Skip to main navigation Skip to search Skip to main content

Induction motor fault diagnosis based on ensemble classifiers

  • Southeast University, Nanjing
  • Case Western Reserve University

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

7 Scopus citations

Abstract

With increasing demand for accurate fault diagnosis of induction motors, traditional methods based on single parameter need amelioration. In this paper, an effective and practical induction motor fault diagnosis algorithm is proposed based on adaptive weighted voting multiple random forest classifiers. Firstly, the vibration signals and stator current signals are obtained and analyzed. The energy features at several characteristic frequencies related to motor faults from each type of signal are extracted and used as input to corresponding random forest classifier. Then clustering analysis is applied to both testing and training samples to determine the weight of each classifier for decision making on diagnostic result. Experimental study performed on induction motor data has verified that the classifier fusion algorithm can improve the diagnostic accuracy.

Original languageEnglish
Title of host publicationI2MTC 2016 - 2016 IEEE International Instrumentation and Measurement Technology Conference
Subtitle of host publicationMeasuring the Pulse of Industries, Nature and Humans, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781467392204
DOIs
StatePublished - 22 Jul 2016
Externally publishedYes
Event2016 IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2016 - Taipei, Taiwan, Province of China
Duration: 23 May 201626 May 2016

Publication series

NameConference Record - IEEE Instrumentation and Measurement Technology Conference
Volume2016-July
ISSN (Print)1091-5281

Conference

Conference2016 IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2016
Country/TerritoryTaiwan, Province of China
CityTaipei
Period23/05/1626/05/16

Keywords

  • ensenble clasifer
  • fault diagnosis
  • induction moter
  • random forest classifier

Fingerprint

Dive into the research topics of 'Induction motor fault diagnosis based on ensemble classifiers'. Together they form a unique fingerprint.

Cite this