Research On Circuit Fault Diagnosis Based On Ant Colony Algorithm

With the fast development of electronic technology, the research and application on circuit diagnosis has been regarded highly by all. The diagnosis using artificial intelligence is one of the hot areas on circuit fault diagnosis. As the important research orientation, Ant Algorithm is studied to use in the fault diagnosis more frequently.The research purpose of this paper is how to find the fewer test vectors to detect and affirm the fault when the complete fault test set is given. And the shortpath-finding characteristic of ant algorithm is helpful to find the relationship of test vectors to find the shortest path for the fault detecting.This paper first introduced the basic theory and model of the ant algorithm, then it analyzed the correlative parameters. At last, it discussed and proved the feasibility for introducing ant algorithm into the circuit fault diagnosis. The basic ant algorithm was easily to trap in the local optimization, aimed at this disadvantages, this paper discuss one improved ant algorithm which used the MMAS strategy, elitist strategy and weighted arranging strategy. According to the characteristic of circuit diagnosis, this paper defined the calculate formula of the parameters in ant algorithm, and presented the circuit fault diagnose model based on ant algorithm, then used it to detect the vedio processing circuit to validate its utility in circuit fault diagnosis.