Modeling the impact of variable renewable energy sources penetration on supply-demand balance: Analysis of France from 2021 to 2025

DEGREE PROJECT IN SUSTAINABLE ENERGY ENGINEERING, SECOND CYCLE, 30 CREDITS

STOCKHOLM, SWEDEN 2021

Modeling the impact of variable renewable

energy sources penetration on

supply-demand balance

Analysis of France from 2021 to 2025

Rafael DE LEON

Abstract

France is planning a strong development of solar photovoltaics (PV) and wind power in the medium term disrupting the power system. This Master Thesis analyzes the impacts of variable renewable energy production on the supply-demand balance from 2021 to 2025 in France. The model used relies on a dynamic programming method.

The analysis is based on the assessment of indicators such as price signals, margins, loss of load duration (LOLD), expected energy not served (EENS) and nuclear drop stop that characterize the supply-demand balance and the security of supply of the electricity system. Wind power and PV are two very different technologies. Their load factor is very sizeable as it characterize their seasonality, variability and predictability and has an impact on all medium-term indicators. Wind power and PV have low marginal costs and their production is seasonal and in anti-phase. With new installed capacity, their added production in the supply-demand balance will substitute first the imports from the interconnections until saturation and then nuclear and thermal power plant production. Prices decrease with the same seasonality as the production and need to be considered when establishing the nuclear planning for the years to come. In addition a re-optimization of hydro power is observed. In terms of security of supply, wind power is more efficient than PV when assessing the reduction of LOLD but both are far from the performance of combined cycle gas turbines (CCGT). Lastly, the lack of nuclear production opportunities increases considerably more with PV due to a very localised production during the day which coincides in summer with periods of low consumption.

Wind power and PV are two distinct technologies and should not be put in the same category when assessing their impact on the power system.

Keywords

Variable renewable energy sources, Supply-demand balance, Load factor, Production,

Prices signals, Loss of load duration (LOLD), Margin, Nuclear power flexibility,

Optimization.

Abstract

Frankrike planerar en stark utveckling av solceller (PV) och vindkraft på medellång sikt för att störa kraftsystemet. Detta examensarbete analyserar effekterna av varierande produktion av förnybar energi på balans mellan utbud och efterfrågan från 2021 till 2025 i Frankrike. Modellen som används bygger på en dynamisk programmeringsmetod.

Analysen baseras på bedömningen av indikatorer som prissignaler, marginaler, förlust av lasttid (LOLD), förväntad energi som inte serveras (EENS) och kärnkraftsfallstopp som kännetecknar efterfrågan och utbudssäkerheten för el systemet. Vindkraft och solceller är två mycket olika tekniker. Deras belastningsfaktor är mycket stor eftersom den kännetecknar deras säsongsvariation, variation och förutsägbarhet och påverkar alla medellångsiktiga indikatorer. Vindkraft och solceller har låga marginalkostnader och deras produktion är säsongsbetonad och i fas. Med ny installerad kapacitet kommer deras extra produktion i utbuds- och efterfrågan att ersätta importen från sammankopplingarna till mättnad och sedan produktion av kärnkraft och värmekraftverk. Priserna sjunker med samma säsong som produktionen och måste beaktas när kärnkraftsplaneringen fastställs för de kommande åren. Dessutom observeras en återoptimering av vattenkraften. När det gäller försörjningstrygghet är vindkraft effektivare än solceller vid bedömning av minskningen av LOLD men båda är långt ifrån prestanda för kombinerade cykelturbiner (CCGT). Slutligen ökar avsaknaden av kärnkraftsproduktionsmöjligheter betydligt mer med solceller på grund av en mycket lokal produktion under dagen som sammanfaller på sommaren med perioder med låg konsumtion. Vindkraft och solceller är två olika tekniker och bör inte placeras i samma kategori när man bedömer deras inverkan på kraftsystemet.

Nyckelord

Variabla förnybara energikällor, balans mellan utbud och efterfrågan, belastningsfaktor,

produktion, prissignaler, förlust av lasttid (LOLD), marginal, kärnkraftsflexibilitet,

optimering.

Acknowledgements

I would like to thank all who have contributed to the production of this Master Thesis.

I owe my gratitude to all those people who made this last six months’ experience possible.

First, I would like to thank all members of the Portfolio and Market Management Office for their very warm welcome. They all were wonderful colleagues and it was a sincere pleasure to work with all of them. A very special thank goes to Vincent Dutoya, who was my supervisor at EDF. Vincent followed my work from the beginning and this Master Thesis would not have been possible without him. My deep thanks also goes to Martin Lacheret for its assistance and patience. Its great knowledge in electricity markets and its advices were essential in the development of this Master Thesis. I am deeply grateful to Damien Loffredo, Léo Granseigne, Sébastien Rondard, Rebecca Chaix, Hoel Langouet, Baudouin Duhem, Daniel Marco, Nathanael Mozzani, Aurélia Esteve, Patrice Nanet and Dimitri Rzepski who welcomed me within the Portfolio Management Team at EDF.

I also would like to thank Anders Malmquist who was my supervisor at KTH for his time, his detailed revision of my work and his precious pieces of advice.

Finally, I am thankful to my family for having supported me all along this Master

Thesis.

