Analytical solution for the problem of frictional-elastic collisions of spherical particles using the linear model

Numerical Analysis of the Relationship between Friction Coefficient and Repose Angle of Blast Furnace Raw Materials by Discrete Element Method

In recent years, the discrete element method (DEM) has been widely used to study the factors affecting the repose angle and calibrate particle parameters for simulations. In this paper, DEM is used to study the effects of the coefficient of rolling and static friction of pellet, sinter and coke particles on the repose angle. By comparison of the results of simulations and physical experiments, the coefficients of rolling and static friction suitable for simulation work are determined. The results demonstrate that repose angle increases with the coefficient of rolling and static friction, but the rate of increase gradually decays, when the coefficient of rolling friction exceeds 0.4 or the coefficient of static friction exceeds 0.35. The coefficient of static friction has a greater impact on the repose angle than the coefficient of rolling friction. The rougher of the base surface, the larger the repose angle of the formed particle piled. It can be concluded that appropriate coefficient of rolling and static friction for simulations can be obtained by the outlined procedure.

Extension of the hard-sphere model for particle-flow simulations

Discrete element methods require appropriate models for particle-particle collisions. Usually, researchers use soft-sphere types of models where the collision dynamics is solved numerically. This makes the simulation computationally expensive. In this paper, however, we show a hard-sphere model that uses ready analytic formulas that relate the pre- and postcollisional velocities of the particles in contact. This hard-sphere model is an extension of an existing model that uses three input parameters. For this, we applied the linear-spring soft-sphere model, where analytic relations can be found. These relations were implemented into the standard hard-sphere model. As a result, we obtain a robust hard-sphere model that is more accurate than the standard one and is still computationally cheap.