A numerical continuous model for the hydrodynamics of fluid particle systems
A numerical continuous model for the hydrodynamics of fluid particle systems
复制标题
流体颗粒系统流体动力学的数值连续模型
DOI：
--
复制
发表时间：
1999
期刊：
影响因子：
--
通讯作者：
J. Caltagirone
中科院分区：
文献类型：
--
作者：
J. Ritz;J. Caltagirone
为了理解流体质点运动中可能出现的流体动力相互作用，人们提出了一种基于控制两种不混相流体运动的方程的原始方法。这些动量方程在流体和固相中都得到了求解。假定固相是具有物理性质的流体相，例如其行为可以与伪刚性颗粒的行为同化。唯一的未知数是两相的速度和压强。采用交错有限体积法和投影法求解非定常二维动量方程。每个粒子的输运用二阶显式格式求解。提出了该方法的物理模型和数值计算方法，并通过圆柱绕流的实验测量和数值计算结果对该方法进行了验证。在大多数情况下，可以观察到良好的一致性。然后应用该方法研究了一个粒子在两个平行壁面之间初始偏离中心的沉降轨迹和相应的尾迹效应。提出并评述了与粒子雷诺数相关的不同粒子轨迹。研究了两体相互作用问题。这种方法允许在合理的时间内模拟稀悬浮液中颗粒的输运。该方法的一个重要特点是计算成本与粒子数量成线性关系。版权所有©1999 John Wiley & Sons， Ltd
SUMMARY In order to understand the hydrodynamic interactions that can appear in a fluid particle motion, an original method based on the equations governing the motion of two immiscible fluids has been developed. These momentum equations are solved for both the fluid and solid phases. The solid phase is assumed to be a fluid phase with physical properties, such as its behaviour can be assimilated to that of pseudo-rigid particles. The only unknowns are the velocity and the pressure defined in both phases. The unsteady two-dimensional momentum equations are solved by using a staggered finite volume formulation and a projection method. The transport of each particle is solved by using a second-order explicit scheme. The physical model and the numerical method are presented, and the method is validated through experimental measurements and numerical results concerning the flow around a circular cylinder. Good agreement is observed in most cases. The method is then applied to study the trajectory of one settling particle initially off-centred between two parallel walls and the corresponding wake effects. Different particle trajectories related to particulate Reynolds numbers are presented and commented. A two-body interaction problem is investigated too. This method allows the simulation of the transport of particles in a dilute suspension in reasonable time. One of the important features of this method is the computational cost that scales linearly with the number of particles. Copyright © 1999 John Wiley & Sons, Ltd.