Support calcuation of critical properties of mixtures in Modular Properties #1336

andrewlee94 · 2024-02-13T21:11:14Z

Replaces part of #977

Summary/Motivation:

This PR adds support for calculating the critical properties of mixtures in the Modular Properties framework, which is necessary to support the new Smooth VLE formulation. This is done by adding a new method to EoS classes which contains instructions for how to calculate the mixture critical properties appropriate to the EoS.

When dealing with systems with multiple phases/EoS, only one EoS can be used thus it is necessary to determine which EoS to use. This PR applies the following steps:

Iterate over all phases and take the EoS associated with the first liquid phase encountered.
If no liquid phases are found, use the EoS associated with the vapor phase (assumes only 1 vapor phase).
If no liquid or vapor phases are found, raise an exception.

Changes proposed in this PR:

Add new unit definitions for specific volumes, and deprecate non-standard MOLAR_VOLUME unit.
Add compress_fact_crit and dens_mol_crit to the standard property set.
Add methods and metadata to core modular properties code to support critical properties as build-on-demand.
Add placeholder build_critical_properties method to EoSBase.
Add build_critical_properties method to IdealEoS. Cubic equivalent WIP.

Legal Acknowledgement

By contributing to this software project, I agree to the following terms and conditions for my contribution:

I agree my contributions are submitted under the license terms described in the LICENSE.txt file at the top level of this directory.
I represent I am authorized to make the contributions and grant the license. If my employer has rights to intellectual property that includes these contributions, I represent that I have received permission to make contributions and grant the required license on behalf of that employer.

codecov · 2024-02-13T21:57:22Z

Codecov Report

Attention: 20 lines in your changes are missing coverage. Please review.

Comparison is base (b7548b9) 77.43% compared to head (c4fff4f) 77.50%.

Files	Patch %	Lines
...erties/modular_properties/base/generic_property.py	84.74%	3 Missing and 6 partials ⚠️
...s/models/properties/modular_properties/eos/ceos.py	87.27%	3 Missing and 4 partials ⚠️
idaes/core/base/property_meta.py	66.66%	3 Missing ⚠️
...dels/properties/modular_properties/eos/eos_base.py	83.33%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1336      +/-   ##
==========================================
+ Coverage   77.43%   77.50%   +0.06%     
==========================================
  Files         390      390              
  Lines       63758    63884     +126     
  Branches    11737    11756      +19     
==========================================
+ Hits        49373    49514     +141     
+ Misses      11841    11799      -42     
- Partials     2544     2571      +27

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

dallan-keylogic · 2024-02-13T23:14:06Z

Most of this PR is good, but I think that adding critical properties to the ideal EoS is fundamentally misguided. There's no point at which the vapor and liquid phases become identical for the ideal EoS. I looked up "Kay's Rule" for calculating T_crit and P_crit for mixtures, and it appears to be used for estimating mixture critical properties to use for calculating Z in a corresponding states chart. For the ideal EoS Z=1 if we have a vapor and Z=0 if we have a liquid, there is no middle ground.

…al_props

andrewlee94 · 2024-02-20T17:27:20Z

@dallan-keylogic This should be ready for review. I've removed the ideal critical properties and added the cubics.

dallan-keylogic

The implementation looks largely fine, though I'm wondering why we're maintaining two separate initialization routines for the modular properties. It probably has to do with the old method of initialization vs the new method of initialization, but having one method call the other or extracting the common code into a function would be much cleaner. That impacts this PR only incidentally, though (where the same typos/stylistic choices were made in both parallel initialization methods).

Potentially more critically, we have four critical properties the user can provide, but they're related through an equation. Thus we have only three degrees of freedom. The user could, potentially, provide inconsistent data. The answer is probably "the user is responsible for understanding which properties are used by their EoS and for providing consistent data", but I thought I should raise the issue.

docs/explanations/components/property_package/general/phase_def.rst

dallan-keylogic · 2024-02-20T20:41:03Z

idaes/core/base/components.py

@@ -246,6 +246,7 @@ def build(self):

        # Create Vars for common parameters
        var_dict = {
+            "compress_fact_crit": pyunits.dimensionless,
            "dens_mol_crit": base_units.DENSITY_MOLE,
            "omega": pyunits.dimensionless,


