简介:ispaperestablishesaformalmodelforhybriddiagnosis,novelfeaturesincluding:(1)Itprovidesaunifiedtheoreticalframeworkforutilizingdevicemodelsandheuristicsindiagnosis,whichnaturallyintegratesalltheimportantcomponentsofdiagnosis-thestructuralandbehavioraldescriptionofdevices,faultmodes,thelowerandupperfaultbounds,faultpossibilitiesandheuristicrules-intoadiagnosticsystem.Devicemodelspredictoutputsfrominputs,heuristicrulesinferthepossibilitiesofcertaincomponentsbeingfaultyfromsymptons,andyetthecombinationofbothconstrainseachotherforustoreducethehypothesisspace.(2)Itpresentsatypicalwayofmodelingbehaviorofdevices,towhichthekeyistheintroductionofI-Ofunctionswithindefiniteinputs/outputs.(3)Itcaneasilybeimplementedoveraforware-chaininginferenceengine.
简介:Ahybridapproachforfuzzysystemdesignbasedonclusteringandakindofneurofuzzynetworksisproposed.Anunsupervisedclusteringtechniqueisfirstlyusedtodeterminethenumberofif-thenfuzzyrulesandgenerateaninitialfuzzyrulebasefromthegiveninput-outputdata.Then,aclassofneurofuzzynetworksisconstructedanditsweightsaretunedsothattheobtainedfuzzyrulebasehasahighaccuracy.Finally,twoexamplesoffunctionapproximationproblemsaregiventoillustratetheeffectivenessoftheproposedapproach.
简介:Theautomaticdetectionoffacesisaveryimportantproblem.Theeffectivenessofbiometricauthenticationbasedonfacemainlydependsonthemethodusedtolocatethefaceintheimage.Thispaperpresentsahybridsystemforfacesdetectioninunconstrainedcasesinwhichtheillumination,pose,occlusion,andsizeofthefaceareuncontrolled.Todothis,thenewmethodofdetectionproposedinthispaperisbasedprimarilyonatechniqueofautomaticlearningbyusingthedecisionofthreeneuralnetworks,atechniqueofenergycompactionbyusingthediscretecosinetransform,andatechniqueofsegmentationbythecolorofhumanskin.Awholeofpictures(facesandnofaces)aretransformedtovectorsofdatawhichwillbeusedforlearningtheneuralnetworkstoseparatebetweenthetwoclasses.Discretecosinetransformisusedtoreducethedimensionofthevectors,toeliminatetheredundanciesofinformation,andtostoreonlytheusefulinformationinaminimumnumberofcoefficientswhilethesegmentationisusedtoreducethespaceofresearchintheimage.Theexperimentalresultshaveshownthatthishybridizationofmethodswillgiveaverysignificantimprovementoftherateoftherecognition,qualityofdetection,andthetimeofexecution.
简介:Motivatedbytheprojectsconstrainedbyspacecapacityandresourcetransportingtime,aprojectschedulingproblemwithcapacityconstraintwasmodeled.Ahybridalgorithmisproposed,whichusestheideasofbi-levelschedulingandprojectdecompositiontechnology,andthegeneticalgorithmandtabusearchiscombined.Topologicalreorderingtechnologyisusedtoimprovetheeffciencyofevaluation.Simulationresultsshowtheproposedalgorithmcanobtainsatisfiedschedulingresultsinacceptabletime.
简介:Ahybridthinningalgorithmforbinarytopographymapsisproposedonthebasisofparallelthinningtemplatesinthispaper.Thealgorithmhasahighprocessingspeedandthestrongabilityofnoiseimmunityandpreservationofconnectivityandskeletonsymmetry.Experimentalresultsshowthatthealgorithmcansolvethethinningproblemofbinarymapseffectively.
简介:Ahybridtriplesystemofordervandindexλ,denotedbyHTS(v,λ),isapair(X,B)whereXisav-setandBisacollectionofcyclictriplesandtransitivetriplesonX,suchthateveryorderedpairofXbelongstoλtriplesofB.AnoverlargesetofdisjointHTS(v,λ),denotedbyOLHTS(v,λ),isacollection{(Y\{y},Ai)}i,suchthatYisa(v+1)-set,each(Y\{y},Ai)isanHTS(v,λ)andallAisformapartitionofallcyclictriplesandtransitivetriplesonY.Inthispaper,weshalldiscusstheexistenceproblemofOLHTS(v,λ)andgivethefollowingconclusion:thereexistsanOLHTS(v,λ)ifandonlyifλ=1,2,4,v≡0,1(mod3)andv≥4.
简介:AnegativecurvaturemethodisappliedtononlinearleastsquaresproblemswithindefiniteHessianapproximationmatrices.Withthespecialstructureofthemethod,anewswitchisproposedtoformahybridmethod.Numericalexperimentsshowthatthismethodisfeasibleandeffectiveforzero-residual,small-residualandlarge-residualproblems.
简介:Earlydefibrillationiscriticaltosurvivalinanypatientwithacardiacarrest.ElectrophysiologistswhoimplantICDs(implantablecardioverterdefi-brillators)testthembyfibrillatingthepatientfollowedbyimmediatedefibrillation.Thisemphasizesthepointthatearlydefibrillationresultsinsurvivalofthesepatients;CPRisdoneonlyifthereisadelayinobtainingthedefibrillatororiftheICDisnotworking.
