简介：Anoveldistributedmodelpredictivecontrolschemebasedondynamicintegratedsystemoptimizationandparameterestimation(DISOPE)wasproposedfornonlinearcascadesystemsundernetworkenvironment.Underthedistributedcontrolstructure,onlineoptimizationofthecascadesystemwascomposedofseveralcascadedagentsthatcancooperateandexchangeinformationvianetworkcommunication.Byiteratingonmodifieddistributedlinearoptimalcontrolproblemsonthebasisofestimatingparametersateveryiterationthecorrectoptimalcontrolactionofthenonlinearmodelpredictivecontrolproblemofthecascadesystemcouldbeobtained,assumingthatthealgorithmwasconvergent.Thisapproachavoidssolvingthecomplexnonlinearoptimizationproblemandsignificantlyreducesthecomputationalburden.Thesimulationresultsofthefossilfuelpowerunitareillustratedtoverifytheeffectivenessandpracticabilityoftheproposedalgorithm.

简介：这篇论文在指导网络调查控制动态代理人的全球行为。代理人是Lyapunov马厩，在一根线被散布，并且通过一个指导网络交流。网络的通讯拓扑学被一张指导的图描绘，控制协议在简单线性分散的反馈法律被设计。我们学习代理人将分别地在下面完成聚集，和批评、分叉的轨道的不同条件。我们在网络下面的动态代理人系统上的调查被扩大到时间延期网络案例。而且，我们与二学习案例在系统的预先指定的虚拟领导人。数字模拟被给并且证明我们的理论结果是有效的。

简介：Frictiondragprimarilydeterminesthetotaldragoftransportsystems.ApromisingapproachtoreducedragathighReynoldsnumbers(>104)areactivetransversalsurfacewavesincombinationwithpassivemethodslikearibletsurface.Fortheapplicationintransportationsystemswithlargesurfacessuchasairplanes,shipsortrains,alargescaledistributedreal-timeactuatorandsensornetworkisrequired.Thisnetworkisresponsibleforprovidingconnectionsbetweenaglobalflowcontrolanddistributedactuatorsandsensors.ForthedevelopmentofthisnetworkweestablishedatfirstasmallscalenetworkmodelbasedonSimulinkandTrueTime.TodeterminetimescalesfornetworkeventsondifferentpackagesizeswesetupaRaspberryPibasedtestbedasaphysicalrepresentationofourfirstmodel.Thesetimescalesarereducedtotimedifferencesbetweenthedeterministicnetworkeventstoverifythebehaviorofourmodel.Experimentalresultswereimprovedbysynchronizingthetestbedwithsufficientprecision.Withthisapproachweassurealinkbetweenthelargescalemodelandthelaterconstructedmicrocontrollerbasedreal-timeactuatorandsensornetworkfordistributedactiveturbulentflowcontrol.

简介：为优化问题的神经网络有的设计类型为更少参数，低寻找空格尺寸和简单结构在另外的网络上有利。在这篇论文，由适当地构造Lyapunov精力功能，当被用来优化在一个关上的凸的集合上定义的连续地可辨的凸的功能时，我们证明了这个网络的全球集中。结果解决网络的广泛的适用性。几个数字例子被给验证网络的效率。

简介：这篇论文调查一个概括复杂动态网络模型的本地、全球的同步与经常、推迟联合。没有假定政变石楠的对称，我们证明一个单个控制器能卡住概括复杂网络到一个同质的答案。一些以前的同步结果被概括。在这篇论文，我们首先讨论怎么由增加仅仅一个控制器卡住一连串的推迟的神经网络到同步解决方案。下次，由使用Lyapunov功能的方法，一些足够的条件为联合系统的本地、全球的同步被导出。获得的结果以LMI被表示，它能被MatlabLMI工具箱高效地检查。最后，一个例子被给说明理论结果。

简介：Inthispaper,weintroduceageneralframeworkfortrackinginleader-followersystemsundercommunicationconstraints,inwhichtheleaderandfollowersystemsaswellasthecorrespondingcontrollersarespatiallydistributedandconnectedovercommunicationlinks.Weprovidenecessaryconditionsonthechanneldatarateofeachcommunicationlinkfortrackingoftheleader-followersystems.Byconsideringtheforwardandfeedbackchannelsasonecascadechannel,wealsoprovidealowerboundforthedatarateofthecascadechannelforthesystemtotrackareferencesignalsuchthatthetrackingerrorhasfinitesecondmoment.Examplesandsimulationsareprovidedtodemonstratesomeoftheresults.

简介：Afuzzyneuralnetworkcontrollerforunderwatervehicleshasmanyparametersdifficulttotunemanually.Toreducethenumerousworkandsubjectiveuncertaintiesinmanualadjustments,ahybridparticleswarmoptimization(HPSO)algorithmbasedonimmunetheoryandnonlineardecreasinginertiaweight(NDIW)strategyisproposed.OwingtotherestraintfactorandNDIWstrategy,anHPSOalgorithmcaneffectivelypreventprematureconvergenceandkeepbalancebetweenglobalandlocalsearchingabilities.Meanwhile,thealgorithmmaintainstheabilityofhandlingmultimodalandmultidimensionalproblems.TheHPSOalgorithmhasthefastestconvergencevelocityandfindsthebestsolutionscomparedtoGA,IGA,andbasicPSOalgorithminsimulationexperiments.ExperimentalresultsontheAUVsimulationplatformshowthatHPSO-basedcontrollersperformwellandhavestrongabilitiesagainstcurrentdisturbance.ItcanthusbeconcludedthattheproposedalgorithmisfeasibleforapplicationtoAUVs.

简介：一个控制器为其在对给定的值作出回应特征的输出快速并且聪明地在焊接进程被控制的弧焊力量来源基于一个PID神经网络(PIDNN)被建议。新方法调和PID控制策略和神经网络聪明地控制焊接过程，因此它有PID控制规则和神经网络的更好的信息处理能力的特点的优点。模拟的结果证明控制器有快反应的性质，低射，快集中和好稳定的精确性，哪个会为焊接进程的控制的要求。

简介：这份报纸考虑为一个关上的线性量系统构造一个直接联合量观察员的问题。建议分布式的观察员由观察员网络在一时间收敛到一致，这平均的振荡器和它被显示出的量泛音的一个网络组成观察员的每个元素在估计量工厂的指定输出的感觉。一个例子和模拟被包括说明观察员网络的性质。

简介：Thispaperproposesanovelideathatclassifiesfaultsintotwodifferentkinds:seriousfaultsandsmallfaults,andtreatsthemwithdifferentstrategiesrespectively.Akindofartificialneuralnetwork(ANN)isproposedfordetectingseriousfaults,andvariablestructure(VS)model-followingcontrolisconstructedforaccommodatingsmallfaults.Theproposedframeworktakesbothadvantagesofqualitativewayandquantitativewayoffaultdetectionandaccommodation.Moreover,theuncertaintycaseisinvestigatedandtheVScontrollerismodified.Simulationresultsofaremotelypilotedaircraftwithcontrolactuatorfailuresillustratetheperformanceofthedevelopedalgorithm.

简介：一个新奇概率的模糊控制系统被建议在控制协议(TCP)联网的传播对待拥挤回避问题。TCP网络的交通测量上的研究证明了包交通展出称为自我类似的长期的依赖性质，它降级网络表演极大地。概率的模糊控制(陆军)系统被用来在网络系统处理自我类似的交通和当模特儿的不确定性的复杂随机的特征。一三维(3-D)会员功能(MF)在PFC被嵌入表示并且描述网络交通的随机的特征。3-DMF延长了传统模糊平面印射并且进一步提供在“fuzziness-randomness-state”之中的空间印射。3-DMF的另外的随机的表示提供PFC处理自我类似的交通的随机的特征的另外的自由。模拟实验证明建议控制方法在随机的环境与传统的控制计划相比完成优异性能。

简介：Inordertocontrolthelarge-scaleurbantrafficnetworkthroughhierarchicalordecentralizedmethods,itisnecessarytoexploitanetworkpartitionmethod,whichshouldbebotheffectiveinextractingsubnetworksandfasttocompute.Inthispaper,anewapproachtocalculatethecorrelationdegree,whichdeterminesthedesireforinterconnectionbetweentwoadjacentintersections,isfirstproposed.Itisusedasaweightofalinkinanurbantrafficnetwork,whichconsidersboththephysicalcharacteristicsandthedynamictrafficinformationofthelink.Then,afastnetworkdivisionapproachbyoptimizingthemodularity,whichisacriteriontodistinguishthequalityofthepartitionresults,isappliedtoidentifythesubnetworksforlarge-scaleurbantrafficnetworks.Finally,anapplicationtoaspecifiedurbantrafficnetworkisinvestigatedusingtheproposedalgorithm.Theresultsshowthatitisaneffectiveandefficientmethodforpartitioningurbantrafficnetworksautomaticallyinrealworld.

