简介:Inthispaper,afuzzyreasoningbasedtemporalerrorconcealmentmethodisproposed.ThebasictemporalerrorconcealmentisimplementedbyestimatingMotionVector(MV)ofthelostMacroBlock(MB)fromitsneighboringMVs.WhichMVisthemostproperoneisevaluatedbysomecriteria.Generally,twocriteriaarewidelyused,namelySideMatchDistortion(SMD)andSumofAbsoluteDifference(SAD)ofcorrespondingMV.However,eachcriterioncouldonlypartlydescribethestatusoflostblock.Toaccomplishthejudgementmoreaccurately,thetwomeasuresareconsideredtogether.ThusarefinedmeasurebasedonfuzzyreasoningisadoptedtobalancetheeffectsofSMDandSAD.TermsSMDandSADareregardedasfuzzyinputandtheterm‘similarity’asoutputtocompletefuzzyreasoning.ResultoffuzzyreasoningrepresentshowthetestedMVissimilartotheoriginalone.Andk-meansclusteringtechniqueisperformedtodefinethemembershipfunctionofinputfuzzysetsadaptively.Accordingtotheexperimentalresults,theconcealmentbasedonnewmeasureachievesbetterperformance.
简介:这份报纸论述了由与一个概念的水文学模型一起集成一个模糊聚类模型和神经网络预报框架的新分类即时洪水。一个模糊聚类模型被用来以洪水山峰和流量深度分类历史的洪水,并且概念的水文学模型为洪水的每个班被校准。背繁殖(BP)神经网络被从模糊聚类模型使用即时降雨数据和产量训练。BP神经网络为即时洪水事件提供了一个快速的联机分类。基于联机分类,水文学模型的一个适当参数集合自动地被选择生产即时洪水预报。不同参数集合连续地在因为即时降雨数据和联机分类结果的变化,预报过程的洪水被使用。建议方法论在辽宁省被用于大集水,中国。结果证明分类框架比传统的非分类的方法提供了更精确的预言。而且,在模糊聚类的不同索引重量的效果也被讨论。
简介:隐藏的Markov建模的侧面(HMMs)广泛地基于古典HMMs被申请了蛋白质顺序鉴定。在侧面HMMs的前面、向后的变量的明确的表达在概率理论的统计独立假设下面被做。我们建议模糊侧面唔定序克服那个假设的限制并且为蛋白质完成改进排列属于一个给定的家庭。建议模型fuzzifies由合并Sugeno的前面、向后的变量模糊措施和Choquet积分,进一步因此延长概括唔。把前面、向后的变量基于fuzzified,我们为侧面建议一个模糊Baum-Welch参数评价算法。强壮的关联和涉及结构使这模糊体系结构基于的蛋白质的顺序偏爱作为造一个给定的家庭的侧面的一个合适的候选人当模特儿,自从模糊集合能比古典方法更好处理无常。
简介:针对车辆在复杂的路况下行驶时,车辆自动变速器的频繁换挡和车辆动力不足的问题,设计了一种基于升、降档模糊控制器基本模块的换挡策略.升降挡模糊控制器模块是以加速度的正负作为控制参量,加速度为正时,基于升档规律曲线的升档模糊控制器起作用;加速度为负时,基于降档规律曲线的降档模糊控制器起作用.通过StateFlow的逻辑来判断哪个控制器起作用.在上述基本模块的基础上,建立了以制动力的大小,弯道大小以及坡度大小作为控制输入参数的模糊修正模块,其输出对上述基本模块的升档模糊控制器和降档模糊控制器的输出进行修正.仿真结果表明:所设计的换挡控制策略具有很好的避免频繁换挡和动力不足的问题.
