Acta Agrestia Sinica ›› 2023, Vol. 31 ›› Issue (12): 3841-3850.DOI: 10.11733/j.issn.1007-0435.2023.12.030

Previous Articles     Next Articles

Research on Detecting Nutritional Quality of Forage Oat Silage by Near Infrared Spectroscopy

TIAN Li-mei1, CHEN Fei1, LIN Yan-li1, NI Kui-kui1, YANG Fu-yu1   

  1. 1. College of Grassland Science and Technology, China Agricultural University, Beijing 100193, China;
    2. College of Animal Science, Guizhou University, Guiyang, Guizhou Province 550025, China
  • Received:2023-07-19 Revised:2023-10-19 Online:2023-12-15 Published:2024-01-03

近红外光谱检测燕麦饲草青贮营养品质研究

田丽梅1, 陈菲1, 林炎丽1, 倪奎奎1, 杨富裕1   

  1. 1. 中国农业大学草业科学与技术学院, 北京 100193;
    2. 贵州大学动物科学学院, 贵州 贵阳 550025
  • 通讯作者: 杨富裕,E-mail:yfuyu@126.com
  • 作者简介:田丽梅(1997-),女,汉族,甘肃榆中人,硕士研究生,主要从事牧草加工与利用研究,E-mail:tianlm0905@163.co
  • 基金资助:
    畜禽饲料饲草精细化加工技术研究(2021YFD1300300);“种草制草养畜”关键技术研发与模式构建(XDA26040000);国家燕麦荞麦产业体系(CARS-07-E-3)资助

Abstract: For improving the analytical efficiency of oat forage nutrient quality testing in production and achieving rapid and accurate prediction of multi-indicator content,in this study,273 oat forage silage samples were collected from Hebei,Gansu,Inner Mongolia,Sichuan,Guizhou,Jiangsu and Shandong province in China,spectrum was obtained using a portable near-infrared instrument,and the quantitative analysis models for oat silage fresh and dried samples were developed by using partial least squares regression (PLS). The results showed that the models of moisture and acid detergent fiber of fresh samples could be used for quantitative analysis,with correction of validation (R2val) of 0.94 and 0.88,respectively,and the ratio of performance to deviation (RPD) was greater than 2.5. The models of water content,acid detergent fiber and crude protein of dried samples could be used for real-time detection and analysis,with R2val of 0.90,0.90,0.88,and RPD were all greater than 2.5. The model of neutral detergent fiber,water soluble carbohydrate,and crude fat were able to achieve the effect of crude estimation,with R2val of 0.76,0.85,and 0.85,and RPD values of 4.56,2.44,and 2.43,respectively,and the accuracy of the model still needed to be further improved. The model of dried samples was better than that of fresh samples,and these results provided important data references for rapid assessment and testing of oat silage nutrition quality in production sites.

Key words: Near-infrared spectroscopy, Forage oat silage, Nutritional quality

摘要: 为提高生产中燕麦(Avena sativa L.)饲草营养品质检测分析效率,实现快速准确预测多指标含量,本研究从中国河北、甘肃、内蒙、四川、贵州、江苏和山东7个省份采集并制作了273份燕麦饲草青贮样本,使用便携式近红外仪器采集光谱,利用偏最小二乘回归法(Partial least square,PLS),建立了新鲜样品和干燥样品营养成分的定量分析模型。结果表明:新鲜样品含水量和酸性洗涤纤维的模型可以用于定量分析,外部验证决定系数(Correction of validation,R2val)分别为0.94和0.88,相对分析误差(Ratio of performance to deviation,RPD)大于2.5;干燥样品的含水量、酸性洗涤纤维和粗蛋白的模型能够用于实时检测分析,R2val分别为0.90,0.90,0.88,RPD均大于2.5,中性洗涤纤维、可溶性碳水化合物、粗脂肪的模型能达到粗估的效果,R2val分别为0.76,0.85,0.85,RPD值为4.56,2.44,2.43,模型精度仍需进一步提高。干燥样品的模型优于新鲜样品,这些结果为生产现场快速评定检测燕麦饲草青贮营养品质提供了重要数据参考。

关键词: 近红外光谱, 燕麦饲草青贮, 营养品质

CLC Number: