简介：设计者被要求计划让未来扩大估计格子的未来利用。有效建模和预报技术，它将高效地使用信息在可用数据，可用资料包含了的这个工具，被要求，以便重要数据性质能被提取并且投射进未来。这研究基于划分算法(MMPA)的多模型建议一个适应方法，为短期的电负担用真实数据预报。格子的利用开始用趋于增加的季节的ARIMA被建模(汽车回归的综合移动平均数)模型。建议方法经过数据使用听说并且当模特儿正常周期的行为电的格子。任何一个ARMA(汽车回归的移动平均数)或州空间的模型能被用于当模特儿的负担模式。象可以出现在夏天或意外差错(停电)期间的意外山峰那样的负担异例也被建模。如果负担模式不匹配负担的正常行为，一个异例被检测，而且，当模式匹配异例的一个已知的盒子时，异例的类型被识别。真实数据被使用，真实盒子基于测量被测试大量希腊公共力量合作S.A，雅典，希腊。过滤算法的应用适应多模型成功地识别正常周期的行为和电的格子的任何不平常的活动。建议方法的表演也与由ARIMA模型生产了那相比。

简介：Awoofer–tweeteradaptiveopticalstructuredilluminationmicroscope(AOSIM)ispresented.Bycombiningalow-spatial-frequencylarge-strokedeformablemirror(woofer)withahigh-spatial-frequencylow-strokedeformablemirror(tweeter),weareabletoremovebothlarge-amplitudeandhigh-orderaberrations.Inaddition,usingthestructuredilluminationmethod,ascomparedtowidefieldmicroscopy,theAOSIMcanaccomplishhighresolutionimagingandpossessesbettersectioningcapability.TheAOSIMwastestedbycorrectingalargeaberrationfromatriallensintheconjugateplaneofthemicroscopeobjectiveaperture.TheexperimentalresultsshowthattheAOSIMhasapointspreadfunctionwithanFWHMthatis140nmwide(usingawaterimmersionobjectivelenswithNA=1.1)aftercorrectingalargeaberration(5.9μmpeak-to-valleywavefronterrorwith2.05μmRMSaberration).Afterstructuredlightilluminationisapplied,theresultsshowthatweareabletoresolvetwobeadsthatareseparatedby145nm,1.62×belowthediffractionlimitof235nm.Furthermore,wedemonstratetheapplicationoftheAOSIMinthefieldofbioimaging.Thesampleunderinvestigationwasagreen-fluorescentprotein-labeledDrosophilaembryo.Theaberrationsfromtherefractiveindexmismatchbetweenthemicroscopeobjective,theimmersionfluid,thecoverslip,andthesampleitselfarewellcorrected.UsingAOSIMwewereabletoincreasetheSNRforourDrosophilaembryosampleby5×.

简介：Wepresentanovelalgorithmforadaptivetriangularmeshcoarsening.Thealgorithmhastwostages.First,theinputtriangularmeshisrefinedbyiterativelyapplyingtheadaptivesubdivisionoperatorthatperformsaso-calledred-greensplit.Second,therefinedmeshissimplifiedbyaclusteringalgorithmbasedoncentroidalVoronoitessellations(CVTs).TheaccuracyandgoodqualityoftheoutputtriangularmeshareachievedbycombiningadaptivesubdivisionandtheCVTstechnique.Testresultsshowedthemeshcoarseningschemetoberobustandeffective.Examplesareshownthatvalidatethemethod.

简介：

简介：

简介：Thispaperproposesanewadaptivepost-filteringalgorithmtoremovecodingartifactsinblock-basedvideocoder.Theproposedmethodconcentratesonblockingandringingartifactsremoval.Forde-blocking,theblockingstrengthisidentifiedtodeterminethefilteringrange,andthemaximumquantizationparameteroftheimageisusedtoadaptthe1Dfuzzyfilter.Forde-ringing,besidestheedgedetection,acomplementaryringingdetectionmethodisproposedtolocatetheneglectedringingblocks,andthegradientthresholdisadoptedtoadjusttheparameterof2Dfuzzyfilter.ExperimentsareperformedontheMPEG-4sequences.Comparedwithothermethods,theproposedoneachievesbetterdetailpreservationandartifactsremovalperformancewithlowercomputationalcost.

