基于EEMD样本熵和GK模糊聚类的机械故障识别

中国机械工程 ›› 2013, Vol. 24 ›› Issue (22): 3036-3040,3044.

基于EEMD样本熵和GK模糊聚类的机械故障识别

王书涛;李亮;张淑清;孙国秀

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

出版日期:2013-11-25 发布日期:2013-11-29
基金资助:
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）

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

Wang Shutao;Li Liang;Zhang Shuqing;Sun Guoxiu

Measurement Technology and Instrumentation Key Lab of Hebei Province,Yanshan University,Qinhuangdao,Hebei,066004

Online:2013-11-25 Published:2013-11-29
Supported by:
National Natural Science Foundation of China（No. 51075349，61077071）;Hebei Provincial Natural Science Foundation of China（No. F2011203207）

摘要/Abstract

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。

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

Abstract:

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 proposed, 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(IMFs) with different time scales. Secondly, IMF 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.

Key words: ensemble empirical mode of decomposition(EEMD), sample entropy, GK-fuzzy clustering, mechanical fault diagnosis

中图分类号:

TH17

王书涛, 李亮, 张淑清, 孙国秀. 基于EEMD样本熵和GK模糊聚类的机械故障识别[J]. 中国机械工程, 2013, 24(22): 3036-3040,3044.

Wang Shutao, Li Liang, Zhang Shuqing, Sun Guoxiu. Mechanical Fault Diagnosis Method Based on EEMD Sample Entropy and GK Fuzzy Clustering[J]. China Mechanical Engineering, 2013, 24(22): 3036-3040,3044.

参考文献

[1]屈梁生, 张海军. 机械诊断中的几个基本问题[J]. 中国机械工程, 2000, 11(1): 211-216.
[2]汤宝平, 蒋永华, 张详春. 基于形态奇异值分解和经验模态分解的滚动轴承故障特征提取方法[J]. 机械工程学报, 2010(5): 37-42.
[3]罗洁思, 于德介, 彭富强. 基于EMD的多尺度形态学解调方法及其在机械故障诊断中的应用[J]. 振动与冲击, 2009(11).
[4]郑旭, 郝志勇, 金阳, 等. 基于EEMD与广义S变换的内燃机噪声源识别研究[J]. 内燃机工程, 2011, 32(5): 68-73.
[5]王宏, Narayanan R M, 周正欧, 等. 基于改进EEMD的穿墙雷达动目标微多普勒特性分析[J]. 电子与信息学报, 2010(6): 1355-1360.
[6]Feng Zhipeng, Song Xigeng,Chu Fulie. Fault Diagnosis Based on lntergration of Cluster Analysis Rough Set Method and Fuzzy Neural Network[J]. Chinese Journal of Mechanical Engineering, 2004, 17(3): 349-352.
[7]雷亚国, 何正嘉, 訾艳阳, 等. 混合聚类新算法及其在故障诊断中的应用[J]. 机械工程学报, 2006, 42(12): 116-121.
[8]Huang N E, Zheng S, Long S R, et al. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Non-stationary Time Series Analysis [J]. Proceedings of the Royal Society of Lond(Series A), 1998, 454(1971): 903-995.
[9]Wu Z H, Huang N E. Ensemble Empirical Mode Decomposition: A Noise Assisted Data Analysis Method[J]. Advances in Adaptive Data Analysis, 2009, 1(1): 1-41.
[10]Richman J S, Moorman J R. Physiological Time-series Analysis Using Approximate Entropy and Sample Entropy[J]. American Journal of Physiology-heart and Circulatory Physiology, 2000, 278(6): 2039-2049.
[11]邹阳, 苗夺谦, 王登. 醉酒者和正常人脑电的样本熵研究[J]. 中国生物医学工程学报, 2010, 29(6): 939-942.
[12]Krishnapuram R, Jongwoo K. A Note on the Gustafson-Kessel and Adaptive Fuzzy Clustering Algorithms[J]. IEEE Trans. on Fuzzy System, 1999, 16(5): 453-461.
[13]黄金杰, 李士勇, 蔡云泽. 一种建立粗糙数据模型的监督模糊聚类方法[J]. 软件学报, 2005, 16(5): 744-753.
[14]The Case Western Reserve University Bearing Data Center Website. Bearing Data Center Seeded Fault Test Data[EB/OL]. [2007-11-27]. http://www.eecs.case.edu/laboratory/bearing/download.htm.