Welcome to Journal of Tea Science,Today is

Journal of Tea Science ›› 2021, Vol. 41 ›› Issue (2): 251-260.doi: 10.13305/j.cnki.jts.2021.02.007

• Research Paper • Previous Articles     Next Articles

Rapid Detection Method of Tea Polyphenol Content in Black Tea Fermentation Based on Electrical Properties

WANG Shenglin1,2, YANG Chongshan2, LIU Zhongyuan2, LIU Shanjian1,*, DONG Chunwang2,*   

  1. 1. School of Agricultural and Food Engineering, Shandong University of Technology, Zibo 255000, China;
    2. Tea Research Institute, The Chinese Academy of Agricultural Sciences, Hangzhou 310008, China
  • Received:2020-11-09 Revised:2021-01-19 Online:2021-04-15 Published:2021-04-13

Abstract: Tea polyphenols are an important evaluation index for the quality of black tea. The quantitative prediction model of tea polyphenol content in the fermentation process was established by combining electrical characteristics detection technology with chemometric method. The changes of electrical parameters during the fermentation process and the influence of different standardized pretreatment methods and variable optimization algorithms on the model were discussed. The results show that the most sensitive electrical parameters to tea polyphenols were Cp, D and X, all of which were concentrated in the low frequency range (0.05-0.10 kHz). In the construction of tea polyphenol prediction model, normalization processing (Zscore) and mixed variable screening (VCPA-IRIV) can effectively improve the performance of the model. The number of variables introduced in the VCPA-IRIV-PLS model was reduced from 162 to 31, and the compression rate reached 80.86%. RMSECV and RMSEP were reduced to 0.630 and 1.116, respectively. Rp and RPD were increased to 0.941 and 2.956. The research results show that the electrical characteristics detection technology is feasible for the rapid non-destructive detection of the content of tea polyphenols in black tea fermentation.

Key words: black tea fermentation, electrical characteristics, tea polyphenols, variable screening, model

CLC Number: