Feature optimization selection and dimension reduction for partial discharge pattern recognition

  • Shi Qiang Wang
  • , Jia Ning Zhang
  • , Hai Yan Hu
  • , Quan Zhen Liu
  • , Ming Xiao Zhu
  • , Hai Bao Mu
  • , Guan Jun Zhang

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

2 Scopus citations

Abstract

Hundreds of features have been extracted from phase resolved partial discharge (PRPD) pattern and PD waveforms to represent and recognize typical defects. Several feature selection and dimension reduction methods for pattern recognition are presented in this paper. Feature selection algorithms including forward feature selection, backward feature selection and floating forward feature selection (FFFS) are adopted to optimally select the features. |Four dimension reduction algorithms such as principal component analysis, linear discriminant analysis, kernel principal component analysis and generalized discriminant analysis (GDA) are used to further reduce the dimension of features. In order to compare the effectiveness of different selection and reduction techniques, PD tests on artificial PD defect models are performed. The results indicate that the FFFS and GDA are the optimal selection and reduction method, respectively.

Original languageEnglish
Title of host publicationCMD 2016 - International Conference on Condition Monitoring and Diagnosis
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages877-880
Number of pages4
ISBN (Electronic)9781509033980
DOIs
StatePublished - 28 Nov 2016
Event2016 International Conference on Condition Monitoring and Diagnosis, CMD 2016 - Xi'an, China
Duration: 25 Sep 201628 Sep 2016

Publication series

NameCMD 2016 - International Conference on Condition Monitoring and Diagnosis

Conference

Conference2016 International Conference on Condition Monitoring and Diagnosis, CMD 2016
Country/TerritoryChina
CityXi'an
Period25/09/1628/09/16

Keywords

  • Partial discharge
  • dimension reduction
  • feature selection
  • pattern recognition

Fingerprint

Dive into the research topics of 'Feature optimization selection and dimension reduction for partial discharge pattern recognition'. Together they form a unique fingerprint.

Cite this