Fault Detection in a Servo hydraulic System in Presence of White Noise, Using an Observer

Occurance of faults in hydraulic systems may cause serious problems in hydraulic machineries, especially where human life is involved. Therefore, online detection of faults is an important issue. In this article, occurance of fault has been detected wing a high-gain observer for an experimental servo hydraulic setup. State-space model of the dynamic system has been derived. Leunberger non-linear observer is designed for the servo hydraulic system. State parameters of the system, such as internal pressures, are estimated by the proposed observer. Observer’s gains and stability conditions are derived based on Lipshitz and Lyapanov methods, respectively. Fault detection has been carried out by multiplicate and residual functions. Using the proposed method, existance of leakage and trashes in the system can be detected in real time. Simulation results show that the application of Leunberger observer can reach a high performance fault detection method for servo hydraulic systems in presence of external noise.