麸炒白术水分和浸出物含量近红外检测模型建立

Establishment of NIR Detecting Model of Moisture and 60% Ethanol Extract Content in Bran-baking Rhizoma Atractylodis

目的基于声光可调滤光器-近红外技术(AOTF-NIR)建立一种简便、快速的麸炒白术水分和浸出物含量检测模型。方法将样品分成校正集和验证集,采用主成分分析法和偏最小二乘法分别建立检测模型,并采用配对t检验检验模型的可靠性。结果麸炒白术水分检测模型的主因子数为4,理论相关系数=0.9834,预测相关系数0.9574,外部验证平均相对偏差为4.99%。浸出物检测模型的主成分数为8,理论相关系数0.9572,预测相关系数0.9088,平均相对偏差为4.87%。验证集10批麸炒白术的水分含量、浸出物含量的预测值和实测值均无显著性差异(P>0.05)。结论成功建立麸炒白术的水分和浸出物近红外定量检测模型,可以准确预测麸炒白术的水分和浸出物含量,为麸炒白术的质量控制提供了一种简便、快速、环保的分析方法。

Objective To establish a method for determining the moisture and 60%ethanol extract content of Bran-baking Rhizoma Atractylodis(BBRA).Methods The collected samples were divided into calibration set and verification set.The detection model was established by principal component analysis and partial least squares respectively,and the reliability of the model was tested by paired t-test.Results The main factor of the moisture detection model of BBRA was 4,the theoretical correlation coefficient was 0.9834,and the prediction correlation coefficient was 0.9574.The external verification average relative deviation was 4.99%.The main component of the BBRA extract model was 8,the theoretical correlation coefficient was 0.9572,the predicted correlation coefficient was 0.9088,and the average relative deviation was 4.87%.The paired t-test results showed that there was no significant difference between the predicted and measured values of the moisture content and the extract content of the 10 batches of the validation set(P>0.05).Conclusion The NIR quantitative detection model of moisture and extract of BBRA has high prediction accuracy and precision.The model is stable and the prediction effect is good.It can accurately predict the moisture and leachate of BBRA,which provides simple,rapid and environmentally friendly method for quality control of BBRA.