The Application Of Fuzzy Neural Network In Vehicle Fault Diagnosis Expert System

The paper briefly introduces the basic theory of fuzzy theory and neural network, and accounts for necessity of combining fuzzy theory with neural network after analysis defects and merits in fault diagnosis respectively. For particularity, complexity and existing incomplete and fuzzy information, it's necessary to use fuzzy theory to retrieve cases, thus a case retrieving method of hierarchical integrating is presented. On the basis of existing research, the paper integrates fuzzy neural network technology with diagnosis expert system based on case-reasoning and rule-reasoning. The method of Automatic fuzzy rules extraction based on fuzzy BP net researches hidden key attributes through deleting redundant linking weight. It transfers weights of FNN into diagnosis guiding operator based on case-reasoning, which plays a important role in selecting similar cases quickly and improving diagnosis validity. And a guiding mechanism in FNN combines case-reasoning and rule-reasoning powerfully. The adoption of multi-pattern diagnosis and their inner integration reach our diagnosis expectation of integrating system.