Alternative ways of coupling particle behaviour with fluid dynamics in mineral processing

Abstract

Research in particulate multiphase hydrodynamics is especially important in mineral processing applications that involve complicated physical processes and complex geometry. Numerous numerical methods have been developed to address the coupling physics. However, there is no single solution for all applications, especially after considering computational costs and model accuracy.

This article presents alternative numerical coupling approaches to capture the physics of interest. A one-way coupling of conventional grid-based CFD with discrete element method (DEM) is applied to simulations of slurry pumps. FLUENT is used to compute the fluid field in the pump, which is imported into our DEM code. The DEM is used to capture dynamics of the detailed behaviour of individual solid particles, including inter-particle collisions and particle impacts into structure (boundaries).

A Lagrangian–Lagrangian multiphase model is proposed for simulation of dense particulate flows in a grinding mill. This work is motivated by developing totally meshfree approaches for particulate flows due to complex and rotating geometry in many pieces of equipment in the mining industry. The particle-based multiphase model is an integration of DEM with smooth particle dynamics (SPH). SPH is extended to the multiphase version where an extra variable, volume fraction, is introduced into the governing equation in order to account for interpenetrating multiphase fluid dynamics. The interaction between the solid particles and the fluid phase consists of three contributions: volume fraction, pressure and drag force. The time stepping scheme can be either explicit Euler for speed-up or lower order Runge–Kutta for more accuracy, and the time step is controlled by the Courant condition. The model is applied to a preliminary study of a semi-autogenous grinding mill.

Keywords:

• DEM
• SPH
• Lagrangian–Lagrangian
• DEM-SPH
• multiphase flow
• particulate flow
• wet mill simulation
• pump simulation

Acknowledgements

The authors' acknowledgements and thanks go to High Fidelity Simulation team of Metso Minerals Optimization Services at Colorado Springs, especially to Dr. Xiangjun Qiu and Dr. Alex Potapov.