Hydraulic actuator circuit fault detection using extended Kalman filter

This paper presents the application of extended Kalman filter (EKF) to closely estimate the states of hydraulic systems. The EKF, working concurrently with the actual system, accepts the input voltage signal to the servovalve and the measurements of the cylinder chamber pressures, and then rebuilds the system states. The spool displacement of the servovalve, the chamber pressures of the actuator and the velocity of the ram are accurately estimated. By comparing the estimated states and the actual measurements, residual signals are generated and then analyzed to report the occurrence of faults. Experiments show that the EKF estimator has the ability to promptly and reliably respond to the changes in the pump pressure as little as 10% of the normal operating pressure.