Incorporating uncertainties in the transition towards a clean European energy system: a stochastic approach for decarbonization paths in the transport sector.

1. IER
Incorporating
uncertainties towards a
sustainable European
energy system: a
stochastic approach for
decarbonization paths
focusing on the
transport sector
Pinar
Korkmaz

2. • Introduction
• Research Question
• Methodology
• Scenario Analyses
6/30/2020Pinar Korkmaz 2
Agenda

3. •Introduction
• Research Question
• Methodology
• Scenario Analyses
6/30/2020Pinar Korkmaz 3
Agenda

4. • EU energy system transition is required to cut the GHG emissions.
• Transport sector itself covers almost the quarter of the GHG emissions in the entire EU.
• Although other sectors have been able to make certain move to reach the targets;
transport sector has not achieved any considerable decline in GHG emissions.
• The deployment of the low-emission alternative energy options for transport needs to be
accelerated.
• This is also identified as one of the priority areas for the action in the EU.
• Biofuels and electric vehicles stand out as mitigation options with their potentials and
technological development so far they have.
6/30/2020Pinar Korkmaz 4
European Energy System Transition & Mitigation Options in Transport
Introduction

5. • Electric vehicles have experienced certain technology learning over the life time of the
technology due to their battery packs.
• According to Schmidt (2017), different cost reduction patterns are possible for the
battery packs of the electric vehicles along with their technology learning.
• Therefore, investigating the learning uncertainties of the EVs is an essential step to
address their role in the decarbonization of the transport sector.
• There is a target defined for the share of the biofuel usage in the transport sector in the
EU level in 2020.
• It is not identified yet how to utilize the existing biomass potential in the EU during the
energy transition.
• How does the decarbonization path does look like with the limited biomass availability
also considering the technology innovation with the EVs.
6/30/2020Pinar Korkmaz 5
Technology Learning-EV and Biomass Potential in Transport
Introduction

6. • Introduction
•Research Question
• Methodology
• Scenario Analyses
6/30/2020Pinar Korkmaz 6
Agenda

7. • To address the decarbonization paths in transport sector during the energy transition
in EU considering the uncertainties;
• Available biomass potential in transport sector,
• Technology learning for the electric vehicles,
• Different reduction targets to cut GHG emissions in the energy system,
• Different resolution times for the considered uncertainties.
6/30/2020Pinar Korkmaz 7
Objective
Research question

8. • Introduction
• Research Question
•Methodology
• Scenario Analyses
• Outlook
6/30/2020Pinar Korkmaz 8
Agenda

9. • Stochastic modeling is a method:
• To make optimal decisions under risk,
• To adress the specific uncertainties.
• Each uncertain parameter is considered to be a random variable,
• Stochastic bottom-up energy system model optimizes the discounted system cost of future State of the Worlds (SOW)
according to weighted average of the given probabilities.
• Objective Function:
6/30/2020Pinar Korkmaz 9
Stochastic Modeling-General aspects
Methodology-Modeling
3 Berglund, C. 2006
Minimize:
Where:
P (t, w): probability of the scenario w in period t
and

10. • The main difference between stochastic modeling and sensitivity analyses is the calculation of the
objection function.
• During the sensitivity analyses, it is not possible to take into account the cost of uncertainty born
because of the uncertainties and to develop a hedging strategy until the uncertainties are resolved.
• The results from the sensitivity analyses might give disputing results which might not be
preferable to give policy relevant messages for the policy makers.
• According to a certain scenario tree, different stages can be determined to adress the uncertainites
at different periods.
• Stochastic analysis determines a hedging and several recourse strategies to deal with the
uncertainties.
6/30/2020Pinar Korkmaz 10
Stochastic Modeling-General aspects
Methodology-Modeling
3 Berglund, C. 2006

