Abstract: As a representative of rapid,convenient,and pollution-free detection technology for organic matter,near-infrared spectroscopy analysis technology can fulfill the requirements of swift and efficient detection. The establishment of a portable near-infrared analysis model of ryegrass multiflora can effectively monitor the nutritional quality dynamics of ryegrass multiflora and make corresponding adjustments to field management to improve the nutritional quality of ryegrass multiflora. In this study,the optimal modeling method was identified by comparing two near-infrared models established under two pretreatments:drying,crushing,and fresh cutting of ryegrass. The findings are as follows:the correlation coefficient (R²) of the six near-infrared models of nutritional quality of ryegrass multiflora samples after drying and crushing pretreatment was better than that of the freshly cut shortened samples. The R² of the near-infrared model of the six nutritional qualities of the dry,crushed,and pretreated samples was higher than 0.81,indicating that the model quality was good and could be used in production practice. The R² of the NIR model with WSC content in the freshly cut pretreatment sample was higher than 0.81,indicating that the model quality was good and could be used in production practice.

Key words: Ryegrass, Near-infrared spectroscopy model, Nutritional quality, Sample pretreatment

摘要： 近红外光谱分析技术作为专门针对有机物快速、便捷、无污染检测技术的典型代表，可以满足快速高效检测需求，建立多花黑麦草（Lolium multiflorum）便携式近红外分析模型可以有效监控多花黑麦草营养品质动态并对田间管理做出相应的调整以提高多花黑麦草的营养品质。本研究通过对多花黑麦草干燥粉碎及新鲜切短两种预处理下建立的2种近红外模型的比较选择最优建模方法。结果表明：多花黑麦草样品在干燥粉碎预处理之后六项营养品质近红外模型的相关系数均优于新鲜切短样品的近红外模型；干燥粉碎预处理样品六项营养品质近红外模型相关系数均高于0.81，模型质量良好可用于生产实践；新鲜切短预处理样品可溶性碳水化合物含量近红外模型的相关系数高于0.81，模型质量良好可用于生产实践。

关键词: 黑麦草, 近红外光谱模型, 营养品质, 样品预处理