摘要
以电路软故障作为研究对象,提出了一种基于差值信号做故障特征的方法。首先将被测电路中各状态的信号波形的数据与标准信号波形的数据作差值,差值作为新的数据组,再进行小波包变换提取特征向量,经主元分析后选取具有代表主要信息的作为综合特征向量,最后送到分类器进行诊断。实验结果表明此方法可以有效地将电路软故障区分开来。
Taking the soft faults of circuit as the object of study, differential signal which may be used for the fault feature extraction is proposed. New data was obtained from differential signals between normal and fault condition firstly. Then wavelet package decomposition and PCA（Principal Component Analysis） were used to extract the feature of the new data. The final characteristic was used by the designed Neural network to recognize the fault. The experiment indicate the method to be very effective.
出处
《国外电子测量技术》
2010年第1期66-68,80,共4页
Foreign Electronic Measurement Technology
关键词
差值信号
特征提取
故障诊断
differential signal
feature extraction
fault diagnosis
