18/09/2025
Published in the CSEE Journal of Power and Energy Systems, this paper introduces an innovative two-stage approach to improve ultra-short-term wind power forecasting accuracy—an essential challenge in smart grids with high wind pe*******on. The authors propose a decomposition-based strategy that first separates the initial wind power data into trend, fluctuation, and residual components, each modeled with tailored prediction algorithms. In the subsequent correction phase, residual errors from the initial prediction are further refined using persistence-based correction techniques. The study conducts both deterministic and probabilistic evaluations of the proposed method and benchmarks its performance against existing models. Simulation results show that the approach significantly reduces prediction errors, offering a practical and effective solution for managing wind power variability in grid operations.
Coauthored by Peng Lu; Zhuo Li; Lin Ye; Ming Pei; Yingying Zheng; Yongning Zhao from China Agricultural University, Beijing, China.
You can check on the whole article on IEEE Xplore by:
https://ieeexplore.ieee.org/document/10520160
All CSEE JPES articles are OA published on IEEE Xplore.
Highly reliable wind power prediction is feasible and promising for smart grids integrated with large amounts of wind power. However, the strong fluctuation fea