Discrete Element Modeling and Simulation of Soybean Seed Using Multi-Spheres and Super-Ellipsoids

The accuracy of particle shape characterization is the key to the study of particle dynamics. This study proposes two modeling methods, super-ellipsoid approach (SE) and multi-sphere approach (MS), to describe the discrete element method (DEM) model of soybean seeds with the goal of improving the accuracy of the discrete element analysis model of soybean seeds. The accumulation process, flow behavior, and mixing characteristics of soybean seeds were studied through experiments and DEM simulation. Three comparative verification tests were carried out to compare the accuracy and potential similarity of the SE particle model and the MS particle model: (1) angle of repose experiment; (2) slope screening experiment; and (3) horizontal rotating cylinder experiment. The simulation and experimental results indicated that when compared with the MS particle model, the soybean seed model established using the super-ellipsoid approach can make the described particle shape more accurate, and also reproduce the accumulation process, flow behavior, and mixing characteristics of soybean seeds more accurately. For the multi-sphere approach, the calculation cost for approximating the DEM model of soybean seeds can be saved when the number of sub-spheres is less, but the accuracy of the particle model will be reduced at the same time. In this paper, the DEM model established using the super-ellipsoid approach can accurately simulate the real movement of soybean seeds. Therefore, the model can provide a new numerical calculation method for the simulation of the particle system in the structure of agricultural machinery.