Skip to main navigation Skip to search Skip to main content

Bearing Fault Diagnosis Based on Adaptive Multiclass-Mahalanobis-Taguchi System

  • Ning Wang
  • , Limin Jia
  • , Zhipeng Wang*
  • *Corresponding author for this work
  • Beijing Jiaotong University
  • Beijing Research Center of Urban Traffic Information Sensing and Service Technologies

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

To diagnose faults of bearings accurately, a fault diagnosis method based on Empirical Mode Decomposition (EMD), Singular Value Decomposition (SVD) and adaptive Multiclass-Mahalanobis-Taguchi System (aMMTS) is proposed in this paper. The condition of the bearing is monitored in real time by sensors. Then, the vibration signal is decomposed by EMD and the features are extracted by using SVD. Then, a novel adaptive Multiclass-Mahalanobis-Taguchi system is proposed for fault diagnosis. The hybrid method based on EMD-SVD and adaptive Multiclass-Mahalanobis-Taguchi system overcomes the shortcomings in the Mahalanobis-Taguchi system in terms of over-fitting and non-adaptive feature selection for fault diagnosis and has some advantages over the traditional auxiliary noise fault analysis method when dealing with nonlinear signal, and can diagnose the bearing fault without manual intervention. The effectivity and feasibility of the proposed method is validated by an experiment.

Original languageEnglish
Title of host publicationProceedings - 2018 Prognostics and System Health Management Conference, PHM-Chongqing 2018
EditorsPing Ding, Chuan Li, Shuai Yang, Ping Ding, Rene-Vinicio Sanchez
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1120-1125
Number of pages6
ISBN (Electronic)9781538653791
DOIs
StatePublished - 4 Jan 2019
Externally publishedYes
Event2018 Prognostics and System Health Management Conference, PHM-Chongqing 2018 - Chongqing, China
Duration: 26 Oct 201828 Oct 2018

Publication series

NameProceedings - 2018 Prognostics and System Health Management Conference, PHM-Chongqing 2018

Conference

Conference2018 Prognostics and System Health Management Conference, PHM-Chongqing 2018
Country/TerritoryChina
CityChongqing
Period26/10/1828/10/18

Keywords

  • Adaptive Multiclass-Mahalanobis-Taguchi System
  • Bearing
  • EMD
  • Fault diagnosis
  • SVD

Fingerprint

Dive into the research topics of 'Bearing Fault Diagnosis Based on Adaptive Multiclass-Mahalanobis-Taguchi System'. Together they form a unique fingerprint.

Cite this