轨道交通电机轴承故障的自适应多尺度散布熵分析与诊断  

Analysis and Diagnosis of Rail Traffic Motor Bearing Based on Adaptive Multi-scale Dispersion Entropy

作　　者：孙建晖[1] SUN Jianhui(Department of Railway Transportation,Liaoning Railway Vocational and Technical College,Jinzhou Liaoning 121000,China)

机构地区：[1]辽宁铁道职业技术学院铁道运输学院,辽宁锦州121000

出　　处：《机械设计与研究》2020年第6期96-99,共4页Machine Design And Research

摘　　要：为提升轨道交通电机轴承故障特征提取效果,提出了一种基于局部特征尺度分解(LCD)和散布熵(DE)相结合的自适应多尺度散布熵(AMSDE)的轴承故障分析与诊断方法。首先,采用LCD对轴承振动信号进行自适应分解,获取原始信号不同尺度下的内禀尺度分量(ISC);其次,计算每个ISC分量的DE值,并选取若干个ISC分量DE值组成特征向量;最后,将该特征向量输入支持向量机(SVM)中进行故障诊断。轴承不同类型和不同程度故障诊断的纵向和横向对比实验结果表明,所提方法能够提升轴承的故障诊断效果,相比其他一些方法,具有一定的优势。In order to improve fault feature extraction effect of rail traffic motor bearing,a fault analysis and diagnosis method of motor bearing based on adaptive multi-scale dispersion entropy(AMSDE)which combines local characteristic-scale decomposition(LCD)and dispersion entropy(DE)was proposed.Firstly,the vibration signal was adaptively decomposed into several intrinsic scale components(ISC)which are in different scales by LCD.And then,the DE of each ISC was calculated and several DE value of ISC was set as feature vector.Finally,the feature vector were put into support vector machine(SVM)to diagnosis the bearing faults.Bearing different fault type and different fault degree diagnosis comparison results from vertical and horizontal show that the proposed method can improve diagnosis effect and has certain superiority when compared with some other methods.

关 键 词：局部特征尺度分解 多尺度 散布熵 特征提取 轴承 

分 类 号：TH17[机械工程—机械制造及自动化]