结合贝叶斯优化及通道注意力的双端优化时序式风功率预测模型OA北大核心CSTPCD

Aimed at the problem of a lack of data-side parameter optimization and model-side structural optimization in the existing wind power time-series prediction model,a double-side optimized time-series wind power prediction model is proposed in this paper. First,Bayesian optimization is used to efficiently search and optimize the data-side parame-ters. Then,channel attention and convolutional neural network are used to construct a feature extraction module to en-hance the learning of the importance of input influencing factors by the model. Finally,the extracted features are accu-rately fitted using a bi-directional long short-term memory model. Results show that the proposed model can capture thechanging trend of wind power under different prediction scenarios and significantly improve the prediction accuracy.