Change computational grid about multiphysics

[AnaMarkH]

@joaofl: You might try the energy2d version in the current pull request. This update should be able to properly deal with modified grid resolutions as you intended right at the beginning.

[charxie]

Sorry, I didn't see these issues until today. For conduction simulations and probably radiation simulations, this should be fine. For convection simulations, this is more a problem due to the butterfly effect in transient simulations. A change of grid size will create impacts far greater than a butterfly as you can imagine. So I would say steady problems should be fine, but I am not sure about transient problems. I am curious how other CFD software deals with this.

[joaofl]

@charxie: Was there any development regarding this? It is a great simulator for dummies and for learning proposes, since going to more complex CFD can cost a lot (Open FOAM for example).