Add a compositional test case for the SSHIFT volume shift - #1588
Conversation
| / | ||
|
|
||
|
|
||
| -- These report times are the ones the reference simulator used for this deck, |
There was a problem hiding this comment.
Now, I am thinking it is not necessary, but let us keep it, we do not do this for cases usually. Maybe the exact time steps might be helpful in some situation.
There was a problem hiding this comment.
original DATA file
with trimmed adjusted time stepping in this PR (shorter)
There was a problem hiding this comment.
🟡 Changes recommended
The new deck uses FULLIMP / which is inconsistent with existing decks and may fail parsing if FULLIMP is treated as a no-record flag keyword.
Once you've addressed the issues Copilot identified, you can request another Copilot review.
Pull request overview
Adds a new compositional Eclipse deck intended to regression-test the SSHIFT volume-shift feature by exercising its effect on phase densities and density-dependent viscosities in a 1D CO₂ flood scenario.
Changes:
- Introduces a new 1D compositional test deck with mixed-sign SSHIFT values to validate both density and viscosity impacts.
- Configures reporting (RPTRST + SUMMARY) and a detailed set of timesteps matching a reference simulator run for step-by-step comparisons.
File summaries
| File | Description |
|---|---|
| compositional/SIMPLE_COMP_SSHIFT.DATA | New compositional deck intended to validate SSHIFT effects on densities and viscosities against a reference simulator. |
Review details
- Files reviewed: 1/1 changed files
- Comments generated: 1
- Review effort level: Lite
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
bska
left a comment
bska
left a comment
There was a problem hiding this comment.
This looks good for the most part, but FULLIMP should not have a keyword terminator (trailing slash).
6450677 to
d93c73f
Compare
Thanks. I updated the DATA file regarding FULLIMP. |
A one-dimensional CO2 flood of a CO2/methane/decane mixture, with volume shifts of mixed sign in PROPS. SSHIFT corrects the volume the equation of state predicts. That changes the oil and gas densities, and since the viscosity correlation uses density, it changes the viscosities too. The case covers both, so it fails if the shift stops reaching either one. The same deck was run in a reference simulator: the gas viscosity and the densities agree to five decimal places, while setting the shifts to zero moves the densities by several percent. The report times are the ones the reference used, so the two runs can be compared step by step.
bska
left a comment
bska
left a comment
There was a problem hiding this comment.
Thanks a lot for the updates. This looks good to me now and I'll merge into master.
A one-dimensional CO2 flood of a CO2/methane/decane mixture, with volume shifts of mixed sign in PROPS.
SSHIFT corrects the volume the equation of state predicts. That changes the oil and gas densities, and since the viscosity correlation uses density, it changes the viscosities too. The case covers both, so it fails if the shift stops reaching either one.
The same deck was run in a reference simulator: the gas viscosity and the densities agree to five decimal places, while setting the shifts to zero moves the densities by several percent. The report times are the ones the reference used, so the two runs can be compared step by step.
This test case is for the PR OPM/opm-common#5322