Abstract: This paper proposed an adaptive digital dynamic surface intelligent control (DSC) algorithm for photovoltaic power generation servo system. By combining radial basis function (RBF) neural network technology, the "differential explosion" problem in traditional backstepping method was overcome and the special requirements for the structure of the control system were released. Therefore, the procedures of the controller design have become more concise and implementable. By quantizing the amplitude of control signal with modified hysteretic quantizer, the chattering phenomenon in logarithmic quantizer was reduced and the pure digital control objective was realized. The ultimately uniformly bounded property of the closed-loop system was proved by designing the Lyapunov functions. The experimental verification was carried out through the StarSim Modeling Tech simulation platform and photovoltaic servo system. The experimental results can demonstrate the feasibility of the proposed control scheme.