results section. Note that
the focus is set on offshore wind farms while offshore RES also
includes ocean/tidal energy which is not assessed here. The
approach for the offshore wind scenario applied in this study is
similar to the one performed for the KARI study which was pub-
lished in the last Federal Development Plan [ELI-3].
The offshore wind potential has been identified via a detailed
approach which started with a database from 4C Offshore [4CO-1]
and in addition considered both geographical constraints
(bathymetry, shipping routes, environmental zones, etc.) and
the latest identified offshore zones in national plans. Along with
the latest known ambitions and announcements, this leads to
the following assumptions:
◆ About 180 GW of radially connected offshore wind is assumed
in the starting grid for 2036. This includes both existing capac-
ity at the end of 2023 as well as projects that are planned to be
commissioned in the coming decade. Care is taken to ensure
coherence with the most recent published national plans and
ambitions, such as those announced at the Ostend North
Sea Summit [FPS-4]. These offshore wind farms are radially
connected to the shore in the starting model. Offshore wind
farms for which no landing point is known (about 80 GW) can
be transformed into hybrid interconnectors by the optimiser
(investing in one additional leg towards another landing point
or offshore hub).
◆ A maximum potential of about 500 GW of total offshore wind
capacity is assumed in Europe in 2036 and 2040. In 2050, a
maximum potential of about 850 GW of total offshore wind
capacity is assumed in Europe.
The investment potentials relate to the upper boundaries of pos-
sible investments for the solver. How much of it will be invested
in by the optimiser is not known beforehand. The theoretical
offshore wind potential is illustrated in Figure 3-13; this outlines
the location of the wind farms and the total capacity potential
per sea basin.
OFFSHORE GRID CONFIGURATION OPTIONS FIGURE 312
Hybrid offshore Multi-terminal offshore hubs
Radial offshore to shore
(home country) Radial offshore to shore
(other countries)
69
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios68
3.1.3.3. THERMAL FLEET
In addition to RES, an important part of the electricity mix today comes from the thermal fleet, which is mainly fuelled by methane,
coal and nuclear. This section outlines the assumed capacities of methane, coal and nuclear in Europe.
Nuclear and coal capacity
In the past, coal-fired plants were a key source of electricity
generation across Europe. However, their role has diminished
in recent years due to rising environmental concerns. Indeed,
several European nations have begun to limit their dependence
on coal-powered electricity, with many announcing phase-outs
within the next decade. Belgium took the lead in 2016 by closing
all coal-powered plants, followed by Sweden and Portugal. Given
the recent energy crisis, some countries like France, the UK, and
Germany delayed their plans to shut down certain coal-powered
plants. Despite this, the majority of European nations are com-
mitted to phasing out coal by 2030 or shortly thereafter (2033
for the Czech Republic and Croatia). Poland is an exception: it is
projected to be among the last European countries to still have
coal-powered capacity in 2036.
Regarding nuclear power, currently the majority (almost 60%)
of Europe's nuclear capacity is centralised in France. Several
countries have made decisions about or are currently debat-
ing whether to prolong the operational lifespan of their nuclear
plants, potentially extending their lifetimes beyond original plans
or legislative limits. In addition, several countries are also consid-
ering building new units (e.g. Poland, Romania, France, United
Kingdom, the Netherlands,...). The Central scenario for nuclear
power is based on current policies, taking into account planned
closures and new units. The capacity is expected to decrease
from about 100 GW in Europe today to about 75 GW in 2050.
While it is assumed that in 2036 the existing nuclear fleet in
France will still be available (60 GW from the older fleet along
with the new European pressurised reactor (EPR) in Flamanville),
the decommissioning of the older fleet will have started by that
point. This will be partially compensated by the commissioning
of new nuclear reactors, leading to a decrease of 20 GW by 2050.
Changes in other countries are also taken into account, notably
new units in Poland, Romania, UK, France….
In order to cover the possibility of a higher total nuclear capacity,
with (for example) the emergence of small modular reactors
(SMRs) in Europe, a sensitivity with 150 GW of nuclear capacity
in the EU is performed reaching 170 GW of nuclear in Europe
(including also UK, NO and CH). This sensitivity is inspired from
the recent communication on the 'Nuclear Alliance' [NUC-1].
COAL AND NUCLEAR CAPACITY IN EUROPE FIGURE 314
75 GW
170 GW
Historical
2036 2040 2050
+
NUC
CEN
350
300
250
200
150
100
50
0
Installed capacity [GW]
Coal Nuclear
2010 2015 2020
Capacity assumed in Europe (incl. UK, NO, CH).
Historical data based on EMBER data.
Gas and oil-fired capacity
Although today gas turbines are mainly fuelled by methane, it is
assumed that hydrogen turbines could enter energy markets in
the future (already a possibility in the model by 2036). In addition
to CH4 and H2 turbines, it is also assumed that CCS technology
will be available for CCGT and biomass.
In this study, the gas thermal fleet is optimised by the tool with
the following logic:
◆ existing units are closed after a certain lifetime;
◆ the model can invest in new units running on hydrogen or
methane (including or not CCS);
◆ the model can choose to invest in CCS or H2 reconversion for
the existing fleet;
◆ the amount of capacity for each country is calculated in a
separated step (adequacy step) that takes into account 200
‘Monte Carlo’ years and the known adequacy indicators (see
also Appendix D for more information).
The results of the optimisation (including the currently installed
capacities) can be found in the results Section 4.6.
All oil-fired capacity for electricity generation is assumed to be
decommissioned from the system by 2036.
SEVERAL STUDIES HAVE DEMONSTRATED THAT ELECTRIFICATION IS THE MAIN PATHWAY TO-
WARDS THE 2050 TARGET OF CLIMATE NEUTRALITY
In the pursuit of achieving net-zero emissions, the electrification of various sectors plays a pivotal role. Electrifying trans-
portation, buildings, and industrial processes can significantly reduce carbon emissions and drive the shift towards
a more sustainable future. In order to assess higher electrification speeds, the ELEC scenario was created assuming
higher shares of electrification in transport, building and industry. Many academic studies do not support the view that
e.g. hydrogen and other gases would be used for transport or residential heating as opposed to electrification. As the DE
scenario does still assume some level of hydrogen and other gases in transport or residential heating, the ELEC scenario
allows to account for those that believe that direct electrification is the most efficient/effective solution.
A non-exhaustive list of studies is provided below:
“The share of electricity in final energy consumption
increases from 23% in 2015 to above 45% in 2040 .. and
up to 57% .. in 2050. This increase is mainly driven by
the uptake of electric vehicles, the penetration of heat
pumps and electrification of low- and medium-temper-
ature industrial processes”
European Commission – ‘Europe's 2040 climate target
and path to climate neutrality by 2050 building a sus-
tainable, just and prosperous society’
“The Commission Communication does not call for a spe-
cific target on electrification. Instead, it acknowledges
that by 2040, electricity will cover 50% of the energy con-
sumed in Europe. Positively, electrification is considered
the main driver of decarbonisation in end-use sectors”
EURELECTRIC – ‘Analysis of Commission communica-
tion on 2040 climate target’ and accompanying impact
assessment
“Electrification is one of the most important strategies
for reducing CO2 emissions from energy towards net zero
emissions by 2050, where the majority of emissions reduc-
tions from electrification come from the shift towards
electric transport and the installation of heat pumps”
IEA – Electrification & ‘Electricity 2024’
“Electricity demand more than doubles in the 3 scenar-
ios. […]Faster electrification in general leads to 9% lower
emissions in 2030 in the residential and commercial sec-
tor, while faster electrification of freight road transport
leads to 75% lower emissions in 2040 in the transport sec-
tor.”
EnergyVille – ‘PATHS2050’
“In addition, it is expected that the current demand for
electricity (~500TWh) will continue to increase towards a
value around ~780 TWh in 2035 [...] e.g. due to increasing
electrification of mobility and a more energy-intensive
mobility, which so far are fossil-fuelled processes today.”
McKinsey –‘Future path |Power supply |Perspectives for
increasing security of supply and economic efficiency
the energy transition in Germany by 2035’
“Europe’s 2040 climate ambition should build more on
renewables, electrification and circularity”
Agora Energiewende – ‘Analysis of Commission com-
munication on 2040 climate target’ and accompanying
impact assessment
“Electrifying everything possible, from transport to indus-
try, will unlock major efficiency gains and emissions
reductions.”
EMBER – ‘Our vision of a clean power system’ & ‘Europe-
an-Electricity-Review-2024’
“A key step to achieving climate neutrality in the Euro-
pean Union is to rapidly shift from fossil fuels to electric
technologies powered by renewable energies”
Adapted from Potsdam Institute for Climate Impact
Research (PIK) – ‘Electrification or hydrogen? Both have
distinct roles in the European energy transition’
“There is a strong and growing consensus that a simul-
taneously growing and decarbonising electricity sector is
necessary to meet declining greenhouse gas emissions
targets”
Energy Policy Columbia University
“Direct electrification is arguably one of the most cost-ef-
fective and reliable ways to decarbonise the European
Union”
Adapted from Electrification-alliance.eu
“Despite the significant attention which hydrogen has
received, independent evidence does not support wide-
spread use of hydrogen for space and hot water heating”
Based on a review of 32 independent studies.
Adapted from Dr. Jan Rosenow, Director of European
Programmes at the Regulatory Assistance Project
(RAP).
"The results of the study indicate significant potential
for the direct electrification of process heat generation,
which could meet 90 percent of the energy demand not
yet electrified by European industry, if fully deployed "
Agora Industry, ‘Direct electrification of industrial
process heat’ June 2024
BOX 3-2
71
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios70
MERIT ORDER FOR METHANE, HYDROGEN, AMMONIA, AND LIQUIDS IN 2040 AND 2050 FIGURE 315
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
0 1000 2000 3000 4000 5000 6000 7000 8000 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000
HYDROGEN
LIQUIDS
METHANE
2040 2050
200
150
100
50
0
200
150
100
50
0
Price [€/MWh]
Price [€/MWh]
200
150
100
50
0
200
150
100
50
0
Price [€/MWh]
Price [€/MWh]
200
150
100
50
0
200
150
100
50
0
Price [€/MWh]
Price [€/MWh]
Generation [TWh]
Generation [TWh]
Generation [TWh] Generation [TWh]
Generation [TWh]
Generation [TWh]
Domestic fossil methane Domestic biomethane Fossil methane import – pipeline
Synthetic methane import – pipeline Fossil methane import – LNG Synthetic methane import – LNG Methane from hydrogen
Liquid from hydrogen Domestic bioliquids Domestic fossil liquids Synthetic liquids import Fossil liquids import
Hydrogen import Ammonia import SMR-H
2
– domestic SMR-H
2
– import
3.1.4. MOLECULE SUPPLY
Beyond the direct electricity supply, this analysis also delves into the sources related to gases and liquids to fulfil the demand for
methane, hydrogen, ammonia, and liquids. A merit order for each of these sources for 2040 and 2050 is included in Figure 3-15. The
prices provided in the merit order do not include a CO2 price or conversion costs since these are scenario dependent. Electrolysers
are not included in this figure since their price depends on the electricity price at which they produce hydrogen in the electricity
model which differs throughout the year. It is important to note that the merit order depicted in the figure assumes no internal
congestion in pipelines and focuses exclusively on each individual molecule (note that the dispatch model accounts for those). For
instance, the merit order for methane presumes the existence of an unlimited internal methane grid and anticipates the unrestricted
utilisation of other molecules, such as hydrogen, independent of the actual hydrogen demand. The rest of this section will explore
each of these sources individually in detail.
Molecule import CENTRAL Molecule import: central scenario based
on literature review and CAPEX of
underlying technologies
Base scenario Options References
MOL
EUR+ MOL
EUR-
Higher and lower cost
for molecule import
Methane
Methane can be sourced either domestically within Europe
or through international imports. Domestic methane produc-
tion encompasses both fossil methane and biomethane. Fossil
methane is primarly sourced from Norway, decreasing a bit by
2050. Conversely, biomethane, distributed throughout Europe
is estimated to increase towards 2040 and 2050. The amount of
methane remains consistent across all scenarios and is aligned
with the TYNDP 2024.
Regarding imports, both pipeline and sea freight imports in the
form of liquefied natural gas (LNG) are taken into account. For
pipeline imports, existing gas pipelines are considered across all
years. Similarly, for LNG, existing terminals today are factored in
for all years.
Additionally, methane can be produced from hydrogen through
a process known as methanation, as detailed in Section 3.1.6. The
model has the flexibility to determine the quantity of methane
derived from hydrogen.
Hydrogen
A supply of hydrogen can be achieved through various methods:
electrolysers, direct hydrogen or ammonia imports, and steam
methane reforming (SMR-H
2
). The amount of electrolyser capac-
ity depends on the optimisation within the electricity model. In
addition, concerning their dispatch, electrolysers can only yield
hydrogen when they are operational within the electricity model.
During periods when the price of electricity is high, electrolysers
may cease to operate, making them unable to supply hydrogen.
During such instances, hydrogen must be supplied either by
SMR-H2, storage or through imports.
Hydrogen can be imported via pipelines or made from imported
ammonia (via terminals). The model has the capability to con-
struct hydrogen pipelines, but the volume of hydrogen produced
in the exporting country is based on TYNDP 2024 data. Although
the production capacity in the exporting country remains con-
sistent across all scenarios, the model's ability to construct import
pipelines can produce variations. For ammonia imports, the max-
imum export volume per exporting country outside of Europe is
based on the global hydrogen flows study from the Hydrogen
Council [HYD-1]. The model can decide on the quantity to use as
ammonia and the amount to convert into hydrogen via ammonia
cracking.
Lastly, hydrogen can be produced through SMR-H2. The SMR-H2
volume per country is derived from the DE and GA scenarios of
the TYNDP 2024. In the DE scenario, this equates to the current
installed SMR-H2 capacity. The SMR-H2 process always includes
CCS; this assumption is aligned with the TYNDP.
Liquids
Just like methane, liquids can be sourced either from domestic
production or through imports. The domestic production of liq-
uids may be either fossil-based or bioliquids. Fossil liquids are pro-
jected to decrease towards 2050. On the other hand, the potential
for bioliquid production in Europe is expected to increase. The
amount of liquids is based on TYNDP 2024 data and remains
consistent across all scenarios.
Given the limited potential for the production of liquids in Europe,
the model primarily relies on imports to fulfil the demand for
liquids. The maximum capacity of imports of fossil and synthetic
liquids is set to infinite to ensure the demand for liquids can be
met.
In addition to imports, liquids can also be produced domestically
from hydrogen via the Fischer-Tropsch synthesis. The quantity
produced through this process is not constrained and can be
determined by the model.
73
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios72
3.1.5.2. NONEXPLICITLY MODELLED EMISSIONS
This study explicitly quantifies all forms of CO2 emissions result-
ing from energy uses. However, as the European objectives are
defined for all GHGs (i.e. not only for CO2) and all types of emis-
sions (including process related emissions), these also need to
be taken into account.
Process emissions
Process emissions are GHGs which are released during the man-
ufacturing or chemical transformation of materials (independent
from energy use). They constitute a critical part of the emissions
profile of many industries, particularly in sectors like cement,
steel, and the production of chemicals. These emissions occur
as a result of chemical reactions (for example, when limestone
is heated to produce lime and CO
2
in the making of cement)
and are hence independent of energy-related emissions, which
are associated with the combustion of fossil fuels for energy.
Process emissions are essential to take into account because
they often represent a significant proportion of an industry's
total emissions and cannot be eliminated by simply switching
to renewable energy or improving energy efficiency. Without
taking these emissions into account, the total GHGs from these
industries would be severely underestimated, meaning that
significant opportunities for reducing emissions would be missed.
These process emissions fall outside of the modelling framework
but must be accounted for. Therefore, a fixed trajectory is used
in line with the points outlined below.
◆ For the metal industry, process emissions are mainly caused
by the reduction of iron ore and coke production. The emis-
sion trajectory is linked with the demand scenarios within the
TYNDP which foresees a switch from blast furnaces to direct
reduced iron ore and/or electric arc furnaces which helps to
reduce these emissions.
◆ For the chemical industry, process emissions are caused
during a variety of processes such as the creation of ethylene,
chloralkali etc. These are assumed to remain constant in the
lead-up to 2050. The production of ammonia, which is cur-
rently carried out via steam methane reforming, is explicitly
modelled and hence excluded from this fixed trajectory.
◆ For the mineral industry, process emissions are mainly
caused by the calcination of limestone during the production
of lime and/or cement. Regardless of how heat is supplied in
this sector, those emissions will persist. A slight reduction in
process emissions is assumed, mainly based on a more effi-
cient use of materials as per [CEM-1].
The final assumed trajectory is included in Figure 3-16. Note that
this includes ‘gross’ emissions; these processes may be fitted
with carbon capture technologies which would reduce the net
emissions (see 3.1.5.3).
HISTORICAL AND ASSUMED PROCESS EMISSIONS FIGURE 316
300
250
200
150
100
50
0
MtCO2
2010 2015 2020
Mineral industry
Other
Chemical industry*
Metal industry
Data for Europe (incl. UK, NO, CH).
*Note that the chemical industry excludes emissions for ammonia production as these are explicitly modeled within this study.
Historical values based on EUROSTAT
2036 2040 2050
Historical
+
2010 2015 2020
228
172
163
156
3.1.5. GREENHOUSE GASES
Greenhouse gases (GHGs) have been a source of significant environmental concern in Europe for many years. These gases (which
include carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O) and F-gases) trap heat in the Earth’s atmosphere and contribute
to global warming and climate change.
This section describes the GHG emission targets assumed in this study, the non-explicitly modelled emissions and the carbon capture,
storage and usage assumed in this study.
Carbon
CO
2
target
non-CO
2
& LULUCF
-90% in 2040 and
net zero 2050
S3 EC scenario
Proposal from the EC for 2040. Lower
target assessed
S3 scenario from the EC impact
assessment (most ambitious). Lower
ambition assessed
Base scenario Options References
-80% at EU level instead of -90% for
2040
Lower ambition on non-CO2 & LULUCF
CO2
80%
non
CO2+
3.1.5.1. GHG EMISSION TARGET
As described in Section 1.2, Europe has set emission reduction
targets for the next decades. Figure 1-2 illustrates the histori-
cal European GHG emission targets and net-zero targets in the
lead-up to 2050. The optimisation considers the targets shown
on Table 3-2 as constraints to be reached.
The target for 2036 is determined by taking the interpolation
between the official EU target for 2030 (i.e. -55% compared with
1990) and the assumed target for 2040. For 2040, no official EU
target exists yet, however in its impact assessment, the European
Commission has recommended reducing the EU’s net GHGs
by 90% by 2040 (compared with 1990) [EUC-12]. As such, this is
the emission constraint used in the central scenario. In order to
assess the impact of relaxing this target, a lower ambition sen-
sitivity of -80% for 2040 (and thus also affecting the 2036 target)
is also studied.
In this study, these targets are assumed to also be applicable
for the UK, Norway and Switzerland, even though they are not
members of the EU. Additionally, international aviation and 50%
of international shipping are included in the scope, following
the methodology used in the impact assessment performed by
the EC [EUC-12].
The consumption of fossil fuels used as non-energetic feedstock
which are not combusted or released as process emissions do
not have their CO2 content included in the emissions. The sys-
tem-wide CO2 price is however also applied on fossil fuels con-
sumed within this sector. Note that in case these products are
burned as waste, their associated carbon intensities are included
in the emissions via the electricity dispatch model.
ASSUMED GHG EMISSIONS REDUCTIONS TARGETS FOR EUROPE TABLE 32
Assumed GHG emissions reductions targets
for Europe (incl UK, CH & NO) – relative to
1990
Base scenario
CO
2
80%
2036 -76% -70%
2040 -90% -80%
2050 -100% (net zero reached)
75
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios74
3.1.5.3. CARBON CAPTURE, STORAGE AND USAGE
In order to comply with emission targets, the model has the
option to invest in a wide range of CO2 abatement options, each
of which has a different cost depending on the specific type of
technology involved and their geography. The methodology is
explained in detail in Appendix C.
This study considers 3 main types of carbon capture (see also
Figure 3-18):
◆ Carbon capture in industry
- In processes: the process emissions detailed in 3.1.5.2 for the
metal, mineral and chemical industry have the potential to
be abated using carbon capture technology.
-
In combustion: methane burned for energy purposes (i.e.
the delivery of heat) could also be fitted with carbon capture
technology. Note that in the case of biomethane, this would
imply negative emissions.
◆ Carbon capture in power generation: the model has the
option to retrofit/install carbon capture technology on power
generation.
◆ Direct-Air Capture (DAC) is a technology that captures car-
bon dioxide directly from the ambient air.
The captured CO
2
then needs to be treated (i.e. through com-
pression, liquefaction, etc.) and transported. These steps and the
potential CO2 pipeline infrastructure are not explicitly modelled
but their assumed cost is added to the different carbon capture
technologies. The final destination is then a potential usage and/
or storage point. Different options are considered;
◆ Use for the creation of synthetic fuels: in this case, CO2 is
typically combined with hydrogen to create a derivative fuel
such as methanol, e-kerosene, e-methane, etc. This new
product is then used as a combustion fuel and thus the CO2 is
emitted back into the atmosphere (if it is not captured again).
◆ Use as feedstock for the creation of materials: in this case,
the CO2 is used as a building block to create chemicals and
final products. One example is the synthesis of methanol
into ethylene and propylene. As long as these products are
not burned, the CO2 emissions are in principle bound into the
final products and not emitted into the atmosphere.
◆ Underground storage: also known as carbon sequestration,
this method permanently stores CO2 to prevent its release into
the atmosphere. Typical storage sources include depleted oil
and gas fields, unmineable coal seams, deep saline aquifers,
and basalt formations.
CARBON CAPTURE, USAGE AND STORAGE OPTIONS CONSIDERED IN THIS STUDY FIGURE 3 18
Industry Combustion Carbon
capture
Carbon
capture CO
2
CO
2
Waste
incineration
Combustion
Process
Power
Direct air capture
Synthethic
fuel
Materials
Underground
storage
Post-treatment
& transport
For each iteration of the molecule model (see Figure 2-6), the
CO
2
module has the option to select different CO
2
abatement
options to invest in for each of the modelled electricity zones. For
each carbon capture type and geographical location, a different
amount of CO2 could be captured at a certain cost, depending
on the CAPEX/OPEX, capture efficiency and electricity prices.
For this study, the consultancy company Compass-Lexecon per-
formed an extensive analysis on the trajectories quantified by the
EC in which they conclude that that the S3 scenario projections of
CH4 and N2O emissions could be perceived as too ambitious. For
this reason, an alternative scenario (‘non-CO2+’) which combines
mitigation measures from the different EC scenarios is developed
to capture a more realistic projection of non-CO2 emissions.
◆ The non-CO2+ scenario assumes the S1 scenario’s projection
for CH4 emissions in the agriculture sector and the S3 scenar-
io’s projection for CH4 emissions in other sectors.
◆ The non-CO2+ scenario assumes the S2 scenario’s projection
for N2O emissions in the agriculture sector and the S3 scenar-
io’s projection for N2O emissions in the other sectors.
The final trajectories for the central and non-CO2+ scenario are
included in Figure 3-17. For more information, see Appendix H.
Non-CO2 GHG gases and LULUCF
Just as for process emissions, non-CO
2
emissions are largely
decoupled from the final energy demand trajectory. Given this,
an assumption is needed regarding the potential evolution of
these GHG emissions. As such, the scenarios studied within the
European Commission’s 2024 Impact Assessment Report [EUC-
12] explores three scenarios that include increasing net emission
reduction ambitions to reach EU targets.
◆ The S1 scenario mainly relies on the Fit for 55 energy trends
with no specific mitigation measures for non-CO2 emissions
or for evolutions in the LULUCF sector.
◆ The S2 scenario builds on S1 and adds substantial reductions
in terms of non-CO2 emissions and significant increases in
carbon removals in the LULUCF sector.
◆ The S3 scenario builds on S2 and adds a fully developed car-
bon management industry by 2040, with sizeable carbon
removals and the deployment of novel technologies.
As the S3 scenario is aligned with the -90% emission target (2040)
used in the base scenario, its trajectory is chosen for the base
scenario. This projects a significant decrease in non-CO
2
GHG
emissions and a relevant increase in LULUCF net emission remov-
als in 2040 and 2050.
HISTORICAL AND ASSUMED NONCO2 EMISSIONS FIGURE 317
CEN non
CO
2
+CEN non
CO
2
+CEN non
CO
2
+
Historical
600
400
200
0
-200
-400
MtCO2_eq
2010 2015 2020
LULUCF
F-gases
Net emissions
CH
4
NO
2
Data for Europe EU27.
Historical values based on EUROSTAT
424
2036 2040 2050 2050
221
105
169
28
134
-40
36
77
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios76
3.1.7. GRID
This section provides further information about the European molecule grid and the high-voltage electricity grid assumed in this study.
3.1.7.1. HIGHVOLTAGE ELECTRICITY GRID
The initial high-voltage network for Europe is based on the TYNDP
2022 IoSN reference grid [ENT-5]. This reference grid represents
a snapshot of the expected grid in 2025. To more accurately rep-
resent flows, most bidding zones that exist today are split into
smaller electrical zones. Flows between electrical zones are mod-
elled using a flow-based approach and take into account the
effects of phase shifting transformers. For Belgium (see Section
3.2.5.), the approved investement in the latest federal develop-
ment plan [ELI-3] are accounted for.
In contrast to the TYNDP 2022 IOSN grid, the present study con-
siders multiple electrical zones in Great Britain. Additionally, rein-
forcements which were approved in the latest Belgian Federal
Development plan have been added to this reference grid. An
overview of the grid used and zonal granularity is depicted in
Figure 3-21.
3.1.7.2. MOLECULE GRID
The starting grid for the methane system is the existing methane
grid in Europe including the LNG terminals and pipelines. This is
defined as capacities between the zones considered in the model
as explained in Section 2.3. No investments (or decomissionings)
are considered for this part of the system.
For hydrogen, no starting grid is considered. New capacities
between the zones assumed in the model (see Section 2.3.) can
be invested in. In addition, it is possible to build pipelines from
North Africa and Ukraine to Europe. Hydrogen can also be pro-
duced via offshore electrolysers or be transformed from ammo-
nia or methane cracking (via steam methane reforming). This is
possible within each zone of the model.
For other liquids, no grid is assumed in the model and Europe is
considered as a single node.
STARTING HIGHVOLTAGE GRID FIGURE 321
AC interconnectors
DC interconnectors
Initially connected offshore wind
farms who cannot be further
invested in
Initially connected offshore
wind farms who can be further
interconnected
Offshore wind farm investment
candidates
Energy hubs
3.1.6. TRANSFORMATION PROCESSES
Transformation processes play a crucial role in the energy sector
by enabling the conversion of one energy source into another.
These processes are incorporated into the model to capture the
transition from one molecule type to another. An overview of
all the transformation processes included in this study can be
found in Figure 3-19.
OVERVIEW OF THE TRANSFORMATION PROCESSES FIGURE 319
CO
2
Liquids
Electricity
Methane
Ammonia
N
H
H
H
Hydrogen
H
Fisher-Tropsh
Haber-bosh
SMR-H2 + CCS
Ammonia cracking
Methanation
Electrolyses
Power generation
Power generation
Each transformation process has its unique efficiency and elec-
tricity consumption. In the molecule model, the transition from
one molecule to another takes into account the efficiency of the
process. A cost is added based on the electricity consumption
of the transformation and the electricity price in the country
where the transformation is occurring. For instance, the cost
of hydrogen production depends on the price of the primary
molecule, adjusted for efficiency losses and electricity costs. An
example of hydrogen production in 2050 derived from imported
ammonia is illustrated in Figure 3-20. For the conversion to liquids,
different efficiencies are used because this depends on the type
of liquid involved. It is also important to note that the electricity
consumption of these processes is added to the demand in the
electricity model.
EXAMPLE OF THE COST OF HYDROGEN
PRODUCTION FROM AMMONIA FIGURE 320
Electricity cost
Efficiency cost
Molecule cost
[€/MWh]
Hydrogen
from
ammonia
import
79
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios78
Dunkelflaute in an interconnected system
Long-lasting periods of dunkeflaute will be particularly challeng-
ing. During these periods, RES will not be able to cover a signifi-
cant part of the load, meaning that the resulting needs will have
to be covered by exchanges and/or dispatchable capacities.
Since European countries are interconnected, one solution for
dealing with low RES generation infeed in one specific country
is to import electricity from other countries, provided that they
are not experiencing the same situation. For the purposes of this
exercise, a level of RES generation from wind and solar sources
that falls below 20% of their installed capacity on a 24-hour aver-
age basis is considered. Figure 3-24 depicts the occurrence of
low RES generation for Belgium, European Member States, and
the Central Western European (CWE) region which includes the
United Kingdom, France, Belgium, Germany and the Netherlands.
Figure 3-24 shows that the number of low RES generation events
across a pool of countries is significantly reduced. For example,
when looking at Belgium on its own, on average, 2 events per
year lasting at least 7 days occur. By contrast, across the CWE,
one event lasting 5 days occurs every year on. At a European
level, there is on average one occurrence every year of an event
which lasts 4 days.
Consequently, the interconnection of different countries effec-
tively mitigates the risk of enduring long-lasting periods of
dunkelflaute, due to the geographical variability of RES. This
underlines the importance of interconnectors in enhancing the
reliability and resilience of renewable energy generation across
Europe.
Regarding the needed dispatchable capacities, an adequacy
study is performed to ensure that each country reaches its legal
criterion (for Belgium, 3h of loss of load expectation, or LOLE).
During this process, the required dispatchable capacities can be
calculated. The adequacy of the system is extensively discussed
in the results chapters 4 and 5.
OCCURRENCE OF LOW RES GENERATION EVENTS PER YEAR AND LENGTH FIGURE 324
20
15
10
5
0
Number of events per year
Length of event [in days]
CWE
1 2 3 4 5 6 ≥7
CWE: UK, FR, BE, DE and NL
The data are averaged over the 200 climate years considered
3.1.8. CLIMATE YEARS
When performing Unit Commitment and Economic Dispatch
across multiple ‘Monte Carlo’ years, the climate impact must be
accounted for. Firstly, this is important because weather-related
variables impact the final amount of energy that is generated
when RES are involved. Secondly, the weather also impacts the
final amount of energy that is consumed: this rises during colder
periods due to the demand for heating.
As part of this study, Elia uses a forward-looking climate database
developed by climate experts at Météo France. This database
serves to incorporate changing climate conditions into the mod-
elling framework. Specifically, Elia utilises a climate database
generated for the target year 2050, employing greenhouse gas
concentration based on the scenario RCP 4.5. More details about
this forward-looking climate database is available in the latest
Adequacy and Flexibility study [ELI-1].
Heating Degree Days
One commonly used metric for assessing how cold the temper-
ature was during a given period is Heating Degree Days (HDD).
HDD can be considered as an indicator of heating needs, which
become crucial as residential heating will increasingly rely on
electricity, with the electrification of heating through the spread
heat pumps. To calculate the HDD, the methodology developed
by SYNERGRID is used; more details about this methodology is
available on the SYNERGRID website [SYN-1]. Note that SYNER-
GRID relies on the temperature measurement at Uccle. However,
in this study, the population weighted average temperature in
Belgium is used.
Figure 3-22 depicts the HDD distribution for both the current
reference period (1991-2020) and the 200 synthetic climate years
for the climate in 2050. The distribution of the current reference
period is derived from the historical temperature available in the
PECD 4.0 database from ENTSO-E. The 200 synthetic climate
years are provided by Météo France. The average HDD for the
current reference period is higher than the average value for the
2050 climate database. This discrepancy can be attributed to the
anticipated warming trend in future climates. As temperatures
rise, the HDD diminishes, representing a reduction in heating
needs.
HEATING DEGREE DAYS DISTRIBUTION FIGURE 322
2,800
2,600
2,400
2,200
2,000
1,800
1,600
Heating Degree Days ≤[-]
Copernicus 1991-2020 Météo-France 2050
2,857
2,492
2,572
2,248
2,275
1,954
2,015
1,754
1,875
1,552
Occurence of Dunkelflaute events
In 2050, the electricity generation landscape will predominantly
rely on RES. As their output, in turn, depends on weather condi-
tions, one of the main challenges is the occurrence of low renew-
able energy generation during one specific day or even during
several consecutive days. This phenomenon is commonly referred
to in the literature as ‘dunkelflaute’.
Currently, there is no consensus on a formal definition for dun-
kelflaute. Elia suggests that low renewable generation periods
should be defined as the hours during which the hourly capacity
factor of wind farms and the hourly capacity factor of photovoltaic
panels is lower than a certain threshold, as defined in [ENE-1]. It
should be noted that only events which last over 24 hours are
taken into account, with the occurrences being averaged out
over 200 climate years.
Figure 3-23 illustrates the annual frequency of dunkelflaute events
in Belgium, defined using different capacity factor thresholds
and time duration thresholds. For example, on average, there
are between 0.5 to 2 events annually in which neither the hourly
capacity factor of wind farms nor the hourly capacity factor of
photovoltaic panels exceeds 25% for three consecutive days. Logi-
cally, the duration and the occurrence of the dunkelflaute periods
is reduced when considering a lower capacity factor threshold.
FREQUENCY OF DUNKELFLAUTE
EVENTS FOR BELGIUM FIGURE 323
40
35
30
25
20
15
10
5
50
20
10
5
2
0.5
0.2
0.1
0
Capacity factor threshold [%]
1 2 3 4 5 6 7
> Persistence time [Days]
Events
per year
The data are averaged over the 200 climate years considered
81
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios80
OVERVIEW OF THE DIFFERENT SUPPLY AND DEMAND OPTIONS FOR BELGIUM FIGURE 325
2036 2040 2050
Options for the transition
FLEX
Im-
ports
Options for adequacy & system management
Options to supply electrical energy
+2 GW / 5-yr +3 GW / 5-yr
+4 GW / 5-yr+4 GW / 5-yr
Options to reduce the electricity demand
PV/onshore wind
Offshore wind in BE
New nuclear
Non-domestic offshore wind
Electricity imports
Demand flex and storage
Thermal generation
Existing nuclear
Far-out baseload RES
Sufficiency & district heating
3 trajectories: Central. High RES. High RES+PV
From 5.8 to 8 GW in all scenarios for 2040 (Repowering MOG1 & 3
rd
zone)
Increasing flex over time. Sensitivity with higher/lower flex (storage/demand flex)
Keeping existing units until 40Y. New units (H
2
. CCS on existing, CH
4
) possible.
Enough to meet adequacy requirements.
20Y option for D4/T3 & 10 Y option for an additional 1 or 2 GW
[0 – 8] GW
[0 – 0.5] GW
[0 – 4] GW
[15 – 45] TWh
[0 – 2] GW
[0 – 8] GW
[20 – 60] TWh
[0 – 8] GW
[0 – 16] GW
[-60 – 60] TWh
(Electricity imports are a result of the other choices)
New
Exis
3.2.2. ENERGY DEMAND
The final energy demand is defined ex ante and is an input to the
model. There is no optimisation between final energy carriers in
the model; however several scenarios/sensitivities are simulated.
As for other European countries, the final energy demand, feed-
stock as well as international aviation and shipping demand for
Belgium is based on the ‘Distributed Energy’ (DE) and ‘Global
Ambition’ (GA) scenarios established and published for the Euro-
pean TYNDP 2024 framework [ENT-1].
In order to reflect a wider range of possibilities in terms of the
evolution of useful energy needs and the level of electrification
(such as outlined in BOX 3-1), one additional demand scenario
was added for Belgium (as for the European scenarios). The ‘ELEC’
scenario assumes full electrification of road transport, full gas
phase-out for heating in buildings and a further electrification
of high-temperature industrial processes.
In addition, as part of the levers that can be activated for Bel-
gium, the ‘sufficiency’ sensitivity (or ‘SUFF’) assumes a significant
impact of consumer and behavioural changes that reduces the
amount of required useful energy, leading to a decrease in final
energy demand for all energy vectors. This scenario is inspired
on the ‘SHIFT’ scenarios published recently by EnergyVille [EVI-1].
Regarding all demand scenarios/sensitivities, the following should
be noted:
◆ the scenarios presume that existing industrial players will
continue to operate in Belgium;
◆ the industrial loads are assumed to remain in their current
locations;
◆ by 2050, data centre consumption is projected to increase to
10 TWh except for the SUFF sensitivity.
3.2. BELGIAN ELECTRICITY SCENARIOS
3.2.1. INTRODUCTION
The main objective of this study is to analyse and quantify the different supply options
for Belgium to meet its electricity consumption after 2036. A large range of options are
identified that could be put in place in the lead-up to 2050.
The ‘current policies’ scenario (also referred to as ‘Central’) is
the common starting point for all Belgian sensitivities.
The current policies scenario relates to the policy measures that
are already put into place by regional and federal authorities.
Those include for instance the acceleration of domestic RES (both
onshore and offshore), the implementation of the CRM, the exten-
sion of the Doel 4 and Tihange 3 nuclear power plants until 2035,
the approval of the federal and regional grid development plans,
etc. These measures are key for the upcoming decade and should
be further pursued in the years to come. This study takes those
measures as a basis and looks beyond (2036-2050).
First, the consumption scenarios are defined. Those are based on
the same storylines as the European ones from the TYNDP2024
for the DE and GA scenario. In addition, as done for the European
scenario, an ELEC scenario is created to reflect the stronger elec-
trification needs brought forward by several studies.
The Central supply scenario assumes that the current policies
and ambitions are put into place:
◆ the draft updated NECP for Belgium (June 2023) for domes-
tic renewables and electrification. This includes the additional
offshore wind in the Princess Elizabeth zone by 2030;
◆ for later years (after 2030), the same growth rates for domes-
tic renewables as for the period up to 2030 are assumed;
◆ 8 GW offshore wind in the Belgian exclusive economic zone
(EEZ) is assumed in all scenarios for 2050;
◆ the extension of the two nuclear reactors - Doel 4 and
Tihange 3 - until the end of 2035;
◆ the increase in flexibility in residential and industrial settings
through storage and additional demand response;
◆ the closure of the older thermal units in Belgium linked to
their age (> 40 years) - the model can, however, choose to
keep them operational in the system if financially beneficial;
◆ no new nuclear;
◆ the grid projects approved under the latest federal devel-
opment plan for Belgium (such as Boucle du Hainaut, Venti-
lus and Nautilus);
◆ no additional non-domestic offshore;
◆ additional onshore interconnectors (on top of those
approved in the latest federal development plan) found in
the European optimisation for Belgium for each horizon;
◆ adequacy is guaranteed for all the scenarios and sensitivi-
ties. For certain sensitivities, this implies the need to add new
thermal capacities in Belgium.
In addition to the current policies scenario (‘Central’ scenario),
sensitivities are developed for additional supply options:
◆ In addition to the three demand scenarios (DE, GA and ELEC),
a sensitivity related to electricity consumption is accounted
for by moderating consumption through sufficiency meas-
ures (SUFF), and another sensitivity by assuming more heat-
ing networks (HEAT).
◆ Regarding the supply options, several sensitivities are applied:
- two additional domestic RES levers can be activated: high
RES (high onshore and high PV) and very high PV (high
onshore wind and very high PV);
- non-domestic offshore can be connected to Belgium with
a certain potential for each target year;
-
new nuclear can be accounted for as from 2036 with a certain
potential for each target year;
- potential nuclear extensions beyond 2035 are considered as
sensitivity options;
-
far-out baseload RES electrically connected to Belgium is
also considered as an option for 2050;
-
other thermal generation (on molecules) can also provide
electricity depending on the European dispatch, however
their capacity is calibrated to ensure an adequate system
for Belgium.
It shoud also be noted that:
◆ Imports/exports are a result of the European hourly economic
dispatch;
◆ Flexibility is linked to demand scenarios, sensitivities are
applied in Belgium (low and high) to assess its impact;
◆ The required thermal generation capacity (based on mole-
cules) is updated to comply with the adequacy requirements.
The different options are depicted in Figure 3-25.
83
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios82
FEEDSTOCK DEMAND FOR BELGIUM FIGURE 327
DE DE DEGA GA GA
Coal
Oil***
Methane**
Ammonia*
Biomass &
waste
Synthetic
Liquid***
Historical
2010 2015 2020
2036 2040 2050
120
100
80
60
40
20
0
Energy demand [TWh]
ELEC scenario is based on DE scenario for energy demand for feedstock (note that the final synthetic liquid demand might differ based on the model dispatch).
* All historical hydrogen consumption is expressed in fossil fuel origin (methane or naptha via reforming or byproduct) except for the hydrogen for fertilisers which is
expressed as ammonia.
** Methane could be fossil, bio or synthetically sourced, which is defined in the model. It excludes methane used in SMR-H
2
for the production of ammonia (shown
seperately as ammonia).
*** Oil and synhtetic liquids are a result of the model dispatch
Historical values based on EUROSTAT
85
74
8989
92 9292
97 98
9393
ELEC ELEC ELEC
International transport
Whereas international transport makes up around 7% of the final
energy demand in Europe, it makes up 20% of Belgium’s energy
demand in 2021. Fuels for international aviation and shipping
used in Belgium amount to about 17 TWh and 80 TWh per year
respectively (calculated as an average over the last ten years)
[EUS-1]. This is mainly due to ship refuelling (bunkering), with
the Port of Antwerp being a significant hub for this in Europe.
The drop due to COVID-19 in 2020 has since recovered for inter-
national shipping, but remains below the pre-COVID average for
international aviation.
The demand for international shipping is assumed in the TYNDP
scenarios to increase for Belgium, mainly driven by the increase
in international shipping. As mentioned above, the share of oil,
synthetic liquids and biofuels is optimised within the multi-en-
ergy model on a European level, as explained in Section 2.3. It is
however important to mention that this consumption is modelled
at European level and those fuels do not necessarily need to be
produced in Belgium.
INTERNATIONAL SHIPPING AND TRANSPORT DEMANDS IN BELGIUM FIGURE 328
140
133
122
140
133
122
140
133
122
DE DE DEGA GA GA
2036 2040 2050
ELEC ELEC ELEC
Oil**
Methane*
Biofuels**
Synthetic
Liquid**
Historical
2010 2015 2020
109
88
160
140
120
100
80
60
40
20
0
Energy demand [TWh]
*Methane could be fossil, bio or synthetically sourced, which is defined in the model
** Biofuels, synthetic liquids and oil are a result of the model dispatch
Historical values based on EUROSTAT
3.2.2.1. FINAL ENERGY DEMAND
Belgium’s final energy demand has remained relatively stable
over the past decade at around 400 TWh. However, the excep-
tional circumstances of 2020, which were primarily caused by
the COVID-19 pandemic, resulted in a drop in demand to approx-
imately 380 TWh. Following a slight rebound in 2021, high energy
prices in 2022 and 2023 prompted a subsequent decrease to
roughly 370 TWh. The primary driver of this reduction was a
decrease in industrial output, particularly from energy-intensive
industries. Additionally, energy use in residential and commercial
buildings also declined significantly.
Similarly to the trend across Europe, the final energy demand is
assumed to decrease significantly (a reduction of between 25%
and 45% by 2050). This reduction can be attributed to a combi-
nation of energy efficiency measures (such as electrification),
behavioural changes and a shift in transport modes. In the SUFF
sensitivity, additional behavioural measures across all sectors
further reduce the final consumption of energy.
The most important factor driving the overall demand reduction
in all scenarios (with varying intensity) is the electrification of
buildings, transportation and industry which carries a higher
inherent efficiency than traditional fossil fuel-based processes
and appliances. In Appendix I, more information can be found
on the share occupied by electricity and other fuels in the final
energy demand for some key demand sectors in the year 2050.
FINAL ENERGY DEMAND FOR BELGIUM EXCL. NONENERGY FEEDSTOCK & INTERNATIONAL
TRANSPORT FIGURE 326
Final energy demand reduction
(electrification, sufficiency and
energy efficiency)
-25% to -45% reduction of final
energy demand
Electricity
Increasing from 24% in 2019
towards
55% to 80% share of final energy
demand
Gaseous Molecules
12% to 38% share of final energy
demand
Coal**
Liquids*
Hydrogen
Methane
Heat
Biomass &
waste
Electricity
Other
GA
DE
ELEC
SUFF
GA
DE
ELEC
SUFF
GA
DE
ELEC
SUFF
360 360
300
320 310
250
310
280
240
260
230 220
2036 2040 2050
2010 2015 2020
Historical
Excluding international aviation & shipping and non-energetic feedstock, including grid losses and refineries.
Note that energy demand for transformations such as power-to-hydrogen and carbon capture are not included. Values are normalised for historical climate while in
the simulations, a forward-looking climate database is used, therefore the simulated demand can differ from these input values.
* Methane & liquids could be fossil, bio or synthetically sourced, which is defined in the model.
** Coal as defined as final energy demand per EUROSTAT (i.e. excluding coal consumed in blast furnaces)
Historical values based on EUROSTAT
450
400
350
300
250
200
150
100
50
0
Final energy demand [TWh]
Not yet included are electricity for
electrolysis and CCS/U (see later)
370
400
Feedstock for non-energy usages
Belgium has the fifth largest feedstock demand in the EU, which
is mainly explained by the presence of the petrochemical cluster
in the port of Antwerp. The latter is one of the largest in the world
and is a hub of activity that includes numerous international
petrochemical companies, with a heavy concentration of refin-
eries, chemical producers, and related industries. The Antwerp
petrochemical cluster is renowned for its integrated value chain. It
transforms crude oil and natural gas into a multitude of chemical
products and plastics.
Figure 3-27 depicts the changes in feedstock demand in this
study. Until 2036, oil products such as naphtha drive the largest
portion of the feedstock demand along with some biomass. In
the lead-up to 2040 and 2050, synthetic liquids could become
viable via (for example) the methanol to olefins process, in which
methanol is converted into ethylene and propylene. As explained
in Section 5.1 , this synthetic feedstock would not be synthetised in
Belgium - it would need to be imported. The demand for ammo-
nia used in fertilisers remains relatively stable compared to today.
The origin of the ammonia is discussed in Section 5.1.
85
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios84
Sufficiency
In addition to the three demand scenarios which are also consid-
ered for Europe (DE, GA & ELEC), an additional sensitivity related
to energy demand is foreseen specifically for Belgium. The SUFF
sensitivity involves the impact of changes in behaviour, lifestyle,
and smarter choices related to technology and design which are
aimed at achieving the same or better results with less energy.
Sufficiency measures could include actions like reducing and/or
improving the use of vehicles (more car sharing, shifting to soft(er)
transport modes…), reducing the set temperature of heating
devices, or reducing the consumption of goods in general.
Introduction to sufficiency
Sufficiency is a concept related to resource use as a whole, but
in the framework of this study, we will focus on the energy use.
It has been described by the IPCC as policies, measures and daily
practices involving sufficiency aims to avoid demand of resources
(energy, materials, water and land) while still ensuring well-being
[IPC-1]. It can be described as guaranteeing a sufficient level of
services (heating, transport, industrial production), while adjust-
ing their nature and quantity to reduce environmental pressure.
Sufficiency is different from energy efficiency. Efficiency implies
reducing the energy used in inputs, while delivering the same
quantity in outputs. Sufficiency is about redefining the means
to deliver the service, or reconsidering the outputs needed. For
example: Driving a smaller car reduces the energy inputs needed
while delivering the same service. This is labelled as a sufficiency
measure. Whereas energy efficiency would mean using the same
car with a more efficient engine, or better aerodynamics reduc-
ing friction and energy losses. Both reduce energy needs, but in
different ways.
Sufficiency policies go beyond temporary voluntary agreements
of energy reduction. The European Sufficiency Database has doc-
umented +350 policies that can implement sufficiency [ENS-1],
and these can be of different natures (economic, fiscal, educa-
tional, structural, etc...).
Sufficiency is often linked to behavioural changes, which would
happen on a voluntary basis. But beyond this, documented also in
the sufficiency database [ENS-1], are more societal and structural
measures that take longer to be implemented (such as fiscal
incentives to reduce the size of cars, or urban planning leading
to a greater modal shift).
It should be noted that sufficiency is a highly cleaving concept.
As it is linked to behaviours and habits, it proves to be a concept
charged with political implications and up to debate. It makes
it an unavoidable debate to have, to establish a credible energy
pathway for Belgium and Europe. This is the reason why it has
been treated as a sensitivity in this study, to explore its potential
impacts on the energy system.
Sufficiency recognised by European and worldwide institu-
tions
Sufficiency is a concept gaining traction in the energy world and
is now documented as a lever to decarbonisation. Here are a few
organisations and studies making use of it in modelling exercises:
◆ EnergyVille recently released a SHIFT scenario in the frame-
work of their PATHS2050 study. The latter leads to a reduction
in electricity generation of 20% leading to lower investment
cost for the energy transition and pushing CO2 reduction
towards 60% by 2030 and 90% by 2040 [EVI-1]. This EnergyVille
scenario serves as basis for the SUFF sensitivity for Belgium
presented in this paper.
◆ The Intergovernmental Panel on Climate Change (IPCC)
[IPC-1] in its 2022 report, estimates that this, alongside other
demand side measures (such as changes in urban planning
and end-use technology) can reduce global GHG emissions
in end-use sectors by 40 -70% by 2050, making it one of the
main levers for mitigating climate change and CO2 emissions.
◆ RTE Futurs Energétiques has explored a low energy scenario
where energy demand is reduced solely through behavioural
changes. They estimate that a 14% reduction in total energy
needs by 2050 compared to the central scenario can be
achieved. This scenario showed to reduce the needs of addi-
tional thermal capacities by 10 GW, a reduction in materials
for battery electric vehicles by more than 30%, a reduction in
flexibility needs of the system and in CO2 emissions. [RTE-1]
◆ The CLEVER study is a European study done with more than
26 partners across the EU delivering a net-zero trajectory for
all EU countries. [CLE-1]
◆ Elia in its latest Adequacy and Flexibility study has showed
that sufficiency measures could reduce the needed volume
for Adequacy by more than 1 GW [ELI-1]
Politically, the concept is also gaining more traction:
◆ During the 2022 energy price crisis, the French energy minis-
try released a sufficiency plan that delivered a 12% reduction
in gas and electricity when compared to yearly consumption
normalised with respect to the temperature [FRG-1].
◆ 70+ European organisations signed a sufficiency manifesto in
March 2024 calling for the EU to manage demand through
sufficiency policies [ACR-1].
So, sufficiency has shown political relevance in France and among
EU organisations. It is also investigated as an energy transition
scenario, where it has shown in simulations its potential to reduce
the use of materials and rare earths, help ensure Security of Sup-
ply and relieve pressure on the grid by curbing the rise in electric-
ity consumption and lower CO
2
emissions (by reducing the needs
for all energy vectors). This underlines the relevance of sufficiency
in climate strategies and the energy transition.
However, as outlined in RTE's `Futurs Energétiques', this concept
is often not well understood and sometimes ill-defined. It is still
to be debated what the impacts and socio-economic costs of
sufficiency could be, and whether or not all behavioural changes
expected and implied by sufficiency measures would be imple-
mented and accepted by the population.
For the quantification of this scenario, the ‘SHIFT’ scenario devel-
oped by EnergyVille is used as inspiration to derive the SUFF
scenario. Starting from the DE scenario, a reduction of end use
demand is applied. For a detailed presentation of the scenario,
please see [EVI-1].
Table 3-3 and Table 3-4 include a comparison of some key demand
drivers in the DE and SUFF demand scenarios.
3.2.2.2. ELECTRICITY DEMAND
Whilst the final total energy demand decreases, in all 3 demand
scenarios (DE, GA and ELEC) a strong increase in electricity
demand can be observed, ranging between +110% and +130%
compared to 2022. The sensitivity SUFF assumes an increase of
+95% compared to 2022.
These trends are comparable to changes across the EU, with a
key difference being the level of electrification in industry which
makes up a relatively important share of Belgium’s energy
demand and which varies greatly between the different demand
scenarios.
The transport sector is assumed to experience the largest elec-
tricity demand increase of all sectors. In 2021 the transport sectors
consumed ~2 TWh of electricity of which most is attributed to
the train subsector. This value is set to increase at least tenfold
by 2050 in all of the scenarios, ranging between 20-33 TWh. Both
the ELEC & SUFF scenario assume a full electrification of road
transport by 2050, however, the measures in terms of reduced
person & freight travel, better loading factors and modal shifts
manage to decrease electricity requirements by around 30%.
Electricity demand in buildings remains relatively stable. On the
one hand, the strong rollout of electric heat pumps (more impor-
tant in DE & ELEC scenarios) increases the electricity demand. On
the other hand, this is compensated by the assumed reduction
in heating needs due to renovations but also due to the high
efficiency of heat pumps and the replacement of old electrical
appliances and heating devices by more efficient ones. On top of
that, the SUFF scenario assumes people will lower their heating
temperature and heat spaces in general.
Today, the industrial demand for electricity mainly stems from
non-thermal workloads such as compressors, machinery, lighting
etc. Nearly all industrial heat is supplied by combustible fuels.
Electrification has a key role to play in order to decarbonise heat
in this sector. The range between the DE, GA and ELEC scenarios
can mainly be explained by the uncertainty linked to the cost
and technical feasibility of electrification of higher temperature
heat processes.
In the GA scenario, combustible fuels remain the key energy
driver, albeit in the form of decarbonised molecules such as biom-
ethane and hydrogen (derivatives). In the DE scenario, most of the
low and medium temperature heat is assumed to be electrified
using already existing technologies. This includes industrial heat
pumps in the food and paper industries, along with the recovery
of derived heat from other industrial processes and e-boilers
in the chemical sector. The direct reduction of iron with meth-
ane (and, in later years, hydrogen) in combination with electric
arc furnaces is assumed to be applied for steelmaking. (Green)
molecules such as biomethane and hydrogen still have a role to
play in some high-temperature heat processes. The ELEC sce-
nario assumes that all industrial heat is mostly electrified in the
form of industrial heat pumps, e-boilers, microwaves, infrared
heaters, induction and resistance heaters in the metal sector,
electric boilers and crackers in the chemical sector, electric arc
furnaces and electrolysis steel in the steel industry and electric
kilns in the cement industry; each of these is considered to be
commercially available and implemented at scale by 2050. In this
scenario, almost no hydrogen is used for process heat and it has
only a limited role to play in some industrial processes such as in
steelmaking as a reducing agent. (Bio-)methane still has a small
role to play for some high-temperature energy uses. Finally, the
SUFF sensitivity also assumes a relatively high electrification rate,
but a lower overall energy demand due to a more resource-effi-
cient and circular economy, leading to a lower need for primary
production of materials.
Note that all scenarios (except SUFF) assume around 10 TWh of
data centre demand by 2050.
BELGIUM’S ELECTRICITY DEMAND IN EACH OF THE DIFFERENT SCENARIOS FIGURE 329
GA GA GADE DE DE
ELEC ELEC ELECSUFF SUFF SUFF
Historical
Transport
Industry
Electrolysis
CCS +
Synfuels
Household
Other
Losses
Tertiary
200
180
160
140
120
100
80
60
40
20
0
ELEC
135%increase
DE
120%increase
GA
115%increase
SUFF
100%increase
2010 2015 2020
2036 2040 2050
125
83
134
141
117
159
165 172
145
178
184
196
168
Values are normalised for historical climate while in the simulations a forward looking climate database is used, therefore the simulated demand can differ from these
input values. Electrolysis, CCS and the production of synfuel is optimised within the model and the associated electricity demand depends therefore on each potential
scenario and sensitivity.
Historical values based on EUROSTAT & Elia internal data
Electricity demand [TWh]
87
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios86
3.2.3.1. ONSHORE RENEWABLES SOLAR
Solar PV has significantly increased over the past five years due to lower installation costs and soaring energy prices.
The future capacities for solar production are determined based on the following:
◆ the Central scenario is an extrapolation of the growth observed over the last three years;
◆ the High scenario assumes a twofold increase in that growth rate;
◆ the Very High scenario anticipates a more than threefold increase in the growth rate.
For the Very High scenario, the meticulous management of oversupply and distribution grids will be necessary. Further details
about this are included in the results section. Both utility-scale and residential-scale PV is considered.
SOLAR CAPACITIES FOR BELGIUM FIGURE 330
Deployment options
in GW/year
Avg. ‘20-’22
Historical
By 2050
=
+1.2
GW/y
=
=
=
Capacity in GW
High
Central
Low
Very high
Solar
+0.45
GW/y
‘10 ‘15 ‘20 ‘25 ‘30 ‘36 ‘40 ‘45 ‘50
97.5 GW
+4
GW/y
65.5 GW
+2.4
GW/y
41.5 GW
+1.2
GW/y
29.5 GW
+0.6
GW/y
1.11.1 3.43.4
5.65.6 11.5 17.5 23.5
29.5
6
12
12
24
16
32
BARRIERS ENABLERS
◆ Supply chain challenges – As global development inten-
sifies, manufacturing capacity and the upstream supply of
input materials could struggle to keep pace with it, leading
to delays and cost inflation. Similar challenges for installation
supply chain and workforce required to install PV. [IEA-4]
◆ Distribution network integration – At moderate levels of
deployment, distribution network investments are expected
to be driven by peaks in winter evening consumption caused
by electric vehicles and heat pumps [FLU-1]. However, at very
high levels of deployment, solar generation might exceed
the network capacity dimensioned for evening peaks in
more and more feeders; this would require additional net-
work capacity, more self-consumption and/or local flexibility.
◆ Land and Space Requirements – for large-scale, ground-
mounted solar farms, substantial land area is needed. This
requirement might lead to competition for land that could
otherwise be used for different purposes, such as agricul-
ture. [SPE-1]
◆ Cost-effective option – Solar panel costs have dropped sig-
nificantly, with installation costs accounting for increasingly
larger shares of the total system cost for end users than the
PV panels themselves [FIT-1].
◆ Prosumer enabler – Small-scale solar is the most decentral-
ised generation technology, allowing individual households
to generate their own energy. It is key for building energy
communities, where consumption is (partially) managed at
a local level through smart use of solar along with electric
vehicles, batteries etc. [EUU-1]
◆ Most modular technology – PV systems are very modular;
individual panels can be produced identically, and com-
bined in a straightforward way to form a larger system. This
allows for standardisation, leading to economies of scale and
bringing down costs.
What can speed up deployment?
◆ Residential incentives – Providing additional incentives (sup-
port to specific consumers for instance) for residential use
cases can increase the attractiveness of installing solar panels
and should convince additional households to install them.
◆ Upgrading DSO grids – In feeders where peak injection
causes curtailment due to voltage rises beyond technical
limits, increasing network capacity can enable their further
deployment [IET-1].
◆ Large & utility-scale PV – As a segment with little develop-
ment in Belgium compared to residential PV, utility-scale PV
could be a way to develop additional capacity at the same
time; Flanders will enforce this through a policy requiring
large consumers (>1GWh annually) to install PV in proportion
to their available roof space [FLG-1]. Agrivoltaics could also be
an option for further PV expansion [KUL-2].
RELATIVE IMPACT OF SUFF VS DE SCENARIO IN 2050 TABLE 33
Relative impact in 2050 between
SUFF and DE Measure
SUFF
DE
Industry
Cement production (kton/y) -39%
Ceramics production (kton/y) -33%
Steel production (kton/y) -22%
Ammonia production (kton/y) -16%
HVC production (kton/y) -20%
Non-ferrous production (kton/y) -11%
Oil refineries production (kton/y) -100%
Transport - Passenger
Evolution passenger travel (passenger-km) -25%
Transport - Freight Evolution freight transport (ton-km) -11%
Building - Residential
Living space per person (m
2
/capita) -16%
Heating setpoint (°C) -7%
Building - Tertiary
Work space per person (m
2
/capita) -16%
Heating setpoint (°C) -7%
DIFFERENCE BETWEEN TRANSPORT USAGES IN THE DE VS SUFF SCENARIO TABLE 34
Transport usage in 2050 Measure
DE SUFF
Transport - Passenger
Modal share – cars 74% 55%
Modal share – rail 9.5% 16%
Modal share – public transport 7% 13%
Modal share – active (bike, walk) 7% 16%
Trip sharing (person/car) 1.5 2.3
Transport - Freight
Modal share – road transport 69% 54%
Modal share – rail 16% 18%
Modal share – inland navigation 14% 28%
Load factor vans & trucks (ton/vehicle) +0% +13%
3.2.3. ENERGY SUPPLY
The energy supply options are defined ex ante for Belgium. In
order to grasp the impact of different developments, several sen-
sitivities are defined for each type of electricity supply in Belgium.
In addition the biomethane domestic supply is also accounted
for in the multi-energy model.
This sections deals mainly with the electricity supply options.
One sheet for each type of supply is discussed that outlines the
different scenarios/sensitivities, barriers and enablers.
A small part will also be dedicated for far-out RES connected in
areas not modelled explicitly in this study, such as North Africa.
The following type of supply for Belgium is discussed:
◆ Onshore renewables
- Solar PV
- Wind onshore
◆ Offshore renewables
- Domestic wind offshore
- Non-domestic wind offshore
◆ Nuclear
- Existing fleet
- New build
◆ Other renewables
◆ Other thermal fleet
NOTE
The analysis in these technology sections is preliminary and not exhaustive; it is based on external sources and
has as sole purpose to illustrate the type of questions that need to be addressed on top of the rather quantitative
approach used in the simulations.
89
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios88
3.2.3.3. OFFSHORE WIND DOMESTIC
Belgium emerged as a front-runner in domestic offshore wind development, successfully completing the first phase of instal-
lations with a capacity of nearly 2,300 MW by 2020. The projected growth includes the commissioning of the Princess Elisabeth
Zone (PEZ) by 2030, which is expected to add a capacity of 3,500 MW.
Post-2030 options for further expansion (both assumed that could be developed for 2040):
◆ The potential repowering of the initial zone (MOG 1), which could augment the capacity by approximately 700 MW;
◆ The exploration of a potential third offshore zone in Belgium, estimated in this study to yield around 1,500 MW.
Implementing both options could elevate the installed offshore wind capacity to 8,000 MW. Studies regarding the feasibility of
expanding Belgian offshore production to a potential of 8,000 MW are currently being carried out by the Belgian government.
Apart from the above measures (repowering of the first offshore wind zone and developing a third offshore wind zone in Belgian
offshore waters), the deployment of offshore floating solar power could also be part of the solution. However, this possibility has
not been evaluated in the present study.
It is important to mention that as a result of the capacity expansion simulations, the repowering of MOG1 and the third offshore
zone is always invested in by the model in all of the sensitivities and simulations and therefore prove to be a cost-effective solution.
DOMESTIC WIND OFFSHORE CAPACITIES FIGURE 332
Offshore wind
LOWCENTRAL
‘10 ‘15 ‘20 ‘25 ‘30 ‘36 ‘40 ‘45 ‘50
Capacity in GW
Historical
HIGH
+1.5GW: new zone in EEZ
+0.7 GW repowering MOG 1
2.3
0.2 0.7
2.3
+3.5 GW: Princess Elisabeth Zone
00
5.8
0.7
5.85.8
1.5
BARRIERS ENABLERS
◆ Limited space available – Belgium has a very small EEZ
which also needs to accommodate (amongst other things):
shipping routes, fishing areas and military uses. [FGV-2]
◆ Legal framework – EU legislation limits the ways in which
the federal government can re-organise and optimise exist-
ing concessions in an effort to repower existing farms. [FGV-
3]
◆ Supply chain challenges – As global development intensi-
fies, manufacturing and installation capacity as well as the
upstream supply of input materials could struggle to keep
pace with it, leading to delays and cost inflation. [RAB-1]
◆ Technological developments in turbine making – Since the
development of the first zones, wind turbines have become
significantly larger; repowering these zones has a potential
capacity increase [FGV-3].
◆ Close to shore – Farms within the Belgian EEZ are relatively
close to the shore, which lowers the costs of bringing the
energy to shore.
◆ Past experience and Belgian expertise – As a pioneer in the
industry [NSS-1], Belgium has amassed significant knowl-
edge and has a robust supply and installation chain estab-
lished.
Key decisions
◆ New zone in Belgian EEZ – Find new offshore zone(s) in the
Belgian EEZ, as the zones which have been identified until
now only allow the country to establish 5.8 GW by 2030. Such
a decision would need to be taken by 2030 to connect the
wind farms by 2040 based on the development of the two
previous zones.
◆ Repowering MOG I – Decide how to proceed with repowering
and adapt the current legal framework accordingly. [FGV-3]
Risks
◆ No space left – Due to Belgium’s small EEZ, following discus-
sions with maritime stakeholders, it is possible that no suita-
ble zone(s) can be found.
◆ Delays in permitting – Permitting delays could occur for
several reasons (environmental concerns for the maritime
environment, NIMBY pushback against onshore grid devel-
opment, etc.).
3.2.3.2. ONSHORE RENEWABLES WIND
Over the past decade, onshore wind capacities have been deployed at a steady pace. However, the rate of installation has slowed
down mainly due to permitting issues.
Future onshore wind capacities are determined based on the latest draft NECP (June 2023) for 2030 which already assumes a
doubling of the installation rate compared to historical levels. Post-2030 assumptions are as follows:
◆ in the Central scenario: a rate of increase which is similar to historical levels;
◆ in the Low scenario: an increase reduced by half beyond 2030;
◆ in the High scenario: a doubling of the trend beyond 2030.
WIND ONSHORE CAPACITIES FIGURE 331
Onshore wind
(NECP)
Historical
+0.35
GW/y
Capacity in GW
‘10 ‘15 ‘20 ‘25 ‘30 ‘36 ‘40 ‘45 ‘50
+0.18
GW/y
Deployment options
in GW/year By 2050
+0.4
GW/y
=
+0.2
GW/y
=
+0.1
GW/y
=
High
13.6 GW
Central 9.6 GW
Low
7.6 GW
Avg. 20-’22
0.7 1.41.4
2.52.5
3.9
5.6
6.6
7.6
1.0
2.0
2.0
4.0
BARRIERS ENABLERS
◆ Permitting / NIMBY – Securing permits for onshore wind
developments can be a significant obstacle [WEU-1], par-
tially due to local residents adhering to a not-in-my-back-
yard (NIMBY) approach.
◆ Supply chain challenges – As global development inten-
sifies, manufacturing capacity and the upstream supply of
input materials could struggle to keep pace with it, leading
to delays and cost inflation. [IEA-6]
◆ Distribution/transmission integration – Plans for grid con-
nections need to be outlined with developers, to ensure the
timely reinforcement of high-voltage distribution [FLU-1]
and transmission grids.
◆ Cost-effective option – Currently, onshore wind is the
cheapest form of renewable generation in Belgium, mak-
ing it a cost-effective option in areas where permitting chal-
lenges can be overcome. [IRE-2]
◆ Can be developed locally – Whilst not as decentralised/
small-scale as residential solar, onshore wind can also be
developed in energy communities which residents can par-
ticipate in [ECF-1].
◆ Advances in spatial planning – A reduction in permitting
obstacles (such as ones related to aviation security [SKY-1])
can speed up the process.
What can speed up deployment?
◆ Accelerating permitting procedures – Lengthy permitting
procedures present a key challenge for onshore wind deploy-
ment; streamlining permitting procedures and making sure
sufficient capacity is available to process them can go a long
way in speeding up deployment.
◆ Spatial planning – NIMBY approaches are a key factor in
lengthening and complicating permitting procedures, mean-
ing that improved spatial planning could mitigate some local
concerns and therefore speed up deployment. Furthermore,
reassessing areas that are currently off limits for wind devel-
opment—like aviation routes [SKY-1] and lands reserved for
other uses—could open up new zones for potential wind
energy generation.
◆ Involvement of local communities – co-development of
wind projects with local communities could ease the permit-
ting and acceptance of new farms. It can potentially lead to
shared economic benefits. [LEC-1]
◆ Technological developments in turbine making – Since
the development of the first wind zones, wind turbines have
become significantly larger; repowering these zones carries a
large potential of capacity increase and yield increase.
