论文部分内容阅读
目的:运用近红外光谱(NIRS)技术对丹参中醇浸出物含量进行快速测定。方法:将近红外光谱与丹参中醇浸出物的含量进行关联,对光谱预处理方法和建模区间进行了考察,采用偏最小二乘法(PLS)建立定量分析模型。结果:56份样品所建立的定量分析模型内部交叉验证决定系数(R2)为0.99814,均方根偏差(RMSEC)为0.236,预测均方根偏差(RMSEP)为1.04;经验证,9份验证集样品真实值和预测值的平均相对偏差为2.60%。结论:近红外光谱法建立的定量分析模型,能够对丹参中醇浸出物的含量进行快速测定。
Objective: To determine the content of alcohol extract in Radix Salviae Miltiorrhizae by near infrared spectroscopy (NIRS). Methods: The content of alcohol extract in Salvia miltiorrhiza was correlated with the near-infrared spectroscopy. The spectral pretreatment method and modeling interval were investigated. Partial least square method (PLS) was used to establish the quantitative analysis model. Results: The internal cross validation coefficient (R2) of the quantitative analysis model established by 56 samples was 0.99814, the root mean square deviation (RMSEC) was 0.236, and the root mean square error of prediction (RMSEP) was 1.04. It was verified that 9 validation sets The average relative deviation between the true value and the predicted value of the sample was 2.60%. Conclusion: The quantitative analysis model established by near infrared spectroscopy can quickly determine the content of alcohol extract in Salvia miltiorrhiza.