Adaptive fault diagnosis for actuator systems based on observer parameters estimation algorithm

An adaptive high-gain observer is presented in this paper for actuator faults dynamic systems. In order to achieve the fault diagnosis and identification by state and fault parameters estimation of the actuator systems, an adaptive observer scheme incorporated with the method of high-gain observer is proposed. For the sake of covering the shortage in diagnosing the fault of actuator lock-in-place, a threshold function is introduced for modifying the designed adaptive observer. Simulation studies verify the effectiveness of the proposed approach.