基于EEMD样本熵和GK模糊聚类的机械故障识别  

Mechanical Fault Diagnosis Method Based on EEMD Sample Entropy and GK Fuzzy Clustering

作　　者：王书涛[1] 李亮[1] 张淑清[1] 孙国秀[1] 

机构地区：[1]燕山大学河北省测试计量技术及仪器重点实验室,秦皇岛066004

出　　处：《中国机械工程》2013年第22期3036-3040,3044,共6页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(51075349;61077071);河北省自然科学基金资助项目(F2011203207)

摘　　要：针对目前各种机械故障诊断方法的局限性,提出了基于总体平均经验模式分解(EEMD)样本熵和GK模糊聚类的故障特征提取和分类方法,建立了一种机械故障准确识别的有效途径。首先,对机械振动信号进行EEMD分解,得到若干不同时间尺度的固有模态函数(IMF)分量。其次,通过相关性分析和能量相结合的准则对IMF分量进行筛选,并将筛选出的IMF分量的样本熵组成故障特征向量。最后,将构造的特征向量输入到GK模糊聚类分类器中进行聚类识别。实验及工程实例证明了该方法的有效性和优越性。Aiming at existing limitations of the various methods for mechanical fault diagnosis ,a new method for fault diagnosis based on EEMD sample entropy and GK fuzzy clustering was pro-posed ,and an efficient paths of mechanical fault recognition was established accurately .First of all , the mechanical vibration signals were decomposed by EEMD into a certain number of intrinsic mode functions（IM Fs） with different time scales .Secondly ,IM F components were chosen by the combined criteria of mutual correlation coefficient and energy analysis ,and the sample entropies of each IMF component composed fault eigenvectors .At last ,the constructed eigenvectors were put into the GK fuzzy clustering classifier to recognize different fault types .The experimental and engineering test demonstrate the efficiency and superiority of this method .

关 键 词：总体平均经验模式分解(EEMD) 样本熵 GK模糊聚类 机械故障识别 

分 类 号：TH17[机械工程—机械制造及自动化]