[Dune] Dune Digest, Vol 92, Issue 18

[davide baroli]

can be useful to use some marker or flag ?


2013/2/20 <dune-request at dune-project.org>

> Send Dune mailing list submissions to
>         dune at dune-project.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         http://lists.dune-project.org/mailman/listinfo/dune
> or, via email, send a message with subject or body 'help' to
>         dune-request at dune-project.org
>
> You can reach the person managing the list at
>         dune-owner at dune-project.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Dune digest..."
>
> Today's Topics:
>
>    1. Topological changes in the grid (Sacconi, Andrea)
>    2. Re: Topological changes in the grid (Oliver Sander)
>
>
> ---------- Messaggio inoltrato ----------
> From: "Sacconi, Andrea" <a.sacconi11 at imperial.ac.uk>
> To: "dune at dune-project.org" <dune at dune-project.org>
> Cc:
> Date: Tue, 19 Feb 2013 18:01:03 +0000
> Subject: [Dune] Topological changes in the grid
>  Hi everyone,
>
> I'd like to ask you a question about the grid.
>
> Suppose you are solving a problem on a given initial grid, and after a
> while some topological changes take place (coalescence, breakups, ...). For
> example, from one connected domain you get two, smaller disconnected ones.
> Suppose in addition that you have a software that removes tangles,
> returning the actual disconnected domain without self-intersections.
>
> I know that a grid can be neither created nor modified; so basically you
> can only have a reference to a grid, initialised by dereferencing a pointer
> to the the same grid type.
> So, if all the objects (function spaces, grid functions, ...) I have in my
> algorithm are defined over the initial one, is there a tricky way to
> continue my computations on the new disconnected domain?
>
> I mean, the new tangle-free grid is returned inside a time loop.
> I have initialised a new pointer to the same grid type by using a grid
> factory (since the new disconnected domain is returned through its
> connectivity, I insert vertices and elements by hand). However, I can't
> assign it to the reference to the old grid ... Please tell me that there is
> a (dirty) workaround to continue the computations inside the time loop
> ........... otherwise I wouldn't know how to proceed without breaking the
> loop ........
>
> Cheers,
> Andrea
>
> __________________________________________________________
>
> Andrea Sacconi
> PhD student, Applied Mathematics
> AMMP Section, Department of Mathematics, Imperial College London,
> London SW7 2AZ, UK
> a.sacconi11 at imperial.ac.uk
>
>
> ---------- Messaggio inoltrato ----------
> From: Oliver Sander <sander at igpm.rwth-aachen.de>
> To: dune at dune-project.org
> Cc:
> Date: Tue, 19 Feb 2013 19:32:25 +0100
> Subject: Re: [Dune] Topological changes in the grid
>  Hi Andrea,
> there is no way to change the connectivity of any Dune grid implementation
> that
> I know.  Not even a dirty trick.  You need to kill your grid object within
> the time loop
> and create a new one using the grid factory.  That by itself is reasonably
> straightforward.
> The problem of course is how to transfer your data from the old to the new
> grid.
> But the solution to that is problem-specific.
> good luck,
> Oliver
>
> Am 19.02.2013 19:01, schrieb Sacconi, Andrea:
>
>  Hi everyone,
>
> I'd like to ask you a question about the grid.
>
> Suppose you are solving a problem on a given initial grid, and after a
> while some topological changes take place (coalescence, breakups, ...). For
> example, from one connected domain you get two, smaller disconnected ones.
> Suppose in addition that you have a software that removes tangles,
> returning the actual disconnected domain without self-intersections.
>
> I know that a grid can be neither created nor modified; so basically you
> can only have a reference to a grid, initialised by dereferencing a pointer
> to the the same grid type.
> So, if all the objects (function spaces, grid functions, ...) I have in my
> algorithm are defined over the initial one, is there a tricky way to
> continue my computations on the new disconnected domain?
>
> I mean, the new tangle-free grid is returned inside a time loop.
> I have initialised a new pointer to the same grid type by using a grid
> factory (since the new disconnected domain is returned through its
> connectivity, I insert vertices and elements by hand). However, I can't
> assign it to the reference to the old grid ... Please tell me that there is
> a (dirty) workaround to continue the computations inside the time loop
> ........... otherwise I wouldn't know how to proceed without breaking the
> loop ........
>
> Cheers,
> Andrea
>
> __________________________________________________________
>
> Andrea Sacconi
> PhD student, Applied Mathematics
> AMMP Section, Department of Mathematics, Imperial College London,
> London SW7 2AZ, UK
> a.sacconi11 at imperial.ac.uk
>
>
> _______________________________________________
> Dune mailing listDune at dune-project.orghttp://lists.dune-project.org/mailman/listinfo/dune
>
>
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.dune-project.org/pipermail/dune/attachments/20130220/f4799de2/attachment.htm>