简介:Thesimulationofaone-dimensionalrivernetworkneedstosolvetheSaint-Venantequations,inwhichthevariableparametersnormallyhaveasignificantinfluenceonthemodelaccuracy.ATrial-and-Errorapproachisamostcommonlyadoptedmethodofparametercalibration,however,thismethodistime-consumingandrequiresexperiencetoselecttheappropriatevaluesofparameter.Consequently,simulatedresultsobtainedviathismethodusuallydifferbetweenpractitioners.ThisarticlecombinesahydrodynamicmodelwithanintelligentmodeloriginatedfromtheGeneticAlgorithm(GA)technique,inordertoprovideanintelligentsimulationmethodthatcanoptimizetheparametersautomatically.Comparedwithcurrentapproaches,themethodpresentedinthisarticleissimpler,itsdependenceonfielddataislower,andthemodelaccuracyishigher.Whentheoptimizedparametersaretakenintothehydrodynamicnumericalmodel,agoodagreementisattainedbetweenthesimulatedresultsandthefielddata.
简介:ResearchersinthepasthadnoticedthatapplicationofArtificialNeuralNetworks(ANN)inplaceofconventionalstatisticsonthebasisofdataminingtechniquespredictsmoreaccurateresultsinhydraulicpredictions.MostlytheseworkspertainedtoapplicationsofANN.Recently,anothertoolofsoftcomputing,namely,GeneticProgramming(GP)hascaughttheattentionofresearchersincivilengineeringcomputing.ThisarticleexaminestheusefulnessoftheGPbasedapproachtopredicttherelativescourdepthdownstreamofacommontypeofski-jumpbucketspillway.ActualfieldmeasurementswereusedtodeveloptheGPmodel.TheGPbasedestimationswerefoundtobeequallyandmoreaccuratethantheANNbasedones,especially,whentheunderlyingcause-effectrelationshipbecamemoreuncertaintomodel.