简介:Amachine,orothertypeof“system”,canoftenbedividedintoseveralsubsystems(components)andthesesubsystemsagaincanbedividedintosevreralsubsystems(secondgeneration).Thisprocessformsasystemtree.Toassessthereliabilityofthemachinebasedondatafromthetrialsofcomponentsofthemachine,virtualsystemmethodisemployed.ItisprovedinthepaperthatthelowerconfidentlimitofthereliabilityofthemachinesetbythevirtualsystemmethodislevelconsistentandasymptoticallyoptimalwhiletheonesetbyLindstrom-Maddensmethodisnot.
简介:LetXbeaBanachspaceand{ej}j=1beasequenceinX.Theauthorshowse∞that{ej}∞j+1isabasicsequenceifandonlyif∑n∞=1rnαnjconvergesforeveryj>1ande∞-∞∞e∑∞n=1rn∑∞j=1αnjej=∑j=1(∑∞n=1rnαnj)jholdsforeverychoiceofscalarvariables{αnj}suchthat∑∞n=1αnjejconvergesforeachn>-1andanychoiceofscalarvariables{rn}suchthat∑∞n+1∑∞f=1rnαnjejconverges.Moreover,someapplicationsabouttheresultaregiven.
简介:在图象之间的颜色转移有效地使用图象的统计信息。我们在场在图象之间的地方色彩转移的一条新奇途径基于简单统计并且局部地线性的嵌入。一个勾略的接口被建议为快速并且容易指定在目标和来源图象之间的颜色通讯。当顺利保存边界时,用户能用抄写员指定本地区域的通讯,它更精确地把目标颜色转移到来源图象,并且展出更自然的产量结果。我们的算法没被限制为一对一的图象颜色转移并且能使用超过一幅目标图象在来源图象在不同区域转移颜色。而且,我们的算法不要求选择在来源和目标图象之间的一样的颜色风格和图象尺寸。我们建议亚采样减少计算负担。与另外的途径作比较,我们的算法在掺输入数据的颜色是好一些的。我们的途径在来源图象保存另外的颜色细节。各种各样的试验性的结果证明我们的途径在来源和目标图象指定地方色彩区域的通讯。并且它表示用户的意愿并且产生视觉效果的更实际、自然的结果。
简介:
简介:Thesymmetriclinearsystemgivesusmanysimplificationsandapossibilitytoadaptthecomputationstothecomputerathandinordertoachievebetterperformance.TheaimofthispaperistoconsidertheblockbidiagonalizationmethodsderivedfromasymmetricaugmentedmultiplelinearsystemsandmakeacomparisonwiththeblockGMRESandblockbiconjugategradientmethods.
简介:ThispaperstudiestheexistenceofperiodicsolutionsfornonautonomousasymptoticallylinearHamiltoniansystems.ByusingtheZ_pindextheorysomemultiplicityresultsfornonautonomoussystemsaregiven,whichgeneralizesomeresultsforautonomoussystemsduetoAmmanandZenhder.
简介:§I.IntroductionInthispaperwearelookingforsolutionsofthefollowingHamiltoniansystemofsecondorder:wherex=(x1,x2)andVsatisfies(V.1)V:R×R2→RisaC1-function,1-periodicInt,(V.2)Visperiodicinx1withtheperiodT>0,(V.3)V→O,Vx→Oas|x2|→∞,uniformlyin(t,x1).
简介:Recurrenteventsdataandgaptimesbetweenrecurrenteventsarefrequentlyencounteredinmanyclinicalandobservationalstudies,andoftenmorethanonetypeofrecurrenteventsisofinterest.Inthispaper,weconsideraproportionalhazardsmodelformultipletyperecurrentgaptimesdatatoassesstheeffectofcovaxiatesonthecensoredeventprocessesofinterest.Anestimatingequationapproachisusedtoobtaintheestimatorsofregressioncoefficientsandbaselinecumulativehazardfunctions.Weexamineasymptoticpropertiesoftheproposedestimators.Finitesamplepropertiesoftheseestimatorsaredemonstratedbysimulations.
简介:1.IntroductionInthispaperwedealwiththeexistenceofperiodicsolutionstothedifferential-delayequations(DDEs)oftheformwherewesupposeinwhatfollowsthatexist.(itisallowedthatshouldbeincreasing.)WehaveLemmaSuppose(H)holdsandn1isaninteger.Ifthereexistsanintege...
简介:AbstractIntheparametertrackingoftime-varyingsystems,theordinarymethodisweightedleastsquareswiththerectangularwindowortheexponentialwindow.Inthispaperweproposeanewkindofslidingwindowcalledthemultipleexponentialwindow,andthenuseittofittime-varyingGaussianvectorautoregressivemodels.Theasymptoticbiasandcovarianceoftheestimatoroftheparameterfortime-invariantmodelsarealsoderived.Simulationresultsshowthatthemultipleexponentialwindowshavebetterparametertrackingeffectthanrectangularwindowsandexponentialones.