关键词: 动态因果图, 模糊推理, 故障诊断

Abstract: The faults can be quickly and reliably diagnosed is important in the operation and maintenance system when a complex control system fails. Aiming at the situation of complex control system with many control signals, strong signal correlation, many signal states and many component failure modes, this paper proposes a fault diagnosis method based on improved dynamic causal graph and fuzzy reasoning, which uses improved dynamic causal graph logic expression. Strong ability, can make multiple（positive, reverse, mixed） fuzzy rules based on the characteristics of causality, effectively overcome the problem that fuzzy logic reasoning can only be caused by the retrospective effect, and at the same time, the dynamic characteristics of causal graphs are introduced into the dynamic update of fuzzy rules, which enhances the real-time performance of fuzzy inference. Finally, the fault diagnosis of the vertical torque motor control process of a certain armored equipment is used as the application background, embedded in the self-developed fault diagnosis platform. This method is used for fault diagnosis test. The test results show that this method can effectively improve the diagnostic efficiency and has higher accuracy, advancement and applicability.

Key words: dynamic causality graph, fuzzy inference, fault diagnosis