Based On Fuzzy Reasoning And Rough Set Of Rotating Machinery Fault Diagnosis

Fault diagnosis has developed in the past decades, which is already from simple observation, data processing, modern diagnostic technology on the basis of the method of sensor ,dynamic testing technology, signal processing and modeling treatment, now to the new stage--- the stage of intelligent diagnosis technology.The research object of this thesis mainly is typical of rotating machinery---fan,do some research in the way which build the diagnosis system . This paper mainly studies rough set reduction,it can let the diagnosis process quicker and more accurate; Constructs the diagnosis system is built, including constructing knowledge base and designing knowledge reasoning strategy,it can identify the typical faults. the system is said, this paper makes the following aspects of the research and discussion:1) introduce the fault symptoms extraction and fault reasoning methods.2)First,with, Database of the fan's symptoms built with access 2000 and frame theory, then VB6.0 can be validly linked to build the database, combining positive reasoning strategy to construct diagnosis system.3)Every common failure characteristic is extracted, then arrange their sequences according to common signs which have one-to-one datitaled,composed of characteristic vector by rough sets theory and reduction, the diagnosis results are faster, more comprehensive after knowledge reduction by MATLAB.4)Fuzzy reasoning diagnosis is conducted on the basis of knowledge base adopting the fuzzy reasoning matching method.The discriminant diagnosis is a kind of uncertainty reasoning by using fuzzy knowledge, fuzzy reasoning matching is based on the theory of fuzzy set theory and the development of fuzzy logic, it handling vague things themselves, suitable for fault knowledge recognition, also accounting for it through the constructed simulation.5)Fault diagnosis system is constructed with VB , when the logical levels is more complex, inevitably facing complex problems such as knowledge search, needing to consider the knowledge organization and improvement of reasoning mechanism, this system has constantly perfect function, knowledge base can be added constantly and maintened, therefore the fault diagnosis system can be more and more accurate with the use of the passage of time with the use of the passage of time.