Analysis of nutrient content of tall fescue hay by near-infrared spectroscopy

In this study, near-infrared reflectance spectroscopy (NIRS) technology was used to construct a near-infrared prediction model of tall fescue hay. A total of 101 samples of tall fescue were collected from Qingyang City, Gansu Province. Wet chemical analysis results and NIRS were combined to establish and validate the prediction model by using modified partial least square (MPLS). Finally the prediction models of dry matter (DM), crude protein (CP), organic matter (OM), neutral detergent fiber (NDF), acid detergent fiber (ADF), ether extract (EE) and crude ash (Ash) of tall fescue hay were established. The ratio of performance to deviation for validation (RPD) values of the prediction models of CP and DM were 3.53 and 2.55, respectively. The prediction models have good prediction effect and can be used to predict the content of components in actual production. The RPD values of the prediction models of OM, NDF, ADF, EE and Ash were 2.17, 2.04, 2.06, 2.06 and 2.02, respectively, and the predicted results can be used as reference for some production.