Ultra-Short-Term Wind Power Forecasting via a Dish-CONET-TimesNet Hybrid Model

View at publisher Summarise with VIRAT
Authors: Haowen Zheng
Cited by: 0
References: 10

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Accurate ultra-short-term wind power forecasting plays a critical role in maintaining the secure and economic operation of modern power systems. However, wind power time series usually exhibit strong non-stationarity, distribution shift, and multi-periodic characteristics, which significantly degrade forecasting accuracy. To address these challenges, this paper proposes a hybrid ultra-short-term wind power forecasting model based on Dish-CONET and TimesNet. First, Dish-CONET is employed to decompose wind power sequences and normalize distribution-shifted components. Then, the processed sequences are fed into TimesNet to capture multi-periodic temporal dependencies through two-dimensional variation modeling. Finally, the forecasting results are reconstructed via inverse normalization. Case studies based on real wind farm data demonstrate that the proposed model outperforms several benchmark methods in terms of prediction accuracy and robustness.

Fields of Study & Subjects

Energy Load and Power Forecasting Solar Radiation and Photovoltaics Time Series Analysis and Forecasting

Publication details

Journal / source
International Conference on Electrical Automation and Artificial Intelligence (ICEAAI)
Pages
318–322
Publication date
16 January 2026
Language
EN