Table of Content

List of figures viii

List of tables x

Nomenclature xi

1 Introduction 1

1.1 Problem . . . . 3

1.2 Master Thesis’ Purpose . . . . 3

1.3 Methodology . . . . 4

1.4 Thesis Framework . . . . 5

1.5 Presentation of EDF . . . . 6

1.5.1 DOAAT (Department) . . . . 6

1.5.2 Portfolio and Market Management Office (GPM) . . . . 7

2 Literature Review 7 2.1 Previous Research . . . . 7

2.1.1 100% Renewable Electricity Scenarios . . . . 7

2.1.2 Modeling Variable Renewable Energy Impacts on Supply-Demand Balance in the Medium-term . . . . 9

2.1.3 Power System Issues Relative to Variable Renewable Energy Integration . . . . 15

2.2 Background and Scope . . . . 16

2.2.1 World Outlook . . . . 16

2.2.2 European Outlook . . . . 18

2.2.2.1 Renewable Energy Deployment . . . . 18

2.2.3 French Power System Outlook . . . . 20

2.2.4 National Context . . . . 20

2.2.5 Demand . . . . 22

2.2.6 Power Generation . . . . 22

2.2.6.1 Nuclear Power . . . . 23

2.2.6.2 Thermal Power plants . . . . 24

2.2.6.3 Hydroelectricity . . . . 25

2.2.6.4 Onshore Wind Power . . . . 25

2.2.6.5 Offshore Wind Power . . . . 27

2.2.6.6 Photovoltaics . . . . 28

2.2.6.7 Bio-energy . . . . 29

2.2.6.8 Electricity mix targets . . . . 30

2.2.7 Grid . . . . 31

2.2.8 Capacity Market Mechanism . . . . 33

2.2.9 Covid-19 impacts on the energy sector . . . . 33

3 Modeling French Electricity Supply-Demand Balance with OPUS 35 3.1 Presentation . . . . 35

3.2 Market and exchanges . . . . 35

3.2.1 Interconnections . . . . 36

3.2.2 Market prices . . . . 36

3.3 Demand . . . . 37

3.4 Supply . . . . 37

3.4.1 Nuclear Power . . . . 37

3.4.2 Thermal Power plants . . . . 39

3.4.3 Hydroelectricity . . . . 39

3.4.4 Load shedding . . . . 40

3.4.5 Variable Renewable Energy . . . . 40

3.4.6 Balancing reserves . . . . 41

4 Variable Renewable Energy Impact Analysis On Supply-Demand Balance 43 4.1 Scope and analytical approach . . . . 43

4.1.1 Context . . . . 43

4.2 First case study : Impact on medium term indicators . . . . 44

4.2.1 Definitions . . . . 45

4.2.1.1 Load Factor . . . . 45

4.2.1.2 Marginal Valorisation (Vmar). . . . 45

4.2.1.3 Margin . . . . 46

4.2.1.4 Expected Energy Not Served (EENS) . . . . 48

4.2.1.5 Loss of Load Expectation (LOLE) . . . . 48

4.2.2 Structure and assumptions . . . . 50

4.2.3 Scenarios . . . . 50

4.2.4 Results and Analysis . . . . 52

4.2.4.1 Load factor . . . . 52

4.2.4.2 Production . . . . 54

4.2.4.3 Supply-demand balance . . . . 55

4.2.4.4 Marginal Valorisations . . . . 56

4.2.4.5 Energy Security . . . . 58

4.3 Second case study: Comparison of VRES with CCGT . . . . 60

4.3.1 Objectives . . . . 60

4.3.2 Assumptions . . . . 61

4.3.3 Scenarios . . . . 61

4.3.3.1 Comparison I . . . . 61

4.3.3.2 Comparison II . . . . 61

4.3.4 Results and Analysis . . . . 63

4.3.4.1 Margin . . . . 63

4.3.4.1.1 Comparison I . . . . 63

4.3.4.1.2 Comparison II . . . . 64

4.3.4.2 LOLE and EENS . . . . 65

4.3.4.2.1 Comparison I . . . . 65

4.3.4.2.2 Comparison II . . . . 66

4.4 Third case study: Impact of VRES on Nuclear power . . . . 67

4.4.1 Definitions . . . . 68

4.4.1.1 Technical Minimum . . . . 68

4.4.1.2 Lack of Production . . . . 69

4.4.2 Objectives and assumptions . . . . 69

4.4.3 Results and Analysis . . . . 70

4.4.3.1 Duration . . . . 70

4.4.3.2 Depth of power . . . . 71

4.4.3.3 Energy . . . . 71

5 Discussions 73 5.1 Sustainability analysis . . . . 73

5.2 Limits of the study . . . . 74

5.3 Future work . . . . 74

6 Conclusions 76

References 78

A Appendix 82

List of Figures

2.1 Main developments in the French generation fleet between 2019 and 2023 and impact on capacity margins [27] . . . . 10 2.2 Capacity margins in the base case of the Mid-term Adequacy Report [27] 11 2.3 Loss of load probability over simulated weeks and Daily load curves

during the cold spell in February 2012 . . . . 12 2.4 Net annual revenue (i.e. market revenue minus variable production costs)

for gas-fired combined cycle and combustion turbine plants from 2015 to 2019 and comparison with fixed cost assumptions [27] . . . . 13 2.5 Planned use of electrolysis to decarbonise the industrial uses of hydrogen

in the longer term [27] . . . . 15 2.6 Renewable energy capacity in the World (Others: Bio-energy,

Geothermal and Marine) [18] . . . . 16 2.7 Renewable energy capacity in Europe (Others: Bio-energy, Geothermal

and Marine) [18] . . . . 18 2.8 Main targets for decommissioning thermal facilities and Carbon

intensities from electricity generation in Europe. [27] . . . . 20 2.9 Capacity margins in the base case and in other scenarios of European

fleets [27] . . . . 20 2.10 Capacity installed and production by sector in France in 2019 [25] . . . . 22 2.11 Onshore wind power capacity : spatial distribution and evolution [30] 26 2.12 PV capacity : spatial distribution and evolution [30] . . . . 28 2.13 Renewable energy capacity in 2019 in France and targets for 2023-2028

[18] . . . . 30 2.14 Renewable energy new installed capacity and growth in France in 2019

(Others: Bio-energy, Geothermal and Marine) [18] . . . . 31 2.15 Transmission lines (red) and distribution lines (green) in France . . . . . 32 2.16 Renewable electricity capacity additions, 2007-2021. [14] . . . . 34 3.1 Provisional schedule for ten-year inspections of the nuclear fleet for the

period 2019-2022 [9] . . . . 38 3.2 Monthly average and quantiles of Onshore wind and PV load factor . . . 41 4.1 Historical and planned installed capacity for Wind power and PV

according to PPE’s targets . . . . 44

4.2 ”Marginal Valorisation” (Vmar) is the price level given when consumption

meets production . . . . 45

4.3 Margin represents the volume of power available subtracted by the consumption . . . . 47

4.4 Loss of Load appears when the market does not ensure the balance between supply and demand . . . . 49

4.5 High, Medium and Low installed capacity trends for onshore wind power and PV for 2020-2025 . . . . 51

4.6 Matrix of 3 x 3 = 9 scenarios [Onshore Wind Power; PV] . . . . 51

4.7 Reference Scenario : MM = [OnWP Medium ; PV Medium] . . . . 52

4.8 OnWP and PV trends for high scenarios . . . . 52

4.9 Monthly Load Factor for PV and Onshore wind power . . . . 53

4.10 Average monthly output power for OnWP and PV for high scenarios . . . 54

4.11 Yearly production balance between high development scenarios and the reference scenario from 2021 to 2025 . . . . 55

4.12 Share of production substituted in supply-demand balance by VRES . . . 56

4.13 Monthly spread of marginal valorisations beetween high scenarios and reference scenario . . . . 57

4.14 Seasonal spread of marginal valorisations beetween high scenarios and reference scenario by season . . . . 57

4.15 Average weekly France Margin deviation during peak hours . . . . 58

4.16 Comparison of LOLD between the high scenarios and the reference scenario. 60 4.17 Comparison I : Margin France deviation for morning and evening peaks for #2 (OnWP), #3 (PV) and #4 (CCGT) compared to #1 (Reference). . . 63

4.18 Comparison II : Margin France deviation for morning and evening peaks for #2 (OnWP), #3 (PV), #5 (CCGT/OnWP) and #6 (CCGT/PV) compared to #1 (Reference). . . . 64

4.19 Comparison I : LOLD and EENS for #1 (Reference), #2 (OnWP), #3 (PV) and #4 (CCGT). . . . . 65

4.20 Comparison II : LOLD and EENS for #1 (Reference), #2 (OnWP), #3 (PV), #5 (CCGT/OnWP) and #6 (CCGT/PV). . . . 66

4.21 Lack of Nuclear production opportunities . . . . 69

4.22 Daily average of the deviation of nuclear power with lack of opportunities between high scenarios and reference scenario. . . . 70

