Photovoltaic Power Prediction Based on RBF Neural Network Optimized by Gray Wolf Algorithm

Abstract

With the vigorous development of photovoltaic power generation, the prediction accuracy of photovoltaic power generation output is getting higher and higher. Therefore, this paper proposes a photovoltaic power prediction model based on the gray wolf algorithm optimized RBF neural network. First, select the similar daily sample data through the gray correlation analysis method; then use the gray wolf algorithm to globally optimize the number of hidden layer nodes of the RBF neural network, and use the global optimal solution obtained by the gray wolf algorithm as the parameter of the RBF neural network. Finally, predict through the optimized RBF neural network. The simulation results show that the RBF neural network optimized by the gray wolf algorithm has greatly improved the prediction accuracy, and has certain significance for the actual photovoltaic power prediction.