麻花艽近红外定量检测模型的影响因素研究

天然产物研究与开发 ›› 2023, Vol. 35 ›› Issue (增刊1): 66-72.

麻花艽近红外定量检测模型的影响因素研究

李佩佩1,2，龙若兰1,2，李朵1,2，冯丹1,2，孙菁1*

1中国科学院西北高原生物研究所青海省青藏高原特色生物资源研究重点实验室，西宁 810008；2中国科学院大学，北京 100049

出版日期:2023-06-28 发布日期:2023-06-13
基金资助:
青海省科研基础条件平台项目（2020-ZJ-T05）；中国科学院仪器功能开发技术创新项目（2022gl09）和青海省重点实验室建设专项（2021-ZJ-Y14）

Analysis of influencing factors to NIR qualitative models of Gentiana straminea Maxim.

LI Pei-pei1,2,LONG Ruo-lan1,2,LI Duo1,2,FENG Dan1,2,SUN Jing1*

1Qinghai Key Laboratory of Qinghai-Tibet Plateau Biological Resource,Northwest Institute of Plateau Biology,Chinese Academy of Sciences,Xining 810008,China;2University of Chinese Academy of Sciences,Beijing 100049,China

Online:2023-06-28 Published:2023-06-13

摘要/Abstract

摘要： 近红外定量检测模型的质量受谱图质量、建模方法等诸多因素影响，本研究以中藏药材麻花艽为例，探讨光谱采集条件及建模集选择对模型的影响。结果表明，扫描次数、分辨率及装样高度对谱图质量无显著影响。随着建模集样本量的增加，模型质量先升后降再升最后趋于稳定；模型质量与样本集含量间无关；而建模集SD值对模型的影响表明，样本量及样品浓度范围对模型质量的影响更大。可见，在进行建模集样本的选择时，样品浓度范围要大，而样本量不是越大越好，在进行低含量化合物建模时可通过转换含量单位进行建模。

关键词: 近红外定量模型, 谱图质量, 建模集, 麻花艽

Abstract:

The quality of NIR quantitative detection models is influenced by many factors such as spectral quality and modelling methods.In this study,the effect of spectral acquisition conditions and modelling set selection on the models were investigated using the Chinese and Tibetan herb Gentiana straminea Maxim..The results showed that the scan times,resolution and sample loading height did not have significant effects on the quality of the spectra.As the sample size of the modelling set increased,the model quality increased firstly,then decreased and stabilized at last.The model quality was found independent of the sample set content.And the effect of SD values on the model indicated that the sample size and the concentration range of the sample had great influence on the model quality.It could be seen that when making the selection of the modelling set samples,the sample concentration range should be large,whereas a larger sample size was not better.And the modelling can be carried out by converting the content units when content was too low.

Key words: NIR quantitative model, spectra quality, modeling set, Gentiana straminea , Maxim.

中图分类号: O657.33

李佩佩, 龙若兰, 李朵, 冯丹, 孙菁. 麻花艽近红外定量检测模型的影响因素研究[J]. 天然产物研究与开发, 2023, 35(增刊1): 66-72.

LI Pei-pei, LONG Ruo-lan, LI Duo, FENG Dan, SUN Jing. Analysis of influencing factors to NIR qualitative models of Gentiana straminea Maxim.[J]. NATURAL PRODUCT RESEARCH AND DEVELOPMENT, 2023, 35(增刊1): 66-72.

[1]	李佩佩1, 2，李朵1, 2，栾真杰1, 2，孟晓萍1，周玉碧1，孙菁1*. 5种秦艽组植物近红外判别模型的建立[J]. 天然产物研究与开发, 2019, 31(增刊1): 40-46.
[2]	秦龙1，王艺璇2，黄燕1，王晓飞1，焦海胜1*. 麻花艽不同部位抗氧化活性研究[J]. 天然产物研究与开发, 2016, 28(12): 1924-1928.