论文部分内容阅读
采用近红外漫反射光谱法对黄芪中毛蕊异黄酮葡萄糖苷和黄芪甲苷的含量进行快速无损检测。以液相色谱质谱联用分析值为参比,采用偏最小二乘法建立黄芪中毛蕊异黄酮葡萄糖苷和黄芪甲苷的定量分析模型。结果显示,毛蕊异黄酮葡萄糖苷近红外光谱经多元散射校正(MSC)+一阶导数+Savitzky-Golay卷积平滑预处理后模型最优,模型参数R2为0.826 6,RMSEP值为0.022 7,校正集R2为0.863 5,RMSEC值为0.019 0;黄芪甲苷近红外光谱经二阶导数+Savitzky-Golay卷积平滑预处理后模型最优,模型参数R2为0.854 8,RMSEP值为0.006 41,校正集R2为0.796 3,RMSEC值为0.007 99。近红外光谱技术结合偏最小二乘法可快速、准确的对黄芪中毛蕊异黄酮葡萄糖苷和黄芪甲苷的含量进行检测。此外,通过主成分分析,发现甘肃产黄芪与其他产地黄芪差异不大,排除甘肃产黄芪后,山西、四川和吉林的样本区分度较高。
The content of calycosin and astragaloside IV in Radix Astragali was determined by near-infrared diffuse reflectance spectroscopy. Using the analytical method of liquid chromatography-mass spectrometry, the partial least squares method was used to establish the quantitative analysis model of calycosin and astragaloside in Radix Astragali. The results showed that the model was optimized by using the MSC + + Savitzky-Golay convolution smoothing pretreatment model. The model parameters were R2 = 0.826 6 and RMSEP = 0.022 7. The calibration set R2 was 0.863 5 and RMSEC was 0.0190. The model was optimized by the second derivative + Savitzky-Golay convolution smoothing method. The model parameters R2 was 0.854 8 and RMSEP was 0.006 41. The calibration set R2 is 0.796 3 and RMSEC is 0.00799. Near infrared spectroscopy combined with partial least squares method can quickly and accurately detect the content of calycosin and astragaloside in Radix Astragali. In addition, through the principal component analysis, found that Astragalus and Astragalus from other producing areas in Gansu little difference, excluding Gansu Astragalus, Shanxi, Sichuan and Jilin samples higher discrimination.