Implementation Of Fault Diagnosis Expert System Based On Rete Algorithm Of Rotating Machinery

Since industrial revolution, the productivity forces and the technology of automation production have been developed rapidly to satisfy the requirement of society. With this background, production equipments present the trendy to become huge-size, more complicated, high-speed and automatic. Once the key units fail, they usually cause major safety accidents and heavy economic losses which even can reach millions of Yuan in petrochemical enterprises. Hence, how to keep unit safety operation turns to be a crucial subject for every large-scale enterprise and the technology of fault diagnosis was developed with this situation. It proved that applying advanced fault diagnosis technology could effectively detect the fault position, analyze the fault reason, improve maintenance efficiency and decrease economic losses. What's more significant, by improving the running status of production line, the probability of major safety accidents could be decreased.In this paper, with rotating machinery taken as research object, incorporated with the theory of expert system, the technology of intelligent fault diagnosis is discussed. The paper consists of five chapters and it is organized as follows. The first chapter mainly introduces the subject source and research significance. Then, the history of fault diagnosis is summarized. It also describes the present situation of expert system's application in this area and the development direction and application prospect of intelligent fault diagnosis technology, then this paper's research content and purpose is followed.In the second chapter, it expounds the history and realization principles of the classic prior to matching algorithm, Rete algorithm with pseudo-code aiming at this paper's research object.The third chapter deeply analyzes fault characteristics and intrinsic motivation of the most familiar faults on rotating machinery including rotor imbalance, rotor misalignment, rotor bow etc. At the last part of this chapter, a symptoms-reason table was established, which forms the basis of the expert system rules base discussed later.According to the above analysis, the fourth part designs and implements the fault diagnosis expert system on rotating machinery with the Rete algorithm as the reasoning algorithm through the Java programming language and Eclipse development platform.Finally, this paper makes a quick review of the research process and looks forward the next step research direction.