Skip to main navigation Skip to search Skip to main content

Fault feature extraction based on kernel principal component analysis for helicopter rotor

  • BUAA

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

Abstract

Considering difficulty in choice of fault feature and deficiency of principal component analysis for helicopter rotor, an effective fault feature choice method based on kernel principal component analysis is presented and realized. A nonlinear mapping from original feature space into high dimensional feature space is realized by calculating inner product kernel function in original feature space. And nonlinear principal components of original feature data are obtained through principal component analysis of mapped data in high dimensional feature space. Experiment result indicated that kernel principal component analysis can not only decrease the dimension of feature vector space, but also decrease the complexity of fault classifier and increase the precision of classification.

Original languageEnglish
Title of host publication2010 Chinese Conference on Pattern Recognition, CCPR 2010 - Proceedings
Pages1006-1011
Number of pages6
DOIs
StatePublished - 2010
Event2010 Chinese Conference on Pattern Recognition, CCPR 2010 - Chongqing, China
Duration: 21 Oct 201023 Oct 2010

Publication series

Name2010 Chinese Conference on Pattern Recognition, CCPR 2010 - Proceedings

Conference

Conference2010 Chinese Conference on Pattern Recognition, CCPR 2010
Country/TerritoryChina
CityChongqing
Period21/10/1023/10/10

Keywords

  • Feature extract
  • Helicopter
  • Kernel principal component
  • Rotor

Fingerprint

Dive into the research topics of 'Fault feature extraction based on kernel principal component analysis for helicopter rotor'. Together they form a unique fingerprint.

Cite this