基于MM-EMD的改进HHT及应用  

Improved HHT based on MM-EMD and its application

作　　者：梁升[1] 王新晴[1] 王东[1] 夏天 钱淑华[1] 

机构地区：[1]解放军理工大学,南京210007 [2]海军工程研究设计局,北京100841

出　　处：《振动与冲击》2012年第20期23-26,37,共5页Journal of Vibration and Shock

摘　　要：针对传统Hilbert-Huang变换假设"信号由IMF(Instinct Mode Functions)分量与普通趋势项构成"这一固有局限,根据油压信号的自身特点,提出了基于MM-EMD(Mathematical morphology and Empirical Mode Decomposition)的Hilbert-Huang变换改进方法,并将其应用于油压信号分析处理中。该方法的基本思想如下:针对存在非IMF、非普通趋势项分量的信号,先用数学形态滤波器分离出该分量,再对剩余信号进行EMD分解与Hilbert谱分析。仿真信号与实测信号表明,新方法能取得比传统Hilbert-Huang变换更好的分析效果。The traditional HHT（Hilbert-Huang transformation） method supposes that a signal is composed of instinct mode functions（IMFs） and an ordinary trend term.Considering this limitation and the characteristics of hydraulic pressure signals,an improved HHT method based on MM-EMD（mathematical morphology and empirical mode decomposition） was proposed and employed to analyze pressure signals.The basic idea of this method was that if a signal contains a component not beloging to instinct mode functions and ordinary trend term,the morphological filter is used to separate this component from the original signal,and the rest signal is analyzed with EMD and HSA（Hilbert spectral analysis）.The results of simulation and the measured signals showed that compared with the traditional method,the new one can have a better analysis effect.

关 键 词：油压信号 数学形态学(MM) 经验模态分解(EMD) Hilbert-Huang变换(HHT) 

分 类 号：TH137.51[机械工程—机械制造及自动化]