Abstract
The mechanical properties of wheat grains were measured and analyzed using discrete element software, which provided crucial data for their processing in a mill. The foundational Hertz-Mindlin model was used as a theoretical framework to evaluate the accumulation angle of wheat grains. The Plackett-Burman, steepest ascent, and Box-Behnken methods were utilized in a series of experiments, with the accumulation angle serving as the dependent variable. Targeting the actual angle of repose in the response surface, the optimal parameters were derived using the regression equations. These included a static-friction coefficient of 0.3 between individual wheat grains, rolling-friction coefficient of 0.04 for wheat-wheat interactions, static-friction coefficient of 0.554 for wheat-tooth roller interactions, collision recovery coefficient of 0.5 for wheat-wheat collisions, collision recovery coefficient of 0.45 for wheat-tooth roller collisions, and rolling-friction coefficient of 0.05 for wheat-roller interactions. Relying on the bonding contact model of Hertz-Mindlin, virtual uniaxial compression tests were conducted to calibrate the wheat grain bonding parameters. A regression equation for the critical load was subsequently generated using the critical load of the wheat grain bonding model as the response variable. The optimal parameters were calculated and incorporated into the EDEM model for computation, which resulted in a relative error of 1.6% between the calculated and observed values, confirming the accuracy and feasibility of the calibration method, suggesting that the calibrated parameters were accurate.