免费文献传递   相关文献

Quality Evaluation of Alfalfa Hay by the Spectroscopic Analysis of Near Infrared Reflectance

利用近红外光谱分析预测紫花苜蓿干草品质



全 文 :!23" !5#
 Vol.23  No.5
! " # $
ACTA AGRESTIA SINICA
   2015$  9%
  Sep.  2015
犱狅犻:10.11733/j.issn.10070435.2015.05.027
ÅÆXYÍÍZ;n[f;<=>H(ð…
;_|
1,
I 82
(1.+¬()°+04¥O,¹-˜™],4¥100193;2.+¬()*+Nñ˜™],4¥100193)
45

èéê
(CP)、+gòóPI(NDF)=sgòóPI(ADF)BÌë5¥ô«¬

¨¬Aiš‹B5Ñ¥ƒKó¥ôõ
(RFV)¥Þèó‘úGBùë,0¬s¥èó‘z€«p(GI)…¥Þè
ó‘úGBùë

ž¥>vwÛ;Ó
(PLS)、Õ<©n3#as±†•–,ð„Hí@WXÐJ#(A}#×Ìo#)
BCP,ADF,NDF,RFV=GIB#asCjéQ。lmPJ:èéêBöJVPkAp(犚2cv)G0.9129,s¤VP+
CjkApG0.901,éQ»µgvì。CP,ADF,NDF,RFV=GIs¤VPRPD–*¦2.5。T˜\üI…-
o
}#G-o
678

#as±†

-o
èéê

ƒKó‘ôõ

èó‘z€«p
9:;<=
:S541    >?@AB:A     >CD=:10070435(2015)05108006
犙狌犪犾犻狋狔犈狏犪犾狌犪狋犻狅狀狅犳犃犾犳犪犾犳犪犎犪狔犫狔狋犺犲犛狆犲犮狋狉狅狊犮狅狆犻犮犃狀犪犾狔狊犻狊
狅犳犖犲犪狉犐狀犳狉犪狉犲犱犚犲犳犾犲犮狋犪狀犮犲
GAOYanli1,SUNYan2
(1.InstituteofAnimalScience,CAAS,Beijing100193,China;
2.InstituteofGrasslandScience,ChinaAgriculturalUniversity,Beijing100193,China)
犃犫狊狋狉犪犮狋:Thecontentsofcrudeprotein(CP),neutraldetergentfiber(NDF),aciddetergentfiber(ADF)
areusedasimportantfactorstoevaluatethehayqualityofharvestedalfalfa.Relativefeedvalue(RFV)is
usedtoevaluatethequalityoffeedinAmerica,butforagegradingindex(GI)isusedinChina.Thequali
typarametersofalfalfahayincludingCP,NDF,ADF,RFV,GIwerepredictedusingfouriertransform
nearinfraredreflectancespectroscopywithPLSregressioninthistest.Thenfivemodelswerevalidatedby
crossvalidationandexternalvalidation.ResultsindicatedthattheFTNIRmodelsofalfalfahayquality
hadconsiderableaccuracyandprecision.ThecorrelationcoefficientofcrossvalidationofCPwas0.9129.
Thecorrelationcoefficientofexternalvalidationis0.901.Therefore,FTNIRcanbeusedtodetermine
thequalityofalfalfahayrapidlyandaccuratelywithoutanychemicalreagent.
犓犲狔狑狅狉犱狊:Nearinfraredreflectancespectroscopy;Alfalfahay;Crudeprotein;Relativefeedvalue;For
agegradingindex
EFGH
:20150506;IJGH:20150611
KLMN

ª‹i)°˜Ä²

”(201403048);A?()ú)•–\AðæIJ‡±(CARS35)´ µ
OPQR

ì5d
(1987),Ë,Ü3Ï2»,ƽ,˜™NLG,NYù´†8TY,Email:gaoyanli.025@163.com;ÇÈÅÆAuthor
forcorrespondence,Email:ctsoffice@163.com
  #as±†•–îNIRS(nearinfraredreflec
tancespectroscopy),Ò±†qr5¦|?±8a
s±†’“
(780~2500nm),#as±†B¾D±
†È¶…†-CH,-NH=-OH!ÌˇÁB[
R8@RÏÐ
[1]。
b*‹pB<+=:9&G•#
asÝ܍ƒžBÏÐ&
,¯
̕#as±†&8
›Ì뒓ð„.A

