Computational Model for Granular Flows based on the Lagrangian Particle Method

TL;DR

This paper introduces a 3D computational model and parallel software for granular flows based on the Lagrangian particle method, capable of simulating transitions between solid, liquid, and gas-like regimes.

Contribution

It extends the Lagrangian particle method to granular flows with a $(I)$-rheology model, enabling accurate simulation of flow regime transitions and parallel computation.

Findings

Validated by experimental data on granular column collapse

Successfully simulated flow transitions to gas-like regimes

Achieved numerical convergence with local polynomial fittings

Abstract

A numerical model and parallel software for 3D simulations of granular flows have been developed based on the Lagrangian particle (LP) method [R.Samulyak, X. Wang, H.-C. Chen, Lagrangian particle method for compressible fluid dynamics, J. Comput. Phys. 362 (2018) 1-19], originally developed for compressible hydrodynamic flows, including free surface / multiphase problems. LP uses local polynomial least square fittings on particle-based stencils that ensure numerical convergence to the prescribed order. The granular flow model implements continuum equations with a $\mu(I)$-rheology closure that is capable of describing two-directional transitions of the flow between various regimes characterized by solid-, liquid-, and gas-like features. The granular flow code has been parallelized for distributed memory supercomputers and validated by comparing 3D simulations to experimental data on the collapse of granular columns. Numerical simulations showing the granular flow transition to a gas-like regime have also been presented.