Adaptive Neural Network Fault Tolerant Control for a Hydraulic System

Abstract

An adaptive control allocation method based on RBF neural network is proposed for a class of hydraulic systems. The goal of this method is to eliminate the output error. Firstly, the actuator fault model of hydraulic systems is established based on the Lyapunov stability principle, an improved adaptive neural network fault-tolerant controller is designed. The simulation results show that the controller has better control performance in the case of partial or combined faults in hydraulic system, can quickly implement fault-tolerant control.