发布时间:2012/7/27 11:23:50 作者:张梅军,韩思晨,王闯,焦志鑫 【字体:
大 中 小】
张梅军,韩思晨,王闯,焦志鑫
(解放军理工大学 工程兵工程学院,江苏 南京 210007)
摘要:针对滚动轴承故障振动信号的非平稳特征和故障征兆模糊性,提出了基于EMD和动态模糊聚类图的轴承故障诊断方法。运用EMD方法提取待诊断的轴承运行状态样本的能量特征指标,应用模糊聚类分析方法对特征参数进行聚类,并作出聚类树状图。结果表明,该方法不需要大量的样本进行学习,且能更直观、准确识别滚动轴承的运行状态。
关键词:EMD分解;动态模糊聚类图;故障诊断
中图分类号:O242.21 文献标识码:A 文章编号:1006-0316 (2012) 07-0001-04
Clustering based on EMD decomposition tree bearing fault diagnosis
ZHANG Mei-jun,HAN Si-chen,WANG Chuang,JIAO Zhi-xin
(Engineering Institute of Engineering Corps,PLA University of Science,Nanjing 210007,China)
Abstract:For the non-stationary feature of a vibration signal of defective rolling bearings and the ambiguity of fault feature, a fault diagnosis method of rolling bearings is proposed using EMD ( Empirical Mode Decomposition ), Dynamic fuzzy clustering graph. Firstly, an EMD method was used to decompose a vibration signal of a rolling bearing. Then those parameters were analyzed by fuzzy clustering algorithm, and plotted amic fuzzy clustering graph. Experiments indicated that This method does not require a large number of samples for learning, and And can more intuitivelt, accurately distinguish the running state of bearings.
Key words:emp iricalmode decomposition ( EMD );dynamic fuzzy clustering graph;fault diagnosis
———————————————
收稿日期:2011-02-29
基金项目:国家自然科学基金资助项目(51175511)
作者简介:张梅军(1958-),女,江苏宜兴人,副教授,硕士生导师,主要研究方向为故障诊断和工程机械动力学。