近红外光谱法快速测定苎麻粗蛋白含量

摘要

文中探讨了应用近红外光谱仪测定苎麻粗蛋白的可行性。以50个样品组成校正集，采用偏最小二乘法（PLS）建立了近红外光谱信息与粗蛋白含量的校正模型，该模型的相关系数为0．98。用该模型对10个样品的粗蛋白进行预测，结果表明，苎麻粗蛋白含量的化学测定值与近红外光谱模型预测值之间存在较好的相关性，预测值与化学值之间的平均相对误差为3．54％。可见，用近红外光谱分析建立苎麻粗蛋白预测模型并测定苎麻粗蛋白含量是可行的。%This paper discussed the feasibility to determine the crude protein of ramie by applying near -infrared spectrometer.The partial least squares (PLS)was used to establish calibration model of the near -infrared spectral information and crude protein content based on 50 samples.The correlation coefficient of the model was 0.98,and 10 ramie samples were used to evaluate the reliability of the cali-bration model.There was a good correlation between the values of the near -infrared predition model and the values of the chemical testing method,and the average relative error was 3.54%.The results showed that using NIR to establish ramie crude protein prediction model to determine crude protein content in ramie was feasible.