文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 工程科技 > 机械/仪表 > Support vector machines-based fault diagnosis for turbo-pump rotor
侵权投诉

Support vector machines-based fault diagnosis for turbo-pump rotor

Support vector machines-based fault diagnosis for turbo-pump rotor

Support vector machines-based fault diagnosis for turbo-pump rotor

940

S.-F.Yuan,F.-L.Chu/MechanicalSystemsandSignalProcessing20(2006)939–952

diagnosisisthelackoffaultsamples,sothemethodswithpoorgeneralisationleadtowrongdiagnosisandsometimesmakefaultdiagnosisalmostimpossiblewhenthefaultsamplesarefew.

Supportvectormachines(SVM)isanewmachine-learningtoolthatisespecially tforclassi- cation,forecastingandestimationinsmall-samplecases.SVMaimsattheoptimalsolutionintheavailableinformationratherthanjusttheoptimalsolutionwhenthesamplenumbertendstowardsin nitelylarge,itsolvessatisfactorilythe‘over- tting’,thelocaloptimalsolution,thelowconvergencerateandsomeotherproblems,especiallyithasagoodgeneralisationevenwhenthesamplesarefew[1].Inthispaper,wehavedevelopeda‘onetoothers’SVMalgorithmforthemulti-classfaultdiagnosis,andclassi ed14kindsoftypicalfaultintheturbo-pumprotorwiththisalgorithm,obtained13optimalseparatingplanes,andrecognised42faultsamples.Thetestresultshowsthatthismethodcoulddiagnosethefaultscorrectlyandeffectively.

2.Supportvectormachines

ThebasicideaofthepatternrecognitionwithSVMistoprojectsamplespaceintoahigh-dimensionaleigenspace.Intheeigenspace,optimalseparatinghyper-planes(orhyper-lines,similarlyhereinafter)oftheoriginalsamplesetcanbefound,butcalculationcomplexitydoesnotincreasealmost.Inthissectionwebrie ydescribetheSVMalgorithmanditsmotivation.AmoredetaileddescriptionofSVMcanbefoundinRef.[1].

Westartfromasimplecaseoftwoclassesthatislinearlyseparable.Assumingthatwehaveadataset:

D¼fðxi;yiÞgði¼1;...;lÞ;

yi2fÀ1;þ1g,

(1)

wherexiisinputsampleandyiisoutputclass,yhastwovalues(À1or+1)thatstandfortwoclasses,listhesamplenumber.Wewishtodetermine,amongalllinearseparatingplanesthatseparatetheinputsamplesintotwoclasses,whichseparatingplanewillhavethesmallestgeneralisationerror.AsshowninFig.1,ringsanddiamondsstandforthesetwoclassesofsamplepointsrespectively;Hisaseparatingplane.H1andH2aretheplanesthatareparalleltoHandrespectivelypassthroughthesamplepointsclosesttoHinthesetwoclasses,thedistancebetweenH1andH2isde nedasmargin.TheoptimalseparatingplanethathasthesmallestgeneralisationerroristheonethatnotonlycorrectlyseparatesallsamplepointsintothesetwoclassesbutalsoleavesthelargestmarginbetweenH1andH2.

IfthetwoclassesarenotperfectlylinearlyseparablewecanstilllookforanoptimalseparatingplanethatmaximisesthemarginbetweenH1andH2andminimisestheclassifyingerror.Theoptimalseparatingplaneisalinearclassi erthathastheclassifyingfunctionasfollows[1]:

()lX

liyixTxiþbði¼1;...;lÞ,(2)fðxÞ¼sgn

i¼1

Fig.1.(a)Aseparatingplanewithsmallmargin.(b)Aseparatingplanewithlargermargin.Abettergeneralisationcapabilityisexpected

from(b).

上一页第2页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email