Abstract: When the surface illumination of the photovoltaic array is nonuniform, its power characteristic curve appears multi-peak phenomenon, which increases the complexity of the maximum power point tracking(MPPT) process. Traditional MPPT method cannot track the global maximum power point(GMPP). Although the artificial intelligence algorithm can solve the photovoltaic MPPT problem in the case of multiple peaks to a certain extent, the implementation process of this kind of algorithm is generally complicated and has certain limitations. In this paper, a photovoltaic global maximum power tracking control strategy for non-uniform lighting conditions is studied. Global maximum power point tracking method for photovoltaic system based on virtual load characteristic disturbance is proposed. The global search mechanism is introduced into the traditional perturbation observation method. When the illumination on the PV array surface is uneven, the location area of the maximum power point on the I-U characteristic curve of the PV array is initially located by the virtual impedance characteristics, and then the traditional perturbation observation method is used to accurately track, which improves the applicability of the traditional MPPT control method. The proposed control method can be used to accurately search the global maximum power point of the PV array, and the dynamic performance is good. Simulation and experiment results testify the feasibility and effectiveness of the proposed algorithm.