Aiming at the problems of photovoltaic power point tracking (MPPT) control based on BP neural network, an improved neural network MPPT controller for photovoltaic power generation system is proposed. Firstly, the working principle and equivalent circuit of the battery in the solar photovoltaic power generation system are analyzed. Then, the combination of BP neural network and fuzzy control is used to realize photovoltaic MPPT control. In addition, the fuzzy controller is improved with adaptive parameters. Finally, based on the BOOST conversion circuit, the matlab/simulink simulation model of the photovoltaic cell was established. The experimental results show that the proposed maximum power point tracking controller has higher accuracy and response speed, and the stability is better.