Principal Component Analysis and Path Analysis of Main Chemical Components and Sensory Quality of White Tea in Different Years
By analyzing the changes of main chemical components and sensory quality of white tea cakes in different years,it provides support for reasonable evaluation of the relationship between storage years and sensory quality of white tea cakes.Taking white tea cakes stored for 2 to 19 years as research materials,the main chemical components were determined,combined with sensory evaluation,and the relationship between main chemical components and sensory quality was discussed by principal component analysis,cluster analysis,correlation analysis,multiple linear regression and path analysis.With the increase of storage years,tea polyphenols,free amino acids and thearubigins showed a downward trend,with the lowest values of 50.71,5.78 mg/g and 6.82 mg/g,respectively.However,total flavonoids and theabrownin showed an upward trend,with the highest values of 62.18 mg/g and 27.09 mg/g,respectively.The results of principal component analysis and sensory analysis showed that white tea cakes stored for 4 years ranked first in the overall quality and the highest total score of 96.8 in sensory quality was considered to have the best quality.Cluster analysis showed that the transformation reaction of tea cake was intense in the second,fourth and sixth years of storage,but it became stable and its sensory quality gradually increased in the eighth to nineteenth years.Path analysis showed that in the changes of sensory quality of white tea cakes in different years,caffeine content was positively correlated with sensory quality of aged white tea cakes,while total flavonoids content was negatively correlated.The sensory quality of white tea cakes stored for 4 years,10 years and 19 years is the best overall,and the contents of caffeine and total flavonoids can be used as direct factors to evaluate the quality of aged white tea cakes.
pancake white teadifferent yearschemical compositionsensory analysiscluster analysispath analysis
万方数据
维普
