Simulation Study of Bidirectional DC/DC Converter for New Energy Vehicles Based on RBF Neural Variable Structure

To make the bidirectional DC/DC converter have better performance and stability, this paper presents a RBF neural network-based variable structure control approach. Mathematically describe the working principle of bi-directional DC/DC converter and construct a variable structure controller using RBF neural network according to the design requirements. The variable structure control is optimized by the RBF neural network algorithm and the chattering phenomenon in the variable structure control is reduced. The simulation model is built and simulated in MATLAB/Simulink, and results of the simulation show that the dynamic response performance and robustness of the control system are improved after using the RBF neural variable structure control optimized voltage controller.