Calibration and Validation of Contact Parameters for DEM Simulation of Mechanically Harvested Fresh Tea Leaves

To enhance the precision of Discrete Element Method (DEM) simulation parameters for the grading of mechanically harvested fresh tea leaves, this study systematically measured the intrinsic physical and basic contact parameters of the Yinghong No. 9 cultivar. Addressing the distinction between primary and secondary contact interfaces during roller screening, the extreme boundary validation method was first employed to determine simplified fixed values for the contact parameters of the secondary component. Based on the measured physical angle of repose of 36.3°, Plackett–Burman screening, steepest ascent, and Box–Behnken tests were conducted sequentially to construct and optimize a second-order regression model relating significant parameters to the angle of repose. The results indicated that the static friction coefficient between tea leaves (0.723), the rolling friction coefficient between tea leaves (0.031), and the static friction coefficient between tea leaves and the PVC roller (0.547) were the key parameters affecting the angle of repose. Verification tests demonstrated that the simulated static angle of repose was 36.9° against the measured 36.3°, yielding a relative error of 1.65%. The simulated dynamic angle of repose in the rotating drum was 39.8° compared to the physical 38.3°, representing a relative error of 3.92%, and the errors in screening efficiency on the grading bench were all less than 5%. These results indicate that the calibrated parameters accurately characterize the material properties of mechanically harvested tea leaves, providing a reliable theoretical foundation for the structural optimization of grading equipment.