A Multi‐Regional Medium‐ to Long‐Term Wind/PV Scenario Generation Method in Provincial Power Grids Considering Spatiotemporal Correlations
IET Generation, Transmission & Distribution · 2026-01 · 20 (1)
DOI: 10.1049/gtd2.70496 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Generation, Transmission & Distribution, online erschienen am 07.11.2026. Verfasst von Xiong Wu und 4 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: State Grid Corporation of China.
Abstract (Originalsprache)
With the increasing penetration of wind and photovoltaic (PV) generation, accurately characterising medium‐ to long‐term uncertainty in renewable power output has become a key issue in power system planning. However, existing scenario generation methods over large geographical areas often fail to adequately capture spatiotemporal correlations. Accordingly, this paper proposes a hierarchical and regionalised framework for wind and PV power output scenario generation considering spatiotemporal correlations. First, ERA5‐based meteorological features are clustered into provincial representative day types using K‐medoids. Then, a spatially representative regional partition is established according to the day‐type time series at each geographical location. On this basis, the generation of multi‐regional day‐type sequences is formulated as a combinatorial optimisation problem with temporal transition consistency and spatial correlation consistency constraints, which is solved by an improved simulated annealing algorithm to obtain multi‐regional typical‐day output labels. Finally, a Wasserstein‐distance‐based conditional generative adversarial network with gradient penalty (WCGAN) is used to generate high‐resolution intraday output profiles for typical days, thereby producing complete medium‐ to long‐term wind and PV power output scenarios. Case studies verify the proposed method's effectiveness in preserving spatial correlations and temporal evolution consistency, demonstrating its potential for planning large‐scale multi‐regional power systems with high renewable penetration.
Förderung
- State Grid Corporation of China
Zitieren
APA
Wu, X., Hao, Y., Jiang, C., Wu, C., & Song, C. (2026). A Multi‐Regional Medium‐ to Long‐Term Wind/PV Scenario Generation Method in Provincial Power Grids Considering Spatiotemporal Correlations. IET Generation, Transmission & Distribution, 20(1). https://doi.org/10.1049/gtd2.70496
Vancouver
Wu X, Hao Y, Jiang C, Wu C, Song C. A Multi‐Regional Medium‐ to Long‐Term Wind/PV Scenario Generation Method in Provincial Power Grids Considering Spatiotemporal Correlations. IET Generation, Transmission & Distribution. 2026;20(1). doi:10.1049/gtd2.70496
BibTeX
@article{Wu2026,
author = {Wu, Xiong and Hao, Yinan and Jiang, Chenlu and Wu, Cong and Song, Changcheng},
title = {A Multi‐Regional Medium‐ to Long‐Term Wind/PV Scenario Generation Method in Provincial Power Grids Considering Spatiotemporal Correlations},
journal = {IET Generation, Transmission & Distribution},
year = {2026},
volume = {20},
number = {1},
doi = {10.1049/gtd2.70496}
}
EUVIMED