91
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios90
Key decisions
◆ Strong political support in the short and long term – It
will require active political follow-up and support from poli-
cymakers to find the right solutions that unlock the offshore
potential fully such as the Belgian government taking a lead-
ing role (proactively shaping and proposing solutions to these
different complexities and barriers).
◆ Cooperation at a regional (sea basin) level together with
other countries as well as on a European level is key. Achiev-
ing such solutions will not be easy given the complexity and it
might take time to eventually get there. It is however impor-
tant not to wait until all elements are solved and evolve pro-
gressively on this matter. A pragmatic approach is required to
ensure that investment decisions for the first hybrid projects
can be taken in the next 1-2 years so that they can be ready by
the assumed time horizon of 2036.
◆ Consider grid and generation together – When taking the
necessary decisions about improved planning, funding and
cost sharing it is important to simultaneously address the
challenges concerning offshore grid infrastructure and off-
shore generation which may require some de-risking mech-
anism in the form of support scheme such as contracts for
difference.
Risks
◆ Challenges related to cost sharing – The system benefit of
an hybrid interconnector can be different for both countries,
not to mention that also third countries can benefit indirectly.
This is hard to quantify, and, given that significant amounts of
money are concerned, this could lead to lengthy negotiations
between the involved countries when deciding how to split
the investment.
◆ Need to involve non-EU countries – In order to harvest the
full potential of the Nort Seas, it will be important to have
all countries of the sea basin cooperate, which includes the
non-EU countries Norway and UK. Next to the points men-
tioned before, their involvement also requires addressing
other barriers such as an improved market design to enable
efficient cross-border trades.
3.2.3.4. OFFSHORE WIND NONDOMESTIC
The term non-domestic offshore wind is used throughout this study to refer to offshore wind farms located in zones outside of
the Belgian EEZ (therefore typically in another country’s EEZ when considering the North Sea or Atlantic Sea) which are directly
electrically connected to Belgium (radially, or via hybrid solutions) and for which a support mechanism is foreseen in which the
Belgian state takes part (like f.i. a CfD), if any. This means that the costs related to both the grid infrastructure and the offshore
wind farms themselves are accounted for in the total system cost for Belgium.
Looking at the commitments made by the nine countries of the North Sea summits in Esbjerg (2022) and Ostend (2023) to
develop 300 GW of offshore renewable energy capacity in the North Seas by 2050, the potential of this non-domestic offshore
wind is enormous and indeed a potential choice for Belgium’s future energy supply. The significant lead times (10-15 years) for
such complex projects require policy makers and relevant stakeholders to provide, in the short term, clear answers to the chal-
lenges and shortcomings in current regulatory and legal framework related to the non-domestic offshore wind that currently
hamper the progress in their development/build out. These challenges are further described in BOX 3-4.
As the graph below shows, we assume in this study that:
◆ at most 4 GW of capacity can be added every 5 years from 2036 onwards (taking into account both the important grid infra-
structure work and complex negotiations with international partners required for these kinds of projects and the progressive
improvement of the regulatory framework at international level).
Several intermediate levels of deployment are analysed for the key scenario years.
NONDOMESTIC OFFSHORE WIND POTENTIAL CAPACITIES FINANCED AND CONNECTED TO
BELGIUM FIGURE 333
30 36 40 45 50
16 GW
12 GW
8 GW
4 GW
Assuming at most possible to
add 4 GW per 5 years from
’36 onwards
BARRIERS ENABLERS
◆ Shortcomings of the national perspective – The develop-
ment of the offshore grid and generation will involve mul-
tiple countries of a sea basin for which the existing bilateral
and individual project-by-project approach based on bot-
tom-up national plans is not a sustainable way forward.
◆ Unprecedented funding needs – The ambitions linked to
harnessing offshore wind in the North Seas will require very
large amount of investment for which current amounts (e.g.
CEF funds) and current way of funding projects individually
will not be effective.
◆ Shortcomings of regulatory framework – The current
regulatory framework is not able to find the most optimal
offshore grid and a right way of sharing costs and bene-
fits which leads to lengthy and inefficient processes with
almost no chance of success.
◆ Large potential – Unlike Belgium’s EEZ, other North Sea
countries have much larger EEZs with a very large potential
for offshore wind development (though further offshore),
with a stated combined ambition of 300 GW by 2050 [FGV-
4].
◆ Improved planning in the Sea Basin – In order to identify
projects which carry most of the value for European society
there is a need to strengthen joint TSO-led planning at sea
basin level whereby synergies across different projects in a
sea basin are considered.
◆ Funding and cost sharing – As outlined in the recent paper
co-written by Orsted and Elia Group, (new) regional funding
and cost sharing mechanisms are needed. Having simple
cost and benefit sharing rules from the start and finding
additional funding streams (institutions, Member States,
third countries and private investors) are key enablers for
more non-domestic offshore/hybrid interconnectors [ELI-9].
OTHER NON-DOMESTIC RENEWABLE OPPORTUNITIES OUTSIDE OF THE SIMULATED PERIMETER
Given the large needs for additional renewable generation
capacity over the next few decades, part of the solution
could be to create interconnectors to other countries which
could develop renewable surpluses, or which have genera-
tion profiles that complement Belgium’s own renewables.
One key example is connecting to offshore wind in other
countries, the focus of the section on non-domestic off-
shore. Extending the scope, there are other regions with
high renewable potential; North Africa is one such example
discussed in more detail below, besides other regions not
discussed in this study (for example Greenland [ENE-1], Ice-
land [ICE-1] and floating wind in the Atlantic Ocean [WEU-2]).
According to IRENA, North Africa’s unique geography and
climate make it a region with immense renewable energy
potential, especially for solar & wind; they estimate the tech-
nical potential at 2300 GW solar and 223 GW wind [IRE-3].
Various European countries are investigating or planning
the construction of HVDC interconnectors to North Africa.
Below is a list of three such projects, ordered by increasing
ambition in terms of distance and capacity:
◆ Elmed
- Connecting Italy-Tunisia
- 0.6 GW across a distance of 220 km [ELM-1]
- Status: financing secured, construction authorised by
Italian government
- Expected to be commissioned in ‘28
◆ EuroAfrica Interconnector
- Connecting Greece-Cyprus-Egypt
- 2 GW (first stage of 1 GW) capacity across a distance of
1400 km [EAI-2]
- Expected to be commissioned in ‘28-’29 [EAI-1]
◆ Morocco-UK Power Project (by Xlinks)
- Connecting UK to Morocco
- 3.6 GW capacity across a distance of 4000 km [REC-1]
-
Status: Raising private capital [XLI-1], conducting public
consultations [XLI-2]
- Ambition to be commissioned by ’30 [REC-1]
The Morocco-UK Power Project by Xlinks most resembles
what it would take for Belgium to connect to Northern Africa
in terms of distance covered and is therefore discussed in
greater detail here.
The interconnector proposed by Xlinks would consist of 4
cables, each 4000 km long, forming a twin 1.8 GW HVDC
system. An agreement has been reached with National Grid
for two connections for a total of 3.6 GW in the UK, Devon
[XLI-3]. Furthermore, Xlinks is additionally investigating the
feasibility of connecting to other markets, such as Germany
[REC-1].
The project goes beyond the interconnector alone by also
planning for 11.5 GW solar & wind power and 22.5 GWh / 5
GW battery storage system in Morocco. That renewable
capacity in Morocco offers several benefits over the same
in the UK [XLI-3]:
◆ Solar: Irradiance is over twice that of the UK, yielding
more energy for the same capacity, and winter days
have more hours of sunlight
◆ Wind: The local Moroccan wind system, the Trade Winds,
is consistent, providing a more stable energy source for
wind turbines
A sensitivity will be assessed in this study by considering ‘far
out RES’ connected to Belgium. As a reference the costs will
be taken from the project by Xlinks.
BOX 3-3
93
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios92
3.2.3.5. NUCLEAR FLEET EXISTING
The Central scenario considers the phase-out of nuclear power in accordance with the law introduced in 2003, which was
amended in 2013 and 2015 to cover the operational lifetime extension of Tihange 1 and Doel 1 and 2. In addition, the Central
scenario also considers the lifetime extension (also referred to later as long-term operation – LTO) of Doel 4 and Tihange 3 for 10
years, as approved by the federal government and ENGIE.
A number of sensitivities are evaluated for the period post-2036:
◆ Prolonging the extension of D4/T3 by an additional 10 years (leading to a total extension of 20 years), which would extend
their operation until the end of 2045.
◆ Extending an additional 1 or 2 GW for 10 years. The specific timing of potential works, restart timing and reactors involved are
not detailed. The sensitivity only assumes those could be present in the target years 2036 and 2040. Several political parties
proposed such an extension ahead of the elections [PLB-1].
It is worth noting that this study does not evaluate the feasibility or other necessary safety, technical or legal measures and
other consequences of extending the operation of these reactors. The primary objective of this study is to assess the impact on
the electricity system.
EXISTING NUCLEAR CAPACITIES EVOLUTION AND POTENTIAL EXTENSIONS CONSIDERED IN THIS
STUDY BEYOND CURRENT FRAMEWORK FIGURE 334
10 15 20 25 30 36 40 45 50
21 2121
42
5.9
Historical capacity of 5.9 GW
across plants in Doel and
Tihange
10y lifetime extension beyond
40y for oldest reactors (D1&2
and T1)
59 59
2121
42
1010 10 10
1010 10 10
Nuclear
phase-out
starts in
’22 and
completes by
end of ‘25
Additional 10y
extension of
D4/T3
10y extension
of additional 1-2
GW reactor(s)
D4/T3 reactors
restart in
winter ’25-’26
for 10y
Initiality planned capacity
Decided extensions
Scenarios to be
decided upon
BARRIERS ENABLERS
◆ Need for political consensus – The multifaceted debates
surrounding nuclear energy add a layer of complexity to the
decision-making process in this field. [UGE-1]. A political con-
sensus is needed and the necessary legislation and other
relevant regulation needs to be adapted.
◆ Technical & safety feasibility – The technical & safety related
feasibility of lifetime extensions needs to be assessed for
each reactor individually and in line with regulation, techni-
cal and grid constraints.
◆ Willingness of the nuclear operator – The operator has
voiced concerns about the further extension of more reac-
tors. [DTI-1]
◆ Most political parties indicated their willingness to pro-
long more nuclear – in the recent elections, several parties
indicated their willingness to further extend reactors beyond
2025. [PLB-1]
◆ No need to secure new locations – Unlike building new
reactors, extending existing ones would not require new
locations to be secured: a complex process involving many
stakeholders.
Key decisions
◆ Additional lifetime extension of D4/T3 – Currently D4/T3 are
set to close in 2035 after a 10-year lifetime extension. Given
the necessary time to reach an agreement and realising LTO
works, a timely decision is needed to allow for an additional
10-year extension until 2045 [DST-1].
◆ Extension of additional reactors – A decision is needed
about whether to extend an additional 1-2 GW of reactors as
some of the reactors are already shut down and preparations
for dismantling are ongoing and others are going to be shut
down in 2025.
Risks
◆ Shared issues across reactors – When designs and compo-
nents are similar, a problem with one reactor might also be
present in other reactors, and require both to be taken offline
for inspection. [WNA-1]
◆ Cost of extension – The cost of extending a reactor is subject
to uncertainty as it triggers discussion at many levels [DTI-2].
◆ Waste management – The handling and storage of nuclear
waste remains a key consideration [NUF-1].
CHALLENGES IN NON-DOMESTIC OFFSHORE DEVELOPMENT
To unlock the potential of integrating non-domestic offshore
into the Belgian supply mix several challenges still need to
be resolved. These challenges include:
A suboptimal national planning approach
◆ Current offshore wind developments are based on a pro-
ject by project approach, developed on a bilateral basis,
whereas the offshore developments have much more a
multi-lateral dimension. This is caused by a bottom-up
planning of individual projects based on national plans
without considering synergies across different projects
in a sea basin suboptimal national planning approach;
◆ Development of non-domestic offshore projects are also
not appropriately considering offshore grid and off-
shore wind generation together;
◆ As a consequence, there is an absence of incentives for
countries which have an excess potential offshore RES
to develop their potential, and for RES countries with a
lack of potential to access these renewable energy sup-
ply sources.
An innapropriate funding framework
◆ The ambitions linked to harnessing offshore wind in the
North Seas will require very large amount of investment
for which the current amounts of European funds (e.g.
CEF funds) is insufficient and current way of funding
projects individually is ineffective.
◆ The benefits of offshore wind projects of non-domes-
tic offshore projects are often broader than solely the
ones of the hosting countries, who are yet expected to
bear (majority) of the costs. Due to today’s absence of
an appropriate mechanism for cost and benefit sharing
between countries, these projects face difficulties to be
developed.
◆ Lack of clarity on ownership of offshore interconnector
projects when multiple parties are involved in the cost
sharing is an issue which must be clarified and improved
in order to facilitate and speed up the development of
these projects.
◆ Given the barriers and issues listed above, countries
with an excess RES potential may have no incentive to
fully unlock their potential and thus the potential of the
North Sea.
Solving these challenges will require active political follow
up and support to find the right solutions that unlock the
offshore potential fully.
Belgium, as a country with limited offshore RES potential,
has to be one of the leading forces to pro-actively shape and
propose solutions to these different barriers. Otherwise, it
will not get access to this source of non-domestic offshore
wind, should it be chosen as a part of the BE future’s energy
supply mix.
Some solutions start to be sketched out as part of the
European debate on this matter, which are summarised
hereunder:
Joint planning & international coordination
◆ There is a need for an improved regional planning where
TSOs of the sea basin identify the most efficient devel-
opment for the offshore grid. An essential part of the
improvement includes also the decision making around
these projects, where the projects delivering most value,
including those belonging to the non-domestic offshore
category, are jointly selected and promoted at regional
level.
◆ The policy makers around a sea basin could initiate
these developments by providing the right mandate
to their TSOs to initiate a more open and collaborative
approach regarding the development of their offshore
potential and the related offshore grid.
◆ The experience gained in the first voluntary initiative
could then help enshrining the approach in legislative
and regulatory development to ensure its sustainability
towards the future.
Joint funding and cost & benefit sharing
◆ A structure providing joint funding for the offshore gen-
eration and infrastructure projects on the sea basin level
may help to support the joint funding of the most inter-
esting projects selected according to the joint planning
described above.
◆ This funding structure should find improved and inno-
vative ways of combining both public and private fund-
ing in order to finance the development of the offshore
infrastructure or de-risking the offshore generation
◆ The challenge on the funding must be combined with
an improved way of sharing cost and benefits of the off-
shore infrastructure and offshore wind, according to the
identified benefits
An inclusive approach around the North Sea
◆ The offshore wind potential in the North Sea can only
be developed in cooperation with non-EU countries, i.e.
UK and Norway. Active engagement where all parties
jointly seek to solutions for the similar ambitions must
be ensured. Their involvement is not solely limited to
the development of these offshore projects, but also
requires addressing other barriers such as improved
market design to enable efficient cross border trades.
BOX 3-4
95
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios94
3.2.3.6. NUCLEAR FLEET NEW BUILDS
The current study also delves into the exploration of new nuclear energy options. These can encompass large-scale nuclear
reactors such as the European Pressurised Reactor (EPR) or smaller, more compact alternatives known as small modular reac-
tors (SMRs).
Even though there are no definitive plans in Belgium at this stage, various stakeholders – including political circles – have under-
scored the necessity to explore this possibility.
The following assumptions, shaped in consultation with stakeholders, were taken into account:
◆ The earliest, an SMR (based on ‘generation 3’ designs) could be operational in 2036. However, achieving this would necessi-
tate immediate action to identify a site, design and conceptualise the project, build, regulate and start to operate the reactor;
◆ Subsequently, it is assumed that a maximum of 2 GW could be operational by 2040 (either 1.6 GW large-scale reactor or/and
several SMRs);
◆ Similarly, for 2050, a maximum of 8.2 GW is examined, which could consist of a mix of SMRs and large-scale units or solely
SMRs.
However, it’s critical to emphasise that this study exclusively assesses the impact on the energy system. It did not delve into the
feasibility, regulatory aspects, types of designs, locations, and detailed timelines.
New nuclear builds will require a long lead time before becom-
ing operational (see also BOX 3-5). In addition new locations for
reactors will need to be found. The graph below shows the most
ambitious deployment timeline considered in this study, whilst
several intermediate levels of deployment are studied for the
target years.
POTENTIAL NEW NUCLEAR CAPACITIES
INVESTIGATED IN THIS STUDY FIGURE 335
05
30 36 40 45 50
20
50
82
0
BARRIERS ENABLERS
(Shared with existing capacity)
◆ Need for political consensus
(For new builds alone)
◆ Potentially strong NIMBY effect – Resistance from local
residents is likely in cases where large numbers of new sites
have to be secured (especially for small modular reactor
(SMR) deployment).
◆ Development Timeline – In the context of large-scale reac-
tors, multiple instances of exceeding the planned timeline
have been observed in Europe. As for SMRs, this technology
has not yet seen widespread deployment across the conti-
nent. SMRs, technology is not yet deployed in Europe [NEA-1].
◆ Potential 4th generation of SMR - Next generation reactor
designs could further improve performance with respect to
safety and waste handling. However, the timing of develop-
ment is uncertain with many designs available. Most sources
indicate that those will not be commercially ready before
2040s [IAE-1].
◆ Valorising heat directly – SMRs could valorise heat directly
in addition to converting it to electricity, through co-location
with industrial clusters or district heating [EUC-13].
◆ Few resources/space needed – SMR require much fewer
input materials [IEA-5] or space [OWD-1] per reactor com-
pared with other new generation options.
Key decisions
◆ Pursue new nuclear builds or not – The first step is to decide
whether to pursue new nuclear builds at all or not; and if so,
which mix of traditional designs versus SMRs is desirable, and
who will construct/operate and how to finance and what will/
should be the state involvement.
◆ Sites for new development – Deciding on new sites early on
is important to ensure the timely preparation of the grid and
other regulatory aspects.
Risks
(Shared with existing capacity)
◆ Shared issues across reactors
◆ Waste management
(For new builds alone)
◆ Delays in securing permits and locations – Securing per-
mits for new locations is a complex process involving many
stakeholders, including local communities. This is a complex
process, and delays could occur. [UKG-1]
Key decisions
◆ Develop a national strategy to support this `new nuclear' sce-
nario and ensure sufficient and qualified suppliers and staff
(i.e. reinforce education programmes for all nuclear profes-
sionals, support R&D programmes where needed…).
Risks
◆ Cost and time overruns in construction – Europe has very
limited recent development experience, and the two most
recent examples (Flamanville [BLO-1] and Olkiluoto 3 [REU-1])
took 17-18 years to construct, more than 3 times longer than
was initially planned and experienced costs overruns.
◆ Technology risk – For new technologies such as SMRs or next
generation reactors, there will be at most a short-term track
record of existing implementations when an investment
decision is taken, as nearly all designs are still in development
stages today [NEA-1, GIF-1].
BOX 3-5
TIMINGS RELATED TO THE DEVELOPMENT OF NON-DOMESTIC OFFSHORE WIND/ INTERCONNEC-
TORS AND NEW NUCLEAR REACTORS
The building of new nuclear reactors is similar to the build-
ing of offshore wind/ interconnectors: they are large, CAPEX-
heavy projects, with long lead times, and require strong
governmental initiative.
The figure below outlines the electricity supplies that can
be achieved with different combinations of nuclear and/or
non-domestic offshore wind. The diagonal lines represent
‘energy isolines’: different combinations of nuclear and off-
shore wind capacity that yield the same amount of elec-
tricity supply.
WHAT CAN BE ACHIEVED IN TERMS OF ELECTRICITY PRODUCTION FOR NEW NUCLEAR
AND NONDOMESTIC OFFSHORE ? FIGURE 336
New
Nuclear new
capacity
Max +3 GW /
5-year
Max +3 GW /
5-year
Max +4 GW / 5-year
Non-domestic offshore capacity
connected to Belgium
2050
120
TWh
90 TWh75 TWh
60 TWh
45 TWh
30 TWh
15 TWh
8 GW
6 GW
4 GW
2 GW
4 GW 8 GW 12 GW 16 GW
2045
2040
Max 2 GW /
2040
Max 0.5 GW /
2036
• 1 GW nuclear is equivalent
to 2 GW offshore in terms of
energy
• New nuclear and foreign
offshore are large
investments that need
to be planned 10 years
beforehand
• Nuclear extensions (beyond
D4/T3 for 10 years) are also
considered
2036
The maximum attainable deployment for each technology by 2036, 2040, 2045 and 2050 is strongly linked to lead times
and technology maturity.
97
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios96
Bonne photo mais nous ne
l'avons pas reçue en haute def
Lead time for new nuclear reactors
There are primarily two candidate technologies for the development of new nuclear reactors: large-scale reactors such as
EPRs and SMRs.
EPRs are generation III+ pressurised water reactors. They
are large, gigawatt-scale reactors, of which only two units
have been built so far in Europe (Olkiluoto 3 in Finland and
Flamanville 3 in France). The figure below outlines the lead
time for these projects (from initial development decision to
their commissioning) along with the lead times for one that
is currently under construction (Hinkley Point C) and three
more that have been announced by European countries
with existing nuclear reactors. Based on this, a lead time of
15+ years seems reasonable, implying that new EPRs would
not be commissioned before 2040.
SMRs are small modular reactors, an early-stage technology
which promises smaller unit sizes, more flexible operation
and improved costs/production ability through the stand-
ardised production of the same design. Though interest
in SMRs is increasing across the globe, most evolutionary
designs are still in their early stages, and it is difficult to
estimate when they could become operational. Tractebel
has announced that a first SMR (generation III) could be
operational in Belgium in 9-12 years, roughly by 2035, pro-
vided that a decision is taken today to pursue it [KNA-1].
This would hold for an SMR ‘generation III’ reactor, based
on current reactor design but smaller sizes. For ‘generation
IV’ SMRs, there are still many uncertainties regarding which
design would be commercially available to be installed in
Europe and when.
LEAD TIME FROM DECISION TO COMMISSIONING FOR SELECTED RECENT EUROPEAN
REACTORS FIGURE 337
Selected recent European reactors [WIK-1]
(incl. still under construction and announced)
23
5
3
9
5
8 5 (est.)
7 (est.)10 (est.)
5 (est.) 8 (est.)
12
5 13
22
17 (est.)
13 (est.)
20
Decision to
pursue unit
Olkiluoto 3
Hinkley Point C
Flamanville 3
Dukovany (new)
Penly (new)
Initial construction
estimate
Construction
delays
Where we
are in ‘24
Lead time for non-domestic offshore wind and (hybrid) offshore interconnectors
The most recent Belgian reference for hybrid offshore inter-
connectors is the currently investigated TritonLink project.
This project aims to connect the energy hubs of Denmark
and Belgium with a ~1000 kilometre-long HVDC link capable
of transporting 2 GW of power. Exploratory studies started in
2020, and current plans have construction starting in 2026-
2027 and its commissioning happening in 2031-2032 [ELI-8].
This implies a timeline of roughly 10 years (down from 12
years for the first link), with a 5-5 year split between the
preparation and construction of the interconnector. This
study assumes for the most ambitious scenario that 4 GW
of capacity comes online every 5 years from 2036 onwards,
which would require starting 2 new projects in parallel every
5 years from today.
3.2.3.7. OTHER RENEWABLES
In addition to solar and wind, the production capacity from hydro-
electricity and biomass is also considered in this study.
When it comes to hydroelectricity, Belgium's capacity and poten-
tial is limited in terms of run-of-river hydroelectricity. This form of
power generation typically involves the use of small hydro units
installed along rivers, harnessing the natural flow of these rivers
to generate electricity. The largest of these facilities in Belgium
is located on the river Meuse in Wallonia. A capacity of around
150 MW is assumed for all future years.
Regarding biomass and waste (not all renewable) for electricity
generation, the country has currently a bit less than 1,000 MW
installed. This amount is kept constant until 2050.
A potential for biomethane is also accounted for Belgium which
can be used by the model. This amounts to around 21 TWh in 2050
based on the TYNDP2024 figures.
3.2.3.8. THERMAL FLEET
The current thermal generation in Belgium (apart from nuclear)
consists of a significant amount of methane-fired generation
both in large-scale units such as CCGTs or OCGT or in smaller
scale units usually used also a combined heat and power (CHPs).
If those units remain in the system, by 2035 there should be:
◆ around 7,300 MW of large-scale gas units;
◆ around 1,600 MW of small-scale gas units.
These figures include the new units being built in the Liège
region. The age distribution of the large-scale units is provided
in Figure 3-38.
Belgium still has some operational oil-fired units (around 100
MW). Those are assumed to be decommissioned by 2035.
EXPECTED AGE OF THE CURRENT
THERMAL GAS FLEET IN BELGIUM IN
2035 FIGURE 338
2,500
2,000
1,500
1,000
500
0
≤=10 10-20 20-30 30-40 >40
Age of the units in 2035
Installed capacity [MW]
99
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios98
Pumped storage
Pumped storage used to be the main storage technology avail-
able in Belgium. The historical installed capacity is equal to 1,224
MW (Coo 1-6 + Plate Taille 1-4). Considering the ongoing change
in the Coo turbine capacity and reservoir volume, this leads to a
total installed capacity of 1,305 MW of pumped storage in Bel-
gium by the end of 2025, along with a total reservoir volume of
6,300 MWh. No additional volume is considered for the future
given the limited potential of this technology in Belgium. This
does not mean that additional projects could not be developed
in the future.
Large-scale batteries
The future volume is calculated based on the existing volume,
the volume contracted in past Belgian CRM auctions, and an
estimation of future potential based on available information
regarding ongoing projects (similar approach as used in the latest
Adequacy & Flexibility study). For the DE/GA scenarios, 50% of the
future potential is accounted for. This percentage is decreased
to 25% in the LFLEX sensitivity, while the full future potential is
considered for the HFLEX sensitivity. These assumptions lead to
the following volumes in 2050:
◆ 3.4 GW in LOW FLEX sensitivity;
◆ 5.4 GW in both DE and GA as well as in ELEC and SUFF sen-
sitivities; and
◆ 9.4 GW in HIGH FLEX sensitivity.
Small-scale batteries
The volume for the future years is calculated based on projec-
tions of the installation rates and the amount of existing assets.
In the DE scenario, ELEC and SUFF sensitivities, the installation
rate is assumed to be 20,000 units/year until 2035 and 30,000
units/year afterwards. In the GA scenario and LFLEX sensitivity,
it reaches respectively 15,000 and 10,000 units/year from 2029 to
2035 and is kept constant afterwards. Finally, in the HFLEX sen-
sitivity the installation rate is assumed to be higher before 2030
and to increase to 50,000 units/year from 2030 to 2050, with an
increasing average capacity from 4.5 to 9 kW/installation. These
assumptions lead to the following volumes in 2050:
◆ 1.4 GW in LOW FLEX sensitivity;
◆ 1.8 GW in the GA scenario;
◆ 2.5 GW in the DE scenario as well as in the ELEC and SUFF
sensitivities;
◆ - 9.2 GW in the HIGH FLEX sensitivity.
FLEXIBILITY ASSUMED FOR EV AND HP IN THE DIFFERENT SCENARIOS AND SENSITIVITIES FIGURE 340
100%
50%
0%
100%
50%
0%
100%
50%
0%
100%
50%
0%
100%
50%
0%
100%
50%
0%
100%
50%
0%
100%
50%
0%
Electric vehicle Heat pump
LOW
FLEX
GA
DE
ELEC SUFF
HIGH
FLEX
Natural
Natural
Natural
Natural Natural
Natural
Natural
Natural
V1X
V1X
V1X
HP1X
HP1X
HP1X
HP1X
V2X
V2X
V2X
V2X
2025 2030 2036 2040 2050 2025 2030 2036 2040 2050
V1X
3.2.4. DEMAND FLEXIBILITY AND STORAGE
This section details the assumptions regarding storage reservoirs
and demand side response in Belgium.
Two categories are considered regarding large-scale storage:
◆ pumped-storage reservoirs;
◆ large-scale batteries.
Three additional categories are considered regarding end user
flexibility:
◆ small-scale batteries (i.e. home batteries);
◆ flexibility from electric vehicles (either via V1X or V2X);
◆ flexibility from heat pumps.
For these assets to offer their flexibility to the system, several
enablers are needed. These enablers were described in the 2023
Adequacy and Flexibility study [ELI-1], and Elia Group’s most
recent viewpoint, 'The Power of Flex' [ELI-7].
The last category consists of market response (large-scale
demand flexibility) available across Belgium (existing and newly
electrified processes).
The flexibility is associated to each demand scenario and sen-
sitivity (from low to high):
◆ GA scenario: it is assumed that end user flexibility is devel-
oped in line with the trend assumed in the latest Adequacy
and Flexibility study [ELI-1]. This scenario is associated with
the GA demand;
◆ SUFF sensitivity: this sensitivity is derived from the DE sce-
nario, considering that the associated electricity demand is
lower, leading mainly to lower market response potential;
◆ DE scenario and ELEC scenario: it is assumed that in a more
electrified residential consumption, more end user flexibility
will be developed including more residential batteries when
compared to the GA scenario. This is associated to the DE and
ELEC demand;
In addition two sensitivities related to the amount of flexibility
are also defined to test the impact:
◆ HIGH FLEX sensitivity: considers much more flexibility from
electro-mobility, residential batteries and large-scale batter-
ies. For 2050, the installed capacity of flexibility is equivalent
to the electricity peak demand;
◆ LOW FLEX sensitivity: end user flexibility is limited in the res-
idential sector and flexibility is mainly provided by large-scale
storage and market response;
The evolution of flexibility associated with demand flexibility and
storage at the Belgian level is summarised in Figure 3-39 for the
different scenarios/sensitivities (DE, GA, ELEC, SUFF, LFLEX and
HFLEX).
OVERVIEW OF STORAGE AND DEMAND RESPONSE CAPACITIES FOR BELGIUM FIGURE 339
Historical data
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
Capacity [GW]
Heat pumps
Electric vehicles
Large-scale batteries
Residential batteries
Pumped Storage
DSM
HFLEX
LFLEX
ELEC
DE
GA
SUFF
SUFFSUFFSUFF LFLEXLFLEXLFLEX
GAGAGA
ELECELECELEC
DEDEDE FLEXFLEXFLEX
2036 2040 2050
2010 2015 2020
The figures given represent capacities (power), but the flexibility of the system is modeled based on its energy content and
other constraints. Therefore, the values provided are an estimate of the flexible power at a certain point in time, as it fluctuates
depending on the availability of the different components.
101
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios100
Flexibility from electric vehicles
Flexibility from electric mobility consists of optimised charging
(V1X) and vehicle-to-grid V2X.
V1X assumes that electric vehicles are combined with unidirec-
tional smart charging technology (without the ability to inject
electricity back into the network) to shift charging to periods with
high RES infeed. The category is split between smart charging
(V1M) and delayed charging (V1H). Further information about the
modelling choice can be found in the latest Adequacy & Flexibility
study performed by Elia.
V2X assumes that electric vehicles are combined with bidirec-
tional smart charging technology to shift their charging away
from periods with higher residual load but also to use the spare
battery capacity to store energy and inject it back to the grid. This
type of charging behaviour is modelled as an additional battery
device that is combined with other battery types. The category is
split between in the market (V2M) or out-of-market (V2H).
The capacity associated with this category depends on the
amount of electric vehicles available and the share of flexibility
associated with natural charging, V1X and V2X. The different flex-
ibility shares by scenario/sensitivity are presented in Figure 3-40.
It should be noted that those capacities are indicative because
the model has a different flexibility available each hour of the day
depending on the amount of cars connected to a charger and
the state of charge of the batteries.
The following volumes are considered for 2050:
◆ 4.7 GW V1X and 1.3 GW V2X for the LOW FLEX sensitivity;
◆ 4.9 GW V1X and 2.7 GW V2X for the GA scenario;
◆ 4.3 GW V1X and 4 GW V2X for the DE scenario and the ELEC
and SUFF sensitivities;
◆ 3 GW V1X and 6.6 GW V2X in the HIGH FLEX sensitivity.
Flexibility from heat pumps
Flexibility from heat pumps across Belgium consists of a mix of
pre-defined (HP0), pre-heated profiles combined with smart
heating (HP1H), where the heat pumps are optimally dispatched
by the model following energy and power constraints (HP1M),
as defined in the latest Adequacy & Flexibility study performed
by Elia. In this study, the HP1H and HP1M are integrated in an
equivalent HP1X category, assuming that the first category would
provide less flexibility than the second one. The associated cat-
egory varies on an annual basis, according to the seasonality of
the heat demand and on an intraday basis, following the daily
heat demand and operating mode of the heat pumps. The val-
ues provided here are indicative for an average winter day. The
model has a different flexible capacity available for each climate
year and day.
The amount of flexibility in GW depends on the amount of heat
pumps considered in the scenario/sensitivity and on the dif-
ferent flexibility shares by scenario/sensitivity are presented in
Figure 3 -40. This leads to the following volumes in 2050:
◆ 0.2 GW in LOW FLEX sensitivity;
◆ 0.5 GW in SUFF sensitivity;
◆ 0.6 GW in the GA scenario;
◆ 0.7 GW in the DE scenario;
◆ 0.9 GW in the ELEC and HIGH FLEX sensitivities.
Market response (including new electrified large consumers)
The volume of market response is the sum of existing usages
(1843 MW, as evaluated in the 2023 Adequacy and Flexibility study)
and newly electrified industrial processes (industrial heat pumps,
e-boilers and direct reduced iron electric arc furnaces) and data
centres. The fraction of flexible demand in the different scenarios
and sensitivities is presented in Table 3-5.
The different assumptions lead to the following volumes in 2050:
◆ 3 GW in LOW FLEX sensitivity;
◆ 4.1 GW in SUFF sensitivity;
◆ 4.3 GW in the DE/GA scenario; and
◆ 5.1 GW in the ELEC and HIGH FLEX sensitivities.
FRACTION OF NEWLY ELECTRIFIED INDUSTRIAL DEMAND CONSIDERED FLEXIBLE UNDER
DIFFERENT SCENARIOS SENSITIVITIES TABLE 35
Demand Type
LFLEX
SUFF
DE
ELEC
GA
HFLEX
P2H – e-boilers 80% 100% 100%
P2H – Heat Pumps 20% 80% 90%
DRI-EAF (Steel) 25% 75% 100%
Data centres 0% 50% 75%
CCS 0% 0% 50%
Electrolysers 100% 100% 100%
Other 0% 0% 0%
3.2.5. BELGIAN ELECTRICITY GRID
1 HTLS (high-temperature low-sag) conductors are a type of overhead power line conductor that can operate at higher temperatures than traditional conductors. This allows
them to carry more current (higher power transfer) without sagging excessively or losing mechanical strength. They are typically used in power transmission lines to increase
their capacity without needing to significantly change the physical infrastructure.
The Belgian electricity grid was analysed across three levels for
the current study:
◆ the horizontal grid, consisting of the backbone, interconnec-
tors and the offshore grid;
◆ the vertical grid which takes the energy from the first level to
the distribution level; and
◆ the distribution level.
This section provides the main assumptions regarding the Bel-
gian starting grid.
Backbone and offshore grids
The 380 kV grid constitutes Belgium’s backbone transmission
grid. Very large consumers, large electricity producers and inter-
connectors are connected to it. The starting grid used for this
voltage level consists of all approved projects in the latest federal
network development plan which was published by Elia [ELI-3].
An overview of the major projects is included in Figure 3-41; it
covers projects such as the Boucle du Hainaut, Ventilus, the Prin
-
cess Elisabeth Island, Lonny-Achène-Gramme and the backbone
with high-temperature low-sag (HTLS1) conductors. In terms of
offshore interconnectors, it includes the Nautilus interconnector
with the UK. It should be noted that the costs presented in this
study are in addition to those associated with approved projects.
REFERENCE BELGIAN HIGH VOLTAGE HORIZONTAL ELECTRICITY GRID FIGURE 341
Substation 380 kV
Axis already reinforced / without potential for
further reinforcement
Axis with remaining potential for further
reinforcement
In the model, Belgium is split into 3 electrical zones: the western,
central and eastern parts of the country. This allows the optimiser
to be able to choose from different projects such as reinforcing
the backbone or new (offshore or onshore) interconnectors with
other countries. An additional optimisation is performed after
the simulation to evaluate the need for internal backbone rein-
forcements based on the flows and overloads of the internal grid
modelled between the three zones.
103
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios102
Regarding onshore interconnectors the optimiser can choose to
reinforce the existing AC grid by means of HTLS reinforcements.
This is the case for the Van Eyck (BE) – Maasbracht (NL) existing
link. In addition, the other potential reinforcements are consid-
ered to be HVDC cables.
It should be highlighted that, except the HTLS possible reinfore-
cements, the expenses related to the reinforcement of the grid
are presumed to be equivalent to the costs of HVDC connections.
For those connections, no limit is imposed in terms of capacity.
Vertical grid
The vertical grid consists of the lower voltages of Elia’s grid (which
ranges from 150kV to 30kV). This grid is used to bring the bulk
power from the backbone to the DSO grids or to certain directly
connected grid users. It is also used by certain generation facili-
ties, storage facilities and renewable sources.
The costs presented in this study are an estimation of the costs
from 2030 onwards. Starting from the approved investment
plans, the study estimates the increase in total peak load per
scenario whilst taking into account that local generation will
reduce the capacity need at TSO level. This is then compared to
the investments based on the peak load hypotheses used for the
network investment plans, translated into an additional need for
transformers, cables & substations and finally yields a high-level,
updated investment need.
Distribution grids
Distribution grids bring the electricity from Elia’s grid to end
consumers. While those are not modelled explicitly, the costs
for the needed reinforcements are estimated, depending on the
chosen scenario.
The costs presented in this study are an estimation of the costs
from 2030 onwards. Starting from the DSO investment plans,
these investments are then rescaled by scenario based on the
expected increase in peak consumption for the load segments
that are expected to be connected at distribution level.
Peak consumption during winter evenings is expected to remain
the peak event that drives network investments, given the spread
of heat pumps and electric vehicles across the low-voltage net-
works. This study assumes that the resulting network reinforce-
ments can also accommodate the growth in PV, except for the
most ambitious scenario concerning PV deployment, where the
PV injection is capped at the level of the network reinforcements.
This will be further explored in the results section.
SCHEMATIC REPRESENTATION OF THE DIFFERENT PARTS
OF THE BELGIAN TRANSMISSION & DISTRIBUTION SYSTEMS FIGURE 342
Centralised Biggest
industrials
Decentralised
Decentralised/
Residential
Big industrials
Small industrials
& residential
Interconnectors
(onshore &
offshore)
Production Load Storage Import/export
380 kV / 220 kV
Horizontal
Grid
Voltage levels
70 kV / 110 kV / 150 kV
Vertical
Grid
MV & LV
30 kV / 36 kV
Distribution
Grid
Biggest
Small
Big
3.3. FINANCIAL ASSUMPTIONS
Financial assumptions play a vital role in determining the optimal
capacities in the model and quantifying the system's total costs.
The model incorporates both variable and fixed costs for the
technologies and infrastructure it utilises and invests in.
In addition to the costs related to the energy system and as
explained in Section 2.4, the costs also include end uses and
all energy vectors. This section offers a snapshot of the key cost
parameters used in this study. For the variable costs (other than
fuel costs), we refer to the most recent Adequacy and Flexibility
study 2023. The cost estimates were initially presented in the
second workshop dedicated to the study, prepared by Compass
Lexecon. These estimates were later adjusted based on feedback
received from various stakeholders during the consultation phase.
The range of costs is intended to capture the known uncertainty
associated with the different technologies. Appendix F provides
more detailed information about the total cost methodology.
3.3.1. APPROACH TO TOTAL COST QUANTIFICATION
This study does not seek to differentiate between various types of
investors or evaluate the methods and time frames for financing
the different technologies. Therefore the technical lifetimes used
for each technology are different from the economical lifetimes
used to evaluate the business case of a certain asset. Similarly,
transfers between consumers and producers (e.g. subsidies) or
welfare for different types of users (consumers, producers) are
excluded from this analysis. The goal of the study is to quantify
the total costs of the Belgian/European system; the base for cost
quantification lies therefore in assessing the fixed and variable
costs required to supply the needed energy.
A few assumptions need to be kept in mind when interpreting
the results:
◆ all cost assumptions and figures are reported in euros (€)
2022 (unless stated otherwise);
◆ all CAPEX figures are expressed in overnight costs (without
financing costs); the financing costs are added afterwards
when applying the cost of capital and accounting for the con-
struction time;
◆ emissions costs are based on the shadow carbon cost calcu-
lated by the model (see also Appendix C);
◆ import/export costs are assumed to be priced at the system
marginal cost for every time step of each energy carrier.
◆ when reporting on the Belgian electricity system costs:
- existing technologies and new installations before 2030
are assumed to be fully depreciated (no CAPEX assumed)
but there is a fixed operating cost (FOM) applied to these
technologies;
- the same holds for the grid in instances where the costs for
the distribution and vertical grids are calculated from 2030
onwards; costs for the backbone and interconnectors are
calculated on top of the approved projects from the latest
federal network development plan. This is further explored
in this section;
-
no replacement CAPEX is assumed for existing technologies
other than thermal generation for electricity; this assump-
tion does not change the comparison between scenarios
as the amount of existing capacities is assumed to be the
same in all the scenarios and sensitivities (except for thermal
generation);
◆ For the quantification of end uses and other vectors costs,
a similar approach is used:
-
For OPEX and CAPEX costs for the molecule system, the
same is applied as for the electricity costs. The OPEX costs are
mainly the imports of molecules needed for Belgium/Europe
while the CAPEX includes the infrastructure requirements;
-
Concerning end uses (industry investments, buildings
and transport), the reader can refer to Appendix F where
more details are provided regarding the methodology. This
includes quantifying the investments that the consumers are
required to make in each scenario (e.g. renovation, purchase
of a car, changing the heating system, etc... ).
Two ways will be used to report on costs:
Annualised approach
The first approach involves selecting a specific target year and
adding up the annual payments and the operational costs of the
system for that year.
◆ The annual payments (or annuities) are derived from past
investments that were needed to set up the required infra-
structure and technologies for a particular scenario. These
annual payments are then annuitised based on a specific
asset lifetime and the Weighted Average Cost of Capital
(WACC).
◆ The yearly operational costs of the system are added to
the annuities as those reflect the yearly variable costs of the
system. Those include the purchase of the fuel of locally pro-
duced power and cost of imports (cost of fuels produced
abroad and imported). In the case of exports, this is then a
revenue.
The annual payments and operational costs can either be com-
bined and presented as absolute figures or expressed relative to
the energy or electricity demand. In the latter case, they would be
reported in terms of €/MWh. This method will be used to compare
the different scenarios between each other.
Sum of overnight CAPEX approach
To determine when investments should be made (for instance,
during which 5-year period), one can also examine the investment
costs required by each scenario. However, this method will only
be used to asses when capital expenditures (CAPEX) would need
to be made for a specific scenario, not for comparing various
scenarios, as it doesn't consider operational expenditures (OPEX).
Additionally, this approach does not offer insight into the cost
of capital or the operational costs, so it should not be used to
compare different scenarios with each other.
105
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios104
3.3.2. INVESTMENT COSTS
Investments costs in different technologies were consulted upon
and discussed with stakeholders. The aim of these cost figures
is to assess the total system costs. It is important to note that a
lot of uncertainty regarding investments costs still resides until
2050. The low and high ranges aim to reflect the known uncer-
tainty; however, new breakthrough technologies or innovations
could change them. This is also the reason why many cases are
simulated and costs quantified with different combinations for
the Belgian sensitivities assessment.
The table in Table 3-6 the overnight CAPEX costs in €/kW. This
means that the figures exclude any financing (e.g. during con-
struction) and those should be added.
In addition to the CAPEX, technical lifetimes are used for all tech-
nologies (and not economic lifetimes) as the study aims to assess
the system costs. Construction times for each technology are
also accounted for.
OVERVIEW OF THE OVERNIGHT CAPEX FOR KEY TECHNOLOGIES TABLE 36
Costs in EUR2022/kW unless stated
otherwise Low Ref High
Overnight CAPEX 2030 2040 2050 2030 2040 2050 2030 2040 2050 Comments
Existing
thermal
CCGT-extension 100 100 100 120 120 120 150 150 150
Very heterogenous
OCGT-extension 80 80 80 100 100 100 120 120 120
CCGT-CCS-refurb 1,250 1,130 1,050 1,280 1,130 1,050 1,320 1,370 1,420
OCGT-CCS-refurb 610 560 470 770 790 530 980 1,030 1,030
CCGT-hydrogen refurb 170 150 150 210 200 200 250 200 200
OCGT-hydrogen refurb 100 100 100 280 280 280 350 330 330
Nuclear-extension-10Y 800 800 800 1,000 1,000 1,000 1,200 1,200 1,200 Ref based on Plan Bureau costs for
the DC2024
New
storage
Battery-2h 350 240 240 530 450 440 720 670 640
Battery-4h 600 500 400 1,000 800 700 1,300 1,200 1,000
New DRES
Wind onshore 1,020 840 770 1,280 1,110 1,030 1,530 1,370 1,300
PV-residential 490 480 360 950 622 500 1200 900 700
PV-utility 390 340 290 640 570 460 900 800 630
New
offshore
Wind offshore fixed 1,500 1,150 810 2,200 1,950 1,600 3,370 2,300 2,180 Includes array cable but excludes
electric plaftorm and cables to shore
Wind offshore floating 2,400 1,950 1,490 2,910 2,720 2,640 5,020 3,100 2,870
New
thermal
CCGT-methane 740 720 700 900 870 850 1050 1000 950
Costs for large units
OCGT-methane 550 500 500 720 700 700 900 850 850
CCGT-hydrogen 930 900 880 1,130 1,090 1,060 1,310 1,250 1,190
OCGT-hydrogen 600 560 560 900 880 880 1,130 1,060 1,060
CCGT-CCS 1,900 1,750 1,750 2,060 1,880 1,870 2,220 2,220 2,220
OCGT-CCS 1,010 930 850 1,390 1,390 1,130 1,760 1,760 1,760
New
nuclear
Nuclear-EPR 6,500 6,500 6,500 7,500 7,500 7,500 10,000 10,000 10,000 Very heterogenous costs. Ref based
on new EPR2 estimates for France.
Nuclear-SMR 7,500 7,000 6,000 9,000 8,500 7,500 12,000 10,000 10,000 Very heterogenous designs and
estimates available
Electrolysis
Elecrolyser-onshore 1,330 1,050 910 1,600 1,260 1,090 1,990 1,580 1,360
Electrolyser-offshore 2,330 2,050 1,910 2,600 2,260 2,090 2,990 2,580 2,360
Converters
(AC/DC)
Onshore 250 250 250 260 260 260 325 325 325
Offshore 550 550 550 590 590 590 700 700 700 Includes the offshore platform re-
quired to hold the converter
HVDC
Onshore 1,680 1,680 1,680 2,210 2,210 2,210 3,800 3,800 3,800
Costs in EUR2022/MW/km
Offshore 1,680 1,680 1,680 2,010 2,010 2,010 3,000 3,000 3,000
3.3.3. COST OF CAPITAL
The overnight costs are accounted for with a certain weighted
average cost of capital (WACC). Three cases are applied for supply
technologies in this study:
◆ Reference: WACC of 7%;
◆ High: WACC of 10%;
◆ Low: WACC of 4%.
The same WACC is applied across all technologies; however, dif-
ferent risk profiles subsist. In order to reflect the differences in
risks, when quantifying the costs, different WACC sensitivities will
be applied depending on the technology.
In addition to the WACC, a cost of debt of 4% is applied during
construction time to reflect the cost of capital incurred during
that period. The longer the construction time, the bigger the
costs incurred.
For grid technologies, a WACC of 6% is applied without sensitiv-
ities on that parameter.
3.3.4. TREATMENT OF NON-DOMESTIC OFFSHORE IN THE COSTS
The grid costs that are accounted for in each scenario/sensitiv-
ity when calculating the total costs of the electricity system for
Belgium include:
◆ All the costs related to the distribution and vertical grids;
◆ All the costs related to the internal backbone grid in Belgium;
◆ All the costs related to the links between Belgium and the
non-domestic offshore wind accounted for in the sensitivities.
Those also include any multi-terminal platforms/island in Bel-
gium
◆ All the costs related to the non-domestic offshore wind farm
up to the capacity considered in the sensitivity/scenario;
◆ Half of the costs for interconnectors from/to Belgium.
This is further summarised in Figure 3-43 for the high-voltage grid.
NONDOMESTIC OFFSHORE COSTS FIGURE 343
Capacity of the links to Belgium
is at least the installed capacity
of the connected wind farm to
ensure the electrons can flow to
Belgium
[x%]: how much of the
costs is considered in
the total system costs
for Belgium
Not considered in the costs
Connection fully paid by Belgium
Non-domestic offshore wind capacity paid by Belgium
Half of the connection paid by Belgium
0%
0%
0%
100%
100%
100%
100%
100%
100%
50%
50%
3.3.5. OTHER COSTS COMPONENTS
Fixed costs (FOM): Each technology is associated with a certain
FOM yearly cost. This is applied to existing and new investments
including infrastructure.
End use costs:
◆ Transport costs: The costs accounted for are those for road
transport only (new vehicles and charging infrastructure).
◆ Building costs: costs for renovation and heating systems are
accounted for. Other costs such as appliance replacement,
cooling are not accounted for.
◆ Industry costs: the cost of replacement of certain industrial
processes by different technologies and and energy carriers
used, cost for carbon capture and storage are also included
here.
Emissions costs: Emissions costs are calculated based on the
shadow CO
2
price calculated by the model. For the Belgian supply
sensitivities, the same CO
2
price is used across all scenarios within
the parent European scenario.
Imports and exports of energy: The exchanges from/to a certain
zone (e.g. Belgium) are also accounted for. Indeed, the molecules
or electricity that needs to be imported (or that is exported) is
priced at the marginal cost of the corresponding energy carrier.
Transfers and subsidies: Fiscal costs, such as taxes, subsidies,
levies, and redistributions are excluded from the system costs.
107
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Scenarios106
3.3.6. COSTS OF IMPORTS OUTSIDE OF EUROPE & DOMESTIC FUELS
Generally, prices from the TYNDP 2024 are used where available.
However, for ammonia imports, the TYNDP assumes a singular
import price. To accurately represent the merit order of imports
and the cost variations when importing from different continents,
a new cost calculation is undertaken. The methodology used to
recalculate the ammonia import cost per continent is depicted
in Figure 3-44.
METHODOLOGY TO CALCULATE IMPORTS COSTS OF MOLECULES OUTSIDE OF EUROPE FIGURE 344
INPUT DATA OUTPUT
Solar profiles of the considered
country
Wind profiles of the considered
country
CAPEX and OPEX of/
- Electrolysers
- Wind turbines
- PV
-Batteries
WACC, lifetime, efficiencies, ...
Installed solar capacity
Installed wind capacity
Installed battery capacity
Levelised cost of hydrogen (LCOH)
[€/MWh]
Optimise the amount of solar,
wind and batteries to install per
MW of electrolyser
Calcute the amount of running
hours of the electrolyser
For each potential export country, solar and wind profiles are ana-
lysed to determine the optimal solar, wind and battery capacity
for a 1 MW electrolyser. These profiles facilitate the calculation of
the corresponding full load running hours for the electrolyser.
These outputs, combined with the CAPEX and OPEX of all tech-
nologies, a corresponding WACC and technical lifetime, enable
the calculation of the levelised cost of hydrogen (LCOH). This
cost is then increased due to the conversion cost of hydrogen to
ammonia and a transportation cost which is proportional to the
distance from the exporting country to Europe.
As a means of validating the derived prices, a comparison was
conducted with other studies. Figure 3-45 illustrates the compar-
ison for ammonia import prices via shipping. It is important to
note that most studies present import prices based on hydrogen
as the end product, while in this case, prices are calculated with
ammonia as the final product. The three scenarios used from
Low to High are in the range of the other studies.
A similar approach is applied for synthetic liquids and synthetic
methane. The primary difference lies in the conversion costs from
hydrogen to synthetic liquids or synthetic methane (which are
higher, primarily due to the increased operating expenses of these
installations). Additionally, for these installations, an extra cost is
added for direct air capture (DAC) systems, as these conversions
necessitate their use.
More detailed merit order curves are provided in Section 3.1.4.
COMPARISON OF AMMONIA IMPORT PRICES WITH OTHER STUDIES’ VALUES FIGURE 345
Elia ESBP2050 - Central
Elia ESBP2050 - High
Clean Hydrogen Partnersships [1]
ENTSOS [2]
Hydrogen Council [3]
Hydrogen Import Coalition [4]
HydrogenInsights [5]
Aurora [6]
Universität Erlangen-Nürnberg [7]
Oxford Energy [8]
Ammonia import prices
2030
2050
250
200
150
100
50
0
€/MWh_NH3
Elia ESBP2050 - Low
Elia ESBP2050 - Central
Elia ESBP2050 - High
Clean Hydrogen Partnersships [1]
ENTSOS [2]
Hydrogen Council [3]
Hydrogen Import Coalition [4]
HydrogenInsights [5]
IRENA [9]
Agora energiewende [10]
Fraunhofer [11)
Entso-G [12]
250
200
150
100
50
0
€/MWh_NH3
Elia ESBP2050 - Low
Range
Range
Average
Average
[1] Study on hydrogen in ports and industrial coastal areas [EUH-3]
[2] TYNDP24 scenarios [ENT-1]
[3] Global Hydrogen Flows 2022 & 2023 [H2C-1]
[4] Shipping sun and wind to Belgium is key in climate neutral economy [H2I-1]
[5] Green hydrogen made in Germany will be cheaper than shipped imports in 2030 [H2I-1]
[6] Renewable hydrogen imports could compete with EU production by 2030 [AUR-1]
[7] The economics of global green ammonia trade [UEN-1]
[8] Renewable hydrogen import routes into the EU [OXF-1]
[9] Global hydrogen trade [IRE-4]
[10] Costs and risks of importing hydrogen derivatives by ship [AGO-1]
[11] Green Ammonia for climate protection [FRA-1]
[12] Learnbook: hydrogen imports to the EU market [ENT-6]
109
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results108
4.
MULTI-ENERGY
EUROPEAN
RESULTS
4.1. Supply and demand per vector 110
4.2. Supply and demand of electricity 119
4.3. Interactions between energy vectors 126
4.4. Management of emissions 129
4.5. European electricity grid 133
4.6. Adequacy and flexibility 139
4.7. System costs across the different scenarios 143
4.8. Key takeaways 146
This chapter provides a summary of the main findings acquired for Europe. Although
the study's primary focus is Belgium’s electricity system, examining the multi-energy
results for the whole of Europe is also crucial.
The results acquired are tied to the assumptions made. Different
assumptions could yield different results. Therefore, the analysis
also includes various scenario variations to enable 'what if' explora-
tions. The assumptions underlying these variations are discussed
in the previous chapter.
The following aspects are analysed:
◆ the supply and demand for each energy vector;
◆ the supply and demand for electricity;
◆ interactions between energy carriers and those between
the electricity system and the other carriers;
◆ carbon management (amount of carbon emitted, stored,
re-used);
◆ electricity grid requirements at the European level across
different scenarios with a focus on offshore development;
◆ adequacy and flexibility requirements;
◆ the costs of the different scenarios.
The chapter then ends with the key takeaways for the electricity
system and for Belgium.
As a reminder, the area investigated by this study comprises all 27 Member States along with Nor-
way, the United Kingdom and Switzerland. References to Europe throughout this study therefore
cover these 30 states. By contrast, specific references to the European Union cover its 27 Member
States only.
111
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results110
4.1. SUPPLY AND DEMAND PER VECTOR
This section outlines the results for the multi-energy supply and
demand balances for each energy vector at a European level.
This covers the final usages per energy vector, but also explores
the usage of these vectors for the generation of power and the
potential synthetisation of derivative fuels. These results are dis-
cussed for the main European demand scenarios (DE, GA and
ELEC) alongside additional sensitivities (where relevant) that
illustrate the impact of exogenous changes that could influence
these balances. Note that positive values are reported as supply
and negative values are reported as demand values; this allows
readers to understand how each energy carrier is supplied and
how they are used at a European level.
4.1.1. METHANE BALANCES
Figure 4-1 depicts the yearly balances for methane. In 2021, meth-
ane was used for a variety of appliances, although its main use was
as a fuel in industry, heating in buildings and as non-energetic
feedstock for the production of hydrogen (and its derivatives) via
SMR-H2. Around 1,500 TWh was consumed for the production of
electricity.
In all simulated scenarios, the demand for methane is seen to
decrease compared with today. This is explained by the key evo-
lutions outlined below:
◆ The decrease in the demand for methane for final end uses
(i.e. usage of methane in buildings, industry and transport)
can mainly be explained by the assumed phase-out of gas
usages such as heating in buildings and process heat in
industry. This effect is stronger in the ELEC & DE scenarios,
which have a stronger focus on electrification than in the
GA scenario. Note that the ELEC scenario assumes that the
demand for methane in end uses is the same as in the DE
scenario.
◆ On the other hand, the need for gas for power generation is
strongly reduced from around 1,500 TWh in 2021 to around
300-750 TWh in 2050. This is mainly explained by the assumed
buildout of RES, even though electricity demand increases in
all scenarios. This will be further explored in Section 4.6.
◆ SMR-H2 (with CCS) for the creation of hydrogen still occupies
a (relatively small) share in 2036, but disappears completely
from 2040 as domestic electrolysis and H2 imports become
more economical.
The trends which emerge on the supply side are outlined below.
◆ The full amount of assumed domestic biomethane poten-
tial is used by 2040 in both scenarios and is able to almost
fully meet the European demand for methane by 2050. How-
ever, this implies increasing the amount from around 40
TWh in 2021 to almost 1,100 TWh by 2050, which could prove
to be challenging. As such, imports of (fossil) methane are
strongly reduced and non-existent in 2050. Note that the
small remaining fossil methane can be supplied domestically
within Europe, mainly by Norway.
◆ The domestic production and import of synthetic methane
(not reported in the figure given it is zero for all scenarios and
horizons) proves to not be economically viable when com-
pared to biomethane and local or imported fossil gas (even
when the cost of CO2 is high).
ANNUAL SUPPLYDEMAND BALANCE FOR METHANE FIGURE 41
2,5602,560 2,5602,560
2,0602,060 1,9101,910 2,1702,170
1,4101,410 1,2001,200 1,3401,340
2036 2040 20502021
Data for Europe (incl. UK, NO, CH).
* Note that NO and UK supplied around 900TWh to the EU-27 in 2021
Historical values based on EUROSTAT and the Hydrogen observatory of the European Commission
[EUH-1]
DE
Domestic fossil*
Domestic bio
Import fossilImport fossil
End use
SMR-H
2
Power
Supply Demand
ELEC
GA DE
ELEC
GA DE
ELEC
GA
+
5,000
4,000
3,000
2,000
1,000
0
-1,000
-2,000
-3,000
-4,000
-5,000
[TWh]
4,7504,750
2,6802,680
4.1.2. HYDROGEN BALANCES
Although no reporting standards exist for the consumption of
hydrogen, it is estimated that in 2022, around 260 TWh of hydro-
gen were consumed/produced in Europe [EUH-3], with the main
source of demand being the production of ammonia for fertilizers
and for the desulfurisation process in refineries. The production
of hydrogen generates a high amount of emissions since 99% of
the time it is made from fossil fuels, mainly via steam methane
reforming (SMR-H2), but also (to a lesser extent) as a by-product
in some industrial processes.
The following changes can be observed on the demand side.
◆ A general increase in the use of hydrogen as an energy fuel,
mainly in industry and road transport (especially for the GA
scenario). The level of the increase varies across scenarios. As
a reminder, the ELEC scenario does not assume any hydrogen
usage in buildings and road transport.
◆ Generally speaking, the higher the electricity demand and
the lower the hydrogen demand are for end uses, the more
hydrogen is used for power generation. This is linked to the
fact that more thermal generation is then required in the
power system to remain adequate. This could change if more
flexibility (storage, demand response) is assumed in the more
electrified scenarios but also if less hydrogen-fired power
generation is assumed.
◆ Almost no synthetic liquids are produced in Europe before
2040 in any of the scenarios. This is mainly explained by the
fact that other parts of the energy system can decarbonise
and reach their emissions targets more cheaply without the
need to decarbonise their consumption of liquids.
In terms of the projected supply, the following changes are
apparent.
◆ Domestic hydrogen production via electrolysis would be able
to deliver around 50% of the assumed hydrogen demand
by 2036 (this implies around 70-120 GW electrolysers to be
installed). This amounts to around 6 Mt-14 Mt, which is close to
the ‘RepowerEU’ target of 10 Mt for 2030 for the EU-27 [EUC-
14].
◆ However, its production stagnates in later years. The GA sce-
nario involves more electrolysis which is driven by the higher
assumed end use demand for hydrogen but also lower end
use demand for electricity. See Figure 4-24 for the localisation
and installed capacities of electrolysers across Europe.
◆ As prices of piped hydrogen and imported ammonia are
assumed to decrease over time, these are favoured over SMR-
H2 and electrolysis to meet the additional demand from 2040
onwards.
ANNUAL SUPPLYDEMAND BALANCE FOR HYDROGEN FIGURE 42
2036 2040 20502022*
Data for Europe (incl. UK, NO, CH).
* Historical values based on the Hydrogen Observatory of the European Commission [EUH-1]
** The historical values for ‘reforming’ do not distinguish between methane or naptha-based and are here grouped together as ‘SMR H
2
’
DE
Supply Demand
ELEC
GA DE
ELEC
GA DE
ELEC
GA
Electrolyser
SMR-H
2
**
End use
Synth. liquids
Power
Ammonia
+
3,000
2,000
1,000
0
-1,000
-2,000
-3,000
[TWh]
Import pipelineImport pipeline
Import ammoniaImport ammonia
260 410
750
1,020
1,060
1,610
1,800
1,900
2,220
2,680
113
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results112
4.1.3. LIQUIDS
Liquids make up most of the energy and non-energy demand in
2021. This holds true for 96% of refined oil products such as diesel,
gasoline, fuel oil, kerosene, bunkering fuels, naphtha, etc. Today,
most of the demand is met by imports from outside of Europe,
or non-EU countries such as Norway and the UK (included as
‘domestic fossil’ in Figure 4-3).
On the demand side, the following changes are apparent:
◆ In the lead-up to 2050, the demand for liquids in road trans-
port and heating (‘other energy’ in the figure) is assumed to
almost entirely disappear, with small levels of consumption
of these remaining in industry. This is mainly assumed to be
driven by the replacement of heating devices and IC engines
by heat pumps and EVs and, to a lesser extent, the use of
direct hydrogen (particularly in the GA scenario).
◆ The demand for liquids for international aviation, shipping
and feedstock remains relatively stable in the lead-up to 2050,
although their supply shifts from oil products to bio and syn-
thetic fuels (see supply side).
On the supply side, the following changes are becoming visible:
◆ Oil products are gradually replaced by bio-liquids; from 2040
onwards, they are replaced by domestic and imported syn-
thetic liquids such as methanol.
◆ Some oil continues to supply areas in 2050, especially aviation
(kerosene), shipping (bunkering fuels) and (to a lesser extent)
feedstock. This will need to be compensated by CCS in other
sectors and/or direct air capture (DAC).
ANNUAL SUPPLYDEMAND BALANCE FOR LIQUIDS FIGURE 43
Aviation
Other energy*
Shipping
Feedstock
Power
+
7,500
6,000
4,500
3,000
1,500
0
-1,500
-3,000
-4,500
-6,000
-7,500
[TWh]
2036 2040 20502021
Data for Europe (incl. UK, NO, CH).
All values are expressed in refined energy products (defined in EUROSTAT as final energy demand)
* Includes the industry, residential, tertiary and domestic transport sectors
** Note that most domestic oil production in 2021 came from NO (980 TWh) and the UK (230 TWh)
Historical values based on EUROSTAT
DE
ELEC
GA DE
ELEC
GA DE
ELEC
GA
Supply Demand
Domestic fossil**
Domestic bio
Domestic synthetic
Import syntheticImport synthetic
Import fossilImport fossil
6,930
3,810 3,810 3,900
2,920 2,920 3,010
1,830 1,830 2,010
4.1.4. IMPORTS FROM OUTSIDE OF EUROPE
Bringing together the energy balances per energy vector pro-
vides an overview of the imports per scenario and target year. In
2021, around 50% of Europe’s primary energy needs were met by
imports, of which most are oil products used in industry, trans-
port and heating. In all scenarios, Europe can reduce its energy
imports drastically both in absolute and relative terms. This is
mainly explained by the following factors.
◆ A reduction in the final energy demand. As explained in
Section 3.1.2.1, Europe is assumed to reduce its final energy
demand by 38%-43% by 2050, which is mainly driven by
energy efficiency measures and electrification.
◆ The electrification of end use does not only reduce its final
energy needs but shifts demand away from liquids and gas-
eous molecules that are mostly imported both today and
across the simulated years. In general, as the level of elec-
trification increases (GA, DE, ELEC respectively) and the level
of renewable production increases across Europe (see next
point), so the requirement for imports decreases.
◆ An increase in the amount of RES in all scenarios is assumed.
The increased production of green electricity reduces the
need for thermal power plants running on coal (which are
assumed to be almost phased out by 2036), methane or
hydrogen (in later years).
◆ The local production of (green) molecules lowers the need
for imported molecules. For example, the assumed increased
potential of biomethane allows the levels of imported meth-
ane to be reduced. The production of (green) hydrogen and
synthetic fuels lowers the need to import (mostly fossil) meth-
ane and oil.
In general, Europe would still need to import around 17% to 22% of
its primary energy needs (compared to 50% today), depending on
the scenario. However, these imports would no longer arrive in the
form of fossil fuels such as coal, oil and fossil gas but would shift
towards green molecules such as piped hydrogen, green ammo-
nia and (to a lesser extent) synthetic liquids such as methanol.
YEARLY ENERGY IMPORTS FROM OUTSIDE EUROPE FIGURE 44
+
10,000
8,000
6,000
4,000
2,000
0
[TWh]
Hydrogen
Liquid synfuels
Ammonia
Coal
Oil
Fossil gas
Import dependence*
2036 2040 20502021
Data for Europe (incl. UK, NO, CH).
Includes international shipping & aviation
* Import dependence expressed as percentage of primary energy demand
Historical values based on EUROSTAT
DE
ELEC
GA DE
ELEC
GA DE
ELEC
GA
51%
9,610
30%
3,900
33%
4,450
31%
4,380
23%
2,820
23%
3,000
27%
3,610 17%
1,970
18%
2,250
22%
2,740
115
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results114
4.1.5. TOTAL PRIMARY ENERGY SUPPLY
The primary energy supply can be determined based on the
final energy demand, the demand for electricity generation,
the demand for transformations between energy vectors and
imports.
In 2021, more than 70% of Europe’s primary energy supply came
from fossil fuels such as coal, oil and fossil methane. In all scenar-
ios, the need for primary energy decreases and is predominantly
met by renewable or low-carbon sources in 2050. Several key
drivers can explain this evolution, as outlined below.
◆ A general assumed decrease in final energy demand as
already explained above.
◆ The role of electrification in reducing primary energy needs
is key, as it mainly replaces fossil fuel-based process such
as heating in buildings and internal combustion engines in
transport. On the one hand, it decreases the direct need for
thermal energy such as methane and/or liquids; on the other
hand, it does increase the need for electricity generation
either via RES or thermal generation. As illustrated in BOX 4-1
the net effect of electrification is a general decrease in pri-
mary energy needs.
