A two-step supervisory fault diagnosis framework (1).pdf

A two-step supervisory fault diagnosis framework (1).pdf

ID:34567276

大小:242.61 KB

页数:10页

时间:2019-03-08

A two-step supervisory fault diagnosis framework (1).pdf_第1页
A two-step supervisory fault diagnosis framework (1).pdf_第2页
A two-step supervisory fault diagnosis framework (1).pdf_第3页
A two-step supervisory fault diagnosis framework (1).pdf_第4页
A two-step supervisory fault diagnosis framework (1).pdf_第5页
资源描述:

《A two-step supervisory fault diagnosis framework (1).pdf》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、ComputersandChemicalEngineering28(2004)21312140Atwo-stepsupervisoryfaultdiagnosisframeworkYvonnePower,ParisaA.Bahri∗SchoolofEngineeringScienceandA.J.ParkerCooperativeCentreforHydrometallurgy,MurdochUniversity(RockinghamCampus),Murdoch,WA6150,AustraliaReceived5December2002;receivedinrevise

2、dform21December2003;accepted20January2004Availableonline4June2004AbstractTechniquesenablingearlydetectionanddiagnosisoffaultsareimportantintheprocessingindustries.Thispaperemphasisesatechniqueforearlyfaultdetectionanddiagnosisbasedondynamicfaultdataandatwo-stepfaultdetectionanddiagnosisfr

3、amework.Theapproachshowsvariousadvantagesoveralternativemethodsincludingpromptfaultdetectionandlocalisation,applicabilitytolarge-scalesystemswithouttheneedforexcessivecomputingresources,andamodulararchitecturethatallowsplantsectionstobetreatedindividually.Intheproposedmethod,thelarge-scal

4、eplantisbrokenupintosectionsandaPetrinetbasedonrealtimedataisusedtolocatetheparticularsectionoftheplantinwhichthefaultoriginates.ThisPetrinetthenactivatessecondaryneuralnetworks,whichdiagnosetheexactlocationofthefaultinthatparticularplantsection.Applicabilityoftheproposedtechniqueisdemons

5、tratedthroughapilotplantcasestudy.©2004ElsevierLtd.Allrightsreserved.Keywords:Faultdetectionanddiagnosis;Petrinets;Neuralnetworks1.IntroductiondiagnosistrainedondynamicfaultdataandtheuseofaPetrinetforfaultdetectioninlarge-scalesystems.FaultdetectionanddiagnosisisanessentialpartoftheInthep

6、ast,Petrinetshavecommonlybeenusedtomodeloperationofanyprocessingplant.Earlydetectionanddiag-discreteeventsystemssuchasbatchprocesses,wherethenosisoffaultsisimportantinprocessingindustrysincealotmaterialisoperatedinfinitequantities(orbatches),ratherofdamageandlosscanresultbeforeafaultpresen

7、tinthethanforfaultdetectionbasedoncontinuoustimedata.How-systemisdetected.Inaddition,itbecomeshardertodistin-ever,evenwithacontinuousprocess,dataiscollectedfromguishtherootcauseofthefaultoncethefaultpropagatesthedistributedcontrolsystem(DCS)atdiscretetimein-throught

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。