简介:AnovelH_∞tracking-baseddecentralizedindirectadaptiveoutputfeedbackfuzzycontrollerforaclassofuncertainlarge-scalenonlinearsystemsisdeveloped.Byvirtueoftheproperfilteringoftheobservationerrordynamics,theobserver-baseddecentralizedindirectadaptivefuzzycontrolschemeispresentedforaclassoflarge-scalenonlinearsystemsusingthecombinationofH_∞trackingtechnique,afuzzyadaptiveobserverandfuzzyinferencesystems.Theoutputfeedbackandadaptationmechanismsarebothrobustandimplementableindeedowingtotheirfreedomfromtheunavailableobservationerrorvector.Allthesignalsoftheclosed-looplargescalesystemareguaranteedtostayuniformlyboundedandtheoutputerrorstakeonH_∞trackingperformance.Simulationresultssubstantiatetheeffectivenessoftheproposedscheme.
简介:Uncertainandhesitantinformation,widelyexistinginthereal-worldqualitativedecisionmakingproblems,bringsgreatchallengestodecisionmakers.Hesitantfuzzylinguistictermsets(HFLTSs),aneffectivelinguisticcomputationaltoolinmodelingandelicitingsuchinformation,havehencearousedmanyscholars’interestsandsomeextensionshavebeenintroducedrecently.However,thesemethodsarebasedonthediscretelinguistictermframeworkwiththelimitedexpressiondomain,whichactuallydepictqualitativeinformationusingseveralsinglevalues.Therefore,itishardtoensuretheintegrityofthesemanticsrepresentationandtheaccuracyofthecomputationresults.Todealwiththisproblem,asemanticsbasisframeworkcalledcompletelinguistictermset(CLTS)isdesigned,whichadoptsaseparationstructureoflinguisticscaleandexpressiondomain,enrichingsemanticsrepresentationofdecisionmakers.Onthisbasistheconceptoffuzzyintervallinguisticsets(FILSs)isputforwardthatemploystheintervallinguistictermwithprobabilitytoincreasetheflexibilityofelicitingandrepresentinguncertainandhesitantqualitativeinformation.Forpracticalapplications,afuzzyintervallinguistictechniquefororderpreferencebysimilaritytoidealsolution(FILTOPSIS)methodisdevelopedtodealwithmulti-attributegroupdecisionmaking(MAGDM)problems.Throughthecasesofmovieandenterpriseresourceplanning(ERP)systemselection,theeffectivenessandvalidityoftheproposedmethodareillustrated.
简介:Aimingtoreducethecomputationalcostsandconvergetoglobaloptimum,anovelmethodisproposedtosolvetheoptimizationofacostfunctionintheestimationofdirectionofarrival(DOA).Inthismethod,ageneticalgorithm(GA)andfuzzydiscreteparticleswarmoptimization(FDPSO)areappliedtooptimizethedirectionofarrivalandpowerparametersofthemodesimultaneously.Firstly,theGAalgorithmisappliedtomakethesolutionfallintotheglobalsearching.Secondly,theFDPSOmethodisutilizedtonarrowdownthesearchfield.InFDPSO,achaoticfactorandacrossovermethodareaddedtospeeduptheconvergence.Thisapproachhasbeendemonstratedthroughsomecomputationalsimulations.ItisshownthattheproposedalgorithmcanestimateboththeDOAandthepowersaccurately.Itismoreefficientthansomepresentmethods,suchastheNewton-likealgorithm,Akaikeinformationcritical(AIC),particleswarmoptimization(PSO),andgeneticalgorithmwithparticleswarmoptimization(GA-PSO).
简介:这篇论文为遥远地察觉到的数据,结果的分类的模糊的度被作为由一个常规算法与那作比较减少聚类的fuzzyc工具描述一个改进算法:也就是说,分类精确性被增加。这被在单个班的水平合并协变性矩阵而非假定全球的完成。从一个爱丁堡郊外的陆地封面的一个模糊分类的实验结果为聚类的fuzzyc工具证实了新算法的改进性能,特别地当模糊也在假定参考书数据被提供时。
简介:Thetraditionaluser-optimalassignmentmodelassumedthatthetrafficinformationisperfect,howerver,thelinktravelcostisAlwaysuncertainorfuzzyinrealsituations.ThepapergivesamodifiedFrank-Wolfemethodforfuzzyuser-optimalroutechoiceproblem.Theshortest-pathalgorithminF-Wissolvedbyanorderrelation(FSA)methodbetweenfuzzytravelcosts.Theresultsofasimpleexampleshowthatthemethodispractical.