Coordination of preventive, emergency and restorative trading strategies under uncertain sequential extreme weather events
Sequential extreme weather events (SEWEs), such as hurricanes and tropical storms occurring in succession, pose significant challenges to local energy (LE) and flexibility (LF) markets. Effective coordination of preventive, emergency, and restorative strategies can mitigate losses during these event...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-04-01
|
| Series: | International Journal of Electrical Power & Energy Systems |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0142061525000511 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Sequential extreme weather events (SEWEs), such as hurricanes and tropical storms occurring in succession, pose significant challenges to local energy (LE) and flexibility (LF) markets. Effective coordination of preventive, emergency, and restorative strategies can mitigate losses during these events, but designing optimal trading strategies for joint LE and LF markets remains complex. This paper introduces a novel trading method to address this challenge. First, a two-layer graph neural network (GNN) is employed to predict the probability distribution of system outages caused by SEWEs. Then, a joint LE and LF market transaction model is developed to optimize multi-stage trading and minimize overall losses throughout SEWEs. To address the uncertainty of SEWEs, a probability forecast-driven distributionally robust joint chance constraint (DRJCC) optimization method is proposed, which is efficiently solvable as a convex conic problem. Finally, case studies conducted on modified IEEE 141-bus and 300-bus systems validate the approach, showing reductions in load shedding and trading costs by up to 15.39% and 42.88%, respectively, compared to single-stage or two-stage strategies. Additionally, the two-layer GNN model achieves a root mean square error of 0.01, demonstrating high accuracy in predicting system outage statuses. |
|---|---|
| ISSN: | 0142-0615 |