基于固定网格小波神经网络的不规则波中船舶横摇运动在线预报（英文）-豆柴文库

您所在位置：网站首页 / 基于固定网格小波神经网络的不规则波中船舶横摇运动在线预报（英文）.docx / 文档详情

在线预览结束，喜欢就下载吧，查找使用更方便

5 金币

下载文档

/ 2

下载提示

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

文本预览

基于固定网格小波神经网络的不规则波中船舶横摇运动在线预报（英文）
Title:Real-timeForecastingofShipRollMotioninIrregularWavesBasedonFixedGridWaveletNeuralNetwork
Abstract:
Shiprollingmotioncanhaveasignificantimpactonvesselstabilityandsafety,particularlyinturbulent,irregularseaconditions.Accurateandtimelypredictionofthismotioniscritical,bothforcrewplanningandtoinformoperationaldecisions.Inthisstudy,weproposeafixedgridwaveletneuralnetwork(FGWNN)foronlineforecastingofshiprollmotioninirregularwaves.Themodelemploystheprincipalcomponentanalysismethodforfeatureextraction,followedbyawavelettransformfordimensionreductionanddynamicfiltering.WetesttheproposedFGWNNmethodusingdatacollectedfromafull-scalecontainervesseloperatinginarangeofwaveconditions.Theresultsshowthatthemodelcanachievehighaccuracyinpredictingshiprollmotioninreal-time,withanR2valueof0.98.Additionally,theFGWNNshowssignificantimprovementinpredictionperformancecomparedtotraditionalwaveletneuralnetworkmodels.TheproposedFGWNNmethodhaspromisingpotentialforuseinreal-timeshipnavigationanddecision-makingsystems.
Introduction:
Shiprollingmotion,amotioninwhichthevesselrotatesaboutitslongitudinalaxis,isacomplexphenomenonthatisinfluencedbyavarietyoffactors,includingwaveheight,waveandshipdirections,shipspeedandsize,andhullshape.Accuratepredictionofrollingmotionisimportantfornavigationsafety,shipdesign,andoffshorestructureengineering,aswellasforcrewcomfortandplanning.Inrecentyears,variousmethodshavebeendevelopedforpredictingshipmotions,includingnumericalsimulationandstatisticalmodels.However,thecomplexityoftheship-waveinteractionandthevariabilityofseaconditionsmakeitachallengingproblemtosolve.
Thewaveletneuralnetwork(WNN)isapowerfulapproachforforecastingshipmotionsinirregularwaves.Inparticular,theWNNhasbeenshowntobehighlyeffectivefornonlineardynamicsystemmodelingandprediction.However,traditionalWNNmodelshaveadisadvantageinthattheperformanceoftendependsontheselectionofwaveletfunctionsandthenetworkstructure,whichcanleadtopoorpredictionaccuracyinsomecases.
Toove