简介：Thispaperaddressestheproblemofcoordinatingmultiplemobilerobotsinsearchingforandcapturingamobiletarget,withtheaimofreducingthecapturetime.Comparedwiththepreviousalgorithms,weassumethatthetargetcanbedetectedbyanyrobotandcapturedsuccessfullybytwoormorerobots.Inthispaper,weassumethateachrobothasalimitedcommunicationrange.Wemaintaintherobotswithinamobilenetworktoguaranteethesuccessfulcapture.Inaddition,themotionofthetargetismodeledandincorporatedintodirectingthemotionoftherobotstoreducethecapturetime.Acoordinationalgorithmconsideringbothaspectsisproposed.Thisalgorithmcangreatlyreducetheexpectedtimeofcapturingthemobiletarget.Finally,wevalidatethealgorithmbythesimulationsandexperiments.

简介：这篇论文基于联网的控制系统为延期系统的一个类集中于稳定性分析和控制器设计的问题。由介绍一些免费矩阵变量，为稳定性分析和基于观察员的控制法律设计的一些标准能被解决线性矩阵不平等获得。一个数字例子也被提供证明建议方法的有效性。

简介：Duetotheenergyandresourceconstraintsofawirelesssensornodeinawirelesssensornetwork(WSN),designofenergy-efficientmultipathroutingprotocolsisacrucialconcernforWSNapplications.Toprovidehigh-qualitymonitoringinformation,manyWSNapplicationsrequirehigh-ratedatatransmission.Multipathroutingprotocolsareoftenusedtoincreasethenetworktransmissionrateandthroughput.Althoughlarge-scaleWSNcanbesupportedbyhighbandwidthbackbonenetwork,theWSNremainsthebot...

简介：这篇论文与导致网络的延期和播送数据退学学生的存在处理联网的控制系统的随机的稳定性。基于Lyapunov途径，为联网的控制系统的吝啬平方的稳定性的足够的条件是传播间隔的顺序被开车由的导出的题目一独立地相等散布了顺序并且由一个有限状态，Markov分别地锁住。稳定控制器相应地以线性矩阵不平等被构造。一个例子被提供说明我们的结果。

简介：Approximatedynamicprogramming(ADP)formulationimplementedwithanadaptivecritic(AC)-basedneuralnetwork(NN)structurehasevolvedasapowerfultechniqueforsolvingtheHamilton-Jacobi-Bellman(HJB)equations.AsinterestinADPandtheACsolutionsareescalatingwithtime,thereisadireneedtoconsiderpossibleenablingfactorsfortheirimplementations.AtypicalACstructureconsistsoftwointeractingNNs,whichiscomputationallyexpensive.Inthispaper,anewarchitecture,calledthe'cost-functio...

简介：Duetoitscharacteroftopologyindependency,topology-transparentmediumaccesscontrol(MAC)schedulingalgorithmisverysuitableforlarge-scalemobileadhocwirelessnetworks.Inthispaper,weproposeanewtopologytransparentMACschedulingalgorithm,withparametersofthenodenumberandthemaximalnodaldegreeknown,ourschedulingalgorithmisbasedonaspecialbalancedincompleteblockdesignwhoseblocksizeisoptimizedbymaximizingtheguaranteedthroughput.Itssuperiorityovertypical...

简介：Thispaperaddressestheimportantintelligentpredictingproblemofperitonealabsorptionrateintheperitonealdialysistreamentprocessofrenalfailure.Astheindexofdialysisadequacy,KT/VandCcrarewidelyusedandaccepted.However,growingevidencesuggeststhatthefluidbalancemayplayacriticalroleindialysisadequacyandpatientoutcome.Peritonealfluidabsorptiondecreasestheperitonealfluidremoval.Understandingtheperitonealfluidabsorptionratewillhelpclinicianstoopthnizethedialysisdwelltime.Theneuralnetworkapproachisappliedtothepredictionofperitonealabsorptionrate.Comparedwithmultivariableregressionmethod,theexperimentalresultsshowedthatneuralnetworkmethodhasanadvantageovermultivariableregression.Theapplicationofthispredictingmethodbased-onneuralnetworkinclinicisinstructive.

简介：这篇论文扩大适应神经网络(NN)控制途径到未知输出反馈的一个类非线性的时间延期系统。追踪控制器的适应产量反馈NN被走技术的背设计。NN准时习惯于近似未知功能依赖者延期。延期依赖者过滤器为州的评价被介绍。支配方法被用来处理光滑的时间延期基础功能。适应的跳技术被采用估计NN近似错误的上面的界限。基于Lyapunov-Krasovskii功能，在靠近环的系统的所有信号的半全球的一致最终的围住的海角被证明。可行性被二个解说性的模拟例子调查。