小麦子粒粗蛋白FT-NIRS分析模型建立的初步研究

Research on Math Model of FT-NIRS Analysis for Crude Protein Content in Seed of Wheat

采用凯氏定氮法测试了216份小麦子粒的粗蛋白含量,用近红外仪采集数据,选择113份建立了数学模型。结果:最佳主成分数(Rank)=6,内部交叉验证均方差(RMSECV)=0.377,决定系数(R2)=96.89。为了验证模型的可靠性,对预测集样品进行预测,结果粗蛋白的预测均方差(RMSEP)=0.950,相对偏差(RSEP,%)=8.40。

Bruker fourier transform near-infrared spectroscopy （FT-NIRS） used 113 seeds to establish math model of crude protein. The result on this model showed that the Rank was 6 and the RMSECV was 0.377 and R2 was 96.89. 54 samples were used to test this model,The result on model validation showed that the RMSEP was 0.950 and the RSEP（%） was 8.40.