Research On Intelligent Diagnosis System For Locomotive Electronic Control Cabinet Based On Virtual Instrument

With the locomotive high speed development, its electronics equipments' function is more complete, constitution is more complex, reliability is requested higher. Therefore it is urgent to diagnose the locomotive electronics equipments. The electronics control cabinet is an important part of locomotive, and it works long time in the situation where high temperature and more vibration. However, because faults occur only once in a while and diagnosis is very difficult, it is very important to diagnose quickly and truly. Using traditional diagnosis methods, we will find the testing period is long and the fixed position is not accurate, even it is hard to satisfy the need of the modern and high rhythm manufacture. So the thesis investigate and study the intelligence diagnosis system firstly based on the virtual instrument technique and the artificial intelligence technique.This thesis introduces virtual instrument technique and diagnosis techniques, such as the fault tree analysis method and expert system. Then we research and analyze the principle of the technique methods and the application in the realm to fault diagnosis. The intelligence diagnosis system of electronics control cabinet combines data acquisition and fault diagnosis. So we design two sub-systems: virtual instrument testing system, fault diagnosis expert system.In the need of data acquisition from the electronics control cabinet, the author select DAQ hardware platform and LabWindows/CVI software to complete the development of the electronics control cabinet virtual instrument test sub-system. User can select to carry on the equipment function test or the single test in testing system. The function test is mainly to complete kinds of working status testing. The single test carries on testing to the functional units could be discovered.After studying the composing and the theory of the locomotive electronic control cabinet deeply, comparing various kinds of fault diagnosis methods, the author decides to use the expert system diagnosis method based on the fault tree analysis. Expert system knowledge base setups by the fault tree model, which can help obtaining diagnosis knowledge and insuring the consistency and complete ofknowledge. The expert system inference engine is realized by CLIPS which is a special expert system tool.