Thank you. It is running now, but I have a question. Pf and Pk are
functions of location (x) and time (t). However, Pd and Pu are only
functions of time (they represent pressure at downstream and upstream
tanks). When I solve this problem, I will get a vector for Pd and Pu (at
last timestep) which varies with location. For example, for Pu,

print Pu

[ 7171802.05029459  7171719.44929532  7171632.68237858  7171538.95796536
  7171435.46595385  7171319.37283423  7171187.81897787  7171037.91833804
  7170866.76056175  7170671.41527712]

Which shows the Pu values at 10 different mesh blocks. Is it related to how
I define my cell variables?

I appreciate your helps.

Bests,

On Fri, May 6, 2016 at 5:21 PM, Daniel Wheeler <[email protected]>
wrote:

> The following runs for me with a few changes.
>
> On Fri, May 6, 2016 at 1:25 PM, Mohammad Kazemi <[email protected]>
> wrote:
> >
> > Pk.constrain([Pd], mesh.facesLeft)
> > Pk.constrain([Pu], mesh.facesRight)
>
> Pk.constrain(Pd.faceValue, mesh.facesLeft)
> Pk.constrain(Pu.faceValue, mesh.facesRight)
>
> > Pf.constrain([Pd], mesh.facesLeft)
> > Pf.constrain([Pu], mesh.facesRight)
>
> Pf.constrain(Pd.faceValue, mesh.facesLeft)
> Pf.constrain(Pu.faceValue, mesh.facesRight)
>
> This makes sense since the Pf and Pk are scalars and [Pd] and [Pu] are
> effectively vectors. Also the face value is required since the
> constraint is on the faces.
>
> I hope this helps.
>
> --
> Daniel Wheeler
> _______________________________________________
> fipy mailing list
> [email protected]
> http://www.ctcms.nist.gov/fipy
>   [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
>



-- 
Mohammad Kazemi
West Virginia University
Department of Petroleum and Natural Gas Engineering
_______________________________________________
fipy mailing list
[email protected]
http://www.ctcms.nist.gov/fipy
  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]

Reply via email to