简介：Inthispaper,weconsiderthesynchronizationofchaoticHindmarshRose(HR)neuronsviaascalarcontrolinput.ChaoticHRneuronscoupledwithagapjunctionaretakenintoconsideration,andanactivecompensationmechanism-basedadaptivecontrolisemployedtorealizethesynchronizationoftwoHRneurons.Assuchanadaptivecontrolisutilized,anaccuratemodelofthesystemisofnonecessity.AsymptoticalsynchronizationoftwoHRneuronsisguaranteedbytheoreticalresults.Numericalresultsarealsoprovidedtoconfirmtheproposedsynchronizationapproach.

简介：

简介：我们建议一条新奇途径叫了适应模糊基于蚂蚁的路由(在远处)一组聪明的代理人(或蚂蚁)造在一双节点之间的路径，并发地探索网络并且交换获得的信息更新路由搁置。路由决定能被模糊逻辑技术基于另外的代理人的行为的一个以前的集合构造的当前的网络状态和知识的本地信息做。模糊逻辑技术允许象路径延期和路径利用那样的多重限制以一个简单、直觉的方法被考虑。在远处超过OSPF，AntNet和ASR的模拟试验表演，中的三个当前很重要的最先进的算法以端对端的延期，包交货，和包落下比率。在远处是为在下一代网络的数据的路由的一种有希望的选择。

简介：Videoobjectsegmentationisimportantforvideosurveillance,objecttracking,videoobjectrecognitionandvideoediting.Anadaptivevideosegmentationalgorithmbasedonhiddenconditionalrandomfields(HCRFs)isproposed,whichmodelsspatio-temporalconstraintsofvideosequence.Inordertoimprovethesegmentationquality,theweightsofspatio-temporalcon-straintsareadaptivelyupdatedbyon-linelearningforHCRFs.Shadowsarethefactorsaffectingsegmentationquality.Toseparateforegroundobjectsfromtheshadowstheycast,lineartransformforGaussiandistributionofthebackgroundisadoptedtomodeltheshadow.Theexperimentalresultsdemonstratedthattheerrorratioofouralgorithmisreducedby23%and19%respectively,comparedwiththeGaussianmixturemodel(GMM)andspatio-temporalMarkovrandomfields(MRFs).

简介：Anoveladaptive–impulsiveschemeisproposedforsynchronizingfractional-orderchaoticsystemswithoutthenecessityofknowingtheattractors’boundsinpriori.ThenonlinearfunctionsinthesesystemsaresupposedtosatisfylocalLipschitzconditionsbutwhichareestimatedwithadaptivelaws.Thenoveltyisthatthecombinationofadaptivecontrolandimpulsivecontroloffersacontrolstrategygatheringtheadvantagesofboth.Inordertoguaranteetheconvergenceisnolessthananexpectedexponentialrate,acombinedfeedbackstrengthdesigniscreatedsuchthatthesymmetricaxiscanshiftfreelyaccordingtotheupdatedtransientfeedbackstrength.AlloftheunknownLipschitzconstantsarealsoupdatedexponentiallyinthemeantimeofachievingsynchronization.Twodifferentfractional-orderchaoticsystemsareemployedtodemonstratetheeffectivenessofthenoveladaptive–impulsivecontrolscheme.

简介：Stimulatedemissiondepletion(STED)microscopyisoneoffar-fieldopticalmicroscopytechniquesthatcanprovidesub-diffractionspatialresolution.ThespatialresolutionoftheSTEDmicroscopyisdeterminedbythespeciallyengineeredbeamprofileofthedepletionbeamanditspower.However,thebeamprofileofthedepletionbeammaybedistortedduetoaberrationsofopticalsystemsandinhomogeneityofaspecimen’sopticalproperties,resultinginacompromisedspatialresolution.Thesituationgetsdeterioratedwhenthicksamplesareimaged.Intheworstcase,theseveredistortionofthedepletionbeamprofilemaycausecompletelossofthesuperresolutioneffectnomatterhowmuchdepletionpowerisappliedtospecimens.Previouslyseveraladaptiveopticsapproacheshavebeenexploredtocompensateaberrationsofsystemsandspecimens.However,itisdifficulttocorrectthecomplicatedhigh-orderopticalaberrationsofspecimens.Inthisreport,wedemonstratethatthecomplicateddistortedwavefrontfromathickphantomsamplecanbemeasuredbyusingthecoherentopticaladaptivetechnique.Thefullcorrectioncaneffectivelymaintainandimprovespatialresolutioninimagingthicksamples.

