The Research On The Used Of Integtated Fault Diagnosis Expert System In Cigarette-making Machine

With the development of the machinery becomes more high-speed, roboticized and complicated, equipment fault diagnosis technology has been getting more attention. The traditional method of fault diagnosis is difficult to meet the needs of complex fault diagnosis. The development of expert system provides a new way for equipment fault diagnosis. With the in-depth application of expert system based on single reasoning method, its limitation becomes more and more abvious. Therefore, it's a new hotspot that integration a variety of reasoning methods.Surrounded the development of high-speed cigarette-making machine fault diagnosis expert system, according to the features of cigarette-making machine fault, from the angle of integrated fault diagnosis, this thesis studies the integrated fault diagnosis mode collaborating the rule-based reasoning(RBR), case-based reasoning(CBR) and artificial neural network(ANN) reasoning, and develops the high-speed cigarette-making machine fault diagnosis expert system.Firstly, the thesis introduces the conception and development status of fault diagnosis technology and expert system, studies the basic structure of expert system. And the necessity of collaborating the expert system and diagnosis technology is discussed.Some relevant reasoning methods have been discussed in this thesis, include RBR method, CBR method and ANN method. The thesis focuses on the reasoning process,key technology and core algorithm of these methods. And the characteristics and the scope of application of these reasoning methods are studied.In the last part, this thesis summarizes some methods of acquiring fault knowledge, and uses them to acquire a lot of fault knowledge. Based on this knowledge, it analyses the characteristics of the fault, presents the perspective of classifying the fault using the tree structure which is a conception of data structure. And the fault of high-speed cigarette-making machine is classified using this method. Through the analysis of the features of various reasoning methods, the integrated fault diagnosis strategy collaborating the RBR, CBR and ANN is presented. the thesis designs the structure and reasoning program of integrated fault diagnosis expert system, researches on the design and realization technology of integrated knowledge base and reasoning mechanism of the expert system. At last, the integrated fault diagnosis expert system is developed with VB6.0 and Microsoft Access. And the system is verified through the examples of failure.