Fault diagnosability qualitative analysis of spacecraft based on temporal fault signature matrix

Fault diagnosability analysis is an important foundation for spacecraft fault diagnosis. Combining signed directed graphs (SDG) with fault temporal information, an improved diagnosability evaluation model, temporal fault signature matrix (TFSM), is proposed in this paper. Multiple-valued evaluation of measuring variables is used in the signature matrix to represent different symptoms and time sequences of several typical faults, which is helpful to distinguish certain faults. Finally, the method is applied to fault diagnosability analysis of spacecraft propulsion system, and the simulation results verify the validity and superiority of the proposed evaluation model.