Advection-Diffusion XML file for ElmerGUI

[hazelsct]

I cobbled together the attached advection-diffusion XML file for ElmerGUI based on heatequation.xml . It has a few limitations though which I don't know how to overcome.

• The solute varname is "Concentration", which means it is limited to just one solute.
• The SIF file requires Solver N have "Equation Advection Diffusion Equation Concentration". But I can't figure out a way to get the "Concentration" word in there, so it requires hand-editing.
• Each concentration variable needs to have "Real" prepended, e.g. one needs to enter "Real 0" for the initial condition concentration. But I don't see a good way to indicate this to the user.
• If coupled with Heat Equation, then two Convection keywords show up in the Equation, e.g. "Convection = Computed\n Convection = Computed". I have not looked at what happens if they're not both the same.

It seems like fixing these requires either GUI features I'm not aware of, or some fundamental changes to the XML format and interpretation. Thoughts?

[raback]

Thanx for the effort!

Fixing the GUI just to one solution is probably a wise solution given the many difficulties one faces when trying to have multiple instances of the same solver.

The SIF file requires Solver N have "Equation Advection Diffusion Equation Concentration". But I can't figure out a way to get the "Concentration" word in there, so it requires hand-editing.

Why could not the word "Concentration" be fixed here as well. Or just use "Advection Diffusion Equation". The string is not of importance, it just has to be unique for each solver.

Each concentration variable needs to have "Real" prepended, e.g. one needs to enter "Real 0" for the initial condition concentration. But I don't see a good way to indicate this to the user.

Here the easy remedy would be to add all the instances of "Concentration" in the SOLVER.KEYWORDS file so their type would not need to be specified (I'll do it the coming week).

If coupled with Heat Equation, then two Convection keywords show up in the Equation, e.g. "Convection = Computed\n Convection = Computed". I have not looked at what happens if they're not both the same.

Yes, this seems like a problem. I can now also better understand your difficulties in having two different convection fields.

-Peter

[hazelsct]

Hi Peter and thank you for the reply.

The SIF file requires Solver N have "Equation Advection Diffusion Equation Concentration". But I can't figure out a way to get the "Concentration" word in there, so it requires hand-editing.

Why could not the word "Concentration" be fixed here as well. Or just use "Advection Diffusion Equation". The string is not of importance, it just has to be unique for each solver.

Indeed, this fixes Concentration as the species name. I tried just "Advection Diffusion Equation", because there's no species name. And I also tried "Advection Diffusion Equation Concentration", but it makes for a very wide tab in the dialogs. I suppose one could do "Advection Diffusion Concentration" but that would be awkward.

Each concentration variable needs to have "Real" prepended, e.g. one needs to enter "Real 0" for the initial condition concentration. But I don't see a good way to indicate this to the user.

Here the easy remedy would be to add all the instances of "Concentration" in the SOLVER.KEYWORDS file so their type would not need to be specified (I'll do it the coming week).

Great, thanks.

If coupled with Heat Equation, then two Convection keywords show up in the Equation, e.g. "Convection = Computed\n Convection = Computed". I have not looked at what happens if they're not both the same.

Yes, this seems like a problem. I can now also better understand your difficulties in having two different convection fields.

It's not a problem -- as long as they're both the same. I'll try using the undocumented method you mentioned in your other post to ElmerSolver, and will see if I can change this and heatequation.xml to make them both work.