Direct and Large-Eddy Simulation of Particle-Laden Flows Using the Point-Particle Approach
暂无分享,去创建一个
In the point-particle approach the particle-fluid interaction is modelled through a force located at the center of the particle, leading to modified equations for the continuous phase, which are solved using numerical methods similar to the ones used in single-phase flows (DNS, LES), together with the tracking of the individual particles. Contrary to single-phase flows, where the only limitation to DNS/LES is the availability of computer resources, it involves a delicate balancing act between two opposing requirements on the grid-cell size. DNS is potentially more accurate from the perspective of the continuous-phase requirements, at the expense of being less accurate from the perspective of the point-particle requirements, and LES the other way around. We present a criterion to estimate the errors associated with the two opposing requirements, and show that, depending on the type of particle and the particle-loading, the best approach can be a compromise solution: high-resolution LES.
[1] John R. Fessler,et al. Particle response and turbulence modification in fully developed channel flow , 1994, Journal of Fluid Mechanics.
[2] P. J. Mason,et al. On the magnitude of the subgrid-scale eddy coefficient in large-eddy simulations of turbulent channel flow , 1986, Journal of Fluid Mechanics.
[3] J. Riley,et al. Equation of motion for a small rigid sphere in a nonuniform flow , 1983 .
[4] Joel H. Ferziger,et al. Computational methods for fluid dynamics , 1996 .