◆ The more electrification is met via RES such as wind and solar
PV, the lower the primary energy needs are since electricity
generation via thermal sources requires more energy input
due to the inherent transformation losses.
◆ The share of fossil fuels used to meet primary energy needs is
reduced to less than 10% by 2050 in all scenarios. Most supply
is made up by renewables such as solar PV and wind, whereas
biomass also has a key role to play both as a direct fuel and as
a feedstock for biomethane and liquids. Imported ammonia
and synthetic fuels also have a role to play in further decar-
bonising the sources used to meet primary energy needs.
PRIMARY ENERGY SUPPLY FOR EUROPE FIGURE 45
Hydro
Wind
Solar PV
Biomass
H
2
& synth fuel -
import
Ammonia - import
Nuclear
Oil
Fossil gas
Coal
81%83%82%
91%93%93%
+
18,000
16,000
14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
[TWh]
2036 2040 20502021
Data for Europe (incl. UK, NO, CH).
Includes international shipping & aviation and non-energetic feedstock demand
Historical data recomputed based on EUROSTAT values
Historical values based on EUROSTAT
DE
ELEC
GA DE
ELEC
GA DE
ELEC
GA
13,000 13,50013,500 14,100
12,30012,300 12,900 13,50013,500
11,60011,600
12,200 12,700
Low carbon
RES
28%
27%
19,100
ELECTRIFICATION REDUCES PRIMARY ENERGY NEEDS AND ALSO CONSTITUTES THE MOST EFFI-
CIENT USE OF ENERGY FOR HEATING AND TRANSPORT
Electrification reduces the need for primary energy but is
also the most efficient energy carrier for certain applica-
tion due to mainly two reasons:
1.
There are significantly more efficient technologies
available for transport and low-temperature heating
compared to conventional technologies. This leads to
a two- to threefold reduction in terms of final energy
required for the same amount of ‘useful’ energy.
2.
Renewable energies predominantly produce electricity,
making it the preferred energy carrier across the entire
chain. Indeed, converting from one energy carrier to
another results in energy losses.
To illustrate these two advantages, the first Figure 4-6
shows the varying efficiencies between energy carriers
(including conversion between carriers and transport
losses). The results show a combined efficiency for three
types of end uses: high- and low-temperature heating,
and transport. The efficiencies in the table are indicative
as they could depend on factors such as distances, types
of technologies, future improvements, and specific cases.
However, the conclusion is clear: direct electrification
should be the preferred option from an energy efficiency
perspective when considering transport and heating
demand (even for high-temperature heating).
EFFICIENCY OF DIFFERENT ENERGY PATHWAYS BY END USE FIGURE 46
Low-carbon
power as
primary energy
source
N.A.: 100%
Electrolysis: 70%
E-fuel prod.: 50%
E-fuel prod.: 50%
90%
NH
3
: 75%
H
2
: 90%
95%
95%
EV: 80%
FCEV: 50%
ICE: 25%
HP: 300%
Hydrogen-fired boiler: 80%
NG-fired boiler: 85%
Elec. heater: 95% EV: 70% HP: 270%
Elec. heater: 90%
FCEV: 25-30%
ICE: 10%
Hydrogen-fired boiler: 40-50%
NG-fired boiler: 40%
Carriers
Conversion
to carrier Carrier
transport
End-use
Road transport Road transportLow-T heating Low-T heatingHigh-T heating High-T heating
Combined efficiency per end-use
Electricity
Methane
Hydrogen
Gasoline
Efficiencies and conversions are provided for indication and can differ depending on the technology, distance and future developments
X X =
Building further on the efficiencies for each carrier, the
required offshore wind capacity is provided if all low-tem-
perature heating in Belgium were to be electrified tomor-
row. This amounts 7 GW offshore capacity. Doing this via
another carrier such as hydrogen would require at least
5 times more offshore wind capacity to be built. This is
illustrated in Figure 4-7.
SUPPLYING THE WHOLE SPACE HEAT IN BELGIUM WITH GREEN POWER FIGURE 47
Low-carbon power as
primary energy source
Conversion to carrier Carrier transport End-use: Low-T heating
ElectricityHydrogen
Not applicable: 100%
Electrolysis: 70%
Heat pump: 300%
H
2
-fired boiler: 80%
Elect . transmission: 90%
H
2
transmission: 90%
7 GW
Offshore wind
36 GW
Offshore wind
64 TWh heat
Primary
space heating
demand (BE'21,
residential)
24 TWh
89 TWh 80 TWh
127 TWh
21 TWh
89 TWh 80 TWh
21 TWh 64 TWh
64 TWh
Efficiencies and conversions are provided for indication and can differ depending on the technology, distance and future developments
Low-carbon power Hydrogen as carrier Heat
BOX 4-1
117
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results116
The same reasoning is then adopted for road transport.
10 GW offshore wind would be required to electrify all the
road transport in Belgium while doing this via another
carrier such as hydrogen would require a twofold offshore
wind capacity. This is further illustrated in Figure 4-8.
SUPPLYING THE WHOLE TRANSPORT BELGIAN FLEET WITH GREEN POWER FIGURE 48
Low-carbon power as
primary energy source Conversion to carrier Carrier transport End-use: Road transport
ElectricityHydrogen
Not applicable: 100%
Electrolysis: 70%
EV: 80%
FCEV: 50%
Elect. transmission: 90%
H
2
transmission: 90%
Efficiencies and conversions are provided for indication and can differ depending on the technology, distance and future developments
10 GW
Offshore wind
22 GW
Offshore wind
25 TWh
Road
transport
primary
demand
(BE ’21)
34 TWh 31 TWh 31 TWh
50 TWh50 TWh56 TWh56 TWh79 TWh
25 TWh
25 TWh
Low-carbon power Hydrogen as carrier Kinetic energy at wheels
31 TWh
BOX 4-1
4.1.6. SUMMARY OF INSIGHTS
Given the assumptions adopted for this study, the results demonstrate certain trends relating to the molecule supply. The key
assumptions that have an impact on the supply and demand for molecules are:
◆ the level of biomethane that can be harvested in the future;
whilst Europe uses around 40 TWh today, the large potential
(>1000 TWh) taken into account for later years is able to meet
most of the methane demand. Uncertainties however remain
around the feasibility to source such amounts;
◆ synthetic liquids, whether produced in Europe or imported,
will depend on their associated costs (but also other elements
not assessed in this study such as policy support, targets for
certain sectors…);
◆ the amount of electrolysers will depend on the level of the
hydrogen demand and price of imported fuels; there is still
potential for the development of renewables which has not
yet been harnessed in some parts of Europe (such as offshore
wind or on the periphery of the continent) that could be used
to domestically produce more hydrogen and its derivatives.
WHAT IF THE PRICE FOR IMPORTING MOLECULES WERE TO BE HIGHER OR LOWER?
Given the uncertainty surrounding the price of importing
green molecules, an impact assessment is used to explore
the impact of different green molecule prices. Two sensi-
tivities are performed at the European level, respectively
assuming higher and lower prices for importing mole-
cules.
What assumptions are changed?
As part of this sensitivity analysis, all green molecule
import prices are adjusted (including those for ammonia,
hydrogen, synthetic liquids, synthetic methane, and syn-
thetic LNG). In the high import price scenario, the central
price is multiplied by a factor of 1.5. Conversely, in the low
scenario, a factor of 0.75 is applied.
What does the analysis indicate?
Molecule prices directly influence the location of mole-
cule production, as evidenced by Figure 4-9 and Figure
4-10. When import prices are lower, more molecules are
imported, leading to a reduction in Europe's electrolyser
capacity. Conversely, higher import prices lead to an
increase in locally produced hydrogen and, subsequently,
an expansion of the electrolyser capacity. Fluctuations in
electrolyser capacity directly impact the installed wind
capacity: an increase in installed electrolysers prompts a
corresponding increase in installed wind capacity. Similar
to other simulations, the installed electrolysers in this sce-
nario also remain close to the coast, implying that there
isn't a significant difference in terms of the onshore grid.
Regarding the molecules themselves, it is hydrogen which
is primarily affected, with a transition from the import of
ammonia to electrolysis and pipeline imports occurring
in the event of high prices. The impact on methane and
liquids is less significant due to the limited alternative
options available for decarbonisation.
AMOUNT OF ELECTROLYSERS
IN EUROPE FOR THE HIGHREF
LOW MOLECULE IMPORT PRICES
SENSITIVITIES
FIGURE 49
TOTAL ELECTROLYSERS GW
MOL
EUR+
REF MOL
EUR-
122
208
102
2050
2040
2036
78
10
14
141
44
23
97
12
-16%
+70%
13
BOX 4-2
119
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results118
SHARE OF DOMESTIC AND IMPORTED GREEN MOLECULES IN EUROPE FIGURE 410
+
0%
share of green molecules [%]*
Domestic
ImportedImported
54%
46%46%
70%
30%30%
43%
57%57%
54%
46%46%
68%
32%32%
70%
30%30%
74%
26%26%
87%
13%13%
48%
52%52%
Data for Europe (incl. UK, NO, CH).
* Green molecules include: hydrogen, synthetic liquids, biomethane, bioliquids, ammonia and e-methane
205020402036
REFREFREF MOL
EUR+
MOL
EUR+
MOL
EUR+
MOL
EUR-
MOL
EUR-
MOL
EUR-
How does this affect the results for Belgium?
In the low import price scenario, there is no change com-
pared to the reference scenario, since for Belgium nearly
all molecules are already being imported in the reference
scenario. However, for the high import price scenario, the
same trend observed across the rest of Europe is notice-
able: a shift from imported ammonia towards pipeline
imports of H2 and a slight increase in electrolysis. In total,
it appears that an electrolyser capacity of 2 GW in Belgium
(compared to none in the reference scenario) could be
viable by the year 2050. The general increase in electroly-
sis in Europe also explains the shift to pipeline imports in
Belgium, since it will become possible to import hydro-
gen made through electrolysis in other countries (see
Figure 4-11). More details about the molecule balances for
Belgium are available in Section 5.1.
HYDROGEN BALANCE FOR BELGIUM WITH LOWREFHIGH PRICES FOR THE DE SCENARIO FIGURE 411
Supply Demand
Electrolyser
SMR-H
2
**
End use
Synth. liquids
Power
Ammonia
70
50
30
10
-10
-30
-50
-70
[TWh]
Import ammoniaImport ammonia
Import pipelineImport pipeline
21
11
23 23
29 31 32 34 39 36
2036 2040 20502022*
Figures presented under current policies supply scenarios (no new nuclear nor new non-domestic offshore)
* Historical values based on the Hydrogen Observatory of the European Commission [EUH-1] for SMR-H
2
REF
MOL
EUR+
MOL
EUR- REF
MOL
EUR+
MOL
EUR- REF
MOL
EUR+
MOL
EUR-
BOX 4-2
4.2. SUPPLY AND DEMAND OF ELECTRICITY
Starting from the demand and supply potentials defined in Chap-
ter 3, a multi-energy dispatch and capacity expansion simulation
(see Chapter 2) is performed. The results of this multi-energy
dispatch in terms of yearly balances for each vector is described
in the previous section. This section delves deeper into the supply
and demand results for the electricity dispatch for the simulation
perimeter (EU27 + NO + CH + UK).
It is important to note that the onshore RES and nuclear capac-
ities are defined ex ante and are not optimised by the model.
Several sensitivities are assessed reflecting different future growth
options (RES+, PV+, NIMBY, NUC150). Offshore wind capacities and
their locations are optimised by the model. Adequacy is always
guaranteed via the calibration of the scenarios, and flexibility is
associated with the different demand scenarios (DE, GA, ELEC).
4.2.1. EUROPEAN ELECTRICITY SUPPLY AND DEMAND
Figure 4-12 shows the annual supply and demand of electric-
ity, averaged over all simulated climate years in TWh. It can be
observed that the electrical demand is assumed to increase sig-
nificantly across all simulated scenarios. This increase in electrical
demand is paired with an even bigger increase in renewables in
all scenarios, leading to a significant reduction in the use of mol-
ecule-fired generation. However, as will be discussed in Section
4.6, thermal capacities are still essential for keeping the system
adequate during moments with low RES infeed.
ELECTRICITY GENERATION BY FUEL TYPE FOR EUROPE FIGURE 412
Historical
Offshore
wind
Onshore
wind
Solar PV
Biomass
Hydro
Nuclear
Gas (CH4
and H2)
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
[TWh]
x3
x4
2010 2015 2020
DE DE DE
ELEC ELEC ELEC
GA GA GA
2036 2040 2050
Increasing the temporal granularity, Figure 4-13 provides a closer
picture of the average daily dispatch in Europe (as if it was a cop-
perplate) for a given climate year for an optimised simulation in
the DE scenario for 2050 with central onshore RES assumptions.
◆ Firstly, the figure demonstrates that there is a significant
period where the demand across the entire simulation perim-
eter could be covered by technologies with a very low mar-
ginal cost and carbon intensity. To enable generation to reach
the electrical load, the construction of electric transmission
(and distribution) capacity is required. The results presented
in Section 4.5 show that the further expansion of the electric-
ity grid is also cost optimal.
◆ Secondly, there are also significant periods of time during
which additional electricity generation is needed to cover the
demand. Demand and production during these periods are
brought into equilibrium in the models through a combined
use of demand flexibility, dispatchable generation capacity
and storage.
◆ Thirdly, at the European level, solar generation and wind
generation are complementary over the year. Indeed, solar
sources produce more electricity in summer and wind farms
produce more in winter. The challenge is therefore to allow
this generation (which is spread across Europe) to be trans-
ported to load centres.
◆ Finally, the load, including and excluding electrolysers, is plot-
ted. Electrolysers typically operate during periods when a sur-
plus of low-cost electricity generation is available. Note that
during some specific periods, transmission constraints can
result in both electrolysers and firm generation operating at
the same time in different geographical zones.
121
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results120
DAILY RENEWABLE AND NUCLEAR GENERATION AND LOAD FOR THE FULL SIMULATION
PERIMETER, DE, CENTRAL RES, OPTIMAL OFFSHORE, 2050 FIGURE 413
1,000
900
800
700
600
500
400
300
200
100
0
Average per day [GW]
Each day from January to December for one climate year
Load
Load excluding
electrolysers
Nuclear
Hydro
Biomass
Solar PV
Offshore wind
Onshore wind
4.2.2. OPTIMAL AMOUNT OF OFFSHORE
The amount of offshore wind found to be economically optimal
by the model is shown in Figure 4-14. The amount (slightly above
300 GW) is far from reaching the total offshore potential that
was defined at European level (>800 GW). When looking at the
results per country, only a few countries utilise their assumed
maximum domestic offshore potential. Belgium is one of these
countries. For the optimisation it is assumed that Belgium could
reach a capacity of 8 GW in the model at the soonest in 2040
due to the maturity of the plans around the development of a
third offshore zone (or repowering of existing zones). As such the
optimiser is constrained to not allow additional investments in
domestic offshore in 2036 on top of the assumed initial capacity
of 5.8 GW. However, given the benefits that were found in the
integration of additional offshore wind in the Belgian energy mix
(see also Section 5.4.2) it would also be beneficial to accelerate
the buildout of domestic offshore in Belgium to the full assumed
potential of 8 GW by or before 2036 (see also BOX 4-3). Across
Europe, 318 GW is installed (with 162 GW in the North Sea) in
the DE scenario by 2050. A similar amount, 316 GW (of which 156
GW in the North Sea), is found to be installed in the GA scenario.
In the ELEC scenario, a capacity of 332 GW (with 168 GW in the
North Sea) is installed.
The model has the potential to invest in dedicated offshore elec-
trolysers fed by offshore wind, provided that this is economically
viable. However, this has not emerged as an economically viable
option in any of the scenarios. The result of the optimisation
shows that all offshore wind that is harvested in the model is
connected to the electricity grid (note that electrolysers on an
offshore platform connected to the offshore electricity grid are
not assessed). However, given the untapped potential of offshore
wind in the simulation, there could be use cases or situations
where offshore sites dedicated to the production of hydrogen
could maybe make sense.
OFFSHORE WIND IN EUROPE ECONOMICALLY OPTIMUM VS TOTAL POTENTIAL FIGURE 414
Before 2036*
Invested in 2040
Existing
Invested in 2050
Invested in 2036
+
900
800
700
600
500
400
300
200
100
0
140
120
100
80
60
40
20
0
8
7
6
5
4
3
2
1
0
Installed capacity [GW]
Installed capacity [GW]
Installed capacity [GW]
Offshore Wind capacity, in GW Offshore Wind capacity, in GW
GB SE DE IT NL IE NO DK FR FI EE ES PL PT GR BE LV
BE
318
Total potential
Potential 2050
EU27 + UK + NO
*Installed capacity assumed to be installed in the period 2023-2035 and therefore assumed preconnected
in the model.
BELGIUM ALWAYS REACHES ITS MAXIMUM DOMESTIC OFFSHORE POTENTIAL ACROSS ALL SCE-
NARIOS
As a result of the optimisation performed for this study,
Belgium always reaches its maximum domestic offshore
potential. Given the benefits of integrating additional off-
shore wind in the Belgian energy mix (see also Section 5.4.2)
shifting the current ambitions of the federal government to
reach 8 GW by 2040 to an earlier date is beneficial.
Figure 4-15 provides an overview of the range of offshore
wind supplied to the optimiser as well as the capacity which
is found to be optimal in all results (which always involves
reaching the full potential when it is allowed to invest in
it). This is the case across all the scenarios and sensitivities
that are performed, and is therefore taken as the basis for
all simulations for Belgium, also given its inclusion in the
current Belgian ambitions.
OPTIMAL OFFSHORE WIND CAPACITY FOUND FOR BELGIUM FIGURE 415
Domestic offshore wind
9
8
7
6
5
4
3
2
1
0
Capacity [GW]
Range given
to optimiser
Maximal capacity always
chosen by the optimiser
Given benefits
of adding more
offshore to the
Belgian energy
mix, further
acceleration of
offshore ambitions
appears beneficial
Before 2036
Existing
2010 2015 2020 2025 2030 2036 2040 2050
BOX 4-3
123
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results122
4.2.3. ZONAL ENERGY MIX
Zooming in from a geographical perspective, Figure 4-16 pro-
vides information about the zonal distribution of the electricity
generation mix in an optimised simulation in the DE scenario for
2050 with central RES and nuclear assumptions. The figure shows
the average energy mix across all simulated climate years. The
differences in energy policy – such as nuclear generation being
more present in France, UK and eastern European countries, the
focus on renewables in Germany, and the use of hydro generation
in the Nordics, Austria and Switzerland – can be clearly observed.
The offshore wind is pooled per sea basin in the figure for ease of
understanding. The size of each pie chart is proportional to the
total generation of each zone.
ZONAL ELECTRICITY MIX WHICH
IS FOUND TO BE OPTIMAL FOR DE
2050 CENTRAL RES AND NUCLEAR
SCENARIO
FIGURE 416
Onshore
wind
Hydro
Solar
Nuclear
CH4
turbines
H2
turbines
Biomass
Offshore
wind
20 TWh
50 TWh
100 TWh
4.2.4. ELECTRICITY FLOWS
The difference in energy mixes along with the correlation and
decorrelation of renewable energies and electrical demand result
in energy flows across the European transmission grid. A map
of the flows is provided in Figure 4-17. The colour of the lines
shows how much electrical energy flows across an interconnector
(absolute flow - average across all the climate years). The links
between zones which are loaded the most bidirectionally are not
necessarily the ones which are loaded the most directionally. An
example are the interconnectors that connect Spain and France:
electricity will flow in a northern direction when significant renew-
able production is available in Spain. Conversely, electricity will
flow southwards during periods of lower renewable generation
in Spain.
The figure also depicts the net positions of each zone:
◆ the green zones are exporting zones (the size of these bub-
bles is proportional to the net exports); it should be noted that
the offshore wind connected to the zone and flowing towards
that zone is also accounted for in the calculation of the net
balance for the zones bordering the seas;
◆ the red zones are importing zones; similar to the exporting
zones, the sizes of these bubbles represents the net amount
of imports.
ABSOLUTE FLOWS AND NET
POSITIONS ACROSS THE ELECTRICITY
GRID FOR THE DE 2050 SCENARIO
WITH CENTRAL RES ASSUMPTIONS
FIGURE 417
5 TWh
20 TWh
40 TWh
15 TWh
50 TWh
80 TWh
Net import
Net export
WHAT ARE THE CONSEQUENCES OF MORE OR LESS ELECTRICITY BEING SUPPLIED FROM ONSHORE
SOURCES ACROSS EUROPE?
On top of the Central RES supply scenario which assumes
an average installation rate of 50 GW/year for PV and 15
GW/year of onshore wind, several sensitivities are simu-
lated in order to assess the impact of the onshore supply
assumptions for electricity. These sensitivities are based
on the DE demand scenario and include the addition
of more solar generation, wind onshore generation or
nuclear generation. At the same time, a sensitivity with
less onshore wind is also accounted for.
The different scenarios are described in Figure 4-18 below.
SENSITIVITIES ON EUROPEAN ONSHORE SUPPLY FIGURE 418
Base scenario Options References
PV: + 50GW/y
Wind onshore:
+15 GW/y
Nuclear: known
plans (new and
phase out)
Onshore supply CENTRAL
Based on TYNDP2024 central
input from TSOs
Sensitivities based on the max
potential for each country
NUC
PV+
RES+
NIM
170 GW nuclear in Europe in 2050
(instead of 75 GW)
2,700 GW vs.
1,600 GW +100 GW/y for PV
+25 GW/y for wind
onshore
+75 GW/y for PV
+10 GW/y for wind
onshore
High costs for the
electricity grid
Acceleration for
domestic PV and
onshore wind
Nimby scenario
(lower onshore &
more expensive
onshore grids)
IMPACT ON EUROPE
Figure 4-19 shows the final electricity mix for the model’s
perimeter for the above sensitivities after optimisation.
All sensitivities with higher assumptions for renewable
generation result in lower levels of molecule-fired gener-
ation. This also further results in lower levels of offshore
wind given that more onshore renewables are assumed
to be present. When considering the NIMBY scenario (less
onshore and more expensive onshore grid), more offshore
wind and slightly less electrolysis is found in the optimal
solution.
EUROPEAN ELECTRICITY MIX FOR EACH EUROPEAN SUPPLY SENSITIVITY IN 2050 FIGURE 419
DE
RES+ PV+ NUC
NIMBY
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
-1,000
-2,000
[TWh]
Onshore wind
Hydro
Solar PV
Nuclear
Gas (CH4 and H2)
CCS/U
Electrolysis
Biomass
Offshore wind
BOX 4-4
125
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results124
In addition, depending on the supply mix, a different
contribution of CCS/U electricity consumption and elec-
trolysis consumption is obtained. This can be observed in
Figure 4-20, where the load factor and optimised capac-
ity of electrolysis is shown. This leads to the observations
outlined below.
1.
Increasing onshore RES and/or nuclear production
while keeping the electrical demand identical results in
higher electrolyser capacities. Conversely, reducing the
onshore low-carbon production leads to lower installed
electrolyser capacities.
2.
Increasing the PV capacity beyond the capacities
assumed in the RES+ scenario results in marginal addi-
tional electrolysis capacity and hydrogen production
and slightly lower load factors for electrolysers. The lim-
ited effect of adding more PV on electrolyser capacity
and production can be attributed multiple factors: first
there is the limited load factor of solar. Secondly there
is the fact that solar PV produces most of its energy
during a limited set of hours in a day in which prices are
on average already lower than during the rest of the day
in the central RES scenario. Therefore, while the total
power of renewable energy that can be harvested by
the electrolysers in these hours increases, the running
hours (for the same volume of electrolysers) remains
relatively stable. Finally, the optimiser also invests in
(slightly) less offshore wind than in the RES+ scenario.
3.
Compared to the sensitivities where RES capacities are
increased (RES+ and PV+), the NUC sensitivity leads to
a small reduction in TWh consumed for electrolysis
for a lower level of installed capacity. This effect can
be explained by the availability of nuclear generation
when renewable production is low, resulting in a higher
level of equivalent full load running hours.
4. The NIMBY sensitivity is the only sensitivity where less
electrolysis capacity is slightly lower compared to the
DE scenario. The capacity factor is also lower given less
excess of onshore renewables.
ELECTROLYSER CAPACITY
AND LOAD FACTOR FOR THE
EUROPEAN SUPPLY SENSITIVITIES
IN 2050
FIGURE
420
200
180
160
140
120
100
80
60
40
20
0
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Electrolyser capacity [GW]
Capacity factor [%]
RES+DE PV+ NUC
NIMBY
Capacity factor [%]
Electrolyser capacity [GW]
HOW DOES THIS AFFECT BELGIUM?
The impact on Belgium in terms of its energy mix is very
limited. In terms of electrolysers, it is only in the NUC sen-
sitivity (with an assumed nuclear capacity of 170 GW in
Europe and 8.2 GW in Belgium), that a volume of 1 GW of
electrolysers is found to be economically viable.
THE RENEWABLE ENERGY POTENTIAL IS NOT EVENLY DISTRIBUTED ACROSS EUROPE
There is an uneven distribution of renewable energy
potential across Europe. Depending on the ratio between
domestic renewable capacity and the demand for elec-
tricity, some areas may have more than enough renewable
energy to meet their own demand, while others will need
to either import or produce their own alternative low-car-
bon energy. The left-hand side of Figure 4-21 shows, for
the optimal DE 2050 simulation for each zone modelled
in the electricity model, the comparison of assumed
low-carbon generation (central RES and nuclear) with
the final electricity demand. On the right-hand side, the
maximum RES potential and central nuclear generation
is compared to the final electricity demand including CCS
and electrolysers. Note that offshore potential is attrib-
uted to the closest onshore zone.
This leads to the following observations:
◆ Areas which have more RES potential than they need
are situated in the north and south of Europe or along
the coastlines. This is mainly driven by the offshore
wind potential in the northern seas and PV potential in
the South of Europe. A lot of potential is available across
the periphery of the continent in areas like Scotland,
Ireland, southern Italy, southern Spain, Finland, etc;
◆ The areas which do not have access to sufficient RES
potential to meet their expected electricity demand are
situated in central-western Europe, with load centres in
the south of England, around Paris, Belgium, the Ruhr
basin and the north of Italy. These are either small areas
which are densely populated or industrial clusters;
◆ Given the assumptions used in this study, Belgium does
not have sufficient domestic RES potential to meet its
expected electricity consumption levels. This will be fur-
ther explored in Chapter 5, in which multiple options
are identified to cover the additional electricity supply
arising from electrification.
RES POTENTIAL VS ELECTRICITY DEMAND FOR THE DE2050 SCENARIO FIGURE 421
DE electricity final demand minus central RES
and central nuclear
With
additional
electricity load
for CCS/P2X
and full RES
potentials
Yearly
excess/
shortage of
low-carbon
domestic
generation
[TWh]
domestic low-carbon
oversupply
domestic low-carbon
undersupply
DE electricity final demand including CCS and
P2X minus RES potentials and central nuclear
-150
-100
-50
0
+50
+100
+150
BOX 4-5
127
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results126
4.3. INTERACTIONS BETWEEN ENERGY
VECTORS
4.3.1. SANKEY DIAGRAMS FOR EUROPE
In the lead-up to 2050, the electrification of final demand is
assumed to strongly increase. The volume of renewable genera-
tion grows significantly while the use of non-decarbonised mol-
ecule-fired generation for electricity generation drops. However,
the need for dispatchable generation remains, especially during
long periods with a reduced volume of renewable generation (see
also Section 4.6.1). In addition to the use of molecules for power
generation, electrical demand is added by electrolysers which are
used to generate synthetic molecules when electricity prices are
low compared to hydrogen prices. This happens typically at times
when a decarbonised electricity supply surpasses demand. An
overview of this is provided in Figure 4-22, in which interactions
between the European electricity, methane and hydrogen sys-
tems are highlighted in the so-called ‘Sankey’ diagram. Another
noticeable change from 2036 is the reduction in liquids for the
final energy and more linkage between the molecules themselves
(ammonia, hydrogen, liquids, methane).
MULTIENERGY SANKEY DIAGRAMS FOR EUROPE FOR DE 2036 AND DE 2050 FIGURE 422
2036 - DE
2050 - DE
4.3.2. INTERACTIONS BETWEEN THE ELECTRICITY SYSTEM AND THE
OTHER VECTORS
Figure 4-23 provides a more detailed picture of the interactions
between the electricity system and the molecule system. There
are basically two processes that involve a coupling:
◆ Power plants using gas (methane or in the future hydrogen);
◆ Production of hydrogen via electrolysis.
Firstly, the use of molecules for electricity generation is expected
to decrease in the later years covered by this study. Indeed, the
amount of gas used for power decreases from around 1,500 TWh
today to values between 500 to 1,000 TWh in 2050 across all of
the scenarios. To contextualise this, currently gas is used to meet
17% of the total electricity demand; this value would decrease to
between 5% and 9% in 2050 (accounting for both methane and
hydrogen).
Secondly, very little electricity is currently used to produce mol-
ecules. In the 2050 scenarios, electrolysis amounts to 330 to 740
TWh, or 5% to 13% in 2050 when compared to the total electricity
demand.
Summing both percentages together provides an indication
of the total coupling between the electricity and molecule sys-
tems (in terms of electrical energy). For the ELEC and DE scenar-
ios, the coupling between the methane and hydrogen system
on the one hand and the electricity system on the other hand
decreases. Starting from 17% in 2023, it decreases to around 13-15%
by 2050. For the GA scenario, a stable/slight increase in coupling is
observed, from 17% today to 18% in 2050. In general, the coupling
between the electricity sector and the methane and hydrogen
sectors remains relatively stable or slightly decreases. More elec-
trified scenarios result in a decrease in coupling. It is important to
note that this reasoning is based on the amount of energy that
is used to transform molecules into electricity or electricity into
molecules and not on the capacity.
INTERACTIONS BETWEEN THE ELECTRICITY AND MOLECULE SYSTEMS FIGURE 423
Gas (CH
4
and H
2
)
Electrolysis
DE DE DE
ELEC
ELEC
ELEC
GA GA GA
2036 2040 2050
2023
20%
15%
10%
5%
0%
-5%
-10%
-15%
-20%
Annual production/consumption relative
to electricity demand [%]
17%
11% 10% 10% 7% 7%
9%
5%
-13%
-14%-14%
-10%-10% -8%
-5%-6%-6%
6%
8%
129
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results128
WHERE ARE ELECTROLYSERS LOCATED THROUGHOUT EUROPE?
For each of the European scenarios and sensitivities, the
electrolyser capacities and locations are optimised on a
zonal level (see Section 2.3). Both onshore and offshore
electrolysers are considered for this optimisation. Figure
4-24 shows how much and where (in both the DE and GA
scenarios) electrolyser capacity is optimally integrated
into the system in the 2036 and 2050 simulations.
In the more molecule-oriented (GA) scenario, a capacity
of 180 GW electrolysers is reached for the entire simula-
tion perimeter in 2050. In the DE simulation, about 120
GW is integrated by 2050. The optimiser chooses to inte-
grate electrolysers in areas with a relatively high share of
available low-carbon energy when compared with the
local electricity demand (see also Figure 4-21). Those are
situated mainly at the close-to-shore zones in areas with
excess wind (Northern Europe) or in areas with excess of
PV (Southern Europe). If those electrolysers were to be
placed further away from those zones, this would result
in additional electricity infrastructure capacity required
across Europe.
Given Belgium’s relatively limited local RES potential
compared to its assumed consumption, no or very little
electrolysers are installed in Belgium. Only the GA sce-
nario results in about 1 GW to be installed in Belgium by
2050.
On top of the visualised capacities, an additional capac-
ity of around 20-30 GW of electrolysers was found in the
Nordics and around 5-10 GW was found in the South of
Europe for 2050.
OPTIMISED ELECTROLYSER CAPACITIES IN THE DE AND GA SCENARIOS FIGURE 424
2036 2050
DEGA
Total: 97 GW
BE: 0 GW
Total: 131 GW
BE: 0.5 GW
Total: 121 GW
BE: 0 GW
Total: 180 GW
BE: 1 GW
BOX 4-6
4.4. MANAGEMENT OF EMISSIONS
4.4.1. CHANGES IN GHG EMISSIONS
It is assumed that Europe reaches net zero by 2050 in all sim-
ulated scenarios. The full scope of GHG emissions is taken into
account either through ex ante assumptions for non-modelled
emissions (such as LULUCF, Non-CO2 and process emissions) or
by explicit modelling and quantification within the model (for
energy-related emissions).
First, the emissions trajectory is defined ex ante for the base
scenario, assuming a reduction of 55% by 2030, 90% by 2040 and
net zero by 2050. The non-CO2 emissions and LULUCF emissions
are defined following the S3 scenario of the EC (see Section 3.1.5
for more information).
The changes in emissions per sector are depicted in Figure 4-25
for both historical and simulated years. Note that all scenarios
follow the same total net emissions trajectory, as this is set as a
constraint in the model. In this view, it can be seen that energy
emissions (i.e. from the combustion of fuels) are almost fully
phased out by 2050. The remaining energy emissions mainly stem
from the use of synthetic fuels; from a ‘net’ perspective, these are
compensated by the fact that their emissions were previously
captured either via carbon capture in industry, power gener-
ation or via direct air capture (DAC). Depending on the source
of feedstock not all biofuels are also assumed to have net-zero
emissions. Emissions such as those from (industrial) processes
and other GHGs such as CH
4
, NO
2
and F-gases constitute less than
25% of total emissions in 2022, but amount to close to 70% of the
remaining emissions in 2050. These are typically hard to abate
even after the application of new technologies or fuel switching
and will also require abatement in the form of CCS and/or LULUCF.
Focusing more specifically on the changes per sector, the power
sector is almost fully decarbonised by 2036. The remaining emis-
sions in the power sector are mainly linked to methane-fired
generation as most coal in Europe is assumed to be phased out
by 2036. By 2050, the buildings and domestic transport sectors
are also nearly entirely carbon free. This is mainly driven by their
electrification and the replacement of fossil fuels towards elec-
trification (more so the case in the ELEC scenario) and because
the remaining gaseous and liquid consumption will have mainly
switched to bio and synthetic alternatives. The main sector in
which emissions will still be high in 2050 is the industrial sector.
This can be attributed to process emissions that do not depend
on the type of fuel used. However, a lot of this remaining CO2 is
compensated by carbon capture (see next section).
CHANGES IN TOTAL GHG EMISSIONS IN EUROPE PER SECTOR FIGURE 425
+
Int shipping**
Int aviation
LULUCF
DAC
Carbon Capture
Non-CO
2
Domestic transport
Buildings
Industry - processes
Power&Heat
Industry - energy*
5,000
4,000
3,000
2,000
1,000
0
-1,000
[MtCO2/y]
2010 2020
Data for Europe (incl. UK, NO, CH)
For UK, because of incomplete data, 2022 emissions data are assumed the same as in 2021
The sectoral split concerns CO
2
emissisions, the non-CO
2
emissions are shown seperately in aggregate
* also includes refineries, agriculture and waste management
** includes 50% of the emissions
Historical values based on European Environment Agency and Department for Energy Security & net-zero for the UK
2050
DE
ELEC
GA
2040
DE
ELEC
GA
2036
DE
ELEC
GA
Positve Netgative
Net emission
131
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results130
4.4.2. CARBON CAPTURE, USAGE AND STORAGE
As explained in Appendix C, the model can choose to invest in
carbon capture, storage and/or usage technologies (CC) if these
are deemed to be financially viable and/or are required to meet
the European emissions target.
Examining the captured carbon more closely (see Figure 4-26)
leads to the observations below about the source of carbon
captured:
◆ Most CC is installed in industry, mainly for the capturing of
process emissions in the mineral, metals and cement sectors,
and also in heat-based processes where CO2 is emitted during
combustion;
◆ Direct air capture (DAC) is a relatively expensive technology
for the abatement of emissions; however, from 2040 onwards
and mainly in 2050, it will be required to compensate for
some emissions that are practically impossible to completely
remove (such as some non-CO2 emissions);
◆ CC in power and SMR-H2 is limited during the whole time
horizon, which can mainly be explained by the relatively low
amount of running hours of gas-fired power plants (which
reduces their economic viability of being fitted with CC).
The captured carbon is either stored or used:
◆ it is mainly stored underground in 2036 while it is also used to
produce feedstock and synthetic fuels in later years;
◆ when the carbon is stored underground or used as feedstock
for the creation of chemical products, this produces negative
emissions; in cases where carbon is used for energetic synfu-
els, the carbon is emitted during combustion.
CARBON CAPTURE, USAGE AND STORAGE FIGURE 426
500
400
300
200
100
0
-100
-200
-300
-400
-500
[MtCO2/y]
DAC
CC Industry
CC Power & SMR-H
2
Capture
Synthetic - as fuel
Synthetic -
as feedstock
Storage-underground
Usage/storage
2036 2040 2050
Data for Europe (incl. UK, NO, CH).
DE
ELEC
GA DE
ELEC
GA DE
ELEC
GA
390
370
300300
170170
130
350
430
WHAT IS THE IMPACT OF HIGHER OR LOWER CO
2
TARGETS?
While the base scenario assumes a 90% reduction by 2040
in carbon emissions across the entire European perime-
ter, two sensitivity analyses are conducted. One assumes
a less ambitious 80% reduction by 2040 (and hence a
reduction of -70% in 2036 instead of -76%), while the other
reflects that non-CO2 emissions may not decrease as pro-
jected, coupled with the possibility that LULUCF may not
be able to offset the emissions as anticipated. This last
sensitivity implies a higher CO2 reduction in the energy
sector. This is further detailed in Appendix H.
Impact of a reduced carbon reduction target on energy
measures
When the carbon reduction target is lowered for the years
2036 and 2040, the new target is generally achieved with
less CC. The following observations therefore hold:
◆ Figure 4-27 reveals that there is a reduced need for car-
bon capture and storage measures. The cap on carbon
storage is not reached in these instances. In 2036, the
only carbon capture incorporated is for SMR-H2 opera-
tions. This is because the application of CCS alongside
SMR-H2 is inherent to the model.
◆ The model opts for reduced investment in wind and
electrolysers, as hydrogen can now be produced via
SMR-H2. The preference for increased SMR-H2 produc-
tion in 2036 and 2040 is directly tied to the carbon
reduction target. Furthermore, by 2040, the need to
produce domestic synthetic liquids diminishes due to
the reduced need to lower CO2 emissions in liquids.
However, in 2050, when the target aligns with the base
scenario of net zero, SMR-H2 is no longer used.
◆ Methane and liquids continue to be entirely based on
fossil fuels as illustrated in Figure 4-28. This increases
Europe’s dependence on countries that supply oil and
gas compared to the base scenario.
This analysis highlights the impact that a 10% reduction in
the greenhouse gas target, from 90% to 80% by 2040, can
have on the EU’s energy system.
Challenges related to achieving targets when non-CO2
emissions do not decrease as anticipated
If non-CO2 emissions do not decrease as anticipated,
achieving the target becomes more challenging for the
energy sector. The following trends are observed:
For methane, the transition to domestic biomethane
occurs from 2036 onwards, rather than only from 2040.
◆ The primary difference lies in the balance of liquids,
particularly in 2040, where more synthetic liquids are
imported to compensate for the increased emissions in
other sectors.
◆ By 2050, an increased amount of underground CCS is
required to offset the unavoidable emissions.
◆ In 2036 and 2040, slightly less carbon capture is
required due to the increased emissions which drive up
the amount of invested wind power and augment the
import of synthetic molecules from outside Europe.
COMPARATIVE OVERVIEW OF CARBON CAPTURE AND STORAGE FOR HIGHER
AND LOWER CO2 TARGETS ON THE ENERGY SYSTEM FIGURE 427
500
400
300
200
100
0
-100
-200
-300
-400
-500
[MtCO2/y]
DAC
CC Industry
CC Power &
SMR-H
2
Capture
Synthetic - as fuel
Synthetic -
as feedstock
Storage-
underground
Usage/storage
Data for Europe (incl. UK, NO, CH).
REF
CO
2
80%
non
CO
2
+REF
CO
2
80%
non
CO
2
+REF
CO
2
80%
non
CO
2
+
100
170 160
80
300 270
340 350
420
2036 2040 2050
BOX 4-7
133
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results132
OVERVIEW OF GREEN AND FOSSIL MOLECULE IMPORTS VS. DOMESTIC PRODUCTION FIGURE 428
SOURCE OF MOLECULES
Green - Domestic
Fossil - Domestic*
+
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Data for Europe (incl. UK, NO, CH)
*Includes SMR-H
2
REF
CO
2
80%
non
CO
2
+REF
CO
2
80%
non
CO
2
+REF
CO
2
80%
non
CO
2
+
Fossil - ImportsFossil - Imports
Green - ImportsGreen - Imports
2036 2040 2050
Impact on investment decisions for the electricity
system
In 2050, both the reduced target scenario and the high
non-CO2 emission scenario result in a grid which is similar
to the grid found in the DE scenario. However, the paths
to achieving this outcome vary significantly, as depicted
in Figure 4-29.
Under the reduced target scenario, grid and offshore
wind investments progress at a noticeably slower pace,
with strongly reduced wind and grid investment in 2036
and 2040. This leads to the need for substantial invest-
ments to be made in offshore wind capacity between
2040 and 2050.
By contrast, under the high non-CO2 emission scenario,
greater wind investment occurs up until 2040. This is to
offset the additional non-CO2 emissions, where the differ-
ence is more significant up until 2040.
In conclusion, the results of the model show that the estab-
lishment of more stringent intermediary targets requires
an acceleration in investments related to the grid, offshore
wind, and CCS (amongst other areas). Despite these efforts,
the final outcome results in a similar situation to those
which develop in less stringent scenarios. However, it's cru-
cial to understand that the carbon emissions over a specific
period are what truly matter, not the 'instantaneous' carbon
emitted in a particular year. Indeed, the journey towards
the final target is just as important as the final target itself.
The difference between -80% and -90% reduction pathways
towards 2050 results in an additional 5 gigatonnes in the
atmosphere in 2050.
INVESTMENTS IN OFFSHORE
WIND PER TARGET YEAR FIGURE 429
This excludes existing wind infrastructure and wind investments prior to 2036.
Investments in 2036 are to be considered on top of pre-connected wind before 2036.
2040-2050
2036-2040
initial-2036
74%
19%
7%
26%
46%
30%
56%
28%
14%
Offshore wind invested per year
REF
CO2
80%
non
CO2+
4.5. EUROPEAN ELECTRICITY GRID
The European electricity grid is optimised as part of the optimi-
sation process followed for each scenario (see Chapter 2 for more
explanation of the methodology adopted). This optimisation pro-
vides insights into what the European high-voltage electricity grid
would look like if it could be optimised by a (simplified) central
planner and in a zonal set-up (big countries are split into multiple
zones to reflect the grid physics). The optimisation minimises the
costs at the European level (no indicator of zonal or country ben-
efits was taken into account), meaning it invests in the different
reinforcements if these reduce European system costs.
4.5.1. IMPORTANT ELEMENTS FOR THE INTERPRETATION OF RESULTS
The following points need to be kept in mind when interpreting
the results:
◆ The geographical granularity and flow-based approach
used in this study raises the possibility of identifying key
trends about cross-border transmission for each scenario.
The results also provide (partial) insights into reinforcements
within each country. It's crucial to note that depending on the
market design (like bidding zones or other rules for cross-bor-
der capacity calculation and allocation) or other constraints
which are unaccounted for, different reinforcement needs
may arise;
◆ Complementing the previous point, additional calculations
are performed to analyse the need for internal reinforcements
(corridors) in Belgium which consider the current market
design. This additional analysis is performed for all Belgian
sensitivities since, depending on the scenario, the need for
reinforcements to the internal backbone emerges and the
need is accounted for in the Belgian system costs. These
results are further discussed in section 5.7;
◆ In addition, while the charts in this section provide graphi-
cal illustrations of potential reinforcements, these reinforce-
ments could deviate from the illustrations shown in the cur-
rent study as they are designed in more detail. However, if the
projects fall within a strategic corridor that was put forward by
the optimisation model and if their potential is confirmed by
the investigations, they are very close to the theoretical opti-
mal solution and their realisation should be pursued;
◆ It is important to note that while a cost optimal solution can
be found, many other constraints are not accounted for in the
model such as spatial constraints, the willingness of coun-
tries/zones to develop projects, NIMBY approaches, routing
constraints, national/zonal benefits, financial constraints, etc;
◆ For Belgium, the ‘current policy’ scenario is used as input
for the European optimisation (no nuclear after 2036). The
amount of non-domestic offshore wind connected to Bel-
gium is not capped and could reach values beyond the max-
imum potential beyond the Belgian EEZ boundaries (see
3.2.3.4);
◆ Finally, when multiple scenarios show the same key trends,
these investments are likely to bring value to society inde-
pendently of the scenario that finally materialises.
135
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results134
4.5.2. OPTIMAL EUROPEAN GRID FOUND
Figure 4-30 depicts the resulting grid for the DE scenario for the
years 2036 and 2050. The following observations can be made:
◆ A large amount of the grid needs to be further developed
over the coming decade. Indeed, the rate of increase in infra-
structure capacity is more than two times higher between
now and 2040 compared with between 2040 and 2050. This
is linked to the increase in electrification and the increase in
renewable generation in Europe (point 1 & 2 in the figure);
◆ A meshed offshore grid is found to be developed by 2036 in
the North Sea, showing that such a solution is interesting
from a European total cost perspective (BOX 4-8 explores the
consequences of doing this via radial connections) (point 3 in
the figure);
◆ Finally, the further construction of connections between the
British Isles and the continent but also across the Baltic and
Mediterranean seas are observed.
EVOLUTION OF THE OPTIMAL GRID AND INTEGRATED OFFSHORE WIND FOR THE DE SCENARIO FIGURE 430
4 4
3 3
3
2
1
3 3
3 3
*+,-
*+,-
,>
Grid topology 2050
Key observations
Grid topology 2036
Capacity evolution
DEDE
New offshore-offshore New offshore-onshore New onshore-onshore
Significant buildout of onshore and offshore grid
on top of the assumed initial capacities already in 2036.
Most of the new grid investments are made by 2040.
Between 2040 and 2050 a further 16% is added compared to the
initial grid.
Significant buildout of meshed offshore grid in the North Sea
(and Baltic Sea) is found as optimal by the optimiser.
The Mediterranean Sea also shows further reinforcement.
Further interconnection of Britain and Ireland
with the continent is found to be beneficial.
4
1
2
2
+62%
+78%
New
Offshore grid
New
Onshore grid
Initial grid
+42%
250
200
150
100
50
0
Equivalent transmission [nominal power . length]
2036 2040 2050
In addition to the DE scenario, the GA and ELEC scenarios are
also assessed. A comparison between the installed capacities of
the grid is depicted in Figure 4-31. A significant expansion in the
installed grid can be observed for all scenarios. More electrified
scenarios obviously require more reinforcements of electrical
grids; however, the need remains high in the less electrified
scenarios.
NEW OFFSHORE AND ONSHORE GRID CAPACITY INVESTED PER SCENARIO FIGURE 431
300
250
200
150
100
50
0
Equivalent transmission [nominal power . length]
initial grid = TYNDP IOSN2022 starting grid
+59% +75%+71% +84%
DE DE DE
ELEC ELEC ELEC
GA GA GA
New
onshore grid
New
offshore grid
16%
12%
9%
+51% +42% +63% +78%+35%
2036 2040 2050
Initial grid
4.5.3. RESULTS OF THE EUROPEAN OPTIMISATION AROUND BELGIUM
Having explored the European grid results above, this section
provides some insights into results from the optimisation for
the grid in and around Belgium. It is important to note that only
cross-border results are shown. Internal grid reinforcements and
more details about the onshore interconnectors being assessed
are explored in Section 5.7. As multiple European scenarios, each
with their own full grid optimisation, are analysed, it is possible to
assess the number of scenarios in which certain reinforcements
appear. Cross-border investments which appear in multiple sce-
narios demonstrate robustness in the face of changes to the
assumptions. Figure 4-32 outlines the investments per border and
the number of different optimisations they occur in for the target
year 2050. It can be observed that the buildout of the offshore
grid occurs in all scenarios. Some level of reinforcement happens
on the east border in all optimisations, although these vary from
1 GW to 5 GW depending on the optimisation. The Belgian grid
results are further discussed in Section 5.7.
BELGIAN GRID INVESTMENTS ACROSS ALL EUROPEAN MAIN SCENARIOS AND SENSITIVITIES FIGURE 432
25
20
15
10
5
0
GW additional transmission
For all European sensitivities
Belgian cross-border grid optimum* (2050)
*The European optimisation starts from the assumption that there is both no limit on the amount of interconnectors and that there is no nuclear in Belgium. Should any of
these assumptions change is it possible that other projects would appear in the optimum.
North
Offshore - NS
Offshore - ATL // South
North East Offshore
- NS Offshore
- ATL //
South
Optimal in at least...
1 optimisation
25% of the optimisations
50% of the optimisations
75% of the optimisations
All optimisations
East
137
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results136
WHAT IF CERTAIN TYPES OF ELECTRIC (OFFSHORE) CONNECTIONS ARE NOT ALLOWED OR MORE
EXPENSIVE IN EUROPE ?
Grid development in different sensitivities
In addition to the main scenarios (DE, GA and ELEC) sev-
eral sensitivities are applied by adding constraints to the
optimisation algorithm. Figure 4-33 provides an overview
of the optimal grid obtained in the RAD, RAD+, 400 GW
and NIMBY sensitivities in 2050.
In all sensitivities, considerable onshore grid develop-
ment is observed. However, in the NIMBY sensitivity, this
onshore expansion is slightly more limited due to the
assumed higher onshore grid costs (and lower onshore
wind assumed). Compensating for this reduction in the
onshore grid, the offshore grid is more developed mean-
ing that, in total, grid volume remains similar to the ref-
erence case in this sensitivity. This shift is also reflected
in wind power with the reduced assumed volume in
onshore wind resulting in higher offshore wind volumes
being integrated. In the radial-only sensitivities, the
model shows a shift to onshore-to-onshore overseas links,
instead of hybrid connections to other countries as in the
reference case.
OPTIMAL GRID FOUND IN DIFFERENT SENSITIVITIES ON OFFSHORE GRID DEVELOPMENT FIGURE 433
3
2
2 2
1
4
4
3
11
1
New offshore-offshore
New offshore-onshore
New onshore-onshore
Significant buildout of onshore is observed in all sensitivities
A significant volume of offshore wind is connected in the
North Sea even when only radial interconnections are allowed
Only radial to country
Connect 400 GW offshore
Radial + to other country allowed
Less onshore wind and higher onshore grid cost
RAD
400
GW
RAD+
NIMBY
If radial connections to other countries are allowed, the
model uses this option to connect Belgium to offshore
wind farms not in its EEZ
In the NIMBY sensitivity, where onshore grid costs were
doubled, slightly less (albeit still significant) onshore
reinforcements are observed
BOX 4-8
The impact of restricting offshore grid buildout on overall system costs
Restricting the expansion of the offshore grid to solely
radial connections, or even more stringently to national
radial connections, leads to a rise in the overall system
cost. This cost increase is mainly driven by two factors.
Firstly, offshore wind integration is less efficient as it is
confined to land-based transportation, rather than being
able to exploit more optimal sea routes. Secondly, allow-
ing only radial connections for new offshore results in
interconnections between onshore zones having to be
made with direct interconnectors instead of being able to
benefit from a meshed offshore grid.
When a forced investment of 400 GW offshore wind is
considered, the cost increase for the total system is rel-
atively limited. This is because additional investments in
offshore wind and grid past the optimum, although they
result in increased costs, also bring significant benefits
which largely compensate for the additional costs. In
general a certain asymmetry is observed in the optimi-
sation where pushing the investment loop to invest past
the optimum results in a less steep increase in total costs
than when investing too little.
Finally, while not presented in the figures, a sensitivity
is performed where the total grid is optimised for 2050
without taking into account investments in earlier years.
Theoretically this should lead to a more optimal grid for
2050 (but not necessarily in total as the intermediate
years are not considered). It is observed that the final grid
configuration remains largely the same.
CHANGE IN COSTS FOR DIFFERENT SENSITIVITIES IN 2050 FIGURE 434
Only radial links allowed
Force investment of 400GW
Optimal amount of
offshore installed:
318 GW
Reference case
Only national radial
links allowed
RAD
All configurations of
offshore allowed
400
GW
RAD+
DE Sensitivities
+29 B€/year
+25 B€/year
+4 B€/year
139
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results138
4.5.4. MAIN TAKEAWAYS REGARDING THE DEVELOPMENT OF THE HIGH-
VOLTAGE GRID
The results for Europe as a whole highlight the following points:
◆ major expansion work across the high-voltage grid is found
to be needed between today and 2050. The biggest need
emerges between 2036 and 2040: +50% compared to the ini-
tial grid. By 2050, the growth of the grid amounts to around
+75% in total;
◆ the need for this grid expansion is experienced in all scenarios
and sensitivities analysed as part of this study. A key driver for
the growth of the grid is the electrification of consumption
and the increase in RES integrated into the system linked to
the net-zero ambitions.
When assessing the impact around Belgium, the following
becomes clear:
◆ the results show that onshore borders towards the north and
east are reinforced (the western border is mainly reinforced
via the sea combined with offshore), making these interesting
areas to explore further;
◆ a major rise in offshore capacity is also found to be optimal
for Belgium (on top of the current policies and considering
no new nuclear).
In terms of offshore grid development, relevant insights that
require further investigation include:
◆ Offshore hybrids and meshing provide greater societal gains
at a European level. When co-optimising on- and offshore
generation and transmission infrastructure, substantial off-
shore meshing is selected by the optimiser. Such meshing,
of course, relies on the availability of the required technology;
◆ Significant offshore reinforcements across the North Sea
appear in the simulations and could constitute a fundamen-
tal building block of the future offshore European network.
Although the exact locations of those reinforcements can
be different across each of the scenarios, they are strongly
present in all of the scenarios that allow for offshore mesh-
ing;
◆ The construction of additional connections from Ireland
and the United Kingdom to the continent, whether through
hybrids or through direct interconnectors, also emerges in
several scenarios. Other corridors also seem to appear in the
Baltic Sea, Atlantic Sea or between islands in the Mediterra-
nean Sea.
4.6. ADEQUACY AND FLEXIBILITY
The adequacy of the power system is guaranteed in all scenarios
via the installation of thermal units (on top of all flexibility that was
defined ex ante for each demand scenario based on the TYNDP
assumptions and on top of nuclear generation). This thermal gen-
eration can be methane-fired or hydrogen-fired. The model also
includes the option to add CCS to power generation. Other ther-
mal generation, such as biomass-fired generation and nuclear,
are defined ex ante in the scenarios. Biomass is kept constant
while a sensitivity is performed with increased nuclear capacities.
4.6.1. REQUIRED THERMAL GENERATION
The amount of thermal capacity across Europe is depicted in Fig-
ure 4-35. This is the capacity required for each zone in the model
to be adequate. The figure shows the historical split between
nuclear, coal, oil and methane-fired units, as well as the amounts
for each of the simulated scenarios and the split between nuclear,
coal, existing methane-fired and new thermal required.
The peak demand included in the graph is the synchronous
European peak, excluding any type of flexibility or storage. The
range provided corresponds to the weather variability (e.g. colder/
warmer winters).
INSTALLED THERMAL CAPACITIES AND PEAK DEMAND RANGES FOR EACH SCENARIO FIGURE 435
Installed capacity/peak in [GW]
1,200
1,000
800
600
400
200
0
2010
Nuclear Coal Oil New thermalExisting (and being built) methane & biomass
ELEC ELEC ELEC
2015 2036 2040 2050
DE DE DE
2020
GA GA GA
European
synchronous
peak load
range
(excluding
demand
flexibility)
The following points can be derived from the figure:
◆ electricity peak variability is expected to increase, linked to
the assumed electrification of heat (compared to the histor-
ical observations). The peak depicted in the graph excludes
any demand flexibility;
◆ the amount of thermal capacity required (assuming that the
flexibility in demand and additional storage is developed) is
expected to slightly decrease from around 500 GW in 2020 to
around 400 GW as from 2036;
◆ depending on the level of electrification, there is a difference
of around 100 GW between the GA and ELEC scenarios;
◆ from 2036 onwards, half of the adequacy requirements are
met by thermal generation, with the other half met by addi-
tional and existing storage, demand flexibility and increased
(on- and offshore) RES and additional interconnectors that
allow an efficient use of RES across borders;
◆ additional new thermal generation emerges in all scenarios.
However, its amount varies depending on the existing fleet
and level of electrification. The level of flexibility and level
of interconnection also impact the amount of dispatchable
capacity required.
The required new capacity could come in the form of new meth-
ane-fired units (with or without CCS), or new hydrogen-fired
units. It could also be supplemented by additional new nuclear
generation units on top of the ex ante assumption made. Tech-
nology-related choices do not need to be taken right now, but
adequacy should be monitored over a 10-year period (as is cur-
rently carried out for most adequacy assessments in Europe). This
will be vital for identifying the need for capacity, the economic
viability of existing (and new) generation, and any necessary sup-
port mechanisms.
Based on the modelling exercise, several insights can also be
provided regarding the technology choices:
◆ The model rarely opts for CCS in power generation due to the
decreasing operational hours needed for power plants. This
makes the high capital expenditure (CAPEX) associated with
CCS less appealing. Moreover, CCS in power generation com-
petes with other CCS applications (like in the industrial sec-
tor), from which more value can be derived.
141
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results140
◆ Extending the lifespan of existing units is consistently seen
as a cost-effective option due to their lower CAPEX require-
ments compared to new units. However, the potential availa-
bility and price of biomethane, synthetic methane and other
applications where methane is used should be factored into
the decision-making process.
◆ The decision between installing new methane or hydrogen
turbines should be made on a case-by-case basis, considering
each unit and location as well as the availability and price of
synthetic or biomethane and required (new) infrastructure.
The model tends to favour the installation of hydrogen units
in most locations, since the potential of domestic low-carbon
methane is being used for other end uses.
4.6.2. GENERATION CHARACTERISTICS OF THERMAL UNITS
In all scenarios, the amount of running hours of different mol-
ecule-fired thermal technologies is expected to decrease com-
pared with today. However, these dispatchable capacities remain
needed to cover sustained periods of low RES infeed. The need
for dispatchable capacities will need to be closely monitored in
each country / across Europe as it will depend on the level of
flexibility that can be activated, the interconnectors, changes
in electricity consumption and the amount of renewables (and
their type) in the system.
Dispatchable capacities do not run very often given their high
marginal cost. Periods during which low amounts of RES are avail-
able mainly occur in winter and are typically driven by low wind
infeed. Indeed, during the rest of the year, electricity consump-
tion is lower and PV generation, wind generation and storage
combined allow consumption to be covered most of the time.
Figure 4-36 shows the duration curve of the share of the load
covered by thermal generation in Europe (produced from meth-
ane or hydrogen):
◆ While dispatchable capacities do not run often, it is possible
that they run at different moments within the year (explain-
ing that there is some thermal generation running around
50% of the time);
◆ The highest hourly share of dispatchable generation found
(compared to the electricity load) is between 25% and 40%
across the different sensitivities;
◆ Dispatchable generation is used less often when considering
more onshore supply.
LOAD DURATION CURVE OF THE LOAD SHARE OF GASFIRED DISPATCHABLE GENERATION IN
EUROPE IN 2050 FIGURE 436
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
Share of the load covered by dispatchable thermal generation [%]
All hours in a year
400
GW
PV+
RES+
NUC
DE
Figure 4-37 depicts the thermal generation distribution over
an entire year for Europe. The seasonality effect can be clearly
observed with the generation happening almost exclusively dur-
ing the winter months. This analysis excludes any requirements
to keep dispatchable generation for other reasons (e.g. redis-
patching, balancing…).
MONTHLY GENERATION OF GASFIRED DISPATCHABLE GENERATION IN EUROPE IN 2050 FIGURE 437
100
90
80
70
60
50
40
30
20
10
0
[TWh]
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
P95
P70
Average
P30
P5
Percentile
Distribution over all climate years and European demand and supply sensitivities.
143
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results142
IMPACT OF MORE FLEXIBILITY IN THE EUROPEAN POWER SYSTEM
The demand flexibility and storage capacities are defined
ex ante in the different scenarios. Those are based on the
TYNDP2024 and are linked to the demand scenarios.
◆ To evaluate the effect of enhanced system flexibility,
a sensitivity analysis is conducted for the DE scenario,
in which the energy content of storage and demand
flexibility is doubled. The main impacts observed are as
follows:
◆ Less thermal generation is required in order to comply
with adequacy at European level. This also results in a
reduced need for CCS visible in Figure 4-39.
◆ Figure 4-38 shows the grid in the high flexibility sce-
nario. As the same amount of offshore wind is installed
(in total 134 GW of new wind capacity), a similar invest-
ment in the offshore grid is observed. However, there
is a decrease in onshore grid investment as additional
renewables can be temporarily stored at their produc-
tion sites, reducing the need for immediate transpor-
tation.
◆ With increased flexibility leading to the more efficient
integration of renewables, there is a decrease in the
installation of electrolysers, with only 105 GW installed
across Europe. This trend is further illustrated in Figure
4-39.
◆ There is a decrease in the amount of curtailed energy
from renewables. As can also be concluded from Figure
4-39, the capacity of renewables in the system is equal;
however, the total generation of renewables is greater
in the high flexibility scenario due to lower curtailment.
NEW OFFSHORE AND ONSHORE GRID CAPACITY INVESTED IN DE AND HFLEX SCENARIOS FIGURE 438
250
200
150
100
50
0
Equivalent transmission [nominal power . length]
Capacity evolutionVS HFLEX
HFLEX
DE
DE
2
2
1
1
+42% +39%
+78%
+74%
2
1
Key observations
In both the DE and HFLEX scenario significant
buildout of grid is observed. Nevertheless slightly
less additional grid is built in the HFLEX sensitivity.
Most of the reduction in grid buildout for the high
flex sensitivity happend onshore, the amount of
offshore grid remains mostly the same and is
necessary to integrate all new offshore.
New
Onshore grid
New
Offshore grid
Initial grid
EUROPEAN ELECTRICITY MIX FOR DE AND HFLEX SCENARIOS FIGURE 439
Electrolysis
Onshore wind
Hydro
Solar PV
Nuclear
Gas (CH4 and H2)
CCS/U
Biomass
Offshore wind
DE HFLEX
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
-1,000
-2,000
[TWh]
~ 3100 TWh~ 3100 TWh~ 3000 TWh~ 3000 TWh
Electrolysis
~ 300 TWh~ 300 TWh~ 400 TWh~ 400 TWh
BOX 4-9
4.7. SYSTEM COSTS ACROSS THE
DIFFERENT SCENARIOS
The total costs of the energy system are presented in this section
and include the capital expenditure costs (CAPEX), operating
expenditure costs (OPEX) and fuel costs associated with the final
energy use of energy carriers and energy-related investments
in three end user sectors (industry, transport and buildings).
Fiscal costs, such as taxes, subsidies, levies, and redistributions
are excluded from the system costs. All investment costs are
annualised over the assumed technical lifetime of the asset. The
annualisation follows a normative depreciation and actualisation
approach, where a constant WACC (with sensitivities) is assumed
invariant of the technology. This approach excludes considera-
tions of contractual structures or market interventions, e.g. price
guarantees, but supports the key objective of the study to com-
pare alternative energy mix scenarios from a system point of
view (in contrast to individual project’s/investors’ point of view).
All costs are expressed in real euros and have been adjusted for
inflation to 2022 values. The annuities stemming from already
made investments are assumed to be the same as the annuities
of the first modelled year. Also see Section 2.4. and Appendix F
for more details.
4.7.1. TOTAL ENERGY SYSTEM COSTS INCLUDING END USES
This section includes OPEX and CAPEX annualised to a WACC
for:
◆ the power system;
◆ the other vectors (molecules);
◆ the end uses investments in industry, transport and
buildings.
In order to compare the different scenarios, the total system costs
in billions of euros per year for the whole of Europe are depicted
in Figure 4-40. The split between the different cost categories
is also included. The values relate to annual spending amounts,
including the yearly CAPEX over the assumed technical lifetime
of each asset and assumed WACC, OPEX, and network costs. Note
that fuel costs are included in the power and molecules costs, but
excluded from end uses. More information on the methodology
used can be found in Appendix F.
Compared with the gross domestic product (GDP) of the entire
region (a bit less than €20,000 billion in 2022 [EUC-15] [NOG-2]
[WBG-1], the costs related to investments and operational costs
of the system (which include investments in energy efficiencies
such as insulation of buildings, acquisition of new cars or charging
infrastructure) would represent around 10% to 15% of the total
GDP of Europe. Note that these costs do not include CO2 prices.
In all scenarios, the cost for end-use sectors generally accounts
for the largest relative share of the total cost. The primary drivers
in these sectors are the costs associated with renovations and
the replacement of heating devices in buildings, as well as the
necessary investments for renewing the vehicle fleet and rolling
out new infrastructure, such as electric vehicle charging stations.
Moreover, the scenarios which involve a higher level of electrifi-
cation require lower annual spending. A higher degree of electri-
fication requires more investment in the power system for grid,
(backup) capacity and operational costs; however, this is more
than compensated for by the fact that the reduced amount of
required domestic and imported molecules reduces the cost of
the molecule system. Additionally, the GA scenario assumes some
end use technologies that are relatively more expensive, such as
fuel cell vehicles and hydrogen heating in buildings.
TOTAL SYSTEM COSTS FOR EUROPE ANNUAL FIGURES FIGURE 440
+
Industry
Buildings
Liquids
Transport
CCS/U
Methane
Ammonia
Hydrogen
Power**
Molecules
End-uses
Power
3,000
2,500
2,000
1,500
1,000
500
0
Total system cost [€ Bn/y]
2024*
Data for Europe (incl. UK, NO, CH)
* 2024 values partially based on Compass-Lexecon estimation of current costs
** Power excludes methane & hydrogen fuels used for power generation, which are reported under ‘methane’ and ‘hydrogen’
2050
DE
ELEC
GA
2040
DE
ELEC
GA
2036
DE
ELEC
GA
2,3002,300
2,5102,510
2,5402,540 2,5802,580
2,7902,7902,8302,830 2,8802,880
2,7302,730 2,7702,770
2,9202,920
145
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results144
4.7.2. ENERGY SYSTEM COSTS ONLY
This includes OPEX and CAPEX annualised to a WACC for:
◆ the power system;
◆ the other vectors (molecules).
Figure 4-41 zooms in on the cost of the energy system only (i.e.
excluding end uses). It can be seen that the cost of the system
does increase even though final and primary energy demand
decreases in general for all scenarios (see previous sections). Sev
-
eral trends can explain the cost evolution:
◆ As explained in Section 3.1.3., electricity demand increases
by 75-95% by 2050, depending on the scenario. Most of the
increase is met by new RES sources, mainly via solar PV and
wind which take on the main share of required investments.
At the same time, new flexible assets such as large-scale bat-
teries and molecule-fired plants are required for adequacy
reasons;
◆ Even though the consumption of molecules decreases in all
scenarios, the costs remain rather stable over time. This is
explained by the fact that these molecules shift from rela-
tively cheap fossil fuels to more expensive molecules such as
ammonia, methanol, biomethane, etc.;
◆ CCS/U is a relatively costly but necessary technology required
to reach European climate targets for which the cost increases
over time;
TOTAL SYSTEM COSTS FOR THE ENERGY SYSTEM IN EUROPE ANNUAL FIGURES FIGURE 441
CCS/U
Molecules
Power**
+
1,200
1,000
800
600
400
200
0
Total system cost [€ Bn/y]
2024*
Data for Europe (incl. UK, NO, CH)
* 2024 values partially based on Compass-Lexecon estimation of current costs
** Power excludes methane & hydrogen fuels used for power generation, which are reported under molecules.
2050
DE
ELEC
GA
2040
DE
ELEC
GA
2036
DE
ELEC
GA
760760 760760 770770 810810
900900 910910
950950
870870 890890
940940
In general it can be observed that the more electrified scenarios
show lower total system costs for the energy subsystem. While
these scenarios require higher spendings in the power sector this
is more than compensated for by the lower spending elsewhere
in the system resulting from the overall lower molecule demand.
