Fix IC in open boundaries and add simple_advection_2d.jl#852
Closed
LasNikas wants to merge 6 commits intotrixi-framework:mainfrom
Closed
Fix IC in open boundaries and add simple_advection_2d.jl#852LasNikas wants to merge 6 commits intotrixi-framework:mainfrom
simple_advection_2d.jl#852LasNikas wants to merge 6 commits intotrixi-framework:mainfrom
Conversation
added 2 commits
July 1, 2025 15:30
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #852 +/- ##
==========================================
+ Coverage 70.47% 70.50% +0.03%
==========================================
Files 106 106
Lines 6893 6897 +4
==========================================
+ Hits 4858 4863 +5
+ Misses 2035 2034 -1
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Collaborator
Author
|
/run-gpu-tests |
svchb
requested changes
Jul 2, 2025
Comment on lines
+3
to
+144
|
|
||
| # ========================================================================================== | ||
| # ==== Resolution | ||
| particle_spacing = 0.01 | ||
|
|
||
| # Make sure that the kernel support of fluid particles at a boundary is always fully sampled | ||
| boundary_layers = 4 | ||
|
|
||
| # Make sure that the kernel support of fluid particles at an open boundary is always | ||
| # fully sampled. | ||
| # Note: Due to the dynamics at the inlets and outlets of open boundaries, | ||
| # it is recommended to use `open_boundary_layers > boundary_layers` | ||
| open_boundary_layers = 6 | ||
|
|
||
| # ========================================================================================== | ||
| # ==== Experiment Setup | ||
| tspan = (0.0, 2.0) | ||
|
|
||
| # Boundary geometry and initial fluid particle positions | ||
| domain_size = (1.0, 0.4) | ||
|
|
||
| flow_direction = [1.0, 0.0] | ||
| reynolds_number = 100 | ||
| const prescribed_velocity = 1.0 | ||
|
|
||
| boundary_size = (domain_size[1] + 2 * particle_spacing * open_boundary_layers, | ||
| domain_size[2]) | ||
|
|
||
| fluid_density = 1000.0 | ||
|
|
||
| sound_speed = 10 * prescribed_velocity | ||
|
|
||
| state_equation = nothing | ||
|
|
||
| pipe = RectangularTank(particle_spacing, domain_size, boundary_size, fluid_density, | ||
| velocity=[prescribed_velocity, 0.0], | ||
| n_layers=boundary_layers, faces=(false, false, true, true)) | ||
|
|
||
| # Shift pipe walls in negative x-direction for the inflow | ||
| pipe.boundary.coordinates[1, :] .-= particle_spacing * open_boundary_layers | ||
|
|
||
| NDIMS = ndims(pipe.fluid) | ||
|
|
||
| n_buffer_particles = pipe.n_particles_per_dimension[2]^(NDIMS - 1) | ||
|
|
||
| # ========================================================================================== | ||
| # ==== Fluid | ||
| wcsph = true | ||
|
|
||
| smoothing_length = 1.2 * particle_spacing | ||
| smoothing_kernel = SchoenbergQuinticSplineKernel{NDIMS}() | ||
|
|
||
| fluid_density_calculator = ContinuityDensity() | ||
|
|
||
| kinematic_viscosity = prescribed_velocity * domain_size[2] / reynolds_number | ||
|
|
||
| viscosity = ViscosityAdami(nu=kinematic_viscosity) | ||
|
|
||
| # Alternatively the WCSPH scheme can be used | ||
| if wcsph | ||
| state_equation = StateEquationCole(; sound_speed, reference_density=fluid_density, | ||
| exponent=1) | ||
|
|
||
| density_diffusion = DensityDiffusionMolteniColagrossi(delta=0.1) | ||
|
|
||
| fluid_system = WeaklyCompressibleSPHSystem(pipe.fluid, fluid_density_calculator, | ||
| state_equation, smoothing_kernel, | ||
| density_diffusion=density_diffusion, | ||
