ZHAO Y J, ZHANG N N, ZHOU M L, et al.Bearing fault diagnosis method based on data mapping and CapsNet[J].Journal of Henan Polytechnic University(Natural Science) ,2024,43(5):108-117.

doi:10.16186/j.cnki.1673-9787.2022040075

Received:2022/04/28

Revised:2023/04/23

Published:2024/07/31

Bearing fault diagnosis method based on data mapping and CapsNet

ZHAO Yunji, ZHANG Nannan, ZHOU Menglin, XU Xiaozhuo, ZHANG Xinliang

School of Electrical Engineering and Automation，Henan Polytechnic University，Jiaozuo 454000，Henan，China

Abstract:Conventional deep learning models adaptively extract fault features from vibration signals to realize end-to-end bearing fault diagnosis.However，the vibration monitoring signal is a very complex non-stationary time series signal，and if the deep network directly takes the original vibration signal as input， the nonlinear coupling effect between the data will greatly affect the efficiency of the model for fault feature extraction.  Objectives To reduce the strong nonlinear coupling effect between fault signals，and to solving the problem of the convolutional neural network on the loss of spatial constraint information so as to improve the performance of bearing fault diagnosis，  Methods a bearing fault diagnosis method based on data mapping and capsule network（CapsNet） was proposed.Firstly，the color space model（color names，CN），which could refine color features in the image processing field，was introduced into the fault data preprocessing to map the original low-dimensional space data to the high-dimensional space and improve the spatial differentiation of the fault data.Secondly，to address the problem of high dimensionality and redundancy of the mapped data that affected the efficiency of fault diagnosis，principal component analysis（PCA） was introduced to extract the main meta-information of the fault data，which reduced the dimensionality of the data.Finally，considering the ability of the capsule network to effectively extract spatial constraint information，CapsNet was used as the backbone network for fault diagnosis to identify and classify fault features.  Results The method was validated using the Case Western Reserve University（CWRU） and Xi’an Jiaotong University （XJTU-SY） bearing datasets，the experimental results showed that the method achieved a fault diagnosis accuracy of more than 98% on both datasets，and the diagnostic performance of the method had certain advantages when compared with other deep learning-based fault diagnosis methods. Conclusions The proposed bearing fault diagnosis method could effectively decouple the fault data，improve the spatial differentiation between the data，and then obtain higher bearing fault diagnosis accuracy.

Key words:bearing fault diagnosis;color names;data space mapping strategy;principal component analysis;capsule network