The Study On Expert System About State Monitoring And Fault Diagnosis Of The Gearbox

As gearbox play an important role in the mechanical drive system, to maintain it better,based on signal analysis multi-expert collaboration expert system about state monitoring andfault diagnosis of the gearbox is developed successfully. It is on the basis of methods on thestate monitoring and fault diagnosis of the gearbox to be analyzed deeply. And the type ofJZQ250 gearbox is the target for expert system's monitoring and diagnosis.The data acquisition system is composed of vibration acceleration transducers, chargeamplifier and NI-DAQ Card from aspect of hardware design of expert system. And fromaspect of software analysis, first of all, the knowledge in Knowledge Base is establishedaccording to characters of signal when faults happen in the gearbox. The knowledge includesthe character of state monitoring which is gained from second analysis of the data and thecharacter of fault diagnosis which is consist of energy proportion spectrum of the data. ThenInference Engine is designed according to artificial intelligence principle. And it is structuredby BP neural network, gray cognate and D-S evidence of information fusion. The validity andcorrectness of three principles applied to fault diagnosis are established by simulation. Fromaspect of software design, the blueprint of the expert system is drawn through the tool ofUML and the method of OOP. Then development of software is achieved with the help of theCASE tool of Rational Rose, Visual Basic6.0 and MATLAB7.0. Final the experiment showsvalidity and correctness of the expert system applied to the type of JZQ250 gearbox.