简介:SIGN-REVERSALPROPERTYFORF-MATRICESANDN-MATRICESXUSong(GraduateSchool,ChinaAcademyofRailwaySciences,Beijing100081,China)Abstra...
简介:ANASYMPTOTICPROPERTYOFTHENUMBEROFSPANNINGTREESOFDOUBLEFIXEDSTEPLOOPNETWORKSYONGXUERONGANDZHANGFUJIAbstract.LetT(C(p,1,q))bet...
简介:OnCopositiveInthispapertheauthorwritesasimplecharacterizationforthebestcopositiveapproximationtoelementsofC(Q)byelementsoffinitedimensionalstrictChebyshevsubspacesofC(Q)inthecasewhenQisanycompactsubsetofrealnumbers.AttheendofthepapertheauthorappliesthisresultfordifferentclassesofQ.
简介:Thispaperreportsthesynthesisofaseriesofpolysiloxanesbearinglongsidechainmeltedatlowtemperature.Thedrug(LNG)releasingpropertiesofthesiciconerubberfilmscontainingthesepolyorganosiloxaneswerestudied.Thereleaseratesofthedrugincreaseasthetemperatureriseupabovethemeltingtemperatureofthesepolyorganosiloxanes.
简介:Biomedicalsegmentedpolyetherurethane(SPEU)elastomersarecompositepoly-merswithmicrophaseseparationstructure.VariousattemptsofintroducingionicgroupsintoSPEUhavebeencarriedoutfortheimprovementoftheirbloodcompati-bilities.Weherereportthemechanicalandanticoagulantproperties.Twoseriesofsulfooctylpendantgroupscontainingsegmentedpolyetherurethaneswithwelldefinedfinestructurehavebeensynthesizedinourlab-oratory.ThesulfooctylgroupswereexactlylocatedatthehardsegmentswhentheSPEUswerepreparedfromthereactionofpolytetramethyleneglycol(PTMG)1000-4,4’diphenylmethanediisocyanate(MDI)oligomerswithN-sulfoalkyldiethanolamine
简介:Silica-boundmercaptoselenaether-containingsilsesquioxaneanditsplat-inumorrhodiumcomplexweresynthesizedfromγ-(2,3-epithiopropyl)propyltrimethoxysi-laneviaring-openingreactionwithbis(2-hydroxyethyl)sclenide,followedbyhydrolysisandimmobilizationonfumedsilica,andthenreactingwithpotassiumchloroplatiniteorrhodiumchlorideinacetoneundernitrogenatmosphere.Itwasfoundthattwonoblemetalcomplexescancatalyzethehydrosilylationofolefinswithtriethoxysilaneeffectively.Theinfluencesoftemperature,theamountofcomplexused,thenatureofolefinoncatalyticactivitywereinvestigated.
简介:Inthispaperthedesignproblemofperfect-reconstructioncosine-modulatedQMFbankshasbeenformulatedasaquadratic-constrainedleast-squares(QCLS)minimizationprobleminwhichallconstrainedmatricesoftheQCLSoptimizationproblemaresymmetricandpositivedefinite.AcostfunctionwhichisaconvexfunctionofdesiredprototypefiltercoefficientsisconstructedsothatthiskindofQCLSoptimizationproblemscanbeefficientlysolved.Soaglobalminimizerofthisproblemcanbeeasilyobtained.Resultsoftwodesignexamplesarepresentedtosupportthederivationsandanalyses.
简介:Cloudmicrophysicalpropertyretrievalsfromtheactivemicrowaveinstrumentonasatelliterequiretheclouddropletsizedistributionobtainedfromaircraftobservationsasaprioridataintheiterationprocedure.Thecloudlognormalsizedistributionsderivedfrom12flightsoverBeijing,China,in2008–09werecharacterizedtoevaluateandimproveregionalCloudSatcloudwatercontentretrievals.Wepresentthedistributionparametersofstratiformclouddroplet(diameter<500μmand<1500μm)anddiscusstheeffectoflargeparticlesondistributionparameterfitting.Basedonthreeretrievalschemeswithdifferentlognormalsizedistributionparameters,theverticaldistributionofcloudliquidandicewatercontentwerederivedandthencomparedwiththeaircraftobservations.Theresultsshowedthattheliquidwatercontent(LWC)retrievalsfromlargeparticlesizedistributionsweremoreconsistentwiththeverticaldistributionofcloudwatercontentprofilesderivedfrominsitudataon25September2006.WethenappliedtwoschemeswithdifferentaprioridataderivedfromflightdatatoCloudSatoverpassesinnorthernChinaduringApril–Octoberin2008and2009.TheCloudSatcloudwaterpath(CWP)retrievalswerecomparedwithModerateResolutionImagingSpectroradiometer(MODIS)liquidwaterpath(LWP)data.TheresultsindicatedthatconsideringaprioridataincludinglargeparticlesizeinformationcansignificantlyimprovetheconsistencybetweentheCloudSatCWPandMODISCWP.Theseresultsstronglysuggestthatitisnecessarytoconsiderparticleswithdiametersgreaterthan50μminCloudSatLWCretrievals.
简介:Basedonthemethodofartificialneuralnetwork,anewapproachhasbeendevisedtopredictthemechanicalpropertyofE4303electrode.Theoutlinedpredicationmodelfordeterminingthemechanicalpropertyofelectrodewasbuiltupontheproductiondata.Theresearchleveragesabackpropagationalgorithmastheneuralnetwork'slearningrule.Theresultindicatesthattherearepositivecorrelationsbetweenthepredictedresultsandthepracticalproductiondata.Hence,usingtheneuralnetwork,predicationofelectrodepropertycanberealized.Forthefirsttime,thisresearchprovidesamorescientificmethodfordesigningelectrode.