Prediction of daily sediment discharge using a back propagation neural network training algorithm: A case study of the Narmada River, India

在线阅读 下载PDF 导出详情
摘要 MostofthestudiesonArtificialNeuralNetwork(ANN)modelsremainrestrictedtosmallerriversandcatchments.Inthispaper,anattempthasbeenmadetocorrelatevariabilityofsedimentloadswithrainfallandrunoffthroughtheapplicationoftheBackPropagationNeuralNetwork(BPNN)algorithmforalargetropicalriver.ThealgorithmandsimulationaredonethroughMATLABenvironment.Themethodologycomprisedofacollectionofdataonrainfall,waterdischarge,andsedimentdischargefortheNarmadaRiveratvariouslocations(alongwithtimevariables)andapplicationtodevelopathreelayerBPNNmodelforthepredictionofsedimentdischarges.Fortrainingandvalidationpurposesasetof549datapointsforthemonsoon(16June-15November)periodofthreeconsecutiveyears(1996–1998)wasused.Fortestingpurposes,theBPNNmodelwasfurthertrainedusingasetof732datapointsofmonsoonseasonoffouryears(2006–07to2009–10)atninestations.Themodelwastestedbypredictingdailysedimentloadforthemonsoonseasonoftheyear2010–11.ToevaluatetheperformanceoftheBPNNmodel,errorswerecalculatedbycomparingtheactualandpredictedloads.Thevalidationandtestingresultsobtainedatalltheselocationsaretabulatedanddiscussed.Resultsobtainedfromthemodelapplicationarerobustandencouragingnotonlyforthesub-basinsbutalsofortheentirebasin.Theseresultssuggestthattheproposedmodeliscapableofpredictingthedailysedimentloadevenatdownstreamlocations,whichshownonlinearityinthetransportationprocess.Overall,theproposedmodelwithfurthertrainingmightbeusefulinthepredictionofsedimentdischargesforlargeriverbasins.
机构地区 不详
出版日期 2019年02月12日(中国期刊网平台首次上网日期,不代表论文的发表时间)
  • 相关文献