Abstract:In order to improve the prediction accuracy of photovoltaic power generation, a prediction model of photovoltaic power generation based on improved BP neural network is proposed. First, such factors as outdoor temperature, light radiation, wind speed and other factors are taken as input layer nodes while AC power is taken as output nodes, RMSE is introduced as indicators to measure the optimal model to determine number of hidden layer nodes, and then BP neural network is used to learn which cuckoo search algorithm is used to optimize BP neural network. Finally, the simulation experiment is used to test its effectiveness. The results show that improved neural network can improve prediction accuracy of photovoltaic power generation, and it has a widespread value.