Calibration of Discrete Element Parameters for Maize Kernels Using the Tavares UFRJ Breakage Model

This study established a stepwise procedure for calibrating the contact parameters and Tavares UFRJ breakage-model parameters required for discrete element method (DEM) simulations of Denghai 605 maize kernels. Kernel–kernel and kernel–carbon-steel contact parameters were calibrated using a cylinder-lifting angle-of-repose test, partitioned-tray restitution tests, and a Box–Behnken design. Single-kernel quasi-static compression tests characterised the distribution and size dependence of specific breakage energy, while unsupported electromagnetic impact tests provided estimates of the damage accumulation coefficient γ and fragment-size distribution parameter b. The estimation of b assumed A = 50%. The calibrated restitution, static-friction, and rolling-friction coefficients were 0.72, 0.18, and 0.06 for kernel–kernel contact and 0.75, 0.28, and 0.08 for kernel–carbon-steel contact, respectively. The simulation produced an angle of repose of 24.91°, with a relative error of 4.27% from the experimental value of 23.89°. Within the observed range, an upper-truncated lognormal distribution described the specific breakage energy (E50 = 386.83 J/kg, σ = 0.4267, and R2 = 0.9897). The size-effect parameters E∞, d0, and φ were 175.25 J/kg, 9.21 mm, and 1.55, respectively (R2 = 0.9381). The calibrated values of γ and b were 5.2 and 0.0253, respectively. The relative error between the simulated and experimental mean breakage forces was 3.17%, and the root-mean-square error of the cumulative particle-size distribution was 4.04 percentage points. Under the tested material and loading conditions, these results provide a parameter set for evaluating subsequent DEM and coupled CFD–DEM simulations.