Fact description based on rough sets for expert system

Abstract

Fault diagnosis expert system is wide used in the maintenance application fields. And knowledge acquisition is the key technology of expert system development. For the equipment maintenance domain, systems typically require a very long period to acquire knowledge, and knowledge is not necessarily correct. To address this issue, we describe the relationship between rough set theory and rule-based description of equipment failures. Then the exclusive rules, inclusive rules and failure images of equipment are built based on the EMES (Equipment Maintenance Expert System) diagnosis model, and the definition of probability rule is put forward. Next, we present the rule-based automated induction reasoning method and resampling methods. We also introduce automated induction of the rule-based description, which is used in our EMES. Finally, the experimental results show our solution gives a very suitable framework to represent processes of uncertain knowledge extraction.