Abstract:  The complex electro-hydraulic servo control system (hydraulicdriven control system) using ordinary PID control has the problem of poor control flexibility and could not achieve the ideal control effect. In order to improve the control characteristics of the electro-hydraulic servo system, a method based on radial basis function neural network (RBF) fuzzy PID control strategy was proposed. Firstly, the state space equation of servo valve controlled hydraulic cylinder was deduced theoretically, and the digital model of hydraulic system was established. Then a fuzzy PID control strategy based on readial basis function (RBF) was designed on the basis of common PID control strategy. The fuzzy control rules were adjusted according to the characteristics of electro-hydraulic servo system. Finally, MATLAB/Simulink simulation was carried out under no-load and load conditions of the electro-hydraulic servo system. The characteristics of the electro-hydraulic servo system based on different control strategies were compared and analyzed, and the superiority of the fuzzy PID control strategy based on radial basis unction (RBF) neural network was verified. The results show that the response speed of ordinary PID control and fuzzy PID control is more than 10s under no-load condition, and the overshoot is large. Moreover, it takes a long time to adjust after loading, and the recovery ability for load interference is poor. The control response speed of RBF neural network fuzzy PID control in noload condition is only 4.23s, the overshoot was reduced to 4.16%; after loading, return to stable state after 2.56s. RBF neural network fuzzy PID control strategy has better antiinterference and strong robustness. It can better meet the control requirements of industrial robot electro-hydraulic servo system.

Key words:  electro-hydraulic servo system; readial basis function (RBF) neural network; fuzzy PID control; MATLAB/Simulink simulation

ZHAO Yan, ZHOU Qin-yuan, SHAO Nian-feng, et al. Electro-hydraulic servo system based on RBF neural network fuzzy PID control［J］.Journal of Mechanical & Electrical Engineering, 2022,39(2):244-249.