论文部分内容阅读
目的应用近红外光谱技术测定藿胆丸中的水分含量。方法首先测定174批藿胆丸水分含量,同时采集其近红外光谱数据,用一阶导数法与S-G平滑法预处理,结合偏最小二乘法建立水分含量的近红外光谱定量分析模型,利用验证集样品验证。结果该模型内部交叉验证决定系数为0.988 17,校正均方差为0.103,内部交叉验证均方差为0.326 16,验证集的预测均方差0.217,平均回收率98.02%。结论该近红外光谱水分定量分析模型稳定,准确可靠,可用于藿胆丸中水分的含量测定。
Objective To determine the moisture content of Hu Dan pill by near infrared spectroscopy. Methods Firstly, the moisture content of 174 batches of Huidan pill was measured and the near-infrared spectroscopy data were collected. The first-order derivative method and the SG-smoothing method were used to pre-process the water content and establish a quantitative analysis model of moisture content by near-infrared spectroscopy. Sample verification. Results The coefficient of cross validation of the model was 0.988 17, the mean square error of calibration was 0.103, the mean square error of internal cross validation was 0.326 16, and the expected variance of validation set was 0.217. The average recovery rate was 98.02%. Conclusion The quantitative analysis model of moisture in near-infrared spectroscopy is stable, accurate and reliable, and can be used to determine the content of water in Kao Dan Wan.