Abstract: Objective: To conduct a digital evaluation of the quality of Duyun Maojian tea infusion. Methods: Firstly, the sensory evaluation method was performed and the content of internal components was determined to define the quality of Duyun Maojian tea infusion. Then, the principal component analysis and correlation coefficient method were comprehensively applied to screen the internal components closely related to the quality of the tea infusion. Finally, an evaluation model of tea infusion quality was established by the methods of partial least squares regression (PLS) and backpropagation artificial neural network (BP-ANN). Results: The cumulative variance contribution rate of the first three principal components in the tea infusion was 97.85%. Seven components reflecting the quality of the tea infusion were screened, including amino acids, tea polyphenols, water extracts, total catechins, (-)-epigallocatechin gallate, (-)-epicatechin gallate, and (-)-epigallocatechin. The comparison of the two established models for predicting the quality of Duyun Maojian tea infusion showed that the determination coefficient of the validation set (R_\mathrmp^2 ) and the root mean square error of the prediction (RMSEP) in the linear PLS method were 0.788 and 1.264, respectively. While in the nonlinear BP-ANN method, they were 0.962 and 0.516, respectively. This indicated that the BP-ANN model had a better stability. Conclusion: The application of the BP-ANN method combined with principal component analysis and correlation analysis has achieved a rapid and accurate digital evaluation of the quality of Duyun Maojian tea infusion, which provides a useful reference for the quality evaluation of other tea infusions.