LJÐ÷ȶ|}K&G
;iz{Ij

Œz{¶¨¸

&GV9y

áT
Ú=Vz…!—rã:
[2]。
#as±†•–•0¬()˜™==<ž¥+
Ñ¥)…)./

ßl

ÑÕ NIRS@ҖÒìi、
¸¶

Jxȶ|/BR:

•9ùB¸¶Ij

z
{Fk=§Œ*ÐëBGY´†‘‚—óHC
*Bž¥Ýç=ãä
。NIRSž¥Å,Ni),¾
DNñ:N/=,NGYGùIjN/

•N
ñ:N/
,NIRS©(ÅGŠY˜™ž¥•z{
Nñàá
[35]、
(;!€}†ËÌâzAÀB‹²R
S!N/

•,NGYGùIjN/

¾DPA•,
!5# ì5d!:*¥#as±†z{Cj-oNnª<+áz

Y7PçIj
[67]、
ü:?Ñ9z
{qìFIj
[8]
!N/

K-o(犕犲犱犻犮犪犵狅狊犪狋犻狏犪L.)GùÃ¥
ô5ÒÅGCDèó‘*¥B‡Ã

³0e*¥
#as±†z{•–ð„-o
。RFV(relativefeedvalue)먬óN=NÞ
XÀ*šOBÜN¶–RÆ.ÅàÎ
[9],
5³0
¨¬./Ñ¥

Ñ¢

”†¼½­OKèó‘ùë
BD×Á‘!

Bèó‘ùë¥k«p

ÒkÛG

ƒK¦ŠYRk¬»èó‘

CÌo#),
uY
èó‘|›<Ü9ùBB–ë
。RFV5ë|›<
Ü9ù
(digestibledrymatter,DDM)BdVB–ë
=­ODDMB–ëÁ.ˆÕÅ。]},RFVBì
Úx¦óNÜ9ùdVB–ë
(drymatterin
take,DMI)=óN+BDDM Ìë(|›ù;zÌë
)。
¼Gÿ
[10]
¼½0¬èó‘*¥BA
h

}èó‘GùGöl:

}AÀ°+G«NÑ
³

•Ïxèó‘Gù¥k«pã:B‡Ã‚

l@
0¬èó‘:ú†*¥B=<

í“ñÎHŠe
èó‘Gù¥kBZ«¬

èó‘z€«p
(grad
ingindexGI)。RFV=GI5³0./Ñ¥¥…È
eèó‘ó¥GùBCD«¬

â^é@;èó
‘ó¥GùB2Yz€\A。UV\üLJ#as
±†•–ð„-oéQ

}#â^E¶»µBK-oÆ=zy

;ŠŒG-oBÔ½

1 STYUV
1.1 [ð0ET
UV&Gx®&4Ò2¶(°0=V‡ñ

•
2011$=2012$K28eGY-o#

‹o#=Ìo#ÜÇx&

§Í5cm。º‚®
¯÷Ü̦/Ü&Gš7

‡1mm§,ël®»+Õ¥

U&Gƒ87„。
1.2 [ðŠ­;naÍZ\>
&G è é ê

è P I z Æ ¼ ½ ¬ ­ ¬ »
GB/T64321994= GB/T 64342006 ; i j k,
ADF=NDFBjk¼½VanSoestòóPIz{
NÓ;i
[11]。
\1 ^š]^ð„
Table1 Alfalfavarietiesusedinthestudy
UGY
Alfalfavarieties
UGY
Alfalfavarieties
UGY
Alfalfavarieties
UGY
Alfalfavarieties
™ã 86266 Ó¬ 34
5mâ +<1™ ÔÕ »ã
JÅ© BARALFA54 ûk íÎ#
¨¬*© ›Z L3750 ª(1™ ALBLE WL323
SITEL ;´(G¬) Queen +<2™
NÞ2™ ACGruzelomd \@ 4Å©
¼½-o1e&GBèó‘?Ñâ ME,DMI,DDM,RFV=
GIõ。1e«¬CjéQG[10,1213]:
RFV=DMI(%BW)×DDM(%DM)/1.29;
DMI(%BW)-RFV=120/NDF(%DM);
DDM(%DM)-RFV =88.9-0.779×ADF
(%DM);
GI=ME(MJ·kg-1)×DMI(kg·d-1)× CP
(%DM)/NDF(%DM);
ME(MJ·kg-1)-GI=4.2014+0.0236×ADF
(%DM)+0.1794×CP(%DM);
DMI(kg · d-1)-GIh°,N = 51.26/NDF
(%DM);
Â+
:GIGèó‘z€«p,œYG MJ;ME
Gèó‘?Ñâ
;DMIGèó‘Ü9ùdVB–
ë
;CPGGèéêÙÜ9ùB;zÈ;NDFG+g
òóPIÙÜ9ùB;zÈ
;ADFGsgòóPI
ÙÜ9ùB;zÈ

±†B§Ñ¥Bð‰G¨¬ ThermoNicolet
ªeL\BÕ<©n3#as±†z{ð
(Antaris
FTNIRAnalyzer),˒“*ÂzÁ­‘=Y2
ÂjH&GÖ

‚Ø4.8cm),­&ÀÁz{¿‘G
TQAnalystv8。B§±†0ËC9ð‰2h。&G
Ë(FOSSªeL\BYï×2üš7Ìël&G
Ö+

ØLP/

Z¦±†ð‚;iç¾

”e&G
–ç¾32ü,ÃCMç¾3ü,”ü–ç¾Áb,Á
.ÒL–±†

±†jkB¾D.Śp˒ç¾
†Ýqr10000~4000cm-1,z‰5(Resolution)
G8cm-1,2[ñ‹(Gain),p½€ÂGlog(1/R)。
1.3 ×o]^aš_
87e&GzGk¬§=VP§2à,Ò+k¬
§71e,VP§16e。VP§&G¼½]&G
<+z{õB§¨qr–WŒ

Œ¥>vwÛ;
Ó
(PLS)ð„éQ,\Ë*¥k¬§§ŒíÉB±
†p½€Â

L¹^XNӆƒžBšpà@

ð„
—ðéQ



§ŠÓ

;iöJVP

KéQ;i
1801
! " # $ !23"
IV

¾DLJȚk¬kAp
(犚2cal)、k¬¬
»0DRMSEC、öJVP¬»DRMSECV。*¥
VP§KéQ;is¤VP

¾DLJȚCj
kkAp犚2p、VP§&G¬»D(SD)8Cj¬
»D
(RMSEP)BÈõRPD。vÌ*¥TQAnalyst
v8¿‘+B±†on:(Spectrumoutlier)=Ù3
Ap
(Leverage)!Úe.ŒÿŒã<éQ,µkv
žéQ

2 lmY;n
2.1 ;<=>9:Y;Š­;nlm
Kk¬§=VP§&G;i‡T<+ázj
k

˒èéê
(CP)、sgòóPI(ADF)、+gò
óPI
(NDF),¼½1.2+BCjéQKž×
RFV=GIõ。
\2 [ðX;Š­;nhi
Table2 Thechemicalanalysesofalfalfahaysamples DM%
àz
Component
&Gp
No.ofsamples
L–õ
Mean/%DM
vwõ
Minimum/%DM
v*õ
Maximum/%DM
¬»D
SD/%DM
CP/%DM
k¤71
Calibration71
VP§16
Validation16
16.39
16.27
10.21
11.21
20.52
19.86
2.66
2.60
ADF/%DM
k¤71
Calibration71
VP§16
Validation16
33.90
33.11
27.57
28.81
41.98
33.11
3.09
2.52
NDF/%DM
RFV
GI
k¤71
Calibration71
VP§16
Validation16
k¤71
Calibration71
VP§16
Validation16
k¤71
Calibration71
VP§16
Validation16
47.16
43.37
124.91
125.21
7.32
7.09
37.28
31.21
88.89
103.37
3.24
4.38
58.81
56.36
168.22
148.95
13.13
11.76
4.89
4.71
16.73
14.39
2.19
1.93
2.2 ×o]^Y³Š
*¥TQAnalystv8z{¿‘zÆKCP,ADF,
NDF,RFV,GIàzk¬§B87eÜN&G;iH†
ÝqrBã<

