Short-Term Prediction of Photovoltaic Power Generation Based on Deep Belief Network with Momentum Factor

Abstract

Accurate prediction of short-term photovoltaic power generation will help the grid dispatching department to make reasonable arrangements to ensure the safe and stable operation of the power system, thereby increasing the proportion of new energy in the power system. In order to improve the accuracy of photovoltaic power generation prediction, a short-term photovoltaic power generation prediction model based on deep belief network with momentum factor is proposed. It focuses on the selection of training samples for short-term prediction models of photovoltaic power generation, short-term prediction models of photovoltaic power generation based on deep belief networks, and optimization of the model by adding dynamically adjusted momentum factors. The actual measured data of the Australian Desert Solar Research Center were used to verify the short-term prediction model of photovoltaic power generation. The simulation experiment results show that the short-term prediction model of photovoltaic power generation established in this paper can effectively characterize the various factors that affect photovoltaic power generation and the measured the complex non-linear relationship between powers has high prediction accuracy.