找回密码
 注册
Simdroid-非首页
查看: 212|回复: 6

[04.流体] 请问有对level set或者phase field比较熟的网友吗

[复制链接]
发表于 2010-1-12 15:52:57 | 显示全部楼层 |阅读模式 来自 天津
本人在做气泡的鼓泡和上升方面模型,可是总是遇到要么计算不收敛,要不得不到正确的仿真结果,有哪位网友对level set或者phase field比较熟的,可以交流下吗,我的qq435725415,谢谢!
发表于 2010-3-15 17:52:15 | 显示全部楼层 来自 北京海淀
Simdroid开发平台
界面厚度,网格尺寸对收敛性影响很大
回复 不支持

使用道具 举报

发表于 2010-4-28 13:58:21 | 显示全部楼层 来自 甘肃兰州
能不能说得具体点?
我也遇到了同样的问题,计算液滴电润湿中的演变过程出现了不收敛。
到底该怎么解决呢?着急ing……
回复 不支持

使用道具 举报

发表于 2010-4-28 14:18:52 | 显示全部楼层 来自 甘肃兰州
我使用的是level set方法。
看到帮助文档里有这个,不知道能否解决收敛问题。
Convergence Problems
If you run into convergence problems, check the following:
•  Is γ approximately as large as the maximum speed of the flow?

•  Is ε constant and equal in all subdomains passed by the interface?  

•  Is ε at least as large as half the maximum element size in the region passed by the interface?

•  Is the cell Reynolds number smaller than 2? If not you should use artificial diffusion. Open the Artificial Diffusion dialog box by clicking the corresponding button on the Sources/Sinks page of the Subdomain Settings dialog box. Click the Navier-Stokes Equations tab. Apply anisotropic diffusion for the Navier-Stokes equations by selecting the Streamline diffusion check box and then selecting Anisotropic diffusion from the list next to the check box. If the convergence problems persist, and increasing the resolution is not possible, apply isotropic diffusion as well.

•  If the effect of surface tension is small, set the surface tension coefficient σ to zero.

•  Make sure you have initialized the level set function as described in the section “Solution Procedure” on page 32.

•  Verify that there is no inflow through any boundaries that has been specified as outlet boundaries. For boundaries that are both inlets and outlets, specify them as inlets.

•  If the pressure is not specified on any boundary, make sure that there is either a condition on the pressure in one point or a condition on the integral of the pressure over the entire fluid domain.

•  If you have a velocity boundary condition, increase it gradually from zero to its actual value during a short initial time interval using the built-in Heaviside smoothing function flc1hs(t-deltat,deltat).
回复 不支持

使用道具 举报

发表于 2010-4-29 21:05:41 | 显示全部楼层 来自 陕西西安
我用两相流做的流体怎么质量不守恒呢,这又是什么原因呢,跟楼主的情况差不多啊,郁闷中。
回复 不支持

使用道具 举报

发表于 2010-4-30 11:07:11 | 显示全部楼层 来自 上海长宁区
试一下细化网格,或者在模式属性(物理量>属性)中选择守恒的方程形式。
回复 不支持

使用道具 举报

发表于 2010-5-3 08:21:15 | 显示全部楼层 来自 美国
phase field的方程不收敛很正常,参数设置非常tricky
回复 不支持

使用道具 举报

您需要登录后才可以回帖 登录 | 注册

本版积分规则

Archiver|小黑屋|联系我们|仿真互动网 ( 京ICP备15048925号-7 )

GMT+8, 2024-3-29 18:53 , Processed in 0.042583 second(s), 14 queries , Gzip On, MemCache On.

Powered by Discuz! X3.5 Licensed

© 2001-2024 Discuz! Team.

快速回复 返回顶部 返回列表