基于可测性分析的容差网络故障定位的神经网络实现

A Neural Network Approach to Fault Location of Tolerance Network Based on Testability Analysis

利用BP网络的分类功能 ,提出一种有容差模拟电路的故障诊断方法 ,此方法基于可测性理论及模糊组概念 ,所用训练样本少 。

Using the category function of BP networks, a method for fault location of analog circuit with tolerance is presented, which is based on testability theory and ambiguity group concept. In the proposed method, only a small number of samples are needed to train the BPNN, and this method is robust.