| smoothing_length, viscosity=viscosity, | ||
| buffer_size=n_buffer_particles) | ||
| else | ||
| fluid_system = EntropicallyDampedSPHSystem(pipe.fluid, smoothing_kernel, | ||
| smoothing_length, | ||
| sound_speed, viscosity=viscosity, | ||
| density_calculator=fluid_density_calculator, | ||
| buffer_size=n_buffer_particles) | ||
| end | ||
|
|
||
| # ========================================================================================== | ||
| # ==== Open Boundary | ||
|
|
||
| velocity_function2d(pos, t) = SVector(prescribed_velocity, 0.0) | ||
|
|
||
| open_boundary_model = BoundaryModelTafuni() | ||
|
|
||
| boundary_type_in = InFlow() | ||
| plane_in = ([0.0, 0.0], [0.0, domain_size[2]]) | ||
| inflow = BoundaryZone(; plane=plane_in, plane_normal=flow_direction, open_boundary_layers, | ||
| density=fluid_density, particle_spacing, | ||
| boundary_type=boundary_type_in) | ||
|
|
||
| open_boundary_in = OpenBoundarySPHSystem(inflow; fluid_system, | ||
| boundary_model=open_boundary_model, | ||
| buffer_size=n_buffer_particles, | ||
| reference_velocity=velocity_function2d) | ||
|
|
||
| boundary_type_out = OutFlow() | ||
| plane_out = ([domain_size[1], 0.0], [domain_size[1], domain_size[2]]) | ||
| outflow = BoundaryZone(; plane=plane_out, plane_normal=(-flow_direction), | ||
| open_boundary_layers, density=fluid_density, particle_spacing, | ||
| boundary_type=boundary_type_out) | ||
| open_boundary_out = OpenBoundarySPHSystem(outflow; fluid_system, | ||
| boundary_model=open_boundary_model, | ||
| buffer_size=n_buffer_particles, | ||
| reference_velocity=velocity_function2d) | ||
|
|
||
| # ========================================================================================== | ||
| # ==== Boundary | ||
| boundary_model = BoundaryModelDummyParticles(pipe.boundary.density, pipe.boundary.mass, | ||
| AdamiPressureExtrapolation(), | ||
| state_equation=state_equation, | ||
| smoothing_kernel, smoothing_length) | ||
|
|
||
| boundary_system = BoundarySPHSystem(pipe.boundary, boundary_model) | ||
|
|
||
| # ========================================================================================== | ||
| # ==== Simulation | ||
| min_corner = minimum(pipe.boundary.coordinates .- particle_spacing, dims=2) | ||
| max_corner = maximum(pipe.boundary.coordinates .+ particle_spacing, dims=2) | ||
|
|
||
| nhs = GridNeighborhoodSearch{NDIMS}(; cell_list=FullGridCellList(; min_corner, max_corner), | ||
| update_strategy=ParallelUpdate()) | ||
|
|
||
| semi = Semidiscretization(fluid_system, open_boundary_in, open_boundary_out, | ||
| boundary_system, neighborhood_search=nhs, | ||
| parallelization_backend=PolyesterBackend()) | ||
|
|
||
| ode = semidiscretize(semi, tspan) | ||
|
|
||
| info_callback = InfoCallback(interval=100) | ||
| saving_callback = SolutionSavingCallback(dt=0.02, prefix="") | ||
|
|
||
| particle_shifting = nothing | ||
|
|
||
| extra_callback = nothing | ||
|
|
||
| callbacks = CallbackSet(info_callback, saving_callback, UpdateCallback(), | ||
| particle_shifting, extra_callback) | ||
|
|
||
| sol = solve(ode, RDPK3SpFSAL35(), | ||
| abstol=1e-5, # Default abstol is 1e-6 (may need to be tuned to prevent boundary penetration) | ||
| reltol=1e-3, # Default reltol is 1e-3 (may need to be tuned to prevent boundary penetration) | ||
| dtmax=1e-2, # Limit stepsize to prevent crashing | ||
| save_everystep=false, callback=callbacks); |
Collaborator
There was a problem hiding this comment.
