Power systems reliability assessment in prospective analyses
1 like825 views
The document presents a reliability assessment of the French power system under various renewable energy penetration scenarios up to 2050. It details a modeling approach, evaluates the technical feasibility of high renewable integration, and identifies potential increases in costs and stability issues due to the shift in power production profiles. Key findings emphasize the importance of new dispatchable power plants and the development of kinetic indicators to assess system robustness.
Power systems reliability assessment in prospective analyses
- 1. Power systems reliability assessment in prospective analyses: the case of France under ambitious renewable energy penetration scenarios Vincent KRAKOWSKI, Nadia Maïzi, Edi Assoumou, Vincent Mazauric MINES ParisTech, PSL Research University, Center for Applied Mathematics (CMA) Chair modeling for sustainable development Sophia Antipolis, France ETSAP Workshop October 22-23, 2015 Sophia-Antipolis, France
- 2. European power production by fuels in 2014 Increasing role of renewable energy in power production ETSAP 2015, October, Sophia-Antipolis 2 / 20 1. Context of the study Global power production by fuels in 2014 Source: REN 21 – Global Status Report 2015 Share of renewable energy in power production per European country in 2014 Source: ENTSOE– Electricity in Europe 2014 20% 2020: 27% 2030: 40%
- 3. OUTLINE 1. Introduction 2. A TIMES model for French power sector 3. Scenarios and results 4. Conclusion and perspectives ETSAP 2015, October, Sophia-Antipolis 3 / 20
- 4. Reference Energy System of TIMES-FR-ELEC ETSAP 2015, October, Sophia-Antipolis 4 / 20 A TIMES MODEL FOR FRENCH POWER SECTOR Time description: • 2012 2050: 13 « annual » periods • Each annual period divided into 84 « infra-annual » sub-periods Ressources Power plants and storage technologies Power grid DemandEmissions
- 5. Modeling VRE in TIMES ETSAP 2015, October, Sophia-Antipolis 5 / 20 A TIMES MODEL FOR FRENCH POWER SECTOR Non-dispatchable No inherent inertia Lower predictability Variability Load curves (wind, PV) No contribution to peak load and reserves Low production events time-slices No contribution to power system reliability Dispersed Work in progress Variable renewable characteristics Modeling technics𝑱𝝎 𝒎 𝒅𝝎 𝒎 𝒅𝒕 = 𝑷 𝒎 − 𝑷 𝒆 − 𝑷𝒍 Y. Rebours, A comprehensive assessment of markets for frequency and voltage control ancillary services. PhD thesis (2008)
- 6. General assumptions used in this study ETSAP 2015, October, Sophia-Antipolis 6 / 20 A TIMES MODEL FOR FRENCH POWER SECTOR COSTS ASSUMPTIONS • Ressources prices (WEO, 2013) • Investment and operational costs of power plants, storage capacities and new lines POTENTIALS ASSUMPTIONS • RES potentials (ADEME, 2012) • New Interconnections: RTE scenarios • Rhythms of installation for new technologies • Extension of current nuclear power plants (40 ans --> 60 ans) DEMAND ASSUMPTIONS • Reference scenario BP RTE (2012: 495 --> 2050: 519 TWh) • Demand elasticity: per sector (- 8% in 2050) • Demand-Response: EV (50%), commercial & residential (10%) POWER CURVES • Wind and solar production (RTE, 2012 data) • Hydropower: seasonal curves • Nuclear power: Availability factor min and max (45% - 95%) CO2 emissions < 2012 value (39 Mt)
- 7. OUTLINE 1. Introduction 2. A TIMES model for French power sector 3. Scenarios and results 4. Conclusion and perspectives ETSAP 2015, October, Sophia-Antipolis 7 / 20
- 8. Description of the 5 studied scenarios ETSAP 2015, October, Sophia-Antipolis 8 / 20 SCENARIOS AND RESULTS Scenarios Years BAU 2020-2030 objectives 60RES 2050 80RES 2050 100RES 2050 Nuclear production constraint 2025 and after NA 50% 50% 50% 50% RES penetration objectives 2020 NA 27% 27% 27% 27% 2030 NA 40% 40% 40% 40% 2035 NA 40% 40% 40% 55% 2040 NA 40% 40% 50% 70% 2045 NA 40% 45% 65% 85% 2050 NA 40% 60% 80% 100% Law on energy transition (2015/08/18) More ambitious targets for RES penetration
- 9. Evolution of the power mix in the 5 scenarios ETSAP 2015, October, Sophia-Antipolis 9 / 20 SCENARIOS AND RESULTS Exports / Demand Elasticity
- 10. Installed capacity in the 5 scenarios ETSAP 2015, October, Sophia-Antipolis 10 / 20 SCENARIOS AND RESULTS x 3 x 2 34 GW 49 GW 54 GW
- 11. Profitability issue for fossil-based power plants ETSAP 2015, October, Sophia-Antipolis 11 / 20 SCENARIOS AND RESULTS
- 12. A completely new power production profile ETSAP 2015, October, Sophia-Antipolis 12 / 20 SCENARIOS AND RESULTS
- 13. Evolution of costs in the different scenarios ETSAP 2015, October, Sophia-Antipolis 13 / 20 SCENARIOS AND RESULTS 0% 5% 10% 15% 20% 25% 30% 35% Totaldiscountedcostofeachscenario comparedtoBAUtotaldiscountedcost
