近红外光谱法预测紫花苜蓿草捆的营养品质和消化率

发布时间：2018-07-24 18:22

【摘要】：优质牧草苜蓿(Medicago sativa)品质的优劣和消化率的高低能在很大程度上影响畜牧业的发展。为探讨近红外光谱技术(NIRS)预测苜蓿草捆中营养成分和消化率的可行性,本试验采集来自我国苜蓿主产区的苜蓿草捆样品229份,利用改进的偏最小二乘法(MPLS),结合不同光谱处理和数学参数设置,建立苜蓿营养品质和消化率的近红外预测模型。结果表明:相对饲喂价值(relative feed value,RFV),中性洗涤纤维(neutral detergent fiber,NDF),酸性洗涤纤维(acid detergent fiber,ADF)和粗蛋白(crude protein,CP)的模型能用于实际含量的分析;半纤维素(Hemicellulose)和干物质体外消化率(In vitro dry matter disappearance,IVDMD)的交叉验证相对分析误差(relative prediction deviation for cross validation,RPDCV)值介于2.5~3之间,能够用于粗略分析,需要对定标集样品进一步扩充和完善以提高预测的准确度。试验初步建立了苜蓿草捆品质的定量分析模型,补充了我国苜蓿草捆营养品质数据库,为苜蓿草产品的生产、流通及动物饲料配方的制定提供了数据支持。
[Abstract]:The quality and digestibility of high quality alfalfa (Medicago sativa) can affect the development of animal husbandry to a great extent. In order to study the feasibility of predicting the nutrient composition and digestibility of alfalfa bales by (NIRS), 229 alfalfa Bale samples from the main alfalfa production areas in China were collected in this experiment. The improved partial least square method (MPLS),) was used to establish the near infrared prediction model of alfalfa nutrition quality and digestibility. The results showed that the models of relative feeding value (relative feed value, neutral detergent fiber (neutral detergent, acid detergent fiber (acid detergent and crude protein could be used for the analysis of actual content. The relative error of cross-validation of hemicellulose (Hemicellulose) and dry matter digestibility (In vitro dry matter appeasance (In vitro dry matter DMD) is between 2.5 and 3, which can be used for rough analysis. The calibration set samples need to be further expanded and improved to improve the accuracy of prediction. The quantitative analysis model of alfalfa Bale quality was established preliminarily, which supplemented the database of alfalfa Bale nutrition quality in China, and provided data support for the production, circulation and formulation of animal feed formula of alfalfa products.
【作者单位】：中国农业大学草地研究所;
【基金】：基本科研业务费“6种牧草品质分析及近红外预测模型建立”(2014XJ005) 农业财政项目“牧草种质资源品质分析及快速测定”(2130135) 公益性行业科研专项“科尔沁沙地苜蓿高效种植技术研究与示范”(201403048-2)资助
【分类号】：S816

【相似文献】