265974 A Numerical Method for Large-Scale Dense Particulate Flows Based On an Improved MP-PIC Method and Extended Lattice-Boltzmann Scheme

Wednesday, October 31, 2012: 1:10 PM
Conference C (Omni )
Radompon Sungkorn, Chemical and Materials Engineering, University of Alberta, Edmonton, AB, Canada and Jos Derksen, Department of Chemical & Materials Engineering, University of Alberta, Edmonton, AB, Canada

         Detailed simulation of large-scale, dense particulate flows is extremely demanding. Prohibitively high computational expenses arise from large numbers of particles and their interactions. In order to be able to gain insight into such flows, numerical methods have been proposed in the literature. A multiphase particle-in-cell (MP-PIC) method offers a satisfactory level of accuracy with reasonable computational expenses [1]. In the MP-PIC method, particles with identical properties (e.g., particle diameter, velocity) are represented by a parcel and tracked in a Lagrangian manner. At the particle volume fraction near close-packing, conditions are imposed to prevent the particle volume fraction beyond close-packing. Effects of particle collisions are modeled based on the local particle distribution and their properties. Recently, O'Rourke et al. [2] proposed a collision procedure based on a Bhatnagar, Gross, and Krook (BGK) approximation to the collision terms in a particle distribution function transport equation. A collision damping time which represents the rate toward the local average particle velocity was introduced.

         We present a numerical method for large-scale dense particulate flows with applications to sedimenting suspensions. The locally averaged conservation equations for the fluid phase are discretized using an extended lattice-Boltzmann (LB) scheme. A MP-PIC method with an improved  collision procedure is used to track parcels of particles. A new collision damping time is derived from the discrete kinetic equation for the particle distribution function. The LB scheme and the MP-PIC method are coupled resulting in the so-called four-way coupling. The computer code for the present method is implemented and parallelized using the message passing interface (MPI) library.

         The present method is used to carry out simulations of dense sedimenting suspensions in a cubic periodic domain. In order to evaluate the accuracy of the present method, the predicted average and variance of particle settling velocity are compared to the experimental data from the literature [3]. In addition, the excellent speedup on a parallel computing platform of the present method demonstrates its potential for simulations of industrial-scale particulate flows.


[1] Snider D. M. An incompressible three-dimensional multiphase particle-in-cell model for dense particle flows. J. Comp. Phys. 2001;170:523-549.

[2] O'Rourke P. J., Snider D. M. An improved collision damping time for MP-PIC calculations of dense particle flows with applications to polydisperse sedimenting beds and colliding particle jets. Chem. Eng. Sci. 2010;65:6014-6028.

[3] Guazzelli E., Hinch J. Fluctuations and instability in sedimentation. Annu. Rev. Fluid. Mech. 2011;43:97-116.

Extended Abstract: File Not Uploaded