【摘要】 利用傅里叶变换红外光谱结合偏最小二乘法对三七主根中3种皂苷Rb₁、Rg₁、R₁的总含量进行预测分析方法研究.测试了7个县12个种植点的100株三七主根样品和纯物质淀粉、皂苷Rb₁、Rg₁、R₁的红外光谱，利用光谱软件计算了每个样品红外光谱的导数光谱，按药典中测量三七有效成分含量的方法测量了100个三七主根样品中皂苷R₁、Rg₁、Rb₁的含量，利用偏最小二乘回归法建立模型研究了三七主根中3种皂苷总含量的预测方法.校正集样品的预测值与测量值的相关系数为0.980 8，验证集样品的预测值与测量值的相关系数为0.985 6，模型性能指数达到了89.9.研究结果表明，红外光谱结合偏最小二乘法建立的模型能有效预测三七主根中3种皂苷的总含量，有望成为生产实践中快速准确预测三七主根主要皂苷总含量的方法.更多还原

【Abstract】 The prediction study of the total contents of three saponins Rb₁,Rg₁ and R₁ in Panax Notoginseng taproot was made by combining Fourier transform infrared spectroscopy （FTIR） with partial least square method.After the infrared spectra of 100 notoginseng taproot samples from 12 planting points in 7 counties,pure starch,Rb₁,Rg₁,and R₁ were tested,the derivative spectra of each sample were calculated by OPUS software.The contents of saponins R₁,Rg₁ and Rb₁ in the 100 samples were first measured according to the method stipulated for measuring the content of active components in Panax Notoginseng in pharmacopoeia,and then predicted with the model established on the basis of partial least squares regression.It is found that the correlation coefficient is 0.980 8 between the predicted value and the measured value of the calibration set and 0.985 6 between them of the verification set with a model performance index of 89.9.The research results show that FTIR combined with partial least-square regression is capable of predicting the total content of saponins in notoginseng taproot and can be used for fast and accurate prediction in production practice.更多还原