4.23 Duration with lack of production opportunities . . . . 71

4.24 Depth of power with lack of production opportunities . . . . 71

4.25 Energy with lack of production opportunities . . . . 72

A.1 Calendar of calls for tenders for renewable energies in France . . . . 82

A.2 Spatial distribution of the load factor of onshore wind and PV . . . . 82

A.3 Spatial distribution of the production of onshore wind and PV . . . . 82

A.4 Low, Medium and High onshore wind power trends compared to RTE’s forecasts . . . . 83

A.5 Low, Medium and High PV trends compared to RTE’s forecasts . . . . . 83

List of Tables 2.1 Comparison of the flexibility abilities of dispatchable power plants [2] . . 14

2.2 Targets set by MAEP adopted in 2018 for installed capacity of hydraulic electricity generation . . . . 25

2.3 Targets set by MAEP adopted in 2018 for installed capacity of onshore wind power generation . . . . 27

2.4 Targets set by MAEP adopted in 2018 for installed capacity of offshore wind power generation . . . . 28

2.5 Targets set by MAEP adopted in 2018 for installed capacity of solar photovoltaic power generation . . . . 30

4.1 Scenarios and characteristics . . . . 63

4.2 Scenario Comparisons . . . . 63

Nomenclature

Acronyms

ADEME Agency for the Environment and Energy Management BAU Business As Usual

CCGT Combined Cycle Gas Turbine CCS Carbon Capture and Storage CRE Energy Regulator Commission CWE Central West Europe

DOAAT Optimisation and Trading Department DPN Nuclear Production Division

DSO Distribution System Operator EDF Electrcité de France

EENS Expected Energy Not Served EPEX European Power Exchange EPR Evolutionary Power Reactor ETS Emissions Trading Scheme EV Electric Vehicule

GHG Greenhouse Gas

GPM Portfolio and Markets Management IEA International Energy Agency

IPCC Intergovernmental Panel on Climate Change IRENA International Renewable Energy Agency LF Load Factor

LOL Loss of Load

LOLD Loss of Load Duration LOLE Loss of Load Expectation LOLP Loss of Load Probability

LTECV Law on Energy Transition for Green Growth NTC Net Transfer Capacity

OnWP Onshore Wind Power

OPUS Optimisation and simulation tool developed by EDF PCN Net Continuous Power

PI Installed Power

PPE Multi-Annual Energy Plan PSH Pumped-Storage Hydro PV Photovoltaic power

RES Renewable Energy Sources

RTE Réseau de Transport d’électricité, French TSO SNBC National Low Carbon Strategy

TIME The Integrated MARKAL-EFOM1 System TM Technical Minimum

TSO Transmission System Operator

TYNDP Ten-Year Network Development Plan Vmar Marginal Valorisation

VRES Variable Renewable Energy Sources Variables

δL

Technical threats at time period i which could force a higher output than expected for certain nuclear units

E

Prudent view of exports capacities at time period i

E

Import capacities at time period i

G

Fatal production including self-production, fatal hydraulics and variable renewable energies at time period i

G

Power output of the system at time period i

L

Available pumping power capacity at time period i L

Need for exports at time period i

L

lack of production opportunities at time period i L

Demand addressed to nuclear at time period i

L

French consumption at normal temperature at time period i L

Power output of the system at time period i

M

Margin France at time period i

M

Margin France with Q1% risk at time period i Parameters

C

Contingencies on the generating fleet K

Annual average availability of a CCGT LF

Annual average load factor

Scenario’s Acronyms

HH High Onshore Wind Power and High PV

HM High Onshore Wind Power and Medium PV

MH Medium Onshore Wind Power and High PV

MM Medium Onshore Wind Power and Medium PV

1 Introduction

Electricity is an electromagnetic phenomenon created by the interaction of particles present in matter that are positively or negatively charged and whose effects can be used to generate energy. Electricity is a secondary energy or energy carrier because it is generated from the transformation of primary energy by means of a conversion system. Primary energy is energy that is directly available in nature, such as fossil fuels, geothermal energy, solar radiation, wind or biomass. For example, the combustion of coal or gas (primary energy) is used in a thermal power plant to produce electricity (secondary energy). As far as sustainable development considerations are concerned, energy sources are frequently classified into two categories: renewable and non- renewable. The first includes solar energy (radiative energy), wind energy (kinetic energy), biomass (chemical energy), hydraulic energy (kinetic energy). In the second, fossil fuels (chemical energies) and nuclear energy are listed. In addition, when it comes to dealing with the issue of greenhouse gases, energy sources are classified into two categories. In the first, those that do not generate CO2 in their use. These include wind, solar, hydro and nuclear energy. In the second, the others. The transportation of the electricity is easy and quick. Power lines deliver electricity from the power station to the areas of consumption. The use of very high voltage makes it possible to limit line losses due to the Joule effect (heat release) or electromagnetic effects (capacitive and inductive effects between the line and the ground and between the lines). Distribution networks deliver electricity directly to end consumers. Some decentralised means of electricity production named local production (wind turbines, photovoltaic panels in private homes) can be directly connected to the distribution network and do not pass through the transmission network. However, electricity is difficult to store in sufficient quantities and at affordable costs to meet our energy needs. Direct solutions require

”resistance-free” conductors called superconductors in which it is theoretically possible to circulate the electricity to be stored without loss, but which are reserved for special applications and for small quantities. Indirect solutions only provide partial, expensive and often local solutions (accumulator batteries, power-to-gas, compressed air, water raised between two dams, etc.). Electricity production must therefore be continuous in order to meet current consumption levels. However, electricity consumption varies according to the time of day. It is therefore necessary to be able to adjust electricity production to consumption in order to maintain the supply-demand balance.

Since the end of the 20th century, in response to the growing scarcity of oil, the

negative climatic and health impacts of carbon energies, the geopolitical stress on the energy system, as well as nuclear accidents, there has been a global reorientation towards renewable energies. As a matter of fact, renewable energies are likely to provide an answer to these concerns. First, the use of primary energy by renewable energies is considered emissions-free and, although greenhouse gases are mainly emitted during their manufacture, transport and installation, the carbon balance of renewable energies is very often better than the one of non-renewable energies. Then, the switch to renewable energy consumption would lead to greater economic and political independence from fossil fuel exporting countries, the avoidance of resource- related conflicts and, the reduction of economic risks associated with energy shortages.

Lastly, the reduction of air pollution and the decentralisation of renewable energies in some territories has a local and short-term beneficial effect on health and on the labour market.

Thus, in the wake of Paris Agreement signed in 2015 at the COP21 summit, France adopted the same year The Law on Energy Transition for Green Growth (LTECV) along with the road-map The National Low Carbon Strategy (SNBC) to reduce its greenhouse gas (GHG) emissions and begin its energy transition. LTECV has set the objective of reducing greenhouse gas emissions by 40% between 1990 and 2030 and to divide them by 4 in 2050 compared to 1990 levels. The assessment of the implementation of the SNBC over the first 2015-2018 period found that the first carbon budget will be exceeded.

Thus, in a context of urgency to act, the Government has raised its ambition by enacting the Energy-Climate Law published on 9 November 2019 aiming to enshrine the objective of carbon neutrality in 2050 in the law. The Multi-Annual Energy Programming (PPE) is an energy policy steering tool coming from the LTECV which defines the trajectory that the government has set itself for the next ten years (2019-2023 and 2023-2028). In terms of electricity production and installed capacity, PPE aims to increase renewable energy capacity to 73.5 GW by 2023, to 101-113 GW by 2028, and reduce the share of nuclear power in electricity production to 50% by 2035. In particular, variable renewable energy sources (VRES) such as onshore wind power and solar photovoltaic (PV) are respectively expected to growth by 7.6 GW and 10.3 GW by 2023 compared to 2019 levels.

Consequently a growing share of renewable energies is expected in France in the short

and medium term which could disrupt the supply-demand balance of the electricity

sector.

1.1 Problem

The multiplication of variable renewable energies will have significant impacts on the power system and on the network that need to be considered when planning the transformation of the power system over the long term.

First, a significant proportion of renewable generation is intermittent as the production corresponds to natural flows, which are not permanently available and whose availability varies greatly without the possibility of control. Some of these energy sources have regular variations, such as tidal power and partly solar radiation, others are less regular, such as wind power. Secondly, increasing the share of variable renewable energy in the electricity mix of a country or region can lead to undesirable effects if it is not accompanied by the necessary measures to manage this variability. The need for flexibility, and in particular for electricity storage, increases in a non-linear way with the penetration rate of renewable energy. It could jeopardize security, stability, reliability, equal access and quality of supply and service. Thirdly, the distribution network, which until now has only had the role of transporting electricity to the end customer, will have to face major challenges. With the development of storage, electric vehicles, decentralized production of renewable energy, and the aspiration of citizens for energy autonomy, the arrival of ”smart grids” will play a key role in optimizing energy production and consumption. Lastly, some European wholesale electricity markets are facing episodes of very low or even negative prices due to integration in the merit order of variable renewable energy technologies characterized by their low marginal costs. Sharp price declines occur especially in periods of abundant renewable electricity production and low demand, situations in which some conventional means of production cannot operate below a technical minimum.

The integration of renewable energies therefore raises various short- and medium- term issues that challenge the supply-demand balance and the security of supply of the electricity system.

1.2 Master Thesis’ Purpose

The objective of this thesis is to assess the impacts of variable renewable energy

production on the supply-demand balance in order to better control them and optimise

the other assets that make up the French electricity system. The study focuses on

the analysis of the indicators such as price signals, Margins, Loss of Load Expectation

(LOLE), Expected Energy Not Served (EENS) and Nuclear drop stop that characterise the supply-demand balance and the security of supply of the electricity system.

A better understanding of the effects of variable renewable energies will allow, on the one hand, a better optimisation of conventional means of production such as nuclear power via the optimisation of maintenance shutdown planning and, on the other hand, to identify and anticipate the possible supply risks to which the energy system will be subjected during its transition to energies such as wind and photovoltaic solar power.

1.3 Methodology

The model used to represent french electricity market requires an optimisation technique to formulate the planning problems of the production units of the power system. This planning has to consider the technical prerequisites of the system and the economy. For this purpose, there are generic methods of optimisation [20]:

• Linear programming: Optimisation techniques for problems characterised by a linear objective function subject to linear equality and/or inequality constraints.

• Integer programming: Opitmisation technique for discrete decision variables.

Used to better value a capacity or resource allocation. This method is applied in finance for portfolio management.

• Enumerative methods and Constraint Programming: the modelling part is separated using Constraint Satisfaction Problems (CSP) from the solution part, whose particularity lies in the active use of the constraints of the problem to reduce the size of the space of the solutions to be run.

• Dynamic programming: consists in solving a problem by breaking it down into sub-problems, then solving the sub-problems, from the smallest to the largest, by storing the intermediate results. Used for process optimisation or stock management.

The objective is to optimise the management of hydraulic reserves, load shedding and

nuclear production. It is a question of deciding between immediate or deferred use

of a reserve. In other words, for a given stock, it is a question of knowing its value if

it is used immediately or if it is kept in order to be used later. For this purpose, the

optimization model used is based on an optimizer using dynamic programming. i.e. an

exact method for solving sequential optimization problems. It is used to determine the optimal operation of stock assets such as Hydro and Nuclear power generation. Within the optimizer any optimal solution is itself based on sub-problems solved locally in an optimal way. The model was applied to three research studies with their own scenarios.

First, VRES new installed capacity impact of supply-demand balance from 2021 to 2025 in France was assessed in order to bring a broad picture and the most important insights.

Within this study, two high VRES capacity development scenarios are compared to a reference scenario to assess the impacts on supply-demand balance indicators such as price signals, Margins and security of supply indicators LOLE and EENS. Then, VRES contributions on the security of supply were compared to Combined Cycle Gas Turbine (CCGT) in terms of energy and installed capacity. A reference and a high development scenario were built for onshore wind, PV and CCGT for the analysis. Lastly, the impact of the integration of VRES on the frequency and intensity of the specific to French Nuclear drop stop phenomena has been assessed by calculating the Technical Minimum (TM) and the Need of Exports during summer.

1.4 Thesis Framework

The report is structured in the following manner: Firstly, a literature review traces

previous research on 100% renewable energy scenarios, on the impact of variable

renewable energy on supply-demand balance in the medium-term and, on the power

system issues relative to variable renewable energy integration. The literature review

is followed by a background and scope presentation of renewable energy in the world,

in Europe and particularly in France. The main power generation units of France are

described with their main characteristics, costs and future trends as well as French

market mechanisms. Secondly, the optimization model OPUS is presented. OPUS is

an integrated optimization model developed by EDF that simulates French electricity

market. It encompasses detailed characteristics of supply assets and demand of France

as well as interconnections with neighboring countries. Thirdly, three research studies

with detailed results are presented: one on VRES new installed capacity impact of

supply-demand balance from 2021 to 2025 in France, another on VRES contribution on

the security of supply compared to Combined Cycle Gas Turbine (CCGT) and a last one

on the impact of the integration of VRES on the frequency and intensity of the specific to

French Nuclear drop stop phenomena. Lastly, a discussion presents the potential issues,

the key insight to take away and suggestions for further work.

1.5 Presentation of EDF

Electrcité de France (EDF) is a french multinational energy utility state-owned at 83.6%.

EDF is the first producer and distributor in France and in Europe. It is the second producer in terms of installed capacity in the world right after China Energy Investment.

EDF manage a diverse portfolio of 122.3 GW in 2019 of which 60% are nuclear power and is present in Eruope, South America, North America, Asia, Africa and the Middle East.

The company is engaged in electricity production, electricity distribution, energy supply as well as optimisation and trading. In 2019, EDF reckoned 164 727 staff members a turnover of 71.3 bn€ and 558 TWh of electricity produced of which 90% come from free- CO2 emitting generation units. For the coming decades EDF has been engaged with different goals. EDF aims to double their renewable energy installed power in the world to 50 GW by 2030. In 2019, EDF CO2 emissions amounts to 55g/kWh produced which is 5 times lower than the European mean of the sector [14]. Furthermore, EDF aims to reach carbon neutrality by 2050.

