大规模模拟电路子网络级故障诊断的神经网络方法

Neural Method of Fault Diagnosis at Subnetwork Level in Large Scale Analog Circuit

本文提出了一种采用双向联想记忆神经元网络诊断法诊断子网络级故障的新方法。在对含不可及节点的子网络级故障诊断的互测试条件检验的基础上,用神经元网络诊断法代替逻辑分析来快速而有效地定位故障子网络,并给出了诊断实例。

Abstrct: A new method of fault diagnosis at subnetwork level with BAMN is proposed in this paper. Based on the MTC test of fault diagnosis at subnetwork level with inaccessible node, a NN method instead of Logical Analysis Method is used to locate faulty subnetwork. And actual example of diagnosis is given in the paper.