英文翻译格式文档格式.docx
- 文档编号:20864679
- 上传时间:2023-01-26
- 格式:DOCX
- 页数:6
- 大小:38.81KB
英文翻译格式文档格式.docx
《英文翻译格式文档格式.docx》由会员分享,可在线阅读,更多相关《英文翻译格式文档格式.docx(6页珍藏版)》请在冰豆网上搜索。
系别软件与服务外包学院.
专业通信网络与设备.
班级通信0901.
学生姓名韩丽司.
学号090969.
指导教师陈佳.
二○一二年二月
Basedonthedatafusionofintelligentfaultdiagnosissystem
1.Primedwords
Multisensordatafusiontechnologywasinitiallymostlyusedinthemilitaryfield,butthecomputer,networkandcommunicationtechnologytherapiddevelopmentthattheapplicationrangeisexpandedgreatly.Inrecentyears,manyscholarsofthedatafusionrulesandstrategytheorytoconductextensiveresearchandimprovement.Whiletheartificialintelligencetechnologyresearchmakesthedatafusiontoimprovetheknowledgeoftargetdecisionheight,theauxiliaryfunctionisgreatlystrengthened.Atthesametime,withtheindustrialtechnologymakeaspurtofprogress,anintelligentfaultdiagnosissystemofdemandinquantityandqualitygreatlyimproved.Astheintelligentfaultdiagnosissystemforthemostbasic,themosteffectiveinformationprocessingtools,multisensordatafusiontechnologydevelopmentwillpromotetheprogressofintelligentfaultdiagnosissystem.
2.MultisensordatafusionandimprovedD-Stheory
Fromamilitaryapplicationperspective,datafusionistomakefulluseofdifferenttimeandspaceofthemultisensorinformationresourcesaccordingtothetimesequence,usingcomputertechnologytoobtainmultiplesensorobservationinformationincertaincriteriatobeautomatic,integratedanalysis,controlanduse,accesstotheobjectconsistencyofinterpretationanddescription,tocompleterequireddecision-makingandestimationtasks,allowingthesystemtoobtainthanitscomponentsthebetterperformanceofH3.Theauthorusesthepresentgenerallyagreethatthepixellevel,featurelayeranddecisionlayerthreelayerfusionstructure.Thedecisionlevelfusiontargetistoachievethetargetsituationdiagnosisandassessment,appliedtothemainBayesianprobabilityreasoningandDSevidencetheory.Thedatafusionmethodtosolvetheuncertaininformationprocessingproblems,D-SmethodwithDempster-Shaferevidentialtheoryasafoundation,itscoreisDempstersynthesisrules,foruncertaininformationexpressionandsynthesisprovidesnaturalandrobustmethod.WillforceSevidencetheoryisusedformultisensorfusion,obtainedfromthesensorrelatedvalueisthetheoryofevidence,itcanconstitutethetargetstoberecognizedpatternsofbelieffunctionassignment,thateachtargetmodelhypothesisofcredibility,eachsensorconsistsofanevidenceofgroup.
MultisensordatafusionisthroughDSunitedrulestocombineseveralevidencegrouptoformanewintegratedevidencegroup,calledtheDSassociationruleswitheachsensorofconfidencefunctiondistributionformedbyfusionofconfidencefunctiondistribution,whichistargetmodedecision-makingprovidecomprehensiveandaccurateinformationofn].Inpracticalapplication,D-Smethodrequiresevidenceofindependenceandevidencecombinationruletheorysupport,andthecalculationofpotentialexiststheproblemofcombinatorialexplosion,soonlythesinglefusionmethodsaredifficulttoobtainidealfusioneffect.
D-Sevidencetheoryhasdoesnotrequireaprioriprobabilityadvantages;
expertsystemhasaproblemdomainknowledge;
fuzzysystemhashigherfuzzylanguageprocessing;
highorderneuralnetworkhasthecapacitytobebig,approximationability,fault-tolerantawiderangeoffeatures,sotheDsevidencetheoryfusionmethodwithmultipledivisioncomplementarytoimprovetheD-smethod,improvethefusionsystemfortargetidentificationaccuracyandreliability,whichmakethesystemhasstrongselflearningabilityandabilitytoadapttotheirenvironment.
3.Intelligentfaultdiagnosissystem
Diagnosissystemofafailuremodeisoftencausedbymultiplefaultsymptom,andafaultsymptomcanbecausedbymultiplefailuremodes,ismany-to-manyform.Sowithoutasensortoensurethatatanytimetoprovidecompleteandreliableinformation,itisusuallyinmultiplesensorbasedonintegrateddiagnosis.Inessence,faultdiagnosissystemistheuseofdiagnosticobjectsystemrunsavarietyofstateinformationandvariouskindsofexistingknowledge,informationprocessing,finallygetonthesystemoperationconditionandfaultconditionofthecomprehensiveevaluationofn3.DatafusionistypicalapplicationsystemisC3Isystem,especiallyinmultipletargettrackingsystem.AccordingtoC3Isystem,faultdiagnosisinformationrequiredforaccesstomorediverse,describediagnosticmathematicalmodeloftheobjectmaybegreaterthanthespacecoordinatesandvelocitycharacteristicsaremorecomplex,thefaultdiagnosticobjectlinkbetween(coupling,backup,transfer)canbetrackedobjectofcoordinatedactionoftherelationstobemoreclose,butcanmakethediagnosisobjectisregardedasasensorthroughthesystematicobservationoftheparticularstatespace,thefaultsignalisthespaceinthespecifictargetsignal,thefaultdiagnosisisbasedonthesignalandtheknowledgebasetodeterminethefaultalarm.