1.5.1 DOAAT (Department)

The Optimisation and Trading Department (DOAAT) is EDF optimisation and trading department in France. Its role is to ensure the physical supply-demand balance at best costs and by minimising risks over EDF perimeter in France.DOAAT have several missions including:

• Hedge physical and financial risks related with supply-demand balance and maximise profits.

• Search and manage at its best the production units flexibility and customer contracts.

• Look for the best energy buying and selling opportunities on the wholesale market through EDF Trading.

• Prepare EDF positions close to regulators.

• Look for and negotiate long term contracts with other energy utilities in the

electricity sector.

1.5.2 Portfolio and Market Management Office (GPM)

The Master Thesis has been held within the Portfolio Management team from the Portfolio and Market Management office (GPM). The team’s goal is to maximise EDF portfolio’s value by optimising EDF assets and commodities. This optimisation task is carried out while managing of financial and physical risks and respecting power plants constraints, contractual agreements and regulations. GPM’s roles is to give a forward- looking insights on supply-demand balance for the french and EDF’s area. This is done with a simulator by computing expected production balance and sells as well as marginal valorisations (vmar). In addition, GPM’s team carries out the physical and financial risks hedging. Thus, EDF, as balance responsible needs to hedge physical imbalances produced in its area and to hedge financially in order to protect EDF’s revenues against commodities and electricity market price volatility.

2 Literature Review

2.1 Previous Research

The goal of the literature review is to present the frame of reference from which is based this Master Thesis. The following section presents a review of existing research in three areas: 100% renewable electricity scenarios, modeling variable renewable energy impact on supply-demand balance in the medium-term and power system issues relative to variable renewable energy integration.

2.1.1 100% Renewable Electricity Scenarios

Following the objectives adopted in the Energy Transition Law in favour of green growth in terms of climate, several prospective studies have been carried out. In view of the carbon neutrality objectives for 2050, studies of 100% renewable system scenarios have been analysed.

Through the study ”Un mix électrique 100% renouvelable?” [1] l’ADEME (Agency for the Environment and Energy Management) has analysed the technical and economical feasibility of very high renewable energy rates in the electric mix. The ThreeMe model has been developed: a macro-economic and multi-sector model which mainly assess energy and environment policies impacts on the GDP and CO2 emissions.

Three scenarios have been drawn up: ” +80% of VRES”, ”100% of VRES” and ”100%

of VRES with lower social acceptability” where the use of the ground by the renewable energy sources has been constrained. In these scenarios, ADEME took into account the different production and consumption assets, their geographical distribution and electric flows. The supply-demand balance was made at the hourly rate. The electrical mixes envisaged by ADEME remain theoretical, since they are built from scratch and do not take into account the current situation nor the scenario to arrive at the result.

The study ”Reprint of Feasible path toward 40–100% renewable energy shares for power supply in France by 2050: A prospective analysis” [33]

analyses to what extent VRES injection impacts power system reliability and which flexible options could aid to connect a large share of VRES. The model used comes from TIMES family which offers a holistic modelling of power system options. The study applied to the case of France different scenarios with share from 0% to 100% by 2050.

The results shows a high development of VRES would need huge investments in new capacities, more flexible options, as well as imports and demand-response management.

If no ad-hoc technologies for the reliability of the grid are installed, the system power would likely be deteriorated.

The study ”Storage cost induced by a large substitution of nuclear by intermittent renewable energies: The French case” [22] explores the impacts of the penetration of VRES when electricity storage is not available. The impacts of high integration of VRES in the energy system are, a sharp decreases in electricity spot prices, a reduction of nuclear electricity output due to the merit order and a reduction of flexible power plant revenues. The reduction of spot prices will jeopardize the recovery in cost installation for all kind of power plants including VRES. In order to ensure the reliability of the system without fossil fuel backups, there will be a need of daily and seasonal electricity storage technologies such as batteries (daily) or power- to-gas (seasonal) converters which costs need to be considered in the costs-analysis of VRES. Reduction in costs of storage are also explored by considering costs of batteries in VRES marginal costs, by improving smart grid tools in order to enhance demand- side management, by coupling VRES with EVs or with the innovation and economy of scale.

Literature has paid higher attention to the exclusion of nuclear power in their models

due to the uncertainty related to their constructions time, their several delays and their

increasing investments costs for latest project such as Flamanville 3 in France, Hinkley

Point C in UK and Olkiluoto 3 in Finland. ”Low-carbon options for the French

power sector: What role for renewables, nuclear energy and carbon capture and storage?” [29] investigates the roles of different low- and negative- emission technologies in France and identifies the costs related to each solution in respect to social values for CO2 emissions. To reach carbon neutrality and cost reduction on the production side, the availability of CCS technologies and the consideration of negative emissions is a key point. If encouraged by policies, negative emissions could have a huge impact on carbon market and on investments attractiveness. Public support on negative emission scheme will be determinant for the implementation of strong business models for negative emission technologies.

A study from McKinsey, ”Net-Zero Europe : Decarbonization pathways and socioeconomic implications” [21] explores the least costly pathway and the technical feasibility to reduce European’s emissions by 55 percent by 2030 compared to 1990 levels and to reach net-zero emissions by 2050. The speed of decarbonization depends on the availability of mature technology and the ability to scale up supply chains.

EU countries could optimize their assets as a whole. In the case of the power sector, Northern EU countries would benefit from 30 to 60 percent more hours of onshore wind than those in the south. Southern countries would benefit from the 1 000 more hours of sunlight they receive each year. Power system may achieve net-zero by 2040.

From this primary energy produced with renewables, 75% would supply the power sector and 25% would be converted in green hydrogen to replace oil use in steel production, aviation and shipping. The power sector would be a central switchboard with the creations of channel and synergies between other sectors. Meeting renewable power demand would require increasing solar capacity from 20 GW per year to 50 GW per year by 2050, wind capacity from 15 GW per yer to 30 GW per yer by 2050, triple interconnections by 2030 and lastly increase battery storage capacity to 25 GW by 2030 and to more than 150 GW by 2050.

2.1.2 Modeling Variable Renewable Energy Impacts on Supply-Demand Balance in the Medium-term

Historically, over the past 15 years, the french power system has had an overcapacity

meaning that the supply fleet was oversized compared to the peaking requirements. For

environmental and economic issues, France has closed almost 12 GW of oil and coal

power plants since 2012. Now it has reached a situation where it closely meets national reliability standard. It still ensures the security of supply of the demand but there is no longer capacity margin according to RTE, the french transmission system operator [27]. Power cuts have already appeared and only caused by identifiable technical or meteorological incidents (storms). In a context where (i) the new Flamanville EPR commissioning has been postponed, (ii) the nuclear fleet is entering in a maintenance and reinvestment program causing long shut down periods for several nuclear reactors, and (iii) all European countries are closing high-emitting power plants, the security of supply has become today an increasingly important topic. Any negative impact will now have a direct impact when assessing the risk level of the power system. ”RTE French Adequacy Report” assess on a five-year period how to implement public energy policy when ensuring compliance with french reliability standard and rules for the good operation the power system.

