allow Vreman model to be used in 2D - this will help with the outflow bc experiments. Also use a more robust way of checking periodicity when Warches is used.
55 lines of code changed in 3 files:
Added example input file to restart methane8patch.ups using the example:rmcrt with
data-onion. Note the instruction at the top of the file.
RMCRT_Test
- removed antiquated pseudoCfd task()
- added areGridsEqual()
This verifies that the input file grid and restart uda grid are the same
on the appropriate levels.
- initializeWithUda()
updated so it will work with data-onion.
75 lines of code changed in 3 files:
Adjust bulletproofing and boundary condition code to allow one to not need
extra cells in the z-direction when doing axisymmetric calculations. Getting
rid of those unneeded extra cells cuts nodal calculations in half.
10 lines of code changed in 3 files:
Make the way interpolators are handled for axisymmetry to be more correct
and more efficient. Results should only change for simulations using GIMP at
this point. Disable use of "fast" cpdi interpolators, as I'm not sure these
ever worked quite right.
593 lines of code changed in 13 files:
changed MIT license back to what it should be
46 lines of code changed in 2 files:
Added Wait time to burning. This will slow the propagation of burning to that found experimentally.
588 lines of code changed in 2 files:
Turning back on RMCRT tests.
1 lines of code changed in 1 file:
Use pointers to the baseclass instead of the base class since this is
a pure base class with pure virtual functions.
3 lines of code changed in 2 files:
Bug fix to correct failing RT. Initialize
method currently doesn't exist for stencil7
variables, so boundFlux is now being initialized
in a getExtraCell iterator.
A new gold standard can be created from this.
10 lines of code changed in 1 file:
Significantly restructured the inlet generation, and added ability to provide a user specified profile.
0 lines of code changed in 2 files:
Turning off RMCRT test until the dust settles.
1 lines of code changed in 1 file:
This should fix the bug showing up in the coal_channel case re:segfault when trying to get the SolidWallTemperature set.
Also, lowered the number of timesteps to 25 for the coal_channel.ups file.
22 lines of code changed in 1 file:
Remove tabs, other formatting.
0 lines of code changed in 2 files:
Remove extra calls to "interp_type" that are no longer needed.
0 lines of code changed in 1 file:
Remove the option of cpGIMP interpolation, which apparently never really worked
anyway.
6 lines of code changed in 1 file:
Add the d_ in front of interpolator.
1 lines of code changed in 1 file:
Use the interpolator->size() function to assign d_8or27, rather than setting
them manually.
2 lines of code changed in 1 file:
Apply speedups to the axisymmetric version of the CPDI interpolator, taking
advantage of the 2D nature of the problem.
311 lines of code changed in 1 file:
Change conditionals to set NGN=NGP=2 for any # of d_8or27 other than 8.
7 lines of code changed in 9 files:
More updates to the RadProp interface. Basic functionality is now there, but this still needs to be vetted.
124 lines of code changed in 4 files:
to change the name for three additional regression tests
4 lines of code changed in 1 file:
Update thread-count specification in Wasatch, removing duplicated (and obsolete) code.
8 lines of code changed in 1 file:
support Stencil7 in compare_uda
6 lines of code changed in 11 files:
updates due to changes in ThreadPool interface.
6 lines of code changed in 1 file:
Make the surface normal calculation less wrong.
3 lines of code changed in 3 files:
Allow multithreading via ExprLib and SpatialOps to be driven from Wasatch input files.
23 lines of code changed in 1 file:
Removing some unused reduction variables.
0 lines of code changed in 3 files:
Reduction variable, totalKineticEnergy, is now optional and turned off
by default. The stdio report of total kinetic energy is turned on by setting
<ExplicitSolver><print_total_ke/> .... </ExplicitSolver>
in the UPS file or by saving out totalKineticEnergy or kineticEnergy
labels.
146 lines of code changed in 7 files:
add three additional regression test for ARCHES solver
12 lines of code changed in 1 file:
Added support for Stencil7 variables. This is incomplete.
Additional support is needed in TypeDescription.
19 lines of code changed in 1 file:
Few updates to integrate radProps into Arches. This still needs work, including converting mass fraction to mole fractions.
14 lines of code changed in 1 file:
Dav: Getting RadProps compiler directive active.
Jeremy: Getting the Arches interface to RadProps compiling. Still needs work to interface with RadProps corretly.
4 lines of code changed in 3 files:
Added bool flags for recording when hypre data structures that are stored
in the data warehouse have actually been created. This fixes a bug when
the hypre solver is specified but no solve actually occurs.
94 lines of code changed in 3 files:
This change updates the Neumann BC for MPM materials so that the user can implement a non-zero heat flux.
9 lines of code changed in 1 file:
output the environment before each sus run.
0 lines of code changed in 2 files:
Rename MonolithicRHS to use .cc as suffix... not sure how it was even being built as the sub.mk file uses the .cc naming...
262 lines of code changed in 2 files:
Comment out not-used/confusing/re-declared 'valParams' variables.
0 lines of code changed in 2 files:
Pulling a few weeds.
0 lines of code changed in 4 files:
Delete the newly added pScaleFactorLabel_preReloc.
0 lines of code changed in 2 files:
Set NGN correctly for d_8or27==18.
1 lines of code changed in 1 file:
Set _noisy_heat_loss to true... ???Is this correct - not sure what is going on with that line??? Fix spacing/indentation.
64 lines of code changed in 2 files:
add mean turbulent velocity profile bc expression (powerlaw). also add ups example for that.
249 lines of code changed in 2 files:
Fix the p.scalefactor bug by relocating the scalefactor variable.
24 lines of code changed in 7 files:
-output problem spec. for the contact model.
-RT:turned on restart checks for riemannMPM_ML
2 lines of code changed in 3 files:
rename a directory.
2 lines of code changed in 1 file:
added multi-level riemann shock tube problem.
1 lines of code changed in 1 file:
Added further checks on CamClay convergence and subcycling for deformation gradient.
60 lines of code changed in 1 file:
Initialize NGN for 8or27==18.
1 lines of code changed in 1 file:
The old mixture fraction solver spec is now optional. Leaving it out of
the inputfile will not solve the extra transport equation, but does
allocate some variables to keep other models working that were expecting
scalarSP. This should be removed at a later date.
135 lines of code changed in 10 files:
added scheduleExMomInterpolated task to scheduleTimeAdvance.
InterpolateParticlesToGrid(0
- initialize gTemperatureRate
InterpolateToParticlesAndUpdate()
- added sources to the particle temperature calculation
This will not change the RT answers.
2L Riemann problem (RR:1) gives identical answers to 1L Riemann.
39 lines of code changed in 1 file:
added contact model
added artificial dampening coefficient to computeAndIntegrateAcceleration
21 lines of code changed in 1 file:
Create an axisymmetric version of the GIMP interpolator. This currently gives
the same answer as the existing GIMP interpolator at reduced computational cost
by taking advantage of the 2D nature of the problem.
495 lines of code changed in 4 files:
fix minor bug in compare using when using -exact option. This was setting the rel_tolerance to 0 and then then taking its log and passing that to cout precision which caused cout to set its precision to default (6). Now it defaults to 16 when the exact option is used.
2 lines of code changed in 1 file:
Renamed "Node27Interpolator" to "GIMPInterpolator"
641 lines of code changed in 6 files:
Add regression test with a momentum source term. Add regression test for one of the RK3SSP verification cases. Modify the Rk3SSP verification cases to run on more patches.
5 lines of code changed in 1 file:
Removing the extra boundary treatment occurring for diffusive fluxes. This was originally done to make the new stencil operations match the old Arches transport code. Since this old method adds extra work at boundaries and sane arguments can be made for this update, I am going ahead and breaking from tradition.
Viva la revolution!
0 lines of code changed in 2 files:
Allow user to specify independently the number
of rays used for divQ and flux calculations.
10 lines of code changed in 2 files:
Created a common initialization method for all property models. This
fixed a bug in how property models were being initialized. This will
cause some answers to change due to the initial condition.
Also fixed a bug for transported variables that are initialized using a
table lookup. In this case, boundary conditions weren't being reapplied
which will also cause some answers to change.
Finally updated some input files to conform to new standard.
43 lines of code changed in 7 files:
remove some leftover code that John put in while reworking the hypre solver interface.
0 lines of code changed in 1 file:
Allow users to specify a source term for the momentum equations. Also allow the input file to disable the pressure solve.
110 lines of code changed in 6 files:
Another small fix that takes the inclined plane simulation further in time.
6 lines of code changed in 1 file:
Fixed restart problem and some pathological cases.
42 lines of code changed in 7 files:
* Setup PC Transport source terms.
* Added an input file test test the PCTransport.
* Added a single patch version of the wall heat transfer model. This is
* probably the way we want to go.
372 lines of code changed in 5 files:
Added line search and a better criterion for stopping Newton iterations.
Runs for at least one test case now.
177 lines of code changed in 17 files:
Made changes to serial MPM enforcing a more physically realistic particle speed limiting scheme. Also changed the material model Arenisca.cc to subcycle a minimum of 4 times, as well as delete particles having a jacobian of the deformation gradient larger than 10. -Dave A.
15 lines of code changed in 2 files:
Added artificial viscosity to CamClay.
33 lines of code changed in 2 files:
RadPropertyCalculator should now be aware of the HAVE_RADPROPS directive.
2 lines of code changed in 1 file:
Fix documentation.
0 lines of code changed in 2 files:
Fix ordering for flux faces.
1 lines of code changed in 1 file:
Removed this directory that was moved to Arches.
0 lines of code changed in 2 files:
Bug fix on automation of solve boundary fluxes.
Code clean up.
21 lines of code changed in 1 file:
Updated version with corrected signs ans simplified convergence checks.
117 lines of code changed in 3 files:
Add two regression tests that verify the SSP-RK3 order of accuracy.
10 lines of code changed in 1 file:
More updates to the wall model. We might want to change this to a
patch-by-patch calculation rather than all-to-all.
232 lines of code changed in 2 files:
Adding begining fragments of the wall model interface.
49 lines of code changed in 4 files:
fix typo in the name of the RK3SSP manuscript reference.
1 lines of code changed in 1 file:
added artificial viscosity to internalForce calculation
46 lines of code changed in 1 file:
schedule and compute the zone of influence label for matl 0.
23 lines of code changed in 3 files:
fix a problem in my recent commit where the parsed modifier expressions were added to ALL variables that have boundary conditions. Change that and only add the modifier to its corresponding variable.
73 lines of code changed in 23 files:
Moved HypreSolverParams.h to AMR directory where it belongs and updated
all #include paths to reflect the change. This helps one's sanity.
HypreSolver.cc
- set do_setup to true on the first timestep. On restarts that wasn't
always the case.
-pruned some sections of code the were #if'd out.
* This should not change the answers.
145 lines of code changed in 21 files:
Report correct info when running UnifiedScheduler in SingleProcessor mode.
7 lines of code changed in 1 file:
Replaced the map that stored fluxes with a stencil7.
We can put the map back in later, but this will
allow for regression testing of the fluxes.
Also, now Ray.cc can handle multiple fluxes per
cell for cells that have more than one boundary face.
183 lines of code changed in 2 files:
Removed all locking mechanisms except for 1.
The cleanup script is called from a cron job and removes all SCI.*
directories.
31 lines of code changed in 11 files: