Abstract: In view of the low grading efficiency and large damage to fresh leaves of the current machine-picked tea fresh leaf grading equipment, a vibrating machine-picking fresh tea leaf classifier was used in this study as the test object, and a new tea leaf viscoelastic material modeling method was proposed. The vibration frequency, vibration direction angle, and amplitude were used as independent variables. The sieving rate and famous tea sieving rate were used as target optimization values. The simulation test was carried out based on the EDEM software. Using the response surface optimization method, the optimized parameters were obtained as the vibration frequency of 29.0 Hz, the amplitude of 9.0 mm, and the vibration direction angle of 34.7°. Under this parameter, the sieving rate obtained by the simulation test was 75.38%, and the sieving rate of famous tea was 95.65%. According to the optimal parameter combination, the simulation and prototype verification test were carried out, and the results show that the sieving rate obtained by the prototype verification test was 71.07%, and the sieving rate of famous tea was 93.26%. Combining the simulation test and prototype test results, the accuracy of the screening rate reached 93.9%, and the accuracy of the screening rate of famous tea reached 97.4%. The simulation optimization parameters based on response surface analysis had high reliability. Under this optimal parameter, the classifier had a better classification effect. This study provided a reference for the optimization of machine-picked tea fresh leaf grading equipment.

Key words: machine-picked fresh tea leaves, vibration classifier, fresh leaves material modeling, response surface optimization