一种自适应小波消噪方法  

Research on an Adaptive Wavelet Denoising Method

作　　者：刘文艺[1] 汤宝平[1] 蒋永华[1] 

机构地区：[1]重庆大学机械传动国家重点实验室,重庆400030

出　　处：《振动．测试与诊断》2011年第1期74-77,130,共4页Journal of Vibration,Measurement & Diagnosis

基　　金：国家自然科学基金资助项目(编号:50875272);国家高技术研究发展计划("八六三"计划)资助项目(编号:2009AA04Z411);霍英东教育基金会11届青年教师基金资助项目(编号:111057);教育部留学回国人员启动基金资助项目(编号:2008-5)

摘　　要：为了消除噪声对被测信号的干扰,有效提取信号中的有用成分,根据信号和噪声小波变换系数的不同特性,在分析了传统阈值方法局限性的基础上,提出了一种自适应小波消噪方法。该方法首先对被测信号进行小波分解,并改进了阈值量化公式,使其具有能量分布自适应的降噪能力;然后,利用类别方差作为判别依据,选取使得类别方差最大和类内方差最小的阈值作为最佳的阈值,并根据每层分解后的小波系数进行自适应的阈值确定;最后,对信号进行重构,通过分解、阈值处理和重构等过程实现小波消噪。仿真信号和轴承故障诊断的实例结果表明该方法可在强噪声背景下消除噪声干扰,有效提取出滚动轴承的早期故障频率。In order to suppress the noise interference and extract the useful fault feature from the measured signal,an adaptive wavelet de-noising method was proposed according to the inverse characteristics of useful signal and noise in different wavelet scales and the limitation of the traditional threshold methods. Firstly,the measured fault signal was decomposed by discrete wavelet transform.Then,the threshold formula was improved to make this method has energy distribution adaptive de-noising ability.At every decomposition level,an optimal adaptive threshold,which made the class variance be the maximal one and the variance within clusters be the minimal one,was chosen according to the class variance.Lastly,the noise in the signal was de-noised by wavelet reconstruction.The method was applied to simulation experiment and bearing fault diagnosis experiment.The results reveal that this method considerably improves the capability of feature extraction and incipient fault diagnosis for rolling bearing in strong noise backgrounds.

关 键 词：小波消噪 小波分解 阈值 自适应 故障诊断 

分 类 号：TG65[金属学及工艺—金属切削加工及机床] TH113[机械工程—机械设计及理论]