基于独立特征选择核Fishier判别分析的电机轴承故障诊断  

Motor bearing fault diagnosis based on individual feature selection kernel fishier discriminant analysis

作　　者：杨斌[1] 李文慧[2] 王畴民 

机构地区：[1]淮阴师范学院数学科学学院 [2]淮阴师范学院城市与环境学院 [3]淮安美妙电子科技有限公司

出　　处：《电子测量与仪器学报》2018年第10期33-38,共6页Journal of Electronic Measurement and Instrumentation

基　　金：江苏省“双创计划”科技副总项目(FZ20180582)资助

摘　　要：为提高电机轴承的故障诊断精度,在深入研究核Fishier判别分析的基础上,对其进行了改进,并提出基于独立特征选择核Fishier判别分析(IFS-KFDA)的电机轴承故障诊断方法。该方法首先从多个角度构建了原始高维故障特征集,在此基础上,利用独立特征选择核Fisher判别分析为轴承每两类故障状态独立选择敏感特征集,使获得的敏感特征对故障状态具有更好的表征作用,同时还有效地排除乐原始混合特征集中的非敏感特征的干扰。轴承故障诊断实例验证了方法的有效性。In order to improve diagnose accuracy of motor bearing, kernel fishier discriminant analysis was researched and improved in this paper and a fault diagnosis method based on individual feature selection kernel fishier discriminant analysis(IFS-KFDA) was proposed. On the basis of original constructed high-dimensional feature set, individual feature selection kernel fishier discriminant analysis was proposed and used to select individual sensitive feature subset for each pair of bearing class. So that the acquired sensitive feature has a better representation to each fault, at the same time effectively eliminates the interference of non-sensitive features of original mixed feature set. The experimental results of motor bearing indicate that the proposed method is effective

关 键 词：独立特征选择 核Fishier判别分析 故障诊断 轴承 

分 类 号：TM307[电气工程—电机]