vz±†C^XNÓ8vz¾E7p
B§Œ

ë£ð„BéêàzB—ðéQk¬kA
pv*

¬»Dvw

¯ÌLJÙ3Ap=+:<²
DõKéQ;iHã<ÃÛ$Honõ

ÕÅHéê
àzBvz—ðéQ

¾DšpCP2]ó。
\3 ;<=>H(9:X;犖犐犚犛`£×olm
Table3 Thecalibrationmodelsofthecomponentsofalfalfahaysamples
àz
Component
ðéšp
Mainparameters
k¬kAp
犚2cal
k¬¬»0D
RMSEC
öJVPkAp
犚2cv
öJVP¬»0D
RMSECV
CP/%DM SG;6;1 0.9852 0.400 0.9129 0.954
ADF/%DM NO;6;1 0.9523 1.08 0.9243 1.12
NDF/%DM SG;7;1 0.9708 1.27 0.9441 1.31
RFV NO;3;2 0.9052 4.06 0.9013 5.31
GI SG;5;1 0.9524 0.69 0.8195 0.73
  ò:šp+:SG:SavitzkyGolayÅ~Ó;NO:W^X;!2eõGE7p;!3eG±†p½€Â,1:Šð×N ;2:Ûð×N
Note:SG:SavitzkyGolayfilter;NO:Noprocessing;thesecondparameterisfactors;thethirdparameterisdataformat,1:Fistderivative
filter;2:Secondderivativefilter
2.3 ×o0š_
GH;ŠŒIVéQBãc

TUVB¥
16eIV&G…KéQBCjim;i¥ô,
¼ëšp Œ ¥ s ¤ I V RPD,C j ƒ @ A p
2801
!5# ì5d!:*¥#as±†z{Cj-o犚2p,Cj¬» 0 D RMSEP。¼ ½ Sinnaeve=
WiliamsBw‡,É RPD*¦ 2.5“,éQâ
¥¦#asCj

*¦3.0“éQŒšïB
Cjim

\4 ×o[flm
Table4 TheresultsofpredictingalfalfahaysamplesbyPLSmodels
àz
Component
&Tpë
No.ofsamples
Cjõ
ANL/%DM
=jõ
LAB/%DM
CjkAp
犚2p
Cj¬»0
RMSEP
s¤VPƒKz{0D
RPD
CP/%DM 16 16.44 16.27 0.901 0.84 3.06
ADF/%DM 16 32.36 33.11 0.919 0.89 2.84
NDF/%DM 16 42.95 43.37 0.912 1.58 2.98
RFV 16 126.61 125.21 0.902 5.13 2.83
GI 16 7.13 7.09 0.892 0.76 2.53
  ò:ANLGéQCjL–õ;LABG=Vºz{L–õ;RPDGVP§&G¬»D(SD)8RMSEPBÈõ
Notes:ANLaverageanalyzedvaluebyNIRS;LABaverageanalyzedvaluebychemicalmethodinlaboratory;RPD=SD/RMSEP
:1 ×o犆犘[f-Y™f-0|63
Fig.1 Relationshipbetweenpredicteddata
andchemicalanalysisofCP
:2 ×o犃犇犉[f-Y™f-0|63
Fig.2 Relationshipbetweenpredicted
andchemicalanalysisofADF
:3 ×o犖犇犉[f-Y™f-0|63
Fig.3 Relationshipbetweenpredicteddata
andchemicalanalysisofNDF
:4 ×o犚犉犞[f-Y™f-0|63
Fig.4 Relationshipbetweenpredicted
andcalculateddataofRFV
:5 ×o犌犐[f-Y™f-0|63
Fig.5 Relationshipbetweenpredicted
andcalculateddataofGI
3 ~Yl
  -oúëì

éêÌëì

E£Ü

,N’;

B¨]

NIRSž¥Å-oš‹˜™
。1976$ Norris!}狋狌犮犪犪狉狌狀犱犻狀犪犮犲犪)=5d(犅狉狅犿狌狊犼犪狆狅狀犻犮狌狊)NÜ
N

eN=Jþ3YWX€ÂG˜™K_,\ü*¥
3801
! " # $ !23"
NIRSÓz{,CjHCP、ADF、NDF、hùF、\s
Ü9ù›<5=Ü9ùB–ë!

