Objective In practical applications the sunlight received by photovoltaic cells may be obstructed causing photovoltaic arrays to operate under partial shading conditions resulting in the occurrence of multiple peaks in the output power of the photovoltaic system. In response to the weak global tracking ability of traditional maximum power point tracking MPPT algorithms and the difficulty in balancing tracking speed and accuracy a combined algorithm of hunter prey optimization HPO and improved perturbation and observation IP&O was proposed. Methods Firstly the HPO algorithm was used to initialize the population and obtain the initial maximum power of the photovoltaic system which was defined as the global optimum of the entire algorithm. Then each update of the output power was compared with the optimal value and the larger power was retained as the global optimum. When the algorithm switching conditions were met the HPO algorithm was considered to have tracked near the maximum power point and the IP&O algorithm was perturbed near the maximum power point until the optimal value was output. When the illumination received by the photovoltaic system changed the HPO-IP&O algorithm could be quickly restarted by restarting conditions. Results To verify the reliability of the proposed algorithm a photovoltaic system simulation model was established in MATLAB / Simulink. The particle swarm optimization PSO whale optimization algorithm WOA and the proposed algorithm were compared under different illumination conditions. Simulation results showed that the HPO-IP&O algorithm did not fall into local extremes and its tracking accuracy was better than that of the WOA algorithm but it was not significantly different from the PSO algorithm. In terms of tracking speed the HPO-IP&O algorithm was faster than the WOA and PSO algorithms and the larger the power the shorter the time used by HPO-IP&O. Conclusion The HPO-IP&O algorithm solves the problems of traditional MPPT algorithms easily falling into local optimum and being unable to balance tracking speed and accuracy. The feasibility and reliability of the proposed algorithm under different lighting conditions are verified by simulation experiments.