To address the issue of difficult and accurate tracking of the maximum power point (MPP) under partial shading in photovoltaic power generation, a control method combining grasshopper optimization algorithm (GOA) based on Levy flight and perturbation observation method (particle swarm optimization, P&O) is proposed in this study. The grasshopper optimization algorithm combines Levy flight and grasshopper behavior rules, allowing the algorithm to have good convergence and quickly find the optimal solution. When the improved GOA searches near the maximum power point of photovoltaic power generation, rapid switching of P&O can be achieved, effectively improving the efficiency of photovoltaic power generation. To test the superior tracking capability of the composite photovoltaic optimization algorithm proposed in this study, static and dynamic shading simulations of photovoltaic energy storage are carried out in Simulink, and compared with algorithms proposed in literature. The research results demonstrate that the composite algorithm proposed in this study effectively addresses the shortcomings of traditional photovoltaic MPP algorithms, such as easily getting stuck in local optimal solutions and slow convergence speed, and has practical value in photovoltaic storage systems.