Discrete element model construction and dehulling simulation verification based on harvested quinoa grains

The dry dehulling process of quinoa grains faces challenges such as poor visibility and unclear dynamic response characteristics, leading to irrational settings of the shelling process and high losses of the final processed product. Simulation methods provide an effective solution to this issue, with the accuracy of the simulation model being crucial. In this study, using the Experts in Discrete Element Modeling (EDEM) simulation software, based on the biological characteristics and mechanical properties of the grains, a method was proposed to construct a discrete-element simulation model for the double-layer bonding of the quinoa grain using the three-axis spatial coordinates method. The Plackett-Burman method, the steepest rise method and the Box-Behnken method were used to simulate the compression test. The bonding parameters of the quinoa grain were calibrated. The parameters were further validated by constructing a mechanical shear test. Finally, the process of the dehulling quinoa grain was simulated in a dry dehulling equipment. The results showed that the relative error between the simulated and actual maximum compression force of the quinoa grains was 0.32%. The breaking shear force in the simulation had a relative error of no more than 5% compared to the actual measurements. In the simulation, the dehulling rate and broken rice rate were 78.54% and 1.38%, respectively, which exceeded the bench test values (75.32% for dehulling rate and 1.24% for broken rice rate) by 3.22% and 0.14%, respectively. The error was within the acceptable range. The calibration of the discrete element model parameters for quinoa grains was accurate, reflecting the mechanical properties differences between the rice and the pericarp effectively, providing a reliable basis for optimizing the design of dry dehulling mechanisms.