Cjim–š
ï
[14]。
¬‚Z¬!˜™H NIRSCj-o=RNNßNÞo@,NBáâázCP,NDF=
ADF!,lmPJ NIRSCj-oNDF,ADF,Ash}†IVDMDBÌë5áB,ˆ
CjADL=DMÃWá[1518]。Ÿg![19]B¥Û
͏Ä#as*±†z{•–

K152e…†
WXB-oz{—ðéQ

•&GLÕN‚¹ƒ!
[20]

NIRSCj-oPJ|}»µBCjèéê

+gòóPI=sg
òóPIÌë

]ðéQBö÷IV=s¤IV
RPD–*¦3。•þ¤ .N/,-oïBJþh°Þ‘

*¥ NIRSÓKâôõ=@ïGùB˜™lmPJ
,NIRS|}»
µ¥¦z{-oNDF=ADF!áz,}†@ïGù+pHõ、*
Ç=wsÌë!«¬

@ïGù$+s=ñsCj
šDs

Ò~«¬C*Ç
、pH=wsCjimš
ï
[2122]。
£s

øK-o
Ç
~!
[23]2009$*¥#as«U±†¸¶FÆ-o

GY‚Æ»µ5ÄÅ100%。•
(;!€CjN/
,Kalenbach•2001$¼½(;
!€jkNÓð„HCjéQ

xÕX³BCjl
m
[24],2011$;a¨!‡¦NIRS=ÉüLë›K
-o[25]。
#as±†ø|}Cjèó‘¥ô«¬GI=
RFV。¨¬¼½ RFV õ¦€

Ò+RFV*¦150GRã€。GI50¬+Æ
¼Gÿl@HNDF,ADF=CPÐ3N/é@Õ
B

â^`\¥ô-o[10]。
*¥Ai
nªBjkz{NÓ¥ô,NGù5Š²rzSC


¬‚¦§¨!
[26]
*¥ NIRS•–KÕ
(犣犲犪犿犪狔)áâGù=JþÕGù;i¥k˜
™

ð„HJþÕBó‘ƒKõFRV8èó‘
Gùz€«pGI«¬B#asÈeéQ。]ð„
BéQ8:ú+./Ñ¥B(VCjéQlmÈ
š

»µ5šì

TUV\üI…*¥#as±†z{ð„-o
lmöï

˜™lmG¸¶jkèó‘CP,NDF,
ADF=Þ.RFV†GIΏHŠYÃGiBN
Ó

ˆ‰•¾…ãöÓKGYGù;i¥ô

u>?
[1] Ñù=,QX,,€.Ç,!.A?#as±†z{•–[M].


+¬j<¡¢
,2000
[2] UUÝ,’0,ɬ3,!.#as±†z{‡Ã8ž¥
[M].4¥:œ.)¡¢,2005
[3] ShafferJA.Estimationofbotanicalcompositioninalfalfa/
ryegrassmixturesbynearinfraredspectroscopy[J].Agrono
myJournal,1990,82(4):669673
[4] WachendorfM,IngwersenB,TaubeF.Predictionoftheclo
vercontentofredcloverandwhiteclovergrassmixtruresby
nearinfraredreflectancespectroscopy[J].GrassandForage
Science,1999,54(1):8790
[5] SmartAJ,SchachtW H,MoserLE,犲狋犪犾.Predictionof
leaf/stemratiousingnearinfraredreflectancespectroscopy
(NIRS)[J].AgronomyJournal,2004,96(1):316318.
[6] É[[,@{%,;Z¬,!.#as±†•–•5dY7Pç
jk+Bž¥˜™
[J].as8ÞÕ~+w,2008,27(2):56
90
[7] ÷z™,@{%,nß,!.WXÌKëcn5dY7PçB#
as±†z{
[J].as,2010,31(7):3944
[8] ÍRN,Éð¬,,!.#as±†•Ij,Nqì4F+
Bž¥
[J].±†+8±†z{,2010,30(5):12431247
[9] RohwederDA,BarnesRF,JorgensenN.Proposedhaygrad
ingstandardsbasedonlaboratoryanalysesforevaluatingquali
ty[J].JournalofAnimalScience,1978,47(3):747759
[10]H„à,¼Gÿ,YðZ,!.èó‘Gù¥k«pB˜™Ah
†Ò;k
[J].N)°+,2004,21(9):5561
[11]f%.ó‘z{†ó‘ùëIj•–[M].4¥:4¥()*
+¡¢
,1993
[12]DunhamJR.Relativefeedvalue measuresforagequality
[C]//Foragefacts.KATATEAES,CES,1998:#41
[13]JeranyamaP.Understandingrelativefeedvalue(RFV)and
relativeforagequality(RFQ)[M].SDSU,CooperativeExten
sionService,2004
[14]NorrisKH,BarnesRF,MooreJE,犲狋犪犾.PredictingForage
QualitybyInfraredReplectanceSpectroscopy[J].Journalof
AnimalScience,1976,43(4):889897
[15]Z¬.#as±†ÓCj-o

