[EPIC] PVT modeling

[francoishamon]

Description

The goal of this issue is to keep track of all developments related to PVT modeling capabilities for CO2 in depleted reservoirs.

Work Scope

Basic first steps to port PVTPackage capabilities in GEOS:

1. Create and (unit) test functions to solve Rachford-Rice equation (@francoishamon)
2. Create and (unit) test functions to model cubic EOS (@francoishamon)
3. Add initialization heuristics for K-values (@dkachuma)
4. Add and (unit) test the negative two-phase flash loop (@dkachuma)
5. Add density calculation for multiphase fluid (@dkachuma)
6. Add immiscible water three phase model

Steps to test the new functions in PVTDriver

7. Create a class derived from MultiFluidBase to read an input and call the new flash functions (@dkachuma)
8. Link MultiFluidBase class to static calculation functions.
9. Allow the user to supply the viscosity values for the constant viscosity model
10. Allow the user to select the mixing rule for the LBC viscosity calculation
11. Allow the user to tune the flash calculation parameters
12. Extend BlasLapackLA::solveLinearSystem to GPU

Steps to make the new code usable for simulations of CO2 injection in depleted reservoirs:

13. Add Soreide Whitson modification to the cubic EOS functions (for the water phase) (@francoishamon)
14. Add binary interaction coefficients
15. Add phase identification in the two-phase flash to identify the water phase
16. Add property calculations for density (CO2 with impurities, etc) and viscosity (LBC and others) (@dkachuma)
17. Compute finite-difference derivatives

Steps to make the code efficient:

18. Modify two-phase flash to do non-iterative K-value flash
19. Compute analytical derivatives
20. Explore Newton bisection in Rachford-Rice

Some tidying tasks

21. Centralise code to convert compositions from mass fractions to mole fractions
22. Simplify fluid model interface to have a single compute for both flow and initialisation
23. Use variadic template parameter for phase types in compositional multiphase fluid model
24. Refactor CubicEOSPhaseModel which has become too large/complex.

Issues in this EPIC

• Add Lohrenz-Bray-Clark viscosity correlation #1726
• Test case crashes with more components #1814
• Multi-phase EOS solver #2379
• Refactored constitutive/fluid folder #2441
• Implemented Rachford-Rice and cubic EOS #2451
• Implement negative two-phase flash #2484
• Implement k-value initialisation #2492
• Implement Cubic EOS derivatives #2520
• Create compositional multiphase fluid  #2533
• Create compositional multiphase fluid #2769
• Compositional multiphase fluid properties #2542
• Compositional tidy up #2661
• Refactor fluid unit tests #2695
• Add phase composition to PVTDriver output #2772
• Use ComponentProperties in calculation functions #2799
• Remove value only compute method from fluid models #2802
• Include water vapourisation in CO2/Brine flash #2811
• Cleanup compositional fluid model #2812
• Implement Spycher-Pruess solubility tables #2820
• Fix Rachford-Rice solver #2854
• Link compositional fluid model to static function implementation #2879
• Remove EOS template parameter from compositional fluid model #3121
• Activate fluid models #3123
• Implement stability test #3136
• Break up multi fluid test #3140
• Allow extra parameters for fluid models #3157
• Remove EOS template parameter from compositional fluid model #3166
• Extend BLAS interface #3173
• feat: Use stability test in flash #3206
• refactor: Add component selector for fluid models #3213
• refactor: Use matrix-matrix interface in LAPACK solve #3234
• feat: Immiscible water model #3236
• feat: Add immiscible water flash model #3237
• feat: Add constant k-value flash #3299
• Add Soreide-Whitson EOS #3309

[paveltomin]

Compute finite-difference derivatives - would you consider using AD instead? Flash is local, so AD operations are much easier.