Publications

Detailed Information

Development of Calibration for Predicting Feed Value of Imported Forage using Near Infrared Reflectance Spectroscopy (NIRS) : 근적외선분광법을 이용한 수입 조사료의 사료가치 측정을 위한 검량식 작성

DC Field Value Language
dc.contributor.advisor김종근-
dc.contributor.author정은찬-
dc.date.accessioned2022-06-08T06:39:34Z-
dc.date.available2022-06-08T06:39:34Z-
dc.date.issued2022-
dc.identifier.other000000170671-
dc.identifier.urihttps://hdl.handle.net/10371/181174-
dc.identifier.urihttps://dcollection.snu.ac.kr/common/orgView/000000170671ko_KR
dc.description학위논문(석사) -- 서울대학교대학원 : 국제농업기술대학원 국제농업기술학과, 2022.2. 김종근.-
dc.description.abstractCurrently, Korea is heavily reliant on imports for forages. However, quality analysis is not performed while forage is imported and distributed to the final purchaser. The reason is that the quality analysis of forage is not legally compulsory, and the analysis procedure is complex, time-consuming, and expensive. Therefore, this study was conducted to develop the calibration equation for a rapid and inexpensive quality analysis for imported hays using near infrared reflectance spectroscopy (NIRS). The representative samples of alfalfa hay (227) and timothy hay (360) were collected nationwide. Furthermore, this study compared sample of intact form (cut into 10 cm) with that of ground form to check the possibility of omitting the drying and grinding process. The partial least squares (PLS) was used as a regression method and several mathematical treatments were conducted to develop more accurate calibration equations. The standard error of cross validation (SECV) was used as an evaluation index of the calibration equation, and the coefficient of determination (R2) was used as an index of the correlation between reference values and NIRS prediction values.
The R2 values of moisture content, acid detergent fiber (ADF), neutral detergent fiber (NDF), crude protein (CP), in vitro dry matter digestibility (IVDMD), total digestible nutrients (TDN) and relative feed value (RFV) on intact alfalfa hay were 0.74, 0.42, 0.68, 0.89, 0.52, 0.68 and 0.54, respectively. The R2 values of moisture content, ADF, NDF, CP, IVDMD, TDN and RFV on ground alfalfa hay were 0.84, 0.90, 0.97, 0.92, 0.95, 0.97 and 0.93, respectively. The R2 values of moisture content, ADF, NDF, CP, IVDMD, TDN and RFV on intact timothy hay were 0.48, 0.95, 0.75, 0.38, 0.61, 0.75 and 0.61, respectively. The R2 values of moisture content, ADF, NDF, CP, IVDMD, TDN and RFV on ground timothy hay were 0.95, 0.98, 0.94, 0.94, 0.84, 0.94 and 0.96, respectively. Intact alfalfa hay had no trend on derivative. However, ground alfalfa hay had the best results on the fourth derivative with the exception of moisture and NDF contents which had the best results on the second and first derivative, respectively. Intact timothy hay had the best results of calibration on the first derivative with the exception of moisture and CP contents which had the best results in the third derivative. However, ground timothy hay had no trend on derivative.
The results of the development of the calibration equations were good and can be used in field. The results of intact form were worse than those of ground form. However, intact form scanning method has a merit which drying and grinding processes can be omitted. Therefore, intact form scanning method can be used in field for quick and rough prediction rather than exact prediction. The calibration results of intact alfalfa hay in this study were poorer when compared to previous reports, most likely because the leaf part of the alfalfa hay was powdered and covered in the scanning cup. This problem could be overcome if the hay is firmly fixed and scanned.
-
dc.description.abstract현재 우리나라는 조사료 수입에 크게 의존하고 있다. 그러나 조사료를 수입하여 최종 구매자에게 도달하기까지 품질분석이 이뤄지지 않고 있는 상황이다. 그 이유로는 조사료의 품질분석이 법적으로 의무화 되어있지 않고, 분석 과정이 복잡하고, 시간이 오래 걸리며 그리고 비용이 많이 들기 때문이다. 따라서 본 연구는 근적외선분광법 (near infrared reflectance spectroscopy; NIRS)을 이용하여 수입 건초의 신속하고 저렴한 품질 분석을 위한 검량식을 개발하기 위해 수행되었다.
알팔파 건초 227 점과 티모시 건초 360 점을 전국에서 수집하였다. 건조 후 분쇄한 시료의 NIRS 검량 결과가 뛰어나지만, 본 연구에서는 건조와 분쇄의 과정을 생략하고 생 시료의 가능성을 확인하고자 두 시료를 비교하였다. 회귀 방법으로 부분 최소 제곱법(partial least squares; PLS)을 사용하였으며 정확한 검량식을 개발하기 위하여 다양한 수처리 기법을 사용하였다. 교차 검증 표준 오차(standard error of cross validation; SECV)가 검량식을 평가하는 지표로 사용되었으며, 결정계수(coefficient of determination; R2)가 습식 분석값과 예측값의 유사도를 나타내는 지표로 사용되었다.
수분 함량, acid detergent fiber (ADF), neutral detergent fiber (NDF), crude protein (CP), in vitro dry matter digetibility (IVDMD), total digestible nutrients (TDN) 및 relative feed value (RFV)에 대한 알팔파 생 시료의 R2의 값은 각각 0.74, 0.42, 0.68, 0.89, 0.52, 0.68 및 0.54 였으며, 알팔파 분쇄 시료에 대한 R2의 값은 각각 0.84, 0.90, 0.97, 0.92, 0.95, 0.97 및 0.93 였다. 수분 함량, ADF, NDF, CP, IVDMD, TDN 및 RFV 에 대한 티모시 생 시료의 R2의 값은 각각 0.48, 0.95, 0.75, 0.38, 0.61, 0.75 및 0.61 였으며, 티모시 분쇄 시료의 R2의 값은 각각 0.95, 0.98, 0.94, 0.94, 0.84, 0.94 및 0.96 였다. 한편 알팔파 분쇄 시료와 티모시 생 시료는 각각 4차 도함수와 1차 도함수에서 가장 좋은 결과를 나타내었지만, 다른 시료들은 도함수의 추세를 보이지 않았다.
검량식의 개발 결과로 보아 현장에서 사용할 수 있을 것이라 판단된다. 생 시료 스캔 방법은 분쇄 시료 스캔 방법 보다 R2의 값이 낮지만 건조 및 분쇄 과정을 생략하고 현장에서 빠르고 대략적인 수입 조사료의 성분 분석을 예측할 수 있었다. 알팔파 생 시료의 경우 잎 부분이 가루가 되어 스캔 컵을 덮는 현상이 발생하기 때문에 추후 연구에서는 알팔파 생 시료를 단단하게 고정하여 휴대용 NIRS를 이용하여 스캔하면 더 나은 검량식 결과를 유도할 수 있을 것으로 판단된다. 한편 분쇄 시료 스캔 방법은 대부분 성분의 R2의 값이 0.9 보다 높아 좋은 정확도를 나타내었다. 소화율에 대한 티모시 분쇄 시료의 결과가 비교적 낮았는데, 이는 샘플의 수분을 5 % 미만으로 조절하여 더 나은 검량식을 개발한 이전의 보고를 따라 추후 연구에서 더 좋은 결과를 얻을 수 있을 것이라 판단된다.
-
dc.description.tableofcontentsAbstract I
Contents III
List of Tables VI
List of Figures VIII
List of Abbreviations X

