集成经验模态分解与相关峭度在滚动轴承故障诊断中的应用_王孝霖.pdf

第 9 期 机械设计与 制造 2015 年 9 月 Machinery Design ＆ Manufacture 61 集成经验模态分解与相关峭度在滚动 轴承故障诊断 中的应用 王孝霖，韩纬华，顾 含 ，胡 存 （63680 部队，江苏江阴 214431） 摘 要：船舶设备 中的滚动轴承其振动信号成分复杂，故障信号易被背景噪声湮没，常规诊断方法难以有效提取轴承故 障信息。为了从复杂的轴承振动信号 中提取弱故障信号，将集成经验模态分解（Ensemble Empirical Mode Decomposition ， EEMD）应用到滚动轴承故障诊断 中，并引入了相关峭度的定义，提出 了一种新的滚动轴承故障特征提取方法。该方法首 先利用 EEMD 将轴承振动信号分解为若干个分量信号，然后根据相关峭度最大原则选取分量信号，提取 出滚动轴承的弱 故障信号。通过对轴承内圈故障的仿真和实验研究验证 了该方法的有效性。 关键词：船舶设备；滚动轴承；振动信号；故障诊断；集成经验模态分解；相关峭度 中图分类号：TH16 ；TH133.3 ；TN911 文献标识码：A 文章编号：1001-3997 （2015）09-0061-04 Application of Ensemble Empirical Mode Decomposition and Correlated Kurtosis to Rolling Element Bearing Fault Diagnosis WANG Xiao-lin ，HAN Wei-hua，GU Han，HU Cun （Unit63680，Jiangsu Jiangyin 214431，China） Abstract ：The vibration signals created by rolling bearing are very complicated ，especially for those bearing applied in ship machinery equipment. Normally the fault signal is easily covered by background noise ，so the traditional methods can ’t effectively extract the weak fault information. In order to solve this problem ，the EEMD and correlated kurtosis is applied to rolling bearing fault diagnosis ，and a feature extraction method is proposed. According to the method ，a group of component signals are obtained by EEMD，then the component signal with maximal correlated kurtosis is selected ，and the faint fault information is extracted. Through simulation and experiment ，the effectiveness of the