EXTRACTION METHOD OF DECISION RULES FOR FAULT DIAGNOSIS BASED ON ROUGH SET THEORY
Dai Li-zhou · Proceedings of the CSEE · 2003
In order to extract simple and effective diagnostic rales from inconsistent diagnostic information, an extraction method of decision rules for fault diagnosis based on rough set theory is proposed. According to the definition of fault diagnosis decision system, a decision network with different reduced levels is constructed to denote the fault diagnosis problems. The diagnostic decision rules at each node are deduced by introducing the coverage factor of diagnostic decision rules. When the decision network is applied to fault diagnosis of a new case, the diagnostic conclusion is obtained by matching the information of the new case with the diagnostic decision rules of the relative network nodes. Even when the information of fault diagnosis is incomplete, the diagnosis results are correct. The availability of this method is proved by a fault diagnosis example of reciprocating machines.