Research on Power Forecasting Model of Wave Energy Generation Based on GRU Neural Network

Wave energy has great fluctuation, which will seriously affect the normal operation of power system when the wave energy power generation system is connected to the grid. Accurate prediction of wave power generation plays an important role in real-time scheduling and control of power system. Aiming at improving the prediction accuracy of wave power generation, this paper takes F-HBA wave power generation device as the research object, and constructs a wave power generation prediction model based on GRU neural network. This section includes three sub-modules, namely, effective wave height prediction module based on GRU neural network, wave period prediction module based on GRU neural network and power conversion module based on F-HBA. Firstly, the wave period and effective wave height are predicted, and then the predicted values of effective wave height and wave period are input into the power conversion model to obtain the predicted wave power generation value. The prediction accuracy of wave power generation model based on GRU neural network is verified by simulation.