Improved CNN-BIGRU Algorithm and its Application based on Self-Attention Mechanism

To enhance the accuracy of wind farm power generation forecasting, optimize energy utilization and reduce system maintenance costs, this paper proposes a wind power prediction framework that integrates convolutional neural networks, bidirectional gated recurrent units, and a self-attention mechanism. Firstly, the convolutional neural network is employed to extract the spatial features in the historical data of the wind farm. These features are then fed into the bidirectional gated recurrent unit network for time series learning, so as to further improve the ability to capture time series changes, and finally the self-attention mechanism weights the model's output, and automatically pay attention to the most important spatiotemporal features in the input data, thereby improving the overall prediction accuracy. The experimental results demonstrate that the proposed model achieves superior forecasting performance, significantly enhancing the precision of wind power prediction.