论文部分内容阅读
采用近红外(NIR)光谱技术结合偏最小二乘法(PLS)测定夏枯草中的水分。运用NIR漫反射光谱技术采集夏枯草的NIR光谱,以烘干法测得的水分含量为参考值,结合PLS法建立夏枯草中水分测定的NIR分析模型,并以未知样品验证。所建模型的校正集内部交叉验证相关系数(R~2)、校正均方差(RMSEC)和预测均方差(RMSEP)分别为0.988 2、0.129和0.134,验证集的NIR预测值与烘干法的参考值之间的差异无统计学意义。
Water content in Prunella vulgaris was determined by near infrared spectroscopy (NIR) spectroscopy combined with partial least squares (PLS). The NIR spectrum of Prunella vulgaris was collected by NIR diffuse reflectance spectroscopy. The moisture content of dried Prunella vulgaris was taken as a reference value, and the NIR analysis model of Prunella vulgaris in water was established by PLS method. The NIR spectrum was verified by unknown samples. The correlation coefficient (R ~ 2), RMSEC and RMSEP of the calibration set within the calibration set were 0.988, 2.0.129 and 0.134, respectively. The predicted NIR predictions were in good agreement with those of the drying method The difference between the reference values was not statistically significant.