基于近红外光谱分析技术的菊花品种识别及3种质量标志物含量快速测定
Identification of Chrysanthemum Varieties and Rapid Determination of Three Quality Markers Based on Near-Infrared Spectroscopy
彭璐 1吴卫刚 1范荫荫 1薄雯映 1李文龙 2徐凌峰 1梅之南3
作者信息
- 1. 河北楚风中药饮片有限公司,河北保定 071200
- 2. 天津中医药大学,天津 301617
- 3. 华中农业大学,湖北武汉 430070
- 折叠
摘要
目的 建立能快速识别不同品种菊花及快速测定其3种质量标志物的方法.方法 分别采用偏最小二乘法和正交偏最小二乘判别分析法建立菊花中绿原酸、木犀草苷和3,5-O-二咖啡酰基奎宁酸的定量分析模型以及菊花品种识别的定性分析模型.结果 所构建的定量分析模型对校正集中3种质量标志物进行预测,相关系数(R)依次为0.9215,0.9404,0.9795,校正均方差(RMSEC)依次为 0.0880％,0.0522％,0.1150％,预测均方差(RMSEP)依次为 0.0911％,0.0452％,0.1760％.经统计学分析,定量分析模型对验证集的预测值与参考值无显著性差异.O2PLS-DA模型可以快速区分不同品种菊花.结论 近红外光谱技术结合化学计量学方法可以实现对菊花3种质量标志物含量的快速测定和不同品种菊花快速识别.
Abstract
Objective To establish a method for the rapid identification of different varieties of Chrysanthemum and the rapid de-termination of three quality markers.Methods The quantitative analysis models for chlorogenic acid,luteolin and 3,5-O-di-caffeoylquinic acid in Chrysanthemum were established by partial least squares(PLS),and the qualitative analysis model for the identification of Chrysanthemum varieties was established by orthogonal partial least squares discriminant analysis(O2PLS-DA).Results The constructed quantitative analysis model predicted the three quality markers in the calibration set with correla-tion coefficients(R)of 0.9215,0.9404 and 0.9795,root mean square error of calibration(RMSEC)of 0.0880％,0.0522％and 0.1150％,and relative standard error of prediction(RMSEP)of 0.0911％,0.0452％and 0.1760％,respectively.Statisti-cal analysis showed no significant difference between the predicted values of the quantitative analysis models for the validation set and the reference values.The O2PLS-DA model can quickly distinguish different varieties of Chrysanthemum.Conclusion Near-infrared spectroscopy combined with chemometrics can achieve the rapid determination of three quality markers in Chrysanthe-mum and the rapid identification of different varieties of Chrysanthemum.
关键词
近红外光谱技术/菊花/偏最小二乘法/正交偏最小二乘判别分析/质量评价Key words
Near-infrared spectroscopy/Chrysanthemum/Partial least squares method/Orthogonal partial least squares discriminant analysis/Quality evaluation引用本文复制引用
基金项目
河北省创新能力提升计划(215A2501D)
湖北省科技重大专项(2021ACA004)
出版年
2024
国家科技期刊平台
NSTL
万方数据