Add compressibility to heat transfer #1051
Conversation
| @@ -454,8 +454,10 @@ HeatTransfer<dim>::assemble_local_system_matrix( | |||
| const DoFHandler<dim> *dof_handler_fluid = | |||
| multiphysics->get_dof_handler(PhysicsID::fluid_dynamics); | |||
|
|
|||
| typename DoFHandler<dim>::active_cell_iterator velocity_cell( | |||
| &(*triangulation), cell->level(), cell->index(), dof_handler_fluid); | |||
| typename DoFHandler<dim>::active_cell_iterator fd_cell(&(*triangulation), | |||
There was a problem hiding this comment.
Is fd meant to mean "fluid dynamics"? Maybe I would use a more meaningful name here.
There was a problem hiding this comment.
I don't mind the fd here to be honest. All the fluid dynamics functions have fd, heat transfers have ht, etc. Right now this is a standard nomenclature in the code, so I would not change that unless we change it everywhere and that owuld be another PR in itself.
| * components and pressure values. | ||
| */ | ||
| template <typename VectorType> | ||
| void | ||
| reinit_pressure(const typename DoFHandler<dim>::active_cell_iterator &cell, | ||
| const VectorType ¤t_solution) | ||
| { | ||
| this->fe_values_fd.reinit(cell); | ||
| this->fe_values_fd[pressure].get_function_values(current_solution, | ||
| pressure_values); | ||
| } |
There was a problem hiding this comment.
Wait why is this a second FE_values reinit? This is very expensive as a call. It would be better to gather pressure when we gather the fluid dynamics instead of calling reinit when we need both...
source/solvers/heat_transfer.cc
Outdated
| scratch_data.reinit_pressure(fd_cell, | ||
| *multiphysics->get_block_solution( |
There was a problem hiding this comment.
It's better to move all the reinits to a single reinit which would be reinit_fluid_dynamics,. Otherwise the call to fe_values reinit will get prohibitively expensive..
There was a problem hiding this comment.
You're right, I'll do so tomorrow!
I'll also check if it is more costly to add an if inside to get the pressure values or if the get_function_values() is not to bad. If it is not too bad (which I don't think is the case), I'll leave it without the if is that okay ?
AmishgaAlphonius
added
Reviewed and waiting for changes
and removed
Ready for review
labels
Co-authored-by: Bruno Blais <blais.bruno@gmail.com>
|
Will merge after CI @AmishgaAlphonius |
blaisb
added
Reviewed and ready to merge
and removed
Reviewed and waiting for changes
labels
Description of the problem
Description of the solution
How Has This Been Tested?
applications_tests/lethe-fluid/heat_transfer_weakly_compressible_flow.prm(.output)applications_tests/lethe-fluid/heat_transfer_vof_weakly_compressible_flow.prm(.output)Future changes
fieldsshould be checked if required in all scratch data that calculates physical properties.Comments