Re: [ff3d-users] Setting boundary conditions!!

[Arun Viswanathan]

Hi Stephane,

I'll try to explain my query a little clearer this
time. Among the 12 cylinders that I have, if we assume
that 1 (or more) cylinders start leaking, how do I
assign this source of leakage as a condition to solve
my PDE? From your reply, I gather that I can use the
Robin condition  (I browsed on this and found out that
it is used to solve elliptic PDEs). Will the Robin
condition be suitable for me to define the source of
leakage?

The condition that you suggested was:
u+dnu(u) = g on <0,1,0>;
Here, is 'g' supposed to represent the rate of escape
of gas from the leaking cylinder? If not, then what is
it? And how can I specify the rate of gas leakage?

The domain specification you suggested was:
domain O = domain(S, outside(<1,0,0>) and
outside(<0,1,0>));
Here, does '<1,0,0>' represent the box (container) and
'<0,1,0>' represent the leaking cylinder? If I have
more than 1 cylinder leaking, can I just use another
'and' to reference that cylinder?

One last question, is there any way for me to
reference different points on a particluar cylinder
(in order to specify the point of leakage more
accurately)?

Thank you Stephane. Best regards,
Arun.


> Le jeudi 27 Mai 2004 00:51, Arun Viswanathan a
> écrit :
> > Hi Stephane!!
> Hello Arun.
> 
> > 1. I will need to add one more boundary condition
> to
> > my code. Among the 12 cylinders, incase 1 (or
> more) of
> > them leak, I would like to pass this as a boundary
> > condition as well. That is, the leak will simply
> act
> > like a source for the entry of the substance into
> the
> > container, and I would like to define the source
> among
> > the boundary conditions. How can I do this??
> I am not sure that I really understand what you want
> to do. If you just mean 
> imposing different boundary conditions to different
> cylinders, you just have 
> to give them different colors. Then you have to
> define your domain this way:
>       domain O = domain(S, outside(<1,0,0>) and
> outside(<0,1,0>));
> for instance. Then you just impose your boundary
> condition in a standard way, 
> for instance:
>       solve (u) in O by M
>       {
>         pde(u) -div(grad(u))=f;
>               dnu(u) = 0 on <1,0,0>; // homogeneous Neuman
>               u+dnu(u) = g on <0,1,0>; // some Robin condition
>       }
> 
> Is that what you are looking for?
> 
> You  can also use the geometry to define functions:
>       domain O2 = domain(S,inside(<0,0,1>));
>       function KiO2 = one(<0,0,1>);
> This later instruction defines a function whose
> value is 1 inside O2 and 0 
> outside...
> 
> > 2. Also, how can I pass this information
> (regarding
> > the source of leak) to the code, without
> hard-coding
> > it. I plan to determine the rate at which the
> chemical
> > escapes from the leak using another formula. How
> can I
> > do that here, can I code it in C?? Or should it be
> > some other language, like Java or Perl?? I tried
> using
> > 'printf' and 'cout' before, they didn't work!! So
> how
> > can I add other code??
> What do you mean. Do you want to modify ff3d or do
> you want to create your own 
> code? cout is working in ff3d, just tell me what you
> really want to do. If 
> you are looking for adding features, you will have
> to learn C++. It could be 
> quite difficult if you are not aware of it, but some
> people already did that.
> 
> Best regards,
> Stephane.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.4 (GNU/Linux)
> 
>
iD8DBQFAuKMdbZyhKhZwdc0RAsMkAKC3YQfeAgjgPehDYb09YBsz9WNWVgCgo0dS
> jcEvyHvQuhdkNCERfVdlEZ8=
> =MI8f
> -----END PGP SIGNATURE-----
> 
> 
> _______________________________________________
> ff3d-users mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/ff3d-users



        
                
__________________________________
Do you Yahoo!?
Friends.  Fun.  Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/