Abstract:Aimed at the problems of slow response speed and large output error of electrohydraulic servo valve control system, the control system was optimized by improved genetic algorithm, and the control effect was simulated and verified. The structure of a new type electrohydraulic servo valve was designed, the dynamic model of the electrohydraulic servo system was established, and the flow motion equation of the hydraulic cylinder was deduced. The improved genetic algorithm was used to optimize the structure of the RBF neural network, and the improved control system of doublestep motor servo valve was simulated and verified by MATLAB, and the control effect was compared with that of the traditional PID. The results show that in the noninterference environment,both the traditional PID control method and improved RBF neural network control method can improve the accuracy of piston rod displacement output,while in the environment with interference,the traditional PID control method has a larger error of piston rod displacement output,the improved RBF neural network control method has a smaller error of piston rod displacement output. The improved RBF neural network control method can suppress the external interference, so as to improve the response speed and output precision of the doublestep motor servo valve control system.