385 201810 BuletinofSoiland WaterConservation Vol.38,No.5 Oct.,2018 >O ) ( =X 8 /,6 +V -. 1,5 2,3,# 3 2, 2,8 4 (1.Z[ G, O 730070;2.Z[ (, O 730070; 3.< CD, O 730000;4. Z( L) 0 G, L 430074). /:[]5 T ' ]!,^ M 2, 8 C D []^6 <J O,)Fb/,( BP bc 9;, 9T']! [])Fb \9 BP b c,!, A 2020, 2025,203054.515 10 5,4.513 10 5,4.512 10 5 hm 2," P; T '] :Z[F ;>Z[? > 6X>a [])Fb $,M5T' ] 01 : ;)Fb ;;J O :A 234:1000 288X(2018)05 0341 06 %5674:F301.24 89: I+,9 8,7,. )Fb R T']!5 [J].,2018,38(5):341 346.DOI:10.13961/j.cnki.stbctb.2018.05.054.Wang Quanxi,Sun Pengju,LiuXuelu,etal.Predictionofcultivatedlandareaandimportanceofinfluencingfactorsbasedon randomforestalgorithm[j].buletinofsoiland WaterConservation,2018,38(5):341 346. 犘狉犲犱犻犮狋犻狅狀狅犳犆狌犾狋犻狏犪狋犲犱犔犪狀犱犃狉犲犪犪狀犱犐犿狆狅狉狋犪狀犮犲狅犳犐狀犳犾狌犲狀犮犻狀犵犉犪犮狋狅狉狊犅犪狊犲犱狅狀犚犪狀犱狅犿犉狅狉犲狊狋犃犾犵狅狉犻狋犺犿 - 犃犆犪狊犲犛狋狌犱狔狅犳犙犻狀犵狔犪狀犵犆犻狋狔, 犌犪狀狊狌犘狉狅狏犻狀犮犲 WANG Quanxi 1,SUNPengju 2,3,LIU Xuelu 2,LIShangze 2,GAOJiancun 4 (1. 犆狅犾犲犵犲狅犳犕犪狀犪犵犲犿犲狀狋, 犌犪狀狊狌犃犵狉犻犮狌犾狋狌狉犪犾犝狀犻狏犲狉狊犻狋狔, 犔犪狀狕犺狅狌, 犌犪狀狊狌 730070, 犆犺犻狀犪 ; 2. 犆狅犾犲犵犲狅犳犚犲狊狅狌狉犮犲狊犪狀犱犈狀狏犻狉狅狀犿犲狀狋犪犾犛犮犻犲狀犮犲狊, 犌犪狀狊狌犃犵狉犻犮狌犾狋狌狉犪犾犝狀犻狏犲狉狊犻狋狔, 犔犪狀狕犺狅狌, 犌犪狀狊狌 730070, 犆犺犻狀犪 ;3. 犔犪狀犱犚犲狊狅狌狉犮犲狊犘犾犪狀狀犻狀犵犪狀犱犚犲狊犲犪狉犮犺犐狀狊狋犻狋狌狋犲狅犳犌犪狀狊狌犘狉狅狏犻狀犮犲, 犔犪狀狕犺狅狌, 犌犪狀狊狌 730000, 犆犺犻狀犪 ;4. 犆狅犾犲犵犲狅犳犘狌犫犾犻犮犕犪狀犪犵犲犿犲狀狋, 犆犺犻狀犪犝狀犻狏犲狉狊犻狋狔狅犳犌犲狅狊犮犻犲狀犮犲, 犠狌犺犪狀, 犎狌犫犲犻 430074, 犆犺犻狀犪 ) 犃犫狊狋狉犪犮狋 :[Objective]Toanalyzetheimportanceofthefactorsthatinfluencethechangeofcultivatedland areainordertopredicttheamountofcultivatedlandarearesources,andtoservicetheprotectionofcultivated land.[methods]taking QingyangCityofGansuProvinceasacasestudy,therandomforestalgorithm wasused toconstructthepredictionmodelofcultivatedlandarea.theresultswerecomparedwiththoseofbpneural networkmodel,andtheimportanceofthefactorsthatinfluencingcultivatedlandareachangewassorted. [Results]Therelativeerrorandroot meansquareerrorofthepredictionresultsoftherandom forest algorithm weresmalerthanthatofbpneuralnetwork,andthepredictionaccuracywashighandtheresults werestable.thecultivatedlandareain2020,2025and2030waspredictedtobe4.515 10 5,4.513 10 5 and4.512 10 5 hm 2,respectively,showingadecreasingtrend.Theimportanceofthe maininfluencing :2018 04 03 :2018 04 17 :<<& XY456(XY ICD (GSAN ZL 2015 045) : I+(1993 ), (L ),< O?, ACDX,CD G E mail:2480115068@qq.com :9 8(1963 ), (L ), <O?,@A,, AX#,GCD E mail:550490919@qq.com