Figure 2.1: Main developments in the French generation fleet between 2019 and 2023 and impact on capacity margins [27]

According to RTE’s models and assumptions capacity margin is expected to be altered over the next five years as shown in figure 2.1. For the next three years to come, are depicted in orange the negative effects on the capacity margin, in green the positive effects and in blue the cumulative effect of positive and negative measures on capacity margin.

• Between 2019 and 2025: National energy ambitions clarify by energy-climate law, will lead to a closure of 5 GW with the closure of 1.8 GW of the two nuclear reactors of Fessenheim and the closure of 3 GW of coal-fired power plants.

• For 2021-2022: The ability to maintain the balance in France will depend on

the commissioning of the new CCGT in Landivisiau, keeping the current nuclear

schedule under control, commissioning two interconnections between Italy and Great Britain, and the compliance of the onshore wind capacity growth.

• From 2022 onwards: after the decommissioning of the last coal power plant such as Cordemais located in Brittany, the problem of maintaining the grid voltage level in Western France would worsen significantly combined with two major risks:

a heavy maintenance for nuclear power fleet with several simultaneous ten-year inspections, and the final shutdown of nuclear power in Germany and the closure of coal-fired power plants in a large number of European countries.

• From 2022-2025: with the commissioning of the Flamanville EPR and the contribution of variable RES, the nuclear maintenance program would be more favorable.

A shown in Figure 2.2, the period 2021-2023 accounts for the majority of the risks : the nuclear fleet maintenance program and the change in European power systems are the main factors requiring attention during this period.

Figure 2.2: Capacity margins in the base case of the Mid-term Adequacy Report [27]

Levers have been identified to improve the security of supply and lead to increase

the margin capacity of at least 1 GW by 2022: (i) Controlling demand, especially

during peak hour, (ii) optimization of nuclear shutdowns periods for maintenance,

and (iii) Maintaining the Cordemais power plant or converting it to biomass. While it seems not be much room to optimize nuclear maintenance schedule, much leeway have been identified in the demand side management prospects on (low-carbon hydrogen production, e-mobility and buildings consumption) with short term actions during winter peaks and the generalization of the information of the system at national level.

Nowadays VRES are becoming more and more competitive even without subsidies. The study ”Impact of renewable resource quality on security of supply with high shares of renewable energies [13] explores how much reserve capacity is required to maintain reliable electricity supply by assessing the indicator loss of load hours (LOLH) and the indicator of the expected energy not served (EENS). Using the LOLH as the single metric for reliable power system planning is not enough. In the LOLH indicator, the amount of electricity not supplied is not part of the metric, only the hours of power undersupply are.

(a) Loss of load probability over the simulated weeks in the base case [27]

(b) Daily load curves during the cold spell in February 2012 [27]

Figure 2.3: Loss of load probability over simulated weeks and Daily load curves during the cold spell in February 2012

The opposite accounts for EENS. Therefore, both indicators are essentials to assess power reliability. In this Master Thesis, both indicators are considered when assessing power reliability. The main risk factor remains the onset of a cold wave, more than nuclear reactor unavailability or windless periods [27] as shown in figure 2.3a. A cold wave similar to the one experienced in February 2012 (figure 2.3b would lead in practically all situations to the use of post-market measures, and potentially even load shedding. Secondly, periods of low availability of nuclear reactors in the heart of winter, regardless of whether they are planned or unplanned, are likely to have a substantial negative effect on electricity supplies. To date, periods of low wind have had less impact.

However, the impact of this kind of events should increase in the long term. Low wind

situations occurring simultaneously in several European countries, thereby reducing the

ability to import electricity from neighbouring countries, should be considered a greater threat than an almost total absence of wind in France alone.

Subsidies given to VRES can be perceived as a market intervention in conventional electricity generation technologies. The study ”Do Intermittent Renewables Threaten the Electricity Supply Security?” [19] investigates the impact of VRES on investment on conventional energy sources. The analysis shows a significant negative effect of VRES on investments in peak-load capacity (mainly gas) while base-load is unaffected (coal). Conventional power plants are essential to the flexibility and reliability of the power system under extreme scenarios such as unfavorable weather conditions, outages in supply or peaks in demand.

Figure 2.4: Net annual revenue (i.e. market revenue minus variable production costs) for gas-fired combined cycle and combustion turbine plants from 2015 to 2019 and comparison with fixed cost assumptions [27]

According to RTE, [27] an analysis of revenues of “semi-base load” and “peak load”

power generation plants in recent years shows high volatility of annual revenues on the energy market due to cyclical energy conditions and the occurrence or non-occurrence of cold waves, which have a major impact on the demand for electricity. For certain years when temperatures were especially mild (2014 and 2015, for instance), annual revenue from gas-fired power plants was considerably below their fixed costs, even coming in at practically zero for combustion turbines as shown in figure 2.4.

In France the introduction of the capacity mechanism, operational since 2017, was

a key factor in sustaining this type of gas-fired power plants and in preserving the

security of supply. In France, Nuclear power accounts each year for more than 75% of

electricity production. With the surge of VRES, nuclear power is challenged: nuclear

is expected to face extreme power output variations as well as an increase number of shut-down/star-up events, even if the nuclear capacity was to be reduced. The study ”Nuclear contribution to the penetration of variable renewable energy sources in a French decarbonised power mix” [4] explores this issue. Three models have been created: First the ”Nuke Profile”, which focus on the nuclear time profiles with increasing shares of wind and solar. Two indicators are used to quantify extreme events : extreme power levels and shut-downs and start-ups needed to follow the load. Secondly the ”Low power model”. Thirdly the ”On/Off” model. Both use simple modeling nuclear dynamics. With high shares of VRES, nuclear would be less dependent on demand profile and becomes rather dependent on VRES profiles. Solar is more challenging than wind due to the peak of production during the midday and would force them to shut down several reactors at midday and restart them for the evening. The number of shut-down/start-up events would increase even with a reduction of nuclear fleet to 40 GWe.

With a rise of the penetration of VRES, the study ”Nuclear and intermittent renewables: Two compatible supply options the case of the French power mix” [5] also explores the flexibility of nuclear power as a contributor of the electricity supply-demand balance. While adding more VRES in the power system decrease the nuclear power load factor down to 40% in the most extreme scenarios, the costs increases could be offset by gradually replacing the plants. In order to be competitive against gas fired power plants as an alternative back-up option, nuclear power needs support. The solution for nuclear output could be to find new outlets such as the production of heat or hydrogen as nuclear has lower ramping rates than gas fired power plants (2.1). This

Table 2.1: Comparison of the flexibility abilities of dispatchable power plants [2]

Technology Start-up time Maximal

Change in 30s (%Pn)

Maximum

Ramp rate

(%/min) Open-cycle gas turbine

(OCGT)

10-20 min 20-30% 20%/min

Combined-cycle gas turbine (CCGT)

30-60 min 10-20% 5-10%/min

Coal plant 1-10h 5-10% 1-5%/min

Nuclear power plant 2h- 2days 5% 1-5%/min

analysis ensure that even if VRES could be curtailed it is preferable to find synergies

with other sectors and profit from the surplus of production. Other alternative back- up solutions could be analysed: load shedding, demand-side response management and storage.

