A multi-source information fusion fault diagnosis for aviation hydraulic pump based on the new evidence similarity distance

Aviation hydraulic pump is one of the key components in aircraft hydraulic system, therefore, high-precision fault diagnosis is essential to improve the reliability and performance of hydraulic pump. A novel multi-source information fusion fault diagnosis method is proposed based on the Dempster–Shafer (D–S) evidence theory, which utilizes the three-level signals from pump level, hydraulic power system level and hydraulic actuation system level. The feature vectors of these three levels are extracted as three bodies of evidences (BOEs) and the fuzzy membership function is employed to construct the basic probability assignments (BPAs) of three BOEs. In order to solve the issue of combining the conflicting evidences, the D–S evidence theory based on the new evidence similarity distance is developed to combine the obtained BPAs. Finally, the making-decision rules are given to diagnose the faults. The diagnosis results validate that the proposed method not only can increase significantly the belief level of supporting the diagnosis target, but also has the ability to diagnose fault of pump correctly even if a sensor is faulty.