简介：Photonic-assistedmicrowavefrequencyidentificationwithdistinctfeatures,includingwidefrequencycoverageandfasttunability,hasbeenconceivedasakeytechniqueforapplicationssuchascognitiveradioanddynamicspectrumaccess.Theimplementationsbasedoncompactintegratedphotonicchipshaveexhibiteddistinctadvantagesinfootprintminiaturization,lightweight,andlowpowerconsumption,instarkcontrastwithdiscreteoptical-fiber-basedrealization.However,reportedchip-basedinstantaneousfrequencymeasurementscanonlyoperateatasingle-toneinput,whichstringentlylimitstheirpracticalapplicationsthatrequirewidebandidentificationcapabilityinmodernRFandmicrowaveapplications.Inthisarticle,wedemonstrate,forthefirsttime,awideband,adaptivemicrowavefrequencyidentificationsolutionbasedonasiliconphotonicintegratedchip,enablingtheidentificationofdifferenttypesofmicrowavesignalsfrom1to30GHz,includingsingle-frequency,multiple-frequency,chirped-frequency,andfrequency-hoppingmicrowavesignals,andeventheircombinations.ThekeycomponentisahighQ-factorscanningfilterbasedonasiliconmicroringresonator,whichisusedtoimplementfrequency-to-timemapping.Thisdemonstrationopensthedoortoamonolithicsiliconplatformthatmakespossibleawideband,adaptive,andhigh-speedsignalidentificationsubsystemwithahighresolutionandalowsize,weight,andpower(SWaP)formobileandavionicapplications.

简介：Theradialplasmapositioncontrolisoneofthebasicplasmacontrolsoftokamak.Inordertomaintainaplasmacolumninthegeometricalcenterofitsvacuumvessel,theexternallyappliedverticalfield(VF)mustbeadjustedtothechangesintheplasmaparameterssuchastheplasmacurrent,poloidalbetaandtheinternalinductance.

简介：Tofindtheoptimalroutingisalwaysanimportanttopicinwirelesssensornetworks(WSNs).ConsideringaWSNwherethenodeshavelimitedenergy,weproposeanovelEnergy*Delaymodelbasedonantalgorithms('E&DANTS'forshort)tominimizethetimedelayintransferringafixednumberofdatapacketsinanenergy-constrainedmannerinoneround.Ourgoalisnotonlytomaximizethelifetimeofthenetworkbutalsotoprovidereal-timedatatransmissionservices.However,becauseofthetradeoffofenergyanddelayinwirelessnetworksystems,thereinforcementlearning(RL)algorithmisintroducedtotrainthemodel.Inthissurvey,theparadigmofE&DANTSisexplicatedandcomparedtootherant-basedroutingalgorithmslikeAntNetandAntChainabouttheissuesofroutinginformation,routingoverheadandadaptation.SimulationresultsshowthatourmethodperformsaboutseventimesbetterthanAntNetandalsooutperformsAntChainbymorethan150%intermsofenergycostanddelayperround.

简介：Thispaperpresentsapressureobserverbasedadaptiverobustcontroller(POARC)forposturetrajectorytrackingofaparallelmanipulatordrivenbythreepneumaticmuscleswithoutpressuresensors.Duetomodelerrorsofthestaticforcesandfrictionforcesofpneumaticmuscles,simplifiedaverageflowratecharacteristicsofvalves,unknowndisturbancesofentiresystem,andunmeasuredpressures,thereexistrathersevereparametricuncertainties,nonlinearuncertaintiesanddynamicuncertaintiesinmodelingoftheparallelmanipulator.Anonlinearpressureobserverisconstructedtoestimateunknownpressuresonthebasisofasingle-input-single-output(SISO)decouplingmodelthatissimplifiedfromtheactualmultiple-input-multiple-output(MIMO)couplingmodeloftheparallelmanipulator.Then,anadaptiverobustcontrollerintegratedwiththepressureobserverisdevelopedtoaccomplishhighprecisionposturetrajectorytrackingoftheparallelmanipulator.TheexperimentalresultsindicatethatthesystemwiththeproposedPOARCnotonlyachievesgoodcontrolaccuracyandsmoothmovementbutalsomaintainsrobustnesstodisturbances.

