关键词: 泵阀并联系统, 单神经元, RBF 神经网络, 滑模控制, 智能控制

Abstract: In light of the valve control system's advantages of fast response and high energy efficiency，a pumpvalve parallel electro-hydraulic position servo system was proposed，and its intelligent control method is studied． Firstly，a single neuron PID controller，which can realize the self-adaptive adjustment of the weight，is designed for the pump-controlled subsystem of the pump-valve parallel system． Secondly，a RBF ( Radial Basis Function) neural network-based sliding mode controller was designed for the valve-controlled subsystem of the pump-valve parallel system． The Lyapunov function was designed to prove the stability of the closed-loop system． Finally，the co-simulation was conducted by using Matlab /Simulink and AMESim． The co-simulation results show that the single neuron PID controller of the pump-controlled subsystem has better speed tracking control performance than the traditional PID controller． The RBF neural network-based sliding mode controller of the valve-controlled subsystem has better position tracking accuracy and stronger anti-interference ability than the traditional PID controller． The proposed intelligent control method can significantly improve the control performance of the pump-valve parallel system.

Key words: pump-valve parallel system, single neuron, RBF neural network, sliding mode control, intelligent control