Thefaultdiagnosissystem,verysuitableforusingthepreviouslydescribedmultisensorfusionstructure,apixellayerislayerofdatafusionforsensorreflectthedirectdata;
featurelayercorrespondingtovariousfaultdiagnosismethodsofdatafusion,theresultsareeffectivedecision;
decisionfusionforintegratedsubsystemsviathefusionruleofcombinationmadethefinaltheresultsoffaultdiagnosisandtroubleshooting.Thethreelayerstructurecorrespondingtothefaultdiagnosissystemofmonitoring,diagnosisanddecisionfunction.Infaultdiagnosissystem
Datafusioninthecertaindegreecanmakethesystemtoobtaintheaccuratestateestimation,increasethedegreeofconfidence,toreduceambiguity,improvediagnosticperformance,improvethemultisensorinformationresourcesutilization.Butwiththedevelopmentofnewtechnology,faultdiagnosissystemisgraduallyintroducedintoartificialintelligencetechnology,themainperformanceis:
theuseofneuralnetworklocaldiagnosis;
theuseofmultipleconcurrentESusingmultipleknowledgeinthefieldofsyntheticinformation;
theuseofadvanceddatabasemanagementtechnologyfordecisionsupportsystemusingreasoning;
learning,sotheautomaticadapttoallkindsoftrend.Inaddition,onthebasisofdatafusion,thefusionlevelsincrease,thedataminingandknowledge(includingrules,methodandmodel)fusion.
4.Basedonthedatafusionofintelligentdiagnosissystem
Fromtheperspectiveofmultisensordatafusion,typicalapplicationexampleistheprocessmonitoringandfaultdiagnosis,andfromtheperspectiveofintelligentfaultdiagnosissystem,usuallyinmultisensordatafusionbasedonintegrateddiagnosis.Basedontheaboveonthemulti-sensordatafusiontechnologyandintelligentfaultdiagnosissystemarediscussed,thefollowingtwotechniquesfororganiccoupling,basedontheestablishmentofamultisensordatafusionofintelligentfaultdiagnosissystemstructureframe.
4.1Workingprinciple
Thesystemiscomposedofinputoutputsystem,asensorsignalacquisitionsystem,signalpre-processingsystem,expertsystemanddecisionfusionsystem.Whenthesystemworks,thefirstuseofmultisensorsignalacquisitionandsignaldatawerepreprocessed(suchassignalfiltering,spectrumanalysis,waveletanalysis,etc.)willbeprocessedinformationanddiagnosticsystemofexpertknowledgebase(rules,methodsandmodelsofknowledge)accordingtocertainrules,andtheneachsub-systemisthelocaldiagnosisresultsareparallelfusionfordecisionfusionsystemforglobaldiagnosis,thefinaloutputdiagnosisresultsandrelevantinformationwillbestoredinthedatabaseandknowledgebasefortheuseofdataminingtechnologyforknowledgediscoveryforthenecessarydataonreserves.
4.2Keytechnology
4.2.1Localdiagnosissystem
Neuralnetworkcanrealizethecomplexnonlinearmapping,inthefieldoffaultdiagnosishasbeenwidelyusedexport].Whenthesystemparametersforthediagnosisofmore,signsofthelargeamountofinformationtimes,duetotheinevitablecontradictionbetweensampleandrandom,ifthehighdimensionalsymptominformationinputatthesametimetothesamenetworkprocessing,willmakethelongtrainingtime,thediagnosisofpoorresults,sometimesevenTocausethenetworkconvergence.Therefore,thehumanbrainindifferentregionswithdifferentinformation.Differentsignalsarealsobytherespectiveneuralnetworkdiagnosis.Sothehighdimensionalsymptomspacedecompositionintolowdimensionalsymptomspace,theprocessmayalsobereferredtoasthelocaldiagnosis.Inaddition,theneuralnetworksystemcaneffectivelysolvetheexpertsystempartofthelimitations,sotheuseoftheneuralnetworkexpertsystem.
4.2.2Decisionfusion
Usingneuralnetworkforlocaldiagnosis,fromeachorseveraldiagnosticparameterscangettheirdiagnosticresults,eachsubsystemisresponsibleforafaultdiagnosis,fromdifferentangles,faultdiagnosis,decisionfusionofthesediagnosticresultsfusion,makesthesubsystemisformedbetweenthe"
consultation"
utmosttoimprovethediagnosisrate.Forpreprocessinginformationfusion,inferenceismoreimportantthannumericalcomputation,shouldbebasedonknowledgeofthetechnologyofexpertsystemandD-Stheoryofevidencecombinationmethodoffusion.
4.2.3Dataminingandknowledgefusion
Systemexistingoperatingstatetorevisetheoriginalsystemknowledgebase,canbemorequickly,moreaccurate,morecomprehensivefaultdiagnosis,thisisthedataminingandknowledgeintegrationissues,dataminingtechniquesininformationfusionsystemwillbecomethenecessarypartof.
5.Theend
Mu
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 英文翻译 格式