- 积分
- 0
- 注册时间
- 2006-2-16
- 仿真币
-
- 最后登录
- 1970-1-1
|
发表于 2011-9-30 12:45:48
|
显示全部楼层
来自 江苏扬州
zsq-w 发表于 2011-9-30 10:04
很奇怪你这些结论从哪里得知的??
1)关于SPH并行计算受限制,可以查看abaqus6.11帮助文件中abaqus analysis user's manual 中的15.1.1中的limitation中的最后一项:
If a large percentage of all nodes in the model are associated with smoothed particle hydrodynamics, the analysis will not scale well if multiple CPUs are used. All PC3D elements with the same section definition are forced to be in the same parallel domain.
我自己也尝试过一个简单的SPH得油箱晃动模型,采用6cpu并行计算,分成六个计算域,其中一个域的节点数量要远远多过其他五个;
2)关于SPH计算需要设置矩形区域的说法,我也是从帮助中下述语句得到的:
A rectangular region is computed at the beginning of the analysis as the bounding box within which the particles will be tracked. This fixed rectangular box is 10% larger than the overall dimensions of the whole model, and it is centered at the geometric center of the model. As the analysis progresses, if a particle is outside this box, it behaves like a free-flying point mass and does not contribute to smoothed particle hydrodynamic calculations. If the particle reenters the box at a later stage, it is once again included in the calculations.
该语句位于abaqus6.11帮助文件中abaqus analysis user's manual 中的15.1.1中的Element\Smoothed particle hydrodynamic domain.
关于这方面,我本人没有做过尝试,只是从字面得到上述观点。
|
|