The Research On Fault Diagnosis Methods In The Analog Circuit Based On Neural Network And Genetic Algorithm

The fault diagnosis of the analog circuits is an advanced synthesis intercrossed subject, which is an application technology absorbing the new theory, technology and method in other subjects and fields. In the past 30 years, it has developed a great deal of theory and approaches. But the approaches are limited to deal with fault diagnosis because of the variety and complexity of the analog circuits, especially the large-scale analog circuits with the tolerance or the soft fault. Along with development of electronic technology, especially VLSI and mixed signal circuits, it brings up new challenge to fault diagnosis of the analog circuits, and requires the new theory and approaches for fault diagnosis of the analog circuits.This paper presents a fast approach of module level fault diagnosis for large-scale tolerance analog circuit based on the neural network and the crossover tearing technology according to conventional circuit decomposition technology in large-scale circuits. The neural networks record the information on each time tolerance, not the each torn module. The neural network can parallel deal with the diagnosis information, and the logic operation can judge the information of the multi-fault. The illustrative simulation shows that it can increase the diagnosis speed and decrease the workload before test.This paper presents a new approach of fault diagnosis for tolerance analog circuit, and especially expounds the soft fault diagnosis in tolerance analog circuit. The article discusses problem of solving the soft fault diagnosis by artificial intelligence according to the invariable general characterization of the component in the circuit. And the paper describes the method of using subsection function in the fuzzy function layer of fuzzy neural network. It can enhance generalization ability of neural network and distill the rules automatically by using the new method.A new approach of soft fault diagnosis for tolerance analog circuit by combining genetic algorithm, fuzzy logic with neural network is presented, to avoid the shortcoming that the result of the neural network may be the local minimum. The genetic algorithm is used to search holistically for the initial network weights. The neural network is used to search partly for the optimization output which is distilled the rules to diagnose the soft fault. Illustrative examples show it is efficient.