Fault diagnosis of aerospace relays based on belief rule base

Abstract

The fault diagnosis of aerospace relays plays a key role in the service of aircraft by virtue of essentiality of them. The requirement of transparency is paramount to a fault diagnosis model. Belief rule base (BRB) is introduced to establish the fault diagnosis model for relays in an easy-to-understand way in this paper. To fully embed the expert knowledge, the fault mechanism of relay is analyzed to generate the failure mode and effects analysis (FMEA). Fault features, fault modes and belief rules are extracted from FMEA by experts to generate belief rules of BRB. The projection covariance matrix adaption evolution strategy (P-CMA-ES) algorithm is employed to reduce the model error caused by the subjectivity of experts. The fault diagnosis of JZX-22F relay is carried out to validate the effectiveness of the proposed method based on BRB and fault mechanism.