Hi Lorenzo,

regarding your point (3):


Let us consider the mass balance equation for one of the phases. Then, for 
outflow BCs the flux across the boundary is computed by using the current 
pressure gradients. For a darcy flux this would mean -(K*gradP)*normal is 
evaluated on the boundary. Afterwards, the flux is turned into a mass flux by 
multiplying with the mobility and the density, for which we typically use a 
simple upwind scheme. Thus, these values are obtained from inside the domain, 
which means that this only works well as long as you really have a flux leaving 
the domain. If the sign flips (essentially an inflow boundary), mobility and 
density are not defined.


With respect to your question what the difference is: you could achieve the 
same thing manually by implementing the above mentioned stuff into the 
neumann() function:

But note that especially for cell-centered schemes the outflow BCs are rather 
tricky to realize and personally I would try to avoid them and restrict myself 
to Neumann and/or Dirichlet BCs where you actually know what exactly happens.


Best wishes,
Dennis



________________________________
Von: Dumux <[email protected]> im Auftrag von lc 
<[email protected]>
Gesendet: Montag, 3. Dezember 2018 12:37:19
An: [email protected]
Betreff: [DuMuX] on IMPES, grid adaptation and BCs

Hello,

I have the following questions:

1) Is it possible to run IMPES algorithm in sequential (not implicit)
mode without considering capillarity forces (pressure)?

2) Is it possible to do some sort of grid adaptation, for example
locally refine the (unstructrured) mesh where the water saturation front
becomes steeper? If yes, how, is there any example?

3) What actually the Outflow BC does? What is the difference with Neumann?


Thank you very much!

Kind regards,

Lorenzo

_______________________________________________
Dumux mailing list
[email protected]
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
_______________________________________________
Dumux mailing list
[email protected]
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Reply via email to