Study on An Intelligence Integration Based Fault Diagnosis Method
On the basis of analyzing each advantages of man and computer in the fault diagnosis, the paper brought forward subject of intelligence integration based fault diagnosis, and then presented intelligent diagnostic system building principles, on which an intelligent integrated diagnostic model was given later. Afterward, concentrating on the three model formed modules, study on the knowledge representation and knowledge reasoning as well as knowledge acquisition was carried on. In the module of knowledge representation, the symptoms set-rule set instance expression method was presented, which has successfully resolved the nontransparent problem in artificial neural networks(ANN). In the knowledge reasoning module, the human mind was introduced, and the comprehension method of Expert System(ES) and ANN was adopted. Then, the combinatorial explosion of single ES and such problems as fuzziness of single ANN diagnostic reasoning are avoided effectively. In the knowledge acquisition module, the automatic study mechanism of ANN is introduced, consequently the intelligence level of diagnosis system is assured.
metasynthesis intelligent diagnosis expert system artificial neural networks
LIU Hui WANG Xiaojian SUN Jinghua ZHANG Junbo
Second Artillery Equipment Academy , Beijing, 100085
国际会议
北京
英文
2007-08-05（万方平台首次上网日期，不代表论文的发表时间）