Photovoltaic power generation prediction based on MEA-BP neural network

Based on the analysis of the affecting factors of the photovoltaic (PV) power generation, this paper presents a prediction model of PV power generation based on BP neural network optimized by the mind evolutionary algorithm (MEA). The model uses the MEA to optimize the weights and thresholds within the BP neural network, which overcomes the shortcomings of conventional learning algorithm which is easy to fall into the local optimal solution and has slow convergence rate. In this paper, the MEA-BP prediction model is compared with the traditional BP prediction model by using the actual PV output data. The results show that the proposed MEA-BP prediction model can predict the PV output more accurately and thus has potential practical value.