Isn't this just another pipe? Can't you base it on pipe_flow_2d.jl?
Collaborator
Author
There was a problem hiding this comment.
I want a clean, simple example for the open boundaries. I've run into issues with trixi_include so many times and ended up with unnecessarily frustrating bugs: you simply forget to enable or disable something. For complexity reasons, it's also better for beginners to have a simple example.
Collaborator
There was a problem hiding this comment.
I think this is more a problem with the structure of our setup files. But anyway I think it would be better to further simplify pipe_flow_2d and base your setup on that.
LasNikas
force-pushed
the
simple-advection
branch
from
ec13ba7 to
7f74e25
Compare
19 tasks
svchb
requested changes
Jul 22, 2025
Comment on lines
+1
to
+2
| # This is a simple advection example with open boundaries. The setup is similar to | ||
| # `periodic_channel_2d.jl`, but uses open boundaries instead of periodic ones. |
LasNikas
pushed a commit
to LasNikas/TrixiParticles.jl
that referenced
this pull request
Jul 29, 2025
Collaborator
Author
|
superseded by #866 |
efaulhaber
added a commit
that referenced
this pull request
Sep 8, 2025
* swap extrapolation * swap density pressure * adapt docs * rm example * fix missing nhs update * implement different mirror methods * add docs * fix tests * fix typos * add docs * add tests * apply formatter * add `average_velocity!` for `BoundaryModelLastiwka` * fix * add interpolation functions * restructure again * fix gpu bugs * remove unnecessary argument * fix gpu again * fix gpu * implement suggestions * fix * add comments * implement suggestions * first prototype: NOT VALIDATED YET * fix avg vel in LastiwkaModel * fix avg in mirroring * fix tests * adapt example file * fix allocations * fix gpu * first GPU working prototype * improve API * fix tests * rm export * fix * fix type stability * implement suggestions * rm ref * format * Rename 'tlsph' to 'place_on_shell' (#814) * rename * format * forgot some * format * naming * forgot some more * fix test * incorporate review comments * format --------- Co-authored-by: Niklas Neher <73897120+LasNikas@users.noreply.github.com> * fix merge bugs * supersede #852 * supersede #850 * fix * rename boundary models * fix? * fix tests * Rename 'tlsph' to 'place_on_shell' (#814) * rename * format * forgot some * format * naming * forgot some more * fix test * incorporate review comments * format --------- Co-authored-by: Niklas Neher <73897120+LasNikas@users.noreply.github.com> * fix characteristics * revise pipe flow example * fix include bug * Rename 'tlsph' to 'place_on_shell' (#814) * rename * format * forgot some * format * naming * forgot some more * fix test * incorporate review comments * format --------- Co-authored-by: Niklas Neher <73897120+LasNikas@users.noreply.github.com> * Combine PST and TVF into a unified framework (#884) * Combine PST and TVF into a unified framework * Require update callback for PST * Fix WCSPH * Update PST only in callback * Fix EDAC * Update docs * Fix alle example files * Fix tests * Fix periodic channel * Fix docs * Update news * Fix tests * Fix example file * add doc strings * implement suggestions * remove duplicated system * rm unnecessary variable * apply formatter * set pressure for WCSPH * setter for open boundary * rename reference values * fix gpu tests * fix sign * implement suggestions * fix unit tests * revise doc strings * add NEWS entry * update NEWS --------- Co-authored-by: LasNikas <niklas.nehe@web.de> Co-authored-by: Sven Berger <berger.sven@gmail.com> Co-authored-by: Erik Faulhaber <44124897+efaulhaber@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The additional example
simple_advection_2d.jlis to quickly test (new) open boundaries and also to benchmark potential new open boundary implementations.