ffc team mailing list archive

[Andy Ray Terrel <aterrel@xxxxxxxxxxxx>]

Wouldn't that ignore the coupling in solving Stokes to begin with?

You want to be able to add an integral operator perhaps something like
the mass matrix (m) with a constant function m(1, p). Added to the
problem.

-- Andy



On Thu, May 28, 2009 at 3:57 PM, Anders Logg <logg@xxxxxxxxx> wrote:
> On Wed, May 27, 2009 at 02:49:36PM -0500, Brandon Chabaud wrote:
>> Hello,
>>
>> I am trying to solve a Stokes flow type problem under the constraint that the
>> integral over the domain of the pressure is zero.  None of the pde demos have
>> an example of this.  How do I impose this condition in fenics?  Thanks.
>
> Have you tried using the normalize() function in DOLFIN? It subtracts
> the average of a vector from a vector so it has zero mean value. You
> can use that on the right-hand side for the pressure to ensure that
> the system has a solution and then again on the solution to obtain a
> solution with zero mean value if necessary.
>
> --
> Anders
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.9 (GNU/Linux)
>
> iEYEARECAAYFAkoe+sIACgkQTuwUCDsYZdGsAgCfXuKWIibVGlWumi6JZ8KMNGvW
> q44An3uPwKuLe9Ooob1v1qdxJDAd+Rub
> =FRiq
> -----END PGP SIGNATURE-----
>
> _______________________________________________
> FFC-dev mailing list
> FFC-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/ffc-dev
>
>