简介：Anadaptivemeshfiniteelementmodelhasbeendevelopedtopredictthecrackpropagationdirectionaswellastocalculatethestressintensityfactors(SIFs),underlinear-elasticassumptionformixedmodeloadingapplication.Thefiniteelementmeshisgeneratedusingtheadvancingfrontmethod.Inordertosuittherequirementsofthefractureanalysis,thegenerationofthebackgroundmeshandtheconstructionofsingularelementshavebeenaddedtothedevelopedprogram.Theadaptiveremeshingprocessiscarriedoutbasedontheposterioristresserrornormschemetoobtainanoptimalmesh.Previousworksoftheauthorshaveproposedtechniquesforadaptivemeshgenerationof2Dcrackedmodels.Facilitatedbythesingularelements,thedis-placementextrapolationtechniqueisemployedtocalculatetheSIF.Thefractureismodeledbythesplittingnodeapproachandthetrajectoryfollowsthesuccessivelinearextensionsofeachcrackincrement.TheSIFsvaluesfortwodifferentcasestudieswereestimatedandvalidatedbydirectcomparisonswithotherresearcherswork.

简介：Amulti-loopadaptiveinternalmodelcontrol(IMC)strategybasedonadynamicpartialleastsquares(PLS)frame-workisproposedtoaccountforplantmodelerrorscausedbyslowaging,driftinoperationalconditions,orenvironmentalchanges.SincePLSdecompositionstructureenablesmulti-loopcontrollerdesignwithinlatentspaces,amultivariableadaptivecontrolschemecanbeconvertedeasilyintoseveralindependentunivariablecontrolloopsinthePLSspace.Ineachlatentsubspace,oncethemodelerrorexceedsaspecificthreshold,onlineadaptationrulesareimplementedseparatelytocorrecttheplantmodelmismatchviaarecursiveleastsquares(RLS)algorithm.BecausetheIMCextractstheinverseoftheminimumpartoftheinternalmodelasitsstructure,theIMCcontrollerisself-tunedbyexplicitlyupdatingtheparameters,whicharepartsoftheinternalmodel.Bothparameterconvergenceandsystemstabilityarebrieflyanalyzed,andprovedtobeeffective.Finally,theproposedcontrolschemeistestedandevaluatedusingawidely-usedbenchmarkofamulti-inputmulti-output(MIMO)systemwithpuredelay.

简介：一个光陷井能在在一个光洞嵌入的一个ultracold原子系统由一块轻地被印，是众所周知的，并且由三块不同的协调的地开车。在光洞共存的三块地中，有一块强烈控制地，经由创造与联系Rabi频率的平方成正比的力量的一个周期的光栅栏的导致electromagnetically的透明性，和另一块地导致巨大的克尔非线性。在这，管理探查域的繁殖的非线性的方程的工作elliptic-soliton答案被考虑，与形成有明确的量数字的一个分离系列的光soliton火车的可能的产生上的强调。这个问题被对待假定周期的光栅栏的二种不同类型并且考虑否定并且detunings(detuning在上面或在回声下面)的正号。结果预言在自我阶段和跨阶段的调整非线性之间的比赛产生不同的量数字描绘的时间的soliton火车模式的一个富有的家庭。

简介：Withtherapidgrowthofwirelessbroadbandtechnologies,suchasWLANandWiMAX,qualitystreamingvideocontentsareavailablethroughportabledevicesanytime,anywhere.Thelayeredmulticastsystemusingscalablevideocodecshasbeenproposedasanefficientarchitectureforvideodisseminationtakingaccountofuserandlinkdiversities.However,inthewired/wirelesscombinedbest-effortbasedheterogeneousIPnetworkswhichprovidemorefluctuationinavailablebandwidthandend-to-enddelay,theperformanceofstreamingsystemshasbeengreatlydegradedduetofrequentpacketloss,resultingfromeitherwiredcongestionorwirelessfading/shadowing.Inthispaper,wepresentareal-timeembeddedpackettrainprobingschemeforestimatingend-to-endavailablebandwidthsoastoaccomplisheffectivecongestionanderrorcontrol.Thisisfacilitatedbyeffectiveclassificationofpacketlosssources,delaytrenddetectionalgorithmandflexibletransmissionrateofpackets.Undertheproperwirelesschannelmodellingandestimation,ourlayeredstructurecanallowappropriatesubscriptionofvideolayersandadaptivelyinsertnecessaryamountofforwarderrorcorrection(FEC)packetssoastoachieveQoSoptimizedsystemforscalablevideomulticasting.