11. • 30 region (EU 28 + NO, CH) model,
• Time horizon: 2010-2050,
• 12 time slices (4 seasonal, 3 day level),
• GHG: CO2, CH4, N2O,
• Country specific differences (characterisation of new power plants, load curves,
availability factors for renewable energy sources),
• EU related policies are implemented such as maximum and minimum shares of energy
carriers in different sectors,
• Main database: EUROSTAT,
• Other pollutants: SO2, NOx, CO, NMVOC, PM2.5, PM10
6/30/2020Pinar Korkmaz 11
TIMES PanEU (The Pan-European Model)
Methodology-Modeling

12. • Introduction
• Research Question
• Methodology
•Scenario Analyses
• Outlook
6/30/2020Pinar Korkmaz 12
Agenda

13. 6/30/2020Pinar Korkmaz 13
Parameter Variations – Deterministic Analysis Scenarios
Scenario Analyses
2050 GHG Reduction
Target
Learning in EVs Biomass Potential
80% according to the
level in 1990
High Learning (HL) High Biomass (HB)
High Learning (HL) Low Biomass (LB)
Low Learning (LL) High Biomass (HB)
Low Learning (LL) Low Biomass (LB)
90% according to the
level in 1990
High Learning (HL) High Biomass (HB)
High Learning (HL) Low Biomass (LB)
Low Learning (LL) High Biomass (HB)
Low Learning (LL) Low Biomass (LB)
Learning uncertainity of the battery packs:
• Learning curve methodology.
• Highest and lowest reduction curves of battery packs
from (Schmidt, 2017) for the EVs.
.
Biomass availability in transport sector:
• 1500 PJ as the maximum potential in 2050 in LB SOWs to be
inline with the EU renewable targets according to The
Renewable Energy Directive (2009/28/EC).

14. 6/30/2020Pinar Korkmaz 14
Parameter Variations–Deterministic Analysis: 80% reduction target
Scenario Analyses
Figure 1: Final energy consumption in transport (without international
aviation and waterborne) for 80% reduction target – Deterministic sensitivity
analysis
Figure 2: Biofuel usage in transport (without international aviation and
waterborne) for 80% reduction target – Deterministic sensitivity
analysis

15. 6/30/2020Pinar Korkmaz 15
Parameter Variations–Deterministic Analysis: Electricity usage in 80% & 90% reduction target and Biomass usage
according to sectors in 90% target
Scenario Analyses
Figure 3: Electricity usage in road transport – Deterministic sensitivity analysis Figure 4: Biomass usage according to sectors – Deterministic sensitivity
analysis for Low Learning High Biomass Scenarios

16. 6/30/2020Pinar Korkmaz 16
Stochastic Analysis : Hedging 2025
Scenario Analyses
Figure 5: Stochastic scenario tree – 80% & 90% reduction target
Figure 6: Difference in different energy carriers consumption relative to
deterministic runs-Hedging strategy in 2020 (as a representative year for the
period between 2018 and 2022) in 80% reduction target

17. 6/30/2020Pinar Korkmaz 17
Stochastic Analysis : Hedging Uncertainty - 2040
Scenario Analyses
Figure 8:Stochastic tree – variation of hedging period with 80%
reduction target
Figure 9: Electricity consumption in car transport – Longer hedging period in 80%
reduction target
Figure 10:Difference in biofuel consumption of transport modes relative to
deterministic runs with longer hedging period in 80% reduction target-
Hedging strategy in 2020 and 2030

18. 6/30/2020Pinar Korkmaz 18
Stochastic Analysis : Combining biomass uncertainty with reduction target uncertainty
Scenario Analyses
Figure 11:Stochastic scenario tree – combining reduction target and biomass
uncertainties
Figure 12: Difference in biomass utilization in different sectors relative to
deterministic runs with High Learning for EVs (PJ) - Hedging strategy

19. Name of the Analysis
Expected value of perfect
information (MEUR)
% Relative to the stochastic
total system cost
80% Reduction 370,978 0.767%
90% Reduction 375,082 0.770%
Longer Hedging 567,542 1.145%
Red. Target & Biomass Unc. 709,181 1.428%
6/30/2020Pinar Korkmaz 19
Stochastic Analysis: Expected value of perfect information
Scenario Analyses
Table 3: Expected value of perfect information
To show the difference in cost between the stochastic approach and deterministic scenario
analyses expected value of perfect information (EVPI) is calculated.
Where:

20. • Policy uncertainty for the decarbonization target has the highest impact between the
studied uncertainties on the development of the transport sector.
• Additionally, decarbonization of car transport is prioritized and the electric cars appear
as no-regret options.
• Longer resolution time for the considered uncertainties accelerates the deployment of
electric vehicles in the hedging period, while it does lower their deployment in the
recourse strategies compared to having shorter hedging period.
• Longer hedging period has an impact on the biomass utilization, by decarbonizing the
aviation in the early periods relative to shorter hedging period.
6/30/2020Pinar Korkmaz 20
Conclusions

21. • European Commission, "Clean Energy," December 2019. [Online]. Available: https://ec.europa.eu/commission/presscorner/detail/en/fs_19_6723.
[Accessed 23 January 2020]
• United Nations, "Paris Agreement," 2015. [Online]. Available: https://unfccc.int/sites/default/files/english_paris_agreement.pdf. [Accessed 1 January
2020]
• European Commission, "IN-DEPTH ANALYSIS IN SUPPORT OF THE COMMUNICATION COM (2018) 773 - A clean Planet for all A European
long-term strategic vision for a prosperous, modern, competitive and climate neutral economy," 2018. [Online]. Available:
https://ec.europa.eu/clima/sites/clima/files/docs/pages/com_2018_733_analysis_in_support_en_0.pdf. [Accessed 30 June 2019]
• European Commission, "COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT; THE EUROPEAN COUNCIL,
THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF TEH REGIONS The European Green
Deal," 12 December 2019. [Online]. Available: https://eur-
lex.europa.eu/resource.html?uri=cellar:b828d165-1c22-11ea-8c1f-01aa75ed71a1.0002.02/DOC_1&format=PDF. [Accessed 23 January 2020]
• European Envrionmental Agency, "EU GHG inventory 1990-2016, proxy GHG estimates for 2017," 2017.
• European Commission, "JRC Science for Policy Report Global Energy and Climate Outlook 2017: How climate policies improve air quality," 2017.
[Online]. Available: https://publications.jrc.ec.europa.eu/repository/bitstream/JRC107944/kjna28798enn(1).pdf. [Accessed 12 January 2020].
• European Commission, "10 Trends Reshaping Climate and Energy," 2018. [Online]. Available: https://ec.europa.eu/epsc/sites/epsc/files/epsc_-
_10_trends_transforming_climate_and_energy.pdf. [Accessed 12 January 2020].
• European Commission, "WHITE PAPER : Roadmap to a Single European Trasnport Area - Towards to a competitive and resource efficient transport
system," 28 03 2011. [Online]. Available: https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2011:0144:FIN:EN:PDF. [Accessed 4 August
2019].
• W. Usher and N. Strachan, "Critical mid-term uncertainities in long-term decarbonisation pathways," Energy Policy 41, pp. 433-444, 2012.
• A. Lehtila and R. Loulou, "Stochastic Programming and Trade-off Analysis," May 2016. [Online]. Available: https://www.iea-etsap.org/docs/TIMES-
Stochastic-Final2016.pdf.
• O. Schmidt, A. Hawkes, A. Gambhir and I. Staffel, "The future of cost of electrical energy storage based on experience rates," Nature Energy , 2017.
7/1/2020Pinar Korkmaz 21
References

22. Vielen Dank!
E-Mail
Telefon +49 (0) 711 685-
Fax +49 (0) 711 685-
Universität Stuttgart
Heßbrühlstraße 49a, 70565 Stuttgart
Pinar Korkmaz
87846
87873
Institut für Energiewirtschaft und
Rationelle Energieanwendung (IER)
Pinar.korkmaz@ier.uni-stuttgart.de