|
发表于 2012-5-16 19:07:01
|
显示全部楼层
原帖由 uesoft 于 2012-5-16 17:55 发表
你讲的当然有道理,熟悉流体和结构的动力学,的确可以解决一些流固耦合问题。你也承认在流固耦合界面上存在困难,我认为这些困难就是因为我们无法计算分子、原子、电子、光子、质子核、中子核、电子核的相互作用的积 ...
I don't think you understood either what I wrote or what you wrote.
1) I didn't say the difficulty is on "interface". That will be 100% computational issue, nothing related to NS applicability. If dT is taken to be small enough, we can follow the moving interface accurately enough. So no real issue.
2) Turbulence difficulty has nothing to do with interaction of particles. You CAN have turbulence within the framework of NS equation; otherwise there is no point of doing DNS. Therefore modeling down to the scale of nuclei, does not help on turbulence issue.
2.1) In that scale, you are not modeling anymore. That's the most stupid approach of computing all
2.2) There's no such thing as "电子核"; this means you lack basic knowledge of physics, and I guess that's why you think physics is answer to everything. We usually turn to the unknowns to look for answers; for you, "电子核" is an unknown.
3) But still, you may be a really good application engineer or manager, regardless of your comments on technical issues.
Cheers,
FIF |
|