Fault diagnosis of rolling element bearings via discriminative subspace learning: Visualization and classification

作者：

Highlights：

• The trace ratio criterion based LDA method is utilized for fault diagnosis of rolling element bearings.

• TR-LDA is also extended to handle the nonlinear datasets confronted in real-world fault diagnosis.

• We evaluate the proposed method by visualizing and classifying the rolling element bearing fault data.

• Simulations results show the superiority of the method in fault diagnosis of rolling element bearings.

摘要

•The trace ratio criterion based LDA method is utilized for fault diagnosis of rolling element bearings.•TR-LDA is also extended to handle the nonlinear datasets confronted in real-world fault diagnosis.•We evaluate the proposed method by visualizing and classifying the rolling element bearing fault data.•Simulations results show the superiority of the method in fault diagnosis of rolling element bearings.

论文关键词：Fault diagnosis,Linear discriminant analysis,Rolling element bearing,Pattern recognition,Vibrations

论文评审过程：Available online 6 December 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.11.026