The power-voltage characteristic curve of photovoltaic system has multiple peaks under partial shade condition. The traditional maximum power tracking method can easily trace to the local maximum power point. To solve such shortcoming，a photovoltaic system Maximum Power Point Tracking（MPPT） algorithm based on adaptive radial basis function neural network is proposed. The model optimizes the extended constants and weights of RBF neural network with adaptive linear algorithm, which overcomes the shortcomings of traditional neural network algorithm with slow convergence speed and poor global optimization. The simulation of adaptive RBF neural network is carried out in MATLAB/Simulink environment. The results show that the proposed algorithm can accurately find the maximum power point of the photovoltaic system when the external illumination and temperature change. Moreover the convergence accuracy and convergence time are greatly improved.