基于EMD的能量算子解调方法及其在机械故障诊断中的应用  

ENERGY OPERATOR DEMODULATING APPROACH BASED ON EMD AND ITS APPLICATION IN MECHANICAL FAULT DIAGNOSIS

作　　者：程军圣[1] 于德介[1] 杨宇[1] 

机构地区：[1]湖南大学机械与汽车工程学院,长沙410082

出　　处：《机械工程学报》2004年第8期115-118,共4页Journal of Mechanical Engineering

基　　金：国家自然科学基金(50275050);高等学校博士点专项科研基金(20020532024)资助项目。

摘　　要：为了提取多分量的AM-FM信号的频率和幅值信息,提出了基于EMD(Empirical mode decomposition)的能量算子解调法,并将它应用于机械故障诊断中。该方法首先采用EMD将多分量的AM-FM信号分解成若干个IMF(Intrinsic mode function)分量之和,然后对每一个IMF分量进行能量算子解调,从而提取多分量的AM-FM信号的幅值和频率信息。对机械故障振动信号的分析结果表明,基于EMD的能量算子解调法能有效地提取机械故障振动信号的特征。An energy operator demodulating approach based on EMD (Empirical mode decomposition) is proposed to extract the frequencies and amplitudes of the multi-component AM-FM signals. Furthermore, the proposed approach is applied to mechanical diagnosis. In this approach, EMD method is used to decompose a multi-component AM-FM signal into a number of intrinsic mode functions (IMF) firstly; then, the energy operator demodulating is applied to each IMF and the amplitudes and frequencies of a multi-component AM-FM signal are extracted. The proposed approach is applied to mechanical fault diagnosis. The analysis results show that the energy operator demodulating approach based on EMD can extract the characteristics of mechanical fault vibration signals efficiently.

关 键 词：AM-FM信号 EMD 能量算子解调 故障诊断 机械 HILBERT变换 

分 类 号：TH17[机械工程—机械制造及自动化]