基于经验模态分解的旋转机械振动信号滤波技术研究  

The Study of EMD-Based Filtering Technique for Vibration Signals in Rotating Machinery

作　　者：胡劲松[1] 杨世锡[1] 吴昭同[1] 严拱标[1] 

机构地区：[1]浙江大学机械工程系,杭州310027

出　　处：《振动．测试与诊断》2003年第2期96-98,共3页Journal of Vibration,Measurement & Diagnosis

基　　金：浙江省自然科学基金资助项目 (编号 :5 0 0 1 0 0 4 )

摘　　要：把经验模态分解方法引入了旋转机械振动信号滤波领域 ,详细地解释了这种方法及其算法。通过对一个带高频毛刺干扰的振动信号的经验模态分解与本征模函数的组合 ,演示了基于经验模态分解的振动信号滤波的方法。把振动信号滤波前后的波形进行了比较 。The main idea of empirical mode decomposition(EMD) is to decompose a time series into a set of intrinsic mode functions(IMF) with different scales.The EMD method was introduced to the field of vibration signals filtering in rotating machinery.The method and algorithm were explained in detail.To illustrate the effect of the EMD based filtering technigue,a vibration signal with high frequency interference was introduced.By using EMD the signal was decomposed into six IMFs denoted by c 1,c 2,c 3,et al.A low pass filtered signal was obtained by the combination of c 3 to c 6 IMFs;a high pass filtered signal was fomed by the combination of c 1 to c 5 IMFs;a band pass filtered signal was also got by the combination of c 3 to c 5 IMFs.The comparison between unfiltered and filtered signal waveforms shows that the EMD based filtering method is fairly effective.

关 键 词：经验模态分解 旋转机械 振动信号 信号滤波技术 EMD 

分 类 号：TH133[机械工程—机械制造及自动化] TN911.7[电子电信—通信与信息系统]