342 38 factorswasrankedas:agriculturalmachinerygeneraldynamics> agriculturalpopulation> GDP> fixed assetinvestment.[conclusion]therandomforestalgorithmissuitableforthepredictionofcultivatedlandarea andcanmeasuretheimportanceoffactorsthatinfluencethechangeofcultivatedlandarea. 犓犲狔狑狅狉犱狊 : 犮狌犾狋犻狏犪狋犲犱犾犪狀犱犪狉犲犪 ; 狉犪狀犱狅犿犳狅狉犲狊狋犪犾犵狅狉犻狋犺犿 ; 狆狉犲犱犻犮狋犻狅狀 ; 犙犻狀犵狔犪狀犵犆犻狋狔 )` /,/ 0 W,XY8 _,6.(8 T `,8] 2017 D P 0 8B 9 M2 Z2 XY 8 M2 6.YP [1], W 6 \] I [2],6 R T' CD ] PR T'CD &" [3] 9 (_ [PQ CD, ( T' - PQ, T' *,! 9 CD : T ', STIRPAT [4] * P 5 [5] 5 5 [6 7] G0$ [8] PLS [9] CD T' 9,$,! * [10] $ [11] P [12],BP bc [13] 2 $F [14] b,! /CD :.? T' Z[. T', CD T'5 ^R CD,9 C D ' _, 9#T'] C D )F(randomforest,RF)b M 5 G C!/ [15],9M 2 WJK,$ G, 9 2 5 O a _!, M5_ T' ]! [16],) F b 7 2 Q [17 18],Z 6 [19] 5 b 6 [20] > 6 5O,CD _ RF b 5O S [21] CD300a_ R," 0,0P TU, CD ^6 J O,)F b 9J O, J O.9,^ 6!,5T',9$G8 C^R68 ] 9 1 -./0 1.1!"K J O(106 20 108 45 E,35 15 37 10 N), < W6, 3< 6 %, U6 P, 885~2082 m b, W M\, 6, 480~660mm, Z Z Y X J O L167>,2015?2.65 10 4, 2.71 10 6 hm 2 )1, `G 1995 2015J O [^) : 1995 2015J O " " YP 51 1995-2015 QXQ 1.2 9 CD^1995 2015 C D, R M,2000 2015M _ J (2001 2016),1995 1999 M _. (1996 2000),W5M ab 2 2.1 6 $\PCD [4 8],
5 I+ : )Fb R T']!5 343 NT' Z[. *:, - \ \ $3, [ ^ ( T ' A CDT', T'! [!,]/ (5 _5T'? T' # ]T' [1], T'V V,Z[$ ST [ ^ _ ; $J O _ (M [!,? T' T' Z[T'3 10T'? T':? ( 狓 1),Z[? ( 狓 2), O ( 狓 3); T ': GDP( 狓 4), a ( 狓 5),Z J? C ( 狓 6),[ ;]( 狓 7),0 [ ;] ( 狓 8);Z[ T ':\ ] 2 ( 狓 9),Z [ F ; ( 狓 10) 2.2 ) ( )Fb Breiman 2001 C!/ [15],5O(randomforestclassi fication,rfc) $ (random forestregression, RFR) b bootstrap, ' 犖 ] ) F 犓 X ' $, X 犓 : a ) F 9 $ L M +, :a 3!(B22)FM,)F! )ab 2 ]! )F $b ab/ [22 23] : (1) bootstrap M 犖 ) F 犓 WM,3 _ : a 1 ',_ 2 M\ ; (2) ',X 9 7 犓 < :a;%0 犕 A, 犕 A ) F 犿 ( 犿 D 犕 `M) 3 ^ 5V, ^ c 犿 5 V 9 ^ 5 V, < a _ ` X W : ; (3)9 M, :a5, 97< :a ; (4) 犓 1 : a, 2.3 % CD / BP b c 3 9;,ab\9 34! 3 &' 3.1 X 8 \ )Fb br ntree mtry M 0,:a<M^5VM +, a b 2 ( X a M 2 ;, ntree 0,! W M,ntree0 7 [M,X( 500 mtry0 2 M 1/3 C D ) F b /,)Fb 9M 2WJK, ^ b 9M $ G )2[, :a500,mtry 3,AY `!, BP bc ', traingd M,90 :1000, " 0 0.5, M 50000, Y 0 0.002, : 0 10,: 0 1,bc0a 10 12 1 b ^ MatlabR2016b, ^J O 1995 2015 C D 9, 1995 20103 ' M,2011 2015 3 `M 52 QX 8 8%K F Q 3.2 X 8 6 / /,9 2011 2015 J O U E, 2020,2025,2030,T " 3 M Q_T'\7., 5 :,., 1, `
