A novel particle subgrid scale modeling of large eddy simulation for swirling particle-laden turbulent flow

Abstract

How to describe the effects of subgrid scale (SGS) gas flow on particle dynamics in large eddy simulation (LES) is a challenge for modeling and simulation of particle-laden swirling turbulent flow. SGS four-way coupling interaction mechanisms of gas-to-particle, particle-to-gas, and particle-particle collision have not been considered fully. A novel particle SGS kinetic turbulent energy-granular temperature (SGS-kp-θp) model is proposed firstly to model this effect on particle motions and dispersions. Second-order moment algorithm to establish the anisotropic two-phase Reynolds stress equations and their closure correlations is utilized. Validations by experiment results and Moissette (MOB) particle dispersion model built-in OpenFoam software are performed with better agreements for mean and fluctuation velocities. Anisotropic behaviors of Reynolds stresses and particle collisions are distinctively, smaller-scale particle fluctuations originating from particle collisions lead to the redistribution and dissipation of Reynolds stresses. Power spectrum density is utilized to analyze statistically the particle number density fluctuations and Reynolds stresses, dominant frequencies are roughly 2.5 Hz and 1.0 Hz marking the internal and the corner recirculation regions. Compared to those of inlet flow regions, normal and tangential Reynolds stresses at development flow regions approximately increased by twice and decreased by 3.0 times. Modeling strategy for multiphase interactions are of great importance for capturing particle dynamics in accurate.