容差模拟电路故障诊断bp神经网络算法——中文翻译(附英文原稿)..pdf

Fault Diagnosis of Analog Circuit s with Tolerances Using Back-Propagation Neural Networks DENG Ying1, HE Yi-gang1, SU N Yi-chuang2 ( College of Electrical and Information Engineering , Hunan Univ , Chang sha 410082, China; 2. Dept of Electronic Communication and Electrical Engineering, Faculty of Engineering and Information Science, University of Hertfordshire, Hat field herts AL10 9AB, United Kingdom) Abstract: This paper proposes a method for analog fault diagnosis using neural networks. The primary focus of the paper is to provide robust diagnosis using a mechanism to deal w ith the problem of element tolerances and reduce testing time. T he proposed approach is based on the k-fault diagnosis method and artificial backward propagation neural network. Simulation result s show that the method is robust and fast for fault diagnosis of analog circuits with tolerances. Key words: artificial neural networks; analog circuit s; fault diagnosis; circuit theory and design Classified index NO: T N062; TN052 Document Symbol : A Ⅰ Introduction For several decades, fault diagnosis of analog circuits, the forefront of modern circuit research, has gained wide attention in the testing arena[ 1 - 7] . However, element tolerances, non-linearities and poor fault models make fault location very complicated. Generally, element tolerances make the parameters of circuit element uncertain and the computational equations of traditional methods complex. The non-linear characteristic of the relation between the circuit and its constituent elements makes it even more difficult to diagnose faults on-line and may lead to false diagnosis. To overcome these problems, a robust and fast fault diagnosis method taking tolerances into account is thus needed. Artificial neural n