摘要
针对电磁超声检测信号含有较多噪声的现象,使用小波分析方法进行了信号去噪。通过建立信号模型,从小波函数的选取、分解层数的选取、阈值函数和阈值的选取等几个方面进行了分析。采用强制去噪、默认阈值去噪、给定阈值去噪三种方法进行小波去噪仿真,比较得出给定阈值去噪方法效果最佳,在保留信号特征值的前提下达到了去除噪声的目的。
For the phenomenon of electromagnetic ultrasonic testing signal containing more noise, the wavelet analysis method is used for signal denoising. The signal model is established and the wavelet denoising is carried out from the several aspects of the selection of wavelet function, the selection of decomposition layers, the selection of threshold function and threshold. Using the compulsory denoising, default threshold denoising and given threshold denoising to com? plete the wavelet denoising simulation. By comparing these three denoising methods, it shows that the given threshold denoising method has the best result which can reserve the feature value of the signal effectively and achieve the purpose of removing the noise.
作者
阮星翔
汪磊
向登峰
RUAN Xing-xiang;WANG Lei;XIANG Deng-feng(Ningbo Institute of Special Equipment Inspection, Ningbo 315048 , China)
出处
《仪表技术》
2019年第4期13-15,20,共4页
Instrumentation Technology
关键词
电磁超声
小波
去噪
electromagnetic ultrasonic
wavelet
denoising
