简介:Spatialcoherenceresonanceinatwo-dimensionalneuronalnetworkinducedbyadditiveGaussiancolourednoiseandparameterdiversityisstudied.WefocusontheabilityofadditiveGaussiancolourednoiseandparameterdiversitytoextractaparticularspatialfrequency(wavenumber)ofexcitatorywavesintheexcitablemediumofthisnetwork.Weshowthatthereexistsanintermediatenoiselevelofthecolourednoiseandaparticularvalueofdiversity,whereacharacteristicspatialfrequencyofthesystemcomesforth.Hereby,itisverifiedthatspatialcoherenceresonanceoccursinthestudiedmodel.Furthermore,weshowthattheoptimalnoiseintensityforspatialcoherenceresonancedecaysexponentiallywithrespecttothenoisecorrelationtime.Someexplanationsoftheobservednonlinearphenomenaarealsopresented.
简介:Inviewofthepoorwatersupplysystem'snetworkproperties,thesystem'scomplicatednetworkhydraulicequationswerereplacedbymacroscopicnodalpressuremodelandthemodelofrelationshipbetweensupplyflowandwatersourcehead.Byusingpump-stationpressureheadandinitialtankwaterlevelsasdecisionvariables,themodelofoptimalallocationofwatersupplybetweenpump-sourceswasdeveloped.Geneticalgorithmwasintroducedtodealwiththemodelofoptimalallocationofwatersupply.Methodsforhandlingeachconstraintconditionwereputforward,andovercometheshortcomingsuchasprematureconvergenceofgeneticalgorithm;asolvingmethodwasbroughtforwardinwhichgeneticalgorithmwascombinedwithsimulatedannealingtechnologyandself-adaptivecrossoverandmutationprobabilitieswereadopted.Anapplicationexampleshowedthefeasibilityofthisalgorithm.
简介:Inthispaper,anewzigzagmethodforplatestructuresandageneticalgorithm(GA)ofdynamicsourceseedspacesaredevelopedandacombinationofthemisusedtodealwithlargescalebuilt-upstructuraloptimization.ThenewGAcombinedwiththezigzagmethodcanworkefficientlywhencopingwithlargescalestructuraloptimizationincludeddisplacementandstressconstraints.ExamplesshowthatthisGAisrobustandcanbeusedformanycomplexstructuraloptimizationproblems.
简介:Tubinecascadeoptimumdesign,thetypicalnon-convexoptimalproblem,haslongbeenadesignchallengeintheengineeringfields.ThenewtypehybridGeneticAlgorithms-wholeannealingGeneticAlgorithmshavebeendevelopedinthispaper,Simulatedannealingselectionandnon-uniformmutationareadoptedinthewholeannealingGeneticAlgorithms.WholeannealingGeneticAlgorithmsoptimalperformancehavebeentestedthroughmathematicaltestfunctions.Onthisbasis,turbinecascadeinversedesignusingwholeannealingGeneticAlgorithmshaebeenpresented.TheB-Splinefunctionisappliedtorepresentthecascadeshape,C-typegridandGodunovschemeareadoptedtoanalysisthecascadeaerodynamicperformance.Theoptimalproblemaimstoobtainancascadeshapefromdifferentinitialcascadethroughthegiventargetpressuredistribution.Theoptimumcascadeshapeisinwellagreementwiththetargetcascadeshape.ThenumericalresultsshowthatthewholeannealingGeneticAlgorithmsarethepowerfuloptimumtoolsforturbineoptimumdesignorothercomplexengineeringdesignproblems.
