Research On Artificial Immune System And The Application In Fault Diagnose

The biological immune system is a parallel adaptive information learning system, which can identify and remove the antigens invading the body. It also has the ability for learning, remember and adjust adaptively to keep the stabilization of the body. Inspired by the biological immune mechanism, a new intelligent information learning system developed gradually. In the recent years, people have constructed many artificial immune model used in different fields based on the biological immune system. Now AIS academic researches mainly focus on artificial immune algorithms and artificial immune network models, AIS application researches mainly focus on computer network security, fault diagnosis and optimize computation, etc.The core content researched in this paper is the design of AIS and its application to fault diagnose. Firstly, some basic concepts, framework and principles of the biological immune system are simply introduced, especially the mechanism that has close relation to fault diagnose. Then introduce the research content, research status and basic theory of the artificial immune system, the framework and flow of some classical immune algorithms are researched and analyzed, especially Negative Selection Algorithm, Clone Selection Algorithm and aiNet immune network. Based on those work, a base framework of artificial immune system is proposed, in which trains the detectors by improved NSA and diagnoses the fault by improved aiNet. In order to improve the efficiency, fuzzy clustering is used to classify the normal data and the detectors, so every normal datum or detector is representative. The simulation experiment about asynchronous electromotor shows the feasibility of the artificial immune system proposed above.Intelligent fault diagnose technology is also discussed in this paper. Expert System, Multi-Agent System and Immune Algorithm should be combined to diagnosis fault.