New imbalanced bearing fault diagnosis method based on Sample-characteristic Oversampling TechniquE (SCOTE) and multi-class LS-SVM

被引:97
|
作者
Wei, Jianan [1 ]
Huang, Haisong [1 ]
Yao, Liguo [1 ,2 ]
Hu, Yao [1 ,3 ]
Fan, Qingsong [1 ]
Huang, Dong [1 ]
机构
[1] Guizhou Univ, Key Lab Adv Mfg Technol, Minist Educ, Guiyang 550025, Guizhou, Peoples R China
[2] Yuan Ze Univ, Dept Ind Engn & Management, Taoyuan 32003, Taiwan
[3] Guizhou Renhe Zhiyuan Data Serv Co Ltd, Guiyang 550025, Guizhou, Peoples R China
基金
中国国家自然科学基金;
关键词
Multi-class imbalanced classification; SCOTE oversampling; Least squares support vector machine; Bearing fault diagnosis; PREDICTION; SMOTE; SUWO;
D O I
10.1016/j.asoc.2020.107043
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In actual industrial production, the historical data sets used for bearing fault diagnosis are generally limited and imbalanced and consist of multiple classes. These problems present challenges in the field of bearing fault diagnosis, for which traditional fault diagnosis methods (e.g., multi-class least squares support vector machine (multi-class LS-SVM)) are not very effective. Therefore, we propose a new multi-class imbalanced fault diagnosis method based on Sample-characteristic Oversampling Technique (SCOTE) and multi-class LS-SVM, where SCOTE is a new oversampling method proposed by us. SCOTE transforms multi-class imbalanced problems into multiple binary imbalanced problems. In each binary imbalanced problem, first, SCOTE uses the k-nearest neighbours (knn) noise processing method to filter out noisy points. Second, samples are trained by LS-SVM, and minority samples are sorted by importance according to the misclassification error of the minority classes in the training sets. Moreover, based on the importance sorting of minority samples, SCOTE performs a sample synthesis method based on the k* information nearest neighbours (k*inn) to address the binary imbalanced problems. Thus, when all the binary imbalance problems are addressed, the multi-class imbalanced problem will also be addressed. The 20 fault diagnosis examples represented by Case Western Reserve University (CWRU) bearing data and Intelligent Maintenance Systems (IMS) bearing data show that the proposed method has higher fault diagnosis recognition rates and algorithm robustness than 8 oversampling algorithms and 8 multi-class imbalanced algorithms. (C) 2020 Elsevier B.V. All rights reserved.
引用
收藏
页数:25
相关论文
共 50 条
  • [31] Fault diagnosis of bearings based on multi-class SVM and D-S evidence theory
    Department of Automobile Engineering, Military Transportation University, Tianjin
    300161, China
    不详
    300072, China
    Qiche Gongcheng, 1 (114-119):
  • [32] SA-CGAN: An oversampling method based on single attribute guided conditional GAN for multi-class imbalanced learning
    Dong, Yongfeng
    Xiao, Huaxin
    Dong, Yao
    NEUROCOMPUTING, 2022, 472 : 326 - 337
  • [33] NEW METHOD FOR BEARING FAULT DIAGNOSIS BASED ON VMD TECHNIQUE
    Bousseloub Y.
    Medjani F.
    Benmassoud A.
    Kezai T.
    Belhamra A.
    Attoui I.
    Diagnostyka, 2024, 25 (02):
  • [34] Fault Diagnosis of Roller Bearing Using Parameter Evaluation Technique and Multi-Class Support Vector Machine
    Susilo, Didik Djoko
    Widodo, Achmad
    Prahasto, Toni
    Nizam, Muhammad
    INTERNATIONAL CONFERENCE ON ENGINEERING, SCIENCE AND NANOTECHNOLOGY 2016 (ICESNANO 2016), 2017, 1788
  • [36] Joint application of feature extraction based on EMD-AR strategy and multi-class classifier based on LS-SVM in EMG motion classification
    Zhi-guo Yan
    Zhi-zhong Wang
    Xiao-mei Ren
    Journal of Zhejiang University-SCIENCE A, 2007, 8 : 1246 - 1255
  • [37] Joint application of feature extraction based on EMD-AR strategy and multi-class classifier based on LS-SVM in EMG motion classification
    Yan, Zhi-Guo
    Wang, Zhi-Zhong
    Ren, Xiao-Mei
    JOURNAL OF ZHEJIANG UNIVERSITY-SCIENCE A, 2007, 8 (08): : 1246 - 1255
  • [38] A New SVM Multi-Class Classification Method Based on Error-Correcting Code
    Wang, Zelong
    Yan, Fengxia
    He, Feng
    Zhu, Jubo
    2008 INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND SECURITY, VOLS 1 AND 2, PROCEEDINGS, 2008, : 584 - +
  • [39] A rolling bearing fault diagnosis method based on interactive generative feature space oversampling-based autoencoder under imbalanced data
    Huang, Fengfei
    Zhang, Kai
    Li, Zhixuan
    Zheng, Qing
    Ding, Guofu
    Zhao, Minghang
    Zhang, Yuehong
    STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2025, 24 (02): : 979 - 997
  • [40] The Study of Transformer Fault Diagnosis Based on Means Kernel Clustering and SVM Multi-class Object Simplified Structure
    Sun, Xiaoyun
    Bian, Jianpeng
    Liu, Donghui
    Li, Zhenquan
    2008 7TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-23, 2008, : 5158 - 5161