Dissipative particle dynamics simulation of flow around spheres and cylinders at finite Reynolds numbers

Dissipative particle dynamics simulation of flow around spheres and cylinders at finite Reynolds numbers
复制标题

DOI:
10.1016/j.ces.2004.04.007
复制
发表时间:
2004-10
影响因子:
4.7
通讯作者:
J. Kim;R. Phillips
J. Kim;R. Phillips
中科院分区:
工程技术2区
文献类型:
--
作者:
J. Kim;R. Phillips

文献摘要

被引文献

相似文献

耗散粒子动力学(DPD)是一种模拟复杂流体流动和其他胶体现象的方法。它是一种介观方法,因为它不依赖于物质的连续水平描述,但也不完全捕获分子水平的细节。因此,它提供了捕捉一定程度的分子水平的细节,同时符合连续流体力学在更大的长度尺度的可能性。我们已经测试了DPD有限雷诺数流动的适用性,通过研究一系列的模型问题,涉及绕流球体和圆柱体。我们的研究是第一个明确考虑DPD模拟中有限惯性的影响。考虑了非移动的物体的绕流以及移动的物体的平移和旋转。对于我们的测试问题,我们表明,在计算可行的条件下,DPD模拟定量准确的雷诺数为50-100。通常,在较高雷诺数下,不准确的物理原因是压缩效应的开始,这可以通过参考DPD马赫数来预测。此外,在我们的DPD实现中,引入了一些新的方法,使计算时间与DPD粒子数成线性比例。它还表明,精度的提高,可以实现利用镜面反射边界条件在固体-流体界面。
Dissipative particle dynamics (DPD) is a method for simulating complex fluid flows and other, colloidal phenomena. It is a mesoscopic method, in that it does not rely on a continuum-level description of matter, but nor does it completely capture molecular-level detail. As such, it offers the possibility of capturing some degree of molecular-level detail, while conforming to continuum hydrodynamics at larger length scales. We have tested the applicability of DPD to finite-Reynolds-number flows by studying a series of model problems involving flow around spheres and cylinders. Our study is the first to consider explicitly the effect of finite inertia in DPD simulations. Both flow around immobile objects and the translation and rotation of mobile objects are considered. For our test problems, we show that under computationally feasible conditions DPD simulations are quantitatively accurate up to Reynolds numbers of 50–100. Typically the physical cause of inaccuracies at higher Reynolds numbers is the onset of compressibility effects, which can be anticipated by making reference to a DPD Mach number. In addition, in our implementation of DPD, some new methods are introduced that result in the computation time scaling linearly with the number of DPD particles. It is also shown that improvements in accuracy can be realized by making use of the specular reflection boundary condition at solid–fluid interfaces.