Page 85 - 《水资源与水工程学报》2024年第5期
P. 85
! 5# ÅÆÇ, :: s½ CNNLSTMAttention ¹ºÈhg¯°g¹ÉJÊ4ËÌ1ÍÄ 8 1
Ð" >ð%<[J]. 9'r5,2021,43(6):8085+97. [J]. '[ÒØ,2022,38(6):59.
[10]MAN Yuanyuan,YANG Qinli,SHAO Junming,etal. [26]F ¥, à É, «ãã, å. wxe'5UVáäËA
EnhancedLSTM modelfordailyrunoffpredictioninthe Ëü%< [J]. ARAS ( Q§B),2021,52
upperHuaiRiverBasin ,China[J].Engineering,2023, (9):1631.
24:229238. [27]ø 6, 4TÐ, 8. /Á Delta >ð{Ç5UVáõ
[11]BIAN Lekang,QIN Xueer,ZHANG Chenglong,etal. ~" ¢£[J]. 9'cô,2012,43(7):1114+46.
Application ,interpretabilityandpredictionofmachine [28]ò), 1 ú, ` e, å. gVáõÓ Delta-DCSI Ë
learningmethodcombinedwithLSTM andLightGBM :a ü>ð[J]. ne..ø( §»-.ÿ),2009,41
casestudyforrunoffsimulationinanaridarea[J].Jour (5):17.
nalofHydrology ,2023,625:130091. [29]KLL, `ê.. 'Ëü>ðä Delta >ðwr
[12]`Lb, q F, P, å.TVGM-LSTM¥èYVC,ç 5EgáõÓ{¢£[J]. á¬.ø,2007,65
UYZk±¢£[J]. ADEÚ¯,2023,39(6):104110. (4):653662.
[13]F_`, F þ, F&, å. /Á LSTM ^|îï{ [30]HOCHREITERS,SCHMIDHUBERJ.Longshorttermmem
WRF/WRF-Hydro¥ è ç U " ø [J]. A R . ø, ory [J].NeuralComputation,1997,9(8):17351780.
2023,54(11):13341346. [31]CHAUDHARIS,MITHALV,POLATKANG,etal.An
[14]GUJiuxiang,WANGZhenhua,KUENJ,etal.Recent attentivesurveyofattentionmodels[J].ACM Transac
advancesinconvolutionalneuralnetworks [J].Pattern tionsonIntelligentSystemsand Technology(TIST),
Recognition,2018,77:354377. 2021,12(5):132.
[15]r(5, J, ½Ä, å. /Á LSTM{tà:ß [32]º , +´. /ÁDÊ{ CNN-LSTM-Attention
sg&V D U V ç U Y Z C " [J]. Ý n ß s, ¾" YV%<[J]. Q!d(=ù-.,2023,19
2021,43(4):11441156. (9):6975.
[16]1Bþ, 2úú, DT, å. /Á Attention ¸_ø [33]KRIZHEVSKYA,SUTSKEVERI,HINTONGE.Ima
LSTM-Seq2seq YV{çUYZ%<[J/OL]. Ýnß geNetclassificationwithdeepconvolutionalneuralnet
s,2023,[20240415].http://kns.cnki.net/kcms/de works [J].CommunicationsoftheACM,2017,60(6):
tail/62.1072.P.20231103.1335.002.html. 8490.
[17]lZ´. ér5UV=jÚ¯äàK1&+cd$Ä{ [34]SRIVASTAVAN,HINTONG,KRIZHEVSKYA,etal.
3[EB/OL].(20190918)[20191015].http://www. Dropout :asimplewaytopreventneuralnetworksfrom
qstheory.cn/dukan/qs/201910/15/c_1125102357.htm. overfitting [J].JournalofMachineLearningResearch,
[18],, 2R, aMT, å. gVA´ïCbï» 2014,15(1):19291958.
1>ð[J]. AR.ø,2024,55(1):112. [35]I, 1µ, ã , å. /Á LSTM-Adam{ú
[19]. Í, «Pî, ,, å. /ÁgVA´ïªä ¸Wá"ÚYV[J]. ¸S§»,2024,41(1):175182.
SWAT YV{V5UVAqr´ï~¢£[J]. AR [36]MORIASIDN,ARNOLDJG,VANLIEW MW,etal.
AS-u+,2024,44(1):915+22. Modelevaluationguidelinesforsystematicquantificationof
[20]aMT, ,, «Pî. /Á Tapio ä LMDI YV{V accuracyinwatershedsimulations [J].Transactionsofthe
5UVö¤Ñ[ADER³¨¿¢£[J]. AD ASABE ,2007,50(3):885900.
EÚ¯,2023,39(4):94101. [37]LEGATESDR,MCCABEJRGJ.Evaluatingtheuseof
[21]cùÈ, ,, ©, å. /Á SWAT{V5UVA “goodnessoffit”measuresinhydrologicandhydroclimat
E12¢£C" [J]. 6Aa\[AR-( Q icmodelvalidation [J]. WaterResourcesResearch,
§B),2023,21(5):862872. 1999,35(1):233241.
[22]x J, c , «Pî, å. /Á SWAT{V5UVAD [38]1 , [, F , å. áõ[shR~ixµA
E=j²J'Cã¡»¢£[J]. ADE[A§ 5UVçUYZ[J]. =j.ø,2017,37(4):12521260.
».ø,2023,34(6):19. [39]p ¹, p
=.2021 vÆÿV5ZåACAv\
[23]^', º, ?, å. /Á SWATYV{V5U ¢£[J]. 9'r5,2022,44(S2):89+12.
V`QcÕçèzfC7¢£[J]. Q [40]c8, 1P<, c? , å. V5çU1~C,
!xÆARAS,2022(6):7683+89.
¢£[J]. 9'r5,2023,45(11):4954.
[24]+0ú, «¤Í, 2 , å. /Á CMIP6 {()Czf [41]G, ë <, ëµg, å. ¥è Encoder-Decoder {
zf|}~[J]. ADEÚ¯,2023,39(2):4049. LSTMçU"øYV%< [J]. jke..ø ( §.
[25]àd[, 1Úu. ¼mYÛÔYVQ+1¿>ð ÿ),2022,55(8):755761.