- 14. Evolution of kinetik reserves deviation for each scenario relative to the minimum 2012 value Renevable Energy penetration causes a serious drop in kinetik reserves ETSAP 2015, October, Sophia-Antipolis 14 / 20 SCENARIOS AND RESULTS
- 15. OUTLINE 1. Introduction 2. A TIMES model for French power sector 3. Scenarios and results 4. Conclusion and perspectives ETSAP 2015, October, Sophia-Antipolis 15 / 20
- 16. Main results • High renewable energy penetration in the French power system seems technically feasible (for a slightly reduced demand) BUT with: A shift in power exchanges with neighboring countries The massive installation of new power plants The need for new dispatchable power plants (biomass, fossil) but which ought to produce very few 11%-31% cost increase compared to a BAU scenario The deterioration of power system’s margins to deal with disturbances ETSAP 2015, October, Sophia-Antipolis 16 / 20 CONCLUSION
- 17. Perspectives: Endogenization of kinetik indicators (1/2) • We aim at building renewable energy penetration scenarios that enable maintaining kinetik reserves above a certain level Making the calculation of kinetik indicator endogenous in the model Discretizing the model: description of each unit of power production and discret investments (Remark: Hkin varies non-linearly with units capacity) Need a good description of the current power system and the disaggregation of future power plants according to their size ETSAP 2015, October, Sophia-Antipolis 17 / 20 CONCLUSION Database Platts (2013) for France 1. Classification List of power plants by kind of fuel, kind of power production and size of the unit 2. Creation of VT file List of process, their commodities in and out , their capacity and the value of Hkin for each 3842 operating units 729 processesFor the BY power system
- 18. Perspectives: Endogenization of kinetik indicators (2/2) • Such an endogenization has already been realized in the CMA for the Reunion Island: It has been showed that intermittent renewable energy share could reach up to 60% without deteriorating too much kinetik reserves ETSAP 2015, October, Sophia-Antipolis 18 / 20 CONCLUSION Source: S. Bouckaert, PhD thesis, CMA, Mines ParisTech, 2013 Kinetik indicator in 2030 for teh different hours of the day for 4 typical days in 3 scenarios
- 19. Perspectives: A new indicator for assessing power systems robustness (1/2) • The kinetik indicator can be seen as an image of the reliability of a power system as it assesses how this power system will be able to meet demand when there is a mismatch between production and consumption. • But the calculation of this indicator is based on the conservation of synchronism over the power system: without synchronism, kinetik reserves from the different power plants could not be aggregated. • Based on the Kuramoto model which depicts the dynamics of non-linear coupled oscillators connected through a network (Kuramoto, 1975), and more recent work (Filatrella et al., 2008, Dörfler and Bullo, 2013) we use a new indicator which assesses the stability of synchronal states of power systems. ETSAP 2015, October, Sophia-Antipolis 19 / 20 CONCLUSION (1) 𝜟P 𝜺 G λ2 G ≤1 Where G is the graph representing the power grid, 𝜀G its edges, P the vector of productions and consumptions and 𝜟P 𝜺 G a norm on the differences between adjacent nodes production/consumption. λ2 G is the algebric connectivity of the graph G.
- 20. Perspectives: A new indicator for assessing power systems robustness (2/2) • The condition (1) can be understood as “the power grid connectivity must dominate the power system heterogenity”. A first analysis using this condition on the current Reunion power system has been conducted: ETSAP 2015, October, Sophia-Antipolis 20 / 20 CONCLUSION
- 21. ETSAP 2015, October, Sophia-Antipolis Thank you for your attention Any questions? Related published or under-publication work: • Krakowski et al. A prospective analysis of contrasted renewable energy penetration targets in French power system. Publication under process • Krakowski et al. Enjeux d’une transition vers une production d’électricité 100% renouvelable en France. Revue de l’Energie, to be published (French only) • Drouineau et al. Increasing shares of intermittent sources in Reunion Island: Impacts on the future reliability of power supply. Renewable and Sustainable Energy Reviews (2015) • Maïzi et al. Future prospects for nuclear power in France. Applied Energy (2014) • Bouckaert et al. Expanding renewable energy by implementing Demand response and Expanding renewable energy by implementing dynamic support through storage technologies Energy Procedia (2014) • Drouineau et al. Impacts of intermittent sources on the quality of power supply: The key role of reliability indicators. Applied Energy (2014) vincent.krakowski@mines-paristech.fr