GSPH module #94
Merged
GSPH module #94
Conversation
This file contains 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
This is just to test everything is working as it was before with the parsed out module. ApproxReimannSolvers will get their own class
added back in eps0 field and all the bc things that make compatible energy work. The compatible might be necessary for RSPH when pairwise DepsDt might have different sign. HLLC solver: gave the linear interface construction its own funcs and now applying linear construction to the pressure field
getting some oscillations on order of the kernel length w/ cfl = 0.25, goes away at cfl 0.5. Possible because using old DvDx DpDx
M corr added to continuity, DpDx and DvDx for approx reimann solver, energy conservation, and conservation of momentum. M corr on coon-mo wont conserve rotational momentum as force no longer acts on line of action
this adds extra dissiplation for high mach # impacts necessary for preventing interpenetration
compatible energy is weighted based on the pairwise DepsDt. Wave speed based on davis/einfeldt min/max strategy for larger Riemann presure when impacting
limiting based on the GSPH velocity gradient created too much diffusion at the FSI. Changed to using the raw SPH velocity gradient to determine velocity slope limiter and then apply that to the GSPH velocity gradient. Pressure gradient changed to the raw SPH version for now (this will need to be adjusted to deal with flows w/ gravitation).
For GSPH we need to allow pairwise internal energy DDts to be opposite signs. To do this we use basically the same compatible energy formulation but the difference between the eval-derivs calculated pairwise work (from DepsDti DepsDtj) and the conservative pairwise work (from acceleration) is used instead of the total pairwise work. This difference is subtracted from DepsDti & j to enforce conversation. The quantity subtracted is weighted by magnitude of DepsDti & j
…tion added, attempted a volume correction seems like using DvDt from past step for DpDx and last stemp DvDx works best for sedov type problems. The volume correction helps with volume conservation but lead to poor performance on sedov shock tests. The density sum with ASPH allowed for lower neighbor counts nPerh~2.5
modificaiton to DrhoDx calc is rolled into this
Mdeti required to be greater than 0.01. Was getting some trouble in wake regions.
these should be less likely to cause overlap troubles
some spatial derivs used to initialize the riemann solver gradient in the initialize method of GSPH were missing from the restart dump and read
jmikeowen
approved these changes
Oct 15, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Convolution - Free GSPH module.
This is currently only for fluids. An HLLC appox Riemann solver replaces aritficial viscosity and creates a interface velocity to resolve contact discontinuities. User can select the limiter and wave speed method. A bunch of options are included for what gradient is used to reconstruct the interface state.