摘要
提出了一种基于小波和人工神经网络的故障检测与诊断的方法 ,该方法利用小波包分解的精确细分的特点 ,分别对正常系统和故障系统的采样信号进行精确特征提取 ,并构造一系列基于信号能量且具有表征系统状态能力的特征向量 ,然后利用人工神经网络分类器对系统在各种状态下的特征向量进行分类决策 。
A method of fault detection and diagnosis based on wavelet and artificial neural networks is proposed in this paper. The eigen-vectors of the normal system and fault system are extracted by virtue of precise subdivision of wavelet packet decomposition, and a series of eigen-vectors based on signal energy are constructed. Then eigen-vector under a certain state can be assorted and decided by use of artificial neural networks sorter and realize fault detection and diagnosis to the system sequentially.
出处
《航天控制》
CSCD
北大核心
2000年第2期64-71,共8页
Aerospace Control
关键词
小波
人工神经网络
故障诊断
故障检测
Artificial neural network\ Fault detection\ Fault diagnosis