344 38!, a b 2011 2015 _ M\9! 3 4,! 9 ; ( 2) 1 QX =8 < 10 4 hm 2. _ )F $ BP bc 2011 44.58 44.70 44.22 2012 44.98 45.05 45.08 2013 45.19 45.18 44.85 2014 45.33 45.23 44.37 2015 45.44 45.23 45.72 2020 45.15 44.50 2025 45.13 44.69 2030 45.12 44.89 3.3 XQ /,6 )Fb T' ]!5, CD)Fb! ;2], c29! ^,; _ 2 ]! )3[,! 9 T'] :Z[F ;( 狓 10)>Z[? ( 狓 2)>GDP( 狓 4)> a( 狓 5)>\]2( 狓 9)>ZJ? C( 狓 6) >[ ; ] ( 狓 7)> 0 [ ; ] ( 狓 8)> O( 狓 9)>? ( 狓 10) 2 QX =8% \9 /% 2011 2012 2013 2014 2015 RFR -0.277-0.165 0.030 0.211 0.459 0.121 BP 0.800-0.214 0.760 2.112-0.617 0.500 1[,) F b _ 15a J O WA,60[$ _ ;BP bc 2011 2015 M2 _, _ 15a ` 2[,)Fb \9 BP b c \ 7 ) Fb, 2015 \9 0.459%, ^.\7,! \9, 0.121,;BP bc, 2014 \ 9 2.112%, 0.500,/)F b A!, 2011 2015 ) F b W, BP bc,/ )F b A! BP b c 9 BP b c +, 0 a M [23], W M 'M,W M M9 W _,WA,! / )Fb 0M,X:a<M a]/^5v M2 M,M, W!, )FG0M,!,A!, 60 53 QXQ /, 3.3.1 X> 4 O023 9:-:!ab T']! 5,Z[?5? Oc2 T' 5 1995 2015 [^5 2 :1995 2005 J O Z [? 0, 2.11 10 6? 0 2.26 10 6?, 0 1.56 10 5?, P, 4.45 10 5 hm 2 4.43 10 5 hm 2, 0.26hm 2,Z[? 0 6 Z OJ^ /0 2005 2015J OZ [?W,2015 Z[?1.91 10 6?, 3.54 10 5?,, 4.43 10 5 hm 2 04.55 10 5 hm 2, 0 1.18 10 4 hm 2,Z [?, W 5 Z, Z 9,_` &, bs0 3.3.2 `a> 4 O023 9:-: T ', 6 X (GDP) a ]! 5 1995 2000,J O GDP,, 3.78 10 9 * 0 5.99 10 9 *, 0 2.21 10 9 *,,, 0 1.34 10 9 *, ;2000 2015,J O
5 I+ : )Fb R T']!5 345 GDP a " P, GDP 02015 6.90 10 10 *, 0 10.18,a 02015 1.22 10 11 *, 0 1.96 10 11 * c :,Z[.P,ZJ X,9 Vb ( 60] ;,.Z J) T,Z[`_,# [8], W 3.3.3 > 4 O023 9:-: Z[T',!9 T']!,Z[F ; 5, \] 2(` PQ 1995 2015J OZ[F ;W,1995 1997 2003 2006, 1995 4.50 10 5 kw 0 2015 1.95 10 6 kw, 0 1.50 10 6 kw, 07.10 10 4 kw J O ` _,.,Z[F ;W0, ZJNT,A ZJ9!,ZJ9V 8 9^6, 4!&! 4.1 T' *, C!PQ, a5 T' #MbG C! M;, ; BP bc,9' M!,A!W )Fb ;C! /,! A! YP M!,WX$ [17] ) F b 9 M!,92 WJK, G,$!MP, _ T'M 3 M,M 5,2,c \;9' G #Mb, )Fb / 60[,)Fb `M5 2]!,M5!ab 2]! 5, 5 T' CD )Fb /b J O,9 T' ]! )Fb 9,( c = [13] E G [24] / b BP bc CD! \; *, 0 a,w MW 0,A!\;,)F b6[ _), CW*:' ^,_15a J O, \; BP bc,)fb $ T '] _), NZ[F ; Z[? GDP T',c( I [25] 9 %T U6 > CD \ 9 &T' +,CD6, W,CDW J O T'] W \,67 ] Z[ $,AZJ9!,(6 PQ CDX? T' T' Z [.5T'_ /, 8*:T',` / 6+CD CD$ K $ _CD6, T' \ P*:T' T'0 & )Fb_ CD,^ I ] T' 4.2 (1)!,)Fb(_ -\ 9 BP bc,2020,2025,2030 54.515 10 5,4.513 10 5,4.512 10 5 hm 2," P;A!,)Fb, BP bc (2)!T'] :Z[F ;>Z[? > 6X >a Z[F ;W 0,YZJ9!,9V 8 9^6 ;Z[?,.?9 X W, " YP [ 8 ] [1]!,bI. R ; 5 [J]. 6(,2006,20(2):146 151. [2] ',? @. @< 1995 2013\]X ( ; Y R T ' [J]., 2017,37(3):167 173. [3] &",,B.< (_ [. PQ[J].,2017,37(3):125 130.
