首页> 外文会议>Conference on Modeling of Casting, Welding and Advanced Solidification Processes >Numerical modeling of macrosegregation during solidification processes using multiple, adaptive finite element grids
【24h】

Numerical modeling of macrosegregation during solidification processes using multiple, adaptive finite element grids

机译:使用多种自适应有限元栅格凝固过程中MacroseGregation的数值模拟

获取原文

摘要

A numerical method based on finite element approximations for the simulation of macrosegregation during solidification processes is presented. In order to describe carefully the boundary layer close to the dendrite tips, where solutal variations are initiated, two different discretizations in space are introduced. A fixed, coarse and global finite element mesh is used at the scale of the casting, whereas a fine mesh evolving with the mushy zone is created by subdivision of the coarse elements which belong to the critical region. For the physical model, average conservation laws for energy, mass, momentum and solute are used. The coupled system of partial differential equations is solved using the two finite element discretizations as follows. At each time step, the resolution of the heat equation at the macroscopic level determines the position of the mushy zone and thus the elements, which need refinement. Fluid flow is then computed on the coarse grid, followed by the resolution within the refined region. Finally, the solute conservation equations are solved by applying a substructuring iterative method using the two discretizations and the coarse and fine approximations of the velocity field. The interpolation and restriction on the boundary of the mushy zone, which allow the passage from the coarse to the fine grid and vice versa, respectively, are done in a conservative way. This adaptive, multiple grid method has been applied to two test cases: fluid flow in a solidifying channel and solute transport in a rectangular cavity. It is shown that a better accuracy can be obtained while limiting the computation time.
机译:提出了一种基于有限元近似的数值方法,用于在凝固过程中模拟MacroseGregation的模拟。为了小心地描述靠近枝形尖端的边界层,其中引发了索特式变化,引入了两个不同的离散化。在铸造的规模中使用固定,粗糙和全球有限元网,而使用属于关键区域的粗元件的细分来产生与糊状区的细网格而发展。对于物理模型,使用能量,质量,动量和溶质的平均保守规律。使用如下的两个有限元分离子来解决部分微分方程的耦合系统。在每个时间步骤中,宏观级别的热方程的分辨率决定了糊状区的位置,从而决定了需要细化的元件。然后在粗网格上计算流体流动,然后在精制区域内的分辨率计算。最后,通过使用两种离散化和速度场的粗糙和精细近似来应用所述子结构迭代方法来解决溶质保护方程。允许从粗栅到细网格的糊状区边界的插值和限制分别以保守方式完成。这种自适应,多电网方法已应用于两个测试用例:在凝固通道中的流体流动并在矩形腔内溶质运输。结果表明,可以在限制计算时间的同时获得更好的精度。

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