In 2050 the difference in Energy system costs between the most
electrified (ELEC) and most molecule-focused (GA) scenarios
amounts to 70 Beur per year. Comparing this to a reduction in
total system costs between these scenarios (see figure 4-40) of
190 BEur it can be observed that the difference in costs in the
Energy sector amounts for more than one third of the reduction.
The other two thirds in reduction result from the reduction in
end-uses costs.
4.7.3. ZOOMING INTO THE POWER SYSTEM COSTS
This includes OPEX and CAPEX annualised to a WACC for:
◆ the power system (including molecules required to run
powerplants).
Figure 4-42 zooms in specifically on the power sector. It can be
seen that CAPEX investments in generation assets will make
up the bulk of power system costs. Even though fuel costs are
strongly reduced due to the decrease in power generation via
molecule-fired power plants (see Section 4.6.), other OPEX costs
(such as for generation assets) do increase. As explained in Section
4.5, the optimisation leads to an expanded onshore & offshore
transmission grid which, however in terms of total power system
costs, these investments remain limited as compared to other
cost components. The main grid expansion costs are located at
the distribution level, mainly due to the electrification of heating
and vehicles at the DSO-level.
TOTAL COSTS FOR THE POWER SYSTEM IN EUROPE ANNUAL FIGURES FIGURE 442
Other
Distribution
Transmission
OPEX & fuel**
CAPEX - Assets
+
600
500
400
300
200
100
0
Total system cost [€ Bn/y]
2024*
* 2024 values partially based on Compass-Lexecon estimation of current costs
** Including methane and hydrogen used for power generation
2050
DE
ELEC
GA
2040
DE
ELEC
GA
2036
DE
ELEC
GA
370370
500500
470470460460
530530 510510
480480
470470450450440440
Comparing the different scenarios it can be observed that the
higher degree of electrification, the more costs are located in
the (electric) power system. As observed in figure 4-40 these
additional costs are more than compensated for by a reduction
in costs in other sectors of the system. The key differentiator for
power system costs are the OPEX & fuel costs. These costs are
strongly related to the use of molecules in electricity generation
which could be further reduced by the integration of additional
flexibility and/or renewable generation in the electricity system.
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Multi-energy European results146
4.8. KEY TAKEAWAYS
As we approach 2050, a significant amount of uncertainty remains
concerning both the supply and demand of each energy vector.
As is currently the case, the demand scenarios to be pursued
towards 2050 for the energy system should be jointly defined
across all energy vectors. However, due to the decreasing interplay
between the electricity and molecule systems in the future (in
share of energy), the design of the electricity system (i.e. supply
and infrastructure) can be largely separated from the molecule
system. Additionally, the points of intersection between the dif-
ferent vectors should be jointly evaluated in terms of location (as
is the case today for power plants).
The question of whether the supply of molecules will be imported
from outside of Europe or produced domestically is a significant
uncertainty. Based on the sensitivities and simulations carried out,
it seems that this will hinge on the prices of imported synthetic
molecules and the economic viability of electrolysers (linked to
the available surplus of low-carbon supply within Europe). Fac-
tors like energy dependency and other geopolitical consider-
ations could also influence this decision, potentially leading to
an increase in domestic production. The simulations reveal that
there is still untapped offshore potential that could be harnessed
for this purpose once the electricity supply has been sufficiently
decarbonised.
It is therefore important to note that without a surplus of elec-
tricity in certain areas, producing hydrogen via electricity in these
areas may not make economic sense due to the fact that there is
insufficient electricity to meet the demand. Areas with electricity
surpluses should be prioritised in terms of the generation of green
molecules. This approach would not only benefit the system by
producing molecules at a lower cost, but could also avoid the
need to invest in additional electricity grid to import the energy
to be used for the production of green molecules.
Even in a future system which is dominated by renewable
energy, dispatchable generation will still be necessary. Although
the required thermal dispatchable capacity remains similar to
today's levels (despite the peak load capacity almost doubling),
the underlying assumption that several newly electrified appli-
ances are expected to be flexible and that more storage will be
installed is key. Greater flexibility (long-term storage, demand
flexibility) could further reduce the need for thermal generation
and may even help to avoid the need to build new units in the
short term. However existing installations are also getting older
and there will be a need to replace those as well. Similarly, new
nuclear generation units could provide the needed thermal gen-
eration (and also provide low-carbon energy) instead of building
new molecule-fired thermal generation units. The results show
that it is typically more beneficial to continue operating existing
units using methane and greening the supply of methane, rather
than switching to hydrogen, as this would require investing in new
infrastructure (power plants, grid). However, such conclusions
should be drawn on a country-by-country and location-specific
basis. Carbon capture and storage (CCS) for power generation is,
however, very limited.
The key takeaways for the Belgian electricity system are the
following:
◆ a potential third offshore zone in the Belgian EEZ (to reach
8 GW in total) is always selected by the optimiser; indeed,
with the zone being situated close to the coast and given the
undersupply of electricity in Belgium, this would be a very
appealing option from a financial point of view;
◆ additional onshore interconnectors are found to be optimal
for Belgium (when optimising those for European system
costs):
-
additional capacity with the Netherlands (HTLS reinforce-
ment);
-
additional capacity with Germany in the DE, ELEC scenarios;
- additional capacity via the offshore interconnectors devel-
oped.
◆ a very limited amount of electrolysers (mostly none) is found
to be optimally installed in Belgium. Up to 1 GW in the GA sce-
nario and up to 2 GW in the high European NUC scenario; this
is an important outcome, since more electrolysers in Belgium
implies more consumption, while the country’s potential for
domestic RES generation is limited (see the next chapter for
more insights about this).
147
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
5.
RESULTS FOR
BELGIUM
5.1. Multi-energy results 149
5.2. Current policies and levers 158
5.3. Electricity demand 164
5.4. Electricity long-term supply options 171
5.5. Transition period (the road to 2050) 189
5.6. Summary of the different levers 194
5.7. Electricity grid 197
5.8. Other key insights 213
5.9. Key takeaways 218
149
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium148
5.1. MULTI-ENERGY RESULTS
The multi-energy vector results for Belgium are explored first.
Similar to the approach for Europe in the previous chapter, the
model is able to combine the supply and demand for each energy
vector. The results presented in this section include the main DE,
ELEC and GA demand scenarios in combination with the central
electricity supply scenario (no new nuclear, ‘central’ onshore RES
and optimised offshore wind). Note that ‘net’ imports and exports
are shown in the figures, meaning that the absolute imported
and exported values can differ. For more information about the
definition and interactions between the different energy vectors,
see Section 2.3.2.
5.1.1. YEARLY METHANE BALANCES
Today, methane is mainly consumed for the provision of heating
in buildings, industrial heat as a feedstock for the creation of
hydrogen (via SMR-H2) and for power generation in gas power
plants. With no natural resources of fossil gas and very limited
biomethane production (~0.1 TWh in 2021), nearly all methane
consumed originates from imports either via pipelines or via LNG.
The trends which emerge on the demand side are outlined below.
◆ In the lead-up to 2050, methane demand decreases, mainly
due to the phase-out of gas usage in end uses such as heat-
ing in buildings, and process heat in industry. This effect is
stronger in ELEC and DE than in GA (in that order). On the
other hand, methane for shipping purposes becomes rel-
atively important (up to 40 TWh in 2050), and becomes the
main source of methane demand for Belgium in 2050.
◆ In 2036, in the absence of nuclear generation and despite the
strong rollout of RES, methane remains important for power
generation in all scenarios, after which (due to the addition of
more RES) this need is reduced in 2050. These required vol-
umes strongly depend on the considered electricity supply
scenario in Belgium. Note that Figure 5-2 demonstrates the
results in a situation where no more nuclear capacity exists
in Belgium and offshore wind connected to Belgium is opti-
mised. For detailed results related to the required gas for
power volumes, see Section 5.4.
◆ The production of hydrogen via SMR-H2 still exists in 2036;
from 2040 onwards, this form of hydrogen production is
replaced by imports and – to a lesser extent – electrolysis (see
next section).
Looking at the supply side, the following changes can be
observed.
◆ It makes economic sense in all scenarios to deploy the full
biomethane potential (21 TWh by 2050); however, this remains
insufficient to meet all domestic demand, subject to the con-
sidered demand scenarios, even by 2050.
◆ Belgium will still need to import vast amounts of methane.
From 2040 onwards, piped methane could be sufficient to
meet demand. Note that this does not necessarily have to
imply fossil methane, as many neighbouring countries with
far more biomethane potential would be able to export this
to Belgium (see Section 4.1.1).
◆ The domestic production or import of synthetic methane
is not deemed economically viable in any of the scenarios
(including on a European scale; see Section 4.1.1).
ANNUAL SUPPLYDEMAND BALANCE FOR METHANE FIGURE 52
148
172
130
174
85
124
Supply Demand
2036 2040 20502021*
Figures presented under the electricity supply scenario assuming no new nuclear, central onshore RES and optimised
offshore wind
*Historical values based on EUROSTAT and the Hydrogen observatory of the European Commission [EUC-14] for SMR-H
2
** The source of pipeline imports is defined on European level and could be fossil, bio or synthetic. By 2050, this is
predominantly bio methane (see European results)
DE DE DE
ELEC ELEC ELEC
GA GA GA
Domestic bio
End use
SMR-H
2
Power
200
150
100
50
0
-50
-100
-150
-200
[TWh]
Import LNGImport LNG
Import pipeline**Import pipeline**
194194
144144
9494
186186
This chapter outlines the results for Belgium. The first part of this chapter explores
the multi-energy carrier results. The focus then shifts to the Belgian electricity system
(supply options, grid and other aspects of the system). It is important to note that
the simulations performed in this study start from 2036 onwards and are based on
the scenarios outlined in previous chapters. For an overview of the different options
simulated for Belgium, Figure 5-1 provides a summary of the different combinations of
sensitivities simulated in this study.
In terms of demand:
◆ 3 scenarios are simulated (DE, GA, ELEC) for all combinations
of RES, non-domestic offshore and nuclear;
◆ 2 additional sensitivities are defined. the 'SUFF' sensitivity is
used to quantify the impact of sufficiency measures on sys-
tem costs while the 'HEAT' sensitivity investigates the impact
of additional district heating in 2050.
In terms of supply options:
◆ For each target horizon, three domestic RES sensitivities are
performed:
- Central RES, which reflects the current policies;
- High RES, which increases the amount of PV and onshore
wind;
-
High RES and very high PV, where the installation rate of
PV is almost quadrupled compared to the central scenario.
◆ Secondly, the combination of non-domestic offshore (or ‘far-
out’ RES) and new nuclear options is simulated according to
the potentials assumed for each target year in this study. For
each target year, several combinations covering the range of
options are simulated.
◆ In addition, for the intermediary period, three sensitivities (for
each of the combinations of non-domestic offshore and new
nuclear) are simulated. These cover the extension of Doel 4 /
Tihange 3 beyond 2036, and an additional 1 or 2 GW on top.
◆ Other sensitivities related to adequacy and flexibility are also
outlined in the study.
Given the large amount of combinations and sensitivities (over
300), the most relevant are depicted in Figure 5-1, with non-do-
mestic offshore wind capacity along the x-axis and new nuclear
capacity along the y-axis.
OVERVIEW OF DIFFERENT OPTIONS SIMULATED FOR BELGIUM FIGURE 51
EXT. EXT.EXT.
HRES HRESHRES HRES
HPV HPVHPV HPV
SUFF SUFFSUFF SUFF
HEAT
far RES
Sensitivities
2036 20402040 2050
0 4 8 12 16 0 4 8 12 160 4 8 12 16 0 4 8 12 16
8
6
4
2
0
8
6
4
2
0
8
6
4
2
0
8
6
4
2
0
GW GWGW GW
GW GWGW GW
Simulated combinations
60 TWh 60 TWh 60 TWh
90 TWh 120 TWh
NEW NEWNEW NEW
30 TWh 30 TWh 30 TWh
2.1 GW 2.1 GW2.1 GW
Doel 4 (D4) & Tihange 3 (T3) nuclear
extension Doel 4 (D4) & Tihange 3 (T3) nuclear
extension
Doel 4 (D4) & Tihange 3 (T3) nuclear
extension
PV +7.2 GW
onshore wind with +1.2 GW PV +12 GW
onshore wind with +2 GW
PV +12 GW
onshore wind with +2 GW PV with +24 GW
onshore wind with +4 GW
PV +16.8 GW PV +28 GWPV +28 GW PV +56 GW
Sufficiency SufficiencySufficiency Sufficiency
Heating networks
Far-out baseload RES
3.1 GW 3.1 GW3.1 GW
D4/T3 + 1 GW additional extension D4/T3 + 1 GW additional extensionD4/T3 + 1 GW additional extension
4.1 GW 4.1 GW4.1 GW
D4/T3 + 2 GW additional extension D4/T3 + 2 GW additional extensionD4/T3 + 2 GW additional extension
151
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium150
5.1.3. YEARLY LIQUID BALANCES
Liquids – namely oil products such as gasoline, diesel, naphtha,
kerosene, bunkering fuels, etc. – make up the main source of the
final energy demand in Belgium. These products are used for
energy purposes such as transport fuels, heating in buildings,
and process heat in industry but are also used as non-energy
feedstock in the chemical sector.
Looking at the future changes on the demand side, the following
observations can be made:
◆ The demand for liquids in road transport and heating ('Other
energy' in Figure 5-4) is almost entirely phased out due to the
near complete phase-out of ICE vehicles and oil heaters.
◆ Liquids used for process heating are assumed to be phased
out completely by 2050 as well, being replaced by either gas-
eous molecules and direct electrification.
◆ International aviation, shipping and feedstock are a relatively
stable source of demand in the lead-up to 2050.
This can be explained by the fact that reducing net emissions of
these demands will still require liquid fuels. In the future these
could however be bio and or synthetic based fuels.
On the supply side the following changes can be observed:
◆ Belgium continues to be significantly reliant on the import
of liquids, there is an observable increase in the use of bio-
based and synthetic liquids at the European level, as detailed
in Section 4.1.3 of the report (not shown in this figure). As a
result, the country's dependence on imported liquids is grad-
ually shifting away from traditional oil and its derivatives. In
other words, while Belgium's relatively high dependency on
imported liquids remains unchanged, the nature of these
imports is becoming more diversified and less reliant on oil-
based products.
ANNUAL SUPPLYDEMAND BALANCE FOR LIQUIDS FIGURE 54
Other energy**
Shipping
Aviation
Feedstock
400
300
200
100
0
-100
-200
-300
-400
[TWh]
Demand
Supply
Domestic bio
Imported*Imported*
2036 2040 20502021
Figures presented under the electricity supply scenario assuming no new nuclear, central onshore RES and optimised offshore wind.
All values are expressed in refined energy products (defined in EUROSTAT as final energy demand)
* The source of imports is defined on European level and could be fossil, bio or synthetic (see European results)
** Includes the industry, residential, tertiary and domestic transport sector
Historical values based on EUROSTAT
DE DE DE
ELEC ELEC ELEC
GA GA GA
201
242
226225
343
207 219
174 178178 185185
5.1.2. YEARLY HYDROGEN BALANCES
It is estimated that in 2022, around 11 TWh of hydrogen were
consumed in Belgium [EUH-3], with the main source of demand
being the desulfurisation process in refineries (grouped under
End use in Figure 5-3) and the production of ammonia for fertil-
isers. The production of hydrogen generates a high amount of
emissions since 99% of the time it is made from fossil fuels, mainly
via steam methane reforming (SMR-H2), but also (to a much lesser
extent) as a by-product in some chemical processes.
The following observations can be drawn by looking at the future
changes on the demand side:
◆ Hydrogen is used to fulfill demand in end uses such as in the
steel sector, for refineries and (to a certain extent) to provide
industrial heat. In the GA scenario, there is also some signifi-
cant demand for H2 in transport.
◆ Hydrogen for power generation carries some importance for
fulfilling peak demands. This effect is stronger in the ELEC
and DE scenarios (respectively) and in later years, due to
the higher degree of electrification. These required volumes
strongly depend on the considered electricity supply scenario
in Belgium. Note that Figure 5-3 depicts the results in a situ-
ation where no more nuclear capacity exists in Belgium and
non-domestic offshore wind connected to Belgium is opti-
mised. For detailed results relating to the required gas for
power volumes, see Section 5.4.
◆ Ammonia and synthetic liquids produced from hydrogen do
not appear to be economically viable in Belgium; instead, the
model chooses to import these molecules (see liquids bal-
ance).
On the supply side:
◆ Belgium will have to import hydrogen, as the demand for H2
generally increases. In later years especially, it becomes more
economically attractive to import hydrogen (via pipelines or
in the form of ammonia) instead of using methane (via SMR-
H2) or electrolysis to produce H2.
◆ Electrolysis in Belgium is rather uneconomical, with some
installed capacity in the GA scenario (1 GW in 2050), mainly
due to the higher demand for hydrogen and lower demand
for electricity in that scenario.
◆ Due to the lower overall demand for hydrogen in Europe in the
ELEC scenario, there is more availability of electrolysis-based
hydrogen in Europe which can be imported via pipelines. In
the DE and GA scenarios, this availability of electrolysis-based
hydrogen is more limited, with Belgium needing to import
hydrogen from outside of Europe (via shipped ammonia).
ANNUAL SUPPLYDEMAND BALANCE FOR HYDROGEN FIGURE 53
2036 2040 20502022*
Figures presented under the electricity supply scenario assuming no new nuclear, central onshore RES and
optimised offshore wind
*Historical values based on the Hydrogen observatory of the European Commission [EUC-14].
DE DE DE
ELEC ELEC ELEC
GA GA GA
Demand
End use
Synth. liquids
Power
Ammonia
70
50
30
10
-10
-30
-50
-70
[TWh]
11 13
23
29
22
31
53
2828
39
65
Supply
Electrolyser
SMR-H
2
*
Import pipelineImport pipeline
Import ammoniaImport ammonia
153
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium152
5.1.5. PRIMARY ENERGY SUPPLY
The primary energy supply can be determined based on the
final energy demand, the demand for electricity generation, the
demand for conversions between energy vectors and imports.
In 2021, around 70% of Belgium’s primary energy supply came
from (imported) fossil fuels such as coal, oil and fossil methane. In
all scenarios, the need for primary energy decreases and renew-
ables are seen to meet more than 50% of its needs. Several key
drivers can explain this change, as outlined below:
◆ A general assumed decrease in final energy demand as
explained above.
◆ The role of electrification in reducing primary energy needs is
key due to the high inherent efficiency of most electrification
technologies. The net effect is a general decrease in primary
energy needs, especially when most of the electricity is sup-
plied via RES.
◆ The higher the amount of electrification met via RES such as
wind and solar PV, the lower the primary energy needs are,
since electricity generation via thermal sources requires more
energy input due to inherent energy losses.
◆ Most supply is made up of renewables such as solar PV and
wind, whereas biomass also has a key role to play both as a
direct fuel and as a feedstock for biomethane and liquids but
also for direct usage in certain sectors. Imported ammonia
(and hydrogen) also have a role to play in further decarbonis-
ing the sources used to meet primary energy needs.
Note that the primary energy supply depends on the considered
electricity supply scenario in Belgium; Figure 5-6 outlines the
results in a situation where no more nuclear capacity exists in
Belgium, under the central onshore RES scenario (see Section
3.2.3 ) and the offshore wind supply is optimised. For detailed
analyses relating to the impact of these electricity supply choices,
see Section 5.4.
PRIMARY ENERGY SUPPLY FOR BELGIUM FIGURE 56
2036 2040 20502021
Figures presented under the electricity supply scenario assuming no new nuclear, central onshore RES and optimised offshore wind
Includes international shipping & aviation and non-energetic feedstock demand
* Non-domestic offshore wind connected to Belgium counted as imports
**Source of imported methane & liquids is defined on European level and could be fossil, bio and/or synthetic (see EU results)
Historical values based on "EUROSTAT
DE DE DE
ELEC ELEC ELEC
GA GA GA
Hydro
Wind
Solar PV
Biomass
Nuclear
Liquids - import**
Methane - import**
Coal
Ammonia - import
Hydrogen - import
Electricity - import*
800
700
600
500
400
300
200
100
0
[TWh]
720
520
570 610
520
570
640
490
530
590
5.1.4. IMPORTS
In 2021, around 80% of Belgium’s primary energy needs were met
via imports, most of which were products based on fossil fuels
such as oil and methane which are used in industry, transport
and heating. In all scenarios, Belgium can reduce its required
energy imports by 2050 in absolute terms, yet it would remain
dependent on imports to meet between 63% and 68% of its pri-
mary energy needs. The following key drivers explain the changes
in fuel imports:
◆ A reduction in the final energy demand also reduces the need
for imports. As explained in Section 3.2.2.1, it is assumed that
Belgium will reduce its final energy demand by 25% to 45% by
2050, which is mainly driven by energy efficiency measures
and electrification.
◆ The electrification of end use does not only reduce its final
energy needs but shifts demand away from liquids and gas-
eous molecules that are mostly imported both today and in
the simulated future years. In general, as the level of electrifi-
cation increases (rising under the GA, DE, ELEC scenarios, in
that order) and the level of renewable production increases in
Belgium, so the requirement for imports decreases.
◆ The required need for thermal (backup) generation greatly
depends on the considered electricity demand scenario. On
the one hand, the increased production of green electricity
via solar PV, onshore and offshore wind reduces the need for
methane and/or hydrogen used in power plants. On the other
hand, the increased electricity demand increases this require-
ment, especially in the ELEC and DE scenarios. For detailed
results, see Section 5.4.
◆ The local production of (green) molecules remains relatively
limited in Belgium. The production of hydrogen and synthetic
methane and/or liquids is (nearly) non-existent in all scenar-
ios. It is only the production of biomethane which slightly
reduces the need for methane imports.
◆ In general, Belgium’s dependence on imports remains rela-
tively elevated. However, these imports would no longer arrive
in the form of fossil fuels such as coal, oil and fossil gas (not
specified in Figure 5-5), but would shift towards including
green molecules such as green ammonia, synthetic liquids
such as methanol and piped hydrogen in the ELEC scenario.
Belgium will also switch from being a net exporter in 2021 to
being a net importer of electricity in the coming years. However,
for the years under consideration, the extent to which this is true
varies greatly based on the electricity supply scenario considered.
Many different combinations of electricity demand, supply and
flexibility and their impact on electricity imports (amongst other
things) are analysed in detail in this Chapter.
TOTAL IMPORTS FOR BELGIUM FIGURE 55
63% 62%62%
66%
72% 73%
71%
75% 75% 75%
700
600
500
400
300
200
100
0
[TWh]
2036 2040 20502021
DE DE DE
ELEC ELEC ELEC
GA GA GA
Ammonia
Liquids***
Coal
Hydrogen
Methane***
Import dependence*
79%
570
390
370
310
430 410
330
460 470
390
Figures presented under the electricity supply scenario assuming no new nuclear, central onshore RES and optimised offshore wind
Includes international shipping & aviation
* Import dependence expressed as percentage of primary energy demand
** Electricity imports depend greatly on the assumed Belgian electricity supply scenario. For a more complete view on the impact on
imports/exports of electricity, see Figures 5-11, 5-21 and 5-22. In this figure, non-domestic offshore wind connected to Belgium is counted
as imports
*** Source of imported methane & liquids is defined on European level and could be fossil, bio and/or synthetic (see EU results)
Historical values based on EUROSTAT
Electricity**
155
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium154
Figure 5-8 focuses on the carbon capture (CC) volumes in Bel-
gium. As it demonstrates, a decent amount of CC would be eco-
nomically viable by 2036, especially for the capturing of process
emissions in the steel and chemical sectors. From 2040 onwards,
CC in the mineral industry (mainly cement) and combus-
tion-based emissions would make economic sense. The decline
in captured emissions between 2040 and 2050 mainly stems
from the decreased combustion of fuels containing carbon. In the
ELEC scenario, there is also less methane combustion in industry
compared with the DE and GA scenarios, explaining why there
is less carbon captured in the ELEC scenario compared with the
DE and GA scenarios (in which more industrial processes are still
fueled by methane combustion).
The model deems the domestic production of synthetic fuels,
which combines hydrogen with captured CO2, to be econom-
ically viable in Belgium in none of the scenarios. This means
that the captured volumes of CO2 in Belgium would need to be
transported to other countries, either for underground storage
or for usage as feedstock for the creation of synthetic fuels and/
or chemical products. This transport requirement is out of scope
of the model used within this study.
CARBON CAPTURE VOLUMES PER SECTOR IN BELGIUM FIGURE 58
20
15
10
5
0
[MtCO2/y]
2036 2040 2050
*Excluding SMR, accounted under ‘SMR-H
2
’
DE DE DE
ELEC ELEC ELEC
GA GA GA
Metals
Chemical*
Mineral
Combustion
SMR-H
2
Industrial
process
emissions
11
8
16
10
16
19
11
13
1515
5.1.6. GHG EMISSIONS AND THEIR MANAGEMENT
The total emissions considered within the scope of this study (i.e.
including international aviation and 50% of international ship
-
ping) amount to around 121 MtCO
2
-eq/y. The change in emissions
per sector and scenario is depicted in Figure 5-7 for both historical
and simulated years.
It is clear that energy emissions (i.e. from the combustion of fuels)
are almost fully phased out by 2050. The remaining energy emis-
sions mainly stem from the international aviation and shipping
sectors. Emissions such as those from (industrial) processes and
other GHGs such as CH4, NO2 and F-gases constitute around
25% of total emissions in 2022, but amount to close to 50% of the
remaining emissions in 2050. These are typically hard to abate
even after the application of new technologies or fuel switching
and will also require abatement in the form of CCS and/or LULUCF.
Focusing more specifically on the changes per sector, the power
sector is almost fully decarbonised by 2050. By 2050, the buildings
and domestic transport sectors are also nearly entirely carbon
free. This is mainly driven by fossil fuels being replaced by elec-
trification (particularly in the ELEC scenario) and because the
remaining consumption of gases and liquids will have mainly
switched to bio and synthetic alternatives. In industry, some emis-
sions persist in 2050. This can be attributed to process emissions
that do not depend on the type of fuel used. However, a lot of
this remaining CO2 is compensated by carbon capture (CC) (see
next section). International transport, which already generated
a relatively important share of Belgian emissions in 2022 com-
pared with other European countries, looks to be the main form
of remaining energy emissions by 2050, as not (all) fuel usages
are switched to green alternatives such as methanol. Ultimately
carbon abatement via carbon capture (CC) and a small amount
of LULUCF helps to lower overall net-emissions.
As can be seen, a net-zero Europe does not necessarily mean
that Belgium needs to reach net zero for its domestic emissions.
Depending on the scenario, Belgium could still emit between
11 MtCO2 and 14 MtCO2 in 2050, implying a relative decrease of
91% to 93% compared with 1990 levels (153 MtCO
2
). The ELEC
scenario generally reaches the highest level of decarbonisation
in all simulated years.
TOTAL GHG EMISSIONS IN BELGIUM PER SECTOR FIGURE 57
150
130
110
90
70
50
30
10
-10
-30
[MtCO2/y]
'
LULUCF
DAC
Carbon Capture - BE
Int shipping*
Net emission
Int aviation
Non-CO
2
Domestic transport
Buildings
Industry - energy**
Power&Heat
Industry - processes
Carbon Capture - EU***
Positve Netgative
2010 2020
Figures presented under the electricity supply scenario assuming no new nuclear, central onshore RES and optimised offshore
The sectoral split concerns CO
2
emissions, the non-CO
2
emissions are shown separately in aggregate
The European weighted average CO
2
intensity is assumed for methane and liquids imported and consumed in Belgium
* Includes 50% of the emissions
** Also includes refineries, agriculture and waste management
*** This includes the CO
2
which was captured within Europe to make synthetic fuels, combusted within Belgium. As such this would
mean a net-zero emission for BE
Historical values based on European Environment Agency
2050
DE
ELEC
GA
2040
DE
ELEC
GA
2036
DE
ELEC
GA
157
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium156
SANKEY DIAGRAMS FOR BELGIUM FOR 2050. FOR CURRENT POLICIES AMBITIONS
CENTRAL RES, 0 GW FOREIGN OFFSHORE INTEGRATED IN BE, 0 GW NUCLEAR. FIGURE 510
2021
2050 DE
2050 GA
In conclusion, this section presents the coupling between the electricity sector and the methane/hydrogen sector in the per-
formed simulations. The coupling of these systems decreases over time in all scenarios. Therefore, it can be argued that the
infrastructure design of electron and molecule systems can be decoupled in Belgium. However, attention should be paid to
the location of the interactions (power plants and electrolysers).
5.1.7. LINK BETWEEN THE MOLECULE AND ELECTRICITY SYSTEMS
Both molecules and electrons are crucial components of the
future energy system. One way to explore the coupling between
the electricity system on the one hand and the methane and
hydrogen system on the other hand is to evaluate the amount
of energy that is exchanged between both systems. Dividing
the total amount of electricity generated using methane and
hydrogen by the electricity demand quantifies the coupling in
the direction from molecules to electrons. To estimate the inter-
action in the other direction, from electrons to molecules, the
energy consumed by electrolysers is divided by the total elec-
tricity demand.
Figure 5-9 depicts these couplings for 2021 and for the target
years assessed in this study. For each of the three main demand
scenarios, all Belgian supply sensitivities are assessed. The sen-
sitivities with the highest total coupling (sum of the coupling in
both directions) and lowest total coupling are presented in the
figure. Several observations can be drawn:
◆ The coupling decreases over time; starting from 27% in 2021,
the total coupling drops to a value of between 20% and 13%
by 2036. This trend continues in the lead-up to 2050, with
the coupling decreasing by 5 to 10 percentage points in each
demand scenario. The main driver of this reduction is the
decreasing use of gas for the generation of power.
◆ Electrolysis occurs in Belgium in the GA scenario only. The
amount of electrolysis is very limited for each of the target
years.
◆ The coupling between the electricity and molecule systems
is for a given supply, on average, the highest in the ELEC sce-
nario. Although more molecules are used in the electricity
system, the electrification assumptions used in this scenario
result in a more efficient use of primary energy and as such
the total primary energy demand for molecules is lower (see
also Section 5.1).
◆ While not depicted in the figure, the sensitivities with the
lowest level of coupling are found to be those with the most
renewable and/or nuclear domestic supply in their assump-
tions. In these sensitivities, the dispatch prefers the use of
these sources as they have lower marginal costs than meth-
ane and hydrogen generation. Conversely, the sensitivities
with the lowest renewable and/or nuclear domestic supplies
rely more on molecules for electricity generation.
COUPLING BETWEEN THE ELECTRICITY AND METHANE SYSTEMS FIGURE 59
ELEC
ELEC
ELEC
DEDEDE GAGAGA
2036
Supply sensitivity with
highest total coupling
Gas (CH
4
and H
2
)
Electrolysis
Supply sensitivity with
lowest total coupling
2021 2040 2050
30%
25%
20%
15%
10%
5%
0%
-5%
-10%
Annual production/consumption relative to
electricity demand [%]
27%
15%
11%
6%
11% 13%
18%
11% 9%
3% 3%
12%
15%
6%
0%0%0%0%0%0%0%
-2% -1%
0%0%0%0%0%
-2% -1%
-3%
-2%
7%
18%
12%
20%
16%
In addition to the previous analysis, the Sankey diagram (Fig-
ure 5-10) depicts the interactions and transformation processes
between energy vectors. The left-hand side of the figure depicts
the primary supply entering Belgium, whilst the right-hand side
depicts the final demand for each vector.
The figure demonstrates that the only interaction between mol-
ecules and electricity in 2021 was the power generation from
methane. This is still the case at the time of writing (September
2024). In the future, hydrogen could also be used for electricity
generation, but overall, the interaction is shown to decrease (in
varying proportions depending on the scenario). Additionally,
some electricity could be used to produce hydrogen. However,
as determined through various sensitivities and results (see also
Figure 5-8), only a very limited amount of electricity (or none at
all) would be used for this in Belgium from a European cost-op-
timal perspective.
While the energy interactions are limited, the capacity (e.g. the
location of power plants or electrolysers) will need to be accounted
for when designing the future electricity grid, just as new power
plants are considered when they are connected to the grid today.
159
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium158
Several observations arise from the Current Policies simulations, as outlined below:
◆ Belgium's net imports are projected to be around 40 to 50
TWh in 2036, increasing to approximately 60 TWh in 2040 and
2050. This is a significant rise when compared with recent
years, during which imports have largely equalled exports
(barring a few exceptions). The increase in imports over time
correlates with a surge in consumption due to electrification,
which outpaces the moderate rise in low-carbon generation
in Belgium under current policy supply evolutions. Further-
more, a noteworthy increase in the quantity of exchanges is
observed in all scenarios, escalating from around 30 TWh to
over 100 TWh in 2050.
◆ The new dispatchable capacity required (on top of the
already assumed existing units that would remain in 2050
and flexibility options) lies between 1,000 MW and 3,000 MW
in 2036 and goes up in 2050 to reach between 8,000 MW and
10,000 MW. This capacity is needed to maintain the adequacy
of the system, despite the increase in flexibility (storage,
demand flexibility) assumed in the Current Policies scenario.
The total thermal capacity required in 2050 lies between 13
GW and 15 GW, on top of the more than 20 GW of flexibility
options assumed in the scenarios.
◆ The dispatchable thermal generation generates around
20-30 TWh electricity from molecules in 2036 and 15-30 TWh
in 2040 and 2050. This amount decreases over time in most
scenarios (when compared to the actuals) alongside the
increase in renewables abroad. However, a certain volume
remains due to the increase in the thermal capacity which is
required and in order to cope with hours during which there
is low RES infeed.
◆ The amount of hours during which curtailment occurs
(due to excess energy) remains limited (<50 hours in all sce-
narios) and the amount of hours with low prices (<€20/MWh)
decreases between 2036 (over 1,000 hours) and 2050 (less
than 1,000 hours). These are hours of curtailment in a ‘per-
fect foresight’ situation where all flexibility and storage is opti-
mally dispatched. The low amount of hours can be explained
by the relatively limited uptake of renewables compared to
the consumption of electricity. The assumed uptake in flexi-
bility (DSM and storage) allows curtailment to be mitigated in
perfect foresight.
Pursuing current policies without additional measures would therefore result in the need for up to 8,000-10,000 MW new thermal
capacity to keep the system adequate and could result in up to 50-60 TWh of net electricity imports and up to 20-30 TWh of electricity
generated from molecules in 2050. The amount of RES curtailment would remain limited, however the amount of hours with low
marginal prices would increase due to the increase of RES abroad.
5.2. CURRENT POLICIES AND LEVERS
Building on the findings outlined in previous sections, the elec-
tricity system impact assessment can be decoupled from the
molecule one. This conclusion is based notably on:
◆ the limited decreasing interactions between the electricity
and molecule systems in the future for Belgium;
◆ the limited impact of European sensitivities on the Belgian
electricity system (similar need for grid, electrolysers, etc.).
The following sections will focus on the electricity system in Bel-
gium. The interactions with the other parts of the system will
be accounted for in the dispatch model (amount of molecules
produced or consumed) and their costs will also be assessed.
The Current Policies scenario starts from the approved policies
and current ambitions. Section 3.2 includes more details about
these. These scenarios were then further quantified through
the optimisation process such that an European optimum was
reached. These scenarios then form the basis for the results in this
chapter. In all scenarios, they include from a supply point of view:
◆ the increase in domestic offshore capacity to 8 GW from
2040 onwards;
◆ the extension of Doel 4 / Tihange 3 until 2035 and no new
nuclear;
◆ no non-domestic offshore wind accounted for as Belgian
supply;
◆ the Central RES scenario for domestic RES (onshore wind,
PV, biomass, hydro and offshore wind in Belgium (or the Bel-
gian EEZ);
◆ the Belgian reliability standard is met (hence each sensi-
tivity will always ensure sufficient capacity to meet adequacy
requirements);
◆ the development of flexibility (storage, demand flexibility),
defined ex ante and stable across the same consumption sce-
narios; this amounts to >15 GW in 2040 and >20 GW in 2050
across the different demand scenarios;
◆ additional onshore cross-border grid capacity found as
‘optimal’ in the European optimisation (see section 4.5.3) in
addition to the approved projects in the latest federal devel-
opment plan (see Section 3.2.5):
-
Reinforcement of the border with the Netherlands, amount-
ing to 1 GW;
-
Reinforcement of the Germany-Luxembourg border,
amounting to between 1 and 4 GW.
5.2.1. RESULTS IN THE ‘CURRENT POLICIES’ SCENARIO
The results related to current policies and trends regarding RES
deployment are depicted in Figure 5-11.
The figure includes the need for thermal capacity, the domestic
electricity mix, imports/exports and indicators of ‘oversupply’
through means of ‘number of hours of curtailment’ and ‘low
marginal prices’. This last indicator provides an indication of the
number of hours during which almost solely RES generation
is running in the model. The figure depicts the three demand
scenarios (DE, GA and ELEC) in a situation where the Central
domestic RES scenario is accounted for and no new nuclear nor
new non-domestic offshore supply for Belgium is considered. The
first row provides an indication of the current system parameters.
ELECTRICITY MIX DASHBOARD FOR CURRENT POLICIES FIGURE 511
(Current policies)
(Current policies)
(Current policies)
Capacity per scenario
Actuals
Nuclear
5.9 GW 0 GW Central
0 GW 0 GW Central
0 GW 0 GW Central
0 GW 0 GW Central
Foreign
offshore DRES Domestic + foreign offshore Exports [-] / Imports [+]
Share hours per yr.
with curtailment
Share hours per yr.
Below 20 €/MWh
Thermal
in GW
’20* 10 20 25 -15 15 0.1-2% 5-8%
-23 67 0.5% 14%
-28 90 0.0% 9%
-34 92 0.2% 10%
-21 65 0.8% 17%
-19 82 0.1% 9%
-19 83 0.3% 10%
-28 78 0.4% 16%
-29 86 0.0% 7%
-31 86 0.1% 8%
39
’36 965
65
65
82
82
82
103
103
103
324
19
27
21
18
31
22
17
30
’36 9 1
’36 9 3
’40 7 6
’40 7 5
’40 7 7
’50 5 9
’50 5 8
’50 510
DE
GA
ELEC
Supply mix in TWh Curtailment Low marginal
prices
Thermal-Existing Thermal-New Foreign offshoreNuclear DRES Thermal Gap-to-max Net imports Gross exports Gross imports
* Based on 5-year average for period 2018-2022 (and range for low price KPI excludes outlier in 2020 to exclude COVID effect)
** Thermal includes biomass, methane and hydrogen fired turbines
161
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium160
5.2.2. DOMESTIC LOW-CARBON SUPPLY AND EXPECTED DEMAND
Figure 5-13 combines the Current Policies scenario with the past
and expected evolution of electricity consumption and genera-
tion in the lead-up to 2050. Electricity consumption (including
CCS, electrolysers and losses) is assumed to more than double
from ~80 TWh today to a bit less than 180 TWh up to almost 200
TWh by 2050. This consumption includes the resulting need for
electricity for electrolysis (if any), grid losses and CCS. Until 2030,
current policies cause electricity generation to follow consump-
tion (though short by ~10 TWh) through renewable growth, the
operational extension of nuclear plants and fossil gas-fired ther-
mal capacity.
From 2036 onwards, which is the primary focus of this study,
Figure 5-13 shows the expected electricity consumption under
various demand scenarios (DE, GA, ELEC) versus the electricity
generation that follows current policies which are in place (i.e.
the renewables due to be installed in line with current trends
and ambitions). Molecule-fired generation is not shown as the
exact generation will depend on the other choices. The following
observations can be made:
◆ A difference of 50-60 TWh in 2036 between consumption
and generation - From 2036 onwards, there is a need for
about 50 to 60 TWh that has to be met by imports or addi-
tional measures, on top of current policies related to renew-
ables.
◆ A difference of 70-90 TWh in 2050 - This need grows over
time with the increasing demand for electricity, but is par-
tially offset by more renewables installed, reaching between
70 and 90 TWh in 2050.
◆ Options to meet this need can include:
- reducing demand – Notably through sufficiency measures,
which can yield a similar outcome whilst employing less
energy-intense methods, by (for example) encouraging ren-
ovation and the recycling of building materials;
-
increasing supply - This can be achieved by increasing
domestic electricity generation or by importing more from
abroad; the next subsection will explore possible supply
levers in more detail.
CHANGES IN THE ELECTRICITY DEMAND AND SUPPLY UNDER CURRENT POLICIES FIGURE 513
* For year X, the 5-year average in the range [X-2,X+2] is shown instead
** Approved policies: Extension of offshore wind in Belgium to 5,8 GW, extension of D4/T3 for 10 years, National/Regional energy climate plans (domestic RES,
electrification, energy efficiency...), CRM
’00* ’05* ’10* ’15* ’20* ’25 ’30 ’36 ’40 ’45 ’50
Focus of the studyHistorical & approved** policies
Historical 5-year averages Result of approved
policies Accounting for the Central scenario
of domestic RES
220
200
180
160
140
120
100
80
60
40
20
0
[TWh]
Thermal
Solar
Wind onshore
Domestic offshore wind
Biomass & Hydro
Existing nuclear
47 47 46 38 39
15 15
6
6 6
6 6 6 6
6
8
4
4
32 34 33 25
25
27 15
23
29
35
41
30
917
19 27 29 29
10 12
16
19 22 25
70-80
70-90
ELEC
DE
GA
50-60
70-80
BOX 5-1
ADEQUACY AND ELECTRICITY SUPPLY ARE NOT THE SAME THING
It is important to note that adequacy and electricity sup-
ply are two different concepts that are linked but sepa-
rate. This study focuses primarily on the electricity mix’s
contribution to maintaining Belgium’s adequacy. Import-
ing a lot of electricity does not mean that the country can-
not maintain the adequacy of its electricity system.
Adequacy is the system’s ability to meet the demand for
electricity. It is measured via different metrics such as the
loss of load expectation (LOLE) and energy not served
(ENS). This implies that there is enough installed capac-
ity across the system in order to cope with consumption.
A system’s adequacy is calculated by accounting for all
types of electricity resources (including imports) and the
ability of other countries to supply the system with elec-
tricity during periods of ‘stress’.
The energy output from the electricity installations,
which is determined by a European economic dispatch,
depends on their availability and variable costs in relation
to other technologies within Belgium and beyond. This
implies that a country could import a significant amount
of electricity annually (due to the economic dispatch from
both domestic and international generation), yet still pre-
serve its adequacy by retaining sufficient capacity within
its borders.
Figure 5-12 illustrates three important points:
◆ the lead times related to the deployment of different
technologies are very different; however, it is important
to note that some technologies are also bound by a
maximum amount that can be installed per year due to
supply chain/workforce constraints;
◆ the indicative contributions of different technologies
to adequacy expressed as a percentage of their nom-
inal capacity;
◆ the energy load factor range (or energy contribution)
of each technology; the load factor depends on the
European economic dispatch for some technologies.
Figure 5-12 also demonstrates that some technologies
mainly contribute to adequacy, while others contribute to
adequacy in a more limited way. Generally, technologies
contributing to electricity supply take longer to develop
than those related to adequacy. This is an important fac-
tor to consider when the core question is related to meet-
ing an energy supply need.
ADEQUACY AND ELECTRICITY SUPPLY ARE TWO DIFFERENT CONCEPTS FIGURE 512
1 year 5 years >10 years
Adequacy
contribution
Energy
contribution
Lead time:
With current
policies in 2050:
Adequacy GAP:
~ 6-10 GW
Energy supply
need:
~ 70-90 TWh
Solar
1%
10%
20-30%
10-50%
40-50%
80%
0% 0%
30-70% 30-70%
5-10% 10-15%
90% 80-90%
Flexibility Storage Onshore
wind Thermal
gas Offshore
wind Nuclear
Flex
All timings are indicative and project specific (size, type, complexity). Those are provide for one project.
Deratings are also indicative and depend on the location, type and the resulting electricity mix in Belgium and abroad.
163
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium162
Figure 5-15 quantifies the maximum potential energy contribu-
tion of each lever (for some split into subcomponents) to closing
the supply need, i.e. on top of what the Current Policies scenario
includes. Per lever, this comes down to:
Supply levers: potential on top of current policies
1. Onshore wind and solar PV
- Onshore wind: going from the Central to the High scenario
(+12 TWh in 2050).
-
Solar PV: from Central to High (+24 TWh) or Very High
(+41 TWh) in 2050. Note that for PV, not all installed capac-
ity will be capable of being evacuated. This will be further
assessed in the simulation results and is already accounted
in these figures.
2. Offshore wind
-
Domestic: all future potential (repowering and potential addi-
tional zone(s)) is already included in Current Policies from
2040 onwards: a total of 8 GW is considered.
-
Non-domestic offshore: 4 GW could be installed every 5 years
from 2036 onwards to reach 16 GW in 2050.
3. Nuclear
-
Lifetime extensions: further extension of 2 GW, 3 GW or 4
GW after 2036.
- New builds: up to 8.2 GW considered in 2050, with 2 GW in
2040.
4.
Imports/exports: Unlike other levers, these are not set in
advance, but are rather the outcome of choices made for the
other levers and the European set-up.
5.
Thermal: Whilst the capacities are set to guarantee adequacy,
the actual dispatch depends on the other levers and the sit-
uation abroad; Figure 5-15 shows a range across simulated
outcomes for different combinations of levers. In 2050, this
could amount to between 5 TWh and 30 TWh.
6. Far-out baseload RES: Far-out RES is considered as a sensi-
tivity on an ad-hoc basis; this would require building inter-
connectors to e.g. Northern Africa, like the UK-Morocco Power
Project is aiming to do. BOX 5-4 provides more details about
the sensitivity.
Demand levers: potential on top of current policies
1.
Sufficiency (demand reduction through behavioural
changes, modal shifts…): since the range of the need only
takes into account the base demand scenarios (DE, GA, ELEC),
moving towards the sufficiency demand scenario represents
an additional lever.
2.
District heating: although not depicted in the figure, a sensi-
tivity analysis is conducted with an increased focus on district
heating. This could potentially reduce the need for additional
supply for the generation of electricity.
OVERVIEW OF QUANTITIES FOR EACH LEVER FIGURE 515
~ 70-90 TWh
supply need 2050
On top of central
domestic RES supply
Range over the demand
scenarios (DE, GA and ELEC)
Maximum for each lever on each time horizon [TWh]
Onshore wind
x2 installation rate
Nuclear
Extend existing
units
Nuclear
Build new units
Solar PV2
x2-x4 installation rate
Molecule-fired
generation
Sufficiency levers
Lower
consumption
Outcome of other
choices
Analysed on ad-hoc
basis
Outcome of the
European dispatch
and type of mol-fired
generation installed
Range observed in the
simulations
Imports
Non-domestic
baseload RES
Build interco & RES
Non-domestic offshore
1
Build interconnectors
& offshore wind
’36 ’40 ’45 ’50 ’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45 ’50
’36 ’40 ’45
’36 ’40 ’45 ’50
3
Range
5912
7 7 7
7 7 7
15
15 15
15
15
15
15
15 15 612 18 24
15
50
70 70 70
10 85
6
13
15
17
15
10
60
TWh
80
TWh 80
TWh
90
TWh
14 GW
10 GW
16 GW
12 GW
8 GW
4 GW
8 GW
6 GW
4 GW
2 GW
98 GW
66 GW
42 GW
4 GW
0,5 GW
4 GW
3 GW
2 GW
+4 GW
+1,5 GW
+3 GW
+3 GW
+4 GW
+4 GW
10 10 20
20 23 25
15 15
15 15
15 15
15 15 18 20 22 22
1. Non-domestic offshore refers to offshore wind capacity installed outside of Belgium’s Exclusive Economic Zone (EEZ) which still counts towards Belgium’s
domestic supply given Belgian financing/support of the wind generation itself.
2. Note that the capacity factor in the highest solar PV scenario is lower, because some PV capacity is curtailed when generation exceeds the limits of what the distribution
network can handle.
5.2.3. OVERVIEW OF THE DIFFERENT OPTIONS TO COMPLEMENT
BELGIUM’S SUPPLY
Even under the most conservative demand scenario, a supply
need of 50-70 TWh remains when considering current policies.
To close this need, additional supply is required on top of current
policies. This study assesses 5 supply levers which are depicted in
Figure 5-14 in addition to sufficiency as a demand lever; these are:
1.
Onshore wind farms and solar PV – These are deployed
as relatively small, decentralised projects that are typically
connected to lower voltage levels and are mostly a regional
matter.
2.
Nuclear plants – Traditionally, these are very large-scale
projects connected to higher voltage levels and require a
high degree of government coordination at the federal level.
There are two main ways to increase their capacity versus the
Current Policies scenario: extend the lifetime of additional
existing reactors and/or build new ones.
3.
Non-domestic offshore wind farms – Much like nuclear,
these are very large-scale projects that require a high degree
of government coordination at the federal level. Since addi-
tional domestic potential is limited (and is already fully used
in the Current Policies scenario), activating these levers means
developing non-domestic wind farms in the North Sea or
elsewhere, connected through (hybrid) transmission systems.
This includes the potential financing/support of these wind
farms in order for them to be counted towards Belgian supply.
4.
Imports/Exports – This is a consequence of the choices made
for the other levers and a result of the European dispatch.
However, the level of dependence on foreign supplies should
be actively chosen by the government, given its significant
implications.
5. Thermal – Molecule-based thermal generation can produce
power in Belgium, which would imply importing the needed
molecules; however, as these thermal generation methods
have high variable costs (compared to other technologies in
the European system), their running hours as an outcome of
the European dispatch would be limited (see Section 4.6.2.
for more details).
Note that this list is not exhaustive: it focuses on levers that can
contribute substantial amounts of energy to the supply need.
Other technologies are also key for integrating renewables and
managing the system. These other technologies include batteries
and other storage options which provide temporal flexibility and
grids which provide spatial flexibility by connecting generation
and loads across the whole continent at different voltage levels.
The concept of far-out baseload RES combines both spatial and
temporal flexibility. Given Europe’s high demand for low-carbon
energy, some projects which are being investigated involve the
building of interconnectors to renewable development in other
continents, such as solar and wind in Northern Africa. These far-out
RES are beneficial in that they can potentially have higher capacity
factors and draughts that are not correlated with domestic ones.
Furthermore, one of the most notable examples, the Morocco-UK
Power Project by Xlinks, considers combining this with significant
storage capacity in Morocco. By using its temporal flexibility, it
could offer a more baseload-like profile for the UK's consumption.
OVERVIEW OF DEMAND, SUPPLY, ADEQUACY AND FLEXIBILITY LEVERS FIGURE 514
Electricity demand Energy supply options Adequacy & Flexibility
> Main drivers > Small-scale > Decentralised temporal
> Centralised temporal
> Spatial adequacy & flexibility = Grids
> Thermal
> Large-scale
Energy import options
Electrification – reduces
the need for total energy
demand
Digitalisation – increases the
electricity consumption
Sufficiency
Dampen increase in
electricity consumption
Energy efficiency
Dampen increase in
electricity consumption
> Levers to mitigate the increase
1
• Requires decisions at regional level
• Continuous investments (some residential)
• Mainly impacts distribution/MV grids
• Residential demand flexibility
• Small-scale batteries
• Large scale batteries
• Industry demand flexibility
• Low/medium voltage grid connecting loads,
storage and generation
• High voltage grid (on & offshore) within and
from/to Belgium connecting loads, storage &
generation
• Requires decisions at federal level
• Large investments & discrete
• Mainly impacts the high voltage grid
Onshore wind and solar PV – Further
onshore RES development
Nuclear – Lifetime extension of
existing reactors & building new ones
Thermal – Enough to be adequate.
Generation depending on EU merit
order
Offshore wind – Domestic but also
non- domestic offshore wind
Imports – Consequence of the choices
above, depending on EU merit order
2
3
6
4
5
165
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium164
MARGINAL ABATEMENT COST CURVE FOR BELGIUM FIGURE 516
2040
MAC (Marginal Abatement Cost) €/tCO
2
1.000
800
600
400
200
0
-200
-400
1.000
800
600
400
200
0
-200
-400
From
To
2030
MAC (Marginal Abatement Cost) €/tCO
2
TransportBuildings Industry Carbon
CO2
CO2
H2H2H2 H2
Oil-fired
boiler
Tertiary
Heat
Pump
Residential
Heat
Pump
Tertiary
Heat
Pump
Residential
Heat
Pump
Tertiary
Hydrogen
Boiler
Residential
Hydrogen
Boiler
Efficient
Residential
Appliances
Residential
Insulation
BEV FCEV BEV FCEV Heat
Pumps
E-Boiler Hydrogen
boiler
Carbon
Capture &
Storage
Direct Air
Capture
Natural gas
boiler Current Energy
Labels Light-Duty
ICE Heavy-Duty
ICE Natural Gas
Heating
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030
Carbon Capture
Techniques
Oil
Cost range based on high-low cost estimates. Appliances: CAPEX difference; Insulation: Façade (high) and roof (low)
Electricity Natural gas Hydrogen Efficiency Carbon capture
CO2
WHAT CAN WE CONCLUDE FROM THE MACC?
A successful climate ambition strategy, for demand-side
applications, could then be based on the insights provided
by this MACC. Notably that:
1.
From an economic point of view, some decarbonisa-
tion options are no-regret options, as they save money
compared with the base case. This is true in the short-
term for oil-fired boilers, energy efficiency in homes and
industry, and in the long-term, for low-temperature heat
in industry (heat pumps).
2. Across all sectors, electrification options prove cheaper
than their green molecules counterpart. Whether for
residential and tertiary heating, person and freight
transport, or industrial heat.
3.
All decarbonisation becomes cheaper in the future. Heat
pumps are a no-regret option in financial terms, with the
largest cost reduction expected for heavy-duty vehicles
and high-temperature heat for industry.
4. Carbon capture options seem interesting compared to
other options. However, they also carry economic and
technological challenges.
Further comments on conclusions and insights are detailed
more extensively in Appendix E.
5.3. ELECTRICITY DEMAND
While the final energy demand is expected to decrease in Europe
and Belgium, the demand for electricity is expected to rise.
Indeed, as explained in previous chapters, the reduction in the
final energy demand is mainly driven by electrification as more
efficient devices are used for transport and heating. The electrifi-
cation of heating and transport can also be analysed by looking at
the cost it takes to remove a certain amount of carbon from the
system when compared with other options. This is discussed in
BOX 5-2 through the marginal abatement cost curves. It is clear
that the electrification of heat or transport is the preferred option
from a carbon abatement cost point of view.
MARGINAL ABATEMENT COST CURVES (MACCS) FOR DEMAND LEVERS
What is a MACC?
Many ways and technologies can be explored to decar-
bonise energy uses. But which option is the most cost
efficient?
A MACC is a tool which is often used as part of decar-
bonisation exercises which aim to identify the cheapest
decarbonisation options within and across sectors and,
ultimately, are aimed at devising a successful climate
strategy. The tool visually represents the cost associated
with achieving each additional unit of pollution reduction,
ordered from the least expensive to the most expensive.
For comparison purposes, the decarbonisation options
are sorted per sector and type of consumption.
This exercise requires several inputs for each technology,
such as CAPEX, OPEX and emissions related to the use
of the assets, as well as those related to the decarboni-
sation option which is being considered (e.g. considering
the Marginal Abatement Cost for a heat pump is different
if it replaces a gas boiler or an oil boiler). All of these are
aligned with data used in the framework of this study’s
inputs and simulations (price of fuels, CO2 prices…).
These inputs can also change over time. For instance, the
cost of hydrogen fuel and technologies is expected to
decrease over time, and electricity is expected to have a
lower CO2 intensity in 2040 than in 2030.
A MACC exercise is most robust when defining range val-
ues for two reasons:
1. the inputs can be prone to uncertainties (the price of
hydrogen in the future is uncertain);
2.
a decarbonisation option can be applied to different
extents. For instance, ‘residential insulation’ could cover
the simple replacement of windows or could involve
the renovation of a whole house.
Elia asked Sia Partners to construct MACCs for the
selected demand-side technologies from a societal per-
spective (which means considering wholesale prices
for fuel costs, which are different from a consumer per-
spective, and considering other investments required for
the decarbonisation option, such as charging points for
electric vehicles). The latter include the different options
for reducing the use of fossil fuels in buildings, transport
and industry, as well as carbon capture techniques. An
explanation of the methodology written by Sia Partners
is included in Appendix E, as well as associated limits and
caveats.
How should a MACC be read?
On the graph, the values in € per tonne of CO2, are dis-
played as bars. Where relevant, a range is indicated with
the whiskers. This range represents high and low values
when there are uncertainties in the costs (e.g. a range
specified in the literature). When the value exceeds
€1,000/tonne, this is indicated with a dashed bar. Some
values are negative, indicating that decarbonising is
cheaper than not doing it.
The chart is split into several sections:
◆ From left to right, decarbonisations are sorted per (i)
sector (buildings, transport, industry and carbon), (ii)
assets to decarbonise (e.g. for buildings: oil-fired boiler,
gas-fired boiler, energy efficiency), and (iii) options
available for each asset (e.g. for gas boiler, you could go
to electric heat pumps or hydrogen boilers).
◆ From top to bottom, the exercise has been realised for
two time horizons: 2030 and 2040. These exercises lev-
erage baseline data from 2024 & 2030 respectively. So
for each decarbonisation option, it is clear how MACC
values are expected to evolve through time.
BOX 5-2
167
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium166
IMPACT OF THE POTENTIAL ASSUMED
The impact of the sufficiency sensitivity is depicted in Figure 5-17.
The sensitivity applied assumes a reduction of around 20 TWh of
electricity consumption and hence includes both shorter term
and longer term measures as described above. It is important to
mention that the socioeconomic cost of the sufficiency measures
is not quantified, as it is still up for debate. Hence the impact
shown is the impact on the electricity system costs only. Certain
measures could require investments or public support (e.g. more
public transportation).
According to the simulations, if all the potential measures
are applied:
◆ The costs of the electricity system are reduced by around €15
to €20/MWh;
◆ The net imports can be reduced by 15 to 20 TWh;
◆ The need for additional capacity for adequacy can be reduced
by 2 to 3 GW.
Sufficiency does not only impact electricity consumption, the
impact on other vectors is also significant, however those were
not assessed in the present study.
IMPACT OF SUFFICIENCY ON KEY SYSTEM INDICATORS FIGURE 517
Impact on costs of the
electricity system in
[EUR/MWh]
Reduction between 15 - 20 €/MWh
Reduction of net imports by 15 to 25 TWh
Reduction of needed capacity by 2 to 3 GW
Costs of sufficiency
measures not
accounted for
Range across
sensitivities
[€/MWh]
[TWh]
[GW]
Range across
sensitivities
Range across
sensitivities
Impact on net
electricity imports in
[TWh]
Impact on needed
capacity for adequacy
in [GW]
2036
2036
2036
2050
2050
2050
2040
2040
2040
0
-5
-10
-15
-20
-25
0
-5
-10
-15
-20
-25
0
-1
-2
-3
-4
5.3.1. SUFFICIENCY AS A LEVER
SUFFICIENCY IN THIS STUDY
The sufficiency sensitivity explored in this study is based on the
EnergyVille SHIFT scenario, part of the PATHS2050 study [EVI-1]
[EVI-2]. It assumes a reduction of 30 TWh in energy demand
compared to the DE scenario in 2050, as described in Section 3.2.2.
Measures were applied in different sectors:
◆ Residential and tertiary: the main measures considered are
(i) optimising space and (ii) reducing the heating setpoint
(by 1°C). Optimising space corresponds to increased co-living
(increasing the average amount of people per house), and
sharing office spaces.
◆ Transport: whether for passenger or freight transport, two
types of measures are applied: (i) modal shift and (ii) increas-
ing occupancy per trajectory (vans load factors or average
passengers per trip). For passenger transport, there is an
increase in the share of public transport and active modes of
transportation (walking and biking), leading to a reduction in
the car modal share.
◆ Industry: it is assumed to increase resource efficiency
through circularity and a change in the means of production
and in the manufacturing model. This assumes optimised
product design, reuse and recycling. The reader should note
that all these measures cannot be implemented from one
day to another. Some can but others require political will and
structural changes.
Out of the 30 TWh of energy demand reduction, around 20
TWh are related to electricity by 2050. But this energy demand
reduction does not happen overnight. Behaviours are expected to
change gradually. Some energy demand reduction can happen
fast (as demonstrated in France in 2022, following the sufficiency
plan), but other measures need support and policies to material-
ise. Hence, a distinction can be made between sufficiency meas-
ures activated in the short-term and the long-term. As already
explored in the most recent Adequacy & Flexibility study:
◆ Certain measures could be implemented in the shorter
term and are linked to behaviour changes. Those are esti-
mated to be around 5 TWh in the 2030. A few examples would
be a decrease in the heating set point, lower speed limits on
highways and a modal shift for short distances (< 1 km). Note
that this is similar to the sufficiency plan established by the
French Government in 2022-23, where the largest gains hap-
pened through lowering heating consumption, and control-
ing lighting waste;
◆ In addition more long-term behaviour changes could lead
to another 5 TWh reduction in electricity consumption in
the 2030. Some examples would be reducing the average
size of cars, and average size of dwellings. These measures
need policies to be incentivised and adopted on a longer time
frame;
◆ The remaining 10 TWh assumed in this study relies on more
systemic changes. These could include notably the use of
circularity in industry. Also, measures related to freight trans-
port (modal shift towards train and boat) require structural
changes in supply chains that do not happen overnight. Here
again, policies are required to encourage and implement
these measures over a prolonged period.
169
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium168
FLEXIBILITY ACTIVATION AND RESIDUAL DEMAND
Next to dispatchable capacity, demand flexibility and storage
(further referred as 'flexibility' in this section) also has a key role
to play in contributing to adequacy (see also BOX 5-6), smooth-
ening fluctuations of demand and allowing a better integration
of intermittent renewables. Figure 5-19 illustrates the impact of
flexibility and storage on the hourly residual demand for the DE
scenario in the year 2050. For this example, the points in the fig-
ure are sorted from high to low residual demand for each hour
of a single simulation year in the case there is complete lack of
demand-side flexibility and storage (black line), the light blue dots
show the impact of activated flexibility and storage options in the
unit dispatch model for each hour on the black line. In general it
can be seen that the activation of demand flexibility and storage
reduces the residual demand in hours with (very) high residual
demand and vice-versa leading to a flattened overall residual
demand curve over the whole period. The assumptions regarding
flexibility in the scenarios (storage and demand flexibility) are
described in Section 3.2.4.
It can be observed that:
◆ Flexibility is activated throughout the year:
◆ When the residual demand is high (ex ante any flexibility
activation), flexibility is mainly used to reduce it. However the
residual demand (in Belgium) is not the sole driver of flexibil-
ity activation. For instance it is possible that the situation is
different in other countries or that the marginal prices allow
to ‘charge’ batteries eventhough the residual demand in Bel-
gium is high. This can be observed when in some hours the
residual demand after flexibility activation is higher than the
one without.
◆ When residual demand is low, flexibility is usually used to
increase it as those are moments where there is an excess of
renewables in the system.
◆ The ‘peak demand’ is not necessarily expected to happen at
maximal residual demand. Indeed with large volumes of flex-
ibility (> 20 GW), it is possible that the ‘peak demand’ after
flexibility activation is observed during other moments and
not necessarily when renewable production is low and con-
sumption (ex ante flexibility activation) high.
◆ The latter observation confirms that representing only an
hourly peak consumption does not provide a complete view
as the hourly peak can happen outside of high consumption
(ex ante flexibility activation). A better proxy for ‘peak demand’
is the daily peak demand as the flexibility is mostly able to
flatten intraday variations to a certain extent.
IMPACT OF FLEXIBILITY ON THE HOURLY RESIDUAL DEMAND DURATION CURVE FIGURE 519
35
25
15
5
-5
-15
-35
Hourly residual load [GW]
Residual demand
- no flex
Residual demand
- incl flex
Moving average
5.3.2. PEAK DEMAND AND FLEXIBILITY
While the additional dispatchable capacity needed was calcu-
lated for each scenario following a similar approach to the one
used in adequacy studies, it is interesting to have a look at the
expected peak demand for the electricity system in Belgium. It
is important to note that in the future, given the large amount
of expected flexible consumption, it is not straightforward to
define something like a peak. Indeed the peak demand of
the system is greatly influenced by the demand flexibility ‘dis-
patch’. This will be further highlighted in this section.
YEARLY PEAK DEMAND
Rather than the absolute hourly peak demand, a proxy value to
analyse the required additional capacity needed on top of the
domestic RES concerns the ‘residual’ demand which is defined
here for each day of the climatic year considered as:
Daily electricity demand – daily solar PV generation– daily
onshore wind generation – daily (domestic) offshore wind gen-
eration
The yearly peak residual demand is then obtained by taking the
day with the highest residual demand for each climate year. Fig-
ure 5-18 shows the distribution of the peak residual demand
(calculated as the day with the highest residual demand over
the entire year) across the 200 forward-looking climate years
used in this study. As such it can be seen that this peak residual
demand increases over time and is obviously higher in the sce-
narios with a higher degree of electrification. The peak demand
ranges between 15 and 25 GW in 2036 depending on the scenario
and weather conditions during winter. By 2050, the peak demand
is projected to increase, reaching between 20 and 30 GW in 2050
(across different scenarios and climate years). Such increase com-
pared to today (around 12-13 GW) is driven by the additional elec-
trification assumed in all scenarios. Another interesting insight
that can be derived from the figure is the increased distribution
of the peak demand for a given scenario over time. As heating
processes become more electrified, the variance between colder
and warmer winters significantly influences the peak demand.
This effect is even more pronounced when considering scenarios
with higher levels of electrification.
DISTRIBUTION OF THE YEARLY PEAK RESIDUAL DAILY DEMAND IN BELGIUM FIGURE 518
P90-max
P10-P90
P10-min
35
30
25
20
15
10
5
0
Peak residual demand [GW]
Average
17
19 20 21
22 22 22
24 25
2036 2040 2050
DE DE DE
GA GA GA
ELEC ELEC ELEC
The boxplots show the distribution of the maximum daily residual demand across 200 climate years
171
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium170
5.4. ELECTRICITY LONG-TERM SUPPLY
OPTIONS
With the expected growth in electricity demand, Belgium has the
opportunity to steer what its future electricity supply mix looks
like. Several supply and demand levers were identified which
could be activated if desired. In this section a quantitative assess-
ment of the effect of each of these levers under a diverse set of
scenarios and sensitivities is explored for indicators within Elia’s
domain of expertise. It should be noted that the final choice of
levers is a complex and multidimensional problem covering not
only costs but also energy independence, the careful considera-
tion of risks, societal dimensions and others, and as such should
include other elements outside of the results presented here.
5.4.1. ELECTRICITY MIX DASHBOARD FOR 2050
Starting from known ambitions for Belgium complemented with
assumptions about growth potentials, a diverse set of supply
sensitivities is identified for each of the target years. A selected set
of sensitivities is shown for the 2050 horizon for the DE demand
scenario in Figure 5-21. This figure gives an overview of the sup-
ply capacities in each considered category on the left and the
results of the European multi-energy dispatch on the right. It is
shown that sensitivities with more domestic renewables, nuclear
or non-domestic offshore supply result in a lower need for ther-
mal capacity, less imports and more moments with low (below
€20/MWh) electricity prices. To paint a more complete picture,
the costs to install and maintain these supply options as well as
required infrastructure (including also TSO and DSO grids), fuel
costs, import costs, export benefits, congestion rents,… should
be taken into account.
