Modelling soil–sweep interaction with discrete element method

• The development of a (3D) discrete element method (DEM) model for the analysis of soil–sweep interaction.
• It implemented computer aided design (CAD) systems to simulate the sweep geometries.
• The parallel bond model of discrete elements approximate draught and soil deformations appropriately.
• With the triaxial test we validated the micro properties.
• Results showed that the comparison between the measured and predicted draught provided good match.

Several analytical methods of the soil–tool interaction have been developed and tested, but they are time consuming and require large effort, which has prevented their widespread use. This paper presents the development of a three-dimensional (3D) discrete element method (DEM) model for the simulation of soil–sweep interaction. The aim was to understand the effects of the sweep rake angle (β) and speed on draught and soil loosening. It implements computer aided design (CAD) systems to simulate the sweep geometry. The DEM model output was validated by comparing simulated and corresponding actual soil bin measurements using a cohesive wet sandy soil. Cohesion of the wet sandy soil was assigned using a parallel bond contact model, where the normal and shear stiffness of the bond, the normal and shear strength, and the size of the connecting geometry were the main parameters. Following the comparison between the simulated and measured draught based on input parameters measured with a direct shear box test, virtual DEM triaxial compression analyses were performed to refine the DEM model parameters including cohesion, internal friction angle, modulus of elasticity and Poisson's ratio, using the Mohr–Coulomb failure criterion.Results showed that the comparison between the measured and predicted draught of a sweep tine with a 30° β provided good match, with rather small error range of 4–15% for selected speed interval of 0.5–2.4 m s−1. A further refinement of the model parameters with the DEM triaxial test led to improved prediction accuracy of draught to be in the range of 4–9%. The displacement vectors of the soil in front of the sweep showed a similar soil failure pattern to a wedge-shape failure. Both soil loosening and draught increased with the travel speed and the sweep rake angle, where the largest porosity (0.489) and draught (4452 N) were calculated for a rake angle of 45° and a tool speed of 4 m s−1. It can be concluded that the developed DEM model is a useful tool to simulate the interaction between soil and sweep tines accurately.