Multi-scale deep intra-class transfer learning for bearing fault diagnosis

被引:207
|
作者
Wang, Xu [1 ]
Shen, Changqing [1 ]
Xia, Min [2 ]
Wang, Dong [3 ]
Zhu, Jun [4 ]
Zhu, Zhongkui [1 ]
机构
[1] Soochow Univ, Sch Rail Transportat, Suzhou 215131, Peoples R China
[2] Univ Lancaster, Dept Engn, Lancaster LA1 4YW, England
[3] Shanghai Jiao Tong Univ, State Key Lab Mech Syst & Vibrat, Shanghai 200240, Peoples R China
[4] Natl Univ Singapore, Dept Ind Syst Engn & Management, Singapore, Singapore
基金
中国国家自然科学基金;
关键词
Fault diagnosis; Deep transfer learning; Multi-scale feature learner; Intra-class adaptation; AUTOENCODER; NETWORK; FUSION;
D O I
10.1016/j.ress.2020.107050
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
The tremendous success of deep learning in machine fault diagnosis is dependent on the hypothesis that training and test datasets are subordinated to the same distribution. This subordination is difficult to meet in practical scenarios of industrial applications. On the one hand, the working conditions of rotating machinery can change easily. On the other hand, vibration data and labels are difficult to obtain to train a specific model for each working condition. In this study, we solve these problems by constructing a novel deep transfer learning model called multi-scale deep intra-class adaptation network, which first uses the modified ResNet-50 to extract low-level features and then constructs a multiple scale feature learner to analyze these low-level features at multiple scales and obtain high-level features as input for the classifier. Pseudo labels are then computed to shorten the conditional distribution distance of vibration data collected under different working loads for intra-class adaptation. The proposed method is validated using two datasets to recognize the bearing normal state, the inner race, the ball and outer race faults, and their fault degrees under four different working loads. The high-precision diagnosis results of 24 transfer learning experiments reveal the reliability and generalizability of the constructed model.
引用
收藏
页数:15
相关论文
共 50 条
  • [31] Bearing fault diagnosis based on DNN using multi-scale feature fusion
    Zhou, Funa
    Zhang, Zhiqiang
    Chen, Danmin
    [J]. 2020 35TH YOUTH ACADEMIC ANNUAL CONFERENCE OF CHINESE ASSOCIATION OF AUTOMATION (YAC), 2020, : 150 - 155
  • [32] Intra-Class Adaptive Augmentation With Neighbor Correction for Deep Metric Learning
    Fu, Zheren
    Mao, Zhendong
    Hu, Bo
    Liu, An-An
    Zhang, Yongdong
    [J]. IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 : 7758 - 7771
  • [33] Multi-Scale Recursive Semi-Supervised Deep Learning Fault Diagnosis Method with Attention Gate
    Tang, Shanjie
    Wang, Chaoge
    Zhou, Funa
    Hu, Xiong
    Wang, Tianzhen
    [J]. MACHINES, 2023, 11 (02)
  • [34] Combining Multi-Scale Wavelet Entropy and Kernelized Classification for Bearing Multi-Fault Diagnosis
    Rodriguez, Nibaldo
    Alvarez, Pablo
    Barba, Lida
    Cabrera-Guerrero, Guillermo
    [J]. ENTROPY, 2019, 21 (02)
  • [35] Fault diagnosis method based on a multi-scale deep convolutional neural network
    Bian, Jingyi
    Liu, Xiuli
    Xu, Xiaoli
    Wu, Guoxin
    [J]. Zhendong yu Chongji/Journal of Vibration and Shock, 2021, 40 (18): : 204 - 211
  • [36] Multi-scale deep neural network for fault diagnosis method of rotating machinery
    Xie, Yining
    Liu, Wang
    Liu, Xiu
    Chen, Deyun
    Guan, Guohui
    He, Yongjun
    [J]. FERROELECTRICS, 2023, 602 (01) : 215 - 230
  • [37] A deep feature extraction approach for bearing fault diagnosis based on multi-scale convolutional autoencoder and generative adversarial networks
    Hu, Zhiyong
    Han, Taorui
    Bian, Jun
    Wang, Ziwei
    Cheng, Liu
    Zhang, Wenlei
    Kong, Xiangwei
    [J]. MEASUREMENT SCIENCE AND TECHNOLOGY, 2022, 33 (06)
  • [38] Bearing fault diagnosis base on multi-scale 2D-CNN model
    Zhang, Jun
    Zhou, Yang
    Wang, Bing
    Wu, Ziheng
    [J]. 2021 3RD INTERNATIONAL CONFERENCE ON MACHINE LEARNING, BIG DATA AND BUSINESS INTELLIGENCE (MLBDBI 2021), 2021, : 72 - 75
  • [39] ADVERSARIAL SUBDOMAIN ADAPTATION METHOD BASED ON MULTI-SCALE FEATURES FOR BEARING FAULT DIAGNOSIS
    Zhou, Yuguo
    Jin, Zhao
    Zhang, Zhikai
    Geng, Zengrong
    Zhou, Lijian
    [J]. MATHEMATICAL FOUNDATIONS OF COMPUTING, 2024, 7 (04): : 485 - 511
  • [40] Bearing Fault Diagnosis Based on Shallow Multi-Scale Convolutional Neural Network with Attention
    Huang, Tengda
    Fu, Sheng
    Feng, Haonan
    Kuang, Jiafeng
    [J]. ENERGIES, 2019, 12 (20)