简介:怪物状态在1D数组被学习了,并且许多不同怪物状态用不同模型被发现了。研究最近被扩大了到2D数组,但是仅仅到他们的阶段模型。这里,我们把它递非局部的联合的影响上的神经原和焦点的2D数组的一个nonphase模型。用广泛的数字模拟,令人惊讶地,我们发现这个系统能显示出以前观察的怪物状态的大多数类型,与以前的模型相对照,在怪物状态的仅仅或一些类型能在每个模型被观察的地方。我们也发现这个模型能显示出象gridding那样的一些特殊像怪物的模式和multicolumn模式,它以前仅仅在阶段模型被观察。进一步,我们在场一条有效途径,即,移开联合的一些连接,产生异构的联合,它导致多样的像怪物的模式并且甚至从一个像怪物的模式导致转变到另外一个。
简介:Anapproachforgeneratinginteractive3Dgraphicalvisualizationofthegeneticarchitecturesofcomplextraitsinmultipleenvironmentsisdescribed.3Dgraphicalvisualizationisutilizedformakingimprovementsontraditionalplotsinquan-titativetraitlocus(QTL)mappinganalysis.Interactive3DgraphicalvisualizationforabstractexpressionofQTL,epistasisandtheirenvironmentalinteractionsforexperimentalpopulationswasdevelopedinframeworkofuser-friendlysoftwareQTLNetwork(http://ibi.zju.edu.cn/software/qtlnetwork).Noveldefinitionofgraphicalmetasystemandcomputationofvirtualcoordinatesareusedtoachieveexplicitbutmeaningfulvisualization.Interactive3DgraphicalvisualizationforQTLanalysisprovidesgeneticistsandbreedersapowerfulandeasy-to-usetooltoanalyzeandpublishtheirresearchresults.
简介:
简介:Thealternatecombinationalapproachofgeneticalgorithmandneuralnetwork(AGANN)hasbeenpresentedtocorrectthesystematicerrorofthedensityfunctionaltheory(DFT)calculation.IttreatstheDFTasablackboxandmodelstheerrorthroughexternalstatisticalinformation.Asademonstration,theAGANNmethodhasbeenappliedinthecorrectionofthelatticeenergiesfromtheDFTcalculationfor72metalhalidesandhydrides.ThroughtheAGANNcorrection,themeanabsolutevalueoftherelativeerrorsofthecalculatedlatticeenergiestotheexperimentalvaluesdecreasesfrom4.93%to1.20%inthetestingset.Forcomparison,theneuralnetworkapproachreducesthemeanvalueto2.56%.Andforthecommoncombinationalapproachofgeneticalgorithmandneuralnetwork,thevaluedropsto2.15%.Themultiplelinearregressionmethodalmosthasnocorrectioneffecthere.
简介:Thebandgap,animportantcharacteristicoftheperiodicstructure,isdispersionrelated,whichcanbedesignedbytailoringthelayoutofmaterialswithintheperiodicmicrostructures.Atypicalexampleofaperiodicstructureisphononiccrystals(PnCs),whicharetraditionallyfabricatedfromtwo-phasematerials.Herein,weinvestigatethetopologiesofperiodicthree-phasePnCs.Themicrostructuresofthethree-phasePnCsareoptimizedusingatwo-stagegeneticalgorithm,andthreecasestudiesareproposedtoobtainthefollowing:(1)themaximumrelativebandgapwidth,(2)themaximumabsolutebandgapwidth,and(3)themax-imumbandgapataspecifiedfrequency.Moreimportantly,thethree-phasematerialprovidessignificantadvantagescomparedtothetypicaltwo-phasematerials,suchasalow-frequencybandgap.Thisresearchisexpectedtocontributehighlytovibrationandnoiseisolation,elasticwavefilters,andacousticdevices.
简介:这篇论文由电压来源在高力量的感应电动机美联储为全部的泛音电流失真(THCD)的最小化论述基因算法(GA)的强大的应用转换ers,基于一个近似泛音模型。也就是说定义一个需要的基础输出电压,当最小化THCD时,最佳的脉搏模式(切换的角度)决心生产基本产量电压。在脉搏宽度调整(PWM)的开始的1/4个时期的三和五切换的片刻的二个盒子的完全的结果波形被介绍。在定子刺激的泛音的存在导致一个阿富汗辅币转矩部件。就如果阿富汗辅币转矩在低频率,他们能在反馈控制系统引起麻烦速度变化,柄疲劳,和不能令人满意的性能的事实而言,第5,第7,第11,并且第13当前的泛音(在五个切换的角度的情况中)在一些预先指定的价值被抑制,减轻低频率的泛音的有害效果。同时,当要求的基本产量电压被维持时,THCD被优化。