小波分析在模拟电路故障特征提取中的应用  

Application of Wavelet Analysis in Extracting Analog Circuits Fault Characteristics

作　　者：李刚[1] 陈建辉[1] 连光耀[1] 卢江雷[1] 

机构地区：[1]军械工程学院,河北石家庄050003

出　　处：《仪表技术》2008年第10期44-46,共3页Instrumentation Technology

基　　金：国家自然科学基金资助项目(60771063)

摘　　要：针对模拟电路故障特征提取,提出了在小波变换Mallat算法的基础上,结合快速傅里叶变换的方法。该方法选用dbN小波,利用Mallat算法将信号分解,得到一系列高频和低频子带,分别对各子带进行快速傅里叶变换,从而得到各子带上信号的频谱。文章给出了采用该方法的基本步骤和应用实例,应用表明该方法可以有效辨别模拟电路参数型故障,准确得到故障特征。Aiming at the extraction of analog circuits fault features, the method based on the Mallat algorithm of wavelet analysis combining with Fast Fourier Transform is presented in this paper. The dbN wavelet is selected to decompose the signal in Mallat algorithm. Then a series of sub-bands in high frequency and low frequency are got. The spectrum of each sub-band by FFT transforming can be achieved separately. It is proved this method can distinguish the parameter fault of analog circuits effectively, and get the fault spectrum feature precisely.

关 键 词：小波变换 MALLAT算法 快速傅立叶变换 

分 类 号：TM930[电气工程—电力电子与电力传动]