基于近红外光谱技术对武夷水仙主要理化成分的快速检测

Rapid Detection of Main Physicochemical Constituents of Wuyi Shuixian Tea Based on Near Infrared Spectroscopy

为实现近红外光谱技术对武夷水仙中水分、水浸出物、茶多酚和咖啡碱4种主要理化成分的同时快速检测。文章采用近红外光谱分析技术结合偏最小二乘法对30份武夷水仙茶叶中4种主要理化成分含量进行定量建模分析,手动选25个样品作为校正级,剩余5个样品作为预测集。实验结果表明:经过光谱预处理和波段筛选,武夷水仙茶叶中4个理化成分的最佳模型的校正集系数(R_(c))和验证集系数(R_(p))均大于0.97,校正集均方误差(R_(MSEC))范围在0.0632~0.230、验证集均方误差(RMSEP)范围在0.118~0.688。通过验证实验显示,武夷水仙中水分、水浸出物、茶多酚和咖啡碱的定量模型效果良好,可实现各成分同时快速高效检测。该方法有助于克服传统检测方法中存在的过程复杂、耗时耗力、污染环境、检测效率低等问题。

In order to realize the rapid detection of moisture,water extract,tea polyphenols and caffeine in Wuyi Shuixian tea by near infrared spectroscopy(NIS).In this paper,NIS and partial least squares(PLS)were used to model and analyze the four main physical and chemical components of 30 Wuyi Shuixian teas.25 samples were manually selected as the correction level,and the remaining 5 samples were used as the prediction set.The experimental results showed that after spectral preprocessing and band screening,the calibration set coefficients(R_(C))and verification set coefficients(R_(P))of moisture,water extract,tea polyphenols and caffeine in Wuyi Shuixian tea were all greater than 0.97.The calibration set mean square error(RMSEC)ranges from 0.0632 to 0.230,and the validation set mean square error(R_(MSEP))ranges from 0.118 to 0.688.The validation test showed that the quantitative models of moisture,water extract,tea polyphenols and caffeine in Wuyi Shuixian tea were effective and could be used for rapid and efficient determination of each component simultaneously.This method is helpful to overcome the problems existing in traditional detection methods such as complicated process,time-consuming and labor-consuming,environmental pollution and low detection efficiency.