Generalize surface normal calc #539
Conversation
Conflicts: src/schemes/fluid/weakly_compressible_sph/system.jl
…e_surface_normal_calc
Conflicts: src/schemes/fluid/entropically_damped_sph/rhs.jl src/schemes/fluid/entropically_damped_sph/system.jl src/schemes/fluid/weakly_compressible_sph/system.jl test/schemes/fluid/fluid.jl
| - `correction`: Correction method of the adjacent fluid system (see [Corrections](@ref corrections)). | ||
| - `viscosity`: Slip (default) or no-slip condition. See description below for further | ||
| information. | ||
| - `reference_particle_spacing`: The reference particle spacing used for weighting values at the boundary. |
There was a problem hiding this comment.
- Please add this to the signature above.
- This is completely unclear to me. Isn't this only used for surface tension? Then why is the surface tension not mentioned here?
There was a problem hiding this comment.
Its a general value that has nothing todo with surface tension beside being only used there at this time.
| surface_tension :: SRFT | ||
| surface_normal_method :: SRFN |
There was a problem hiding this comment.
Wouldn't it make sense to store the normal method inside the surface tension to avoid cluttering the struct?
There was a problem hiding this comment.
While the surface normals are only used for surface tension, at this time, they are never actually called from the actual surface tension model. The surface normal calculation steps are always called from the systems.
There was a problem hiding this comment.
But it is just a struct defining the method, right? There is no per-particle data stored that depends on the system, right? Then I don't see a reason why we couldn't store the normal method in the surface tension. Seems more intuitive to me:
surface_tension = SurfaceTensionWhatever(surface_normal_method=ColorWhateverNormals())There was a problem hiding this comment.
The surface normal calculations can be used in other things than the surface tension.
- Tracking of the interface
- Heatflux between phases
- Free surface boundary conditions
Moving it to surface tension just makes it seem like it somehow depends on anything in surface tension but it actually doesn't. Also you need to actually declare surface normal methods for the solids/boundaries as well but they don't have a surface tension. Declaring normal methods for solids/boundaries is part of #666 since they are needed for the wetting model. These are than also different for elastic solids and rigid solids.
| number_density :: Int64 | ||
| color :: Int64 |
There was a problem hiding this comment.
Why do these need to be here? They are only used when using surface tension, no? Then they would be better off in the cache.
There was a problem hiding this comment.
They are currently only needed when a surface normal method is used.
There was a problem hiding this comment.
The problem is though that these values are needed for all systems but not all systems have a surface normal or surface tension model activated.
…vchb/TrixiParticles.jlOpen into generalize_surface_normal_calc
Depends on #599
extracted from #584