1. Introduction 1
1.1 Study background 1
1.2 Aims of research 3
2. Literature review 4
2.1 The definition of near infrared reflectance spectroscopy (NIRS) 4
2.2 The history of NIRS 4
2.3 The process of obtaining calibration 5
2.3.1 Sample collection and analysis 5
2.3.2 Measurement of spectra and mathematical treatment 6
2.3.3 Calibration development 7
2.3.4 Validation of calibration 10
2.4 Evaluation of feed value of forage using NIRS 10
2.4.1 Crude protein 10
2.4.2 Fiber 11
2.4.3 Digestibility 11
3. Material and method 13
3.1 Collection of samples 13
3.2 Spectrum acquisition using NIRS 13
3.2.1 Intact sample 13
3.2.2 Ground sample 14
3.3 Chemical analysis 14
3.3.1 Preparation of chemical analysis 14
3.3.2 Moisture content 14
3.3.3 Crude protein content 15
3.3.4 Fiber content 15
3.4 Digestion analysis 16
3.4.1 In vitro dry matter digestibility 16
3.4.2 Total digestible nutrients 17
3.4.3 Relative feed value 17
3.5 Development of calibration and validation 17
4. Results and discussion 19
4.1 Development of calibration for alfalfa hay 19
4.1.1 Chemical composition of alfalfa hay 19
4.1.2 The characteristics of NIR spectra of alfalfa hay 20
4.1.3 Development of calibration and validation for alfalfa hay 22
4.1.3.1 Moisture 22
4.1.3.2 Fiber 24
4.1.3.3 Crude protein 28
4.1.3.4 In vitro dry matter digestibility 30
4.1.3.5 Total digestible nutrients 32
4.1.3.6 Relative feed value 34
4.2 Development of calibration for timothy hay 36
4.2.1 Chemical composition of timothy hay 36
4.2.2 The characteristics of NIR spectra of timothy hay 37
4.2.3 Development of calibration and validation for timothy hay 39
4.2.3.1 Moisture 39
4.2.3.2 Fiber 41
4.2.3.3 Crude protein 45
4.2.3.4 In vitro dry matter digestibility 47
4.2.3.5 Total digestible nutrients 49
4.2.3.6 Relative feed value 51
5. Conclusion 53
6. Bibliography 54
7. Abstract in Korean 60
-
dc.format.extentX, 63-
dc.language.isoeng-
dc.publisher서울대학교 대학원-
dc.subjectNIRS, alfalfa, timothy, imported hay, calibration-
dc.subject.ddc631-
dc.titleDevelopment of Calibration for Predicting Feed Value of Imported Forage using Near Infrared Reflectance Spectroscopy (NIRS)-
dc.title.alternative근적외선분광법을 이용한 수입 조사료의 사료가치 측정을 위한 검량식 작성-
dc.typeThesis-
dc.typeDissertation-
dc.contributor.AlternativeAuthorEunchan Jeong-
dc.contributor.department국제농업기술대학원 국제농업기술학과-
dc.description.degree석사-
dc.date.awarded2022-02-
dc.identifier.uciI804:11032-000000170671-
dc.identifier.holdings000000000047▲000000000054▲000000170671▲-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share