It'd be nice to one day rename this "acentric factor", but that can wait for another day.

dallan-keylogic · 2024-02-20T20:43:57Z

idaes/core/base/tests/test_components.py

+                "compress_fact_crit": 1,
+                "dens_mol_crit": 55,
                "pressure_crit": 1e5,
                "temperature_crit": 500,


This is a good example of the perils of inconsistent data. We say Z_crit = 1, but P_crit/(rho_mol_critRT_crit) ~= 0.4376. It probably doesn't matter for this test, though.

idaes/models/properties/modular_properties/base/generic_property.py

dallan-keylogic · 2024-02-20T21:39:24Z

idaes/models/properties/modular_properties/base/generic_property.py

+            elif self.params.get_phase(p).is_vapor_phase():
+                vap_phase = p


Perhaps initialize vap_phase as None, then if you find a second vapor phase have a warning/exception?

dallan-keylogic · 2024-02-20T22:34:36Z

idaes/models/properties/modular_properties/eos/ceos.py

+        def rule_A_crit(m):
+            am_crit = getattr(m, "am_crit")
+            return EoS_param[ctype]["omegaA"] * (
+                Cubic.gas_constant(m) * m.temperature_crit
+            ) ** 2 == (am_crit * m.pressure_crit)
+
+        m.add_component("A_crit", Constraint(rule=rule_A_crit))
+
+        def rule_B_crit(m):
+            bm_crit = getattr(m, "bm_crit")
+            return EoS_param[ctype]["coeff_b"] * Cubic.gas_constant(
+                m
+            ) * m.temperature_crit == (bm_crit * m.pressure_crit)
+
+        m.add_component("B_crit", Constraint(rule=rule_B_crit))


Can't these be Expressions? Or is the idea to avoid divisions.

No, these are the actual constraints we need to solve to find T_c and P_c.

dallan-keylogic · 2024-02-20T22:50:51Z

idaes/models/properties/modular_properties/eos/tests/test_ceos_PR.py

+        m.params = GenericParameterBlock(
+            components={
+                "a": {
+                    "parameter_data": {
+                        "omega": 0.1,
+                        "compress_fact_crit": 1,
+                        "dens_mol_crit": 1,
+                        "pressure_crit": 1e5,
+                        "temperature_crit": 100,
+                    }
+                },
+                "b": {
+                    "parameter_data": {
+                        "omega": 0.2,
+                        "compress_fact_crit": 1,
+                        "dens_mol_crit": 1,
+                        "pressure_crit": 2e5,
+                        "temperature_crit": 200,
+                    }
+                },
+                "c": {
+                    "parameter_data": {
+                        "omega": 0.3,
+                        "compress_fact_crit": 1,
+                        "dens_mol_crit": 1,
+                        "pressure_crit": 3e5,
+                        "temperature_crit": 300,
+                    }
+                },
+            },


Here we might start to run into problems with the critical properties not being consistent. Knowing how the cubic EoS works, I imagine what happens is that T_crit and P_crit influence the mixture critical properties while rho_mol_crit and Z_crit do not.

Z_crit and dens_mol_crit are only used for initialization (at this point) so it won't really matter too much.

andrewlee94 · 2024-02-21T14:13:34Z

Regarding having the two initialization routines, it is backward compatibility for the older approach. As they are in separate classes and the new approach also has a few things the old one does not, I did not want one to call the other. It would be possible to have them use a single, common method for the main initialization routine, but at the time I wanted the new version to package everything it would need neatly.

dallan-keylogic

LGTM

agarciadiego

LGTM

docs/explanations/components/property_package/general/phase_def.rst

andrewlee94 added 4 commits February 13, 2024 10:10

Adding volume_mass as unit and renaming molar_volume

819453a

Hooks for mixture critical properties

9e3e098

Hook for link ot EoS for critical properties

027ad83

Adding critical properties to ideal EoS

4af1dda

andrewlee94 requested a review from dallan-keylogic February 13, 2024 21:11

andrewlee94 self-assigned this Feb 13, 2024

andrewlee94 added enhancement New feature or request Priority:Normal Normal Priority Issue or PR property packages Issues dealing with properties WIP deprecation Add deprecation to codebase and/or remove deprecated feature labels Feb 13, 2024

