A Vienna rectifier is used as the research object, and an adaptive voltage outer loop sliding mode control algorithm based on the approximation rate is analyzed for its traditional voltage outer loop sliding mode variable structure control invariance and sensitivity to system parameter perturbation. By effectively combining the RBF neural network with the sliding mode control algorithm, the algorithm also adds the midpoint potential balance control to the design of the RBF neural network adaptive voltage outer-loop sliding mode control algorithm. It uses the RBF neural network for adaptive approximation of the voltage outer-loop nonlinear system. This can effectively reduce the switching gain, weaken the jitter and enhance the anti-interference capability of the system. Lastly, simulation analysis and experimental tests are conducted to verify the effectiveness of the proposed control algorithm. The algorithm is compared with the traditional sliding mode control algorithm and the PI control algorithm, and the results show that the use of this voltage external loop control algorithm can provide fast tracking of the target value of the DC output voltage and balanced midpoint potential. This improves the dynamic and static performance of the system and enhances its anti-interference capability. This work is supported by the National Natural Science Foundation of China (No. 51777120).