Hi Olivier,
I fixed up the script a bit
https://gist.github.com/wd15/030fd25b9f19ef0ec4ec
It at least runs without errors now. The main change was the replacement of
the tuple of CellVariables passed to the diffusion and convetion term with
single CellVariables, but with the the shape of rank 1 tensors (vectors).
Passing tuples of CellVarialbes in FiPy to the terms doesn't really work
correctly. In the diffusion term it invokes a biharmonic (fourth order
diffusion term) and in the convection term it only works if each of the
entries is single valued (not an array or variable of any kind).
The other thing is that I changed the CellVariables to FaceVariables since
it makes sense to evaluate the coefficients directly at the faces rather
than interpolating from the cell centers. I noticed that the velocity
depends on 1 / R so I had to finesse this for R=0 faces, you may want to
think about that a bit.
I have no idea whether the results are correct, but at least you can do
some analysis now.
Cheers,
Daniel
On Wed, Sep 24, 2014 at 2:04 AM, olivier atteia <[email protected]>
wrote:
> Dear all,
> i set a problem with a cylindrical2D grid, it is composed, in the vertical
> direction of two layers, in the lower one there is a velocity from the
> center well to the outside,in the upper, the velocity is negligible. A
> concentration is injected in a well located at the center.
>
as it is in a porou medium, there is dispersion, which is proportional to
> velociaty. This dispersion has two different compounds in the radial and in
> the vertical direction. For this, i use two cellVariable (D2 and D2) and i
> set the coeff in the diffusion term as (D1.faceValue,D2.faceValue)
> The BC at the outlet for the concentration is a null gradient
> I tried two BC for the injection well : 1, where=mesh.faceLeft and
> where=[0] it does not change
>
> my problem :
> when i run the solve command, iget this :
> ValueError: The truth value of an array with more than one element is
> ambiguous. Use a.any() or a.all()
>
> the file is enclosed
> thanks
> olivier
>
> _______________________________________________
> fipy mailing list
> [email protected]
> http://www.ctcms.nist.gov/fipy
> [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
>
>
--
Daniel Wheeler
_______________________________________________
fipy mailing list
[email protected]
http://www.ctcms.nist.gov/fipy
[ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
