Page 77 - 《水资源与水工程学报》2024年第5期
P. 77

!35 "!5 #                         & ' ( ) & * + , -                               Vol.35No.5
               2024 $ 10 %              JournalofWaterResources&WaterEngineering                 Oct.,2024

            DOI:10.11705/j.issn.1672-643X.2024.05.09


                'x CNN-LSTM-Attention ýwv³²?ë²


                                        ýöA’^q{‘a}


                                                               1,2
                                                     1,2
                                           1
                                                                          1
                                    nvw , xyz , { W , n|} , ~€                     3
                    (1. ùÄe. AR[c¶.T, 56 ùÄ 450001;2. 56ªAWXYZ[AXhÚ¯!~©è^_`,
                            56 ùÄ 450001;3. Q!hK\KÐUtTš‹hK\KQo, ;; Ut 264000)
                 H I: @ú™Ð‘.lYVð~XhiUV{çUYZÿ‘, •V5UV@Ê, ‹Œ/ÁWŽ›%èÙ
                 (CNN)、 c`ÕGHèÙ(LSTM) äÓ¦2¸_(Attention) { CNN-LSTM-Attention ¥èYV, c•¼yøŽ
                 ð, °èû&¡!~¥èYQš'( CMIP6 Q{ BCCCSM2MRáõYQÎCÁ¼yӜ, l†ÁUV{çU
                 YZä" , ›|š¼yБ.lYV{YZÿ‘。°±²o:CNN-LSTM-Attention YVéV5UV²É
                 š£{çUYZk±, YZÿ‘µøÁ,ŸÐ‘.lYV, øXkû¿‰(NSE) @ 0.883, µ>J45(RMSE)
                 @ 2.317, ´µpð45(MAE) @ 1.098; º›áõ~Ӝi, V5UVé 2025—2050 v{vçU1µÈÉ×n
                 ~mÌÍÃʦ»‘še, Æ,é SSP12.6 Ӝi, çU1~mäʦ»‘‘。%<*@Б.lYVé9A
                 ¨¿ï1'ŽYZÑV{l†úÙ¶hÉ, Î@UVáÑ{ADEÈ&R†äÞúÙ-.BCD½。
                 JKL: çUYZC" ; Б.lYV;CNN-LSTM-Attention; áõ~; V5UV
                 MNFOP:TV213.4   QRSTU:A    QVWP:1672643X(2024)05007309
                RunoffsimulationandfuturemultiscenariopredictionintheQinheRiver

                             BasinbasedontheCNN-LSTM -AttentionModel

                                                                   1,2
                                                                                               3
                                                                                    1
                                       1
                                                    1,2
                         ZHANGShuqi,ZUOQiting ,ZANGChao ,ZHANGLekai,BAYinji
                    (1.SchoolofWaterConservancyandTransportation,ZhengzhouUniversity,Zhengzhou450001,China;2.Henan
                   InternationalJointLaboratoryofWaterCycleSimulationandEnvironmentalProtection ,Zhengzhou450001,China;
                        3.YantaiCenterofCoastalZoneGeologicalSurvey,ChinaGeologicalSurvey,Yantai264000,China)
                 Abstract:Toenhancetheaccuracyofdeeplearningmodelsinsimulatingwatershedrunoffunderchan
                 gingenvironmentalconditions,acoupledmodelofconvolutionalneuralnetwork(CNN),longshortterm
                 memory (LSTM)andAttentionmechanismwasconstructedforthestudyoftheQinheRiverBasin.Inte
                 gratedwithmultipleoptimizationalgorithmsandmultiplescenariosinBCC-CSM2-MRclimatemodel
                 fromtheCoupledModelIntercomparisonProjectPhase6(CMIP6),thismodelwasappliedtowatershed
                 runoffsimulationandprediction.Itssimulationaccuracywasthencomparedwiththatofvariousdeep
                 learningmodels.TheresultsdemonstratethattheCNN-LSTM-Attentionmodelexhibitssuperiorper
                 formanceinsimulatingrunoffintheQinheRiverBasin,withNashSutcliffeefficiencycoefficient(NSE)
                 of0.883,rootmeansquareerror(RMSE)of2.317,andmeanabsoluteerror(MAE)of1.098,outper
                 formingotherdeeplearningmodels.Notably,theannualrunoffoftheQinheRiverBasinfrom2025to
                2050showsaslowdecreasingtrendwithsignificantfluctuationsunderdifferentclimatechangescenarios ,
                 especiallyintheSSP12.6scenario.Thisstudyprovidesnewinsightsintotheapplicationofdeeplearn
                 ingmodelsinintelligentsimulationofhuman-waterrelationshipsandoffersareferentialvalueforsubse

                !"#$:20240327; %&#$:20240714
                '()*: ! " # $ % & ' ( ) * (2021YFC3200201); Q ! § » - Š & + Ö Ù 5 6 % < T Ö Ù R S % < ) *
                        (2024HENYB01); Q!hK\KÐ)*(DD20220885)
                +,-.: c¨è(1999—), H, 56ùÄ9, u;%<=, ÞߍfAB.CADE%<。
                /0+,: çT[(1989—), H, ;;;•9, u;, §»P, %<>?@+,DEçè\K[߅çè| 。
   72   73   74   75   76   77   78   79   80   81   82