Hi Wolfgang,

 

I know that the typical treatment of the interface matching conditions is 
along the lines that you noted. However, it is not clear to me how to 
follow these guidelines in the particular case that I am working on, which 
is described in the file that I attached to my previous email. In 
particular, in my case,

 

a)    The matching conditions involve gradients of velocity and pressure, 
and not just the velocity and pressure fields. 

b)     The matching conditions involve time-dependent fields, so probably 
the explicit form of the constraints has to be updated each time step.

 

In order to address point a), I thought that maybe I should define 
additional degrees of freedom that correspond to the components of each one 
of the gradient terms that appear in my matching conditions (e.g. for the 
pressure gradient add DOFs P_i_x, P_i_y and P_i_z). I would then define the 
constraints with respect to those new DOFs, and would also have to assemble 
the corresponding equations (e.g. P_i_x = dP_i/dx).

 

Is this the recommended way to address point a), or is there another way 
which is better?

 

Thank you,

Oded

On Friday, December 2, 2016 at 5:18:22 PM UTC-5, Wolfgang Bangerth wrote:
>
> On 12/02/2016 01:13 PM, Oded Yaakobi wrote: 
> > Now I encounter another aspect of my problem that I don’t know how to 
> tackle – 
> > the matching conditions on the interface between the domains of the 
> Stokes 
> > flow in the droplet and around it. 
> > 
> > 
> > 
> > Attached is a description of the problem in detail. I would be happy to 
> know 
> > if you or anyone else has some advice. 
>
> Typically, if you convert the equations into the weak form, you multiply 
> the 
> equation by a test function and then you integrate (on each of the 
> subdomains) 
> by parts. This yields terms that live on the boundaries of the subdomains, 
> including the interface between the subdomains. This is where you then 
> need to 
> use the interface conditions. 
>
> In your code, this will lead to integrals over internal faces where these 
> faces sit at the interface. 
>
> Best 
>   W. 
>
> -- 
> ------------------------------------------------------------------------ 
> Wolfgang Bangerth          email:                 [email protected] 
> <javascript:> 
>                             www: http://www.math.colostate.edu/~bangerth/ 
>
>

-- 
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see 
https://groups.google.com/d/forum/dealii?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"deal.II User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to