ELECTRICITY MIX DASHBOARD FOR 2050 DE SCENARIO FIGURE 521
Capacity per scenario
Nuclear
Foreign
offshore DRES
Thermal
in GW Domestic + foreign offshore Exports [-] / Imports [+]
Share hours per yr.
with curtailment
Share hours per yr.
Below 20 €/MWh
Supply mix in TWh Curtailment Low prices
0 GW
8.2 GW
4 GW
0 GW
0 GW
8 GW
16 GW
16 GW
Central
Central
Central
High
High
High
High
High
Central
Central
5103
136
60
59
28
27
59
58
58 59
58 58
30
29
102
135
102
132
101
132
103
136
5
5
5
5
5
5
4
5
4
9
3
9
3
7
7
22 92 0.2% 10%-34
20 74 2.7% 14%-46
15 62 0.2% 15%-55
14 46 2.5% 19%-69
13 64 0.3% 14%-52
12 50 3.1% 19%-67
11 65 0.7% 14%-52
10 50 4.3% 18%-66
534 0.8% 20%-73
520 4.5% 24%-87
Thermal-Existing Thermal-New Foreign offshoreNuclear DRES Thermal Gap-to-max Net imports Net exports Gross exports Gross imports
IMPACT OF MORE HEATING NETWORKS IN 2050
As requested by several stakeholders, a sensitivity assum-
ing more heating networks is assessed. Heating networks
are currently very limited in Belgium. In the DE scenario,
3.5 TWh of district heating is already assumed (compared
to almost nothing nowadays). This sensitivity identifies
the impact of additional district heat (such as heating
networks) replacing part of the residential heat pumps. In
their study, EnergyVille identified a total technical poten-
tial of 24 TWh by 2050, with up to 8 TWh being found to
be economically viable according to their model [EVI-1].
The sensitivity considered in this study assumes 15 TWh
in order to test a reasonable extreme. Such heating net-
works could be either supplied via direct heat or com-
bined with heat pumps. The way in which this heat would
be supplied is not analysed. The same stands for the costs.
The costs of developing such infrastructure and the costs
of upstream heat generation are not accounted for in the
impact assessment. This is an important assumption that
should be accounted for when looking the results.
More heating networks could lead to several benefits:
◆ Reductions in the peak demand for the electricity sys-
tem and hence the adequacy requirements and local
grid requirements;
◆ Re-use of waste heat (e.g. industrial heat) that would
otherwise remain unused;
◆ Such a solution could be interesting notably for densely
populated areas.
The main impact on the electricity system is reduced
electricity demand (if the heating networks are replacing
electrified heat), hence impacting imports and adequacy
needs. A smaller part is also attributed to the local grids
(lower peak demand).
Figure 5-20 summarises the results for electricity sys-
tem costs, net imports and adequacy requirements. It
is however important to mention that the additional
costs related to the heating networks (deployment, con-
struction) and to the supplied heat are not accounted
for. Those should be accounted for when performing a
full cost-benefit analysis of the solution (which was out
of scope of this study). One can conclude that given the
large benefits, such technology and solutions should be
further investigated.
IMPACT OF ADDITIONAL DISTRICT HEATING ON THE ELECTRICITY SYSTEM COSTS FIGURE 520
Costs of heating
networks and
upstream heat not
accounted for
From 3.5 TWh district heating
to 15 TWh in 2050
0
-1
-2
-3
-4
-5
-6
-7
-8
Impact on costs of the
electricity system in
[EUR/MWh]
Impact on net
electricity imports
[TWh]
Impact on needed
capacity for
adequacy in [GW]
Reduction of system
costs by 6 to 7 EUR/MWh
Reduction of net
electricity imports
by 2 to 4 TWh
Reduction of adequacy
need by 0 to 1.5 GW
BOX 5-3
173
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium172
NET IMPORTSEXPORTS AND THERMAL GENERATION FOR THE DE SCENARIO IN 2050 FIGURE 522
Net Import [+] /Export [-] Thermal generation [TWh]
2050
CENTRAL RESHIGH RES
11
10
7
7
5
5
12
11
10
9
16
15
13
12
11
10
17
15
14
13
22
20
18
17
15
14
13
-16
-12
-40
-39
-67
12
-17
-12
-41
37
8
12
-17
-15
-44
35
5
10
-19
58
28
34
3
7
-23
8 GW
8 GW
8 GW
8 GW
4 GW
4 GW
4 GW
4 GW
6 GW
6 GW
6 GW
6 GW
2 GW
2 GW
2 GW
2 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
4 GW
4 GW
4 GW
4 GW
8 GW
8 GW
8 GW
8 GW
12 GW
12 GW
12 GW
12 GW
16 GW
16 GW
16 GW
16 GW
5.4.2. IMPORTS, EXPORTS AND THERMAL GENERATION
First, one can assess the amount of net imports/exports for each
of the options in 2050. Those are shown in Figure 5-22 and depend
on the amount of domestic RES (comparison between the upper
and lower charts) and on the amount of non-domestic offshore
and new nuclear considered. The same is shown for the thermal
generation (sum of methane and hydrogen-fired turbines). While
the analysis is based on the DE scenario, similar trends can be
observed in the GA and ELEC demand scenarios.
A few observations can be made:
◆ Increasing the amount of domestic RES (from Central to High
RES) can reduce the net imports by around 30 TWh (no matter
what level of non-domestic offshore or nuclear is accounted
for). However the decrease in thermal generation is much
more limited (1 to 2 TWh). This can be attributed to the fact
that the amount of thermal capacity needed is relatively simi-
lar in the Central and High RES and that thermal generation is
dispatched when there are low volumes of renewables in the
European system (during winter);
◆ Increasing non-domestic offshore and assuming that the
generation is counted as ‘Belgian’ allows net imports to be
decreased by the same amount as the offshore generation. A
similar conclusion can be made for nuclear generation.
◆ More non-domestic offshore or nuclear allows the ther-
mal generation in Belgium to be reduced as the amount of
installed thermal capacity required to keep the system ade-
quate decreases. This effect is higher with more nuclear given
its higher contribution to adequacy than wind offshore.
◆ In the Central RES scenario, installing either 16 GW of non-do-
mestic offshore or 8 GW of new nuclear results in Belgium net
importing around 10 TWh of electricity. Combining the above
with a higher domestic RES scenario, Belgium becomes a
next exporter of around 20 TWh.
◆ In the high domestic RES scenario, Belgium could have a net
import close to zero if it installs at least either 8 GW non-do-
mestic offshore or 4 GW of nuclear or a combination thereof
(4 GW offshore and 2 GW nuclear). In such a case, the thermal
generation would produce around 15 TWh.
175
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium174
ELECTRICITY SYSTEM COSTS IN €MWH FOR THE DE SCENARIO FIGURE 523
8 GW
4 GW
6 GW
2 GW
0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
119
118
118
120
117
117
114
123
117
112114
115 110
8 GW
4 GW
6 GW
2 GW
0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
114
113
112
114
114
108
110
114
115
109110
109 107
HIGH RESCENTRAL RES
2050 - DE - reference costs
Onshore wind
x2 installation rate
-9 to -2€/MWh
Solar PV
x2 installation rate
COSTS AND WACC SENSITIVITIES
In order to grasp the impact of different cost assumptions for
new nuclear and non-domestic offshore, Figure 5-24 shows the
impact of:
◆ Higher costs for new nuclear generation (€10,000/kW instead
of €7,500/kW). This also includes a higher WACC (10% instead
of 7%) to reflect that higher risks for big projects could rise the
investor’s demanded return. Other parameters such as con-
struction time are left unchanged. This was assumed to be
7 years for an SMR and 9 years for a large-scale unit (see BOX
3-5 for more information on recent construction durations);
◆ High offshore and grid costs (€2,200/kW instead of €1,600/
kW for non-domestic offshore wind). The WACC was not
increased given that the risks associated to wind and grid
projects (usually regulated) are more limited than for nuclear
projects.
The variation on costs is applied on the figure as follows:
◆ In the chart on the left, the cost increase of offshore wind and
grids is applied;
◆ In the chart on the right, the cost increase of new nuclear is
applied.
It can be observed that when considering high costs for non-do-
mestic offshore wind and grids but keeping the reference
assumptions for nuclear costs:
◆ There is a benefit observed to do non-domestic wind when
there is no nuclear in the system. The impact is limited to €-1/
MWh;
◆ If there is nuclear in the system, the cost increases by €2/
MWh (when 4 GW nuclear) to €9/MWh (when 8 GW nuclear).
It can also be observed that when considering high cost
assumptions for new nuclear and keeping the assumptions
for offshore wind and grids as the reference:
◆ The costs are higher in any of the sensitivities. The cost
increases by €20/MWh when there is no offshore to €33/MWh
where there is 16 GW of non-domestic offshore;
◆ The impact of changes in cost assumptions on the total sys-
tem costs is therefore higher for nuclear than for non-domes-
tic offshore.
5.4.3. ELECTRICITY SYSTEM COSTS OF THE DIFFERENT OPTIONS
The different supply options can be compared in terms of total
electricity system costs. Those costs include all costs of the elec-
tricity system:
◆ CAPEX annuities of supply options, for all investments
required from 2030;
◆ fixed operation and maintenance costs of supply options;
◆ CAPEX and OPEX of the grid (DSO, TSO);
◆ variable costs (fuel costs, CO2 costs, hydrogen generation ben-
efits and imports/exports costs/benefits).
The costs are expressed in €/MWh in order to give the reader a
value that can be compared between scenarios of different final
electricity demand. However, these costs are not to be confused
with the ‘electricity price’ paid by the consumer, nor the ‘whole-
sale’ price on bulk power markets. Indeed the ‘wholesale’ price
is driven by the marginal costs of the system (which only include
variables costs and other opportunity costs of flexible devices).
In addition, the electricity price paid by consumers can differ
depending on the different types of costs included in the final bill
(grid, taxes, wholesale…). The total system cost considered in the
next sections is therefore the sum of the yearly costs (annuities
and variable costs) divided by the total electricity consumption.
Different scenarios will be assessed in terms of cost assumptions
for the investments made. Indeed, there are many uncertainties
regarding the future cost of supply options. As explained in Sec-
tion 3.3, two types of sensitivities are performed– those will be
clearly indicated in the results:
◆ Low/Ref/High costs of investments (CAPEX), which differ for
each technology;
◆ Low/Ref/High WACC (cost of capital) assessed (4%-7%-10%).
REFERENCE COSTS AND WACC
As a starting point one can depict the total electricity system costs
in €/MWh for the reference CAPEX and WACC (7%) assumptions
taken for each technology. Similar to the previous section three
levers are shown together: Central/High RES on two separate
charts, the level of non-domestic offshore on the x-axis and the
level of new nuclear generation on the y-axis. The results are
shown in Figure 5-23.
It can be seen that the cost of the system ranges between €123/
MWh and €107/MWh depending on the different sensitivities
assessed. A few observations can be made:
◆ The High domestic RES scenario allows the system costs to
be reduced by €2 to €9/MWh. The reduction is more pro-
nounced for sensitivities with less non-domestic offshore or
new nuclear. Indeed, in those cases, Belgium imports a lot
of its electricity and the increase in domestic RES avoids the
most expensive imports;
◆ In the Central domestic RES scenario, no new nuclear or no
new offshore wind is always more expensive compared with
scenarios where one of these options (or a combination) is
used, even when accounting for the additional investment
costs. This means that reducing the amount of net imports
from the Current Policies scenario allows Belgium to reduce
its total electricity system costs.
◆ Installing more non-domestic offshore always reduces total
system costs, in any of the combinations except in the ‘high
domestic RES’ and 8 GW new nuclear case where going from
8 GW to 16 GW offshore increase the costs by €1/MWh. New
non-domestic offshore wind impacts the costs as follows:
-
€-2/MWh (8 GW new nuclear) to €-13/MWh (no new nuclear)
in the Central RES scenario;
- €-7/MWh (no new nuclear) to €1/MWh (8 GW new nuclear)
in the High domestic RES scenario.
◆ Installing new nuclear reduces total system cost if there is
no new non-domestic offshore. Otherwise it is always more
expensive from a system cost point of view. New nuclear
impacts the costs as follows:
-
€-4/MWh (no non-domestic offshore) to €7/MWh (16 GW
non-domestic offshore) in the Central RES scenario;
-
€0/MWh (non-domestic offshore) to €8/MWh (16 GW
non-domestic offshore) in the High RES scenario.
177
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium176
IMPACT ON ELECTRICITY SYSTEM COSTS FOR HIGH NONDOMESTIC OFFSHORE
AND HIGH NUCLEAR COST ASSUMPTIONS. FIGURE 524
2050 - DE - Central RES
145 144 143
137 136
133 127 124
124 121
123 115 110
119 117 117
118 117
120 114 112
118 114
123 115 110
120 124 129
122 126
122 121 124
122 124
124 122 123
Nuclear WACC 7%; overnight CAPEX 7 500 €/kW
Offshore wind WACC 7%; overnight CAPEX 1 600 €/kW
Reference grid costs
Nuclear WACC 7%; overnight CAPEX 7 500 €/kW
Offshore wind WACC 7%; overnight CAPEX 2 200 €/kW
High grid costs
Cost increase
for new nuclear
Cost increase for
non-domestic
offshore wind
Nuclear WACC 10%; overnight CAPEX 10 000 €/kW
Offshore wind WACC 7%; overnight CAPEX 1 600 €/kW
Reference grid costs
NEW
NEW NEW
NEW
8 GW
8 GW 8 GW
4 GW
4 GW 4 GW
6 GW
6 GW 6 GW
2 GW
2 GW 2 GW
0 GW
0 GW 0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
0 GW 0 GW4 GW 4 GW8 GW 8 GW12 GW 12 GW16 GW 16 GW
IMPACT OF HIGHER HVDC CONVERTER COSTS
Due to the fact that the EU high voltage electricity grid is pro-
jected to nearly double in size by 2050 compared to its current
state, the supply chain for high voltage electric components is
under stress. This strain on the supply chain results for example
in increased prices for HVDC convertors. As it is very uncertain
for how long this situation will endure, to account for different
grid cost evolution scenarios two additional sensitivities involv-
ing increased HVDC convertor costs were conducted for 2036
and are outlined below. For these sensitivities, the cost of off-
shore HVDC converters are doubled compared to the reference
costs, while onshore converters experience a 25% cost increase.
Impact on the EU high voltage electricity grid optimum for 2036
First, the EU optimum was recalculated using aforementioned
high convertor costs throughout Europe. From these results it
was observed that only slightly less new offshore wind capacity
(around 6 GW less compared to the reference cost scenario) was
developed in Europe. For Belgium the domestic offshore poten-
tial remained however fully used, reaching the 5.8 GW by 2036
(the increase to 8 GW in the BE EEZ is only included as an invest-
ment candidate as of 2040). The Nautilus hybrid interconnector,
connected to the Princess Elisabeth Island, remains present in
the cost-optimum as well.
In addition, 2 GW of hybrid interconnectors connecting non-do-
mestic offshore wind in the North Sea to Belgium were found
to be cost-optimal for 2036. While overall less offshore wind
farms and hybrid interconnectors are developed in Europe com-
pared to the reference costs scenario (about half as much kilo-
meter cables (GW*km, see also figure 4-31) of offshore HVDC is
installed in the high costs case compared to the reference cost
case), it was observed that it remains cost-optimal on system
level for Belgium to further develop domestic offshore wind
capacity, create an offshore energy hub, and connect far-out
offshore wind through hybrid interconnectors to the country.
Effect on the cost effectiveness of the BE domestic supply mix
The impact of the increased grid cost on the estimated total
electricity system cost for Belgium (per MWh) is shown the fig-
ure in this box for different domestic supply mixes for 2040 and
2050. It can be seen that even with such high cost estimates for
HVDC converters, the connection of additional non-domestic
offshore wind on top of the 8 GW of offshore wind in the Bel-
gian EEZ remains cost-effective, indicating a robustness of the
investments to this type of unexpected event. The tipping point
at which investing in additional non-domestic volumes stops
reducing system cost is however slightly shifted downwards,
especially with high volumes of nuclear generation in Belgium.
IMPACT OF HIGH HVDC CONVERTER COSTS ON TOTAL ELECTRICITY SYSTEM COST FOR BELGIUM
2050 - Distributed Energy - Cenral dRES - Total Electricity System cost for Belgium in €/MWh
2040 - Distributed Energy - Cenral dRES - Total Electricity System cost for Belgium in €/MWh
105 108
106 110
107 107
Offshore HVDC: 5 90 EUR/kW
Onshore convertor: 260 EUR/kW
Offshore HVDC: 1125 EUR/kW
Onshore convertor: 325 EUR/kW
NEW NEW
8 GW 8 GW
4 GW 4 GW
6 GW 6 GW
2 GW 2 GW
0 GW 0 GW
0 GW 0 GW
4 GW 4 GW
8 GW 8 GW
12 GW 12 GW
16 GW 16 GW
HVDC offshore
convertor cost x2
106 108
106 109
107 107
106 108
105 109
108 108
HVDC offshore
convertor cost x2
119 117117
NEW
8 GW
4 GW
6 GW
2 GW
0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
118 117
120 112114
118 114
123 110115
119 123120
NEW
8 GW
4 GW
6 GW
2 GW
0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
120 121
120 118119
120 119
123 117118
- Investing in additional non-domestic offshore wind, on top of 8 GW in the BE EEZ is still beneficial for Belgium with respect to the electricity system costs
- However the tipping point at which investing in additional non-domestic volumes stops reducing system cost is shifted, especially with high volumes of
nuclear generation
179
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium178
TOTAL YEARLY ELECTRICITY SYSTEM COSTS FOR BELGIUM SPLIT BY COMPONENT
FOR THE DE CENTRAL DOMESTIC RES SCENARIO FIGURE 525
30
25
20
15
10
5
0
-5
B€/year
Non-domestic
offshore wind
New nuclear units
0 GW 16 GW 8 GW 0 GW 16 GW DE - Central RES
Net total costs
Grid costs
Non-domestic
offshore
Nuclear
Flexibility
Domestic
RES
Fuel (nuclear,
molecules)
& CO
2
Imports/
exports
Benefits
Costs
Backbone and ICs
Offshore grid
Regional
DSO
Adequacy
(thermal
fixed costs)
0 GW 0 GW 4 GW 8 GW 8 GW
OPEX FIXED (CAPEX & FOM)
TOTAL YEARLY ELECTRICITY SYSTEM COSTS FOR BELGIUM SPLIT BY COMPONENT
FOR THE DE HIGH DOMESTIC RES SCENARIO FIGURE 526
30
25
20
15
10
5
0
-5
B€/year
Non-domestic
offshore wind
New nuclear units
0 GW 16 GW 8 GW 0 GW 16 GW DE - High RES
Net total costs
Grid costs
Non-domestic
offshore
Nuclear
Flexibility
Domestic
RES
Fuel (nuclear,
molecules) &
CO
2
Imports/
exports
Benefits
Costs
Backbone and ICs
Offshore grid
Regional
DSO
Adequacy
(thermal
fixed costs)
0 GW 0 GW 4 GW 8 GW 8 GW
OPEX FIXED (CAPEX & FOM)
COSTS COMPONENTS OF THE DIFFERENT SENSITIVITIES
The total electricity system costs presented in the previous sec-
tions combine all types of costs of the system. In order to assess
the different components and their size, it is possible to split the
costs by component.
Figures 5-25 and 5-26 present the costs for the DE scenario with
Central RES and High RES assumptions. Five sensitivities are
chosen to show the differences between OPEX and fixed costs
and the different component sizes. Those cover the range of the
previous charts with non-domestic wind offshore on the x-axis
and new nuclear on the y-axis.
The components are split by category:
◆ Variable costs:
-
Import/export costs are related to the amount of spending to
buy electricity abroad or to sell this electricity. The net costs
are reported. Those can be negative if the revenues from
exports are higher than the costs of imports;
-
Fuel and CO
2
costs are linked to the fuel used in thermal
generation (nuclear, methane or hydrogen) as well as costs
for CO2 abatement (if any).
◆ Fixed costs (annuities for investments from 2030 and fixed
costs of operation):
-
Domestic RES fixed costs are annuities and FOM related
to PV, wind onshore, hydro, biomass and wind offshore in
Belgium;
-
Adequacy (thermal) are the annuities and fixed costs related
to thermal power plants other than nuclear;
- Flexibility costs are annuities and fixed costs related to stor-
age and demand flexibility;
- Nuclear are the costs related to new nuclear installations;
-
Non-domestic offshore are the costs related to offshore wind
outside of the Belgian EEZ;
-
Grid costs are all costs related to the grid (DSO costs, regional
grid costs (30-70-150 kV), backbone (380 kV and onshore
interconnectors) and offshore grids.
The costs components are very different depending on the
scenario:
◆ The first sensitivity (0 GW non-domestic offshore and 0 GW
nuclear or ‘Current Policies’) incurs the highest OPEX costs,
accounting for 50% of the total costs. This indicates that these
costs are associated with purchasing resources from abroad,
such as fuel or electricity;
◆ The other sensitivities have lower OPEX costs but a higher
share of CAPEX (up to 80 % of the costs).
A few more observations can be made:
◆ As identified previously, the combination with the lowest sys-
tem costs (for the reference cost assumptions) is the one with
16 GW non-domestic offshore and no nuclear. This is indi-
cated by the red dot on the graph;
◆ The part related to grid investments is between 7% (no
non-domestic offshore) and 20% (16 GW non-domestic off-
shore) for the TSO related costs (which include regional, back-
bone, interconnectors and offshore grid);
◆ DSO related costs make up around 18% of the total costs
across all supply scenarios. This share is rather stable across
demand scenarios as well as it is mainly linked to residential
electrification;
◆ Adequacy related costs (fixed costs of power plants) are rather
limited, up to 6 % in the scenario with the highest need for
thermal generation and less than 1 % in the lowest one. Note
that the variable costs related to the operation of those power
plants are included in the fuel costs in the chart;
◆ Both the High and Central domestic RES scenarios display
a similar trend. However, in the High RES scenario, benefits
from exporting electricity can be observed in all combina-
tions shown in the graph, except for Current Policies.
181
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium180
INVESTMENTS OVER TIME
Depending on the chosen pathway, the amount of CAPEX spent
can differ. Indeed, some scenarios rely more on OPEX as illustrated
in the previous paragraphs and others more on fixed costs (invest-
ments). Figure 5-27 illustrates the amount of overnight CAPEX
spent depending on the chosen pathway. This only includes
overnight costs and no financing costs. In addition, it does not
account for costs incurred during construction. Indeed, for some
technologies, the costs could begin to accrue well before the
different installations are commissioned.
A few interesting observations can be made:
◆ The scenarios with the highest CAPEX share are the ones
where the highest amount of overnight CAPEX needs to be
spent. The scenarios differ in terms of amount to be spent
every 5 years;
◆ While yearly grid costs (when compared to the total annual
system costs) represent less than 20% for DSOs and less than
10% for TSOs (see previous charts), the share of the amount
of total CAPEX to be spent is around 50% if current policies
are pursued. This share is lower if new nuclear is considered
in the system.
It is to be noted that the different supply options (e.g. non-do-
mestic offshore wind versus new nuclear) are developed with a
different timing, as reflected in the chart with respect to invest-
ment costs. This also results in benefits of those investments
materialising differently over the full time horizon, which is to
be taken into account when comparing different options and
pathways. Another noteworthy observation is that aiming for the
most cost-effective scenarios (those with lower total system costs,
including both fixed and variable costs) will necessitate higher
capital expenditure (CAPEX). This means that not only will there
be a need to source the funding for these investments, but it will
also be crucial to try and minimize financing costs as much as
possible. The similar observation can be made for CAPEX to be
spent for grids. Reaching the best scenarios from a total system
costs point of view will require more investments into the grids.
OVERNIGHT YEARLY CAPEX SPENT EVERY 5 YEARS FOR SELECTED PATHWAYS DE CENTRAL RES FIGURE 527
12
10
8
6
4
2
0
Overnight CAPEX B€/year
Non-domestic
offshore wind
New nuclear units
0 GW 16 GW 8 GW 0 GW 16 GW
Backbone and ICs
Offshore grid
Regional
DSO
31-35 36-40 41-45 46-50 31-35 36-40 41-45 46-5031-35 36-40 41-45 46-50 31-35 36-40 41-45 46-50 31-35 36-40 41-45 46-50
0 GW 0 GW 4 GW 8 GW 8 GW
Grid costs
Non-domestic
offshore
Nuclear
Flexibility
Domestic
RES
Adequacy
(thermal
fixed costs)
DE - Central RES
CONSUMER AND PRODUCER SURPLUS
When considering scenarios with a high share of net imports such
as the Current Policies scenario, the total amount of inframarginal
rent received by Belgian producers is lower compared to situa-
tions with more Belgian electricity supply. In order to illustrate
the effect of having more electricity generation in Belgium, the
Current Policies scenario is compared with the scenario with 16
GW non-domestic offshore and 8 GW new nuclear. Figure 5-28
aims to illustrate the impact on the inframarginal rents and con-
sumer payments when international fuel prices increase by 1.5x
or 2x. The figure (provided as illustration of the effect) gives the
following insights:
◆ When international fuel prices increase, the cost for electricity
increases. Indeed, the import costs will increase in moments
where the marginal unit is molecule-fueled thermal genera-
tion. In addition, when this type of generation is marginal in
Belgium it will result in an increase in operational costs of the
system. Such effect is observed in both cases but the absolute
impact is higher in case of less domestic generation as the
prices are then more correlated to fuel prices;
◆ In the situation where Belgium would have more domes-
tic generation, the increase in costs for consumers can be
compensated by the increase in ‘revenues’ for the producers
whose marginal costs are less or not at all affected by vari-
ations in fuel prices. Such a situation can help dampen the
electricity final price paid by consumers if part of that increase
for producers could be captured and redistributed.
ILLUSTRATION OF THE IMPACT ON INFRAMARGINAL RENTS AND CONSUMER PAYMENTS OF
INCREASING INTERNATIONAL FUEL PRICES IN DIFFERENT ELECTRICITY SUPPLY SITUATIONS FOR
BELGIUM. FIGURE 528
40
30
20
10
0
-10
-20
-30
-40
-50
B€/year
0 GW
‘normal conditions’ International fuel
prices x1,5
International fuel
prices x2
Total infra-
marginal rents
Consumer
payments
Net
16 GW 0 GW 0 GW16 GW 16 GW
0 GW 8 GW 0 GW 0 GW8 GW 8 GW
Non-domestic
offshore wind
New nuclear units
183
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium182
FAR-OUT BASELOAD RES CONNECTED TO BELGIUM
Another option outlined in the scenarios is ‘far-out’
baseload RES that would be connected to Belgium. This
assumes an HVDC link from North Africa or other regions
which would be linked to Belgium being able to provide
quasi-baseload renewable generation. This is further
explained in BOX 3-3 in the scenarios chapter.
Similar to new nuclear and new non-domestic offshore,
the sensitivity can be shown on the X/Y axis figure. In this
case, the new nuclear y-axis was replaced by baseload
RES connected to Belgium.
A few observations can be made:
◆ In the reference cost scenario, the option is similar
to new nuclear. The CAPEX costs are assumed to be
€9,000/kW in the reference case and are based on the
project brought forward by Xlinks (see [XLI-3]) and a
WACC of 7%. A small decrease in total system costs can
be observed for the scenario without new non-domes-
tic offshore and a limited amount of far-out baseload
RES. However, an increase in costs can be observed if
there is non-domestic offshore wind integrated into the
Belgian energy system. This conclusion is similar to the
case of new nuclear;
◆ When using high CAPEX cost assumptions (€12,000/
kW) the option becomes less interesting; further
increasing the WACC as well to reflect higher risks
associated to such a project increases system costs
even more.
The option of ‘far-out baseload RES’ can prove interesting
if the cost and the risk associated with the project can be
kept low.
IMPACT OF ‘FAROUT’ BASELOAD RES ON THE SYSTEM COSTS FOR BELGIUM FIGURE 529
148 147 147
140 139
135 129 127
125 122
123 115 110
132 130 130
128 127
127 121 119
121 118
123 115 110
119 117 117
118 117
120 114 112
118 114
123 115 110
High costs €12,000/kW
WACC 7%
Ref costs €9,000/kW
WACC 7%
WACC increase for
far-out baseload RES
Cost increase for far-
out baseload RES
High costs €12,000/kW
WACC 10%
8 GW
8 GW 8 GW
4 GW
4 GW 4 GW
6 GW
6 GW 6 GW
2 GW
2 GW 2 GW
0 GW
0 GW 0 GW
0 GW 4 GW 8 GW 12 GW 16 GW
0 GW 0 GW4 GW 4 GW8 GW 8 GW12 GW 12 GW16 GW 16 GW
Far-out baseload
RES
EUROPEAN BASELOAD RES CONNECTED TO BELGIUM
Although not directly examined in this study, such an
analysis could potentially evaluate the advantages of,
for instance, linking Belgium to a European region with
extensive deployment of low-carbon energy sources. This
large-scale local deployment could prove its benefits due
to regional attributes such as space availability, public
approval, and load factor. Given that the cost assump-
tions used in figure 5-29 were based on regions external
to Europe, the distance needed for electric transmission
could be lower. As such the benefits associated to such a
project could be higher. Such projects could become rel-
evant for Belgium depending on the actual evolution of
Europe's energy landscape and future technology costs.
BOX 5-4
5.4.4. IMPACT OF ADDITIONAL DOMESTIC RES AND PV
The assessed impact of additional domestic RES concluded that
it can reduce system costs and imports but doesn’t considerably
affect thermal capacities (installed and generated energy). This
section delves deeper and explores the impact on the cost of
renewable energy sources and the effect of having even more
photovoltaic (PV) generation in the system than in the High RES
scenario (i.e. the Very High PV sensitivity scenario – going from
65 to 97 GW by 2050). Figure 5-30 shows the total system cost for
the central RES, high RES and very high PV scenarios (where the
PV was capped to the estimated DSO peak) under reference and
low CAPEX assumptions in a similar format as presented before:
with the amount of non-domestic offshore integrated on the
x-axis and the amount of new domestic nuclear on the y-axis.
The results shown for high PV are the ones where the evacua-
tion capacity is capped to the maximum peak demand for the
distribution grid (see also BOX 5-5 for further explanation). Note
that if this was not the case, the system costs should be further
increased with the additional distribution grid reinforcement that
would be required to integrated additional PV on the local grids.
IMPACT ON SYSTEM COSTS OF INTEGRATING ADDITIONAL RENEWABLES IN THE BELGIAN
ELECTRICITY SYSTEM ON TOP OF THE CENTRAL RES ASSUMPTIONS FOR THE REFERENCE AND
LOW CAPEX ASSUMPTIONS. FIGURE 530
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
4 GW
4 GW
4 GW
4 GW
4 GW
4 GW
6 GW
6 GW
6 GW
6 GW
6 GW
6 GW
2 GW
2 GW
2 GW
2 GW
2 GW
2 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
CENTRAL RES HIGH RES VERY HIGH PV
REFERENCELOW PV CAPEX
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
4 GW
4 GW
4 GW
4 GW
4 GW
4 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
12 GW
12 GW
12 GW
12 GW
12 GW
12 GW
16 GW
16 GW
16 GW
16 GW
16 GW
16 GW
119
116
114
109
117
109
123
120
114
110
114
106
117
114
115
110
117
108
114
112
110
105
111
104
110
108
107
102
106
97
€
It can be observed that:
◆ Going beyond the central RES assumptions to the high RES
assumptions always results in a decrease in total system cost.
This decrease in costs is higher for the sensitivities with low
amounts of non-domestic offshore integrated into the Bel-
gian electricity system and low amounts of new nuclear. This
intuitively makes sense as in the scenarios with lower domes-
tic generation electricity prices will tend to be higher and
more expensive imports can be replaced by domestic RES
production.
◆ While the benefits of installing additional solar PV diminish
with increased domestic supply, they decrease at a slower
rate with an increase in non-domestic offshore supply com-
pared to an increase in domestic nuclear. Indeed, the vari-
able nature of wind and its seasonal complementarity with
solar (more wind is produced on average in winter while solar
energy production peaks in winter) means that they impact
each other’s benefits less.
◆ Increasing the PV assumptions even further to the very high
PV assumptions is beneficial (or neutral) when using low
CAPEX assumptions. However, when reference CAPEX num-
bers are used the results are positive only if no additional
nuclear is installed.
185
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium184
MANAGING A MASSIVE PENETRATION OF SOLAR PV
One of the sensitivities assessed in this study models a
massive increase in PV capacity, reaching almost 100 GW
in 2050. Without considering any grid limitations, the
production of PV would be directly proportional to the
installed capacity.
When considering a very high penetration of solar PV,
the peak generation could become higher than the con-
sumption peak on which distribution networks are dimen-
sioned. This raises the question of whether it is more ben-
eficial for society to limit photovoltaic production in a set
of overproduction hours or to invest in accommodating
the full photovoltaic production. This study explores this
effect by examining a scenario where the photovoltaic
production is capped to ensure no reinforcements have to
be made to the DSO grid to accommodate these large vol-
umes of photovoltaics on top of the investments already
needed for the increase of the peak demand. A simplified
reasoning is applied and is further explained below.
Two scenarios are used when considering very high PV
generation:
◆ One where there is no capping assumed;
◆ One where the generation is capped (before the dis-
patch) and associated to additional local storage. This
is the scenario that is accounted for when providing
results for the high PV scenario.
By capping the production, the aim is to limit the cost
associated to the additional strengthening of the DSO
network for overproduction. The capping considers a
threshold production for the photovoltaics on a national
level and is applied to the assumed residential PV installed
capacity. Above this threshold, it is assumed that excess
solar production can be stored in small-scale batteries
dedicated to reducing curtailment which were added to
the very high PV capped sensitivity. Any energy which
cannot be evacuated through the DSO grid or absorbed
into batteries is curtailed. It is assumed the battery is dis-
charged as soon as possible but without exceeding the
energy evacuation threshold. This optimisation happens
before the dispatch and aims to maximise the PV gen-
eration.
In the case of ‘no capping’, additional costs should be
accounted for in order to integrate the additional PV
capacity. This amount was not calculated however it
could significantly raise the costs of the system. Figure
5-31 presents three distinct cases for one climate year of
the Météo France climate database 2050. Case 1 repre-
sents the situation where neither curtailment nor storage
is necessary to respect the threshold. Case 2 represents
the situation where the battery capacity is sufficient to
prevent any curtailment while still respecting the thresh-
old. Case 3 represents the situation where the size of the
battery falls short in preventing curtailment.
COMPARISON OF PV GENERATION IN DIFFERENT SITUATIONS: NO CURTAILMENT
OR STORAGE, SUFFICIENT BATTERY CAPACITY TO AVOID CURTAILMENT, AND
NECESSARY CURTAILMENT TO RESPECT MAXIMUM OUTPUT LIMITATION FIGURE 531
1
0.9
0.8
0.7
0.6
0.5
0.5
0.4
0.3
0.2
0.1
0
Photovoltaic capacity factor [-]
0 4 8 12 16 20 0 4 8 12 16 20 0 4 8 12 16 20
Hour of the day [-]
PV production Storage charge Storage discharge Curtailment
Case 1 :
Photovoltaics production does not exceed the
threshold
Case 2 :
Photovoltaics production exceeds
the threshold, but battery storage is
sufficient to avoid curtailment
Case 3 :
Photovoltaics production exceeds the threshold,
battery storage is not sufficient to avoid
curtailment
With substantial volumes of PV generation in Europe/
Belgium, other system aspects also need to be consid-
ered and enhanced. Managing the short-term variability
(due to i.e. prediction and forecasting errors), is one such
aspect that will become increasingly critical with the
heightened penetration of RES, particularly PV given its
decentralized nature. This element was not examined in
this study, yet it is crucial for the future electricity system
and must be factored in.
Finally, Figure 5-32 shows the comparison of the yearly
photovoltaics production for the very high PV scenario
without and with a production cap for 2036, 2040 and
2050. Note that the amount of assumed storage is based
on the installed PV capacity at 1 GW per 3 GW of differ-
ence between the nominal power of the PV and the DSO
accomodation capacity. Obviously, if the storage size is
increased, the curtailed energy is reduced.
The maximum output limitation is introduced to limit
additional reinforcements of the DSO grid, but the results
obtained are close to the results of the market optimisa-
tion. Indeed, some peak PV production is also curtailed as
a result of the market dispatch in the very high PV sce-
nario as seen in Figure 5-32.
◆ In 2036, the required curtailment as an effect of the
capping is negligible: it is only required for an average
of 75 hours per year over 200 climate years. With battery
storage, solar curtailment is minimal. The energy cur-
tailed through the market dispatch process of the very
high PV sensitivity without capping is also negligible.
◆ By 2040, the impact of the capping grows. Curtailment
is applied for an average of 450 hours per year over 200
climate years, resulting in 1.7 TWh of curtailed energy,
or 3% of total production. Note that due to the market
dispatch in the very high PV sensitivity without cap-
ping, finally a similar overall curtailment occurs com-
pared to the scenario with capping.
◆ In 2050, capping has a more important effect, and
would be activated for an average of 1075 hours per
year over 200 climate years. This leads to around 11 TWh
of curtailed energy to respect the maximum threshold.
When no local storage is added, around 3 TWh is addi-
tionally curtailed. Finally, the market dispatch in the
very high PV sensitivity with capping curtails 4 addi-
tional TWh. Note that due to the market dispatch in the
very high PV sensitivity without capping, finally roughly
the same overall curtailment occurs compared to the
scenario with capping.
CURTAILMENT OF PV GENERATION IN THE VERY HIGH PV SENSITIVITY WITHOUT
CAPPING LEFT AND WITH CAPPING RIGHT. FIGURE 532
100
80
60
40
20
0
Photovoltaics yearly production [TWh]
Curtailment
Curtailment if no
storage
Market curtailment
(simulation output)
Photovoltaics
production after
curtailment if no
additional storage
2036
Very high PV - No maximum constraint Very high PV capped to maximum constraint
2040 205020362040 2050
3 TWh 3 TWh
15-20 TWh 17-20 TWh
Total PV production
BOX 5-5
187
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium186
RAMPING
Finally, due to the fluctuating nature of renewable generation,
integrating more renewables into the system causes an increase
in the variability of the electricity residual demand. Flexible tech-
nologies capable of smoothening out these fluctuations will
therefore become key in the future. The 90th percentile ramps
over one and three hours are indicators which can be used to
quantify supply fluctuations in the system. The selected percen-
tile and indicators are chosen to provide an idea of the increased
need for intraday flexibility or ramping between hours. However,
it should not be inferred that these indicators should be used for
system dimensioning.
It should be noted that the simulation set-up used is based on a
perfect foresight model. As such, it does not inherently capture
forecast errors in renewable production, consumption or varia-
tions due to power plant outages. These forecast errors increase
the need for flexibility and as such need to be considered when
dimensioning and managing the actual system. Similarly, the
concept of 'short-term flexibility', as evaluated in the Adequacy
& Flexibility study, is not assessed in the current study.
The resulting 1-hour and 3-hour ramping ranges over all domestic
supply scenarios is presented in Figure 5-34. It can be observed
that:
◆ Ramping tends to increase over time with the scenario with
the most domestic renewables showing the largest ramping.
This is mainly linked to the increase in renewables and more
particularly PV;
◆ The ramping rate is significantly reduced in 2050 in the high
PV capped (HPVC - where the PV generation is capped to a
certain value) when compared to the high PV (HPV - where no
capping is applied on PV). This highlights the impact of PV on
the rampings of the residual load (see also BOX 5-5 for more
information on how the HPVC scenario was constructed);
◆ The nominal capacity of storage assumed is higher than the
1h ramping for all scenarios and time horizons. This means
that it is capable of covering most of the 1h ramping (if the
state of charge allows it);
◆ The 3-hour ramping is higher than the 1-hour ramping. This is
mainly due to morning and evening rampings (linked to PV
and consumption). However for the 3-hour ramping many
other technologies are able to provide such a service for the
system;
◆ The figure only compares the rampings to the installed stor-
age in the scenarios. This therefore excludes other types of,
interconnectors or demand flexibility.
◆ The figure demonstrates that the flexibility of the electricity
system will be needed to cope with intraday variations in the
residual demand.
HOURLY AND 3HOURLY RAMPINGS OF THE RESIDUAL DEMAND FIGURE 534
1h ramping - 90th percentile 3h ramping - 90th percentile
Domestic (excludes non-domestic offshore) residual ramping (90
th
percentile for each scenario) calculated in perfect foresight: demand minus domestic renewables including
dispatch of demand flexibility.
This does not account for flexibility needs due to short term deviations (forecast errors, outages...).
25
20
15
10
5
0
25
20
15
10
5
0
2020-25 2020-252036 20362040 20402050 2050
[GW/hour]
[GW/hour]
Nominal installed storage
Nominal installed
storage
HPV
HPVC
HRES
CEN
HPV
HPVC
HRES
CEN
5.4.5. ADDITIONAL INDICATORS
RES CURTAILMENT AND LOW MARGINAL PRICE SITUATIONS
Next to the costs per MWh several other indicators characterising
the electricity system can be extracted from the economic dis-
patch simulation. In this section indicators related to moments
where electricity prices are low are explored further.
First of all, indicators such as the number of hours where RES is
curtailed in the market dispatch and the number of hours with
low prices give an idea of how much renewable energy was pro-
duced by renewable energy sources but not consumed and how
many hours exist where flexible technologies could benefit from
very low electricity prices.
The number of hours with RES curtailment and with marginal
prices below €20/MWh are presented in Figure 5-33. It can be
seen that increasing the amount of domestic RES results in both
more hours with low marginal prices and more RES curtailment.
Capping the solar production in the Very High PV scenario (as
described in BOX 5-5) results in approximately a halving (depend-
ing on the scenario) of the hours with RES curtailment in the mar-
ket dispatch while keeping the number of hours with marginal
prices below €20/MWh approximately the same.
HOURS WITH RES CURTAILMENT AND MARGINAL PRICES BELOW 20 €MWH FOR SUPPLY
SENSITIVITIES PERFORMED FOR THE DE SCENARIO IN 2050 FIGURE 533
60
375
550
850
2050
2000
1210
1600
75
390
550
850
2400
2320
1720
2060
30
285
370
680
1980
1900
1230
1630
15
220
320
660
1600
1500
890
1220
15
220
320
660
1950
1900
1300
1680
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
4 GW
4 GW
4 GW
4 GW
4 GW
4 GW
4 GW
4 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
16 GW
16 GW
16 GW
16 GW
16 GW
16 GW
16 GW
16 GW
Hours with RES curtailment Hours with marginal prices <20€/MWh
2050 - DE scenario
CENTRAL RES
High RES
High PV (capped)
High PV
189
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium188
5.4.6. TOTAL SYSTEM COSTS (ALL VECTORS)
The previous sections analysed the electricity system costs only.
Those costs included the costs from other vectors to produce
the needed electricity or the benefits incurred from producing
molecules from electricity. This method allows different electricity
supply options to be compared.
In order to grasp the total energy system costs, an exercise was
completed to evaluate what the other aspects of the Belgian
energy system would cost. This also includes the cost of end-use
investments. As explained in Section 2.4, those costs can be split
into three main categories:
◆ End uses investments (e.g. investments in electro-mobility,
energy efficiency…):
-
Those are the investments made by the end users of the
energy and include the cost for acquisition of new cars,
charging infrastructure, heating devices or building reno-
vations.
◆ Energy vector – molecules (oil, methane…):
- Costs related to molecule infrastructure (grid and transfor-
mation processes);
-
Costs related to the molecule supply (e.g. imports, domestic
production but excluding the fuel used for electricity gener-
ation and fuel generated from electricity);
◆ Energy vectors – electricity:
-
Electricity grid costs (offshore, interconnectors, backbone
(high voltage grid within a zone), regional grid (also called
‘vertical’ grid: 150-70-36-30 kV), DSO grids);
- Electricity supply capital expenditure (CAPEX) costs (invest-
ments and fixed costs of production facilities);
-
Electricity supply operational expenditure (OPEX) costs (fuel
used to generate electricity).
Other types of benefits or costs (e.g. socioeconomic impact,
employment …) are excluded from the analysis. The main assump-
tions behind the costs are further detailed in Section 3.3.
Figure 5-35 compares the annual costs of these components in
2050 for the three main demand scenarios and ‘Current Policies’
electricity supply scenario. It can be observed that:
◆ Electricity system costs would only represent between 19%
and 23% of the annual spending of the energy system by 2050;
◆ Even though the consumption of molecules is strongly
reduced in all demand scenarios, the costs remain important.
This can be explained by the fact that by 2050 most (imported)
molecules consist of relatively expensive green mole-
cules such as ammonia, biomethane and synthetic liquids
when compared to (cheaper) fossil fuels today;
◆ In all scenarios, the cost for end use sectors generally accounts
for the largest relative share of the total. The primary drivers
in these sectors are the costs associated with renovations and
the replacement of heating devices in buildings, as well as
the necessary investments for renewing the vehicle fleet and
rolling out new infrastructure such as electric vehicle charg-
ing stations in the transport sector. In industry, the costs are
mainly driven by the replacement of fossil fuel based appli-
ances by electric/green molecule-based alternatives and the
investment in carbon capture technologies;
◆ In general, the scenarios which involve a higher level of elec-
trification result in lower total costs. A higher degree of elec-
trification requires more investments in the power system for
grid, (backup) capacity and operational costs; however, this
is more than compensated for by the fact that the reduced
amount of required domestic and imported molecules
reduces the cost of the molecule system. Additionally, the GA
scenario assumes some end use technologies that are rela-
tively more expensive than their electrical counterparts, such
as fuel cell vehicles and hydrogen heating in buildings.
TOTAL SYSTEM COST FOR BELGIUM, EXPRESSED IN B€YEAR FIGURE 535
ELEC
!"#$%&'""%#"%(#&%)**"! "+#%&'""%#"%(#&%)**" !",-%&'""%#"%(#&%)**"
2050
DE GA
120
100
80
60
40
20
0
B€/year
Industry*
113
104
97
Buildings
Transport
Molecules
Power**
End-uses
Results for the ‘Current Policies' electricity supply scenario (no new nuclear, non-domestic offshore connected to BE
and central onshore RES trajectory)
* Includes the cost for carbon capture
** Includes the cost of methane and H2 used for power generation (excluded in molecules)
5.5. TRANSITION PERIOD
(THE ROAD TO 2050)
5.5.1. ELECTRICITY MIX DASHBOARD FOR 2036 AND 2040
Where the previous section focused on the long-term (2050) elec-
tricity options for Belgium and the choices that are to be made,
this section focuses on the period leading up to 2050. Studying
the intermediate period is crucial as it paves the road to 2050. Not
taking action early enough may result in the exclusion of certain
options for 2050. In addition, access to affordable, clean energy
is as crucial in the period leading up to 2050 as it is after 2050.
During this period, Belgium also has access to some options which
will likely not be available in 2050. For example, the extension of
existing nuclear plants will not be possible beyond a certain date.
Similar to the analysis for 2050, an overview of the different supply
options considered and their effect is shown in Figure 5-36 for
the DE scenario in 2036 and Figure 5-37 for 2040. The figures only
show some of the possible combinations. Extension options for
existing nuclear units (beyond 2036) are discussed in Section 5.5.3.
ELECTRICITY MIX DASHBOARD FOR 2036 DE SCENARIO FIGURE 536
Capacity per scenario
Nuclear
Foreign
offshore DRES
Thermal
in GW Domestic + foreign offshore Exports [-] / Imports [+]
Share hours per yr.
with curtailment
Share hours per yr.
Below 20 €/MWh
Supply mix in TWh Curtailment Low prices
Thermal-Existing Thermal-New Foreign offshore
Nuclear DRES Thermal Gap-to-max Net imports Net exports Gross exports Gross imports
0 GW
0.5 GW
0 GW
0 GW
4 GW
4 GW
Central
Central
High
High
High
High
Central
Central
9
9
9
9
9
9
9
9
3
2
1
3
2
1
2
2
3
3
3
3
65 24 67 0.5% 14%
1.2% 16%
0.5% 16%
65 0.5% 15%
-23
-24
62
61 1.3% 17%
-26
-28
59
57 0.6% 16%
-28
-29
54 1.2% 18%
53 1.3% 18%
-31
-33
65 23
73 24
14
14 73 23
65 24
73 23
65 2214
73 2214
14
14
14
14
ELECTRICITY MIX DASHBOARD FOR 2040 DE SCENARIO FIGURE 537
Capacity per scenario
Nuclear
Foreign
offshore DRES
Thermal
in GW Domestic + foreign offshore Exports [-] / Imports [+]
Share hours per yr.
with curtailment
Share hours per yr.
Below 20 €/MWh
Supply mix in TWh Curtailment Low prices
90 0.0% 9%
-28
81 0.6% 12%
-35
73 0.1% 13%
-35
63 0.8% 16%
-41
77 0.1% 11%
-34
67 0.8% 14%
-39
82 0.1% 10%
-31
73 0.9% 13%
-38
65 0.2% 14%
-39
59 1.1% 17%
-42
Thermal-Existing Thermal-New Foreign offshoreNuclear DRES Thermal Gap-to-max Net imports Net exports Gross exports Gross imports
0 GW
2 GW
1 GW
0 GW
0 GW
4 GW
8 GW
8 GW
Central
Central
Central
High
High
High
High
High
Central
Central
7
7
7
7
7
7
7
7
7
7
6
4
4
3
6
4
4
3
5
5
82 21
21
18
17
18
18
16
15
19
18
98
29
29
16
15
29
29
81
98
81
98
81
97
81
98
14
7
14
7
14
14
191
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium190
5.5.2. ELECTRICITY SYSTEM COSTS
REFERENCE COSTS AND WACC
Similarly to Section 5.4.3, the total electricity system costs in €/
MWh for the reference CAPEX and WACC (7%) assumptions are
shown for the three target years. Figure 5-38 first shows the results
for the Central RES scenario on top. The level of non-domestic
offshore wind is shown on the x-axis and the level of the new
nuclear generation on the y-axis. The bottom graphs depict the
results for the High RES scenario. Note that for this figure no
nuclear extensions are shown, those are presented in Section 5.5.3.
In general:
The cost range increases as the target year is moved further into
the future. This intuitively makes sense as in 2050 the range of
possible supply mixes is bigger than in earlier years. It should how-
ever not be forgotten that not investing early enough may result
in supply options in 2050 becoming unreachable. In addition,
investing early means that there are more years where society
can benefit from the reduction in costs. Therefore, the decisions
made for 2036 may seem less impactful from the figure but they
certainly are when looking at the long-term numbers.
For 2036 with reference costs and WACC:
It can be observed that the total system costs range between
€98/MWh and €100/MWh depending on the different sensi-
tivities assessed. A few observations can be made (from a costs
point of view):
◆ High domestic RES never results in an increase in costs and,
depending on the level of new nuclear and non-domestic off-
shore wind, allows the system costs to be reduced by up to
€2/MWh.
For 2040 with reference costs and WACC:
It can be observed that the costs range between €102/MWh and
€108/MWh depending on the different sensitivities assessed.
Based on those a few observations can be made:
◆ High domestic RES never results in an increase in costs and,
depending on the level of new nuclear and foreign offshore
wind, allows allows the system costs to be reduced between
€2 to €4/MWh.
◆ In the Central domestic RES scenario, no new nuclear or
no new offshore wind is always more expensive than doing
one or a combination of both. This means that reducing the
amount of net imports allows Belgium to reduce its total elec-
tricity system costs.
◆ Investing in non-domestic offshore wind is always more
interesting than not doing it if no new nuclear is installed. If
new nuclear is installed the benefits of installing additional
offshore are not as clear.
The results for 2050 were discussed in Section 5.4.3 and are not
discussed again here.
TOTAL ELECTRICITY SYSTEM COSTS FOR THE DE SCENARIO IN €MWH. FIGURE 538
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
4 GW
4 GW
4 GW
4 GW
4 GW
4 GW
6 GW
6 GW
6 GW
6 GW
6 GW
6 GW
2 GW
2 GW
2 GW
2 GW
2 GW
2 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
2036 2040 2050
CENTRAL RESHIGH RES
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
4 GW
4 GW
4 GW
4 GW
4 GW
4 GW
8 GW
8 GW
8 GW
8 GW
8 GW
8 GW
12 GW
12 GW
12 GW
12 GW
12 GW
12 GW
16 GW
16 GW
16 GW
16 GW
16 GW
16 GW
119
114
118
113
118
112
120
114
117
114
117
108
114
110
100
98
107
104
106
102
106
103
107
103
105
103
106
103
100
99
100
98
123
114
117
115
112
109
114
110
115
109
110
107
99
99
108
104
106
102
105
102
99
98
99
98
5.5.3. NUCLEAR EXTENSIONS
The possible extension of existing nuclear plants was considered
in the intermediate period. For both 2036 and 2040 an extension
of 0, 2, 3 and 4 GW was simulated, combined with different levels
of non-domestic offshore wind integrated into the Belgian elec-
tricity system. In addition, a sensitivity on the costs was assessed
where the cost of the extension was raised from €1000/kW to
€1200/kW and the WACC was increased from 7% to 10%. It is
important to repeat that a decision about nuclear extensions
should consider a broader set of criteria than only costs (feasibility,
safety, regulation, grids, socioeconomics, etc.).
The results are presented in Figure 5-39. It can be seen that the
extension of existing nuclear:
◆ Makes sense from a cost point of view in all studied sensitivi-
ties under the central cost assumptions;
◆ Results in the biggest reduction in total cost for the sensitiv-
ities with higher levels of imports. In these sensitivities the
additional electricity generation is used to meet a large part
of the domestic load, resulting in benefits for consumers as
well as for producers being captured in Belgium. This can also
be observed in the reduction in import volumes;
◆ Results in a significant adequacy benefit, resulting in a lower
volume of new thermal that needs to be installed to ensure
that the security of supply criteria are met. The reduction in
costs that this entails are included in the presented numbers;
◆ Results in bigger benefits in 2040, where the growth of elec-
trical demand results in higher levels of import if no addi-
tional domestic supply is built on top of the central buildout
of renewables;
◆ Still makes sense or is neutral from a cost point of view when
higher cost assumptions are used.
NUCLEAR EXTENSION COSTS, NET IMPORTS AND NEW THERMAL NEEDED FOR THE DE SCENARIO FIGURE 539
2036
2036
2040
2040
Total electricity system costs [€/MWh] Total electricity system costs [€/MWh]
1000 EUR/kW - 7%
Net Import Net Import
1000 EUR/kW - 7%1200 EUR/kW - 10%
New thermal need New thermal need
1200 EUR/kW - 10%
99 10899 10898 10598 105
4GW
4GW 22 TWh 39 TWh11 TWh 14 TWh
4GW
4GW
3 GW
3 GW 26 TWh 44 TWh
0 GW 0 GW
0 GW 0 GW
0 GW
14 TWh 19 TWh
3 GW
3 GW
2GW
2GW 31 TWh 55 TWh0.4 GW
-2 to -5
-2 to -5
-2 to -4
-9 to -12
-7 to -10
-5 to -7
4.0 GW 2.4 GW
3 GW 1.4 GW
2 GW 0.4 GW
19 TWh
-4 to -7
-4 to -7
-3 to -5
-11 to -14
-8 to -11
-6 to -8
25 TWh
2GW
2GW
0 GW
0 GW 43 TWh 62 TWh
2.5 GW 6.1 GW31 TWh 37 TWh1.7 GW 4.5 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
0 GW
4 GW
4 GW
8 GW
8 GW
4 GW
4 GW
8 GW
8 GW
€ €
extension of
impact when
extending…
impact when
extending…
extension of
193
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium192
5.5.4. ADEQUACY
In order to ensure security of supply criteria are met in all the sen-
sitivities, it is sometimes necessary to add firm backup capacity.
The cost of this additional capacity is already taken into account
in the electricity system costs presented earlier in this chapter.
Figure 5-40 shows the remaining capacity needs (excluding addi
-
tional domestic supply of nuclear or non-domestic offshore; i.e.
in the ‘Current Policies’ scenario) for each demand scenario and
target year.
DERATED CONTRIBUTION OF THERMAL AND FLEXIBILITY OPTIONS TO ADEQUACY FIGURE 540
30
25
20
15
10
5
0
Derated capacity [GW]
2000 2005 2010 2015 2020 2025 2030 2036 2040 2045 2050
Period covered by
the CRM and by
theAdequacy &
Flexibliity study
Simulated period
Oil
Coal NUC
0 to 3 GW
1.5 to 7 GW
6 to 10 GW
ELEC
DE
GA
SUFF
Flex
R
Flex I
Flex I
Molecule
fired
BAT & PSP
BAT
Gas
≈ 4.5 GW
≈ 2 GW
≈ 11 GW Assumed new Flex in
the Central scenario
Existing Flex & Storage
estimated in 2025
Existing thermal on
gas with assumed
decomissionings (old
units)
Assumed new thermal
in the Central scenario
(no nuclear, no new
non-domestic offshore
wind). However, this
can be additional flex
and storage aswell.
Derated need in Central RES scenario (no
non-domestic offshore wind, no nuclear)
but with solar PV, onshore wind and
interconnection contribution
Year 2045 is not simulated.
Legend : Flex R = flexibility in the residential sector; Flex
I = flexibility in industry sector; BAT = batteries; PSP =
Pump-Storage Plant
New derated capcity assumed.
Existing derated capacity estimated in 2025 for Flex & Storage,
and with assumed decommissioning for the thermal fleet.
Several observations can be made:
◆ The need for firm capacity increases towards the later years.
This increase is mostly driven by the increased electrical
demand given the electrification assumptions used in this
study;
◆ The assumed existing domestic thermal capacity still present
by 2050 in the central scenario covers (only) approximately 4.5
GW of the total need of 23.5 to 27.5 GW. In the supply sensitiv-
ities the addition of new nuclear, extension of existing nuclear
or addition of renewable energy is taken into account;
◆ The growing need for firm capacity is partially compensated
by the assumed growth in installed batteries and the unlock-
ing of flexibility in the building, transport and industrial sec-
tors. In 2050 flexibility covers 13 GW of the firm capacity needs
which would otherwise need to be met with other sources;
◆ Depending on the load scenario a higher or lower volume of
firm capacity is needed. The SUFF scenario where behavioral
changes result in a lower electricity demand has the lowest
requirements. The ELEC scenario where the electricity con-
sumption is highest out of all demand scenarios results in the
highest need for firm capacity.
ENERGY-LIMITED RESOURCES’ ADEQUACY DERATING FACTORS
As flexibility and RES increase across the system, the ade-
quacy derating factors (the contribution of storage and
energy limited resources to adequacy) are expected to
decrease. Their exact contribution will depend on the
amount of energy limited resources that are installed in
the future system (both in Belgium and abroad). In order
to illustrate the impact that the amount of flexibility in the
system will have on the contribution to adequacy (i.e. the
derating factor), Figure 5-41 provides indicative values for
several storage durations as well as ranges depending on
the scenarios accounted for in terms of installed flexibility.
With the flexiblity assumptions used in the DE scenario
the derating of batteries decreases towards later time
horizons. The decrease over time can be mainly explained
by the higher share of RES in the system and more flexi-
bility at European level.
In the 0 GW battery in Belgium sensitivity it can be
observed that the derating factor for storage remains rel-
atively high from 2036 to 2050. Nevertheless a decrease
can be observed which is driven by the same elements as
in the DE scenario.
In the High FLEX sensitivty the installed capacity of bat-
teries represents up to 40% of the average daily peak load
and the derating factor for a 4h large-scale battery is equal
to 25% for 2036/2040. In 2050, this derating decreases to
about 15%, as the ratio of installed batteries capacity to
average daily peak load increases to more than 50%. This
means that 6.3 GW of 4h storage could be used to fill 1
GW of adequacy GAP. For 2h storage, the needed volume
would be close to 10 GW.
CONTRIBUTION OF STORAGE TECHNOLOGIES TO ADEQUACY FIGURE 541
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
100%
50%
0%
Derating Factor [%]
Ratio [%]
Decreasing the
battery volume
in Belgium to 0
DE
Increasing
the flexibility
in Belgium to
HFLEX
It means 1GW of
adequacy GAP
can be filled by :
4.1 GW of
batteries
2036 2040 2050
Ratio storage* volume on average hourly peak load [%]
DE HFLEX
Battery 2h Battery 4h Battery 6h Battery 8h
6.3 GW of
batteries
*Storage including large-scale batteries, pumped-storage, small-scale batteries and V2G
BOX 5-6
195
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium194
5.6. SUMMARY OF THE DIFFERENT LEVERS
As observed in the sections above, decisions about domestic supply can never be fully decoupled. For example, the presence of more
onshore renewable generation has an impact on the benefits obtained when adding new non-domestic offshore, nuclear generation
or the extension of existing nuclear. Each of the supply and demand levers impact in some way or another the benefits of the others.
Nevertheless, it is possible to identify general trends by assessing the effect of the levers on each of the sensitivities.
FOR 2050:
Figure 5-42 shows the impact on system costs of each lever per
MWh of demand in the DE scenario when applied to the current
policy scenario (circle). To represent the effect of the lever under
other sensitivities (more nuclear, more non-domestic offshore, ...)
the supply lever was applied on a diverse set of sensitivities and
analysed for each. The average of the resulting effects is shown
using the diamond. By including both indicators, a view of the
scenario robustness is provided for assessing trends.
In addition to assessing the effect of the levers for the reference
costs, the effects of each lever are also analysed when the costs for
the selected lever turn out to the high or low assumptions instead
of the central costs (again for the set of supply and demand sen-
sitivities). This also allows for a robustness check of trends with
respect to changes in costs.
Finally, the figure depicts the impact on net imports and firm
capacity needs (to respect security of supply criteria) of each
sensitivity. The effect of these changes is taken into account
when calculating the system costs. It should be noted that the
district heating and sufficiency levers are assessed but no cost
is accounted for the buildout of heating networks nor for the
application of sufficiency measures. These costs should also be
considered when assessing the desirability of these options.
The figure offers several interesting insights when looking at the
costs (and associated sensitivities), net imports and firm capacity
requirements:
SUMMARY OF THE LONGTERM IMPACT OF THE DIFFERENT DEMAND
AND SUPPLY LEVERS DE 2050 FIGURE 542
Impact on
Net imports
in TWh
Impact on
Firm capacity
needs in GW
Impact on Electricity system Costs in €/MWh
Impact of a given sensitivity when applied to current policies scenario ( ) or average impact across supply sensitivities ( )
Sensitivities
2050
DRESLarge-scale Demand
Reference High costs Low costs
The district heating & sufficiency sensitivities only
consider potential benefits, cost implications are
not assessed
District heating
15 TWh
-20.0
-8.3 -4.6 -10.3
-10
-6.5
-6.1
-11.1
-10.9
-18.3
+3.3
+12.6
+10.4
+22.2
-2.6
-1.6
-5.3
-7.9
-12.3
-2.2
-2.4
-4.3
-7.0
-15.0
-4.8 -1.1 -6.8
-3 -1.0
-2.0
0
0
-3.8
-3.8
-7.8
-1.6
-2.4
-15
-29
-21
-21
-47
-45
-25
-51
-5.4
-3.9
-3.5
-3.9
-7.1
-11.8
+6.6
+15.1
+12.8
+29.2
+1.0
+4.8
-0.1
-4.1
-5.8
+0.4
+0.1
+2.9
-6.0
Sufficiency
-20 TWh
High DRES
High DRES
+PV
Far out baseload
RES 4 GW
New nuclear
4 GW
New nuclear
8.2 GW
Non-domestic
Offshore 8 GW
Non-domestic
Offshore 16 GW
No-regret levers:
◆ Increasing the ambitions for domestic RES to the values
assumed in the High DRES scenario is beneficial even when
the costs turn out to be higher. From a cost point of view, this
lever is thus a no-regret decision.
Minimum-regret levers:
◆ Increasing the ambitions for domestic PV even further (High
DRES + PV) is beneficial except when costs turn out at the
high estimate. Therefore, pushing the development of PV
beyond even the High DRES assumptions should be con-
sidered as a minimum regret while requiring a monitoring
of the costs. The integration of large amounts of PV into the
distribution grids and the flexibility requirements associated
to large volumes of PV are two aspects requiring important
attention when applying this lever;
◆ The integration of 8 up to 16 GW of non-domestic offshore
wind results in overall net benefits except for the scenario
where costs turn out to be at the high estimate, similar to
High DRES + PV. On top of this, the integration of non-domes-
tic offshore wind results in a significantly decreased reliance
on electricity imports.
Levers that can be advantageous depending on the refer-
ence sensitivity:
◆ The desirability of integrating new nuclear and/or far-out base-
load RES depends on the scenario in the central costs case.
If the high costs materialise these options appear to be not
financially attractive. Inversely, when using low-cost assump-
tions these options are beneficial in all scenarios. It should be
noted that these options do have a significant impact on the
net imports needed in Belgium and the required amount of
other firm capacity that is to be installed, and can have other
societal advantages as well.
Levers whose costs need to be assessed further, but look
promising:
◆ Both the district heating (DH) and sufficiency (SUFF) sensi-
tivities result in lower electrical demand and therefore end
up costing less overall. The costs associated with these levers
were however not assessed in this study and as such remain
to be further investigated.
FOR 2040 BUT ALSO REPRESENTATIVE OF THE INTERMEDIARY PERIOD:
In 2040 an additional lever – the extension of existing nuclear –
is considered. The results are shown in the same format as they
were for 2050 in Figure 5-43. When interpreting these results, it
should be kept in mind that depending on the decisions made
for the intermediary period some options for 2050 might become
unattainable.
The following trends can be observed:
No-regret levers:
◆ Similarly to 2050, high domestic RES remains a no-regret
lever.
◆ The extension of existing nuclear appears to be a no-regret
lever from a cost point of view even if high cost estimates
are used. In addition, this lever, like the installation of new
nuclear, has a significant impact on firm capacity needs and
net imports. However other aspects not covered in this study
(such as safety, regulation, grids, etc) should be taken into
account as well.
Minimum-regret levers:
◆ The integration of 8 GW of non-domestic offshore wind is
beneficial except when high cost assumptions are used. On
top of this, the integration of non-domestic offshore results
in a significant decreased need for foreign electricity imports.
Levers that can be advantageous depending the reference
sensitivity
◆ From a cost perspective, the buildout of additional new
nuclear appears only to be beneficial in some sensitivities.
Levers which are advantageous only if low cost assump-
tions are used:
◆ Increasing the solar capacity ambitions to levels beyond those
assumed in the High DRES scenario seems to be beneficial
only if costs lower than the reference costs are assumed.
Levers whose costs still need to be assessed further:
◆ As for 2050, the SUFF sensitivity is assessed for 2040, similarly
showing significant benefits if it were to materialise. These
benefits are to be compared to cost calculations. The costs
calculations for the implementation of SUFF are not per-
formed in the framework of this study.
