Dual-Channel Wind Power Forecasting Model Using Squeeze and Excitation Network

Accurate wind power forecasting plays an increasingly important role in the field of wind power forecasting but still intractable in practice. In order to improve the accuracy of wind power forecasting, a novel model based on squeeze and excitation network (SENet) embedded in a dual-channel is proposed. The dual-channel mechanism integrates convolutional neural network and gated recurrent unit, both of which can be used without interfering with each other. SENet can increase the attention of each important features, thus improving the network efficiency. Compared with a single model, the proposed model has better feature extraction performance with the root mean square error of 29.9. The test results of wind power data based on Kaggle platform show that the proposed method outperforms several deep learning forecasting methods.