A Combined Short‐Term Wind Power Forecasting Method Considering Multi‐Timescale Icing Hybrid Features
IET Renewable Power Generation · 2026-01 · 20 (1)
DOI: 10.1049/rpg2.70321 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Renewable Power Generation, online erschienen am 31.07.2026. Verfasst von Wenjie Ye und 2 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Icing disasters pose a great challenge to the safe operation of wind power and the new type of power system with wind power as the main body. Therefore, this paper proposes a combined short‐term wind power forecasting method considering multi‐time‐scale icing hybrid features. Firstly, the time‐series generative adversarial network (TimeGAN) algorithm is used to generate samples of meteorological data and power loss under icing conditions. Then, based on the physical mechanism of icing, multi‐time‐scale icing hybrid features that fully integrate the parameters of the wind turbine model and meteorological parameters are extracted. Furthermore, the main hyperparameters of the eXtreme Gradient Boosting (XGBoost) model are optimised and selected by the improved particle swarm optimisation (IPSO) to establish a wind power icing loss prediction model. Finally, the stacking ensemble learning is used to construct the optimal combination of multiple algorithms based on normal weather samples, and the icing loss period power prediction is accurately obtained in combination with the wind power icing loss prediction model. The experimental results prove that the method proposed in this paper can effectively improve the accuracy of wind power prediction under icing disasters.
Zitieren
APA
Ye, W., Dai, Z., & Tu, M. (2026). A Combined Short‐Term Wind Power Forecasting Method Considering Multi‐Timescale Icing Hybrid Features. IET Renewable Power Generation, 20(1). https://doi.org/10.1049/rpg2.70321
Vancouver
Ye W, Dai Z, Tu M. A Combined Short‐Term Wind Power Forecasting Method Considering Multi‐Timescale Icing Hybrid Features. IET Renewable Power Generation. 2026;20(1). doi:10.1049/rpg2.70321
BibTeX
@article{Ye2026,
author = {Ye, Wenjie and Dai, Zemei and Tu, Mengfu},
title = {A Combined Short‐Term Wind Power Forecasting Method Considering Multi‐Timescale Icing Hybrid Features},
journal = {IET Renewable Power Generation},
year = {2026},
volume = {20},
number = {1},
doi = {10.1049/rpg2.70321}
}
EUVIMED