Federated Learning-Based Adaptive Multi-Head Attention Model for Wind Power Forecasting
Big Data and Cognitive Computing · 07.05.2026 · 10 (5) · 147
DOI: 10.3390/bdcc10050147 ↗Kurz gefasst
Fachartikel in der Zeitschrift Big Data and Cognitive Computing, online erschienen am 07.05.2026. Verfasst von Yihua Zhu und 6 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National Key Research and Development Program of 657 China.
Abstract (Originalsprache)
Enhancing the accuracy of short-term wind power forecasting helps mitigate the adverse impacts of prediction errors on grid dispatch. Wind power exhibits a significantly nonlinear dependence on multiple influencing factors. However, existing methods struggle to effectively resolve multi-dimensional feature redundancy and multi-scale non-stationary evolutionary characteristics inherent in far-offshore wind power forecasting tasks. This leads to bottlenecks such as insufficient feature discriminability and temporal dependency focus shift under complex marine environments, ultimately limiting further improvements in prediction accuracy. To address these challenges, this paper proposes a federated learning-based adaptive multi-head attention model for wind power forecasting (Fed-AMHA). The proposed framework operates as follows: First, each wind farm client utilizes a Bidirectional Long Short-Term Memory (BiLSTM) network to model input sequences bidirectionally, capturing long-term temporal dependencies. Subsequently, linear projection and parallel one-dimensional convolution operations are introduced to mine multi-scale local temporal features from each time step and its neighborhood. Building upon this, channel attention and multi-head temporal feature attention mechanisms are stacked. The model adaptively adjusts the weights of different time slices and feature channels by learning the importance of each channel to the forecasting task. The central server then aggregates the model parameters uploaded by the clients via averaging, enabling cross-site collaborative training without directly sharing raw data. Simulation results based on public datasets and actual wind farm data under various short-term forecasting scenarios demonstrate that the proposed model consistently achieves lower prediction errors and superior stability compared to existing forecasting models under identical settings.
Förderung
- the National Key Research and Development Program of 657 China
Zitieren
APA
Zhu, Y., Luo, C., Wu, K., Yu, J., Hu, B., Huang, L., & Xiao, B. (2026). Federated Learning-Based Adaptive Multi-Head Attention Model for Wind Power Forecasting. Big Data and Cognitive Computing, 10(5), 147. https://doi.org/10.3390/bdcc10050147
Vancouver
Zhu Y, Luo C, Wu K, Yu J, Hu B, Huang L, et al. Federated Learning-Based Adaptive Multi-Head Attention Model for Wind Power Forecasting. Big Data and Cognitive Computing. 2026;10(5):147. doi:10.3390/bdcc10050147
BibTeX
@article{Zhu2026,
author = {Zhu, Yihua and Luo, Chao and Wu, Ke and Yu, Jiawei and Hu, Binjiang and Huang, Lei and Xiao, Bitao},
title = {Federated Learning-Based Adaptive Multi-Head Attention Model for Wind Power Forecasting},
journal = {Big Data and Cognitive Computing},
year = {2026},
volume = {10},
number = {5},
pages = {147},
doi = {10.3390/bdcc10050147}
}
EUVIMED