基于小波原理的电磁声发射信号降噪研究  

De-noising research of electromagnetic acoustic emission signal based on wavelet theory

作　　者：杨素梅[1] 陈宏艳[1] 张闯[2] 刘素贞[2] 

机构地区：[1]河北工业大学计算机科学与软件学院,天津300401 [2]河北工业大学电磁场与电器可靠性省部共建重点实验室,天津300130

出　　处：《计算机工程与设计》2014年第4期1447-1451,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(51077036;51307043);河北省自然科学基金项目(E2012202026;E2012202048);河北省高校优秀青年基金项目(Y2012006)

摘　　要：针对电磁声发射信号中存在的噪声干扰问题,提出了基于小波原理的电磁声发射信号降噪方法。根据需求不同,分别以特定频带的小波系数所占信号的能量百分比最小和最大为寻优依据,对信号依次使用两次自适应小波包分解与重构法;针对分解与重构后各小波包各分量的特点,对其使用新阈值函数、改进的阈值进行阈值化处理,重构得到降噪后的电磁声发射信号。MATLAB实验结果表明,该降噪方法对电磁声发射信号降噪效果明显,保留了原始信号的细节特征,有助于对信号的进一步分析,具有很好的实用价值。To solve the noise problems of the electromagnetic acoustic emission signal, de-noising method about electromagnetic acoustic emission signal based on wavelet theory is proposed. According to different needs, with a specific frequency band energy of wavelet coefficients of signal minimum or maximum optimization based on the percentage, adaptive wavelet packet decomposi- tion and recomposing method is used to deal with the signal in turn for two times. Then, based on the characteristics of each wavelet packet component, the new threshold function is used to denoise the wavelet packet components. MATLAB experimen- tal result proves that the de-noising method has an obvious de-noising effect, maintains the details of the characteristics of the original signal, and has the very good practical value.

关 键 词：小波包 自适应小波 小波阈值 降噪 电磁声发射 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]