Fault diagnosis of a hydraulic actuator using neural network

This paper presents the failure analysis and fault simulation of the electro-hydraulic servo (EHS) system for the positioning of the control valves. An idea for the realisation of a fault diagnosis system is given, and one of the possible faults occurring in the electro hydraulic system is considered in detail. On the basis of the detailed mathematical model of the entire electro hydraulic system, model-based fault detection and isolation algorithm for the considered case is presented. The method of non-linear system identification using neural network models is applied for fault modelling. The algorithm has been demonstrated by simulation using the signals measured in the cases of occasional failure occurrence in that part of the system.