简介：Inordertostudycavitationcharacteristicsofa2-Dhydrofoil,themethodthatcombinesnonlinearcavitationmodelandmixed-iterationisusedtopredictandanalyzethecavitationperformanceofhydrofoils.ThecavitationelementsarenonlinearlydisposedbasedontheGreenformulaandperturbationpotentialpanelmethod.Atthesametime,themethodthatcombinescavityshapeforfixedcavitylength(CSCL)iterationandcavityshapeforfixedcavitationnumber(CSCN)iterationisusedtoworkoutthethicknessandlengthofhydrofoilcavitations.Throughanalysisofcalculationresults,itcanbeconcludedthatthejumpofpressureandvelocitypotentiallyexistbetweencavitationendareaandnon-cavitationsareaonsuctionsurfacewhencavitationoccursonhydrofoil.Incertainanglesofattack,thecavitationnumberhasanegativeimpactonthelengthofcavitations.Andunderthesameangleofattackandcavitationnumber,thebiggerthethicknessofthehydrofoil,theshorterthecavitationslength.

简介：Inthenormaloperationcondition,aconventionalsquare-rootcubatureKalmanfilter(SRCKF)givessufficientlygoodestimationresults.However,ifthemeasurementsarenotreliable,theSRCKFmaygiveinaccurateresultsanddivergesbytime.ThisstudyintroducesanadaptiveSRCKFalgorithmwiththefiltergaincorrectionforthecaseofmeasurementmalfunctions.Byproposingaswitchingcriterion,anoptimalfilterisselectedfromtheadaptiveandconventionalSRCKFaccordingtothemeasurementquality.Asubsystemsoftfaultdetectionalgorithmisbuiltwiththefilterresidual.Utilizingaclearsubsystemfaultcoefficient,thefaultysubsystemisisolatedasaresultofthesystemreconstruction.Inordertoimprovetheperformanceofthemulti-sensorsystem,ahybridfusionalgorithmispresentedbasedontheadaptiveSRCKF.Thestateanderrorcovariancematrixarealsopredictedbythepriorifusionestimates,andareupdatedbythepredictedandestimatedinformationofsubsystems.Theproposedalgorithmswereappliedtothevesseldynamicpositioningsystemsimulation.TheywerecomparedwithnormalSRCKFandlocalestimationweightedfusionalgorithm.ThesimulationresultsshowthatthepresentedadaptiveSRCKFimprovestherobustnessofsubsystemfiltering,andthehybridfusionalgorithmhasthebetterperformance.Thesimulationverifiestheeffectivenessoftheproposedalgorithms.