以速溶红茶和速溶乌龙茶为研究对象,利用近红外光谱检测技术结合化学计量方法,开发了一套速溶茶内部成分含量在线检测系统.该系统根据实时采集的速溶茶光谱数据自动检测速溶茶中茶多酚、茶氨酸、咖啡碱成分含量.结果表明,比较分析不同预处理方法后,选择使用均值中心化-偏最小二乘法回归(Mean centering-partial least squares regression,MC-PLS)对速溶茶的茶多酚进行定量分析,校正集和预测集的R2分别为0.9965、0.9981;使用MC-PLS对速溶茶咖啡碱进行定量分析,校正集和预测集的R2分别为0.9985、0.9986;使用NONE-PLS对速溶茶茶氨酸进行定量分析,校正集和预测集的R2分别为0.9991、0.9998.建立的速溶茶内部成分含量分析模型与检测系统可以实现速溶茶中茶多酚、茶氨酸和咖啡碱的快速无损检测.
Rapid Detection of Internal Component Content in Instant Tea Based on Near-Infrared Spectroscopy
Taking instant black tea and instant oolong tea as research objects,combining near-infrared spectroscopy detection technology with stoichiometric methods to develop an online detection system for the content of internal components in instant tea.The system automatically detects the content of tea polyphenols,theanine,and caffeine in instant tea based on real-time collected spectral data.The results indicate that after comparing and analyzing different preprocessing methods,choosing to use mean centering-partial least squares regression(MC-PLS)for quantitative analysis of instant tea polyphenols,The R2 values of 0.9965 for the correction set and 0.9981 for prediction set,respectively.Quantitative analysis of instant tea caffeine using MC-PLS,the R2 values of 0.9985 for the correction set and 0.9986 for prediction set,respectively.Quantitative analysis of instant theanine using NONE-PLS,the R2 values of 0.9991 for the correction set and 0.9998 for prediction set,respectively.The established internal component content analysis model and detection system can achieve rapid non-destructive detection of tea polyphenols,theanine,and caffeine in instant tea.
万方数据
维普
