Sensor Fault Detection and Isolation of Electromechanical Actuator Based on Structural Residual Parity Space

Abstract

Aiming at the current popular fault methods’ disadvantages on diagnosing sensor faults existing in electromechanical actuator, this paper presents a fault diagnosis approach based on dynamic parity space. Firstly, we derive the mathematical model of EMA and establish the state space equation. Secondly, a sensor fault diagnosis approach based on structural residual parity space is introduced. Lastly, the sensor fault such as current sensor fault, motor rotary encoder fault and LVDT fault is detected and isolated based on structural residual, which prove that the proposed fault diagnosis approach is effective in diagnosing the sensor fault of electromechanical actuator.