Granular Flow in Silo Discharge: Discrete Element Method Simulations and Model Assessment

Abstract

Discharge dynamics of granular particles from a flat-bottomed silo is studied using both continuum modeling and three-dimensional (3D) discrete element method (DEM) simulations. Using DEM, the influence of microscopic parameters (interparticle friction coefficient, particle-wall friction coefficient and particle coefficient of restitution) and system parameters (orifice width) on the discharge rate is quantified. The spatial extent of different regimes (quasi-static, intermediate and inertial) of granular rheology are quantified using a regime map previously established from DEM data of homogeneously sheared granular flow. It is shown that all three regimes of granular rheology coexist during silo discharge, and the intermediate regime plays a significant role in discharge dynamics. A quantitative comparison between results of continuum and DEM simulations is performed by computing discharge rates, solid velocities, and solid stresses for a threedimensional (3D) flat-bottomed silo. It is found that the three constitutive models investigated in this study overpredict the discharge rate when compared to DEM data. Contour plots of the error in solid stress prediction are compared with the spatial extent of different regimes of granular rheology to deduce that it is inaccurate modeling of the intermediate regime that is responsible for overprediction of the discharge rate. ABSTRACT: Discharge dynamics of granular particles from a flat-bottomed silo is studied using both continuum modeling and three-dimensional (3D) discrete element method (DEM) simulations. Using DEM, the influence of microscopic parameters (interparticle friction coefficient, particle−wall friction coefficient and particle coefficient of restitution) and system parameters (orifice width) on the discharge rate is quantified. The spatial extent of different regimes (quasi-static, intermediate and inertial) of granular rheology are quantified using a regime map previously established from DEM data of homogeneously sheared granular flow. It is shown that all three regimes of granular rheology coexist during silo discharge, and the intermediate regime plays a significant role in discharge dynamics. A quantitative comparison between results of continuum and DEM simulations is performed by computing discharge rates, solid velocities, and solid stresses for a three-dimensional (3D) flat-bottomed silo. It is found that the three constitutive models investigated in this study overpredict the discharge rate when compared to DEM data. Contour plots of the error in solid stress prediction are compared with the spatial extent of different regimes of granular rheology to deduce that it is inaccurate modeling of the intermediate regime that is responsible for overprediction of the discharge rate. Disciplines Aerodynamics and Fluid Mechanics | Mechanical Engineering Comments Reprinted with permission from INTRODUCTIONModeling and prediction of granular flows in nature and in technological applications is a challenging problem of considerable socioeconomic i...

Numerical Investigation on Granular Flow from a Wedge‐Shaped Feed Hopper Using the Discrete Element Method

“…When different size particles are produced during the granulation process and the hopper is filled with such particles, in order to improve the discharge performance, an accurate prediction of the discharge flow pattern, rate and stability is highly important . Anand et al studied the effects of some single variables on the discharge rate and concluded that the effect of particle size on the discharge rate was minimal.…”

Implementing discrete element method for large-scale simulation of particles on multiple GPUs

“…It has been validated as an efficient method in physical, geographic or engineering (Cheng et al, A c c e p t e d M a n u s c r i p t Implementing Discrete Element Method for Large-scale Simulation of Particles on Multiple GPUs 5 2003; Ferdowsi et al, 2014;Peters et al, 2005;Vidyapati & Subramaniam, 2013). In the most conventional formulation of DEM, the Voigt model is applied at each contact point in both the normal and tangential directions.…”

Constitutive modeling of dense granular flow based on discrete element method simulations

“…Figure 2.1(b) shows that the discharge rate predicted from continuum simulations deviates considerably (more than 80%) from the Beverloo correlation (Beverloo et al, 1961) and DEM data. The discharge rate from a silo is controlled by the complex rheology of granular flow due to co-existence of different flow regimes (Vidyapati and Subramaniam, 2012b). This complex behavior makes it very difficult to formulate a comprehensive macroscale theory for granular stress tensor, which can describe all the flow regimes.…”

“…This phenomenon of regime transition has been observed in many industrial and practical applications such as discharge of granular particles from silos or hoppers. It also affects the discharge rate from these devices (Vidyapati and Subramaniam, 2012b). However, the mechanisms of regime transition have not been fully understood in spite of many studies in both engineering and physics communities.…”

“…Continuum models for granular flow need to accurately predict global flow characteristics in practical devices, such as in hopper/silo discharge, chute flow and densephase pneumatic conveying (Sundaresan, 2001). However, prediction of discharge rate obtained using existing continuum models for even a simple flow such as silo, differs considerably from correlation to experimental data (Beverloo et al, 1961) and 3D DEM (discrete element method) simulations (Vidyapati and Subramaniam, 2012b). In order for continuum models to predict global flow characteristics accurately, they need accurate constitutive models that can perform well in all the regimes (inertial, intermediate and quasi-static) of granular flow.…”