Abstract: This paper proposes a maximum power point tracking algorithm that combines genetic algorithm and GRU neural network（GA-GRU-MPPT）to address the problem of photovoltaic power generation systems struggling to maintain maximum power point output in the face of rapidly changing external environmental factors. Based on the constructed maximum power point prediction model,this algorithm optimizes the parameters of the GRU neural network using genetic algorithm. Considering the correlation of the data,the previous moment’s solar cell temperature,solar irradiance,maximum power point voltage,as well as the current moment’s solar cell temperature and solar irradiance,are taken as input variables for the prediction model,with the output being the current moment’s maximum power point voltage. Simulation results for three different climate scenarios show that the tracking accuracy of this algorithm can reach99%,significantly improving the energy conversion efficiency of solar photovoltaic systems.