基于自适应多尺度散布熵的滚动轴承故障诊断方法  

Fault Diagnosis Method of Rolling Bearings based on Adaptive Multi-scale Dispersion Entropy

作　　者：李从志 郑近德[1] 潘海洋[1] 刘庆运[1] LI Congzhi;ZHENG Jinde;PAN Haiyang;LIU Qingyun(School of Mechanical Engineering,Anhui University of Technology,Ma’anshan 243032,Anhui China)

机构地区：[1]安徽工业大学机械工程学院,安徽马鞍山243032

出　　处：《噪声与振动控制》2018年第5期173-179,共7页Noise and Vibration Control

基　　金：国家自然科学基金资助项目(51505002);国家重点研发计划资助项目(2017YFC0805103);安徽省高校自然科学研究重点资助项目(KJ2015A080)

摘　　要：针对滚动轴承振动信号的非平稳、非线性特性,将一种新的衡量时间序列复杂性和不规则程度指标——散布熵(dispersion entropy,DE)引入到滚动轴承非线性故障特征提取,提出一种基于经验模态分解与DE相结合的自适应多尺度散布熵滚动轴承故障诊断方法。首先,采用经验模态分解对振动信号进行分解,得到若干不同尺度的本征模态函数;其次,计算每个本征模态函数的DE值;再次,将得到的DE值作为特征向量输入到基于支持向量机建立的多故障分类器进行训练和识别。最后,将提出的滚动轴承故障诊断方法应用于试验数据分析,结果表明,提出的方法能准确地识别滚动轴承故障类型。Aiming at the non-stability and non-linearity of the vibration signals of rolling bearings,a new parameter called dispersion entropy(DE),which measures the complexity and irregularity of time series,is introduced into the nonlinear fault feature extraction of the rolling bearings.On this basis,the fault diagnosis method of rolling bearings based on the adaptive multi-scale dispersion entropy(AMDE),which combines empirical mode decomposition(EMD)with the DE,is proposed.First of all,the EMD is used to adaptively decompose the vibration signal of the rolling bearings into several intrinsic mode functions(IMF)in different scales.And the DE values of the obtained IMFs are calculated.Then,the DE values are trained and recognized by inputting them as the feature vectors into the multi-fault classifier on the base of support vector machine(SVM).Finally,the proposed rolling bearing fault diagnosis method is applied to test data analysis.The results show that the proposed method can accurately identify the fault types of rolling bearings.

关 键 词：振动与波 经验模态分解 多尺度 散布熵 滚动轴承 故障诊断 

分 类 号：TN911.7[电子电信—通信与信息系统] TH165.3[电子电信—信息与通信工程]