197
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium196
SUMMARY OF THE IMPACT OF THE DIFFERENT LEVERS FOR THE TRANSITION PERIOD DE 2040 FIGURE 543
Impact on
Net imports
in TWh
Impact on
Firm capacity
needs in GW
Impact on Electricity system Costs in €/MWh
Sensitivities
2040
DRESLarge-scale
Reference High costs Low costs
The sufficiency sensitivity only consider potential
benefits, cost implications are not assessed
Demand:
Sufficiency -20 TWh -20.0
-4.2 -2.5
-6.7 -8.5
-3.3
-5.1
-15 -2.0
0.0
0.0
-2.0
-1.6
-1.9
-3.8
-15
-19
-11
-12
-25
-23
-12.5 -15.9
-6.9
-4.5
+0.4 +3.3
+4.9
+1.6
-1.1
-2.7
-7.7
-14.4
-15.0
-2.9 -1.2
-5.6 -7.4
-2.0
-3.8
-10.3 -13.7
-4.7
-1.7
+1.1 +1.7
+6.1
+3.7
+0.2
-0.5
-6.6
-12.2
High DRES
High DRES +PV
New Nuclear
2 GW
Non-domestic
Offshore
8 GW
Extension
Nuclear 2 GW
Extension
Nuclear 4 GW
Impact of a given sensitivity when applied to current policies scenario ( ) or average impact across supply sensitivities ( )
IN CONCLUSION
Using the framework developed in this study, the effect of several
levers on the energy costs, import needs and need for firm capac
-
ity of the Belgian electricity system are assessed. The acceleration
of domestic onshore RES to the high levels assumed in this
study appear beneficial in all cases. As such, it is highly advisable
to investigate how these levels of renewables could be attained.
Furthermore, the extension of existing nuclear, if all related
aspects (which are not further developed in this study) allow it,
also presents an interesting opportunity to reduce system costs
in the period leading up to 2050. The buildout of non-domes-
tic offshore wind integrated into the Belgian energy system
also shows great promise in terms of reducing system costs and
should be further investigated. For the buildout of new nuclear
the results are less straightforward and depend to an important
extent on the cost assumptions taken.
Finally, the impact of sufficiency measures and district heating
show important potential to further reduce system costs, but it
should be noted that the costs associated with implementing
these measures are not calculated in this study and as such addi-
tional analysis is to be performed.
Finally, Elia would like to reiterate that the insights provided are
based about and limited to the area of expertise of Elia. Making
decisions about the future energy system is a complex multi-di-
mensional exercise where not only system costs but also the
implications for other societal aspects such as the environment,
public acceptance and budgetary constraints are to be taken
into account. It is the full prerogative of policymakers to define
the principles of the Belgian electricity supply mix for 2050, and
the pathway to reach it.
5.7. ELECTRICITY GRID
Electricity grids are an essential part of the energy system. They
support society in its transition to net zero by facilitating:
◆ the integration of onshore and offshore renewable genera-
tion and other low-carbon electricity sources into the energy
system;
◆ the efficient exchange (import and export) of green electricity
with neighbouring countries;
◆ large-scale electrification, both on an industrial (e.g. steel sec-
tor, data centres, etc.) and residential level (e.g. transportation,
residential heating, etc.);
◆ the integration of the necessary (physical) means for ade-
quacy, such as storage systems and backup generation.
Based on the insights gained from this study, this section explores
the potential impact on the future evolution of the Belgian
electricity grid and the concrete infrastructure projects for grid
reinforcement that are to be expected. The development of the
Belgian electricity system consists of three key pillars which are
supported by a strong foundation, as follows:
◆ Pillar 1: Development and integration of the offshore net-
work: this involves an efficient integration of offshore renew-
able electricity into the energy system as well as the (offshore)
exchange of electricity between countries.
◆ Pillar 2: The further development of onshore intercon-
nectors: this allows for the onshore exchange of electricity
between neighbouring countries, thus enabling the efficient
use of renewable electricity on a European scale.
◆ Pillar 3: The creation of hosting capacity: this involves devel-
oping sufficient and suitably located grid capacity to connect
new consumption, production, and storage systems to the
grid, via the different substations, and pertains to grid devel-
opment at the backbone level, at regional transmission level,
but also at distribution level.
◆ The foundation: The development of a strong and robust
internal backbone grid enables power flows resulting from
the three key pillars to be securely managed across the whole
Belgian electricity system.
SCHEMATIC REPRESENTATION OF DEVELOPMENT OF THE ELECTRICITY GRID
SOURCE: ELIA’S FEDERAL DEVELOPMENT PLAN 20242034 FIGURE 544
On the road to a Net Zero Society
An energy efficient electricity system that ensures sustainability, affordability
and security of supply and facilitates the decarbonisation of each secor
Improving
sustainability
in network
development
Development and
integration of the offshore
network
Further development
of the onshore
interconnections
Creation of hosting
capacity
Optimisation of the existing
potential
Realisation of the
missing links Ensuring system stability Ensuring system flexibility
Reinforcement and expansion of the internal backbone 380 kV
The impact of the future energy system on grid development is
assessed based on the expected benefits at the European level,
and expressed as additional needs on top of the grid development
projects that were approved in the context of:
◆ Elia’s federal development plan 2024-2034 [ELI-3] approved
by the Belgian Minister of Energy on 05/05/2023 (see also
Figure 3-41 on the reference grid for the present study, which
includes the projects approved in this latest federal develop-
ment plan).
◆ The Walloon region adaptation plan [ELI-11];
◆ The Brussels capital region investment plans [ELI-12];
◆ The Flemish region investment plan [ELI-13].
In the following sections, grid development needs are provided
for each of the four development categories, and broken down
into three types of development measures:
199
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium198
No-regret infrastructure measures
Certain grid infrastructure investments are considered to be
no-regret measures and are resilient to changes to the choice of
energy sources for the Belgian electricity supply mix. Typically,
their main drivers are the electrification of demand and the devel-
opment of domestic RES. These investments should be prioritised
and implemented without delay to avoid any potential setbacks
in the energy transition process. Therefore, the first steps leading
to their realisation should be taken today. Additional studies may
be required to further detail and optimise the exact configuration
and scope of the concrete grid infrastructure projects, factoring
in technology constraints, practical feasibility, environmental
impact, etc.
Minimum-regret infrastructure measures
These are grid development projects that provide substantial ben-
efits to society across a wide array of scenarios and assumptions,
while posing minimal financial risks to the other remaining sce-
narios. Generally speaking, not proceeding with these measures
would entail bigger risks than further pursuing them. The exact
timing and order of these developments, however, is subject to
policy decisions.
Policy-dependent infrastructure measures
Several infrastructure needs for the Belgian grid strongly depend
on the electricity supply mix selected, which is thus mainly driven
by policy choices. Swift decision-making regarding the future of
Belgium’s energy system is required to anticipate the realisation
of these grid developments in a timely manner.
It should be noted that this study points towards overarching
needs regarding the development of strategic corridors and
therefore does not indicate specific grid development projects.
Concrete infrastructure developments (projects) that fall within
these strategic corridors should be further investigated through
more detailed assessments. Investigations into tangible grid
development projects will typically cover a wide range of aspects,
such as their technical feasibility, integration into the existing
transmission network, routing, integration into market mech-
anisms, stability aspects, environmental aspects, etc. alongside
more detailed economic studies and should be performed with
all of the involved TSOs.
When further developing such concrete project proposals, they
might deviate from the graphical illustrations shown in the cur-
rent study. However, if the projects fall within a strategic corridor
that was put forward by the optimisation model and if the poten-
tial is confirmed by the detailed investigations, they are very close
to the theoretical optimal solution and their realisation should
be pursued without further ado.
More detailed investigations will now be carried out to estab-
lish concrete project concepts that address the different
needs. If these more detailed studies reconfirm the antici-
pated benefits for society, the projects will be presented to the
authorities, for approval in the upcoming federal development
plan 2028-2038.
Some concrete examples of ongoing investigations regarding
new grid development projects that Elia is involved in, are pro-
vided in separate boxes throughout the following sections.
5.7.1. DEVELOPMENT AND INTEGRATION OF THE OFFSHORE NETWORK
The North Sea holds a significant amount of potential in terms of
the generation of offshore renewable electricity, namely through
large amounts of offshore wind potential. This study demon-
strates that the large-scale integration of this offshore wind pro-
duction into Belgium’s electricity system may be an important
lever for the decarbonisation of Belgium’s electricity mix. Indeed,
depending on the scenario and policy choices, the results demon-
strate important benefits linked to the integration of non-domes-
tic offshore wind in Belgium, on top of a total Belgian domestic
offshore production which lives up to its total potential of 8 GW.
Of course, the electricity generated offshore needs to be trans-
ported back to the Belgian mainland as efficiently as possible.
This section provides insights regarding future developments
to the Belgian offshore grid and its links to an integrated North
Sea offshore grid.
5.7.1.1. MAXIMISE OFFSHORE RES GENERATION
IN THE BELGIAN PART OF THE NORTH SEA
The results of this study clearly outline the benefits that are linked
to the development of Belgium’s full domestic offshore wind
potential of 8 GW and its connection to the Belgian electricity
system, for all considered scenarios and from 2040 onwards.
This indicates that the development of an additional 2.2 GW of
domestic offshore production through the repowering of the
first offshore wind zone and/or development of a third Belgian
offshore wind zone1 will be a no-regret option by 2040 (should it
prove to be feasible).
Existing connection capacity to the Belgian mainland, through
the MOG and the Princess Elisabeth Island, will not be sufficient to
connect this additional 2.2 GW to the Belgian grid. An additional
(third) Belgian offshore node will be required. This third node
could also be used to connect additional offshore (hybrid) inter-
connectors, on top of a first hybrid interconnector (see further),
and could ultimately be linked to the second offshore node, i.e.
the Princess Elisabeth Island. However, the latter is dependent
on policy choices.
The possibility of allocating supplementary areas to the develop-
ment of offshore wind is to be investigated as part of the publica-
tion of the new Marine Spatial Plan 2026-2034. The latter should
be finalised in 2025 and adopted on 20 March 2026 at the latest
[FPS-5].
Therefore, depending on the results of the various feasibility stud-
ies, a further increase in Belgian domestic offshore production
to a total of 8 GW and its integration into the Belgian trans-
mission grid is a no-regret grid development decision in the
lead-up to 2040. The necessary studies regarding its develop-
ment should be anticipated and the first steps towards its reali-
sation should be undertaken from today onwards. Furthermore,
in this framework, a Belgian strategy relating to the repowering
of the first offshore zone should be elaborated as soon as possible
(such as shown in BOX 4-3).
5.7.1.2. CONNECTING A FIRST BATCH OF NON
DOMESTIC OFFSHORE WIND VIA HYBRID
INTERCONNECTORS
The results of the current study indicate that investments in
hybrid offshore interconnectors (an interconnectors connected
to a foreign offshore wind zone) appear cost-efficient for comple-
menting Belgium’s energy supply. As offshore (interconnector)
projects take a long time to develop, it is important to continue
working towards the development of a first batch of non-domes-
tic offshore wind hybrid interconnectors.
An example - as stated in the federal development plan 2024-2034
- is the hybrid interconnector between Belgium and Denmark,
which is named TritonLink. Acknowledging that projects of this
size may encounter several hurdles, Elia is actively investigating
several offshore hybrid interconnector opportunities, as described
in BOX 5-7.
Furthermore, all scenarios confirm the benefit of creating an
Offshore Energy Hub by physically connecting the DC assets
of the Princess Elisabeth Island with a supplementary hybrid
interconnector. The latter is of course also to be realised in
HVDC-technology, and thus depends on the level of maturity of
this DC coupling technology, namely for DC circuit breakers. This
project was approved in the federal development plan 2024-2034
[ELI-3], on the condition of its technical maturity. Figure 5-45
below illustrates this concept.
THE OFFSHORE ENERGY HUB ON THE PRINCESS ELISABETH ISLAND FIGURE 545
1
st
hybrid
interconnector
with non-domestic
offshore wind
2 GW
Princess Elisabeth
Island
3.5 GW Wind
3.5 GW
TBD
Offshore Onshore
Nautilus
1.4 GW
Offshore Energy Hub
AC/DC Converter
DC busbar
AC Substation
5.7.1.3. FURTHER INVESTIGATIONS INTO
OFFSHORE HYBRID INTERCONNECTORS ARE
MINIMUMREGRET MEASURES
Although the scenarios show different possible decarbonisation
pathways for Belgium, it must be acknowledged that offshore
wind farms are a mature and proven low-carbon technology
and are seen as a cornerstone of Europe’s future green electricity
supply. Hybrid offshore solutions and offshore hubs are often
the most cost-efficient approach for incorporating far-offshore
locations into the Belgian electricity mix. However, the lead time
for the realisation of such projects can be as long as 10 to 15 years.
Without pre-empting upcoming energy policy decisions, further
investigations into offshore hybrid interconnectors should thus be
considered as a minimum risk, especially when they are aligned
with the strategic corridors identified in the simulations. This
will ensure that a first batch of mature projects are ready in the
project portfolio, which can then swiftly be continued once a
final decision is taken.
At this point, in addition to the TritonLink project, it is too early
to identify the specific offshore hybrid projects that would best
serve Belgian and European interests (see also Section 4.5). Col-
laboration with international partners is essential for identify-
ing promising options and establishing the necessary organi-
sational structures and agreements to successfully implement
the chosen projects. Therefore, further detailed investigations
with these partners have been launched within the framework
of the above-mentioned conclusions in order to elaborate a set of
key project candidates, which will allow for further decision-mak-
ing. These investigations are presented in BOX 5-7. The concrete
developments will have to be approved in the next federal devel-
opment plan if commissioning before 2040 is envisioned.
1. A third lever: deployment of offshore floating solar power, is also under investigation. This has not been explored in this study.
201
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium200
ONGOING OFFSHORE HYBRID INTERCONNECTOR INVESTIGATIONS
This box provides an overview of all the concrete inves-
tigations into offshore hybrid interconnectors that Elia
is currently engaged in. These investigations are clearly
aligned with the results of this study. Although it is far
from certain that all these projects will (need to) be real-
ised, each option must be investigated in a timely manner
in order to facilitate timely decision-making.
Belgium Netherlands
At the North Sea Summit held in Ostend 23 April 2023,
Elia and TenneT signed a memorandum of understand-
ing (MoU) relating to the investigation of the socioeco-
nomic welfare potential (for Belgium, for the Netherlands
and for Europe as a whole) of an electrical interconnector
between the Netherlands and Belgium that would also be
connected to an offshore wind farm (possible realisation:
between 2035 and 2040).
The results of the current study confirm the need to fur-
ther investigate the reinforcement of the interconnection
capacity between Belgium and the Netherlands (both
on- and offshore). This emphasises the importance of the
ongoing investigations.
Belgium Norway
On 15 November 2023, a memorandum of understanding
(MoU) was signed by Elia and Statnett. The MoU covers
investigations and cooperation relating to one or more
potential electrical hybrid interconnectors that would
benefit both countries. This could improve security of
supply, diversify energy sources, facilitate the integration
of wind power across sectors and create positive climate
effects.
The offshore area which is being considered is the Sørvest
F area (an extension of the Sørlige Nordsjø II area) in Nor-
wegian waters. The overall offshore length of the inter-
connector would amount to a total of ~1,000 km.
Source: www.equinor.com
The development and commissioning of the potential
BE-NO hybrid interconnector will be aligned with the
timeline and size of relevant and anticipated wind farm
allocations and wind developer plans. The investigations
covered by the MoU are in line with the political deci-
sion-making process and timings set by the Norwegian
government. This potential project could provide an
appropriate solution for the strategic North – South off-
shore corridor, which is identified in this study as a key
building block of offshore development in the North Sea.
Currently, the examination of several concepts is ongoing.
Statnett (NO) has signed similar MoUs with Amprion (DE),
Energinet (DK), National Grid (UK) and TenneT (DE).
Belgium Germany
Besides the investigations that cover onshore connection
options between Belgium and Germany (see BOX 5-10),
an offshore hybrid interconnector that would link Bel-
gium and Germany via the North Sea is also to be investi-
gated as a possible solution for the strategic North-South
corridor. The German and Belgian governments signed a
joint statement during the Energy Council that was held
on 30 May 2024, agreeing to further explore this idea. The
next step will involve discussions being held between the
involved TSOs and determining the scope of the study.
Ireland United Kingdom Belgium
The governments of Ireland, the United Kingdom and
Belgium signed a letter of intent (LOI) covering electricity
interconnectors during the Bruges Offshore event on 15
May 2024. This LOI is a clear example of the willingness
of the respective nations to enhance regional cooperation
and to move from a current bilateral perspective to more
regional, multi-lateral cooperation.
Elia is committed to supporting the investigations initi-
ated by different governments. Acknowledging that the
results of this study demonstrate that a renewable energy
corridor stretching from Ireland through the UK to Bel-
gium holds potential, Elia will thus work alongside the
Irish and British TSOs to investigate optimal infrastructure
options to find fitting solutions for this strategic corridor.
This work will also consider possible offshore routes, inte-
grating offshore wind from Ireland and the UK into the
transmission system of each country. Plans for this work
still need to be established, but the objective is to deliver
input for the North Sea Summit being held in June 2025.
BOX 5-7
5.7.1.4. POLICYDEPENDENT EVOLUTION OF THE BELGIAN OFFSHORE GRID
In several scenarios, the North Sea emerges as a major source of
electricity production for Belgium, since it is seen to play a crucial
role in achieving the continent’s climate goals. In situations where
policymakers decide that more electricity generated in the North
Sea will be tapped to complement Belgium’s electricity mix, on
top of the currently planned offshore projects, offshore electricity
hubs in Belgian waters will become indispensable, given the
scarcity in coastal landfall points and cable routes to the inland
transmission system. These hubs, which differ in number and size
depending on the scenario being explored, bundle renewable
energy produced in the North Sea and allow this energy to be
transported in bulk to Belgian load centres.
Furthermore, these hubs would not only be vital for the decarbon-
isation of Belgian society but would also be an essential building
block for the decarbonisation of Europe. Although it is a relatively
small country with a short coastline, Belgium’s central position
is of strategic importance for the buildout of offshore renewable
energy and its transportation back to shore and across the con-
tinent. With its access to the North Sea and proximity to several
electricity markets (France, UK, Netherlands, Germany and Lux-
embourg), such offshore hubs would allow for the very efficient
trade of renewable electricity between these different markets,
benefitting customers throughout Europe.
Furthermore, creating offshore transport capacity for bulk trans-
port between these different countries allows existing onshore
transmission infrastructure to be bypassed, so reducing onshore
reinforcement needs.
Belgian offshore electricity hubs could serve as vibrant energy
roundabouts, powering an offshore backbone integrating
offshore wind from (for example) France, Belgium, the United
Kingdom, the Netherlands, Norway or Denmark.
Beyond the technical challenge of realising a grid which spans
the entire North Sea, other barriers exist and are currently slow-
ing down the development of offshore hybrids. Elia is actively
working on alleviating these barriers and proposing solutions to
them through publications such as the ‘Making Hybrids Happen’
[ELI-9] paper that was jointly developed with Ørsted, and the OTC
expert paper II [OTC-1].
203
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium202
OFFSHORE TSO COLLABORATION (OTC)
The OTC is an informal group of offshore TSOs from nine
Member States and third countries which border the
North Seas. Its purpose is to accelerate the development
and implementation of an offshore grid and to support
the realisation of ambitious political goals in the best pos-
sible way. The OTC is working on the implementation of
an offshore hybrid grid in the North Seas that includes
hybrid interconnectors, energy hubs and hydrogen infra-
structure and is based on the political declarations of the
North Sea summits in Esbjerg and Ostend.
Elia plays a very active role in its work, since harnessing
the offshore renewable potential of the North Sea is of vital
importance for the decarbonisation of Belgian society.
During the offshore summit held by the Belgian govern-
ment on 15-16 May 2024 in Bruges, the second OTC expert
paper was launched. A third expert paper is planned to
be delivered during the North Sea Summit being held in
June 2025.
The paper covers both the key projects that will under-
pin the development of a North Seas grid in the long
term and an initial grid map of hybrid projects. The most
advanced projects amongst those mentioned above are
already depicted in the map, such as TritonLink, Nautilus,
Belgium-Netherlands and Belgium-Norway.
The paper also includes a series of policy recommenda-
tions, each of which falls into one of three areas:
1. supply chain enhancement;
2. market conditions framework;
3.
cost sharing and funding of infrastructure related to
offshore hubs and hybrid projects.
The OTC plays a vital role in offshore grid development
in the North Seas as it fulfils the role of the missing link
between conceptual long-term studies and the creation
of a short- to medium- term coherent multi-lateral project
portfolio. Indeed, current approaches, which have mainly
resulted in bilateral agreements and follow-up, need
to be replaced by a multi-lateral approach, allowing all
impacted TSOs to contribute during the early stages of
infrastructure development.
In order to achieve this, the offshore project investigations
mentioned above will be fed into the OTC investigations
along with project investigations from other TSOs (that
Elia is not involved in). This will ensure that projects are
not only developed for bilateral benefits, but that they are
assessed on a more holistic level, allowing the most opti-
mal combination of all of these projects to be selected.
BOX 5-8
5.7.2. THE FURTHER DEVELOPMENT OF ONSHORE INTERCONNECTORS
This study highlights that onshore interconnectors continue
to play a crucial role in ensuring the efficient dispatch of elec-
tricity throughout Europe, regardless of Belgium’s dependence
on imports. Strengthening the interconnection capacity across
certain borders is highly cost effective and continues to yield
significant benefits for society. Therefore, the further develop-
ment of additional Belgian interconnectors (on top of those that
have been approved in the federal development plan 2024-2034)
remains beneficial. This is critical for Belgium’s full integration
into the European electricity market and is a crucial component
for a net-zero energy system.
Although the further reinforcement of Belgian interconnectors
remains beneficial, the borders to be prioritised appear to depend
on the electricity mix chosen for the country and neighboring
countries. Figure 5-46 provides an overview of the interdepend-
ency between the benefits of reinforcing a specific border in
relation to energy mix choices.
IMPACT OF ENERGY MIX CHOICES ON NEED FOR INTERCONNECTORS FIGURE 546
4GW
2GW
0GW
8GW
16GW
12GW8GW0GW 4GW
6GW
8GW
6GW
4GW
2GW
0GW
16GW
12GW8GW0GW 4GW
More interesting
in case of less
offshore generation
directly connected
to BE
Invariant of
the offshore/
domestic
generation
More interesting in case of
more generation in Belgium
Note: this is the need for XB reinforcement calculated as the marginal benefit reducing European electricity costs
in a zonal setting. The impact on the Belgian costs can be different and should be further investigated.
NEW
8
6
4
2
0
GW
0 4 8 12 16
GW
Legend
Lower
need
Higher
need
Three main trends clearly emerge:
1.
Point-to-point interconnectors with the Netherlands become
more attractive in situations where lower levels of offshore
generation are connected to Belgium. This is logical, since
such interconnectors would give Belgium access to the vast
domestic offshore potential of the Netherlands.
2.
Interconnectors with Germany and Luxembourg are more
beneficial in situations where Belgium’s local electricity sup-
ply is increased. Indeed, in such situations, these intercon-
nectors offer more opportunities to sell power produced in
Belgium to areas with limited access to renewable energy.
3.
The need for interconnection capacity between Belgium and
France is not significantly determined by the energy mix
choices in Belgium.
The following sections link these general trends and study results
with concrete onshore interconnector projects. As mentioned
previously, regarding possible offshore projects, hybrid solu-
tions are more attractive, as they offer a cost-effective solution
for integrating offshore wind into the system whilst increasing
cross-border capacity.
5.7.2.1. REINFORCEMENT OF THE
INTERCONNECTION CAPACITY WITH THE
NETHERLANDS
The reinforcement of Belgium’s north border will prove to be
essential for the decarbonisation of Belgian society through
imports of green electricity from northern Europe. Indeed, the
current study reveals that the reinforcement of cross-border inter-
connection capacity with the Netherlands is a no-regret meas-
ure in all considered scenarios. The expected benefits increase
as domestic (renewable) electricity generation decreases, but
the business case remains positive in all scenarios. This is partly
because additional cross-border reinforcement can be done in a
very cost-efficient way, by reinforcing the existing axis between
the substations of Van Eyck (Belgium) and Maasbracht (the Neth-
erlands) through the replacement of the existing conductors
with high-performance conductors (HTLS conductors), which
more than double the transport capacity on that particular
axis.
These results confirm the analysis that was carried out as part
of the federal development plan 2024-2034; more information
about this project can be found under section ‘4.3.2. Versterking
205
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium204
Van Eyck - Maasbracht (NL)’ [ELI-3]. As discussed in BOX 5-7, the
realisation of this project is being investigated in line with the
existing MoU that was signed by Elia and TenneT.
A further increase in the exchange capacity between Belgium
and the Netherlands, on top of the no-regret reinforcement of the
Van Eyck – Maasbracht axis, may be beneficial in multiple future
scenarios. Indeed, the business case for an additional cross-bor-
der link between Belgium and the Netherlands will, for example,
become increasingly attractive when:
◆ domestic production in Belgium is lower, meaning that Bel-
gium will depend more on imports from other countries;
◆ technology will not prove to be sufficiently mature for real-
ising an offshore meshed DC grid, requiring far-out offshore
renewable electricity to be imported more via the onshore
grids.
Such developments may lead to a positive business case for an
additional link between Belgium and the Netherlands from 2036
onwards.
5.7.2.2. REINFORCEMENT OF THE
INTERCONNECTION CAPACITY WITH
LUXEMBOURG
Luxembourg is one of the European countries that lacks a direct
sea connection. It will depend on green electricity imports from
its neighbours as it transitions to net zero.
A clear strategic corridor identified in this study is the corridor
that crosses the south-east of Belgium. Although this is not a
newly identified need, it is the first time that the corridor features
so strongly in the results. Indeed, cross-border reinforcement
between Belgium and Luxembourg is shown to be beneficial
across all scenarios. It can thus be considered as a no-regret
investment. However, more detailed work is needed to assess the
feasibility of different technological options, as this will influence
the costs. Synergies with other cross-border projects, for Belgium
and for other countries, may also prove to be possible through a
more holistic analysis approach across multiple borders.
PENTALATERAL STUDY
On 1 March 2023, an MoU was signed between five TSOs, of
four countries: RTE (France), Amprion (Germany), Trans-
net (Germany), Creos (Luxembourg) and Elia (Belgium).
The resulting study, which is currently being developed, is
therefore referred to as the Pentalateral Study.
The study aims to identify additional interconnector needs
and internal reinforcements where relevant between Bel-
gium, Luxembourg, France, and Germany after 2040. It
also seeks to identify potential project candidates that
should be included in future international planning pro-
cesses (TYNDP) and national development plans.
The project candidates that are relevant for Belgium
relate to further onshore reinforcements between Bel-
gium and France on the one hand and Belgium and Lux-
embourg on the other.
The authors of the study intend to work towards a joint
feasibility report which focuses on a common vision of
grid development projects after 2040 (cross-border and
internal reinforcements where relevant). This common
vision is to be finalised in mid-2025, so ensuring that the
results can be taken into account during the next TYNDP
and subsequent national development plans.
BOX 5-9
5.7.2.3. REINFORCEMENT OF THE
INTERCONNECTION CAPACITY WITH GERMANY
ALEGrO, the first interconnector built between Belgium and
Germany, was commissioned in November 2020. It was realised
with HVDC technology and has an exchange capacity of 1 GW in
both directions. The optimisation model selects ample additional
interconnection capacity between Belgium and Germany, on
top of ALEGrO, in all scenarios but the Global Ambition scenario.
Within the Distributed Energy scenario, an additional cross-bor-
der capacity of up to 2 GW, on top of ALEGrO, proves to be bene-
ficial by 2040, which is even increased to 3 GW by 2050. As shown
in Figure 5-46, the higher the level of domestic electricity supply
and/or non-domestic offshore wind in Belgium, the more attrac-
tive this reinforcement becomes.
Whereas the study shows that the additional cross-border capac-
ity is beneficial in almost all scenarios, the nature of its behaviour
in the market depends on the market situation and configuration
in both countries. For example, depending on the connection
points and the market configuration in Germany, cross-border
exchanges may tend towards a position of net exports of renewa-
ble electricity to the energy-intensive Ruhr area, or rather towards
a position of net imports of renewable electricity from the high
amounts of German renewable generation capacity. In any case,
across all scenarios, the flows between both countries remain
relatively well balanced in both directions, showing a clear ben-
efit for both countries in the effective integration of renewable
generation on a European scale.
SECOND INTERCONNECTOR BETWEEN BELGIUM AND GERMANY
A second HVDC interconnector between Belgium and
Germany was provisionally included in section 4.3.3 of the
federal development plan 2024-2034, to give readers an
idea of its estimated potential (Tweede interconnector
België-Duitsland [ELI-3]). The results of the current study
confirm the potential identified in the framework of the
federal development plan.
Building on the results presented above and on the
extensive experience gained through the construction
and operation of their first shared interconnector, Elia
Transmission Belgium and Amprion recently intensified
their exploratory work related to a second electricity inter-
connector and launched a concept development study.
The study will be finalised before the end of 2024 and
will put forward an optimal capacity for the underground
HVDC link (1 vs 2 GW), a detailed cost-benefit analysis, an
indication of the connection points with the transmission
systems in both countries, an idea of the necessary rein-
forcements of the internal backbones in order to host the
new interconnection capacity and, finally, an initial time-
line relating to its realisation, with an expected commis-
sioning in 2037-2038.
BOX 5-10
5.7.2.4. REINFORCEMENT OF THE
INTERCONNECTION CAPACITY WITH FRANCE
The border between France and Belgium is already very well
developed. The further reinforcement of onshore cross-border
capacity is not directly selected by the optimisation model of the
current study; however, implicitly, the model increases cross-bor-
der capacity between Belgium and France through an offshore
route via offshore French and Belgian nodes. Should the feasibility
of such offshore development prove to be difficult, or undesir-
able for other reasons, the realisation of an additional onshore
interconnector may prove to be beneficial and desirable. As the
need for Belgian-French cross-border reinforcements seems to be
unaffected by choices relating to the Belgian energy mix, the fea
-
sibility and impact of the different (onshore and offshore) options
need to be explored. The further reinforcement of cross-border
capacity between Belgium and France is currently being studied
as part of the same Memorandum of Understanding, elaborated
for the launch of the pentalateral study (see BOX 5-9).
207
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium206
BOX 5-11
5.7.3. THE CREATION OF HOSTING CAPACITY
The development of domestic RES and the increasing electrifica
-
tion of the industrial, mobility and heating sectors carries impor-
tant implications for the electricity system. As the electrification
of local end use and the buildout of domestic RES is expected to
primarily unfold over the next 10 to 15 years, taking swift action
now in order to create hosting capacity across the grid in a timely
manner is of the utmost importance.
Although the realisation of new infrastructure is essential, it is
not the only lever that needs to be deployed in order to create
sufficient hosting capacity. It should be emphasised here that
unlocking flexibility on both the consumer and producer sides
is the first way to swiftly create hosting capacity by efficiently
managing congestions. However, as flexibility for congestion
management is not the central focus of this chapter (and study)
it will not be explored in any detail.
Furthermore, the smart selection of locations for new grid con-
nections for new large load offtake facilities, such as big data
centres and (ultimately) power-to-X plants, plays a crucial role in
ensuring the grid is used in the most efficient manner. Bringing
production and large loads closer together shortens the distance
electricity has to travel from generation to load, reduces losses
and also reduces the need.
Therefore, the smart selection of sites for large load centres is not
just about finding a location that meets the needs of individual
consumers, but it is also related to the optimisation of the grid
on a holistic system level.
GRID HOSTING CAPACITY
In order to guide grid users (either new grid users or those
who want to increase their offtake) to suitable grid con-
nection points, Elia launched a Grid Hosting Capacity
map [ELI-10] at the end of 2023. The map provides a sim-
plified indication of the grid hosting capacities that are
still available on top of capacity reservations/allocations/
low-voltage connected evolutions and also takes into
account planned grid infrastructure development.
The map also provides users with information about the
capacity for flexible connections. This refers to the max-
imum percentage of permissible yearly energy curtail-
ment relative to the yearly total generated or consumed
energy. This map effectively serves the purpose of guid-
ing new electrical loads to suitable locations.
ELIA GRID HOSTING CAPACITY MAP FIGURE 547
In this vein, depending on the scenario, new sites which consume
a lot of electricity and that have a certain degree of freedom with
respect to their choice of location for site development, can be
best placed close to the landing sites of offshore wind or nuclear
production sites. And, vice versa, the integration of new offshore
wind can actively support an efficient use of the grid by being
connected near to the large load centres in Belgium.
This not only benefits individual consumers through potentially
shorter connection lead times, but also contributes to a less
expensive and more efficient and sustainable electricity grid
for everyone.
Although both levers mentioned above are indispensable, new
transmission infrastructure remains essential across the three
parts of the Belgian electricity system (see Figure 5-48) to cre-
ate the required hosting capacity for new loads, production and
storage:
1.
The horizontal grid is the Belgian backbone (380 kV and
220 kV) to which interconnectors, centralised production and
large industrial loads and storage are connected (typically
>300 MVA). The required evolutions of the horizontal grid are
discussed in Section 5.7.3.1.
2. The vertical grid is supplied via transformers connected to
the backbone, and provides connection opportunities for big
industrial loads, decentralised production and storage (>25
MVA) and supplies the distribution grid. This network will, over
time, evolve to voltage levels of 110 kV and 150 kV, increasing
both its transport capacity and efficiency.
3.
The distribution grid supplies small industrial grid users and
residential loads and is used to connect small decentralised
production and storage (<25 MVA).
SCHEMATIC REPRESENTATION OF THE DIFFERENT PARTS OF THE BELGIAN TRANSMISSION &
DISTRIBUTION SYSTEMS FIGURE 548
Centralised Biggest
industrials
Decentralised
Decentralised/
Residential
Big industrials
Small industrials
& residential
Interconnectors
(onshore &
offshore)
Production Load Storage Import/export
380 kV / 220 kV
Horizontal
Grid
Voltage levels
70 kV / 110 kV / 150 kV
Vertical
Grid
MV & LV
30 kV / 36 kV
Distribution
Grid
Biggest
Small
Big
With regard to the vertical and distribution grids, this study shows
that the increase in electricity consumption drives reinforcement
needs, which are significant in all scenarios. To match the required
load evolution in each scenario, significant investments should
be made for:
◆ The realisation of new substations to connect the new loads
to the three parts of the system;
◆ Increasing the transformation capacity from the 380 kV
and 220 kV grids (the ‘horizontal’ grid) to the underlying,
regional transmission grids of 30 kV to 150 kV (the ‘vertical’
grid);
◆ The reinforcement of the 110 kV and 150 kV transmission
grids - this relates to the reinforcement or upgrade of over-
head lines and new underground cables;
◆ Increasing the transformation capacity from the vertical
grid to the distribution grids;
◆ The reinforcement of the distribution grids.
These investments depend on the distribution of electricity con-
sumption across the different elements of the grid. A grid user
connected to the distribution grid will also impact the vertical
grid, whilst a grid user connected to the horizontal grid will not.
For some segments, the connection points are easy to forecast
(EVs and residential heating are mainly expected to be connected
to the distribution grid), whilst for other industrial load segments,
the connection points will depend on the size and type of the
industrial assets involved, which is more difficult to predict.
209
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium208
Across the different scenarios, the impact of decentralised gener-
ation on the grid reinforcement of the vertical system is assumed
to be rather limited. The main impact is due to the level of pen-
etration of decentralised and residential production (onshore
wind, solar) which will limit the peak grid loading and therefore
slightly limit the need for reinforcement. The need for additional
grid reinforcements for the creation of hosting capacity will thus
be higher in the scenarios with more focus on centralised produc-
tion (nuclear production, offshore production) or more focus on
imports from neighbouring countries, and lower in the scenarios
with more focus on decentralised production.
Realisation of new substations
Historically speaking, the 380 kV backbone was mainly con-
structed for the transportation of bulk energy. Grid users con-
nected to the 380 kV system were limited to large production
units and very few large loads. Given the electrification of society
- and, more specifically large industrial loads - and the emergence
of large batteries and data centres, an increasing number of grid
users will be connected directly to the 380 kV system in the future.
To facilitate such connections, the creation of hosting capacity
hubs at the 380 kV level is essential. As part of a first phase, this
will also make it possible to free up capacity on the underlying
networks for the electrification of smaller grid users.
The federal development plan 2024-2034 identified the need for 5
supplementary substations at the 380 kV level from 2024 to 2034,
for currently existing areas with large loads: Antwerp, Hainaut,
the Ghent area and Albert Channel. These substations will help
to connect direct grid users to the grid and will also increase the
transformation capacity to the lower voltage levels.
However, not all new loads are limited to the existing industrial
areas. For example, batteries, power-to-X and data centres can
be much more flexible in terms of their location. Supplementary
hosting capacity hubs at the 380 kV level are thus to be expected
in the coming decades.
For example, when observing Figure 3-39, a significant number
of large-scale battery volumes (ranging from 3.4 GW in the LFLEX
sensitivity, to 5.4 GW in DE, GA, ELEC and SUFF scenarios/sen-
sitivities and up to 9.4 GW in HFLEX sensitivity) appear by 2050.
Current development projects of large-scale batteries in Belgium
hold a wide variety of power capacity, from tens of MW up to 750
MW (and possible higher). If we assume an average of 300 MW
per battery, around 11 connections in the 3.4 GW case, around 18
connections in the 5.4 GW case, and up to 31 connections in the
9.4 GW case would be required for the grid connection of large-
scale batteries alone.
Although it is very difficult to predict exactly how many new
substations would be required at the 380 kV level, taking into
account the example above, the emergence of new large loads
such as data centres and the increased need for transformation
capacity in the vertical system (see below), an additional need for
5 to 10 supplementary hosting capacity hubs at the 380 kV level
could be expected by 2050 on top of the identified needs for load
clusters in the federal development plan. Consequently, new sub-
stations across the vertical and distribution grids are also needed
to connect more grid users and free up capacity at lower levels.
Increasing the transformation capacity from the 380 kV grid
Here, two different aspects must be combined. Firstly, the increas-
ing loads across the vertical and distribution systems will require
more transformation capacity from the horizontal grid. Secondly,
the increasing flows across the horizontal grid will drive the need
to split up the vertical grid into many smaller electrical areas
to limit the size of the flows across the lower voltage networks
(e.g. instead of having one entire 150 kV network covering all
of Belgium, in the future several smaller, separate 150 kV grids
will exist, each covering a specific regional area). Taking the N-1
criteria and short-circuit limitation of the assets into considera-
tion, those two aspects will lead to the need to install new large
transformers (>300 MVA).
Reinforcement of the 110 kV and 150 kV transmission grids
The regional hosting capacity will have to be created for industrial
loads and large decentralised production sites and storage. The
reinforcement of overhead lines and new underground cables
will be required to cope with the higher flows necessary for sup-
plying the increased load. These lines and cables typically have
a capacity of 200-300 MVA.
Increasing the transformation capacity to the distribution
grids & reinforcement of the distribution grids
The electrification of transport, residential heating and small
industrial loads will impact distribution grids at low-voltage and
medium-voltage levels. This implies huge investments in the
distribution grids: low- and medium-voltage underground cables,
new MV substations, new transformers.
To cope with the load increase across the distribution grid,
additional transformation capacity from the vertical grid will
be required. A typical 150 kV/MV transformer has a capacity of
50 MVA.
Time horizon
As the creation of hosting capacity is a must for the decarboni-
sation of society, a large part of the required grid reinforcements
is a no-regret as they are needed across all scenarios. For hosting
capacity, the larger impact is felt across the distribution grid,
where the needed investments are 2-4 times higher than for the
vertical grid. Still, this remains a challenge for the vertical grid,
across which the needed projects are more difficult to realise and
take more time. A typical investment in the vertical grid takes 5
to 10 years to be realised.
New infrastructure investments must be robust enough to meet
future needs throughout their typical lifetime of 40 years. There-
fore, the investments that are planned for 2036 must already
take changes that will happen by 2050 into consideration. This
is why the next ten years will be crucial, since a large part of the
investments will have to be realised in this period.
5.7.4. THE DEVELOPMENT OF A STRONG AND ROBUST INTERNAL
BACKBONE GRID
The grid developments discussed in the previous sections will only
be possible if a robust, reliable and meshed internal backbone
grid is in place. Indeed, the internal backbone grid must be capa-
ble, with a high degree of reliability, to provide sufficient hosting
capacity for an increased offtake of electricity from the 380 kV
grid and below, facilitate an increased amount of international
exchanges of electricity stemming from new (on- and offshore)
interconnectors, and enable the integration of large quantities
of new renewable or low-carbon generation into the system.
Today, the backbone grid consists of a partially meshed 380 kV
grid. Several reinforcement projects have already been approved
in previous federal development plans, which have been or are
currently actively being developed. Important ongoing rein-
forcement projects include the HTLS reinforcement of a large
part of the 380 kV backbone grid and the new Ventilus and
Boucle du Hainaut corridors. These planned reinforcements
are of the utmost importance for allowing further evolutions
across the backbone grid, as well as for enabling the connection
of new offshore developments and new onshore interconnectors.
Without these projects, very few of the projects described in
this chapter will be possible, obstructing Belgium’s path to
decarbonisation.
5.7.4.1. FULFILLING THE REMAINING POTENTIAL
FOR THE REINFORCEMENT OF THE EXISTING
380 KV FLUX AXES
The currently approved backbone reinforcement projects com-
prised in the federal development plan 2024-2034 include the
reinforcement of most of the existing 380 kV flux axes in Belgium,
mainly through the replacement of the current conductors with
new HTLS conductors, which potentially more than doubles the
transport capacity of these axes. For two axes, a reinforcement
decision is yet to be made. In light of the results of the current
study, the reinforcement of these axes is proven to be a no-regret
measure:
◆ reinforcement of the Zandvliet – Doel – Mercator corridor;
◆ reinforcement of the south-east Gramme – Brume –
Villeroux– Aubange corridor.
More information about these projects is included in the federal
development plan 2024-2034 [ELI-3] in Sections 4.5.1.2. and 4.5.1.3.
It should be kept in mind that those sections describe indicative
solutions which need to be further explored and fleshed out.
The solution which will be adopted may thus deviate from the
one described in the development plan. The reinforcement of
the Doel-Zandvliet connection with HTLS conductors has been
identified as being of the highest priority and a concrete project
will swiftly be launched.
211
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium210
FULFILLING THE REMAINING POTENTIAL FOR REINFORCEMENT OF THE EXISTING 380 KV
FLUX AXES IS A NOREGRET MEASURE FIGURE 549
Substation 380 kV
Axis already reinforced / without potential for
further reinforcement
Axis with remaining potential for further
reinforcement
5.7.4.2. REINFORCING THE WESTEAST
CAPACITY OF THE INTERNAL BACKBONE
The currently planned west-east corridors - Ventilus and Boucle
du Hainaut - are essential measures to:
◆ increase the capacity from the Belgian coast to the load cen-
tres across the country;
◆ allow for the integration of the planned offshore production
(up to 5.8 GW);
◆ create additional hosting capacity for growing offtake and
decentralised renewable production;
◆ increase reliability & security of supply for the respective
regions;
◆ to ensure competitive and affordable energy prices by facili-
tating increased market exchanges with the UK and France;
However, in the lead-up to 2050, more offshore capacity might
be integrated into the Belgian backbone grid (for example the
evolution towards the full Belgian potential of 8 GW of domestic
offshore production, and the development of non-domestic off-
shore wind hybrid interconnectors). The planned backbone grid
with Ventilus and Boucle du Hainaut will not be able to host such
additional connections on the coastline.
For the further integration of offshore connections beyond
the Princess Elisabeth Zone and the Nautilus interconnector,
additional west-east transport capacity will need to be realised
across the backbone grid.
As part of a first phase, and up until a certain number of connec-
tions is reached, this additional west-east capacity could be real-
ised by connecting these future offshore wind farms individually
to strong 380 kV nodes situated deeper inland. This approach has
already been applied in the design of the TritonLink project, for
which a connection point in the Ghent area has been proposed.
When several offshore developments materialise, a holistic
approach will be required to design the future backbone grid
evolutions as well as to define an optimised set of connection
points. The design will be influenced by multiple factors, such
as local increases in the electrical load (which will increase the
local hosting capacity for connecting additional offshore develop-
ments), or an uptake of nuclear generation (which will decrease
the capacity for additional offshore hybrid interconnectors).
Further exploration should lead to the definition of the optimal
approach and configuration for further increasing the backbone’s
west-east transport capacity.
It is important to note that the timely realisation of the Ventilus
and Boucle du Hainaut projects, and the resulting meshed AC
backbone grid consisting of three closed 380 kV loops, is an
essential prerequisite for the further west-east development
of the backbone.
5.7.4.3. POLICYDEPENDENT REINFORCEMENTS
On offshore generation
As mentioned above, increased offshore ambitions with larger
amounts of (far-out or domestic) offshore wind production or
hybrid interconnectors that are to be integrated into the Belgian
energy system will put a lot of pressure on west-east transport
capacity needs in the Belgian backbone grid.
On prolonging existing nuclear generation
A further uptake of other centralised (i.e. nuclear) production in
Belgium may give rise to additional needs for internal transport
capacity, as well as the further development of cross-border inter-
connectors. If the extension of more than 2 GW of existing nuclear
units is planned, the electrical infrastructure around the current
nuclear sites needs to be prepared. Belgian nuclear phase-out
plans since 2003, the arrival of additional grid users nearby, and
changes in European legislation have reduced the grid hosting
capacity for such extensions.
On new nuclear
Identifying potential future new nuclear sites is an essential step.
This involves preparing the backbone grid infrastructure at the
most probable locations among those sites and integrating them
into the overall Belgian backbone grid.
These evolutions will lead to the need for a holistic approach to
the planning of further backbone grid reinforcements. This may
give rise to overlay corridors or grids, for which the optimal config-
uration and sizing will be highly dependent on the exact number
and locations of offshore developments, centralised onshore pro-
duction facilities, and onshore interconnectors. A clearly defined
roadmap for Belgium’s future energy system is essential for the
timely anticipation of the further reinforcement of the Belgian
backbone grid as Belgium approaches net zero in 2050.
A HOLISTIC APPROACH TO BELGIAN SPATIAL PLANNING
Our research has revealed a powerful insight: if policymak-
ers decide on the development of non-domestic offshore
wind for Belgium, this will require substantial amounts
of transmission capacity to connect the offshore hubs to
the onshore grid. Given Belgium’s limited coastal landfall
points and scarce cable routes to the inland transmission
system, the planning of cable routes on a project-by-pro-
ject basis lacks efficiency.
We propose a more integrated and holistic approach to
spatial planning - one that fosters collaboration between
different sectors and authorities. This approach will allow
the required energy corridors to be promptly identified
and allocated with environmental and social implications
being considered early on in the infrastructure realisa-
tion phase. Forward-thinking approaches such as this will
streamline permitting procedures, foster public accept-
ance and ultimately pave the way for the successful inte-
gration of large-scale offshore renewable energy into our
existing energy system.
For situations which involve high amounts of domestic
nuclear power generation and high amounts of domestic
RES, the amount of required offshore-onshore transmis-
sion capacity remains limited. Yet the need for a holis-
tic approach to spatial planning remains imperative in
a densely populated area such as Belgium. Indeed, the
location of new nuclear sites demands thoughtful consid-
eration and might have a significant impact on required
electricity corridors.
BOX 5-12
213
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium212
5.7.5. AN OVERVIEW OF NO-REGRET, MINIMUM-REGRET AND POLICY
DEPENDENT EVOLUTIONS TO BE ENVISAGED FOR THE BELGIAN OFF- AND
ONSHORE GRID
3
B
C
Reinforcement and expansion
of internal backbone 380 kV
Reinforcement corridor
Zandvliet-Doel-Mercator
No regretMinimum regretPolicy dependent
Reinforcement South-East
Gramme-Brume-Villeroux-Aubange
Direct Connection to strong
nodes
Additional Backbone Corridors
New offshore electricity hubs could serve
as vibrant energy roundabouts
Scarcity in coastal landfall
points and cable routes to
the inland transmission
system make offshore hubs
indispensable
Belgium-Netherlands
Belgium-
Luxembourg
First hybrid interconnector
with far-out wind Flexibility for congestion
management & strategic site
selection for new large grid users
Supplementary hosting capacity
hubs on 380 kV in areas where new
loads emerge, such as batteries,
power-to-X or data centres
Strong influence of location of load &
generation makes it very difficult to
make estimations. Strategic energy
policy choices need to be made by
the government in order to timely
anticipate new needs.
Expanding grid for industrial
clusters: Hosting capacity hubs for
electrification of industry on 380 kV
in known load areas
Belgium-Germany
Within the
Decentralized Energies
scenario, an additional
cross-border capacity
of up to 2 GW proves to
be beneficial by 2040.
Belgium-Netherlands
Need increases if a meshed
offshore grid is not possible
Belgium-France
Onshore options if
offshore routes prove
to be not feasible
Strengthening distribution &
regional transmission grids,
incl. transformation capacities
Maximize offshore
RES generation in
the Belgian part
of the North Sea,
if feasible
Anticipate by swiftly
executing the
reinforcements of the
backbone identified as
conditional in the Federal
Development Plan
Develop project candidates for strategic
offshore corridors, allowing further
decision making
As part of a first phase, and up until
a certain number of connections
is reached, this additional west-
east capacity could be realised by
connecting such future offshore wind
farms or hybrid interconnections
individually to strong 380 kV nodes
situated deeper inland.*
The optimal configuration and sizing
will be highly dependent of the exact
number and locations of offshore
developments, centralized onshore
production facilities, and onshore
interconnectors
Offshore
Energy
Hub
Nautilus
+ 2.2 GW
Development and integration
of the offshore network Further development of the
onshore interconnections Creation of hosting capacity
* Creation of 3 closed loops in the Belgian Backbone by means of Ventilus & Boucle Du Hainaut remains fundamental in all scenario’s
5.8. OTHER KEY INSIGHTS
5.8.1. MATERIAL NEEDS AND OTHER ENVIRONMENTAL ASPECTS
The objective of this subsection is to quantify the material require
-
ments for the Central onshore RES scenario of DE and GA for
Europe. The material needs have been calculated for all the
countries included in this study, as illustrated in Figure 5-50, and
specifically for Belgium, as shown in Figure 5-51. The evaluated
material requirements include the amount of copper, cobalt,
graphite, lithium, and nickel used by the power sector (e.g. wind
farms, photovoltaics, gas units), electrolysers, and electric vehicles.
Note that electric vehicles in this context include passenger cars,
vans, busses, and heavy trucks. However, it should be noted that
the materials required for the grid are excluded from the anal-
ysis. In order to help the reader to quantify the increased needs
in materials, Table 5-1 shows the world production capacity in
kilotonnes for 2023 [USG-1].
WORLD PRODUCTION CAPACITY TABLE 51
Copper Graphite Nickel Cobalt Lithium
Current production (kilotonnes) 27 000 1 600 3 600 230 180
Despite the scale differences between Europe and Belgium, Fig-
ure 5-50 and Figure 5-51 exhibit comparable trends. The demand
for copper is primarily driven by electric vehicles, offshore wind
(less impactful for Belgium), onshore wind and the photovoltaics.
As for cobalt, lithium, nickel and graphite, the demand is largely
fuelled by electric vehicles.
For all the materials considered, the additional requirements rep-
resent a significant portion of the current production. This surge
is largely attributed to the expected evolution in mobility choices.
For instance, under the DE scenario, the demand for lithium is
expected to increase more than eightfold by 2050, compared to
the projected demand in 2024. Similarly, the demand for cop-
per is projected to be more than 4 times by 2050 the projected
demand in 2024.
The estimates are subject to changes in technological advance-
ments, particularly those related to EVs, where battery types and
energy density are undergoing significant evolutions. Nonethe-
less, the substantial increase in material needs highlights that
material availability will be a critical factor for the success of
the energy transition. This is a global challenge that must be
addressed to ensure a smooth transition towards sustainable
energy sources.
The demand for materials in Belgium alone is projected to double
or even quadruple, but this amount is relatively small when com-
pared to the rest of Europe, a logical outcome given Belgium's
proportional size. The requirement for materials is crucial in the
energy transition and should be closely monitored at both Euro-
pean and global levels to ensure these materials are available and
can be recycled and reused in the future. As for Europe, the main
impact is driven by mobility.
Considering the minimal impact of supply technologies on
the total material requirements examined (excluding copper),
material usage is not expected to be a crucial criterion. However,
it's important to note that this analysis does not consider other
types of materials (such as rare earth metals, for example). This is
undoubtedly an aspect that requires further monitoring.
215
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium214
ESTIMATED USAGE OF SELECTED MATERIAL USAGE OF THE DIFFERENT SCENARIOS FOR EUROPE FIGURE 550
+
!
COPPER
LITHIUM
NICKEL
COBALT
GRAPHITE
1,200
1,000
800
600
400
200
0
250
200
150
100
50
0
250
200
150
100
50
0
1,600
1,400
1,200
1,000
800
600
400
200
0
2,000
1,800
1,600
1,400
1,200
1,000
800
600
400
200
0
Yearly needs [kilotonnes]Yearly needs [kilotonnes]
Yearly needs [kilotonnes]
Yearly needs [kilotonnes]
Yearly needs [kilotonnes]
EVs
EVs
EVs
EVs
Wind offshore
Batteries
Wind onshore
Solar PV
Other
Other
Batteries
2036
2036
2036
2036
2036
2024
2024
2024
2024
2024
2040
2040
2040
2040
2040
2050
2050
2050
2050
2050
EVs
Solar PV
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
ESTIMATED USE OF SELECTED MATERIALS FOR BELGIUM FIGURE 551
COPPER
GRAPHITE
LITHIUM
NICKEL
COBALT
35
30
25
20
15
10
5
0
50
45
40
35
30
25
20
15
10
5
0
60
50
40
30
20
10
0
8
7
6
5
4
3
2
1
0
7
6
5
4
3
2
1
0
Yearly needs [kilotonnes]
Yearly needs [kilotonnes]
Yearly needs [kilotonnes]
Yearly needs [kilotonnes]
Yearly needs [kilotonnes]
EVs EVs
EVs EVs
EVs
Wind offshore
Wind onshore
Solar PV
Other
Other
Batteries Batteries
Solar PV
2036
2036
2036
2036
2036
2024
2024
2024
2024
2024
2040
2040
2040
2040
2040
2050
2050
2050
2050
2050
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
GADEEL
217
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium216
5.8.2. LONG-DURATION ENERGY STORAGE
While this study covers the most important technologies for
2050 as they are known today, unforeseen technological break-
throughs could happen in the lead-up to 2050 which enable a
transition at a lower cost. One potential such breakthrough is
the advent of cheap long-duration energy storage (LDES). Such
storage types have energy contents that enable them to work
on an interday or even multi-day basis (see also Figure 5-52).
LDES therefore has the potential to assist during longer periods
of shortage where their shorter-duration counterparts do not.
LDES is, like shorter duration storage, capable of smoothening
out variations in electricity supply on a short-term basis (see also
5.4.5 - Ramping) but also has the potential to do this on an inter-
day or even multi-day basis. To provide some insights on the
behaviour of such technologies in the energy system an ex-post
optimisation based on electricity prices was performed. Figure
5-52 shows the amount of hours a storage system with a given
energy content and efficiency are discharging in the DE scenario
with central renewables. The amount of hours during which the
storage is discharging gives an idea of how actively the storage
was used. A first thing that can be noted is that storage systems
with a higher efficiency are more actively used than their lower
efficiency counterparts. Given their higher efficiency and thus
lower losses, they are able to take advantage of smaller price
variations and thus are used more often. A second observation is
that, for a given efficiency, the longer the charge/discharge time
the more hours the battery is used. Indeed, batteries capable
of delivering their nominal power for more consecutive hours
are capable of charging and discharging for longer durations
of low and high prices. Finally, a more limited increase in hours
of charging per year per additional hour of storage is seen once
battery sizes reach 4 to 8 hours, depending on the efficiency of
the batteries. This indicates that a significant chunk of battery
operation happens during price phenomena with a duration of
4 to 8 hours. In this case, solar PV generation causes the effect.
In addition to the figure, simulations are performed for storage
up to a duration of 100h in which the slowly increasing amount
of hours of discharging when compared to the hours of storage
is confirmed.
LONG DURATION ENERGY STORAGE AMOUNT OF CHARGING VERSUS HOURS OF STORAGE
FOR DIFFERENT EFFICIENCIES IN THE DE SCENARIO WITH CENTRAL RES IN 2050. FIGURE 552
3500
3000
2500
2000
1500
1000
500
0
90%
80%
70%
60%
50%
40%
Hours discharging per year [h]
0 5 10
Hours of storage [h]
Storage efficiency
15 20 25
5.8.3. MARGINAL COSTS AND PRODUCTION COSTS
The focus throughout this report is on finding the energy sys-
tem which reaches GHG emission targets with the lowest costs
and as such maximises the benefits for society. The way these
benefits and costs are distributed throughout society will ulti-
mately depend on the market design and financial incentives
that are put in place. One question one might ask is: ‘Would
the growing share of renewables in the electricity mix, which
have a very low operational cost, result in a lot of moments with
zero or very low marginal prices?’. To explore this question, the
hourly energy mix obtained in the DE 2050 scenario for a given
Monte Carlo year is subdivided into technologies with a low
operational cost (renewable and low carbon production such as
nuclear), technologies with a with high marginal cost (such as gas
based generation), flexibility (which discharges when marginal
costs are higher and charges when marginal costs are lower)
and imports. The hourly results are then sorted from moments
with the highest marginal prices to lowest marginal prices. The
results for the DE 2050 scenario sorted by marginal prices in
Belgium and for the domestic renewable generation (excluding
hybrids) of Belgium and its neighbouring countries (Netherlands,
Luxembourg, Germany and France) is shown in Figure 5-53. Sev-
eral interesting observations can be made:
◆ Gas-based generation is running somewhere in Belgium or
its neighbouring countries 35% of the time (while represent-
ing less than 6% of the energy mix for the region and Monte
Carlo year shown). This indicates that marginal prices are in
the high range at these moments in at least one of the con-
sidered zones.
◆ 75% of the time flexibility is used to reduce demand or increase
supply, implying that marginal prices are high enough to war-
rant the activation of these technologies.
◆ Finally, marginal prices are not guaranteed to be zero in the
remainder of the hours as the charging of storage or produc-
tion of hydrogen through electrolysers may be the technol-
ogy setting the price.
◆ As a result, it can be concluded that marginal prices are likely
to remain positive for a significant amount of the year even in
a renewables-dominated electricity system.
Finally, it should be noted that having a significant share of the
production producing at near-zero marginal cost means that for
a given hour the marginal cost could be significantly higher than
the average operational costs.
ENERGY MIX AS A SHARE OF DEMAND FOR BELGIUM AND NEIGHBOURING AREAS FRANCE,
GERMANY, LUXEMBURG, NETHERLANDS SORTED BY MARGINAL PRICE IN BELGIUM FIGURE 553
200%
150%
100%
50%
0%
-50%
-100%
Share of denmand [%]
High marginal cost (CH
4
& H
2
)
High marginal cost units produce in 35% of all hours
yet represent less than 6% of the energy produced
most expensive hours least expensive hours
Flexibility
Import
Low marginal cost (RES and low carbon)
219
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Results for Belgium218
5.9. KEY TAKEAWAYS
Building further on the key take aways of Chapter 4, this Chapter explored first the multi-energy results for Belgium in order to give
the overall picture. The core of the Chapter treats the different supply options (and demand levers) identified for Belgium. In addition,
a focus was made on the required infrastructure. Finally, other implications of the scenarios such as the material used were touched
upon. Several key insights can be extracted from the various analyses, and these are outlined below.
THE MULTIENERGY PERSPECTIVE
All investigated scenarios assumed that Belgium’s final energy
demand is estimated to decrease by 25 to 45% by 2050. This
decrease is mainly enabled through an increased energy effi-
ciency: notably the electrification of energy consumption, which
is energetically more efficient than the use of molecules. This
results in significant challenges for the electricity system but
also creates opportunities for Belgium as half of the country’s
electricity supply in the lead-up to 2050 is still to be defined.
The main take aways on system level for Belgium align closely
with those on a European scale. However, there are several unique
aspects that need to be considered for Belgium:
◆ Belgium’s domestic renewable energy potential is limited
and will not be able to cover the entire energy demand;
◆ Belgium has been, and will continue to be, a significant
energy importer, although this is expected to decrease in the
future as domestic renewable energy generation increases
and final energy demand decreases;
◆ The mixes for the different energy vectors were provided
in this chapter in order to evaluate their annual supply bal-
ances, even though this was not the central focus of the study.
The results highlight that there are still many trade-offs to be
made between the type of molecules to be used as carrier as
well as their source;
◆ The coupling between the electricity and molecule sec-
tor decreases in all studied sensitivities compared to today.
Therefore, it can be argued that the infrastructure design of
electron and molecule systems can be decoupled in Belgium.
However, attention should be paid to the location of the inter-
actions (power plants and electrolysers);
◆ Estimated total costs for the different demand scenarios show
that the most electrified scenario leads to the lowest total
system costs (all energy vectors and end-uses included).
ELECTRICITY SYSTEM IMPACT
Flexibility
The dependence of renewable energies on weather conditions
results in the electricity system’s supply becoming increasingly
volatile. The development of and access to different modes of
flexibility as well as the European integrated electricity market will
be key to manage this volatility. Both will be essential to manage
the energy system in the most cost-efficient way and to limit the
economic curtailment of RES.
Adequacy
The tools for managing adequacy were put in place by the outgo-
ing government. The need for similar tools will be felt throughout
the analysed horizon. However, the relative contribution of ade-
quacy measures to the overall cost of the future energy system
is rather limited, and technical solutions can be deployed within
a relatively short timeframe.
Grid development
No-regret infrastructure investments in distribution, local trans-
missions grids, industrial clusters and are typically linked to the
electrification of demand and the development of domestic RES.
In addition onshore interconnections were found to be beneficial
in all studies scenarios. These investments should be prioritised
and implemented without delay to avoid any potential setbacks
in the energy transition.
Other important grid investments, particularly in the Belgian
backbone and offshore grid, depend significantly on policy deci-
sions regarding Belgium’s electricity mix. Timely decisions on the
future energy landscape are crucial to dimension and implement
the necessary changes to the electricity grid.
CHOICES RELATED TO BELGIUM’S ELECTRICITY SUPPLY AND DEMAND
The ‘current policies’ scenario (assuming the current genera-
tion development plans including additional domestic offshore,
flexibility assumptions, nuclear phase-out and thermal capacities
to ensure adequacy) results in:
◆ Net imports of around 60 TWh in 2050 with a strong increase
of both imports and exports;
◆ Total thermal capacity requirements of between 13 and 15 GW
in addition to almost 20 GW of other flexible capacities (stor-
age, demand response…);
◆ Renewable domestic generation amounts to around 100
TWh, while thermal generation amounts between 15 to 30
TWh.
When looking at the ‘current policies’ scenario, the difference
between domestic electricity supply and demand (excluding
thermal generation and imports) is between 70 and 90 TWh
in 2050. Several options to face this challenge were identified.
Leveraging their combined potential, Belgium has the means
to cover its growing demand.
Demand levers
The moderation of energy consumption (sufficiency) holds a great
deal of potential in terms of keeping system costs under con-
trol. Sufficiency is predominantly related to changes in human
behaviour. The main challenge of this measure lies in citizen
acceptance, particularly when individuals believe that changes
in their behaviour will lead to a loss of comfort. It was found that
sufficiency measures have the potential to reduce the total
system costs by 15%. Behavioural changes typically do not hap-
pen overnight but rather evolve gradually. As such, some energy
demand reduction can happen relatively fast, but other measures
need support and policies to materialize over time.
Small-scale supply options
When accounting for the total system cost, maximising the
development of domestic renewable energy (onshore wind,
PV panels and offshore wind in the Belgian EEZ) is demonstrated
to be part of a cost-optimal solution for Belgium in all scenarios.
While Belgium’s domestic RES generation can contribute sig-
nificantly to the electricity supply mix, this alone will not suffice.
Several additional options are however available for meeting the
country’s increasing electricity demand.
Large-scale supply options
As a large-scale energy source, non-domestic offshore wind
appears to be a cost-effective supply option for Belgium. None-
theless, the scaling up of offshore wind development requires a
step change in international coordination, joint planning, and
funding. The benefits of non-domestic offshore wind therefore
need to be weighed against other supply options, such as the
development of new nuclear generation units or connecting far-
out baseload RES. While new nuclear plants are a viable solution,
this option carries its own challenges related to areas including
safety, complexity, and financing. Important elements linked to
these options are the cost assumptions, the time to market and
the risk profile (technological, financial, environmental, etc.) of
each technology.
Imports versus domestic generation
One key policy choice that should be made relates to finding
the right balance (over time) between relying on electricity
imports and undertaking domestic investments in electricity
supply. Numerous considerations have to be made. These include:
affordability considerations, opportunities for the redistribution of
costs and benefits, agility in the face of uncertainties, resilience
against supply shocks,
international cooperation to ensure a coordinated approach to
offshore development, risks related to budget and timing over-
runs, private-public partnerships for financing, funding, etc.
221
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix220
6.
APPENDIX
Appendix A - KARI dispatch and investment
electricity model 221
Appendix B - Molecules and liquids model 225
Appendix C - Carbon capture, utilisation and
storage model 227
Appendix D - Adequacy electricity model 228
Appendix E – Marginal Abatement Cost Curve
methodology 230
Appendix F – Total cost methodology 235
Appendix G – Schematic view of the model 243
Appendix H – Non-CO2 emissions methodology 244
Appendix I – Details on energy demand 246
Most commonly used abbreviations 247
References 248
APPENDIX A - KARI DISPATCH AND
INVESTMENT ELECTRICITY MODEL
The KARI electricity model was developed and used by Elia Trans-
mission Belgium over several years. Amongst other it was used
as part of the Federal Development Plan which was published
in 2023 [ELI-3]. Its initial goal was to perform an identification of
system needs in terms of both European and Belgian electricity
grid infrastructure reinforcements. This included determining
an appropriate quantity and strategy for integrating offshore
RES into the system as well as suitable cross-border transmis-
sion capacity increases. The objective of the study was to satisfy
European and Belgian policy targets as much as possible whilst
adhering to relevant system constraints. As part of this current
study, the KARI model and tools have been further developed
in order to include the interactions with other energy vectors as
explained in appendices B and C.
The KARI model uses the open-source Antares simulation soft-
ware (also detailed further in the appendices).
INTRODUCTION
The KARI model is used to find an optimal electricity system in
Europe, with the following in mind:
◆ ‘optimal’: the goal is not to find a unique optimal system but
one possible optimal system alongside others because the
final results are influenced by many factors. The aim is thus
to identify which general trends remain valid across different
assessed futures.
◆ ‘electricity system’: the objective is much broader than identi-
fying the need for new offshore infrastructure. Offshore infra-
structure, onshore infrastructure, offshore RES capacities, the
location of electrolysers, thermal capacities, CO2 emissions,
total system costs, etc… are all intrinsically linked to each
other. An integrated approach is therefore required, which
takes into account the electricity system as a whole.
◆ ‘in Europe’: the challenge of integrating RES into the system
is first and foremost a European one. In order for Belgium
to support the achievement of European objectives, and
because Belgium is an important European crossroads, its
grid must be conceived within and developed in a European
context.
Starting with a reference grid and a set of general assumptions
(regarding national targets, standard costs, offshore RES poten-
tial…) and scenario-specific assumptions (regarding the produc-
tion mix, electricity consumption, flexibility means), KARI grad-
ually and optimally integrates additional offshore RES, expands
the grid of the electrical system until a given goal is achieved.
Constraints can be added to the optimisation to reflect certain
realities (e.g. limited capacity increases per decade along a given
corridor) or in order to compare different network development
strategies (for example, no onshore developments or offshore
hybrid systems). Of course, when interpreting the study’s results,
attention should be paid to the criteria which cannot be factored
into the analysis – for example, political support, environmental
changes or public acceptance.
It is also very important to keep in mind that the solution found
by the optimiser is one possible outcome (which minimises total
costs) but that other solutions, close to optimum are also possi-
ble and could be easier to implement due to other constraints
(political support, acceptance…).
