A Novel Method for Aircraft Actuator Fault Diagnosis Based on Causal Graph and T-S Fuzzy Model

Proper assessment of the health status of the flight control system actuators, a critical system of the aircraft, is a prerequisite for proper aircraft operation. Timely and accurately determining the status of aircraft actuators help reduce the impact of faults. Obtaining knowledge related to fault information from actuator operation data and applying fault knowledge for diagnostic decisions is a crucial step in determining actuator status. In this paper, a novel fault diagnosis method is proposed by combining the causal inference method with the T-S fuzzy model. The proposed method improves the diagnostic decision performance of actuator faults by improving the fault knowledge acquisition process. The causal inference approach helps to obtain causal graph expressing the relationship between aircraft actuator parameters and states from actuator operational data. A subset of parameters with strong correlation with actuator states can be filtered out. The T-S fuzzy model was used to extract fuzzy diagnostic rules. Experience based on actuator simulation data verifies that the model can be effectively adapted to the actuator’s diagnostic logic and is competent for actuator troubleshooting tasks.