简介:它为几乎所有足够地大的甚至整数n被证明那,主要可变方程n=p_1+p_2,p_1∈P_γ是可解决的,与13/15<γ≤1,在此P_γ={p|p=[m~(1/γ)],为整数m和主要p}是Piatetski-Shapiro素数的集合。
简介:Inthispaper,westudyquantumteleportationofatomicstatesviaahybridentangledstate(HES)involvinganatomandacavityfield.Andweinvestigatehowtoimplementcontrolledphase(CP)gatesbetweenatomicinternalstatesandcoherentstatesofcavityfield.Wealsodiscussthemethodsofdistinguishingcoherentstates|±α>inacavity.Finally,abriefdiscussionaboutthefeasibilityofthisschemeinexperimentispresented.
简介:在在中国和另外的国家引起两个的超级米饭的大进步在最近的年里被取得了。然而,在超级米饭繁殖有三个主要问题:1)超级米饭变化仍然是稀罕的;2)展出的很超级的米饭变化缩小适应性;并且3)当前的繁殖理论在米饭生长模型上强调了太多,但是他们在为米饭繁殖的指导是不切实际的。根据作者ther冰繁殖上的经验,包括三步的繁殖策略(繁殖的超级父母,超级杂交稻繁殖和超级杂交稻采种)被建议,并且各个的目的走,达到目标的关键技术详细被阐明。杂交稻的超级父母应该所有农学的特点地展出优秀性能,与收益或水池能力在地区性的试用到达杂交稻控制的水平。超级杂交稻联合应该满足下列标准:好米饭质量,宽改编,耐倒伏性,到主要昆虫和疾病的电阻,和收益上面在国家、省的地区性的试用在控制变化上超过了8%。达到目标,技术策略例如选择父母的最佳的联合,增加选择压力,对理想的植物类型,优秀生理性状和所有农学的特点的和谐给予更多的注意,应该被强调。采种的收益应该为联合分别地从不育系的早季节、中间季节的类型导出的超级杂交稻到达3.75t/ha和5.25t/ha。为提起采种产量的主要技术包括了选择最佳采种地点,并且由父本的集中的耕作改进花粉的数量,用有大水池能力和好outcrossing特征的不育系。根据在超级杂交稻上的三步的繁殖的技术,二更起来米饭父母,包括父本996并且ther瞬间(相片)遗传因子的男性无菌[T(P)公克]线C815S,被引起。而且,超级早杂交稻联合,Luliangyou996,能在YangtzeRiver的中间、更低的活动范围被用作一个双季节早熟稻变化,中国,被使用超级米饭引起作为父本的变化996,并且有比在地区性的试用的控制变化,两个湖南Provinceand说的高产的几杂交稻联合作为母
简介:Solution-drivenmeshadaptationisbecomingquitepopularforspatialerrorcontrolinthenumericalsimulationofcomplexcomputationalphysicsapplications,suchasclimatemodeling.Typically,spatialadaptationisachievedbyelementsubdivision(hadaptation)withaprimarygoalofresolvingthelocallengthscalesofinterest.Asec-ond,less-popularmethodofspatialadaptivityiscalled'meshmotion'(radaptation);thesmoothrepositioningofmeshnodepointsaimedatresizingexistingelementstocapturethelocallengthscales.Thispaperproposesanadaptationmethodbasedonacombinationofbothelementsubdivisionandnodepointrepositioning(rhadaptation).Bycombiningthesetwomethodsusingthenotionofamobilityfunction,theproposedapproachseekstoincreasetheflexibilityandextensibilityofmeshmotionalgorithmswhileprovidingasomewhatsmoothertransitionbetweenrefinedregionsthanispro-ducedbyelementsubdivisionalone.Further,inanattempttosupporttherequirementsofaverygeneralclassofclimatesimulationapplications,theproposedmethodisde-signedtoaccommodateunstructured,polygonalmeshtopologiesinadditiontothemostpopularmeshtypes.
简介:Hybridbraidsofpolyglycolide(PGA)andchitosanwerepreparedbythethree-yarnbraidingmethodfromPGAandchitosanfiberbundles.Thesebraidswereinvitrodegradedbyincubatingtheminphosphatebufferedsaline(PBS)atpH7.4and37℃for5weeks.ResultssuggestedthatPGA/chitosanhybridbraidsdegradedsignificantly.ScanningelectronmicrographsshowedthatchitosanfibersinthePGA/chitosanhybridbraidwithabout750%PGAinweight(PGA75/chitosan)wereshapedintogel-likeafter5weeks,butthoseinthehybridbraidwithabout250%PGAinweight(PGA25/chitosan)didnotchange.After5weeks,theultimatetensileloadsofPGAandPGA75/chitosanbraidslostalmostcompletely,butthoseofchitosanandPGA25/chitosanbraidsremainedaround14N.ThePGA/chitosanhybridbraidswithhigherinitialultimatetensileloadwouldhavepotentialapplicationsintendon/ligamenttissuereconstruction.