+¬()*+
,2009
[16]Z¬,Éð¬,,!.FTNIR±†Ójk-oB6²Gù«¬[J].±†+8±†z{,2007,27(7):1308
1311
[17]Z¬,Éð¬,,!.#as±†ÓjkIázB˜™
[J].±†+8±†z{,2008,28(5):1045
1048
[18]Z¬,Éð¬,,!.#as±†Ójk[J].±†+8±†z{,2008,28(2):317320
[19]Ÿg,H%+,Öá,!.-oéQBð„
[J].±†+8±†z{,2009,29(12):32503253
[20]¹ƒ,Éð¬,Z¬,!.#as“*±†ÓCj-o4801
!5# ì5d!:*¥#as±†z{Cj-oNDZáâôõ
[J].±†+8±†z{,2008,28(12):2826
2829
[21]¶3â.#as±†•–¥k-o

+¬()*+
,2006
[22]¶3â,âªË,Éð¬,!.#as±†Ójk-oe&Báâôõ
[J].±†+8±†z{,2007,27(7):1304
1307
[23]Ç~,Éð¬,Hã2,!.#as±†-oFÆNӘ™
[J].±†+8±†z{,2009,29(2):386388
[24]KalenbachRL,RobertsCA,TeuberLR,犲狋犪犾.Estimation
offaldormancyinalfalfabynearinfraredreflectancespectros
copy[J].CropScience,2001,41(3):774777
[25];a¨,áS˜,¼åj.‡¦#as±†•–8ÉüLë›
B[J].±†+8±†z{,2011,31
(6):15101513
[26]¦§¨.#as±†•–•ÕáâGù=JþÕGù¥
kB˜™
[D].´=µR:‚œc()*+,2008

%&( , -
欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍欍

XâßATUcV

XâY×WåcV

9«„ã


+¬Y)

5ë()¤¾

+¬()°+0Å9°+˜™]=+¬Y
7Â*ƒX¾B|¬g

Ä)g

•–gY)°•#

9³¬kY

}i)NB/³A

ÃÃÅ»‡g

:=g=†“g

æçi)qr

*õÅ9

äA



£h





,N

RYY8

Y7
›G!

馑*

•–=¥

•ÆK_

1€Y7X

(áæ)B@N=•–»È

1€()°˜

ú.¤w»È

*+Ä()0—˜:

(ÄÄ)Y=.*():ú(áÆ



*16U,”#8®,|$96®。¬‚ÀŠ™:CN11-4413/S,¬<¬»™:ISSN1671-
895X,|¬1ñ¯°–|±™,´ |ÓA¸çט[¤±™,™F”#h 3®。
¯@?™
:82-132
ñ ³:(100081)4¥¶+@Ä*¸12™ +¬Y)˜[¤
³ ¹:010-82105796(˜[¤)010-82105795(.“@i¤)
 ::010-82105796ø³:www.chinaseedqks.cn
Email:chinaseedqks@caas.cn chinaseedqks@163.com
5801