Page 88 - 《水资源与水工程学报》2023年第4期
P. 88

!34 " ! 4 #                       & ' ( ) & * + , -                               Vol.34No.4
               2023 $ 8%                JournalofWaterResources&WaterEngineering                 Aug.,2023

            DOI:10.11705/j.issn.1672-643X.2023.04.10


                   'a ASWPD-BO-GRUdh/Œ>æCZ



                                                mwx, Vyz, ?{|
                        ( 81I& @JK•–&P ÐàÁ—@AK8B.¬6°Í&ѾÛå, /0 01 710054)
                 H I: ?uX†‰dZ[”ÁZ, Äïo"#·z,¨‰d[”qrL¢c­nÛù×^ƒ„, uhÄLso'x
                 #’¬(›¨·z(ASWPD) WJJ·qV(BO) ^/Ôá•a3(GRU) †‰dZ[”qr(ASWPD-BO-GRU)。
                 ö“, J­ ASWPDo®N†‰dZâfg}¶·z, €êc­nÛù×^tu—¥å 4 †{oÇÈ^·zÛf
                 g, Σ¤[”IZ; $ô, J­ BOq…·zô^Ûfgo¬^ GRUqrõûù; ±®, oÿ†Ûfg}¶[”, Ÿ
                 [”¯—{ì&x¥h†‰dZ[”¯—。ŸuhÄLs^qr¬­xÀdeKkÄ@A„†‰dZ[”¢, Ä
                 K GRU、BO-GRU、WPD-BO-GRUqr( 'x"#·z¿Do®N†‰dZâfg_V}¶·z^[”qr)
                 ^[”¯—}¶oº。¯—ï˜:ASWPD-BO-GRUqr^Sà. Wù (NSE)? 0.89, €¾«¬­¢[”ÁZ
                 ±X, A˜ ASWPD-BO-GRUqr€A·z^tu—þÿûX^[”ÁZWkY^~VQi。
                 JKL: †‰dZ[”; #’¬]B·zs!; (›¨·z; JJ·qV; /Ôá•a3
                                +
                                       +
                 MNOPQ:TV121 4;P338 .2   7RSTU:A    7VWQ:1672643X(2023)04008408
                      MonthlyrunoffpredictionmodelbasedonASWPD-BO-GRU

                                      TANGMingze,YANGYinke,ZHANGJingwen
                            (KeyLaboratoryofSubsurfaceHydrologyandEcologicalEffectsinAridRegion,Schoolof
                                   WaterandEnvironment ,Chang’anUniversity,Xi’an710054,China)
                 Abstract:Toimprovetheaccuracyofmonthlyrunoffpredictionandtoaddresstheproblemthatthecon
                 ventionaldecompositionintegratedrunoffpredictionmodelsincorrectlyusesfuturedata,agatedrecurrent
                 unit (GRU)monthlyrunoffpredictionmodel(ASWPD-BO-GRU)basedonselfadaptationstrategy
                 waveletpacketdecomposition(ASWPD)andBayesianoptimization(BO)isproposedanddeveloped.
                 First,inordertoreducethepredictiondifficultytheoriginalmonthlyrunofftimeseriesisdecomposedu
                 singASWPD ,bywhichfourrelativelyregulardecomposedsubseriesareobtainedwithoutusingfutureda
                 ta.Then,thehyperparametersoftheGRUmodelcorrespondingtothedecomposedsubseriesareopti
                 mizedusingBO.Finally ,themonthlyrunoffpredictionresultsareobtainedbypredictingeachsubseries
                 andsummingandreorganizingthepredictionresults.Theproposedandestablishedmodelisappliedto
                 thepredictionofmonthlyrunoffatYingluoxiaHydrologicalStationintheHeiheRiverBasin,andthepre
                 dictionresultsarecomparedwiththoseofGRU,BO-GRU,andWPD-BO-GRUmodels(models
                 basedontheconventionaldecompositionideawhichdecomposestheoriginalmonthlyrunofftimeseriesas
                 awhole ).TheresultsshowthattheNash-Sutcliffeefficiencycoefficient(NSE)ofASWPD-BO-GRU
                 modelis0.89 ,whichhasthehighestpredictionaccuracyintheexampleapplication,indicatingthatthe
                 ASWPD-BO-GRUmodelhashigherpredictionaccuracyandstrongergeneralizationabilitywithcorrect
                 decomposition.
                 Keywords:monthlyrunoffprediction;selfadaptationdecompositionstrategy(AS);waveletpacketde
                 composition(WPD);Bayesianoptimization(BO);gatedrecurrentunit(GRU)

                !"#$:20230301; %&#$:20230405
                '()*: 81I&¢üXÀ'u%;‘øX)C()*(300102292903); ¢!%&P~pKËâÚÁ¡!"&Ѿ
                        Ûå±%'()*(SKLLQG1933); /0œ#$%&'w;<ÌÊ)*(2017JM4018、2021SF497)
                +,-.: =ÿ(1998—), ., /0013, ß5;<8, ‚@É@A@CD=Î^;<。
                ¥¦+,: Q÷%(1977—), ., /0âI3, 45, ª67, ‚@É@A@CD=Î^;<。
   83   84   85   86   87   88   89   90   91   92   93