In order to account for the effect of particle existence on gas-particle turbulence flow in large-eddy simulation (LES),a new gas-particle turbulent kinetic energy subgrid-scale (SGS) turbulence model is established,and the effect of particle wake is also considered in gas turbulent kinetic energy SGS turbulence model.Simulation of gas-particle turbulence flow in backward-facing step is carried out by LES using present model and by unified second-order moment (USM) model.The prediction statistical results including mean velocity and fluctuation velocity by LES using present model are in reasonable agreement with the experimental results.It is shown that present model is with higher calculating accuracy than USM model,which indicates that the turbulent kinetic energy SGS turbulence model is suitable.
Turbulence model of kg-εg-kp-εp-kpg-θ is proposed.In the model,the two-phase velocity correlation turbulent kinetic energy k pg is modeled by transport equation.To close this turbulence model,algebraic expressions of two-phase Reynolds stresses and two-phase velocity correlation variable are established by considering both gas-particle interaction and anisotropy.This turbulence model is used to simulate dense gas-particle flow in a riser and in a downer.The predicted results show the core-annulus flow structure observed in the riser and the skin effect of particle concentration in the downer.The present model gives simulation results in much better agreement with the experimental results than those obtained by kg-εg-kp-εp-θmodel which is simply closed using a semi-empirical dimensional analysis.
