A GVSAO-Optimized CNN-BiGRU-Attention Network for Short-Term Wind Power Forecasting
- Authors
-
-
Qing Lv
-
Xinyu Zhai
-
Rundong Zhou
-
Qiang Li
-
Xiancong Wu
-
- Abstract
- This paper suggests a deep learning model for predicting wind in the short term by combining the Good Point and Vibration Snow Ablation Optimizer (GVSAO), Convolutional Neural Network (CNN), Bidirectional Gated Recurrent Unit (BiGRU), and an attention mechanism. The convolutional neural network (CNN) is employed to capture spatiotemporal feature representations from the time series data, while the bidirectional gated recurrent unit (BiGRU) effectively learns relevant temporal dependencies through forward and backward propagation. To further enhance the representational capacity of both models, an attention mechanism is introduced to highlight critical patterns within the time series, thereby guiding the CNN and BiGRU to focus on the most informative features during the learning process. In addition, the GVSAO algorithm is employed for optimal parameter selection to enhance the accuracy of the model. Experimental verification demonstrates that this model achieved a MAPE of 0.11725, an RMSE of 0.73206, and an R2 of 0.9255. These results demonstrate that the proposed model surpasses existing mainstream approaches and further confirm its superiority in predictive performance.
- References
- Cover Image
-
- Downloads
- Published
- 2026-09-29
- Issue
- Vol. 28 No. 3 (2026)
- Section
- Articles
- License
-
Copyright (c) 2026 Journal of Control Engineering and Applied Informatics
This work is licensed under a Creative Commons Attribution 4.0 International License.
