EUVIMEDEUVIMED entdecken
Crossref

Short-Term Wind Power Prediction Based on TimeGAN-SHAP Fusion Under Extreme Weather Conditions

Ran Wang, Haibin Li

Advances in Engineering Technology Research · 24.03.2026 · 16 (1) · 856

DOI: 10.56028/aetr.16.1.856.2026 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advances in Engineering Technology Research, online erschienen am 24.03.2026. Verfasst von Ran Wang und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Forecasting wind power is essential for effective power system management and the integration of renewable energy sources. Nevertheless, the accuracy of traditional forecasting methods is significantly impaired by unusual data distributions and complex feature interrelations caused by severe weather conditions. In response to this issue, this research presents a short-term wind power forecasting method that combines Temporal Generative Adversarial Network (TimeGAN) with SHapley Additive exPlanations (SHAP). In this approach, TimeGAN is used to produce high-quality synthetic wind power data to enhance the volume of training samples, while SHAP serves to evaluate feature significance, enabling the identification of key influencing factors and the optimization of model inputs. Experimental findings indicate that the introduced method achieves a reduction in Mean Squared Error (MSE) exceeding 15% when compared to conventional models during extreme weather events. Additionally, it effectively identifies the main factors influencing wind power generation. By striking a balance between prediction accuracy and the interpretability of the model, this approach offers an innovative technical solution for forecasting wind power under extreme weather, providing crucial support for the secure and stable functioning of power systems characterized by high wind energy integration.

Zitieren

APA

Wang, R., & Li, H. (2026). Short-Term Wind Power Prediction Based on TimeGAN-SHAP Fusion Under Extreme Weather Conditions. Advances in Engineering Technology Research, 16(1), 856. https://doi.org/10.56028/aetr.16.1.856.2026

Vancouver

Wang R, Li H. Short-Term Wind Power Prediction Based on TimeGAN-SHAP Fusion Under Extreme Weather Conditions. Advances in Engineering Technology Research. 2026;16(1):856. doi:10.56028/aetr.16.1.856.2026

BibTeX
@article{Wang2026,
  author = {Wang, Ran and Li, Haibin},
  title = {Short-Term Wind Power Prediction Based on TimeGAN-SHAP Fusion Under Extreme Weather Conditions},
  journal = {Advances in Engineering Technology Research},
  year = {2026},
  volume = {16},
  number = {1},
  pages = {856},
  doi = {10.56028/aetr.16.1.856.2026}
}
Kostenlos starten →