OPTIMISED VARIABLES
As illustrated inFigure A-1, the optimisation is based on pre-fixed
scenarios (depicted in the upper half of the figure, shaded grey),
with the remaining factors optimised by the tool (depicted in
the lower half of the figure, shaded light orange): offshore wind,
electricity network (onshore and offshore), other thermal fleet,
power-to-X.
Regarding the electricity network and offshore wind farms, the
type of connections for offshore wind farms that can be invested
in by the KARI model are shown in Figure 3-13 (Section 3.1.3.2),
which complements the onshore or offshore point-to-point inter-
connectors between two zones.
Generic input data per link type (HVDC link versus AC reinforce-
ments, onshore versus offshore cabling, distance, etc.) is used
for the pan-EU system, following a standard cost approach. This
inherent approximation of reality allows the optimisation to jointly
and fairly consider all available options.
STARTING GRID AND CANDIDATES
The starting grid in the KARI model is based on ENTSO-E’s TYNDP
2022. Every two years, ENTSO-E undertakes an identification of
system needs (IoSN): a grid study exercise performed , amongst
other things, to detect the grid infrastructure reinforcements
which need to be carried out in a timely manner for the power
system as a whole. As part of this, a zonal reference grid is used
for each target year, based on input from TSOs. This means, for
example, that the onshore network in Belgium is already signif-
icantly reinforced in the starting grid (i.e. HTLS backbone). The
assumptions regarding the starting grid can be found in the
scenarios chapter of this study.
The zonal approach (including the use of Kirchhoff laws), in which
countries are divided in several zones, enables internal grid bot-
tlenecks to be reflected more accurately. For long-term stud-
ies, such an approach allows for an overall optimal system to be
found independently from current market design rules. This is
relevant, since market rules can still evolve within the considered
time horizon. This approach therefore proposes a long-term sys-
223
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix222
tem-wide optimum. Such an approach is also the one followed
in other long-term studies, including the TYNDP zonal IoSN or
the e-Highways 2050 study [EUC-10].
The offshore wind potential is identified via a detailed approach,
starting with a database from 4C Offshore [4CO-1], and consider-
ing both geographical constraints (bathymetry, shipping routes,
environment zones, etc.) and latest identified offshore zones in
national plans. The maximum offshore RES capacity which might
be electrically integrated in Europe for each target time horizon
is outlined in Section 3.1.3.2.
All possible offshore wind farm candidates are aggregated into
offshore wind farms with a standard size of 2 GW each. Existing
wind farms are modelled with their rated capacities. A distinction
is made between standard offshore wind farms and radially-con-
nected hubs. Radially-connected hubs consist of several offshore
wind farms that converge onto a central offshore location (or
hub) which is connected to the onshore grid. Six hubs are placed
in the starting grid in order to reflect current ambitions. One of
them is located in the Belgian part of the North Sea, simulat-
ing the Princess Elisabeth Island (which will be connected to 3.5
GW of offshore wind capacity). Offshore hubs are assumed to be
able to welcome, without extra cost on the hub, other (hybrid)
interconnectors and/or wind farms. Other hubs can be added
to the model.
Four ‘types’ of standard offshore wind farms are considered, as
follows (and illustrated in Figure 3-12).
◆ Existing offshore wind farms which are already radially con-
nected to their respective countries.
◆ Pre-connected offshore wind farms that are due to be
radially connected to their respective countries. These are
assumed to be ready by 2036 and are thus present in the ini-
tial model for 2036.
◆ Pre-connected offshore wind farms that are due to be radi-
ally connected to their respective countries but are assumed
to be potential candidates for hybrid interconnectors. These
are assumed to be ready by 2036 and their radial connections
are thus present in the initial model for 2036. The optimiser
can, however, decide to take these into account as hybrid
interconnectors.
◆ Offshore wind farm candidates. These will only be present if
the optimiser invests in the offshore wind farm and its associ-
ated connection (either a radial or hybrid connection).
The cost of the offshore wind farms is also considered by the
algorithm and a distinction is made between fixed-bottom wind
turbines and floating wind turbines. The distance to shore is also
reflected in the considered costs.
GOAL FUNCTION AND CONSTRAINTS
Different optimal solutions can be sought out for the different
time horizons, in accordance with the used goal function. The
goal function used for KARI as part of this study is the lowest
total system cost (including operational costs (including CO
2
emissions), wind and transmission annuity). The model interacts
with the ‘molecule’ model and the investments are made in order
to ensure that the CO2 targets are met.
FIXED AND OPTIMISED VARIABLES IN THE KARI MODEL FIGURE A1
O
P
T
I
M
I
S
E
D
B
Y
T
H
E
T
O
O
L
Consumption
Demand response
Storage
Power-to-X location
(heat, molecules)
Onshore and offshore
electricity network
Offshore wind
Other thermal
Nuclear
Onshore wind
PV
Hydro
F
l
e
x
i
b
i
l
i
t
y
G
e
n
e
r
a
t
i
o
n
D
e
m
a
n
d
G
r
i
d
CONSTRAINTS
Some general constraints are imposed in order to better reflect
reality and ensure that the study’s outputs are realistic. Amongst
other constraints, the connections to/from a radially pre-con-
nected offshore wind farm can ‘only’ evolve into a hybrid system
by connecting one additional leg to another hub or country. The
goal is to reflect typical space limitations on offshore platforms.
However, no limitations are placed on known energy hubs, since
it is assumed that space limitation is less of an issue and/or better
planned out for them. The direct (hybrid or radial) connections
linking offshore RES to the onshore network are limited to real-
istic lengths; in other words, they are connected to coastal or
near-coastal areas. Similarly, the pace at which possible capacity
increases occur in each corridor is limited to adequately reflect the
lead time of such projects. It is assumed that the required tech-
nological progress will have been achieved by then, allowing the
realisation of envisaged multi-terminal systems. In other words,
no specific technological constraints related to the development
of multi-terminal systems are considered.
KARI AND THE ADEQUACY MODEL USE THE OPENSOURCE ANTARESSIMULATOR
The Antares-Simulator (hereafter after ‘Antares’) is an open-source
hourly electricity market simulator which was developed by RTE
[ANT-1] and has been used by Elia Transmission Belgium to per-
form the simulations for both adequacy and economic assess-
ments. In addition, the output of the tool is also used as input
for assessing the flexibility means. Antares is a UC/ED model
as it calculates the optimal unit commitment and generation
dispatch from an economic perspective; in other words, it min-
imises the generation costs of the system while respecting the
technical constraints of each generation unit. The dispatchable
generation (including thermal and hydroelectric generation,
storage facilities and demand side response) and the resulting
cross-border market exchanges constitute the decision variables
of the optimisation problem.
Antares simulates each year by solving fifty-two weekly optimi-
sation problems in a row along the whole European perimeter
for each 'Monte Carlo' year. This results in an hourly dispatch over
the whole year for all technologies implemented in the model,
considering all generation, storage and market response capac-
ities as well as interconnection flows.
KEY MODELLING ASSUMPTIONS TO KEEP IN MIND WHEN INTERPRETING THE RESULTS
It is important to highlight several modelling assumptions in
order to correctly interpret the results. These are outlined below
and need to be kept in mind when analysing the results.
◆ Perfect weekly foresight is considered for renewable gen-
eration, consumption and unit availability (known one week
in advance following an ex ante draw). This also means that
storage, hydro reservoirs and thermal dispatch are optimised
knowing all of this in advance. In reality, this is not the case,
as forecasting deviations and unexpected unit and intercon-
nection outages can happen and need to be covered by the
system. In line with the ERAA methodology, for each market
zone, in order to cope with such events, part of the capacity
is therefore reserved for balancing purposes and cannot be
dispatched by the model.
◆ Simulations of the market are performed on the basis that all
the energy is sold and bought on an hourly basis. Integrat-
ing long (i.e. capacity markets) and/or real-time markets (i.e.
balancing markets) in such a model is not straightforward.
Forward markets are assumed to act as financial instruments
which anticipate day-ahead/real-time prices. Depending on
the trading strategy and actual market conditions, an arbi-
trage value may exist between different time frames.
◆ The model minimises the total cost of generation (including
energy not served) across the whole simulated system.
◆ A perfect market is assumed (no market power, bidding
strategies...) as part of the model’s scope. The optimisation
solves all the system (i.e. the whole geographical perimeter)
at once.
◆ Energy Limited Resources (ELR) such as pumped storage
units, batteries and demand side response, modelled as
‘in-the-market’, are dispatched/activated in order to mini-
mise the total cost of operation of the system. In reality, they
could be used to net a certain load in a smaller zone or to
react to other signals. The modelling approach also assumes
that price signals are driving the economic dispatch of those.
During times of scarcity, ELRs (such as storage or demand
response) can be dispatched in different ways. In this respect,
the default ‘shedding policy’ in Antares (i.e. ‘shave peaks’ see
[ANT-1]), is used in the simulations. This 'shedding policy' aims
to minimise the depth of the ENS, in line with the reliability
standard calculation.
◆ Prices calculated in the model are based on the marginal
cost/activation of each unit/technology while considering the
modelled network constraints and their shadow prices.
◆ The efficiency of each thermal unit is considered as fixed and
independent of the loading of the unit. In actual fact, effi-
ciency is a function of the generated power.
◆ Each bidding zone is considered a copper plate. This means
that internal grid limitations within a bidding zone are not
considered. In practice, some units can be re-dispatched in
order to limit congestion on a grid.
◆ Offshore hybrid interconnectors (i.e. interconnectors which
combine both offshore wind and market-to-market con-
nections) are modelled under the assumption that the wind
farms which are connected to the interconnector are in a sep-
arate bidding zone.
225
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix224
The Antares-Simulator is an open-source software devel-
oped by RTE. It is a sequential ‘Monte Carlo’ simulator
designed for short- to long-term studies related to large
interconnected power grids. It simulates the economic
behaviour of a given transmission-generation system
over the period of one year and on an hourly basis.
Elia Transmission Belgium (hereafter Elia) has been using
the software for more than 10 years. It is used for perform-
ing the simulations that are used as part of the capacity
mechanism calibration in Belgium (Strategic Reserves
and more recently the market-wide CRM) and has been
used for Elia’s adequacy and flexibility studies since their
launch in 2016.
Antares has been used in several studies across Europe,
including studies undertaken by ENTSO-E, which uses
it as market modelling software. These ENTSO-E studies
include:
◆ the pan-European Resource Adequacy Assessment
(ERAA) that ENTSO-E publishes every year [ENT-4];
◆ the assessment related to the ten-year network devel-
opment plan (TYNDP, [ENT-3]) that ENTSO-E publishes
every two years.
Moreover, Antares is used as the reference market model-
ling software as part of many other European projects and
national assessments. In addition to the adequacy studies
performed by Elia and the economic assessment of the
Belgian federal grid development plan, the tool has been
used for (non-exhaustive list):
◆ the ‘Bilan Prévisionnel’, published by RTE [RTE-2], which
assesses the adequacy of France’s system for the years
2023 to 2035;
◆ RTE’s analysis of trends and perspectives across the
energy sector (transition to low-carbon hydrogen in
France or integration of electric vehicles into the power
system) [RTE-2];
◆ RTE’s Energy pathways 2050 (‘Futurs énergétiques
2050’) [RTE-1];
◆ the OSMOSE project [OSM-1];
◆ the Cigré Working Group C1.35: Global Electricity Net-
work Feasibility Study [CIG-1];
◆ e-Highway 2050, which aims to develop a grid planning
methodology [EUC-10];
◆ MedTSO studies [MED-1];
◆ Litgrid Adequacy Assessment [LIT-1];
◆ APG (Austrian TSO) Electricity stress test for the security
of supply in winter 2022-23 [APG-1].
For the creation of annual scenarios, Antares can be pro-
vided with ready-made time series or can generate them
through a given set of parameters. Based on this input
data, a panel of ‘Monte Carlo’ years is generated through
the association of different time series (randomly or as set
by the user). Then, an assessment of the supply-demand
balance for each hour of the simulated year is performed
by subtracting wind and solar generation from the load,
by managing hydro energy with a heuristic approach
and by optimising the dispatch and unit-commitment
of thermal generation clusters, storage and demand side
response. The main goal is to minimise the total cost of
generation for all interconnected areas.
Finally, RTE international (RTE-i) has developed a users
club for Antares. This gathers different users together to
enhance the application, provide training and support for
it, and guide its development. TSOs which are members
of this club: APG, Elia, EMS, Swissgrid, SEPS, IPTO, ELES,
MAVIR, MEPSO, ESO, OST.
BOX A-1
APPENDIX B - MOLECULES AND LIQUIDS
MODEL
The KARI model is linked to a molecules and liquids model. One
of the main improvements identified through discussions with
stakeholders was the need to incorporate other energy vectors
into the model. This is because, to accurately determine if a region
can achieve carbon neutrality, all types of energy carriers need
to be modelled. To fully understand the interplay between these
markets, Elia Transmission Belgium has developed a ‘molecules’
model. Unlike an electricity model where supply and demand
need to match at every hour, a gas-based model allows for flexibil-
ity due to the variable compression of gas within pipelines. There-
fore, the molecules model is designed to operate on a (twice-)daily
basis rather than an hourly one. To simplify the model, a year's
worth of data is condensed into one week. This means that one
hour in the model equates to 52 hours in a full year, with average
values calculated over these 52 hours blocks.
The molecules model is comprised of four molecule types: meth-
ane, hydrogen, ammonia, and liquids. An overview of these is
provided in Figure 2.5. Each type can be transformed into another,
following the conversion processes outlined in Section 3.1.6.
For methane, Belgium, France, Germany, and the Netherlands are
modelled individually, while other countries are grouped together
based on their geographical proximity. The existing methane grid
and storage capacities for each country are used. Additionally,
existing import pipelines from outside the EU and existing LNG
import terminals are incorporated. All data has been aligned and
verified with Fluxys. This methane grid remains consistent across
all time horizons.
Methane can then be converted into hydrogen through the use
of steam methane reforming (SMR-H2). In this part of the model,
every European country is modelled individually. In the 2036
scenario, there are no existing hydrogen pipelines or storage
facilities. Therefore, the hydrogen aspect of the model is subject
to optimisation during each iteration, with potential investments
in hydrogen pipelines (either within the EU or for imports), stor-
age facilities, or offshore electrolysers in the North Sea. Once an
optimal hydrogen grid is established in 2036, this grid serves as
the starting point for future years.
Hydrogen can also be obtained through the imports by sea and
the subsequent cracking of ammonia. However, this method is
only feasible for countries that are not landlocked.
In the model, ammonia is treated separately. The ammonia node
is designed to meet all of Europe's ammonia needs, which allows
for the immediate use of imported ammonia. Additional ammo
-
nia imports can then be transformed into hydrogen to satisfy
the demand for hydrogen. Potential import regions include the
Middle East, Africa, Australia, and South America. The associated
costs of these imports are determined by factors such as ship-
ping distance, the price of renewable energy in the country of
origin, and the expenses involved in conversion. Further details
regarding these costs can be found in Section 3.3.5. Additionally, in
scenarios where there is an excess of hydrogen, it can be domes-
tically converted into ammonia to meet the demand.
Finally, the model includes a category for liquids, which is further
divided into four sub-categories: aviation, feedstock, shipping,
and others.
The 'aviation' category encompasses kerosene, e-kerosene, and
biofuels (sustainable aviation fuels, or SAF). 'Feedstock' includes
naphtha, methanol, and biofuels. The 'shipping' category covers
bunkering fuels, methanol, and biofuels. The 'other' category
comprises traditional fuels (mazut, gasoline, and diesel), synthetic
fuels (e-diesel, methanol, etc.), and biofuels (bio-ethanol).
Each of these fuels have unique efficiencies when it comes to
creating synthetic variants from hydrogen and varying import
costs. Therefore, they are individually modelled.
227
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix226
ADDITIONAL INFORMATION ON THE IMPORT COSTS
Section 3.1.4 previously provided an overview of the merit order
utilised in the model. However, this did not factor in a CO2 price,
leading to fossil fuels occupying the initial positions in the merit
order. The molecule model incorporates an optimal CO2 price to
achieve the specific carbon reduction goals for a given year. This
inclusion of a CO
2
price effectively displaces fossil sources from the
merit order. Figure B-1 offers a comparative example of a merit
order for 2050, both with and without a CO2 price.
MOLECULES AND LIQUID MERIT ORDER BASED ON CO2 PRICES FIGURE B1
METHANE
HYDROGEN
LIQUIDS
CO2 PRICE OF 0 €TONCO2CO2 PRICE OF 400 €TONCO2
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000
200
150
100
50
0
Price [€/MWh]
200
150
100
50
0
Price [€/MWh]
200
150
100
50
0
Price [€/MWh]
200
150
100
50
0
Price [€/MWh]
200
150
100
50
0
Price [€/MWh]
200
150
100
50
0
Price [€/MWh]
Generation [TWh]
Generation [TWh]
Generation [TWh] Generation [TWh]
Generation [TWh]
Generation [TWh]
Domestic fossil methane Domestic biomethane Fossil methane import – pipeline
Synthetic methane import – pipeline Fossil methane import – LNG Synthetic methane import – LNG Methane from hydrogen
Hydrogen import Ammonia import SMR-H
2
– domestic SMR-H
2
– import
Liquid from hydrogen Domestic bioliquids Domestic fossil liquids Synthetic liquids import Fossil liquids import
APPENDIX C - CARBON CAPTURE,
UTILISATION AND STORAGE MODEL
The overall goal function in the present study is a minimisation
of the total European system costs subject to a maximal GHG
emission target. To ensure GHG emission targets are met at the
lowest cost the model has the possibility to invest in a wide variety
of options who have the potential to reduce emissions. These
options include:
◆ The reduction of the use of fossil fuels (offshore, reconversion
to hydrogen turbines, etc…);
◆ The production and/or import of decarbonised molecules;
◆ Carbon capture, utilisation and sequestration .
Note that other options to reduce greenhouse gas emissions such
as increasing the efficiency of appliances, heating efficiently using
renewable electricity or energy sufficiency are not optimised but
assessed through scenarios and sensitivities. In addition, it should
be noted that if the model would not be able to reach the targeted
emissions reduction, a very high cost is attributed to the carbon
emitted above the limit.
The optimal activation of carbon removal options is orchestrated
by the carbon capture, utilisation and storage model. This model
collects the carbon emissions from each of the other models and
has the capability to monitor the total GHG emissions and link
this information to (both direct and indirect) decisions regarding
infrastructure developments. More specifically the model can
invest directly in technologies such as carbon capture infrastruc-
ture for process emissions and other transformation processes
but also in Direct Air Capture (DAC). Furthermore, indirectly the
model can influence the dispatch in the other models by setting
a carbon price. This is illustrated in Figure C-1.
INTEGRATION OF CARBON CAPTURE AND STORAGE MODEL IN THE TOOLCHAIN FIGURE C1
Electricity model
H
2
model
H
2
H
2
H
2
H
2
H
2
H
2
H
2
Liquids model
Electric and (non-CO2) molecule modelling Carbon capture and storage model
Modelled
Price data
(to rank
options
by cost)
Selected
options
and
CO
2
price
CH
4
model
CH
4
CH
4
CH
4
CH
4
CH
4
CH
4
CH
4
Combine modelled emissions with
unmodelled emissions (process emissions,
LULUCF, carbon stored in materials, non-CO
2
emissions...)
Select most cost-effective options to reach
the emissions target taking into account
technical constraints on the max amount of
CO
2
stored
Options activated in the carbon model itself:
- CCS for thermal power generation
- CCS for process emissions per sector
(cement, chemicals, ...)
- Reconversion of thermal power generation
to H
2
- Investments in direct air capture
underground storage of carbon
Options activated by the carbon capture and
storage model in the Electric and (non-CO
2
)
models though the carbon price:
- Import of green molecules
- Increase the price of carbon-intensive
electricity generation
- Investment in offshore wind
- Investments in interconnections between
regions with available low-carbon supply
options to regions with carbon emissions.
- Investment in direct-res offshore hydrogen
generation
It is important to note that the CO2 infrastructure is not modelled explicitly. However the costs of the different carbon capture options
are assumed to integrate those costs.
229
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix228
APPENDIX D - ADEQUACY ELECTRICITY
MODEL
The goal of the adequacy electricity model is to evaluate the
required capacity to comply with the adequacy criteria at Bel-
gian and European level. This indicator is calculated based on
an iterative process at European level that is required to meet
the assumed reliability standard of each modelled country. The
required capacity is defined as the thermal capacity layer needed
on top of all capacities (generation, storage, demand response)
and the grid configuration assumed in each scenario.
This model is performed at bidding zone level. As the geographic
resolution is lower than the KARI model, it allows to perform a full
adequacy assessment such as performed in Elia’s adequacy and
flexibility studies. The adequacy assessment is performed by sim-
ulating a large amount of ‘Monte Carlo’ years, based on the same
forward-looking climate database used in the rest of the study.
More information on the adequacy assessment can be found
in the Adequacy and Flexibility study performed by Elia [ELI-1].
The methodology associated to the adequacy electricity model
requires three steps to be followed, as illustrated on Figure D-1.
This methodology is performed for each time horizon and sce-
nario.
STEP 1: AGGREGATION
The first step consists of aggregating the zonal KARI electricity
model (see Appendix A) into an adequacy model. The electricity
consumption, generation, storage and demand response of each
zone from the KARI electricity model before and after invest-
ment optimisation are aggregated at bidding zones level. For the
model after optimisation, the additional investments in the grid
and in offshore wind are also integrated. The adequacy models
obtained are therefore an image of the KARI electricity model
with a lower resolution.
STEP 2: EUROPEAN ADEQUACY LOOP
Once the input from the zonal KARI electricity model are inte-
grated in the adequacy electricity models, an European adequacy
loop is performed. This assessment aims to ensure that all mod-
elled countries meet their reliability standard. In order to simplify
the assessment, all countries are assumed to be compliant with
their reliability standard (RS) (or to 3h of LOLE if RS is not available).
This step is performed as follows:
1. Future possible states (‘or ‘Monte Carlo’ years) covering the
uncertainty of the generation fleet and HVDC (technical fail-
ures) and weather conditions (impacting RES generation and
demand profiles due to thermosensitivity effects) are defined.
2.
An hourly European market simulation for each ‘Monte Carlo’
year is performed in order to identify for each bidding zone
structural shortage periods, i.e. moments during which the
electricity production in the market is not sufficient to satisfy
the electricity demand. The model allows a quantification of
the amount of hours during which the system is not adequate
for each future state.
3. If at least one country does not meet its reliability standard,
capacity is added or removed. This iterative process aims to
assess the additional capacity needed (100% available) in each
bidding zone to satisfy the 3 hours LOLE criteria.
4.
At the end of the process, a potential GAP is identified for
each bidding zone in both the adequacy electricity models
before and after optimisation.
STEP 3: ADEQUACY GAIN
Finally, the needed capacity can be used to quantify the adequacy
benefits. This can be performed by looking at the required capac-
ity defined in step 2 between two models. This indicator can be
calculated either as the additional capacity savings at European
level between the two models or as the equivalent benefit in M€
for the system to avoid building additional capacity. The same
analysis can be performed at Belgian level.
METHODOLOGY OF THE ADEQUACY ELECTRICITY MODEL FIGURE D1
Hourly European market simulation
for each ‘Monte Carlo’ year
Additional capacity required in Europe to ensure adequacy
Capacity
added/removed
if not compliant
Amount of scarcity hours
(LOLE) for each country h < ? SoS level
of each country
Definition of ‘Monte Carlo’ years
Additional capacity required
in Europe to ensure
adequacy
(after optimisation)
Additional capacity required
in Europe to ensure
adequacy
(before optimisation)
Adequacy model
(after optimisation)
Adequacy model
(before optimisation)
Model
(after optimisation)
Model
(before optimisation)
KARI electricity models
Adequacy electricity models
Adequacy indicators
GW Additional capacity savings at Belgium and European level
M€ Conversion of these savings into equivalent system benefit
STEP 1
Aggregation
STEP 2
EU
Adequacy Loop
Investment
optimisation
STEP 3
Adequacy Gain
ZONAL BIDDING ZONE
231
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix230
APPENDIX E – MARGINAL ABATEMENT
COST CURVE METHODOLOGY
This text was prepared by Sia Partners who performed the work
related to the calculation of the MACC (Marginal Abatement Cost
Curve).
INTRODUCTION
Steering the Belgian energy landscape towards climate neutrality
has proven to be a complex endeavour: it requires both a myriad
of policy measures as well as the implementation of multiple stra-
tegic abatement levers. Understanding the interactions between
these levers, both positive and negative, adds an additional layer
of complexity to the decarbonisation mission.
The MACC (Marginal Abatement Cost Curve) method is recog-
nised as a widely used framework in decarbonisation exercises.
While it may seem deceptively simple, the MACC can be valuable
because it quantifies the associated implementation costs. Its
visualisation provides guidance on where to allocate resources
first in a merit-order like way.
Setting up a robust MACC is a resource-intensive process that
demands a well-structured methodology. It starts by identifying
different sectors and their corresponding baseline emissions,
ensuring that data are sufficiently granular. Next, a non-exhaus-
tive list of abatement levers is compiled through a combination
of both internal expertise and literature review. The abatement
levers are then further assessed for each target year to estimate
their abatement costs, considering a wide range of input param-
eters.
Subsequently, the different abatement levers are plotted in what
is called the MACC, with the most cost-effective options posi-
tioned on the left and, towards the right, increasingly expensive
measures for each sector. However for this exercise, the MACC is
presented by sector and measures instead of sorting the different
them by a merit-order.
Although numerous technologies can be incorporated in the
MACC, this study focuses solely on demand-side solutions. This
does not mean that supply-side options are overlooked, but rather
that they have already been thoroughly examined in the report
through an extensive set of sensitivity analyses and other metrics
(such as LCOE).
CAVEAT
While the MACC offers an intuitive framework for understanding
different decarbonisation options, it comes with some limitations.
First, a MACC is tailored for marginal emissions reductions. This
marginal focus might prioritise incremental improvements over
strategic shifts.
Second, the interdependencies among emission reduction meas-
ures pose challenges. Decisions in one sector can affect outcomes
in others, complicating strategy development and imposing the
need for integrated approaches that consider intersectoral syn-
ergies and trade-offs.
Moreover, the assumption of uniform costs regardless of transi-
tion speed overlooks the higher costs associated with rapid transi-
tions in hard-to-abate sectors or slow transitions in easier-to-abate
sectors. Early consideration of each abatement lever becomes
crucial to mitigate long-term costs effectively. In this regard, tech-
nological advancements can also further complicate the MACC
approach, as evolving technology costs can significantly alter the
cost dynamics of certain emission reduction measures.
Finally, it is crucial to note that some solutions within a category
may exhibit high-cost heterogeneity due to unique characteristics
(e.g. renovation of dwellings). Consequently, the MACC should not
be viewed in isolation but should be used as an analytical tool
in combination with the other analyses carried out in this study.
IDENTIFIED CATEGORIES AND LEVERS
This Sia Partners’ analysis, commissioned by Elia, focuses on Bel-
gian demand sectors that play a pivotal role in national emission
reduction efforts. Through an intensive collaboration between Elia
and Sia Partners, different sectors were identified and quantified
based on their greenhouse gas (GHG) emission contribution as
well as their potential for implementing effective abatement
levers, drawing upon insights and data from Elia’s BluePrint study.
The identified sectors are:
◆ Buildings,
◆ Transportation,
◆ Industry,
◆ Carbon Capture Techniques.
Within each sector, a range of levers has been identified to effec-
tively target emission reductions. The non-exhaustive list of abate-
ment levers include:
◆ Electrification,
◆ Switch to green molecules,
◆ Technological advancements such as energy efficient appli-
ances,
◆ Innovative solutions such as carbon capture and storage.
By focusing on these specific categories and levers, a decarbon-
isation strategy is aimed for that maximises emission reductions
while optimizing resource allocation.
MACC CATEGORIES AND ABATEMENT
LEVERS TABLE E1
Category Abatement levers
Buildings
Energy efficient household appliances
Insulation (roof & facade)
Heat pumps (residential & tertiary buildings)
Hydrogen boilers (residential & tertiary
buildings)
Transportation
Light duty BEV
Heavy duty BEV
Light duty FCEV
Heavy duty FCEV
Industry
Efficiency improvement
Heat pumps (low temperature)
E-boiler (medium & high temperature)
Hydrogen boiler (medium & high
temperature)
CO2
Carbon capture
Carbon capture & storage
Direct air capture
METHODOLOGY FOR CONSTRUCTING THE
MACC 2030 AND 2040
Sia Partners’ methodology to construct a MACC for the years 2030
and 2040 runs in parallel with projections for those years, lever-
aging baseline data from respectively. Constructing a MACC for
each target year generally involves several steps. In each of these
steps, distinct methodologies and insights are utilised to tailor
the analysis to the specific abatement categories, being build-
ings, transportation, industry and carbon capture. Nonetheless,
a number of common phases can be identified.
The initial phase involves pinpointing GHG-intensive activities
within each category. This step requires an assessment of both the
emission intensity and the current cost structure, encompassing
CAPEX and OPEX. Understanding the inherent costs is critical for
evaluating the economic impact of emission reduction efforts.
Subsequently, the potential of various abatement levers is evalu-
ated using both in- and outputs from Elia’s BluePrint study, which
explores different options for achieving net-zero emissions. This
analysis considers factors such as energy prices and infrastructure
development costs. By assessing the potential of installing new
technologies and their associated costs, insights in their emission
reduction effectiveness can be acquired.
After identifying the potential of each abatement lever for every
target year, the Net Present Value (NPV) of implementing this
lever relative to maintaining the baseline activity is calculated.
This involves comparing the costs of adopting the new tech-
nology over time using a discount factor (predefined Weighted
Average Cost of Capital (WACC)). To determine the MACC, the
NPV then is divided by the discounted amount of the emission
abatement potential1.
1 A discount rate is applied to future GHG abatement, reflecting a preference for implementing the abatement lever earlier and a risk that future abatement may not occur at
the projected pace.
2 The analysis deliberately excluded hybrid vehicles as an abatement lever. This decision was driven by the focus on prioritizing measures with the most significant impact on
emission reductions.
Societal perspective
Throughout the MACC exercise, Sia Partners assumes a system
perspective, having the interest of society as its guiding principle.
This approach translates to the use of wholesale energy prices
(electricity, oil, natural gas and hydrogen) instead of retail prices
as input parameters.
Furthermore, the integration of the carbon price in the energy
parameters provides a nuanced understanding of the economic
implications of emission reduction strategies.
Buildings
For buildings, the analysis focuses on electrification, green hydro-
gen and energy efficiency.
In evaluating electrification, heat pumps were compared to tradi-
tional natural gas and oil-fired boilers. This comparison assesses
the potential benefits of transitioning to electric heating systems.
Subsequently, to explore the abatement lever of green hydrogen,
natural gas boilers were used as a baseline to examine the feasibil-
ity and effectiveness of hydrogen as an alternative energy source
in residential and tertiary heating, considering both emission
intensities and economic parameters.
Last, the analysis extends to energy efficiency, with a focus on
using current energy labels to identify potential improvements.
For insulation, for instance, a distinction is made between facade
and roof insulation. Additionally, for household appliances, con-
sideration is given to both low and high investment cost options,
recognizing the variety in costs. While using averages provides
an easy manner to tackle variety, the study can be extended to
(or solely focus on) examining additional factors and variations
in energy efficiency, such as different building types or specific
household appliances.
Transportation
As regards transportation, the analysis focuses on two key abate-
ment levers for both light-duty and heavy-duty transport: elec-
trification2 and green molecules.
In evaluating these options, various transport-related cost factors
were considered including operational expenditure (OPEX), cap-
ital expenditure (CAPEX) and fuel costs. Specifically for battery
electric vehicles (BEV), the costs of changing batteries as well as
infrastructure costs were examined. For fuel cell electric vehicles
(FCEV), the infrastructure cost was taken into account.
The assessment encompassed not only the initial costs but also
the long-term operational implications, ensuring a thorough
understanding of the economic considerations associated with
each option.
Industry
For industry, it is essential to acknowledge its inherent hetero-
geneity. This sector is composed of a diverse array of subsectors,
each with its own unique characteristics and challenges. Within
this complex landscape, the general focus was on three strategies:
electrification, green molecules and energy efficiency.
233
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix232
Electrification was tackled by evaluating the potential of
◆ heat pumps for low-temperature processes,
◆ e-boilers for mid and high-temperature processes.
Additionally, green molecules, in particular hydrogen boilers for
mid and high-temperature processes, were considered for com-
parison with electrified technology.
Energy efficiency was integrated as another lever. Drawing from
both EU targets and literature, emphasis was placed on further
improving energy efficiency to reduce overall emissions in the
industry sector. It is important to note that an interval was con-
structed for 2030, considering the high uncertainty of investment
costs.
The potential for each lever was derived from the BluePrint study
and meticulously analysed together with energy consumption
data from Belgian industrial processes. This methodology facili-
tated the assessment of each abatement lever within the intricate
operational landscape of industry. Deviating from this broader
scope to conduct a more granular, subsector-specific analysis
could potentially dilute the transparency of the BluePrint study's
overarching message.
Carbon capture
Last, two techniques for carbon capture were scrutinised:
◆ Carbon Capture and Storage (CCS),
◆ Direct Air Capture (DAC).
It is important to acknowledge the uncertainties surrounding
these techniques from their technological feasibility to their
economic viability, as the deployment of carbon capture poses
multifaceted challenges, encompassing both business-oriented
hurdles as well as regulatory and public considerations.
Furthermore, the data considered focused on the total cost of the
system, encompassing both the costs associated with CCS tech-
nology and ex-post expenses (i.e. costs related to compression,
transportation and storage of CO2 after capture).
3 This representation differs somewhat from the merit-order MACC most commonly found in literature. This visualisation has the advantage that it categorises the different
levers by sector, hence, gives a quick overview of potential actions by sector.
GRAPHICAL REPRESENTATION
The methodologies employed in constructing the MACC for 2030
and 2040 aim to provide a robust framework for analysing various
decarbonisation options. Some abatement levers, nonetheless,
demonstrate a range to indicate that performing a MACC analysis
should not be considered as an exact science.
The MACC visualisation, hence, needs to be carefully interpreted.
If one tends to compare e.g. CCS with heat pumps, the significant
differences in initial investment and operational expenses should
be noted. CCS requires substantial upfront and operational costs,
while heat pumps generally incur a lower upfront investment.
Heat pumps also achieve less CO2 reduction compared to CCS.
This limitation highlights the complexity of comparing technol-
ogies in isolation.
The results of the analysis are visualised in the MACC
3
, presenting
the identified abatement levers together with their associated
costs. This visualisation serves as a valuable tool for stakeholders,
facilitating informed decision-making and strategic planning
by offering options to achieve significant emission reductions.
MARGINAL ABATEMENT COST FOR SELECTED DEMAND LEVERS FOR 2030 AND 2040 FIGURE E1
204 0
MAC (Marginal Abatement Cost) €/tCO
2
1.000
800
600
400
200
0
-200
-400
1.000
800
600
400
200
0
-200
-400
From
To
2030
MAC (Marginal Abatement Cost) €/tCO
2
TransportBuildings Industry Carbon
CO2
CO2
H2H2H2 H2
Oil-fired
boiler
Tertiary
Heat
Pump
Residential
Heat
Pump
Tertiary
Heat
Pump
Residential
Heat
Pump
Tertiary
Hydrogen
Boiler
Residential
Hydrogen
Boiler
Efficient
Residential
Appliances
Residential
Insulation
BEV FCEV BEV FCEV Heat
Pumps
E-Boiler Hydrogen
boiler
Carbon
Capture &
Storage
Direct Air
Capture
Natural gas
boiler Current Energy
Labels Light-Duty
ICE Heavy-Duty
ICE Natural Gas
Heating
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030 & 2040
Excessive
Mac projection
for 2030
Carbon Capture
Techniques
Oil
Cost range based on high-low cost estimates. Appliances: CAPEX difference; Insulation: Façade (high) and roof (low)
Electricity Natural gas Hydrogen Efficiency Carbon capture
CO2
INSIGHTS IN THE MACC FOR 2030 AND 2040
Examining the MACC for 2030 and 2040 reveals significant
shifts in required abatement levers over time. In the short term
(2030), energy efficiency measures emerge as highly cost-effec-
tive options. Notably, efficient appliances and industry-focused
energy efficiency initiatives demonstrate significant potential for
emission abatement. Additionally, the transition to electrification
shows promise in achieving considerable emission reductions
in the short and medium term, particularly in the transport and
residential sector.
Specific insights into individual abatement options shed further
light on their evolving economic viability and potential impact. As
the wholesale price of natural gas currently is relatively low, gas-
fired boilers continue to be an interesting option for residential
heating. Natural gas prices, however, are expected to rise more
sharply than wholesale electricity prices over time, leading to elec-
trified heat pumps becoming more attractive alternatives. The
transition from oil-fired boilers to heat pumps already presents
a viable option due to a.o. the lower coefficient of performance
(COP) of oil-fired boilers.
Furthermore, BEV, encompassing both light and heavy-duty vehi-
cles, demonstrate significant promise as abatement options.
Light-duty BEV are already proving to be economically attractive,
driven by their declining investment costs and increasing energy
efficiency. Meanwhile, advancements in battery technology and
charging infrastructure are bolstering the potential of heavy BEV,
positioning them as viable alternatives for emission reduction.
As we shift our focus towards 2040, notable changes in the cost
dynamics of abatement options become apparent. One signifi-
cant trend is the decreasing MAC of most levers. This reduction
can be attributed to anticipated increases in energy prices which
amplify the financial impact of energy loss, resulting in a lower
MAC. Investments in insulation, for instance, become more eco-
nomically attractive, offering a means to mitigate energy loss
and reduce overall expenses. The MAC of other options, however,
may increase as efficiency levels approach their pareto optimum,
resulting in diminishing financial attractiveness. This is the case
for efficient appliances. These trends underscore the importance
of adaptive strategies and ongoing innovation to navigate evolv-
ing market dynamics and maximise the effectiveness of emission
reduction efforts.
As regards the green molecule, it currently faces challenges, par-
ticularly due to its elevated price. While technological advance-
ments, market developments and infrastructure expansion are
anticipated to enhance the feasibility and adoption of this solu-
tion over time, its higher MAC compared to electrified options
in transport and buildings suggests that it is and will remain
more expensive.
235
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix234
Carbon capture techniques which aim to reduce carbon emis-
sions by capturing and storing carbon dioxide, encounter both
economic and technological hurdles, making them less attractive
solutions in the short term.
Last, the MACC highlights the relevance of carbon pricing mech-
anisms in guiding investment decisions and strategic planning.
By integrating carbon prices, stakeholders can prioritise cost-ef-
fective abatement measures that align with carbon pricing
mechanisms, ensuring efficient allocation of resources towards
emission reduction goals.
CONCLUSION
The collaboration between Elia and Sia Partners in constructing
the Belgian MACC highlights the importance of addressing decar-
bonisation challenges. By examining abatement options from
a demand-side perspective, the analysis provides an informed
ranking of the identified emission reduction options.
In this analysis, energy efficiency measures emerge as cost-effec-
tive options in the short term, alongside the promising trajectory
of electrification with battery electric vehicles and heat pumps.
Moreover, in this exercise, electrification of industrial heat pops up
as an interesting option compared to a switch to green molecules.
This can be attributed to a.o. the high price for green hydrogen
and its largely lacking infrastructure. Additionally, carbon capture
initiatives show relatively promising MAC values compared to
other abatement options, but significant hurdles persist in the
form of high investment costs and regulatory uncertainty.
The insights gained from the Belgian MACC 2030 and 2040 anal-
ysis, reinforced by a confirmatory literature review, may offer
valuable guidance for policymakers and industry leaders. This
robust foundation provides a framework for informing strategic
decisions and driving meaningful progress towards reaching cli-
mate goals and thus a more sustainable and resilient future for all.
APPENDIX F – TOTAL COST METHODOLOGY
This section was elaborated by Compass Lexecon to provide more
insights on the total cost methodology that was developed. The
presented methodology was used to calculate the total costs
for Europe and Belgium. For the Belgian results, the DSO costs
approach was refined in bilateral discussions with DSO’s and the
historical investment annuities were not taken into account (as
they do no influence the results there).
F.1. GENERAL INTRODUCTION
The Cost tool quantifies total energy system costs for the scenar-
ios developed in Elia’s ‘Belgian Electricity System BluePrint for
2035-2050’ future energy scenarios study. The tool covers multiple
energy carriers and end-uses over the 2024-2050 period for both
EU-27 plus UK, Norway and Switzerland and for Belgium.
Energy system costs are defined as the sum of :
◆ Capital expenditure (CAPEX) for both production and distri-
bution, and final consumption
◆ Operating expenditure (OPEX) excluding fuel costs for both
production and distribution, and final consumption
◆ Fuel costs for both production and final consumption
Fiscal costs, such as taxes, subsidies, levies, and redistributions
are excluded from the energy system costs calculation. A similar
approach is used for public decision-making in the EC report of
2024 [EUC-9] and IEA & NEA report of 2020 [IEA-1], which analyses
long-term impacts of energy policies and options for infrastruc-
ture development. The system cost calculation methodology
and detailed assumptions were presented and challenged in two
stakeholder workshops (13.11.2023 and 13.12.2023). At the end of
2023, the consultation process was opened for stakeholders to
provide detailed feedback on the cost methodology and tech-
no-economic assumptions, and written answers were given in
spring 2024.
F.2. STRUCTURE OF THE COST TOOL
The Cost tool models both the costs associated with the produc-
tion and distribution, and the final consumption of the energy
carriers, which are defined as electricity (power system) and mol-
ecules (hydrogen, ammonia, methane, and liquids).
On the production and distribution side the costs of production,
storage and network for each energy carrier are modelled in
details (see section 6.6.3) based on inputs from Elia’s models.
On the consumption side, end user costs modelled cover energy
carriers switching costs across three sectors: transport, buildings,
and industry - these sectors being the most energy intensive and
requiring material investments to meet climate targets [EUC-9].
Both energy production and consumption system costs are linked
through end users fuel costs which reflect the levelised energy
production system cost of electricity and molecules.
Figure F-1 illustrates the Cost tool calculation process.
In addition to modelling system costs, the Cost tool also quantifies
material needs for a selection of key minerals (aluminium, copper,
cobalt, graphite, lithium and nickel) needed in power systems.
Material needs are calculated by multiplying material intensity
per technology with projected production capacity, electrolyser
capacity, and EVs rollout.
237
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix236
FLOWCHART OF THE COST TOOL’S SYSTEM COST METHODOLOGY FIGURE F1
Data on
final energy
demand
Data on
electricity
production
Data on
molecules
production
capacity
End user fuel
prices
End user Fuel
costs
Unit cost
library
OPEX
OPEX
Lifetime and
WACC by
application
Elia input CL input
Direct cost
inputs
Total system costs
Cost
Annualised
CAPEX
Annualised
CAPEX
Distribution
costs
Balancing
costs
Transmission
costs Energy
production
fuel and
impost
costs
Levelised cost
Energy
production
system costs
Energy
consumption
system costs
INPUTS
The cost tool is fed by Elia’s models (KARI, Multi-E, …) that provide
necessary inputs such as future energy demand, capacities of
different technologies and import costs of different fuels among
other (see Table F-1 for details). These inputs are combined with
assumptions on unit costs and material intensity of different
technologies, which are gathered from wide variety of sources,
such as EC’s impact assessment [EUC-9], IEA reports ([IEA-1], [IEA-
2], [IEA-3]) and other studies. Sensitivities on key parameters can
be assessed to test results robustness.
ELIA’S MODELLING INPUTS INTO THE
COST TOOL TABLE F1
Elia’s input Description
Data on final
energy demand
Existing and simulated final consumption
of energy in TWh by end-use and energy
carrier, and assumptions on future trends,
such as efficiency improvements, shares of
road transport fuels, and shares of industrial
processes.
Data on molecules
production
Simulated production, storage, and
distribution of molecules in MWh.
Data on electricity
production
The existing and simulated capacity of power
production technologies in MW.
Direct cost inputs
Various simulated costs, including cost of
imports and some precalculated CAPEX and
OPEX costs (based on the consulted costs) in
the molecule and power sector.
OUTPUTS
All investment costs in the Cost tool are annualised over the
assumed technical lifetime of the asset. Annuities depend on
the user-defined Weighted Average Cost of Capital (WACC). The
annualisation follows a normative depreciation and actualisation
approach, where a constant WACC (with sensitivities) is assumed
invariant of the technology. This approach excludes considera-
tions of contractual structures or market interventions, e.g. price
guarantees, but supports the key objective of the study to com-
pare alternative energy mix scenarios from a system point of view
(in contrast to individual project’s/investors’ point of view). The
same approach is applied in several studies, such as RTE report
of 2022 [RTE-1] and IEA & NEA report of 2020 [IEA-1]. All costs are
expressed in real euros and have been adjusted for inflation to
2022 values.
239
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix238
F.3. ENERGY PRODUCTION SYSTEM COSTS OF DIFFERENT ENERGY
CARRIERS
ELECTRICITY
The power sector’s system costs consist of CAPEX, variable and
fixed operating costs, fuel, transmission, distribution, balanc-
ing and import costs for the whole power sector. Of these costs,
variable operating costs, fuel costs, import costs and grid costs
are directly provided by Elia. The other elements are calculated
in the Cost tool. A comprehensive benchmarking exercise and
stakeholder consultation was performed to identify and collect
data on the unit costs of different power generation technologies.
Figure F-2 illustrates the methodology applied to the power sec-
tor.
FLOWCHART OF THE POWER SECTOR COSTS FIGURE F2
Existing
capacity Existing
capacity CAPEX
unit costs CAPEX
OPES +
Fuel costs
FOM
unit costs
VOM + Fuel
costs
Grid costs
Import
costs
Simulated
generation
Balancing
unit cost Balancing
costs
Grid costs
Import
costs
WACC &
lifetime
EV and electrolyser
energy demand
EV and
electrolyser
energy
demand
Distribution
costs
Data on final
energy
demand
Direct cost
inputs
Data on
electricity
production
CL input
Data on final
energy
demand
Direct cost
inputs
Elia input
Future
capacity
For each scenario, Elia’s modelling framework provides infor-
mation on the existing electricity generation capacity in 2023
and on the total future capacities of the years 2036, 2040, and
2050 by technology type. The total installed capacities of the
years in between are linearly interpolated. For technologies with
a net increase in capacity, the existing already installed capacity
is assumed to wind down over the assumed technical lifetime
of each technology following a polynomial reduction function.
The yearly difference between the total installed capacity and
what remains at the time of the existing capacity is assumed to
be newly installed.
CAPEX for newbuilt capacity is computed at yearly granularity by
multiplying technology’s yearly increment in installed capacity
with their associated costs (annualised), and adding annualised
costs stemming from assets built in previous years. For existing
capacity, the CAPEX consists of the annuity of the remaining
capacity. The initial investment costs are assumed to be incurred
linearly at the start of each construction year during a technolo-
gy-dependent construction period and to be financed through
debt at an assumed interest rate. Finally, these investment costs
are annualised over the lifetime of each technology using the
WACC as the discount rate.
The fixed operating costs are calculated by multiplying each year’s
technology-specific total installed capacity with FOM unit costs.
Balancing costs are assumed to be driven by variable renewable
energy sources (vRES) and calculated as onshore wind and solar
annual generation multiplied by an assumed fixed balancing
cost per unit of energy produced. Net electrical import costs for
the EU as whole are assumed to be zero, while import costs for
Belgium are based on Elia’s modelling.
For distribution costs the unit costs related to vRES generation,
and the electrification of transport and buildings are estimated by
regressing the projected distribution network investments on the
projected consumption of electric vehicles, heat pumps (proxied
by household electricity demand net of appliances demand), and
the power generation of solar PV and onshore wind. The esti-
mated unit costs are in turn multiplied by the amount of electric-
ity demand (transport, buildings) and supply (vRES). Distribution
costs not related to vRES or electrification of transport and build-
ings are assumed to make up half of the total distribution costs.
The Cost tool includes three sensitivities (low, central, high) for
costs and lifetimes for each technology. These sensitivities adjust
the assumed unit costs, WACCs, costs of debt and technical life-
times of different technologies.
MOLECULES
Production, imports, storage, and distribution costs are modelled
for the following energy carriers: hydrogen, ammonia, methane
and liquids. For each molecule different production methods and
import modes are considered. This means that, for example, the
production of fossil, bio and synthetic oils are treated separately.
Model inputs only cover years 2036, 2040 and 2050. The years
in between are linearly interpolated, while the preceding years
are extrapolated.
Figure F-3 illustrates the molecules cost calculation methodology.
FLOWCHART OF THE POWER SECTOR COSTS FIGURE F3
Molecule
demand WACC &
lifetime
Capacity
factor Production
capacity CAPEX
CAPEX unit
costs
OPEX unit
costs OPEX
Import
costs Import
costs
Stored
molecules Capacity
factor Storage
capacity Storage
unit costs Storage
costs
Distribution
costs
Distributed
molecules Capacity
factor Distribution
unit costs
Distribution
capacity
Direct cost
inputs
Data on
molecules
production
Elia input CL input
Some costs, including total import costs, are directly provided
by Elia’s model. The import costs sum the costs of each import
source, and thus don’t consider the marginal price of imports.
Other costs are calculated in the Cost tool and typically involve
converting yearly molecule consumption (in MWh) to yearly
production/storage/transport capacities (in MW) by assuming a
capacity factor. OPEX costs are calculated by multiplying these
capacities by operating unit costs of each technology, while
CAPEX costs are calculated by multiplying the yearly increase in
capacities and by capacity unit costs and then annualizing this
investment cost with assumed lifetime and WACC.
241
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix240
F.4. ENERGY CONSUMPTION SYSTEM COST OF END-USE SECTORS
The three end-use sectors representing the largest shares of
energy demand have been included in the modelling [EUC-9].
For these three sectors, transport, industry and buildings, the
following energy related costs are modelled: annualised CAPEX
costs, yearly OPEX cost, and fuel costs associated with different
energy carriers.
Fuel costs are calculated using the prices that the final energy
consumers face. These prices include distribution costs and firm
markups but exclude taxes and subsidies. The starting values
of these end user prices are based on research on current con-
sumer fuel prices, while their projected trajectory is indexed to
the levelised cost outputs of the electricity and molecule models.
Figure F-4 presents a high level overview of the end user cost
methodology.
FLOWCHART OF END USER COSTS FIGURE F4
Final energy
demand
Historical fuel
prices
Historical
capacities /
usage
WACC &
lifetime
CAPEX
unit costs CAPEX
OPEX
OPEX unit
costs
Future capacities
/ usage
Fuel costs
End user
fuel prices
Sector specific
assumptions
Future levelised
costs
Elia input CL input Cost tool input
Data on
final energy
demand
TRANSPORT
For the transport sector, CAPEX and OPEX costs are only calcu-
lated for road transport which accounts for the largest cost com-
ponent of transport [EUC-9]. Road transport includes private cars,
busses, vans and trucks. Fuel costs are calculated for all modes of
transportation, including aviation and maritime transport.
Historical data, sourced from Eurostat’s and OECD’s transport
statistics, is used as a starting value for the size of road transport
fleet and for the total number of passenger and freight kilometres
travelled. Assumptions are made about the occupancy rates of
passenger cars and of the load factor of freight vehicles. Elia’s
inputs include assumptions about the future growth of passen-
ger and freight transport needs, which by keeping the average
yearly vehicle kilometres travelled constant, is used to model the
future size of the road transport fleet. Elia’s inputs also include
assumptions about the shares of vehicles by motor type. These
are used to further divide the total fleet into ICE, electric, gas and
hydrogen vehicles.
A part of the vehicle fleet is assumed to be renewed each year.
Additionally, an assumption is made about the number of electric
and hydrogen charging stations per new vehicle. The number
of new vehicles is then multiplied by the unit cost of new vehi-
cles and infrastructure to obtain CAPEX costs. These investment
costs are then annualised over assumed lifetime using WACC as
discount rate. OPEX costs are obtained by multiplying the total
number of kilometres travelled by unit costs. Finally, total fuel
costs for the transport sector are calculated by multiplying the
yearly total final demand of energy in transport with end user
fuel prices.
INDUSTRY
Steel, chemistry, and refineries were chosen as industries to be
modelled in detail, while the rest of industrial sector is treated
as an aggregate in the “others” category. Elia’s inputs are used
to estimate how much industrial processes switch their energy
carriers, for example estimating how much natural gas-powered
processes are being electrified. Additionally, a fixed amount of
industrial capacity is assumed to be replaced each year.
Elia’s inputs include the proportions of total energy demand
attributed to various industrial processes within a sector. For
instance, they include the share of the energy demand of scrap
processing within the steel industry. CAPEX costs are derived
by multiplying the total amount of capacity being switched or
replaced yearly by these process shares and by unit costs for each
process given by EC’s impact assessment. These CAPEX costs are
then annualised. Heating costs are treated similarly.
OPEX costs are calculated by multiplying the total capacity of
each process by OPEX unit costs and fuel costs by multiplying
the total demand of energy with end user fuel prices.
BUILDINGS
For residential buildings, Elia provides an estimate of the future
number of households. By assuming a trend for household sizes
and the average number of square feet for single resident, the
number of households is converted into the total amount of living
space. For services, Eurostat’s population projection is used to
model the total population of EU and Belgium, while an assump-
tion on the average service space per person is used to estimate
the total number of service space.
The average efficiency improvements of renovated or new con-
structions compared to existing buildings is provided by Elia. The
yearly rate of renovations is set so that the mix of existing, refur-
bished and new buildings match Elia’s model of total household
and service energy consumption. CAPEX costs are derived by
multiplying the number of residential and service space being
renovated or built by the respective renovation and construction
unit costs.
Heating costs are calculated by combining Elia’s data on heating
type shares and energy consumption with per MWh unit costs.
Fuel costs are calculated by multiplying the total demand of
energy with end user fuel prices.
F.5. MATERIAL NEEDS
The assessed material needs include the amount of aluminium,
copper, cobalt, graphite, lithium and nickel used by the power
sector, electrolysers and electric vehicles. These materials are
identified widely to be crucial for decarbonising the economy
and were chosen on the basis of being included in assessments
conducted by the EC in 2023 [EUC-8], RTE in 2022 [RTE-1], IEA in
2021 [IEA-2] and IRENA in 2022 [IRE-1].
Key sources for per unit material needs are studies elaborated by
RTE [RTE-1] and IEA ([IEA-2], [IEA-3]) which, among other sources
used, quantify how much of the selected materials are needed
for each MW of capacity for each power sector technology and
for every MWh of energy for electrolysers and EVs. These per unit
material needs are then multiplied by additional power and elec-
trolyser capacity and EV consumption to derive the total material
needs. The unit per material needs are based on current estimates
of future technologies, which might not fully reflect the potential
of future technological breakthroughs on material needs.
Figure F-5 demonstrates the material need calculations.
FLOWCHART OF MATERIAL NEEDS FIGURE F5
Increase in power
production capacity
New EVs
Increase in electrolyser
capacity Material
intensity data
Power sector
results
Molecule results
Transport results
Material needs
CL input Cost tool input
243
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix242
F.6. COMPARING THE RESULTS TO PREVIOUS STUDIES
The final system costs were benchmarked against results of sim-
ilar studies by EC [EUC-9] and Institute Rousseau [ROU-1]. Due
to scope and methodological differences, comparing pure head-
line results is not possible, therefore the benchmarking exercise
was limited to comparing investment needs. The benchmark-
ing exercise showed the magnitude of the Cost tool’s modelled
investment needs to be in line with these studies. See the meth-
odological and scope comparison of the Cost tool and the two
studies in Table F-2.
For material needs, comparisons were made against EC [EUC-9]
and RTE [RTE-1]. As these reports for the most part only report
material needs related to some specific use case or time frame,
relevant adjustments were made to the Cost tool to ensure com
-
parability. Additionally, RTE only reports the material needs for
France. When comparing against it, RTE figures have been scaled
up using France’s share of EU’s GDP. Comparisons show that
estimated material needs are in line with alternative studies.
METHODOLOGY AND SCOPE COMPARISON OF COST TOOL AND TWO REPORTS TABLE F2
Cost Tool Rousseau Institute’s Road to Net Zero EC’s Impact Assessment
General information
Purpose Quantify total system costs and material
needs
Main question is to quantify public
investment needs
Comprehensive study on the effects of
climate change. Includes a small section
on “Evolution of the energy system and
associated raw material needs”
Costs included CAPEX + OPEX + Fuel costs Investment needs Total energy system costs (includes
different approaches)
Geography EU & Belgium EU (extrapolated from the study of
7 member countries) EU
Time frame 2024-2050 2024-2050 2030-2050
Sectoral scopes
Transport
CAPEX and OPEX for road transport
only. Fuel costs for all transport modes.
Infrastructure CAPEX for EV and FCEV
recharging.
Whole transport system, including
public infrastructure
Expenditures on vehicles, rolling stock,
aircraft and vessels plus recharging and
refuelling
infrastructure
Industry
Steel, chemicals and refineries
separately. Other industries as an
aggregate.
Steel and aluminium, cement and
glass, olefins & aromatics, ammonia
and chlorine, sugar, paper-cardboard
production, methanol production
Unclear. Likely all industries.
Buildings Renovation and heating system costs
for residential and tertiary sector
Costs associated with public program to
renovate housing and tertiary building
stock, including costs of subsidised
loans and subsidised technical advice
Investment needs for residential and
services.
APPENDIX G – SCHEMATIC VIEW OF THE
MODEL
Figure G- 1 gives a schematic view of the model.
SCHEMATIC VIEW OF THE MODEL FIGURE G1
;
SUPPLY
DEMAND
DomesticImports
Airborne CO
2
CCS
Permanent
storage
Captured CO
2
Potential
CCS
Demand
shiftintg
Storage
Offshore
wind
Onshore
wind
Electricity
imports
PV
Nuclear
Hydro
potential
CCS
potential
CCS
potential
CCS
potential
CCS
potential
CCS
CHP
incinerator
E-boilers
Demand shifting
Demand shedding
Low-T HEAT
High-T HEAT
Heat pumps
H
2
turbine Methanation
Biomethane
CH4 storage
Aquifer, …
Biomethane
LNG
terminals Ammonia
terminals
CH
4
pipelines
CH
4
power
plant
potential
CCS
CHP
Direct
RES H
2
Industry
End uses
End uses Industry Industry Industry
Transport
Bioliquids
Fossil
liquids
Fertilizers
Other
end-uses
Hydrogen
to liquids
Liquids
terminals
Other
end-uses
Other
end-uses
Other
end-uses
H
2
pipelines
H2 storage
steel tank,
cavern,
permeable rock
electrolyser
DAC
CH
4
NH
3
Liquids
Elec
Coal
H2
SMR-H
2
cracking
Haber-Bosch
potential
CCS
potential
CCS
potential
CCS
E-boilers
Low-T Gas boilers
Gas boilers
Biomass
& waste
Steam
turbine
potential
CCS
Heat demand not
explicitly modelled
but final demand for
heat and associated
flexibility modelled.
245
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix244
APPENDIX H – NON-CO2 EMISSIONS
METHODOLOGY
Elia Transmission Belgium (hereafter Elia) asked Compass Lex-
econ to provide support on key questions regarding assumptions
used for CH4, N2O and F-gas emissions as well as emission sinks
from the LULUCF sector currently used by Elia in its Blueprint
study. This section briefly analyses for each type of emissions key
emission sources, implemented and planned measures to unlock
reductions, associated challenges and alternative scenarios based
on the EC scenarios and mapping possible evolutions at both
EU and BE levels.
As basis Elia chose to take as reference the S3 scenario from the
recent EC’s 2024 Impact assessment, which is the most optimistic
in terms of LULUCF emissions. In addition, Elia asked Compass
Lexecon to provide an alternative scenario to be performed as sen-
sitivity. The following text has been written by Compass Lexecon
for Elia and provides the background for the alternative scenario.
INTRODUCTION
The EC’s 2024 Impact Assessment report explores three scenar-
ios with increasing net emission reduction ambitions to reach
EU targets.
1. The S1 scenario, which mainly relies on the ‘Fit for 55’ energy
trends, with no specific mitigation measures for non-CO
2
emissions or for evolutions in the LULUCF sector.
2.
The S2 scenario, which builds on S1 and adds substantial
reductions in non-CO
2
emissions and significant increases
in carbon removals in the LULUCF sector.
3.
The S3 scenario, which builds on S2 and adds a fully developed
carbon management industry by 2040, with sizable carbon
removals and the deployment of novel technologies.
Elia is assessing whether to rely on the S3 scenario, which projects
a significant decrease in non-CO2 GHG emissions and a relevant
increase in LULUCF net emission removals in 2040 and 2050, as
the basis for its projections of non-CO
2
emissions and LULUCF
net removals.
However, achieving significant reductions in CH4 and N2O emis-
sions could be challenging mainly due to inertia in the agricul-
ture sector, which is the most important source of non-CO2 GHG
emissions. Persisting challenges, mainly associated with cattle
farming and fertiliser applications, have led to slow CH4 and N2O
emission reductions both in the EU and in Belgium. Implemented
and proposed policies have had limited impact in accelerating
the rate of reductions.
As the S3 scenario’s projections of CH
4
and N
2
O emissions could be
perceived as too ambitious, an alternative scenario SA combining
mitigation measures from the different EC scenarios is developed
to capture a more realistic projection of non-CO2 emissions.
4. The SA scenario assumes the S1 scenario’s projection for CH4
emissions in the agriculture sector and the S3 scenario’s pro-
jection for CH4 emissions in the other sectors.
5.
The SA scenario assumes the S2 scenario’s projection for N2O
emissions in the agriculture sector and the S3 scenario’s pro-
jection for N2O emissions in the other sectors.
FGASES
For F-gases emissions, significant reductions are expected due
to already existing and planned measures. F-gases emissions
have been decreasing at increasingly faster rates in recent years
in the EU and in Belgium due to regulations affecting their use
in heating and cooling and industrial applications. However, a
complete phase-out seems more plausible to be achieved by 2050
rather than 2040, due to some monitoring issues that would take
time to be completely resolved. The SA scenario assumes the S1
scenario’s projection for F-gases emissions in 2040 and the S3
scenario’s projection for F-gases emissions in 2050.
LULUCF SECTOR
In the LULUCF sector, the GHG removal potential remains uncer-
tain as the removals are mainly due to forest land, while activities
associated with croplands, grasslands, wetlands and settlements
have resulted in emissions.
Historical LULUCF net emission removals have been decreasing
ever faster, both in the EU and in Belgium. Main reasons are chal-
lenges associated with deforestation and forest degradation, as
well as urban expansion and agricultural activities for food and
energy. While LULUCF policies and regulations, both at EU and
national levels, aim at restoring the sector’s carbon sink potential,
there are long lag-times for measures to show results and lead
to a reversal of trends.
The challenges to increase the LULUCF net removals in the EU
and in Belgium are well captured in the EC’s S3 scenario – albeit
with some ambitious assumptions. The alternative scenario
SA therefore heavily relies on the S3 scenario. The SA scenario
assumes the S1 scenario’s projection for net removals from crop-
lands and the S3 scenario’s projection for net removals from the
other land uses.
The SA and the S3 scenarios are associated with high levels of
uncertainty regarding the evolution of the carbon sink potential
of the different land uses, especially forest land removals and
whether grasslands and croplands become sources of removals
rather than emissions.
PROJECTED EVOLUTION OF CH4, N2O AND FGASES EMISSIONS BY SECTOR FIGURE H1
700
600
500
400
300
200
100
0
21
18
15
12
9
6
3
0
2021
Emissions in EU-27
20212030 20302040 20402050 2050
S1 S2 S3 SA S1 S2 S3 SAS3 SA S3 SA
CH4 Agriculture CH4 Waste CH4 Energy and Transport CH4 Industry and other
N2O Agriculture N2O Waste N2O Energy and Transport N2O Industry and other
F-gases Industry and other
[MtCO
2-
eq]
Emissions in Belgium
Source: CL Analysis based on EC Impact Assessment report
PROJECTED EVOLUTION OF LULUCF EMISSIONS AND REMOVALS FIGURE H2
200
100
0
-100
-200
-300
-400
-500
2
1
0
-1
-2
-3
-4
2021 20212030 20302040 20402050 2050
S1 S2 S3 SA S1 S2 S3 SAS3 SA S3 SA
Settlements/ Other land Harvested wood products (HWP) Wetlands Cropland
Grassland Forest land Total net removals
[MtCO
2-
eq]
Emissions in EU-27 Emissions in Belgium
Source: CL Analysis based on EC Impact Assessment report
Overall, the SA scenario differs from the EC scenario, especially regarding non-CO2 emissions from the agricultural sector, with pro-
jections for non-CO2 emissions in other sectors and for net removals in the LULUCF sector being relatively similar.
SA NONCO2 EMISSIONS PROJECTIONS
FOR EU27 TABLE H1
[MtCO
2
-eq]
SECTOR GAS
S3 SA S3 SA EC
SCENARIO
RETAINED
IN SA
2040 2040 2050 2050
Agriculture CH4171 206 157 197 S1
N2O 90 118 82 102 S2
Waste CH451 51 29 29 S3
N2O 4 433S3
Energy and
transport
CH418 18 14 14 S3
N2O 6 644S3
Industry and
other
CH41111S3
N2O 3 333S3
F-gases 1 911S1
Total 345 416 294 354 —
SA LULUCF EMISSIONS AND REMOVALS
PROJECTIONS FOR EU27 TABLE H2
[MtCO
2
-eq]
SECTOR
S3 SA S3 SA EC
SCENARIO
RETAINED
IN SA
2040 2040 2050 2050
Forest land -252 -252 -290 -290 S3
Grassland -9 -9 -7 -7 S3
Cropland -43 -8 -22 -7 S1
Wetlands 21 21 21 21 S3
Harvested wood
products -46 -46 -46 -46 S3
Settlements/ Other
land 13 13 10 10 S3
Total net removals -317 -282 -333 -318 —
247
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix246
APPENDIX I – DETAILS ON ENERGY
DEMAND
Figure I-1 shows the share of electricity, hydrogen, methane
and other vector in different sectors (transport, industry, etc.)
assumed in the different scenarios (ELEC, DE, GA) in Belgium in
2050. Electrification plays a key role in different sectors and its
effect is more pronounced in the DE and ELEC scenarios.
DE
GA
ELEC
DE
GA
ELEC
DE
GA
ELEC
Electricity
100%
99%
91%
86%
85%
85%
68%
45%
25%
25%
62%
58%
73%
84%
59%
47%
65%
86%
35%
76%
65%
14%
24%
1%
9%
14%
15%
32%
38%
42%
55%
75%
75%
15%
41%
16%
35%
27%
53%
Hydrogen Methane Other
GA
DE
DE
ELEC
Cars
Trucks
Buildings
Food
Chemicals
(energetic)
Other
Industry
Steel
0%
0%
100%
GA
ELEC
DE
GA
ELEC
DE
GA
ELEC
MOST COMMONLY USED ABBREVIATIONS
• AC: Alternative Current
• AdeqFlex’23: Adequacy and Flexibility Study for Belgium over
the horizon 2024-34, published in June 2023.
