Progress in particle-based multiscale and hybrid methods for flow applications
Progress in particle-based multiscale and hybrid methods for flow applications
复制标题
DOI：
10.1007/s10404-016-1729-y
复制
发表时间：
2016-04
影响因子：
2.8
通讯作者：
Tom-Robin Teschner;L. Könözsy;K. Jenkins
中科院分区：
文献类型：
--
作者：
Tom-Robin Teschner;L. Könözsy;K. Jenkins
这项工作的重点是回顾过去 20 年在流体力学领域出现的基于粒子的多尺度和混合方法。我们考虑五种已建立的粒子方法：分子动力学、直接模拟蒙特卡罗、格子玻尔兹曼方法、耗散粒子动力学和平滑粒子流体动力学。结合多尺度和混合应用对每种粒子方法进行了一般描述。对长度尺度分离的分析表明，当前的多尺度方法只能跨越 数量级的尺度，并且需要对复杂几何形状和并行代码优化进行进一步的工作以增加分离。强调了方法之间的相似之处并讨论了组合。列出了每种粒子方法的优点、缺点和应用，以供参考。
This work focuses on the review of particle-based multiscale and hybrid methods that have surfaced in the field of fluid mechanics over the last 20 years. We consider five established particle methods: molecular dynamics, direct simulation Monte Carlo, lattice Boltzmann method, dissipative particle dynamics and smoothed-particle hydrodynamics. A general description is given on each particle method in conjunction with multiscale and hybrid applications. An analysis on the length scale separation revealed that current multiscale methods only bridge across scales which are of the order ofand that further work on complex geometries and parallel code optimisation is needed to increase the separation. Similarities between methods are highlighted and combinations discussed. Advantages, disadvantages and applications of each particle method have been tabulated as a reference.