Page 115 - 《水资源与水工程学报》2022年第6期
P. 115
! 33 " ! 6 # & ' ( ) & * + , - Vol.33No.6
2022 $ 12 % JournalofWaterResources&WaterEngineering Dec.,2022
DOI:10.11705/j.issn.1672-643X.2022.06.14
KÈD~{¦3ÉËbpq|}
1 1 1 1,2
~ , )¡¢ , P £ , ¤¥
(1. Îyw8-.c6 Nd(}ãgh6i, 6Î _ 712100;2. Îyw8-.c6
MnwNeghuvÆ$%5|, 6Î _ 712100)
F G: M«/µr2¯YcOnouNÄvç&J, MNQR-6du¸ëØÙ, u ARMA、 8Û
GM(1,1) P BP ËÖÖhç], }~#᪽n:、 "o、 Y]cn 3 !OuNwÄvç, ¥ço
䬩ª\9, ¤uÐoç¥ 2021-2025 I#᪽nuNw©ªÄv。¬®: ¥à¯Y#
᪽n:O,9ª½nuNwÐoÄvçM ARMAç,6ª½nÐoÄvçM8Û
GM(1,1) ç,16 ª½nÐoÄvçM BP ËÖÖhç; ¥à"oO,5 "ouNwÐoÄv
çM ARMAç,3 "oÐoÄvçM8Û GM(1,1) ç, }@"oÐoÄvçM BPËÖÖh
ç; ¥àY]cnO, yK 6 nuNwÐoÄvçM BP ËÖÖhç, "K 4 nÐoÄvçM8
ÛGM(1,1) ç;2021-2025 ¯YI#᪽nuNwwléÎ。&J¬;Mr2¯YuNw
Ê]¸ëìí。
HIJ: uNwÄv;ARMAç; 8Û GM(1,1) ç;BP ËÖÖhç; r2¯Y
KLDMN:TV213.4 OPQRS:A OTUN:1672643X(2022)06011109
PredictionmodelsofregionalannualtotalwaterconsumptioninmainlandChina
1
1
1,2
1
ZHOUYuming,CHENTianyue,GUOQing,SONGSongbai
(1.CollegeofWaterResourcesandArchitecturalEngineering,NorthwestA&FUniversity,Yangling712100,China;
2.KeyLaboratoryofAgriculturalSoilandWaterEngineeringinAridandSemiaridAreas,Ministryof
Education ,NorthwestA&FUniversity,Yangling712100,China)
Abstract:InordertoimprovelargescalewaterconsumptionpredictionmodelsformainlandChinaand
provideatechnicalsupportforthescientificutilizationofwaterresources,theannualtotalwatercon
sumptionpredictionmodelsofprovincialadministrativedivisions ,riverbasinsandgeographicalregions
wereestablishedbasedontheprinciplesofARMA ,grayGM(1,1)andBPneuralnetworkmodel.The
optimizedresultsofthemodelswerestatisticallyanalyzed ,andtheselectedoptimalmodelswereusedto
predictthetotalwaterconsumptionfrom2021to2025.Theresultsshowthatfortheprovincialadminis
trativedivisionscale ,theoptimalpredictionmodeloftotalannualwaterconsumptionisARMAmodelin
nineadministrativedivisions ,grayGM(1,1)modelinsixadministrativedivisions,andneuralnetwork
modelin16administrativedivisions.Attheriverbasinscale ,theoptimalpredictionmodelisARMA
modelinfivebasins,grayGM(1,1)modelinthreebasins,andneuralnetworkmodelintheYangtze
Basin.Forthescaleoflargegeographicalregions,theoptimalmodelofthesixnorthernregionsisthe
neuralnetworkmodel ,andthatofthefoursouthernregionsisthegrayGM(1,1)model.Thewatercon
sumptioninthenextfiveyearswillbegenerallystable.Thesefindingsareexpectedtoprovidesomesta
tisticalreferencefortotalwaterconsumptionmanagementinChina.
Keywords:annualtotalwaterconsumptionprediction;ARMAmodel;grayGM(1,1)model;BPneu
ralnetworkmodel;mainlandChina
!"#$:20220126; %$:20220623
'()*: 2345-678,1(52079110); Îyw8-.c6c6G-.tå,1(S202110712560)
+,-.: GÙ(2001-), ?, 6ÎÎ;C, ÏJã-G, &JKLMNOPNQR。
/0+,: ®Ú(2001-), ?, 6ÎÎ;C, ÏJã-G, &JKLMNOPNQR。