现代电子技术Issue(13)：50-52,3.
基于局部均值分解和K近邻算法的滚动轴承故障诊断方法
Fault diagnosis method based on LMD and KNN algorithms for rolling bearing
蔡锷 1李春明 1刘东民 1谭晓伟1
作者信息
- 1. 长安大学 汽车学院，陕西 西安 710064
- 折叠
摘要
Abstract
The roling bearing fault diagnosis method which combines the algorithms of local mean decomposition(LMD) and K nearest neighbor(KNN)is proposed. LMD is applied to decomposing vibration signals of rolling bearing. The fault infor⁃mation is involved in different production functions(PF)components. The eigenvalue of each PF component is extracted in the two aspects of time⁃domain and frequency⁃domain. Dimension reduction of the obtained high dimension eigenvalue is proceeded by principal component analysis(PCA). Finally,state classification of the samples is realized with KNN method in lower dimen⁃sional space. The experimental results show that different fault types of rolling bearing samples can be diagnosed correctly.关键词
滚动轴承/局部均值分解/K近邻算法/特征提取/故障诊断Key words
rolling bearing/LMD/KNN algorithm/feature extraction/fault diagnosis分类
信息技术与安全科学引用本文复制引用
蔡锷,李春明,刘东民,谭晓伟..基于局部均值分解和K近邻算法的滚动轴承故障诊断方法[J].现代电子技术,2015,(13):50-52,3.
