Representing Cross-border Trade in Long-term Power System Planning Models with Limited Geographical Scope

1. Representing Cross-border Trade in Long-term Power
System Planning Models with Limited Geographical Scope
Tim Mertens, Kris Poncelet, Jan Duerinck & Erik Delarue
ETSAP workshop, Paris
7 June 2019
1 / 17

2. Planning models with a limited geographical scope
Recent studies regarding long-term planning have focused on the impact of:
• the temporal/spatial resolution
• the level of technical detail
However, less attention has been drawn to the impact of the model’s geographical scope.
2 / 17

3. Planning models with a limited geographical scope
Recent studies regarding long-term planning have focused on the impact of:
• the temporal/spatial resolution
• the level of technical detail
However, less attention has been drawn to the impact of the model’s geographical scope.
Often long-term planning models are designed for a specific country or region, e.g.,
TIMES-Belgium, which requires a proper representation of cross-border trade with
neighbouring regions.
• Extend the geographical scope beyond the focus region, e.g. NREL’s RPM
• Define exogenous import and export functions/processes
2 / 17

4. Planning models with a limited geographical scope
Recent studies regarding long-term planning have focused on the impact of:
• the temporal/spatial resolution
• the level of technical detail
However, less attention has been drawn to the impact of the model’s geographical scope.
Often long-term planning models are designed for a specific country or region, e.g.,
TIMES-Belgium, which requires a proper representation of cross-border trade with
neighbouring regions.
• Extend the geographical scope beyond the focus region, e.g. NREL’s RPM
• Define exogenous import and export functions/processes
How to properly design and use these import and export functions?
2 / 17

5. Content
• Introduction
• Methodology
• Illustration
• Wrap-up
3 / 17

6. Content
• Introduction
• Methodology
• Illustration
• Wrap-up
3 / 17

7. Methodology
Make assumptions regarding the capacity mix and electricity demand in the neigbouring
countries.
• Existing studies/scenario ananlyses
• Communicated policy targets
• ...
The methodology adopted for representing cross-border trade can be summarized in the
following three steps.
1 Construct import/export functions.
2 Include the obtained functions from 1 in the optimization model (and solve the
optimization model).
3 Perform an ex-post cost reallocation.
4 / 17

8. 1. Construct import/export functions
Construct import and export functions that:
• Reflects the potential of other countries to facilitate
electricity imports (import curve)
• Reflects the willingness-to-pay of other countries for
electricity exports (export curve)
5 / 17

9. 1. Construct import/export functions
Construct import and export functions that:
• Reflects the potential of other countries to facilitate
electricity imports (import curve)
• Reflects the willingness-to-pay of other countries for
electricity exports (export curve)
Due to varying demand and intermittent RES, the
process is repeated for every time step.
D
ImportExport
p
q
5 / 17

10. 2. Planning model formulation
Min Investment cost+Generation cost + Import cost - Export revenue
Subject to:
• System constraints (e.g., supply-demand balance)
• Policy constraints (e.g., RES targets, emission
prices)
• Technical constraints (e.g., generation limits)
• Cross-border trade constraints
Import cost =
i∈I t∈T
Pimp
i · importi,t
Export revenue =
e∈E t∈T
Pexp
e · exporte,t
D
ImportExport
p
q
6 / 17

11. Export revenue in objective function
DAp
q
DBp
q
A B
7 / 17

12. Export revenue in objective function
DAp
q
f
DBp
q
f
A Bf
7 / 17

13. Export revenue in objective function
DAp
q
f
DBp
q
f
A Bf
7 / 17

14. Export revenue in objective function
DAp
q
f
DBp
q
f
A Bf
7 / 17

15. Import cost in objective function
DAp
q
f
DBp
q
f
A Bf
8 / 17

16. 3. Ex-post cost reallocation
DAp
q
f
DBp
q
f
A Bf
9 / 17

17. 3. Ex-post cost reallocation
DA
λA
p
q
f
DB
λB
p
q
f
A Bf
9 / 17

18. 3. Ex-post cost reallocation
DA
λA
p
q
f
DB
λB
p
q
f
A Bf
9 / 17

19. 3. Ex-post cost reallocation
DA
λA
p
q
f
DB
λB
p
q
f
A Bf
9 / 17

20. 3. Ex-post cost reallocation
DA
λA
p
q
f
DB
λB
p
q
f
A Bf
9 / 17

21. 3. Ex-post cost reallocation
The Import/export curves trigger the correct investment decisions, however the objective
function does not represent the true cost for the modeled country.
• Objective function
• Import cost is underestimated while the export revenue is overestimated
• Total welfare increase due to cross-border trade is allocated to the modeled country.
• Ex-post cost reallocation
• Traded electricity is valued at the locational electricity price (pay-as-cleared/marginal
pricing)
• Total welfare1
due to cross-border trade are split up in (i) profits for exporting
country, (ii) avoided costs for importing country and (iii) a congestion rent.
1
not including cost reductions due to more efficient investment decisions
10 / 17