• ANTARES: A New Tool for Adequacy Reporting of Electric Sys-
tems (simulator used in this study)
• CAPEX: Capital Expenditure
• CCGT: Combined Cycle Gas Turbine
• CC: Carbon Capture
• CCS: Carbon Capture and Storage
• CCUS: Carbon Capture, Utilisation and Storage
• CfD: Contract for Difference
• CH: Switzerland
• CRM: Capacity Remuneration Mechanism
• CWE: Central West Europe
• DAC: Direct Air Capture
• DC: Direct Current
• DE: Distributed Energy (scenario)
• DRES: Decentralised RES
• DSM: Demand Side Management
• DSO: Distribution System Operator
• DSR: Demand Side Response
• EC: European Commission
• EEZ: Exclusive Economic Zone
• ELEC: ‘Electricity’ scenario
• EU: European Union
• ENTSO-E: European Network of Transmission System Operators
for Electricity
• ENTSO-G: European Network of Transmission System Operators
for Gas
• EPR: European Pressurised Reactor
• ERAA: European Resource Adequacy Assessment (study from
ENTSO-E)
• ETB: Elia Transmission Belgium
• ETS: Emission Trading System
• EV: Electric Vehicle
• FCEV: Fuel cell electric vehicle
• GA: Global Ambition (scenario)
• GHG: Greenhouse Gases
• HDD: Heating Degree Days
• HFLEX/LFLEX: High/Low Flexibility scenario
• ICE: Internal Combustion Engine
• KARI: Elia zonal electricity model
• LDES: Long Duration Energy Storage
• LNG: Liquefied Natural Gas
• LULUCF: Land-Use, Land Use Change and Forestry
• HP: Heat Pump
• HPV: High PV scenario
• HPVC: High PV capped scenario
• HTLS: High Temperature Low Sag Conductors
• HVDC: High Voltage Direct Current
• MACC: Marginal Abatement Cost Curve
• NECP: National Energy and Climate Plan
• NIMBY: Not In My Backyard
• NO: Norway
• OCGT: Open Cycle Gas Turbine
• OPEX: Operational Expenditure
• P2X: Power to X
• PV: Photovoltaïcs
• RES: Renewable Energy Sources
• RTE: Réseau de Transport d'Electricité (French transmission
system operator)
• SMR: Small Modular Reactor (nuclear)
• SMR-H2: Steam Methane Reforming (hydrogen)
• SUFF: ‘Sufficiency’ scenario
• TSO: Transmission System Operator
• TYNDP: Ten Years Network Development Program (study from
ENTSO-E)
• UK: United Kingdom
• V1X: Electric Vehicles with unidirectional smart charging tech-
nology
• V1H: V1X charging optimised with a local signal
• V1M: V1X charging optimised with a market signal
• V2X: Electric Vehicles with bidirectional smart charging tech-
nology
• V2H: Vehicle-to-Home
• V2M: Vehicle-to-Market (equivalent to Vehicle-to-Grid, or V2G)
• WACC: Weighted Average Cost of Capital
249
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix248
REFERENCES
INSTITUTION CODE WEBSITE LINK
Association of Cities and Regions [ACR-1] https://www.acrplus.org/images/position-papers/2024_03_Sufficiency_Manifesto.pdf
Agora energiwende [AGO-1] https://www.agora-industry.org/news-events/costs-and-risk-of-importing-hydrogen-derivatives-by-ship
Antares Simulator [ANT-1] https://antares-simulator.org/
Austrian Grid Power [APG-1] https://www.apg.at/stromnetz/sichere-stromversorgung/
Aurora [AUR-1] https://auroraer.com/media/renewable-hydrogen-imports-could-compete-with-eu-production-by-2030
Bloomberg [BLO-1] https://www.bloomberg.com/news/articles/2024-05-07/edf-gets-regulatory-approval-to-start-long-delayed-
nuclear-power-plant-in-france
Plan National Energie Climat Belgique [BEL-1] https ://www.plannationalenergieclimat.be/fr/le-pnec-c-est-quoi?plan-actualise-2023
CEMBUREAU – The European Cement
Association [CEM-1] https://cembureau.eu/media/kuxd32gi/cembureau-2050-roadmap_final-version_web.pdf
CLEVER energy scenario [CLE-1] https://clever-energy-scenario.eu/
De Standaard [DST-1] Engie spreekt zich uit over de energietransitie: ‘Kernenergie is een engagement voor de lange termijn’
https://www.standaard.be/cnt/dmf20231116_97607252
De Tijd [DTI-1] https://www.tijd.be/ondernemen/milieu-energie/engie-staat-voor-tijdperk-waarin-kernenergie-slechts-bijrol-
speelt/10528792.html
De Tijd [DTI-2] https://www.tijd.be/ondernemen/milieu-energie/verborgen-factuur-van-nucleaire-verlenging-duikt-op-
electrabel-vraagt-half-miljard-euro-aan-regering/10548531.html
Iceland Review [ICE-1] https://www.icelandreview.com/economy/uk-energy-company-announces-planned-windmill-farm-in-icelandic-
waters/
e-CIGRE [CIG-1] https://www.e-cigre.org/publications/detail/775-global-electricity-network-feasibility-study.html
Elia [ELI-1] https://www.elia.be/en/news/press-releases/2023/06/20230629_pressrelease_adeqflex
Elia [ELI-2] https://www.elia.be/en/electricity-market-and-system/adequacy/capacity-remuneration-mechanism
Elia [ELI-3] https://www.elia.be/fr/infrastructure-et-projets/plans-investissements/plan-de-developpement-
federal-2024-2034
Elia [ELI-4] Electricity scenarios for Belgium towards 2050 by Elia Group
https://issuu.com/eliagroup/docs/20171114_elia_4584_adequacyscenario
Elia [ELI-5] https://www.elia.be/en/users-group/the-horizontal-electricity-system-think-thank
Elia [ELI-6] https://www.elia.be/en/news/press-releases/2022/11/20221110_elia-launches-its-first-academic-board-in-
collaboration-with-belgian-universities
Elia [ELI-7] https://www.elia.be/en/news/press-releases/2023/11/20231121_thepowerofflex
Elia [ELI-8] https://www.elia.be/en/infrastructure-and-projects/infrastructure-projects/tritonlink
Elia [ELI-9] https://issuu.com/eliagroup/docs/elia_group_x_orsted_-_making_hybrids_happen
Elia [ELI-10] https://www.elia.be/en/customers/connection/grid-hosting-capacity
Elia [ELI-11] https://www.elia.be/-/media/project/elia/elia-site/company/publication/studies-and-reports/investment-plans/
plan-dadaptation-for-the-wallon-region/20240131_plan-dadaptation-wallon-2024-2034-definitif.pdf
Elia [ELI-12] https://www.elia.be/-/media/project/elia/elia-site/infra-and-projects/investment-plans/bxll-capital/20220427_
planinvbxl_2022-2032_fr.pdf
Elia [ELI-13] https://www.elia.be/-/media/project/elia/elia-site/company/publication/studies-and-reports/investment-plans/
investement-plan-for-the-flemish-region/investeringsplan_vlaams-gewest_2022-2032.pdf
Elmed [ELM-1] https://elmedproject.com/#sezione-chi-siamo
Energies (peer-reviewed, open access
journal) [ENE-1] Complementarity assessment of South Greenland katabatic flows and West Europe wind regimes, Radu D et al.,
2019 https://www.sciencedirect.com/science/article/abs/pii/S0360544219304529
Energies (peer-reviewed, open access
journal) [ENE-1]
A Brief Climatology of Dunkelflaute Events over and Surrounding the North and Baltic Sea Areas, Li B, Basu S,
Watson SJ, Russchenberg HWJ, 2021
https://www.mdpi.com/1996-1073/14/20/6508
Energie-suffizienz [ENS-1] https://energysufficiency.de/en/policy-database-en/
ENTSO-S [ENT-1] https://2024.entsos-tyndp-scenarios.eu/
ENTSO-S [ENT-2] https://2024.entsos-tyndp-scenarios.eu/scenario-storylines-for-tyndp-2024/
ENTSO-E [ENT-3] https://tyndp.entsoe.eu/
ENTSO-E [ENT-4] https://www.entsoe.eu/outlooks/eraa/
ENTSO-E [ENT-5] https://eepublicdownloads.blob.core.windows.net/public-cdn-container/tyndp-documents/TYNDP2022/public/
system-needs-report.pdf
INSTITUTION CODE WEBSITE LINK
ENTSO-G [ENT-6]
Learnbook: hydrogen imports to the EU market, 2023
https://www.entsog.eu/sites/default/files/2023-12/European%20Clean%20Hydrogen%20Alliance%20TD%20
RT_Learnbook%20Hydrogen%20Imports%20to%20EU%20market_20231219.pdf
EuroAfrica Interconnector [EAI-1] https://www.euroafrica-interconnector.com/at-glance/project-timeline/
European Climate Foundation [ECF-1] https://europeanclimate.org/resources/europeans-support-new-wind-and-solar-projects-in-their-local-area/
European Commission [EUC-1] https://managenergy.ec.europa.eu/publications/eu-achieves-20-20-20-climate-targets-55-emissions-cut-2030_en
European Commission [EUC-2] https://climate.ec.europa.eu/eu-action/climate-strategies-targets/2030-climate-targets_en
European Commission [EUC-3] https://climate.ec.europa.eu/eu-action/climate-strategies-targets/2040-climate-target_en.
European Commission [EUC-4] https://climate.ec.europa.eu/eu-action/climate-strategies-targets/2050-long-term-strategy_en
European Commission [EUC-5] https://energy.ec.europa.eu/topics/renewable-energy/renewable-energy-directive-targets-and-rules/renewable-
energy-targets_en
European Commission [EUC-6]
Implementing the REPowerEU action plan: investment needs, hydrogen accelerator and achieving the
biomethane targets
eur-lex.europa.eu/legal-content/EN/TXT/PDF/ ?uri=CELEX :52022SC0230&from=EN
European Commission [EUC-7]
Belgium NECP FACTSHEET (europa.eu)
https://commission.europa.eu/document/download/73fd2378-8c49-4840-b0cb-22a5548ee541_
en?filename=Factsheet_Commissions_assessment_NECP_Belgium_2023.pdf
European Commission [EUC-8] https://single-market-economy.ec.europa.eu/sectors/raw-materials/areas-specific-interest/critical-raw-materials/
critical-raw-materials-act_en
European Commission [EUC-9] Impact Assessment Report, 2024
https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52024SC0063
European Commission [EUC-10] https://ses.jrc.ec.europa.eu/eirie/en/data-collection/e-highway-2050-europes-future-secure-and-sustainable-
electricity-infrastructure-e
European Commission [EUC-11] https://energy.ec.europa.eu/topics/renewable-energy/solar-energy_en
European Commission [EUC-12] Recommendation for 2040 emissions reduction target
https://ec.europa.eu/commission/presscorner/detail/en/ip_24_588
European Commission [EUC-13] https://energy.ec.europa.eu/topics/nuclear-energy/small-modular-reactors/small-modular-reactors-explained_en
European Commission [EUC-14] https://energy.ec.europa.eu/topics/energy-systems-integration/hydrogen_en
European Commission [EUC-15] https://ec.europa.eu/eurostat/statistics-explained/index.php?title=National_accounts_and_GDP
European Union [EUU-1] Energy prosumers in Europe – European Environment Agency
https://op.europa.eu/en/publication-detail/-/publication/496ccfd9-3d4c-11ed-9c68-01aa75ed71a1
European Hydrogen Observatory [EUH-1] https://observatory.clean-hydrogen.europa.eu/hydrogen-landscape/end-use
European Hydrogen Observatory [EUH-2] https://observatory.clean-hydrogen.europa.eu/hydrogen-landscape/production-trade-and-cost/hydrogen-
production
European Hydrogen Observatory [EUH-3] https://observatory.clean-hydrogen.europa.eu/hydrogen-landscape/end-use/hydrogen-demand
European Hydrogen Observatory [EUH-4] https://www.clean-hydrogen.europa.eu/media/publications/study-hydrogen-ports-and-industrial-coastal-areas-
reports_en
Eurostat [EUS-1]
Simplified energy balances
https ://ec.europa.eu/eurostat/databrowser/view/nrg_bal_s/default/table?lang=en&category=nrg.nrg_quant.
nrg_quanta.nrg_bal
Eurostat [EUS-2] Reporting instructions – Hydrogen – Annual data (V1.1 2023)
https://ec.europa.eu/eurostat/documents/38154/16135593/Hydrogen+-+Reporting+instructions.pdf/
EnergyVille [EVI-1] https://energyville.be/en/blogs/a-paths2050-shift-scenario-towards-a-climate-neutral-2050-for-belgium-what-
role-can-low-energy-demand-play/
EnergyVille [EVI-2] https://perspective2050.energyville.be/paths2050
EnergyVille [EVI-3] https://perspective2050.energyville.be/sites/energyoutlook/files/inline-files/Full-Fledged%20Report_1.pdf
Federal Planning Bureau [FPB-1] https://www.dc2024.be/index_nl.html
Federal Public Service - Economy (of
Belgium) [FPS-1] https://economie.fgov.be/nl/themas/energie/energie-cijfers/belgian-energy-data-overview
Federal Public Service - Health, Food
Chain Safety and Environment (of
Belgium)
[FPS-2] https://www.health.belgium.be/en/environment/seas-oceans-and-antarctica/north-sea-and-oceans/our-sea-
nutshell
Federal Public Service - Economy (of
Belgium) [FPS-3] https://economie.fgov.be/sites/default/files/Files/Energy/North-Sea-Offshore-Wind-Repowering-Eastern-Wind-
Zone.pdf
251
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix250
INSTITUTION CODE WEBSITE LINK
Federal Public Service - Foreign Affairs,
Foreign Trade and Development
Cooperation (of Belgium)
[FPS-4] https://diplomatie.belgium.be/en/policy/policy-areas/highlighted/ostend-north-sea-summit-lays-foundation-
worlds-largest-green-power-plant
Federal Public Service - Health, Food
Chain Safety and Environment (of
Belgium)
[FIT-5] Federal Public Service - Health, Food Chain Safety and Environment (of Belgium)
Fraunhofer [FRA-1] https://www.iee.fraunhofer.de/en/presse-infothek/press-media/2022/green-ammonia-for-climate-protection.html
Financial Times [FIT-1] Global glut turns solar panels into garden fencing option
https://www.ft.com/content/2ea6bf6d-04e9-453b-a35f-cd6431cfc7bf
Flemish Government [FLG-1] https://www.vlaanderen.be/zonnepanelen/verplichting-zonnepanelen-voor-gebouwen-met-hoge-
elektriciteitsafname
Florence School of Regulation [FSR-1] Five reflections on clean hydrogen’s contribution to European industrial decarbonisation from 2024 to 2030
https://fsr.eui.eu/publications/?handle=1814/76564
Fluvius [FLU-1] https://over.fluvius.be/sites/fluvius/files/2023-12/investment-plan-2024-2033.pdf
Ministère de la transition énergétique
(France) [FRG-1]
Sobriété énergétique
https://www.ecologie.gouv.fr/sites/default/files/121023_DP_Sobriete%20energetique_un%20an_apr%C3%A8s_
VF.pdf
Generation IV International Forum [GIF-1] https://www.gen-4.org/gif/jcms/c_216241/gif-annual-report-2022
Hydrogen Import Coalition [HIC-1] https://www.fluxys.com/en/projects/hydrogen-import
Hydrogen Council [H2C-1] https://hydrogencouncil.com/wp-content/uploads/2023/11/Global-Hydrogen-Flows-2023-Update.pdf
Hydrogen Insight [H2I-1] https://www.hydrogeninsight.com/production/green-hydrogen-made-in-germany-will-be-cheaper-than-
shipped-imports-in-2030-study/2-1-1477657
International Energy Agency and Nuclear
Energy Agency [IEA-1] https://www.iea.org/reports/projected-costs-of-generating-electricity-2020
International Energy Agency [IEA-2] https://www.iea.org/reports/the-role-of-critical-minerals-in-clean-energy-transitions
International Energy Agency [IEA-3] https://www.iea.org/reports/batteries-and-secure-energy-transitions
International Energy Agency [IEA-4] https://iea.blob.core.windows.net/assets/d2ee601d-6b1a-4cd2-a0e8-db02dc64332c/
SpecialReportonSolarPVGlobalSupplyChains.pdf
International Energy Agency [IEA-5] https://www.iea.org/reports/the-role-of-critical-minerals-in-clean-energy-transitions/mineral-requirements-for-
clean-energy-transitions
International Energy Agency [IEA-6] https://www.iea.org/reports/renewable-energy-market-update-june-2023/is-there-enough-global-wind-and-
solar-pv-manufacturing-to-meet-net-zero-targets-in-2030
International Atomic Energy Agency [IAE-1] Technology roadmap for small modular reactor deployment
https://www-pub.iaea.org/MTCD/publications/PDF/PUB1944_web.pdf
IET [IET-1] Methods and strategies for overvoltage prevention in low voltage distribution systems with PV
https://ietresearch.onlinelibrary.wiley.com/doi/full/10.1049/iet-rpg.2016.0277
Litgrid and RTE International [LIT-1] https://www.rte-international.com/rte-international-to-evaluate-the-lithuanian-power-system
L’Echo [LEC-1] https://www.lecho.be/economie-politique/belgique/wallonie/en-wallonie-l-eolien-devra-se-developper-avec-la-
participation-des-citoyens/10524901.html
IRENA [IRE-1] https://www.irena.org/Energy-Transition/Technology/Critical-materials
IRENA [IRE-2] https://www.irena.org/Publications/2023/Aug/Renewable-Power-Generation-Costs-in-2022
IRENA [IRE-3] https://www.irena.org/Publications/2023/Dec/North-Africa-policies-and-finance-for-renewable-energy
IRENA [IRE-4] https://www.irena.org/Energy-Transition/Technology/Hydrogen/Global-hydrogen-trade
Knack [KNA-1] https://trends.knack.be/nieuws/duurzaamheid/tractebel-smr-mogelijk-operationeel-tegen-2035/
KU Leuven [KUL-1]
Techno-Economic Analysis of HVAC, HVDC and OFAC Offshore Wind Power Connections, Hardy et al., 2019
https://kuleuven.limo.libis.be/discovery/search?query=any,contains,LIRIAS2862630&tab=LIRIAS&search_
scope=lirias_profile&vid=32KUL_KUL:Lirias&offset=0
KU Leuven [KUL-2] https://iiw.kuleuven.be/apps/agrivoltaics/projects.html
Mediterranean Transmission System
Operators [MED-1] https://med-tso.org/events/ess-wg-antares-expansion/
National Grid ESO [NGE-1] https://www.nationalgrideso.com/future-energy/future-energy-scenarios-fes
Nuclear Energy Agency [NEA-1] https://www.oecd-nea.org/jcms/pl_90816/the-nea-small-modular-reactor-dashboard-second-edition
North Sea Foundation [NSF-1] https://offshore-coalition.eu/publications/north-sea-foundation-publishes-an-analysis-of-potential-risks-to-north-
sea-nature-posed-by-offshore-wind-energy-transition
Nuclear Alliance meeting – Ministère de
la transition écologique (France) [NUC-1] https://www.ecologie.gouv.fr/sites/default/files/nuclear%20alliance%20statement_VEN.pdf
Nucleair Forum [NUF-1] https://www.nucleairforum.be/publicaties/opiniepeiling-2021-belg-genuanceerd-maar-positief-tegenover-
kernenergie-en-nucleaire
North Sea Summit [NSS-1] https://northseasummit23.be/en/belgium/
INSTITUTION CODE WEBSITE LINK
Open Service for Market European
eXpertise (OSMOSE) [OSM-1] https://www.osmose-h2020.eu/
Offshore TSO Collaboration [OTC-1] https://issuu.com/eliagroup/docs/expert_paper_ii_offshore_tso_collaboration
Our World in Data [OWD-1] https://ourworldindata.org/land-use-per-energy-source
Oxford Energy [OXF-1] https://www.oxfordenergy.org/publications/renewable-hydrogen-import-routes-into-the-eu/
Rabobank [RAB-1] https://www.rabobank.com/knowledge/d011354306-the-bottlenecks-challenging-growth-in-the-eu-offshore-
wind-supply-chain
Recharge News [REC-1] https://www.rechargenews.com/energy-transition/we-wont-dump-uk-for-germany-25bn-morocco-europe-
green-power-mega-project-denies-switch/2-1-1611204
Reuters [REU-1] Germany raises offshore wind capacity target to 20 GW by 2030, 2020
https://www.reuters.com/article/idUSS8N2CH01M/
Dr. Jan Rosenhow [ROS-1]
‘Efficient & Smart Electrification’ Dr. Jan Rosenow, Director of European Programmes at the Regulatory
Assistance Project (RAP) (World Energy Outlook Guest Speaker Series)
https://www.worldenergy.org.tr/wec-mondays-dr-jan-rosenow/
Institut Rousseau [ROU-1] Road 2 Net Zero EN - Institut Rousseau (institut-rousseau.fr), 2024
https://institut-rousseau.fr/road-2-net-zero-en/
Réseau de Transport d’Electricité
français [RTE-1] Futurs énergétiques 2050 : les chemins vers la neutralité carbone à horizon 2050 | RTE (2022)
https://www.rte-france.com/analyses-tendances-et-prospectives/bilan-previsionnel-2050-futurs-energetiques
Réseau de Transport d’Electricité
français [RTE-2] https://www.rte-france.com/en/analyses-trends-and-perspectives/projected-supply-estimates
Skeyes [SKY-1] https://press.skeyes.be/skeyes-and-belgian-defence-support-wind-energy-by-expanding-permitted-sites-for-
wind-turbines
Solar Power Europe [SPE-1] https://www.solarpowereurope.org/insights/outlooks/global-market-outlook-for-solar-power-2023-2027/detail
Synergrid [SYN-1] https://www.synergrid.be/fr/centre-de-documentation/statistiques-et-donnees/degres-jours
Universität Erlangen Nürnberg [UEN-1] The economics of global green ammonia trade – ‘Shipping Australian wind and sunshine to Germany’, Egerer, J.
et al., 2022
U Gent [UGE-1] Wat valt er te leren uit meer dan 30 jaar nucleaire controverse in België (en waarom trekt men zo weinig lessen)?
https://www.ethiekenmaatschappij.ugent.be/wp-content/uploads/2012/07/EM_91-2006-Laes-et-al.pdf
UK Government [UKG-1] https://environmentagency.blog.gov.uk/2023/03/28/permitting-a-new-nuclear-power-station-sizewell-c/
UK Government [UKG-2] https://www.ons.gov.uk/economy/grossdomesticproductgdp
United Nations Framework Convention
on Climate Change (UNFCC) [UNF-1] https://unfccc.int/process-and-meetings/the-paris-agreement
U.S. Geological Survey [USG-1] Mineral commodity summaries 2024
https://pubs.usgs.gov/periodicals/mcs2024/mcs2024.pdf
World Bank Group - Norway [WBG-1] GDP (current US$) - Norway
https://data.worldbank.org/indicator/NY.GDP.MKTP.CD?locations=NO
Wikipedia [WIK-1] List of nuclear power stations - Wikipedia
Wind Europe [WEU-1] https://windeurope.org/newsroom/press-releases/wind-energy-permitting-is-improving-but-governments-still-
have-work-to-do/
Wind Europe [WEU-2] https://windeurope.org/newsroom/news/floating-wind-is-making-great-strides/
World Nuclear Association [WNA-1] https://world-nuclear.org/information-library/country-profiles/countries-a-f/france#Stresscorrosion
4COffshore [4CO-1] https://www.4coffshore.com/
Xlinks [XLI-1] https://xlinks.co/xlinks-first-updates-guidance-on-construction-costs-and-strike-price-for-the-morocco-uk-
power-project/
Xlinks [XLI-2] https://xlinks.co/public-consultation-announced-on-updated-proposals-for-xlinks-morocco-uk-power-project/
Xlinks [XLI-3] https://xlinks.co/morocco-uk-power-project/
253
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix252
FOM FIXED O&M COSTS
Costs in EUR2022/kW per year except for HVDC where costs are expressed in EUR2022/
(MW*km*year)
Low Ref High
2030 2040 2050 2030 2040 2050 2030 2040 2050
Existing thermal
CCGT-extension 25 25 25 30 30 30 35 35 35
OCGT-extension 20 20 20 23 23 23 27 25 25
CCGT-CCS-refurb 47 47 47 57 57 57 63 63 63
OCGT-CCS-refurb 42 42 42 50 50 50 55 55 55
CCGT-hydrogen refurb 30 30 30 35 35 35 40 40 40
OCGT-hydrogen refurb 25 25 25 28 28 28 32 32 32
Nuclear-extension-10Y 120 120 120 150 150 150 170 170 170
New storage Battery-2h 10 8 5 16 14 11 22 20 17
Battery-4h 18 16 15 24 22 20 35 30 25
New DRES
Wind onshore 17 12 10 31 26 17 44 40 25
PV-residential 11 10 9 18 13 11 37 32 26
PV-utility 8 6 6 13 11 10 18 17 14
New offshore Wind offshore fixed 37 35 32 60 50 40 75 69 63
Wind offshore floating 37 35 32 70 60 50 85 79 73
New thermal
CCGT-methane 25 25 25 30 30 30 35 35 35
OCGT-methane 20 20 20 23 23 23 27 25 25
CCGT-hydrogen 30 30 30 35 35 35 40 40 40
OCGT-hydrogen 25 25 25 28 28 28 32 32 32
CCGT-CCS 47 47 47 57 57 57 63 63 63
OCGT-CCS 42 42 42 50 50 50 55 55 55
New nuclear Nuclear-EPR 120 110 100 150 140 140 160 160 160
Nuclear-SMR 80 80 80 140 130 120 150 150 150
Electrolysis Elecrolyser-onshore 10 5 5 15 13 10 22 18 15
Electrolyser-offshore 15 10 10 20 18 15 27 23 20
Converters
(AC/DC)
Onshore 1.5% of CAPEX
Offshore 1.5% of CAPEX
HVDC Onshore 2.5% of CAPEX
Offshore 2.5% of CAPEX
MARGINAL COST OF ELECTRICITY GENERATION
Eur/MWh on average over simulation, incl CO
2
cost excl CCS benefits
2036 2040 2050
low high low high low high
Existing thermal
CCGT-extension 90 - 130 150 - 225 95 - 145 175 - 260 145 - 220 175 - 260
OCGT-extension 135 - 160 225 - 270 150 - 175 255 - 305 220 - 260 255 - 300
CCGT-CCS-refurb 120 195 130 220 190 220
OCGT-CCS-refurb 155 255 170 285 250 285
CCGT-hydrogen refurb 170 270 190 220 200 240
OCGT-hydrogen refurb 230 365 255 295 270 325
Nuclear-extension-10Y 30 30 30 30 30 30
New thermal
CCGT-methane 90 150 95 175 145 175
OCGT-methane 135 225 150 255 220 255
CCGT-hydrogen 170 270 190 220 200 240
OCGT-hydrogen 230 365 255 295 270 325
CCGT-CCS 120 195 130 220 190 220
OCGT-CCS 155 255 165 285 245 285
New nuclear Nuclear-EPR 30 30 30 30 30 30
Nuclear-SMR 30 30 30 30 30 30
Important note: Ranges are provided for some technologies as the marginal cost depends on the associated marginal fuel costs and CO
2
which
are optimized by the model
255
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix254
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be Photo by
Belgian electricity
s
ystem Blueprint for
2035-2050
Reasoned opinion by VITO/EnergyVille
AUTHORS:
DATE:
Pieter Vingerhoets September 2024
Wouter Nijs
TECHNICAL
LIFETIME CONSTRUCTION
TIME WACC
years years annual rate [%]
Ref Ref Low Ref High
2030 - 2040
- 2050
2030 - 2040 -
2050
2030 - 2040
- 2050
2030 - 2040
- 2050
2030 - 2040
- 2050 Comments
Existing
thermal
CCGT-extension 15 0 4 7 10
OCGT-extension 15 0 4 7 10
CCGT-CCS-refurb 15 1 4 7 10
OCGT-CCS-refurb 15 1 4 7 10
CCGT-hydrogen refurb 15 1 4 7 10
OCGT-hydrogen refurb 15 1 4 7 10
Nuclear-extension-10Y 10 3 4 7 10
New storage Battery-2h 15 1 4 7 10
Battery-4h 15 1 4 7 10
New DRES
Wind onshore 30 1 4 7 10
PV-residential 30 1 4 7 10
PV-utility 30 1 4 7 10
New offshore Wind offshore fixed 30 3 4 7 10
Wind offshore floating 30 3 4 7 10
New thermal
CCGT-methane 25 3 4 7 10
OCGT-methane 25 3 4 7 10
CCGT-hydrogen 25 3 4 7 10
OCGT-hydrogen 25 3 4 7 10
CCGT-CCS 25 3 4 7 10
OCGT-CCS 25 3 4 7 10
New nuclear
Nuclear-EPR 60 9 4 7 10 For first of a
kind a 10 year
construction
time is taken
Nuclear-SMR 60 7 4 7 10
Electrolysis Elecrolyser-onshore 20 2 4 7 10
Electrolyser-offshore 20 2 4 7 10
Converters
(AC/DC)
Onshore 40 4 - 6 -
Offshore 40 4 - 6 -
HVDC Onshore 40 4 - 6 -
Offshore 40 4 - 6 -
257
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix256
2
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Table of Contents
Table of Contents .......................................................................................................................... 2
Reasoned opinion ......................................................................................................................... 4
Annex 1 Overview of the methodologies Used in Elia's Blueprint Study ................................. 6
Annex 2 Evaluation of methods, results and data and suggestions for future work ............... 8
Annex 3 Renewable potential and deployment ......................................................................... 11
Annex 4 Results in the broader context of energy transition research .................................. 12
Annex 5 Sources used in Elia's Blueprint study....................................................................... 16
3
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Elia utilises highly
sophisticated tools that
enable the modelling of
various scenarios and
interactions within the
European energy system.
This Blueprint, along
with other modelling
work, clearly
demonstrates that a
significant expansion of
the electricity grid is
required, extending
beyond current plans.
The results of Elia’s
Blueprint were and will
be of great use to our
PATHS coalition.
Like EnergyVille’s
PATHS2050 exercises,
the Blueprint shows that
the energy transition will
significantly reduce our
dependency on foreign
energy. We will not
necessarily produce all
the electricity we need in
Belgium.
Wouter Nijs
Senior expert
Pieter Vingerhoets
Senior expert
Pieter Lodewijks
Programme manager
Gerrit Jan Schaeffer
General manager
“
”
“
“
”
”
“
”
259
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix258
4
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Reasoned opinion
Introduction
The purpose of this reasoned opinion is to provide an evaluation of the "Belgian electricity system Blueprint for 2035-2050"
conducted by Elia Transmission Belgium. We conduct a critical analysis of the study by thoroughly evaluating its assumptions,
verifying the credibility of sources, assessing the soundness of the methodology, and ensuring the coherence of arguments. We
also check for logical consistency in the conclusions, all within the broader context of energy transition research. The focus of this
analysis is on the energy system modelling and not on the grid infrastructure modelling, nor on the materials analysis.
This reasoned opinion is authored by EnergyVille, a collaborative research partnership between KU Leuven, VITO, Imec, and
UHasselt. As a neutral partner with no vested interests in the policy outcomes, EnergyVille's analysis is grounded in objective
scientific principles, ensuring that the evaluation is both unbiased and thorough. The goal is to support policymakers and
stakeholders by providing a clear, evidence-based assessment of the study's quantitative methodologies and results. EnergyVille
cross-checked the main mechanisms of the modelling suite, but was not granted access to the models, so a detailed validation of
the model was not conducted.
Context of the Blueprint study
The "Belgian Electricity System Blueprint for 2035-2050," conducted by Elia Transmission Belgium, offers a detailed analysis of
potential pathways for Belgium's energy future. The study aims to guide policymakers, industry stakeholders, and the public in
achieving a low-carbon energy mix by 2050. Elia created the Blueprint to provide a comprehensive, data-driven framework for
navigating Belgium's energy transition, focusing on various scenarios to meet future energy demands while reducing greenhouse
gas emissions.
As Belgium's Transmission System Operator (TSO), Elia focuses on ensuring that the grid infrastructure can support new energy
sources and future energy needs. The study provides a long-term vision for Belgium’s electricity system, covering the period up
to 2050. This is critical given the long lead times required for infrastructure development, technological adoption, and regu latory
changes. By balancing economic and environmental goals, the study offers valuable insights into cost-effective and sustainable
strategies for Belgium's energy system, helping the country achieve its long-term climate targets.
Analysis of scenarios
The study explores a wide range of scenarios, including varying levels of electrification, reliance on domestic versus imported
energy, and the role of new technologies like nuclear power and offshore wind. The inclusion of over 300 sensitivities for Belgium’s
electricity demand and supply highlights the study's thoroughness. It effectively addresses the uncertainties inherent in long-term
energy planning. The key insights, such as the necessity of defining Belgium's future energy mix to avoid costly scenarios, are
well-supported by the data presented.
Evaluation of methodology
While the analysis does not model demand options endogenously, the energy modelling in the study includes several beyond
state-of-the-art features. Key advancements include multi-energy integration, allowing for optimized interactions between
electricity, hydrogen, methane, and liquids; flow-based zonal modelling, which enhances the accuracy of grid constraint
simulations across Europe; and an iterative optimization process that consistently refines electricity dispatch alongside molecule
and carbon capture models. Additionally, the comprehensive carbon management approach, which optimizes carbon abatement
investments, and detailed offshore grid modelling, focusing on hybrid interconnectors and energy islands, are particularly crucial
for future European energy landscapes.
The study utilizes a multi-energy model that includes electricity, hydrogen, methane, and other energy vectors. The quantitative
analysis is built around three core scenarios for energy demand—Global Ambition (GA), Distributed Energy (DE), and Increased
Electrification (ELEC)—which are further subdivided into various sub-scenarios and sensitivities. This approach allows for a
detailed examination of how different combinations of policy decisions and technological developments might influence Belgium's
energy landscape. The use of different scenarios and time horizons (2036, 2040, and 2050) adds depth to the analysis, allowing
for a nuanced understanding of potential outcomes.
The sources used in Elia’s Blueprint study are diverse, credible, and well-aligned with the broader landscape of energy research
and policy. By incorporating input from established industry bodies, consulting firms, and academic institutions, the study ensures
a robust foundation for its scenarios and conclusions. Among the most referenced organizations in the report are the European
Commission, ENTSOs, Federal Public Services from Belgium, International Energy Agency, IRENA, European Hydrogen
Observatory, EnergyVille, KULeuven, Xlinks and Eurostat.
5
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Economic implications
The study provides a clear comparison of the total system costs associated with different energy mixes. The study's quantitative
analysis reveals that the cost of different energy pathways varies significantly. For instance, the total electricity system cost per
megawatt-hour (€/MWh) for Belgium ranges from €110 to €145, depending on the scenario and the mix of technologies
implemented. The analysis suggests that while non-domestic offshore wind appears more cost-effective than new nuclear power,
the differences are small and the actual decision should consider various risks, including financial, technological, and regulatory
uncertainties. This balanced presentation of options is a strength of the study, as it avoids prescribing a one-size-fits-all solution
and instead provides a framework for informed decision-making.
Scenario results on electrification and hydrogen
The Blueprint study presents three scenarios: DE, GA, and ELEC. The DE and GA scenarios, based on TYNDP assumptions,
continue to assume the use of methane in buildings. In contrast, EnergyVille’s scenarios emphasize a greater adoption of heat
pumps, which are technically capable of meeting all heating needs. In the DE and GA scenarios, there remains a significant
reliance on methane for steel production, even up to 2050. The ELEC scenario, developed also with input from EnergyVille,
features strong electrification and aligns more closely with the PATHS2050 analyses. In EnergyVille’s scenarios, energy use in
cars, trucks, and buildings leans more toward technologies that utilize electricity due to their much higher efficiency. The Blueprint
study includes a range of scenarios and options, providing a comprehensive analysis of the impact of various electrification
pathways and enabling readers to draw their own informed conclusions.
Total hydrogen supply amounts to 1,900 to 2,700 TWh (including pipeline, ammonia and electrolysers). Electricity based hydrogen
generation is limited in Europe, around 500 TWh and often self-consumed. The study estimates 121-180 GW of electrolyser
capacity, which is moderate compared to other sources, primarily due to access to ammonia imports. While the study optimizes
a hydrogen grid, detailed analysis of it is not provided. A graph indicates substantial electrolyser capacity in the Netherlands and
Northern Germany, regions with high local hydrogen demand, abundant onshore renewable supply, and access to a robust
offshore wind network. In Belgium, however, the role of electrolysers remains limited.
Suggestions for future work
For future work, it's suggested to refine the balance between domestically produced and imported synfuels, evaluating the
economic rationale behind ammonia-based production versus direct synfuel imports. Incorporating scenarios with higher synfuel
imports could provide a comprehensive view of different supply chain impacts. Additionally, introducing a higher electrification
scenario could address uncertainties in biomethane availability. Other key sensitivities include the impact of ammonia terminal
capacity distribution, reduced natural gas use in steel production, and temporary CO2 storage.
Transparency
The study’s development involved significant stakeholder engagement, including consultations with academic experts, industry
partners, and regulatory bodies. This collaborative approach enhances the credibility of the findings and ensures that the study
reflects a wide range of perspectives. Overall, the choice of sources significantly contributes to the reliability and relevance of the
Blueprint study. Further transparency in data accessibility and a deeper summary of external findings could enhance the ability of
stakeholders to critically engage with the study's content. We understand and support that Elia is considering publishing additional
information on their website, within the limits of confidentiality. To improve transparency even more, the Elia Blueprint may provide
explanations in the future of CO2 price optimization and sequential optimization techniques. This additional information would
enhance transparency and collaboration.
Conclusion
Overall, the "Belgian Electricity System Blueprint for 2035-2050" is a comprehensive and well-reasoned study that provides
valuable insights for policymakers. Its strength lies in the detailed scenario analysis and the balanced consideration of economic,
technological, and infrastructural factors. As with any scenario analysis, it is crucial to remain mindful of the assumptions and
sensitivities of each scenario to ensure well-informed policy recommendations.
In summary, the study is a significant contribution to Belgium’s energy planning, offering a detailed roadmap that balances
ambition with practicality. It provides a strong foundation for informed decision-making, though continuous updates and
transparency will be crucial as the energy landscape evolves.
261
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix260
6
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Annex 1 Overview of the methodologies Used
in Elia's Blueprint Study
The "Belgian Electricity System Blueprint for 2035-2050" by Elia Transmission Belgium employs a range of sophisticated
methodologies to analyse Belgium's potential energy futures. These methodologies are central to the study's conclusions and
recommendations, making their evaluation critical to understanding the study's robustness and reliability.
Multi-energy system modelling
The study utilizes a multi-energy model that integrates electricity, hydrogen, methane, and other energy carriers. The model
operates with hourly granularity for electricity and daily granularity for other energy vectors, providing detailed insights into the
interactions between different parts of the energy system. The study’s cost-benefit analysis evaluates the economic implications
of different energy pathways, considering both capital expenditure (CAPEX) and operational expenditure (OPEX) for various
technologies. It also accounts for externalities such as greenhouse gas emissions.
The multi-energy modelling approach is highly comprehensive, reflecting the interconnected nature of modern energy systems.
This methodology allows the study to capture complex interactions between different energy carriers, providing a more holistic
analysis than electricity-only models. By incorporating both direct and indirect costs, the analysis provides a balanced view of the
economic trade-offs associated with different energy strategies. The consideration of externalities, such as the social cost of
carbon, adds depth to the analysis and aligns with best practices in sustainability assessments.
The Blueprint energy system model suite
The table provides a detailed overview of the Kari model used in the Elia's Blueprint, describing its purpose, scope, features,
inputs, outputs, limitations, and applications.
Category Description
Purpose The model suite includes a unit commitment model used to simulate and optimize the European energy
system, taking into account grid constraints, flow patterns, and interactions between different energy vectors.
Scope
The electricity part covers the entire European electricity system on a zonal basis, including over 100 onshore
zones and more than 400 potential offshore wind farms. It also evaluates over 25,000 potential transmission
candidates.
Key features
Flow-based constraints: The model incorporates reduced equivalent grid models to accurately capture
electricity flows and grid bottlenecks within and between countries.
Zonal and offshore focus: Detailed modelling of both onshore zones and offshore wind farms, considering
hybrid interconnectors and offshore grid developments.
Inputs
Electricity demand and supply profiles: Hourly data for electricity demand and renewable generation (wind,
solar) across Europe.
Grid infrastructure: Existing and potential grid reinforcements, including HVDC links and AC interconnectors.
Technology costs: CAPEX, OPEX, and efficiencies of various generation and storage technologies.
Outputs
Optimal dispatch and investments: The model provides optimized dispatch schedules and investment
decisions for generation and grid infrastructure.
Electricity prices and carbon costs:
The model estimates marginal electricity costs and carbon prices based
on the optimization outcomes.
Limitations The model suite relies on a zonal approximation, which may not capture all local grid constraints and
dynamics.
7
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Modelling approaches beyond State-of-the-art
Though not covering all trade-offs within the entire energy system, several aspects of the energy modelling can be considered
beyond state-of-the-art:
Multi-energy integration
One of the most advanced features is the multi-energy integration, where the modelling framework explicitly incorporates different
energy vectors such as electricity, hydrogen, methane, and liquids. This allows for a comprehensive simulation of interactions
between these energy systems, optimizing the overall energy mix for both cost and emissions.
Flow-based zonal modelling
Another cutting-edge aspect is the use of flow-based zonal modelling for the electricity system. This method captures grid
constraints and flow patterns with a high level of granularity across Europe, providing more accurate modelling of interconnections
and the physical realities of the grid, particularly in a continent-wide context.
Iterative optimization process
The iterative optimization process linking the electricity dispatch model with molecule and carbon capture models represents a
sophisticated approach. It ensures that electricity dispatch and the consumption of other energy vectors are consistently optimized,
reflecting real-world interactions and the dynamic nature of energy systems.
Comprehensive carbon management
The comprehensive carbon management approach is also notable. By integrating a carbon emissions model that enforces GHG
emission targets and derives an associated carbon price, the model optimizes investments in carbon abatement technologies
such as CCS, CCU, and Direct-Air Capture (DAC), showcasing a forward-thinking strategy for managing and reducing emissions
within the energy system.
Detailed Offshore grid modelling
Finally, the detailed modelling of offshore grid developments, including potential hybrid interconnectors and energy islands,
reflects a cutting-edge focus on optimizing offshore renewable energy sources and their connection to the grid. This is particularly
crucial for the future European energy landscape.
Together, these features demonstrate that the Elia Blueprint's energy modelling is at the forefront of current capabilities, pushing
the boundaries of how energy systems are analysed and optimized in terms of integration, accuracy, and comprehensiveness.
Scenario analysis
The study employs scenario analysis to explore various potential futures for Belgium's energy system. It examines three primary
scenarios: Global Ambition (GA), Distributed Energy (DE), and Increased Electrification (ELEC). Each scenario considers different
levels of electrification, energy demand, and the balance between domestic production and imports.
Conclusion
The methodologies used in Elia's "Belgian Electricity System Blueprint for 2035-2050" are generally robust and align with best
practices in energy system modelling. The study’s scenario analysis, multi-energy modelling, and cost-benefit analysis provide a
comprehensive and detailed exploration of Belgium’s energy future. However, the study could benefit from increased transparency
in its modelling assumptions. Overall, the methodologies provide a solid foundation for the study's conclusions and make it a
valuable resource for policymakers navigating Belgium's energy transition.
263
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix262
8
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Annex 2 Evaluation of methods, results and
data and suggestions for future work
Following tables outlines the key areas where the Elia Blueprint excels in clarity and where further explanation, or expansion could
enhance the overall comprehensiveness and utility of the report. These suggestions aim to improve transparency, reproducibility,
and the practical application of the Blueprint's findings and methodologies. The first table outlines the key methodological
approaches used in the Elia Blueprint, highlighting areas that are well-explained and those that require further clarification.
Recommendations for improving the methodological transparency and rigor are also provided.
SCENARIOS and
METHODOLOGY Details
Clear
• Multi-energy capacity expansion and dispatch model: Clearly presented steps, including
scenario definition and optimization.
• The analysis covers emissions from the international aviation and shipping sectors.
• The study assumes that feedstock (non-energy purposes) would fall under the Emissions
Trading System (ETS) mechanism, meaning CO2 prices would apply for the use of fossil
fuels. In the case of feedstock, this implies that the life cycle emissions of fossil-derived end
products are taken into account.
• Adequacy study: well-defined process for ensuring adequacy in the power system,
particularly the simulation of climate years.
Limited explanation
in the report
• Marginal abatement cost curve (MACC) methodology: general approach provided, but
sector-specific data collection and assumptions are less clear.
• Improved explanation regarding CO2 price (optimizing CO2 price to reach carbon neutrality)
• Sequential optimization process: concept is introduced but underlying mathematical or
computational techniques are not fully elaborated.
Suggestions for
future work
• Refine the explanation and balance between domestically produced and imported synfuel.
In the current approach, synfuels are produced in Europe using hydrogen from imported
ammonia, hydrogen pipelines, or electrolysis. This raises questions regarding the economic
rationale behind preferring ammonia-based synfuel production over the direct import of
synfuels. The model selects this approach based on the assumed spread between
ammonia and synfuel prices which may not be optimal. However, the cost spread between
ammonia and synfuel would be improved, the impact on the power system will likely be
minimal whether synfuels are produced domestically or imported.
• The model could also incorporate alternative scenarios where a significant share of
synfuels is imported directly rather than being produced domestically from imported
ammonia. This would allow for a more comprehensive evaluation of the impacts of different
supply chain configurations.
• Include a higher Electrification scenario due to the uncertainty surrounding the availability
of biomethane. Rationale: almost 1,100 TWh of biomethane is used in all scenarios. As the
availability of biomethane remains uncertain, exploring higher electrification scenarios will
provide a clearer understanding of how Belgium can meet its energy needs with lower
reliance on biomethane.
• Other sensitivities on some key assumptions:
• Impact of different capacity distribution of ammonia terminals, not in line with their
domestic demand
• Impact of lower natural gas use in steel production.
• Impact of CO2 release that is only stored temporarily in materials
9
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
The following table presents an overview of the results from the Elia Blueprint, indicating which aspects are clearly reported and
where additional detail or case studies would be beneficial. It includes suggestions for further analysis to deepen the understanding
of energy system interactions and costs.
RESULTS Details
Clear
• Energy system costs: clearly presented breakdown of total system costs, including CAPEX,
OPEX, and fuel costs.
• Supply and demand balances: clear visual representations provided for methane, hydrogen,
and other energy vectors.
• 4.2.4 Electricity flows. The report makes clear that the links between zones which are loaded
the most bidirectionally are not necessarily the ones which are loaded the most directionally. It
is made clear how large annual imports or exports (average directionally flows) are in each
Belgium (though in an unexpected location in the report, see “electricity mix dashboard”). In the
grid chapter (5.7.2), it is explained that a further onshore reinforcement between France and
Belgium is not selected, instead increasing capacity via offshore nodes in the Atlantic.
• Information is available from the multi-region model on the type of methane supplied in the long
term. Similar to electricity, methane is flowing freely and the exact source abroad cannot be
identified. A proxy for the fossil share of methane in Belgium is the reported EU mix..
Limited
explanation in
the report
• 5.1.4. Imports for Belgium. There is limited explanation on the source of electricity imported in
the different scenarios, though results imply that most of the electricity imports are non-domestic
offshore wind connected to Belgium. Regarding electricity imports, it is understandable that you
cannot simply identify the exact source and the technology abroad used to generate it. Without
additional runs, it is not possible to link additional production units responsible for the extra
imports into Belgium.
• Location of electrolysers. The report explains that the optimizer places electrolysers primarily in
close-to-shore zones with excess wind in Northern Europe and excess solar PV in Southern
Europe. It is clearly stated that local hydrogen demand plays a crucial role. There is however a
limited number of electrolysers in Southern Europe and it is unclear how results and costs for
additional electricity infrastructure would change in case of massive rollout of PV.
• Further elaborate on the role of flexibility within the energy system, including demand-side
response and storage options. Explain why residential batteries have a smaller share in storage
or demand response compared to the European results.
• Further elaborate on the costs of the methane system. The report explains that the starting grid
for the methane system is the existing methane grid in Europe and that no investments (or
decomissionings) are considered. Still, the maintenance of this large network has a cost and
this cost is not begin reported.
265
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix264
10
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
The last table summarizes the clarity and limitations of the data presented in the Elia Blueprint, focusing on energy demand,
production data, and cost inputs. It also provides suggestions for enhancing the transparency and detail of the data used in future
work.
DATA Details
Clear
• Energy demand data: well-documented existing and projected final consumption of energy
in TWh by end-use and energy carrier. Demands from the DE and GA scenarios are linked
to TYNDP24.
• Power production data: clearly outlined existing and projected capacity of power production
technologies in MW.
• Clear overview of lead times (construction times) for new nuclear reactors as well as
construction delays. Construction delays are however not accounted for in the modelling.
Limited explanation
in the report
• Flexibility volumes from electric vehicles. Limited explanation on the sources of the 4.3 GW
V1X (optimised charging) and 4 GW V2X (vehicle-to-grid) for the DE scenario and the
ELEC and SUFF sensitivities.
• Relative impact of sufficiency versus the DE scenario in 2050: some (smaller) deviations to
the original EnergyVille study; unclear explanation why values differ.
• The study could provide clearer documentation regarding the amount of carbon required to
produce synfuel or to use it in material production.
• Unclear if waste treatment is included in the CAPEX of new nuclear.
Suggestions for
future work
• Open data: consider making the raw data and modelling tools available to stakeholders
and the public for increased transparency, collaboration and reproducibility
11
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Annex 3 Renewable potential and deployment
Renewable potentials
Wind onshore and solar are in all scenarios based on the trajectories that were brought forward during the public consultation of
the TYNDP2024 scenarios (with adaptations for the PV trajectories reflecting recent growth trends), except in the PV+ scenario.
The PV+ scenario increases the PV capacity for 2050 on top of the RES+ scenario: PV +100 GW/year.
The assumptions for biomass capacity are fixed at approximately 5 GW, based on the expected capacity projected for 2030.
The offshore wind potential is identified via a detailed approach, starting with a database from 4C Offshore [4CO-1], and
considering both geographical constraints (bathymetry, shipping routes, environment zones, etc.) and latest identified offshore
zones in national plans. In the North Sea basin, the potential amounts to 360 GW.
Comparing wind and solar PV projections
In comparing the "Belgian Electricity System Blueprint for 2035-2050" with the European Commission's impact study on the
Climate Target 2040 and the REPowerEU plan, follwing differences emerge, particularly concerning the projected capacities for
wind and solar energy.
1. Wind energy projections:
In the DE scenario (Central RES), which includes the EU27 plus Norway, the UK, and Switzerland (EU+), the projected total wind
capacity for 2040 is 830 GW, consisting of 550 GW from onshore wind and 280 GW from offshore wind. By 2050, the total wind
capacity is expected to increase to 940 GW, with 620 GW from onshore wind and 318 GW from offshore wind. When focusing
solely on the EU, the total wind capacity is projected to be around 700 GW in 2040 and approximately 800 GW in 2050.
Due to data availability, the below statements refer to the combination of onshore and offshore wind capacities.
o 2040: The Blueprint study projects a total wind capacity of 700 GW by 2040. This is significantly lower than
the European Commission’s Climate Target 2040, which estimates around 900 GW of wind capacity by the
same year.
o 2050: Similarly, the Blueprint’s projection for wind energy by 2050 is 800 GW, whereas the EU “Climate Target
2040” study anticipates a much higher capacity of around 1200 GW. This indicates that the Blueprint study is
more conservative in its wind energy assumptions compared to the European Commission’s more ambitious
targets. The Blueprint economically optimizes the build-out of offshore wind in Europe, primarily focusing on
the local generation and treatment of hydrogen. Similarly, in the scenario runs for the European Commission,
the deployment of various technologies is driven by cost optimization. However, likely due to differences in
cost assumptions—particularly regarding electrolysers and synfuel imports—there is a significantly higher
reliance on electrolysis-based hydrogen production in the European Commission’s scenarios.
o The REPowerEU plan, part of the European Commission’s strategy to accelerate the transition to renewable
energy, also emphasizes a higher wind capacity compared to the Blueprint study.
2. Solar PV projections:
o The solar PV projections in the Blueprint study align closely with those in the European Commission’s Climate
Target 2040 study, though the Blueprint’s estimates for 2050 are slightly lower. This suggests that both studies
share similar expectations for the growth of solar energy, albeit with minor variations.
The "Belgian Electricity System Blueprint for 2035-2050" adopts a more conservative stance on wind energy capacity compared
to the European Commission’s Climate Target 2040 and REPowerEU studies.
267
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix266
12
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Annex 4 Results in the broader context of
energy transition research
In the next sections, EnergyVille authors compare energy results from the Blueprint study with other studies. Some results have
been benchmarked within the study. Chapter F.6. of the Blueprint study discusses a comparison of the final system costs and
material needs from the study with previous studies by the European Commission (EC) and Institute Rousseau, noting that while
direct comparisons are limited due to differences in scope and methodology, the results show that the investment needs and
material needs estimated by the Cost tool align with those from the other studies, with necessary adjustments made for scope
and country-specific data (e.g., scaling RTE’s material needs for France to an EU level).
Benchmark the final energy demand with EU 2040 Climate Target Impact Assessment
Demands from the Elia DE and GA scenarios are linked to TYNDP24. In Figure 1, the comparison is presented of the TYNDP DE
scenario with the European Commission's IA S31 scenario for the year 2050. There are differences in the predicted distribution
of energy sources between the two scenarios, especially in how they treat renewables, hydrogen, and synthetic fuels.
In terms of electricity, the TYNDP DE scenario projects a slightly higher consumption (3932 TWh) compared to the European
Commission IA S3 scenario, which estimates 3695 TWh. This suggests that TYNDP DE places a slightly greater emphasis on
electricity as a key energy carrier in the future, although the difference between the two projections is not substantial.
Figure 1: Final Energy Consumption for the EU of the TYNDP DE and EC IA S3 scenarios in 2050
Notes: Final Energy Consumption excludes energy branch, international shipping, ambient heat, non-energy and includes aviation.
Within TYNDP, additionally part of energy branch is included (as some are reported under industry). Hydrogen includes ammonia
production for TYNDP scenarios. DE scenario synfuels and biomethane are distributed under methane and liquid demands.
Renewables* only includes some biofuels and biomass in the DE scenario but also include RFNBO fuels in the EC IA scenario. Solid
fuels have been removed because very low.
The contribution of hydrogen and synthetic fuels is significantly higher in the European Commission IA S3 scenario, which
projects 1110 TWh, compared to 824 TWh in the TYNDP DE scenario. This indicates that the Commission anticipates a larger
role for these fuels in decarbonizing the energy system, especially in sectors that are harder to electrify.
The difference in methane is mostly due to the different categorisation. For liquids, TYNDP DE estimates 511 TWh, while the
European Commission IA S3 scenario predicts a lower consumption of 356 TWh. The difference could be partially explained by
1 “COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT REPORT Part 3 Accompanying the Document COMMUNICATION FROM THE
COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS
Securing Our Future Europe’s 2040 Climate Target and Path to Climate Neutrality by 2050 Building a Sustainable, Just and Prosperous Society,” 2024,
https://climate.ec.europa.eu/eu-action/climate-strategies-targets/2040-climate-target_en#documents.
13
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
the inclusion of Renewable Fuels of Non-Biological Origin (RFNBO) in the Commission's renewables category, which might
otherwise fall under liquids or gas in other scenarios.
For district heating, TYNDP DE projects 597 TWh, which is significantly higher than the 359 TWh predicted in the European
Commission IA S3 scenario. This suggests that TYNDP DE expects a more widespread adoption of district heating as a means
of delivering energy for heating and cooling purposes.
Finally, the final energy consumption (FEC) in TYNDP DE is higher, at 6817 TWh, compared to 6450 TWh in the European
Commission IA S3 scenario. The Commission's scenario envisions a lower overall final energy consumption, indicating a greater
emphasis on energy efficiency.
Evolution of the primary energy demand in the Blueprint’s scenarios
First, we compare the primary energy consumption of the European energy system in 2019 with the various scenarios Elia
scenarios (Elia DE, Elia GA, Elia ELEC and Elia High RES). The role of renewable energy sources expands significantly. Wind
energy grows to between 18-22%, a substantial increase compared to 2019. Solar PV also rises, reaching 13-20%. Hydropower
remains steady at around 5%. This substantial increase in renewables underscores their central role in replacing fossil fuels in
Europe’s future energy system. In addition to renewables, new low-carbon fuels such as imported hydrogen, synthetic fuels
(synfuels), and ammonia play a crucial role in the 2050 scenarios. These fuels contribute between 10-20% of the total energy
mix, providing key solutions for decarbonizing sectors that are more difficult to electrify. Nuclear energy remains an important
part of the energy mix, although its share decreases slightly to between 9-11%. Biomass and waste feedstock continue to play
a significant role, contributing 24-26%, which is more than a doubling compared to 2019.
Overall, the comparison shows that the share of fossil fuels is reduced to less than 10% by 2050 in all scenarios. This sharp
reduction is driven by several factors. First, electrification reduces the need for primary energy by replacing less efficient fossil
fuel-based processes, such as heating and internal combustion engines. Second, the increasing role of renewables, particularly
wind and solar PV.
Figure 2: Primary energy supply for Europe in 2019 and 2050, source: Elia Blueprint.
Notes: Data for Europe (incl. UK, NO, CH). Includes international shipping & aviation and non-energetic feedstock
demand. Historical values based on EUROSTAT.
10%
0%
37%
6% 7%
4%
10%
25%
2%
3%
12%
26% 26% 24%
25%
10%
10% 9% 11% 9%
18% 20%
17%
10%
3%
19% 18%
20%
22%
14%
13%
15%
20%
3% 5% 5% 5% 5%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2019
(Elia data)
Elia
DE
Elia
GA
Elia
ELEC
Elia
High RES
Hydro
Solar PV
Others (geo, ambient heat,…)
Wind
Imported H₂ & synfuels & ammonia
Nuclear
Biomass and waste feedstock
Gas
Oil
Coal
269
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix268
14
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Benchmark primary energy demand with EU energy system analyses
In Figure 3, a comparison is presented of the energy mixes in different scenarios. Following sources have been used in this
comparison: European Commission2, Béres et al3., TYNDP244 scenario results and Neumann et al5. Absolute numbers are difficult
to compare because the Blueprint study results include UK, Norway and Switzerland. For that reason, we use the share of the
different energy vectors.
In the hydro category, the difference between Elia scenarios (5%) and other scenarios is minimal. Most scenarios, including the
European Commission IA S3, TYNDP DE, TYNDP GA, Béres et al., and Neumann et al., report similar contributions, ranging from
3% to 5%, showing general agreement on the role of hydro energy.
For solar PV, the Elia ELEC scenario projects 15%, which is aligned with the European Commission IA S3 scenario. The Elia
High RES scenario (20%) is more in line with other scenarios like TYNDP DE (23%) and TYNDP GA (18%). The studies Béres
et al. (26%) and Neumann et al. (27%) projects even higher Solar PV contributions. In terms of wind energy, Elia’s wind energy
projection is moderate. The Elia scenarios projects 20-22%%, which is lower than the 30% in the European Commission IA S3,
40% in TYNDP DE, and 37% in TYNDP GA. Neumann et al. anticipates the highest wind energy share at 42%, while Béres et al.
projects a much lower 16% (but this is very scenario dependent).
Figure 3: Primary energy supply for Europe (left) and EU (right), comparison based the Elia ELEC scenario.
Notes: For simplicity, we added imported hydrogen, synfuels and ammonia into one category. Includes international
shipping & aviation and non-energetic feedstock demand. Where available (Elia, EC, Béres et al.), data on biomass
feedstock have been used instead of bio energy products. Béres et al. and Neumann et al. include ambient heat used by
heat pumps in the gross available energy (as Eurostat does).
2 “COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT REPORT Part 3 Accompanying the Document COMMUNICATION FROM THE
COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS
Securing Our Future Europe’s 2040 Climate Target and Path to Climate Neutrality by 2050 Building a Sustainable, Just and Prosperous Society.”
3 Rebeka Béres et al., “Will Hydrogen and Synthetic Fuels Energize Our Future? Their Role in Europe’s Climate-Neutral Energy System and Power System
Dynamics,” Applied Energy 375 (December 1, 2024): 124053, https://doi.org/10.1016/j.apenergy.2024.124053.
4 ENTSO-E and ENTSOG, “TYNDP 2024 Scenario Report,” TYNDP 2024, May 2024, https://2024.entsos-tyndp-scenarios.eu
5 Fabian Neumann et al., “The Potential Role of a Hydrogen Network in Europe,” Joule 7, no. 8 (August 2023): 1793–1817,
https://doi.org/10.1016/j.joule.2023.06.016.
4%
10% 10%
4% 5% 5% 3%
3%
4%
24%
25%
20%
18% 20%
15%
12%
11% 9% 14%
3%
5% 13%
17% 10%
6%
10%
15%
20%
22% 30%
40%
37% 16%
42%
3%
5%
13%
15% 20% 15%
23% 18%
26%
27%
5%
5%
4%
3% 3% 3%
3%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Elia
ELEC
Elia
High RES
European
Commission
IA S3
TYNDP
DE
TYNDP
GA
Béres et al.
JRC-EU-
TIMES
Neumann
et al.
PyPSA
Hydro
Solar PV
Others (geo, ambient heat,…)
Wind
Imported H₂ & synfuels & ammonia
Nuclear
Biomass and waste feedstock
Gas
Oil
Coal
15
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Biomass and waste feedstock is estimated at 24-25% in the Elia scenarios, making it the highest among the projections. Most
other scenarios, like the European Commission IA S3 (20%), TYNDP DE (18%), TYNDP GA (20%), and Béres et al. (15%),
project lower shares for biomass. Neumann et al., with only 12%, is the most conservative. This suggests Elia ELEC places a
stronger emphasis on biomass as part of the future energy mix.
The imported hydrogen, synfuels, and ammonia category shows significant differences. Elia ELEC projects 17%, while most
other scenarios, including the European Commission IA S3 and Neumann et al., do not include this energy source. TYNDP DE
and TYNDP GA project smaller shares of 6% and 10%, respectively, and Béres et al. projects 15%. This indicates that Elia ELEC
envisions a major role for hydrogen and synfuels in decarbonizing the energy system. The High RES scenario from Elia only
reaches 10% of imported hydrogen, synfuels, and ammonia
Nuclear energy is projected to make up 9-11% of the energy mix in the Elia scenarios, slightly lower than the 14% projection in
the European Commission IA S3 scenario. In contrast, TYNDP DE and TYNDP GA project much lower nuclear shares, at 3% and
5%, respectively, while Béres et al. anticipates 13%. For the TYNDP scenarios, the low share of nuclear is not related to the
capacity but to the participation in the electricity market. Neumann et al. projects no contribution from nuclear energy.
For gas, the Elia ELEC scenario projects 3%, which is close to the 4% projection by the European Commission IA S3 scenario.
The scenario High RES is in line with TYNDP DE, TYNDP GA and Neumann et al., which project no gas usage. Béres et al.
anticipates a very small 1%.
Oil in the Elia scenarios is expected to make up 4-10%. In most scenarios, a high share of oil is used for non-energy uses
(feedstock) to produce materials.
In summary, the Elia ELEC scenario emphasizes biomass and hydrogen/synfuels more than most other scenarios, while
maintaining a moderate stance on wind and solar PV. It anticipates a moderate role for nuclear energy and minimal use of
fossil fuels like gas and oil, aligning with general trends in other scenarios but with some key differences in emphasis on specific
energy sources.
Benchmark with PATHS2050 for Belgium
In all three scenarios (DE, GA, and ELEC), electricity demand shows a significant increase, ranging from +110% to +130%
compared to 2022. The Electrification Scenario in PATHS2050 (2022 version) has a similar shift towards electrification, leading
to a 2.3-fold increase in electricity use by 2050 compared to 2020. The biggest difference is methane imports and usage, that is
mostly consumed by industries and by ships.
271
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix
BELGIAN ELECTRICITY SYSTEM
BLUEPRINT FOR 2035-2050
Appendix270
16
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Annex 5 Sources used in Elia's Blueprint study
The "Belgian Electricity System Blueprint for 2035-2050" by Elia Transmission Belgium is a comprehensive study that relies on a
diverse array of sources to inform its assumptions, methodologies, and conclusions.
Diversity and Range of Sources
The Blueprint study incorporates a wide range of sources, including academic research, industry reports, regulatory documents,
and data from well-established energy organizations. Key sources cited include:
• European network transmission system operators for electricity and gas (ENTSO-E and ENTSO-G): The study
references the Ten-Year Network Development Plan (TYNDP), which provides demand scenarios (TYNDP 2024) that
are supported by stakeholders from each country and provides data for the European grid (TYNDP 2022).
• International Energy Agency (IEA): The study uses data and projections from the IEA, a globally recognized authority
on energy statistics and policies.
• Peer-reviewed papers and reports: References to peer-reviewed articles, such as those published in the Energies
journal, ensure that the study’s scientific and technical assumptions are grounded in validated research.
This diversity of sources strengthens the credibility of the study by integrating multiple perspectives and ensuring that the analysis
is informed by a broad spectrum of expertise.
Consistent with TYNDP 2024, but also going beyond
The alignment of the study’s scenarios with the TYNDP 2024 is particularly important. This ensures that the Blueprint’s scenarios
are consistent with broader European energy strategies, making the findings relevant not just for Belgium. While aligning to
TYNDP 2024, Elia’s study also differs from it:
• A scenario is created to allow more electrification in vehicles and heating. Only how demand is being served is changed,
not the actual demand level of the activity.
• Unlike the TYNDP demand scenarios, which combine all hydrogen demands into a single final demand category, this
study treats energy and non-energy hydrogen demands separately. This approach allows for the consideration of
alternative strategies, such as importing synthetic fuels or using bio/fossil energy vectors, which may be more
economical than producing or importing hydrogen directly.
Details on the Belgian energy demand in 2050
Specifically for Belgium, Appendix I outlines nicely the projected shares in final energy of electricity, hydrogen, methane, and other
energy carriers by sector (transport, industry, etc.) for 2050. Electrification (always on the left) plays a pivotal role across sectors.
The category “Other” includes among others district heat and liquid fuels. As such, Appendix I provides critical assumptions
underlying the modelling, offering essential context that readers need to consider to fully grasp the magnitude and implications of
the energy transition outlined in the report.
Figure 4: Details on the Belgian energy demand in 2050. Source: Elia Blueprint
17
EnergyVille • Thor Park 8310-8320, 3600 Genk - Belgium • info@energyville.be • energyville.be
Use of consulting firms and Expert review
Elia’s study also engaged reputable consulting firms such as Compass Lexecon and Sia Partners, which provided inputs on cost
assumptions and marginal abatement cost curves. The involvement of these firms is a positive aspect, as they bring specialized
knowledge in economic modelling and scenario analysis.
Transparency and data availability
The study includes detailed references and provides transparency in the sources of its data. For instance, it cites specific studies
and documents from regulatory bodies and academic publications, making it easier for stakeholders to verify the information and
methodology used. Chapter 3.3 covers financial assumptions including investment costs, cost of capital.
Conclusion
The sources used in Elia’s Blueprint study are diverse, credible, and well-aligned with the broader landscape of energy research
and policy. By incorporating input from established industry bodies, consulting firms, and academic institutions, the study ensures
a robust foundation for its scenarios and conclusions. However, further transparency in data accessibility and a deeper summary
of external findings could enhance the ability of stakeholders to critically engage with the study's content. Overall, the ch oice of
sources significantly contributes to the reliability and relevance of the Blueprint study.
Registered offices
Elia Transmission Belgium,
Boulevard de l’Empereur 20
1000 Brussels, Belgium
Contact
info@elia.be
Graphic design
www.chriscom.be
Editor
Frédéric Dunon
WE WOULD LIKE TO THANK EVERYONE
WHO CONTRIBUTED TO THIS REPORT
