首页> 外文期刊>Computational Mechanics >A comparison of staggered solution schemes for coupled particle–continuum systems modeled with the Arlequin method
【24h】

A comparison of staggered solution schemes for coupled particle–continuum systems modeled with the Arlequin method

机译:用Arlequin方法建模的颗粒-连续体耦合系统的交错求解方案比较

获取原文
获取原文并翻译 | 示例
           

摘要

This contribution aims at a systematic investigation of staggered solution schemes for the computation of coupled domains having different resolutions in space, a problem frequently arising in multi-scale modeling of materials. To couple a standard finite element domain with a high resolution atomistic or coarse-grained, i.e. particle-based domain, a so-called bridging domain is considered. In this handshake region a total energy, which is the sum of the weighted energies of both domains, needs to be formulated. Interactions in the particle domain are modeled by potential functions, e.g. a harmonic potential in the simplest case or the Lennard-Jones potential to consider also anharmonic interactions between the particles. The main goal is to separate the computation of finite element and particle domains as much as possible, amongst others to calculate the different domains on several CPUs. In the present work, the governing equations of the coupling method are presented. The energy functions of continuum, particle domain and bridging domain are recapitulated and the coupling constraint is set up. For the sake of simplicity, these relations are reformulated for the case of a one dimensional system. On the one hand, this system is computed monolithically without any separation of domains. On the other hand, various staggered solution schemes are derived systematically. The relevant equations of each scheme are given in detail together with the sequent iteration steps. All staggered schemes are investigated qualitatively, e.g. by their convergence behavior, as well as quantitatively by comparing the staggered solutions with the monolithic solution.
机译:该贡献旨在对交错求解方案的系统研究,以计算具有不同空间分辨率的耦合域,这是在材料的多尺度建模中经常出现的问题。为了将标准有限元域与高分辨率的原子域或粗粒度域(即基于粒子的域)耦合,考虑了所谓的桥接域。在此握手区域中,需要制定总能量,这是两个域的加权能量之和。粒子域中的相互作用是通过潜在的函数来建模的,例如最简单情况下的谐波电位或Lennard-Jones电位也要考虑粒子之间的非谐相互作用。主要目标是尽可能地将有限域和粒子域的计算分开,以在多个CPU上计算不同的域。在本文中,给出了耦合方法的控制方程。概括了连续体,粒子域和桥域的能量函数,并建立了耦合约束。为了简单起见,对于一维系统,重新定义了这些关系。一方面,该系统是整体计算的,没有任何域的分离。另一方面,系统地推导了各种交错的解决方案。给出了每种方案的相关方程式以及后续的迭代步骤。对所有交错方案进行定性研究,例如通过它们的收敛行为,以及通过比较交错解决方案和整体解决方案进行定量分析。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号