Research On Fault Diagnosis System Based On Fuzzy Petri Nets

With the rapid development of science and technology,the equipment used inindustry or life is more and more advanced. It is important that the key equipmentcan operate in good state so as to improve the industrial economic benefits,even thesafe of people’s life. Therefore,it is necessary to research fault diagnosis methods.The diagnosis knowledge has some features such as uncertainty and inaccurate.Howto express and reasoning the diagnosis knowledge is a key problem which needed tobe solved in the fault diagnosis system.Petri net, as a graphical modeling tool, which can represent concurrence,asynchrony in intuitive way.It has very important theoretical value and paracticalsignificance to use Petri net into fault diagnosis.Fuzzy Petri nets, as a very importantdevelopment of Petri net,not only have the describe ability,but also inherite the greatfuzzy reasoning ability.It is very suitable to use fuzzy Petri nets for expressing andreasoning fuzzy information.Based on the character of fault diagnosis,the Petri net is applied to the field offault diagnosis. Research work in the thesis focus on:1.Introduce the fault diagnosis system,the knowledge of Petri net and fuzzyPetri net.2.A method based on the fuzzy Petri net is put forward by analyzing theapplication limits of traditional Petri net in the fault diagnosis fields.An effectiveway is devised for computing fuzzy tokens ,which has the ability to express fuzzyinformation. We can use this method to solve the problem of fault propagationchosen.At last,a transition fired algorithm and a backward-reasoning algorithm is putforward ,which can applied for knowledge reasoning and find out fault propagationroutes.3.Use the example of elevator fault diagnosis to confirm the feasibility andvalidity of this method.