The Research For Fault Diagnosis Expert System Of Air Compressor System

To identify and diagnose the faults of oxygen manufacture equipment is the ultimate guarantee to ensure the oxygen manufacture equipment work well. But it is difficult to find and obviate the faults timely only by manpower due to the complicacy of the equipment. Along with the development of the computer technology, a new method which combines expert system and fault diagnosis technology is used to solve the problem.In this paper, it wants to explore an effective method to build a complicated mechanism fault diagnosis expert system.Based on the theories of fault diagnosis technology and expert system, it ascertain that the design of the knowledge base and inference engine is core part of the expert system. a fault tree analysis is used to analyze all possible faults which the oxygen manufacture equipment will appear when it is working. Then a production rule is used to describe the base. It constructed each rule to be a node. Then a linked list can be made on the basis of these nodes. Finally, it can finish the construction of the knowledge base from these linked list. So the construction of the linked list is the one of the knowledge base. To maintain the base, it just needs to add or delete or modify the nodes of the linked list. Obviously, it is more convenient to build and maintain the knowledge base compared with the tree data structure. The inference engine performs the forward and backward mixed inference strategy and width-first search strategy. Thus make a small search space and avoid the creation of the combination explosion. At the same time, the inference can avoid searching the knowledge base repeatedly so that it advances the inference efficiency.