Improved coarse-grained CFD-DEM methods for simulating rough spherical particles in fluidized beds

Coarse-grained discrete element methods (CGDEM) have received much attention due to their effective reduction of computational effort. However, compared to discrete element methods (DEM), coarse-grained discrete element methods exhibit lower accuracy in modeling rough, inelastic spherical particles. This is partly because conventional coarse-graining strategies focus only on energy dissipation due to inelasticity. In this article, the impact of particle rotational inertia is incorporated into the translational energy dissipation in inter-particle collisions. A new formula for calculating the normal recovery coefficient is derived according to the strategy that dissipation in coarse-grained models is equal to real particles. The validation was conducted in both a bubbling bed and a spout-fluid bed. It turns out that collision modeling, which considers friction and roughness dissipation (CMF/R), provides better accuracy than CGDEM with the traditional coarsening strategy. Notably, in the 0.15 m height of the spout-fluid bed, CMF/R model shows a smaller relative error compared to DEM simulation experiments. The accuracy and versatility of the new model in calculating normal recovery coefficients were verified by the simulation results.