In the case of local shadows, the power voltage (P-V) characteristic curve of photovoltaic (PV) arrays presents multiple extreme points, which is difficult to achieve target using traditional maximum power point tracking (MPPT) methods. Therefore, an improving cuckoo search with adaptive incremental conductivity (ICS-AINC) maximum power tracking algorithm based on the combination of the cuckoo algorithm and conductivity increment method is proposed. In the ICS algorithm, ’Levy flight and random walk preferred location updates adopt convergence factors to adaptively adjust its step size. Through the study of its boundary value conditions, when the cuckoo flies to the boundary value, dynamic boundary value processing is adopted to reduce its search times. When local optimization is carried out, the judgment conditions are switched to the adaptive conductance increment method for global maximum point tracking, and the maximum power point tracking speed is improved by using zoning variable step size, simultaneously utilizing incomplete partial differential theory to optimize steady-state oscillations. Finally, simulation was conducted to verify that the algorithm can quickly and accurately track the maximum power point with and without shadows, effectively improving the output efficiency of the photovoltaic array.