The production of ”green” hydrogen at low costs is becoming a new challenge regarding the struggle against greenhouse gas emissions. This issue is explored as a high share of VRES is expected to be reached in France by 2050 leading to periods with large surplus of electricity at low prices. The study ”Adapting the French nuclear fleet to integrate variable renewable energies via the production of hydrogen: Towards massive production of low carbon hydrogen?”[3] analyses the potential of nuclear hydrogen production in term of volume, costs and emissions. Increased penetration of low variable cost VRES will result in diminished load factors for nuclear.

Energy carriers such as hydrogen could be a solution to modulate nuclear power output.

Figure 2.5: Planned use of electrolysis to decarbonise the industrial uses of hydrogen in the longer term [27]

With the expected hydrogen market growth driven by mobility uses, industrial uses and injections in the gas network, opportunities are emerging for the nuclear operator.

Hydrogen market driven by the use of electrolysis to decarbonise the industrial uses of hydrogen in the longer term (2.5) would lead to a two-fold increase by 2050.

2.1.3 Power System Issues Relative to Variable Renewable Energy Integration

The study ”Embedding power system’s reliability within a long-term Energy

System Optimization Model: Linking high renewable energy integration

and future grid stability for France by 2050” [28] employs the TIMES-FR model

through an intern indicator based on kinetic reserves into an optimisation model.

Results show that additional back-up and flexible options enable to ensure and maintain a stable grid while increasing the share of renewable energy sources up to 65% without jeopardizing the reliability of the system. In the case of a 100% renewable energy scenario, roughly 84% of the hours computed comply with grid reliability constraints.

In a scenario which guarantees a fully reliable 100% renewable energy power mix the aggregated capacity would double compared to the Business-As-Usual (BAU). Therefore, the power sector would increase its reliance on storage technologies, interconnections, demand-response and massive installation of new power plants in order to satisfy grid reliability constraints. Fixed operation and maintenance costs (Fixed O&M) are proportional to the installed capacity, then a threefold rise in annual costs is expected for the 100% renewable energy scenario compared to the BAU scenario.

Flexibility in power systems is the capability to provide supply-demand balance, control stability in extreme scenarios, and cope with disruptions and uncertainty on supply and demand sides. There are four different categories of flexibility: flexibility for power, energy, transfer capacity and voltage. The study ”Challenges of renewable energy penetration on power system flexibility: A survey” [15] explores the impact of variable renewable energy sources penetration on power system stability, small signal stability and frequency stability. Besides the uncertainty of demand side, there also lies uncertainty on the power generation side. Increasing demand in flexibility due to VRES outputs and forecast errors.

2.2 Background and Scope

2.2.1 World Outlook

(a) Capacity by energy source in 2019

(b) Growth from 2015 to 2019 (c) New installed capacity in 2019

Figure 2.6: Renewable energy capacity in the World (Others: Bio-energy, Geothermal

and Marine) [18]

All countries that approved the Paris Agreement in 2015 undertook a commitment to act to keep global warming under 2°C by 2100 and by stepping up efforts to avoid exceeding 1.5°C [30]. Changing the energy sector is key to a sustainable future. According to IEA, energy efficiency and renewable energy are the two key solutions to enable the global energy transformation in order to accelerate a reduction in CO2 emissions. Since 2015 an important development of VRES and particularly onshore wind and PV as shown in figure 2.6c. VRES are dominating the global market for new power generation capacity.

Since 2019 renewable power generation is growing faster than overall power demand while fossil fuel electricity generation decreased. Solar PV and Wind are increasingly the cheapest source of electricity in many markets and most renewables power sources will be fully cost competitive within the next decade. Despite varied transitions paths, all regions would see higher shares of renewable energy use. Europe poised to reach 70-80% shares in its total energy mix by 2050.

IRENA describes the four energy pillars for the future of energy:

Electrification: While VRES are increasing, so is energy demand. The electrification of end uses will drive increased power demand to be met with renewables. The share of renewable energy in global final energy consumption has increased only slightly since 2010 staying around a threshold of about 10%. The electrification in end-uses like heat and transport would exceed 50% with the increasing number of EVs, and heat pumps.

Increased Power system flexibility: Their integration requires an adaptation of

the infrastructures and the management of the electrical system. Electricity grids

were originally designed to carry centrally generated electricity in one direction only,

from production to consumption. The development of grid-connected decentralized

renewable energies will increasing bidirectional operation of the electricity networks and

profoundly change the structure, planning and operation of the power system. In the

future, the role of the networks will be not only to distribute the electricity produced,

but also to collect all decentralized production. Smart Grids [23] appear as one of the

possible solutions to facilitate the integration of renewable energies while guaranteeing

security, stability, reliability, equal access and quality of supply and service. In addition,

long term and short term storage will be important for adding flexibility, and the amount

of stationary storage would need to increase sharply as well as demand-side flexibility

and sector coupling. IRENA estimates that smart solution, such as smart charging EVs,

can significantly facilitate the integration of VRES by leveraging storage capacity and the

flexibility potential of the demand side [16].

Conventional renewable sources: Hydropower can bring important synergies to the energy system which include using hydropower to counteract the short-term variability of wind and solar generation, and seasonal complementarities in resource patterns. Bioenergy will become increasingly vital in end-use sector as it will remain a significant source of fuel for power and heat generation in industry and as a fuel used in transport [16].

Green Hydrogen: Hydrogen can offer a solution for energy sectors and technologies that are hard to directly electrify. Green hydrogen, produced by renewable electricity trough eletrolysis, has its costs falling fast. In the next few years in locations with favourable low-cost renewable energy green hydrogen will become cost competitive with

”blue hydrogen” which is produced from fossil fuels combiend with carbon capture and storage (CCS). As costs fall further, green hydrogen will be competitive with blue hydrogen in many location within the next 5 to 15 years. Hydrogen can be processed further into hydrocarbons or ammonia, which can then help to reduce emissions in shipping and aviation. A hydrogen commodity trade is nascent, the development of hydrogen could lead to geopolitical implications as well as further accelerating the demand for renewable power generation.

2.2.2 European Outlook

(a) Capacity by energy source in 2019

(b) Growth from 2015 to 2019 (c) New installed capacity in 2019

Figure 2.7: Renewable energy capacity in Europe (Others: Bio-energy, Geothermal and Marine) [18]

2.2.2.1 Renewable Energy Deployment For more than two decades EU has

been at the forefront of global renewable energy deployment. The adoption of long-

term goals and supporting policy measures has resulted in strong growth in renewable energy consumption across the region. In October 2014, The European Council agreed on a set of energy and climate goals for the period up to 2030, including a minimum target of 27% for the share of renewable energy consumed in the EU. In 2015, following this commitment, the Energy Union Framework strategy aimed to make the EU the world leader in renewable energy. Afterwards the EU ratified the Paris Agreement which established the target to limit the rise in global temperatures this century below 2°C compared to pre-industrial levels and to pursue the efforts to maintain the rise below 1.5°C. For the crucial 2020-2030 period, the European Commission submitted in 2016 the ”Clean Energy for All Europeans” package which request a regulatory framework to support renewable energy deployment. This package assess and identify cost-effective renewable energy options across the region in order to meet or exceed the 27% renewable share target by 2030. Since the adoption of these targets, key renewable technologies such as solar PV and offshore wind have achieved spectacular cost reductions. European energy policy has developed strongly since the beginning of the 2000s. [30] In particular, several European texts have set targets for limiting greenhouse gas emissions, increasing energy efficiency and increasing the energy generated from renewable sources. The climate energy package, adopted under EU presidency in 2008, set ”3x20” targets for 2020:

• -20% GHG emissions

• 20% improvement in energy efficiency

• 20% renewable energy in final EU energy consumption.

