Abstract:

The traditional fault-diagnosing method cannot locate the fault due to the complexity in structure and operationenvironment of a complex system. In order to overcome this difficulty , the method of case-based reasoningand the Petri net theory are introduced in the fault diagnosis , and a CRPN (Case-based Reasoning Petri Nets) modelis established based on the two above-mentioned methods. The key problems of the CRPN model , such as thetranslation rules and the case characteristics , are then investigated. By using a two-layer CRPN model , the difficultyin modeling and the fuzzy in reasoning process during the fault diagnosis of a complex system can be effectivelyovercome. Finally , as an example , the proposed method is applied to a system of aircraft to reveal the process offault diagnosis based on the CRPN model. The results show that the proposed method is effective.

Key words: complex systems, fault diagnosis, case-based reasoning, Petri net