andrewlee94 added 8 commits February 15, 2024 13:06

Initial working cubic critical properties

d795063

Adding basic initialization

59525d6

Refining crit prop initialization

9d06c2e

Finishing initialization

a7a0e92

Fixing pylint and removing ideal crit props

ea8ff6e

Merge branch 'main' of https://github.com/IDAES/idaes-pse into critic…

692abbb

…al_props

Docs and extra error message

66db51f

Test for new error message

46253e3

andrewlee94 changed the title ~~[WIP] Support calcuation of critical properties of mixtures in Modular Properties~~ Support calcuation of critical properties of mixtures in Modular Properties Feb 20, 2024

andrewlee94 marked this pull request as ready for review February 20, 2024 17:26

andrewlee94 requested review from eslickj, lbianchi-lbl and ksbeattie as code owners February 20, 2024 17:26

andrewlee94 added 2 commits February 20, 2024 12:39

Fixing typo

16520bd

Fixing Sphinx and pylint errors

7bdff70

dallan-keylogic requested changes Feb 20, 2024

View reviewed changes

Address PR review

c1a1f75

dallan-keylogic approved these changes Feb 21, 2024

View reviewed changes

andrewlee94 requested a review from agarciadiego February 21, 2024 14:41

agarciadiego approved these changes Feb 21, 2024

View reviewed changes

docs/explanations/components/property_package/general/phase_def.rst Outdated Show resolved Hide resolved

Fixing typo

c4fff4f

andrewlee94 enabled auto-merge (squash) February 21, 2024 15:41

andrewlee94 merged commit 94b98f7 into IDAES:main Feb 21, 2024
39 checks passed

andrewlee94 deleted the critical_props branch February 21, 2024 16:49

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Support calcuation of critical properties of mixtures in Modular Properties #1336

Support calcuation of critical properties of mixtures in Modular Properties #1336

andrewlee94 commented Feb 13, 2024 •

edited

Loading

codecov bot commented Feb 13, 2024 •

edited

Loading

dallan-keylogic commented Feb 13, 2024

andrewlee94 commented Feb 20, 2024

dallan-keylogic left a comment

dallan-keylogic Feb 20, 2024

dallan-keylogic Feb 20, 2024

dallan-keylogic Feb 20, 2024

dallan-keylogic Feb 20, 2024

andrewlee94 Feb 21, 2024

dallan-keylogic Feb 20, 2024

andrewlee94 Feb 21, 2024

andrewlee94 commented Feb 21, 2024

dallan-keylogic left a comment

agarciadiego left a comment

		elif self.params.get_phase(p).is_vapor_phase():
		vap_phase = p

Support calcuation of critical properties of mixtures in Modular Properties #1336

Support calcuation of critical properties of mixtures in Modular Properties #1336

Conversation

andrewlee94 commented Feb 13, 2024 • edited Loading

Replaces part of #977

Summary/Motivation:

Changes proposed in this PR:

Legal Acknowledgement

codecov bot commented Feb 13, 2024 • edited Loading

Codecov Report

dallan-keylogic commented Feb 13, 2024

andrewlee94 commented Feb 20, 2024

dallan-keylogic left a comment

Choose a reason for hiding this comment

dallan-keylogic Feb 20, 2024

Choose a reason for hiding this comment

dallan-keylogic Feb 20, 2024

Choose a reason for hiding this comment

dallan-keylogic Feb 20, 2024

Choose a reason for hiding this comment

dallan-keylogic Feb 20, 2024

Choose a reason for hiding this comment

andrewlee94 Feb 21, 2024

Choose a reason for hiding this comment

dallan-keylogic Feb 20, 2024

Choose a reason for hiding this comment

andrewlee94 Feb 21, 2024

Choose a reason for hiding this comment

andrewlee94 commented Feb 21, 2024

dallan-keylogic left a comment

Choose a reason for hiding this comment

agarciadiego left a comment

Choose a reason for hiding this comment

andrewlee94 commented Feb 13, 2024 •

edited

Loading

codecov bot commented Feb 13, 2024 •

edited

Loading