计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (1): 211-214.DOI: 10.3778/j.issn.1002-8331.2010.01.063

• 工程与应用 • 上一篇    下一篇

基于小波-神经网络的数字电路IDDT故障诊断

余长庚,雷 加   

  1. 桂林电子科技大学 电子工程学院,广西 桂林 541004
  • 收稿日期:2008-07-21 修回日期:2008-10-14 出版日期:2010-01-01 发布日期:2010-01-01
  • 通讯作者: 余长庚

Fault diagnosis using IDDT for digital circuit based on wavelet analysis and neural networks

YU Chang-geng,LEI Jia   

  1. School of Engineering,Guilin University of Electronic Technology,Guilin,Guangxi 541004,China
  • Received:2008-07-21 Revised:2008-10-14 Online:2010-01-01 Published:2010-01-01
  • Contact: YU Chang-geng

摘要: 利用小波变换与神经网络相结合的方法,采用“能量-故障”特征提取方法和BP算法,提出了一种基于小波分析和神经网络的数字电路瞬态电流IDDT故障诊断方法。该方法首先采样电源到地的瞬态电流IDDT,然后通过小波分析提取电路的故障特征向量,最后输入到神经网络进行故障诊断。经过计算机软件对故障进行仿真,结果表明使用小波-神经网络的数字电路IDDT方法行之有效。

关键词: 数字电路, 故障诊断, 神经网络, 小波分析

Abstract: A method of fault diagnosis for digital circuits of transient power supply current(IDDT) based on the combination of neural network with wavelet transformation is presented,using the method of drawing energy-fault feature and BP algorithm.The proposed method samples transient current of circuit,then uses wavelet analysis as a tool to extract the best optimal feature information.The feature is applied to neural network and the fault is detected and localized.The simulation result of fault shows that the proposed method can classify the faults well.

Key words: digital circuits, fault diagnosis, neural networks, wavelet analysis

中图分类号: