Short-term photovoltaic power prediction based on T-S fuzzy neural network

Abstract

Due to the meteorological factors, such as weather conditions, irradiance, ambient temperature and wind speed, as well as the non-meteorological factors, such as the temperature and installation location of components and parts, the output power of photovoltaic power generation system is characterized by strong intermittency, volatility and uncertainty, and low forecast accuracy of photovoltaic power generation. Based on the historical power generation data and the actual meteorological data of the photovoltaic system, the short-term prediction of the photovoltaic power generation is carried out by using the T-S fuzzy neural network prediction model. Finally, the prediction results of T-S neural network and traditional BP neural network are compared. The results show that the prediction accuracy of PV output power is improved by using the T-S fuzzy neural network prediction method. The average error percentage between the predicted result and the measured value is controlled within 8%. The algorithm can be effectively used for short-term power forecasting of photovoltaic systems.