Bearing fault diagnosis method under unbalanced data distribution

被引：0
|
作者：
Cao J. [1 ,2 ,3 ]
He Z.-D. [1 ]
Yu P. [1 ,3 ]
Wang J.-H. [1 ,3 ,4 ]
机构：
[1] College of Electrical & Information Engineering, Lanzhou University of Technology, Lanzhou
[2] Engineering Research Center of Urban Railway Transportation of Gansu Province, Lanzhou
[3] Engineering Research Center of Manufacturing Information of Gansu Province, Lanzhou
[4] National Experimental Teaching Center of Electrical and Control Engineering, Lanzhou University of Technology, Lanzhou
来源：
Jilin Daxue Xuebao (Gongxueban)/Journal of Jilin University (Engineering and Technology Edition) | 2022年 / 52卷 / 11期
关键词：
Class-Balance loss function; convolutional neural network; fault diagnosis; rolling bearing; unbalanced data distribution;
D O I：
10.13229/j.cnki.jdxbgxb20210374
中图分类号：
学科分类号：
摘要：
Aiming at the problem that the unbalanced distribution of data in the fault diagnosis of rolling bearings will reduce the model's diagnostic ability，a 1DCNN with Width Kernel of First Layer（WKFL-1DCNN）is proposed. WKFL-1DCNN firstly uses a larger kernel in first-layer to extract fault features，and adds a BN（Batch normalization） layer after the alternate convolution layer to adjust the data distribution；then uses the Class-Balanced loss function instead of the Cross-Entropy loss function to offset the impact of the data imbalanced distribution on the network. Experiments show that the improvement method in this paper can effectively improve the performance of WKFL-1DCNN in unbalanced fault diagnosis，and its fault diagnosis ability is better than other comparison algorithms. © 2022 Editorial Board of Jilin University. All rights reserved.
引用
收藏
页码：2523 / 2531
页数：8
相关论文
共 25 条
  • [1] Tang D H, Hee J K., A survey on deep learning based bearing fault diagnosis, Neurocomputing, 335, pp. 327-335, (2019)
  • [2] Szegedy C, Ioffe S, Vanhoucke V, Et al., Inceptionv4, inception-resnet and the impact of residual con⁃ nections on learning[J/OL]
  • [3] Mikolov T, Karafiat M, Burget L, Et al., Recurrent neural network based language model[C] ∥Inter⁃ speech, Conference of the International Speech Com⁃ munication Association, pp. 1045-1048, (2015)
  • [4] Zhong Hui, Kang Heng, Lyu Ying-da, Et al., Image manipulation localization algorithm based on channel attention convolutional neural networks, Journal of Jilin University(Engineering and Technology Edi⁃ tion), 51, 5, pp. 1838-1844, (2021)
  • [5] Xia M, Li T, Liu L Z, Et al., Intelligent fault diagno⁃ sis approach with unsupervised feature learning by stacked denoising autoencoder, IET Science, Mea⁃ surement & Technology, 11, 6, pp. 687-695, (2017)
  • [6] Zhang W, Li C H, Peng G L, Et al., A deep convolu⁃ tional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load, Mechanical Systems and Signal Processing, 100, pp. 439-453, (2018)
  • [7] Shao H, Jiang H, Zhao H, Et al., An enhancement deep feature fusion method for rotating machinery fault diagnosis, Knowledge-Based Systems, 119, pp. 200-220, (2017)
  • [8] Levent E., Bearing fault detection by one-dimensional convolutional neural networks, Mathematical Prob⁃ lems in Engineering, 2017, pp. 1-9, (2017)
  • [9] Pan J, Zi Y Y, Chen J L, Et al., LiftingNet: a novel deep learning network with layerwise feature learning from noisy mechanical data for fault classification, IEEE Trans Ind Electron, 65, 6, pp. 4973-4982, (2018)
  • [10] Chen Xiao-lei, Sun Yong-feng, Li Ce, Et al., Stable anti-noise fault diagnosis of rolling bearing based on CNN-BiLSTM, Journal of Jilin University(Engi⁃ neering and Technology Edition), 52, 2, pp. 296-309, (2022)
123