COMSOL 3.5中的crosswind diffusion的一点改进

TBE_Legend

Hi, in 3.5 we improved the stabilization (crosswind diffusion) of convective transport problems.

The new stabilization will discard some unphysical solutions. One example is a transient incompressible flow problem (Navier Stokes) with initial condition u=0 with a boundary condition u=1 at t=0. The physical equivalence would be to send a shockwave through an incompressible medium. The incompressible Navier-Stokes equations does not permit a solution to that problem.

Niklas Rom
COMSOL
[email=niklas%40comsol.com]niklas@comsol.com[/email]

--- In [email=COMSOL_Users%40yahoogroups.com]COMSOL_Users@yahoogroups.com[/email], "pierre Infelta"
<pierre.infelta@...> wrote:
>
> I had problem with 3.5 as well. It looks like there is going to be a
> fix up for this, version 3.5 a.
> The problem seems to be that 3.5 uses a linear solver when it should
> use a non linear solver.
>
> --- In [email=COMSOL_Users%40yahoogroups.com]COMSOL_Users@yahoogroups.com[/email], "Omar Al-Abbasi" <omabbasi@>
> wrote:
> >
> > Hi everyone,
> >
> > I am facing a problem with COMSOL 3.5 in solving transient case of
> > Navier-Stokes equations, which I had no problem in 3.4. COMSOL 3.5
> is not
> > able to find the starting time step to march the solution.
> > Appreciating any suggestions.
> >
> >
> > Regards,
> >
> > Omar
> >
>

zhoulon

经常见你发出一些最新的版本消息什么的，很是热情，感谢一下。

TBE_Legend

原帖由 zhoulon 于 2008-12-17 19:56 发表
经常见你发出一些最新的版本消息什么的，很是热情，感谢一下。

这倒也不是，我经常关注这个人的email，因为他是COMSOL公司的，他给出的信息一般人是不知道的，关键还可信。