In addition, in 2018, The EU adopted targets for 2030 such as the reduction of the EU’s GHG emissions by at least 40% 2030 compared to 1990 level as well as reach at least 32% renewable energy in energy consumption. GHG emissions will be achieved thanks to the European Emissions Trading Scheme (ETS). A new European governance regulation now requires each Member States to publish a ten-year integrated national energy-climate plan, which match in France with the Multi-Annual Energy Plan (PPE) and with the Low-Carbon National Strategy (SNBC).

The simultaneous decommissioning of many production facilities in Europe is a key

issue. The French power system now relies on imports from neighbouring countries

to cope with the high-est demand peaks. This contribution of interconnections to

ensure the security of supply is part of economic optimisation and pooling of production

(a) Main targets for decommissioning thermal facilities in Europe.

(b) Carbon intensities from electricity generation in Europe in the base case in 2025.

Figure 2.8: Main targets for decommissioning thermal facilities and Carbon intensities from electricity generation in Europe. [27]

resources on a European scale. It does, however, mean that the availability of capacities in neighbouring countries needs to be closely monitored to be able to study the security of supply as shown in figure 2.9 [27].

Figure 2.9: Capacity margins in the base case and in other scenarios of European fleets [27]

2.2.3 French Power System Outlook 2.2.4 National Context

The Law on Energy Transition for Green Growth (LTECV) set the energy policy

framework in France until 2050 [30] . The LTECV embraces European commitments

and proposes ambitious national targets in terms of energy.

By 2020:

• 23% of energy consumption should be from renewable sources.

By 2030 :

• -40% GHG emissions compared to 1990

• -20% of finale energy consumption compared to 2012

• -40% of primary fossil energy consumption compared to 2012

• +27% energy efficiency

• 40% of electricity generation from renewable sources which will be discussed later in this chapter.

• A 5-fold increase of heating and cooling from renewable sources and recovery sources in heat networks compared to 2012.

LTECV targets are covered over the next 10 years by the Multi-Anual Energy Plan (PPE).

This national roadmap from the french government outlines the priority actions for the energy sector in the mainland. This roadmap is split in two periods of 5 years (2019- 2023) and (2024-2028). At the end of the first period, the second period is reviewed and a new 5-years period is added afterwards. The PPE is regulated according to the law for ”the energy transition and the green growth” enacted in 2015 and to the law for ”the energy and the climate” implemented in 2019.

The PPE contains clear targets and guidelines for the following segments :

• Security of supply.

• Energy efficiency enhancement and the reduction of primary consumption an particularly fossil-fuel consumption.

• Renewable energy and recovery energy development and operation.

• Balanced development of grids, storage, demand side energy management in order to promote local production, smart grids and self-consumption.

• Customer purchasing power preservation

• Competitive energy prices.

The law has set a target of 40% of renewable energy in electricity production in 2030.

France has initiate a major evolution of the electrical system with a strong acceleration

of renewable energy sectors. In 2016 the PPE has set a lower and a higher objective for installed capacity of renewable electricity to be reached by 2030:

• Lower PPE Objective : 71 GW

• Higher PPE Objective : 78 GW

2.2.5 Demand

In France, in 2019, gross consumption corresponding to consumption in France, including Corsica and losses, amounted to 473.7TWh, slightly down compared to 2018 (-0.5%) and relatively stable over the last ten years [25].

There is a structural slowdown in electricity consumption in France, which is also observed in most European countries. This is due to a diffusion and reinforcement of energy efficiency actions within buildings and on the performance of equipment. It is also due to a slowdown in economic and demographic growth. And finally to the structural evolution of economic activity, which tends to become tertiary, with services consuming four to five times less electricity than the industrial sector at an equivalent level of GDP.

2.2.6 Power Generation

Figure 2.10: Capacity installed and production by sector in France in 2019 [25]

Total electricity production in France amounted to 537.7 TWh in 2019, a decrease of 2%

(11 TWh) compared to 2018. Renewable energies provide more than 21% of electrical

energy total despite a drop in Hydropower production of more than 12% compared

to 2018. Wind power production has indeed increased sharply compared to 2018 (+

21.2%), as does solar production, a significant increase of +7.8% [25].

2.2.6.1 Nuclear Power In France, nuclear installations consisted of 58 nuclear power reactors until February 2020. With the closure of Fessenheim’s first reactor on 22th of February 2020 and the closure of the second reactor on 30th of June 2020, France now relies on 56 nuclear power reactors. With a current total capacity of 61 370 MW, Nuclear is the main type of power generation in France. Nuclear power is divided in three main categories of reactors sorted by capacity. There are 34 reactors of 900 MWe, 20 reactors of 1 300 MWe and 4 reactors of 1 450 MWe. It’s worth noting that a new last generation reactor (EPR) of 1 650 MWe in Flamanville is scheduled to be commissioned by 2023.

In 2019, 379.5 TWh were produced from nuclear reactors representing 70.6% of total electricity generation. It has decreased by 3.5% compared to 2018 level (393.2 TWh).

For 2020 several outages due to the Covid-19 outbreak have decreased the nuclear availability and therefore the energy produced from nuclear power is expected to be close to 335 TWh which is less than 2019 by 12.25%.

In terms of costs, the full economic costs calculated by The Court of Auditors estimated the production costs for existing power plant at 61.6€/MWh. Nuclear costs are broken down in different parts. First, variable OM costs that are function of the level of energy production equivalent to marginal costs. Marginal cost for nuclear power plants depends mostly on their fuel constraints and on their maintenance schedule. For the cheapest and the less constrained, the value is of the order of 10€/MWh. Nuclear power plant are therefore called very soon in the merit order : before fossil-fuel power plants and just after variable renewable energy as shown in figure 4.2. Then, fixed OM costs depends on fixed operating expenditures and current and future investments. Fixed OM costs are estimated to represent on average 33€/MWh. Last costs include decommissioning costs and waste management costs.

The Intergovernmental Panel on Climate Change (IPCC) and estimates that the carbon

impact is on average at 12gCO2/kWh. Most emissions are associated with the

exctraction, conversion and enrichment of uranium (49%) and with construction,

operation and decomissioning (40%). In the case of radioactive materials and waste

produced by nuclear reactors, a safe and sustainable final destination is sought to prevent

and limit the burden for future generations and for the protection of human health, safety