346 38 [4], [,b <,. STIRPAT [J].Z[,2014,30(4):204 211. RT'[J].Z[,2008,24(12): [15] Breiman L.Random forests[j].machine Learning, 196 200. 2001,45(1):5 32. [5] a,;@, A,. 6 [16], &, /,.) F C D / (?. \ P P Q C D:^ < [J]. a(,2012,26(3):32 38. [J].6(,2011,25(1):74 79. [17], M,?. ) F [6] @,. 5 5 " O R 7 [J].8,2014,30(1):34 38. R ;5 [J]. G,2014,34(1):54 59. [18],L, Z. ) F Q [7] 2,` ;. 5 5 C D [J].Z [ F,2016,47(10):338 2 P ( ;[J]. 6 ( 345.,2012,26(3):105 110. [19] ;, &,,. )FbZ 6 [8],, E,. G 0 $ 5O C D [J].Z [ F,2016,47(1): 6? T ' 5 [J], 297 303.,2016,36(1):136 142. [20] *,, B`,. b )Fb [9],!,)<. 15aT' 5 b6 5O[J].Z [,2017, 5 [J].,2013,27(7):55 61. 33(14):250 257. [10] 9 B,7,9.R ; ; [21],9N,G. 300 _ CDR[J].,2011,31(1):207 210. R5 [J].6 (,2012,26(2): [11],7,9. $ 94 101. [J].6 (,2013,27(8): [22] 8,J, M,. )F 1 5. 2CD[J]. M,2015,33(6):6 10. [12],c!,?. P [23],,$,.MATLAB bc43 V 7 CD: 1985 2010< 5 5 [M]. G: G H,2013. [J].6(,2015,29(8):185 189. [24] E G,[3J,7 J,.6 M2 [13] c=,,7 %. BP bc6v 9;CD[J].,2010,24(5):13 18. > CD[J].Z[(6,2007, [25] I, J,@ @,. %T U6M2 28(4):30 34. YR ; 5 :^ [J]. 6 [14] %,S,<,. F M2 $F Z[CD,2009,27(3): 245 248. 櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍櫍 ( 335 ) [14] SantisiS,CappeloS,Catalfamo M,etal.Biodegra dationofcrudeoilbyindividualbacterialstrainsanda mixed bacterialconsortium [J].BrazilianJournalof Microbiology,2015,46(2):377 387. [15] JianliangXue,Yang Yu,YuBai,etal.MarineOil Degrading MicroorganismsandBiodegradationProcess ofpetroleum HydrocarboninMarineEnvironments:A review[j].currmicrobiol,2015,71:220 228. [16]?!, 5I.4IMG a[m]. G: H H,2001. [17] G,R.#0W 9 X [J].,2010,29(1):6 11. [18] JamgocianRP,YapijakisC.Pilot scalecomparisonof bioventingvs.hydrogenperoxideinmaintainingefec tiveaerobicinsitu,bioremediation[j].studiesinen vironmentalscience,1997,66:365 377. [19] MerkovaM,Zalesak M,RinglovaE,etal.Degrada tionofthesurfactantcocamidopropylbetainebytwo bacterialstrainsisolatedfrom activated sludge[j]. InternationalBiodeterioration& Biodegradation,2018, 127:236 240.