Short-term Wind Power Prediction Model Based on BOWA-MCNN-BIGRU-Attention
This project aims to establish a new short-term forecast model for wind power generation by combining multi-scale convolutional neural network and bidirectional selective neural network to address the scientific problem of drastic change of wind turbine power due to extreme climate in the current grid load forecasting.The"black spider"optimization method is used to realize the adaptive adjustment of the super-parameters of the network,which effectively overcomes the problem of the model's accuracy being affected by the severe meteorological environment.It is proved that this method has higher forecasting accuracy and better generalization compared with the traditional method.
万方数据
