简介:Platestructuresareemployedasimportantstructuralcomponentsinmanyengineeringapplications.Hence,assessingthestructuralconditionsofin-serviceplatestructuresiscriticaltomonitoringglobalstructuralhealth.Modalcurvature-baseddamagedetectiontechniqueshaverecentlygarneredconsiderableattentionfromtheresearchcommunity,andhavebecomeapromisingvibration-basedstructuralhealthmonitoringsolution.However,computingerrorsarisewhencalculatingmodalcurvaturesfromlateralmodeshapes,whichresultfromunavoidablemeasurementerrorsinthemodeshapesasidentifiedfromlateralvibrationsignals;thismakescurvature-basedalgorithmsthatusealateralmeasurementonlytheoreticallyfeasible,butpracticallyinfeasible.Therefore,inthisstudy,long-gaugefiberBragggratingstrainsensorsareemployedtoobtainamodalcurvaturewithoutanumericaldifferentiationprocedureinordertocircumventthecomputingerrors.Severaldamageindicesbasedonmodalcurvaturesthatweredevelopedtolocatebeamdamageareemployed.Bothnumericalandexperimentalstudiesareperformedtovalidatetheproposedapproach.However,althoughpreviousstudieshavereportedrelativesuccesswiththeapplicationofthesedamageindicesonasimplebeam,onlyonedamageindexdemonstratedthecapabilitytolocatedamagewhenthestiffnessofthelocalregionchangednearthesensor.
简介:以便完成聪明的车辆的侧面的控制,基于云模型使用双性人认知的模型并且遮蔽推理,解决决定问题聪明的车辆的侧面的控制质、量。由驾驶车辆获得很多个实验数据,根据数据的概念分类数据并且修理云控制器的输入和输出变量,设计聪明的车辆的云控制器的控制规则,并且遮蔽并且修理云控制器的参数:期望,熵和亢奋的熵。以便验证云控制器的有效性,联合模拟站台基于Matlab/Simulink/CarSim被建立。试验性的分析显示出那:司机侧面的控制器基于云模型能完成需要的角度追踪,并且完成好控制效果,它也验证那一系列心理活动象感到那样,认知,计算,决定等等模糊、不明确。
简介:Tomakeuseofthepriorknowledgeoftheimagemoreeffectivelyandrestoremoredetailsoftheedgesandstructures,anovelsparsecodingobjectivefunctionisproposedbyapplyingtheprincipleofthenon-localsimilarityandmanifoldlearningonthebasisofsuper-resolutionalgorithmviasparserepresentation.Firstly,thenon-localsimilarityregularizationtermisconstructedbyusingthesimilarimagepatchestopreservetheedgeinformation.Then,themanifoldlearningregularizationtermisconstructedbyutilizingthelocallylinearembeddingapproachtoenhancethestructuralinformation.Theexperimentalresultsvalidatethattheproposedalgorithmhasasignificantimprovementcomparedwithseveralsuper-resolutionalgorithmsintermsofthesubjectivevisualeffectandobjectiveevaluationindices.
简介:AbstractBackground:Intracranial aneurysm (IA) is a serious disease. Analyze and review the cases of anterior circulation ruptured IA by supraorbital lateral keyhole approach, and summarize the experiences of this approach.Methods:Retrospective analysis of 16 cases of ruptured anterior circulation IA in our department from January 2019 to June 2020, CT angiography (CTA) was performed before operation. Analyzing the IA's parameters by 3D-CT reconstruction. The IA was clipped by supraorbital lateral keyhole approach combined with the 3D-skull reconstruction. Extraventricular drainage was performed before craniotomy. Intraoperative neurophysiological monitoring was performed during the operation. After operation, fluorescein angiography and vascular ultrasound were performed to check the clipping effect. Intracranial pressure monitor was performed postoperatively. CTA was reexamined one week after operation. The modified Rankin Scale (MRS) was performed 6 months after operation.Results:There were 7 males (43.8%) and 9 females (56.2%), and the average age is 52.31 ± 11.12 years old. Among them, 11 patients (68.8%) were anterior communicating artery aneurysms and 5 (31.2%) were middle cerebral artery aneurysms. All patients were out of hospital within 10 days without any death, without cerebral infarction, cerebrospinal fluid leakage and neurological impairments. About mRS score, after 6 months follow-up, 8 cases (50%) had 0 point, 4 cases (25%) had 1 point, and 4 cases (25%) had 2 points.Conclusions:For ruptured anterior circulation IA, the supraorbital lateral keyhole approach combined with ventricular drainage, intraoperative electrophysiological monitoring, and intraoperative vascular ultrasound is a safe and minimally invasive treatment. The application of reconstruction clipping can reconstruct the diameter of parent vessel and reduce the recurrence rate of IA.
简介:Thispaperconstructsanewkindofblockbasedbivariateblendingrationalinterpolationviasymmetricbranchedcontinuedfractions.Theconstructionprocessmaybeoutlinedasfollows.Thefirststepistodividetheoriginalsetofsupportpointsintosomesubsets(blocks).Thenconstructeachblockbyusingsymmetricbranchedcontinuedfraction.FinallyassembletheseblocksbyNewton’smethodtoshapethewholeinterpolationscheme.OurnewmethodoffersmanyflexiblebivariateblendingrationalinterpolationschemeswhichincludetheclassicalbivariateNewton’spolynomialinterpolationandsymmetricbranchedcontinuedfractioninterpolationasitsspecialcases.Theblockbasedbivariateblendingrationalinterpolationisinfactakindoftradeoffbetweenthepurelylinearinterpolationandthepurelynonlinearinterpolation.Finally,numericalexamplesaregiventoshowtheeffectivenessoftheproposedmethod.
简介:AnadaptiveinternalmodecontrolisproposedtoeliminateeffectivelyperiodicdisturbancewithuncertainfrequencycausedbyinputerrorangleofPIGA(PendulousIntegratingGyroAccelerometer).Anadaptivealgorithmwithperiodicdisturbancefrequencyidentificationonlineisappliedandtheinternalmodelcontrollerparametersareadjustedtoeliminatedisturbance.Thentheconvergenceofthisalgorithmandthestabilityofthesystemareprovedbytheaveragingmethod.SimulationresultsverifytheproposedschemecaneliminateperiodicdisturbanceandimprovethetestprecisionforPIGAeffectively.
简介:Inthispaper,wepresentasimplebutpowerfulensembleforrobusttextureclassification.Theproposedmethodusesasingletypeoffeaturedescriptor,i.e.scale-invariantfeaturetransform(SIFT),andinheritsthespiritofthespatialpyramidmatchingmodel(SPM).Inaflexiblewayofpartitioningtheoriginaltextureimages,ourapproachcanproducesufficientinformativelocalfeaturesandtherebyformareliablefeaturepondortrainanewclass-specificdictionary.Totakefulladvantageofthisfeaturepond,wedevelopagroup-collaborativelyrepresentation-basedstrategy(GCRS)forthefinalclassification.Itissolvedbythewell-knowngrouplasso.Butwegobeyondofthisandproposealocality-constraintmethodtospeedupthis,namedlocalconstraint-GCRS(LC-GCRS).Experimentalresultsonthreepublictexturedatasetsdemonstratetheproposedapproachachievescompetitiveoutcomesandevenoutperformsthestate-of-the-artmethods.Particularly,mostofmethodscannotworkwellwhenonlyafewsamplesofeachcategoryareavailablefortraining,butourapproachstillachievesveryhighclassificationaccuracy,e.g.anaverageaccuracyof92.1%fortheBrodatzdatasetwhenonlyoneimageisusedfortraining,significantlyhigherthananyothermethods.
简介:Hemocytecountsandphenoloxidase(PO)activitywereexaminedafterhemolymphbeingincubatedindopamine(DA),noradrenaline(NE)andserotonin(5-HT).Resultsshowedthatallthethreebiogenicamines(BAs)hadasignificantimpactontotalhemocytecount(THC),differentialhemocytecount(DHC),andintracelluarandextracelluarphenoloxidase(PO)activity.AmongtheseBas,DAhadthestrongesteffectontheaboveparameters,whereas5-HThadtheleasteffect.PreincubationwithD1receptorantagonistSCH23390,D2receptorantagonistSulpirideand1:1admixtureofthetwocouldsignificantlyinhibittheeffectofDAontheseparameters.SCH23390showedastrongerinhibitoryeffectthanSulpiride,andtheadmixtureexhibitedthestrongesteffect.Theseresultssuggestedthatthechangeofhemocytecountandactivationofprophenoloxidase(proPO)systeminLitopenaeusvan-nameihemocytecanberegulatedbyBAs,andDAmodulatesthetwoparametersviaitsreceptors.
简介:AlthoughGaussianrandommatricesplayanimportantroleofmeasurementmatricesincompressedsensing,onehopesthatthereexistotherrandommatriceswhichcanalsobeusedtoserveasthemeasurementmatrices.Hence,Weibullrandommatricesinduceextensiveinterest.Inthispaper,wefirstproposethel2,qrobustnullspacepropertythatcanweakentheD-RIP,andshowthatWeibullrandommatricessatisfythel2,qrobustnullspacepropertywithhighprobability.Besides,weprovethatWeibullrandommatricesalsopossessthelqquotientpropertywithhighprobability.Finally,withthecombinationoftheabovementionedproperties,wegivetwoimportantapproximationcharacteristicsofthesolutionstothelq-minimizationwithWeibullrandommatrices,oneisonthestabilityestimatewhenthemeasurementnoisee∈Rnneedsapriori‖e‖2≤,theotherisontherobustnessestimatewithoutneedingtoestimatetheboundof‖e‖2.TheresultsindicatethattheperformanceofWeibullrandommatricesissimilartothatofGaussianrandommatricesinsparserecovery.
简介:AnovelH_∞tracking-baseddecentralizedindirectadaptiveoutputfeedbackfuzzycontrollerforaclassofuncertainlarge-scalenonlinearsystemsisdeveloped.Byvirtueoftheproperfilteringoftheobservationerrordynamics,theobserver-baseddecentralizedindirectadaptivefuzzycontrolschemeispresentedforaclassoflarge-scalenonlinearsystemsusingthecombinationofH_∞trackingtechnique,afuzzyadaptiveobserverandfuzzyinferencesystems.Theoutputfeedbackandadaptationmechanismsarebothrobustandimplementableindeedowingtotheirfreedomfromtheunavailableobservationerrorvector.Allthesignalsoftheclosed-looplargescalesystemareguaranteedtostayuniformlyboundedandtheoutputerrorstakeonH_∞trackingperformance.Simulationresultssubstantiatetheeffectivenessoftheproposedscheme.