Discrete particle simulation in horizontally rotating drum: Uncertainty quantification of granular material physical parameters

This work deals with stochastic calculations for the dynamics of granular flows in rotating cylindrical drums, using the discrete element method (DEM). The drum is 100 mm in diameter and 35% filled with 3-millimetrediameter spheres rotating at various speeds. The deterministic model is extensively validated through reported studies in terms of the dynamic repose angle and spatial velocity fields. The stochastic simulation study focuses on the uncertainties related to the physical particle characteristics that simultaneously influence the granular flow interactions, such as friction coefficients, restitution coefficients, Young's modulus, Poisson's ratios and others input parameters. The uncertainty propagation from the model input parameters to the output stochastic solution is quantified using non-intrusive chaos expansion. Throughout this work, the study focuses on the behaviour of particles within a rotating drum, as many industries use this type of configuration for their processes. This method allows for the determination of stochastic solutions from a set of deterministic simulations. The simulation variables (such as particle velocities and dynamic angles of repose) are characterized in terms of the stochastic mean, probability density function and error bar. For the uncertainty propagation, the random friction coefficient is the most dominant input variable for the granular flow dynamics in a horizontal drum. (C) 2018 Published by Elsevier B.V.