22. Content
• Introduction
• Methodology
• Illustration
• Wrap-up
11 / 17

23. 2-country example
A (BE) B (NL)
Static greenfield optimization for different RES shares
Three cases
• A+B (co-optimized case)
• A (isolated)
• A + import/export curves
12 / 17

24. Performance of methodology
RES 50 A + B A (isolated) A + Import/Export curves
Cost for country A
(no congestion rents included)
[billion EUR]
5.854 5.917 5.854
Error [%] - + 1.06 + 0.0
Computation time [s] 22.23 2.25 4.31
If the import/export curves are constructed based on the optimal capacity mix for country B.
• We get the same solution for country A as would be obtained in the multi-country
optimization.
• The computation time can be reduced substantially
This method reduces the cross-border trade issue to making accurate exogenous
assumptions about the power system in the neigbouring countries (without the need for
co-optimization).
13 / 17

25. Ex-post cost reallocation
The objective function overestimates the
cost reductions due to cross-border trade.
We need to compensate the objective value for:
1 The congestion rent in case of congested
transmission line.
2 Profits and avoided costs that are actually
contributing to welfare increases in
neigbouring countries. 0 10 20 30 40 50 60 70
RES share [%]
0
1
2
3
4
5
6
7
8
Totalcostreduction[%]
Objective value
Market-based cost allocation
14 / 17

26. Wrap-up
Limiting the geographical scope of long-term planning models requires correctly representing
cross-border trade.
The proposed methodology has the benefit of:
1 Reducing the cross-border trade issue to making (accurate) assumptions about the
surrounding power systems.
2 Correctly exogenizing the countries excluded from the scope of the model, hereby
improving computational tractability.
There is a need to reallocate country-specific costs.
Future work:
• Perform a proper case study focusing on CWE-system
• Perform sensitivity analyses with Belgian TIMES model.
• Include stochasticity in current approach
15 / 17

27. References
[1] Devogelaer, D., Duerinck, J., Gusbin, D., Marenne, Y., Nijs, W., Orsini, M., & Pairon, M. (2012). Towards
100% renewable energy in Belgium by 2050. Belgium: FPB, ICEDD, VITO, 156.
[2] Balyk, O., Andersen, K. S., Dockweiler, S., Gargiulo, M., Karlsson, K., Næraa, R., Petrovic, S., Tattini, J.,
Termansen, L. B., & Venturini, G. (2019). TIMES-DK: technology-rich multi-sectoral optimisation model of the
Danish energy system. Energy Strategy Reviews, 23, 13-22.
[3] Poncelet, K., Delarue, E., Six, D., Duerinck, J., & D’haeseleer, W. (2016). Impact of the level of temporal
and operational detail in energy-system planning models. Applied Energy, 162, 631-643.
[4] Krishnan, V., & Cole, W. (2016). Evaluating the value of high spatial resolution in national capacity
expansion models using ReEDS. In 2016 IEEE Power and Energy Society General Meeting (PESGM) (pp. 1-5).
IEEE Power and Energy Society General Meeting (PESGM).
[5] Mai, T., Drury, E., Eurek, K., Bodington, N., Lopez, A., & Perry, A. (2013). Resource planning model: an
integrated resource planning and dispatch tool for regional electric systems (No. NREL/TP-6A20-56723).
National Renewable Energy Lab.(NREL), Golden, CO (United States).
[6] Georgiou, P. N. (2016). A bottom-up optimization model for the long-term energy planning of the Greek
power supply sector integrating mainland and insular electric systems. Computers & Operations Research, 66,
292-312.
[7] Biggar, D. R., & Hesamzadeh, M. R. (2014). The economics of electricity markets. John Wiley & Sons.
16 / 17

28. Total welfare increase due to cross-border trade
Welfare gains due to CBT increase for
increasing RES shares:
• More efficient RES investments - RES
potential varies geographically
• Smoothing of variability - correlation
effect of generation profiles and demand
profiles
Total welfare gains are split between:
• Welfare increase for country A
• Welfare increase for country B
• Congestion rent
0 10 20 30 40 50 60 70
RES share [%]
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Totalwelfareincrease[%]
Total
Welfare increase allocated to A
Welfare increase allocated to B
Congestion rent
17 / 17