Fault Detection of Simulated Servo Hydraulic System by Utilizing Artificial Neural Network (ANN)

Maintenance reliability and efficiency in industrial hydraulic systemsoperation has become a point ofconcern for the professionals in maintenanceengineering. one practical approach in this regard is the realization of symptomsof early stage malfunctioning in fluid power systems after which maintenanceplanning and preventive means would follow upon a reasonably accurate andsubsequently acceptable determination. among the highly reliable sourcesproviding such convenience, artificial neural network (ann) stands ahigh chanceof success neural network method has been used to detect faults occurring in mosthydraulic systems. these faults could be related to supply pressure, effective bulkmodulus and total leakage. the simulated system in this study consists of hydraulicservo valve, double acting cylinder and a spring that resists piston movement. twomain reasons causing this system to have a nonlinear behavior are hydraulic servovalve and compressibility effect of hydraulic fluid.the neural network approach inthis investigation comprises of an efficient use in nonlinear systems and requiresadvance knowledge about the system behavior under faulty conditions andassumptions about the type and severity of faults likely to occur. neural networkstrained with different training algorithms are investigated. after training thenetwork, the system was examined for different inputs and obtained results were compared.