近红外光谱技术快速测定首乌丸水分含量

Determination of moisture in Shouwu pills by near-infrared spectroscopy

目的应用近红外光谱技术测定首乌丸中水分的质量分数。方法烘干法测定115批首乌丸样品水分的质量分数,采集其近红外光谱数据,经一阶导数法与S-G平滑法预处理,结合偏最小二乘法建立测定首乌丸中水分质量分数的近红外光谱定量分析模型。结果该模型内部交叉验证决定系数为0.944,校正均方差为0.121,内部交叉验证均方差为0.205,验证集的预测均方差为0.127。结论该近红外光谱水分定量分析模型稳定,准确可靠,可用于首乌丸中水分质量分数的测定。

Objective To determine the content of moisture in shouwu pills by near-infrared spectroscopy. Methods At the beginning, the moisture content of 115 samples were determined with the method of drying under oven. The samples′near-infrared spectrum data were collected and pretreated by the methods of first-derivative and Savitzky-Golay filter. At last, the NIR quantitative model for measuring moisture content in shouwu pills was established by partial least squares regression （ PLS） method. Results The correlation coefficient, the root mean square error overall of calibration and that of cross validation of the calibration model were 0. 944, 0. 121, and 0. 205, respectively. The root mean square error overall of prediction was 0.127. Conclusion The NIR spectroscopy analysis model is steady, accurate and reliable, which could be used to evaluate moisture′s content of shouwu pills.