Remove reaction path degeneracy adjustments in surface families #655
Conversation
bjkreitz
force-pushed
the
clean_kinetic_trees
branch
from
f9be4eb
to
a5f453d
Compare
Regression Testing ResultsWARNING:root:Initial mole fractions do not sum to one; normalizing. Detailed regression test results.Regression test aromatics:Reference: Execution time (DD:HH:MM:SS): 00:00:01:10 aromatics Passed Core Comparison ✅Original model has 15 species. aromatics Passed Edge Comparison ✅Original model has 106 species.
Observables Test Case: Aromatics Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! aromatics Passed Observable Testing ✅Regression test liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:14 liquid_oxidation Failed Core Comparison ❌Original model has 37 species. Non-identical kinetics! ❌
kinetics: liquid_oxidation Failed Edge Comparison ❌Original model has 202 species. Non-identical kinetics! ❌
kinetics:
Observables Test Case: liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! liquid_oxidation Passed Observable Testing ✅Regression test nitrogen:Reference: Execution time (DD:HH:MM:SS): 00:00:01:28 nitrogen Passed Core Comparison ✅Original model has 41 species. nitrogen Passed Edge Comparison ✅Original model has 132 species.
Observables Test Case: NC Comparison
✅ All Observables varied by less than 0.200 on average between old model and new model in all conditions! nitrogen Passed Observable Testing ✅Regression test oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:26 oxidation Passed Core Comparison ✅Original model has 59 species. oxidation Passed Edge Comparison ✅Original model has 230 species.
Observables Test Case: Oxidation Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! oxidation Passed Observable Testing ✅Regression test sulfur:Reference: Execution time (DD:HH:MM:SS): 00:00:00:57 sulfur Passed Core Comparison ✅Original model has 27 species. sulfur Failed Edge Comparison ❌Original model has 89 species.
Observables Test Case: SO2 Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! sulfur Passed Observable Testing ✅Regression test superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:00:40 superminimal Passed Core Comparison ✅Original model has 13 species. superminimal Passed Edge Comparison ✅Original model has 18 species. Regression test RMS_constantVIdealGasReactor_superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:02:21 RMS_constantVIdealGasReactor_superminimal Passed Core Comparison ✅Original model has 13 species. RMS_constantVIdealGasReactor_superminimal Passed Edge Comparison ✅Original model has 13 species.
Observables Test Case: RMS_constantVIdealGasReactor_superminimal Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_constantVIdealGasReactor_superminimal Passed Observable Testing ✅Regression test RMS_CSTR_liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:05:52 RMS_CSTR_liquid_oxidation Passed Core Comparison ✅Original model has 37 species. RMS_CSTR_liquid_oxidation Failed Edge Comparison ❌Original model has 206 species.
Observables Test Case: RMS_CSTR_liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_CSTR_liquid_oxidation Passed Observable Testing ✅beep boop this comment was written by a bot 🤖 |
|
I ran the methanol model. The behavior for me was pretty similar. comparing the two edge models for methanol I found there reactions were significantly different, due to using different training reactions: 
If I have time I will run without the training reactions, but removing the rules does change the averaging for the trees, sometimes by quite a bit. I can't say if this is a good or a bad thing, given my model is not super different, but I suspect it is bad, because we have less specificity when we match the rules. I know our current design philosophy is to move towards using rules instead of training data, but currently we do have training data populated, so I'm hesitant to introduce something that will make that worse. Would it be too difficult to just leave the groups.py the way they are, and just remove the rules? or would that be even worse than our current setup? |
|
Thanks for the thorough review, Chris. I never thought about keeping the nodes in the trees since I haven't really populated the trees very well. However, there is no downside to keeping them and just removing the rate rules. I'll make the changes. |
ChrisBNEU
force-pushed
the
clean_kinetic_trees
branch
from
a5f453d
to
f1901c3
Compare
Regression Testing ResultsWARNING:root:Initial mole fractions do not sum to one; normalizing. Detailed regression test results.Regression test aromatics:Reference: Execution time (DD:HH:MM:SS): 00:00:01:09 aromatics Passed Core Comparison ✅Original model has 15 species. aromatics Passed Edge Comparison ✅Original model has 106 species.
Observables Test Case: Aromatics Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! aromatics Passed Observable Testing ✅Regression test liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:13 liquid_oxidation Failed Core Comparison ❌Original model has 37 species. liquid_oxidation Failed Edge Comparison ❌Original model has 202 species.
Observables Test Case: liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! liquid_oxidation Passed Observable Testing ✅Regression test nitrogen:Reference: Execution time (DD:HH:MM:SS): 00:00:01:29 nitrogen Passed Core Comparison ✅Original model has 41 species. nitrogen Passed Edge Comparison ✅Original model has 132 species.
Observables Test Case: NC Comparison
✅ All Observables varied by less than 0.200 on average between old model and new model in all conditions! nitrogen Passed Observable Testing ✅Regression test oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:31 oxidation Passed Core Comparison ✅Original model has 59 species. oxidation Passed Edge Comparison ✅Original model has 230 species.
Observables Test Case: Oxidation Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! oxidation Passed Observable Testing ✅Regression test sulfur:Reference: Execution time (DD:HH:MM:SS): 00:00:00:58 sulfur Passed Core Comparison ✅Original model has 27 species. sulfur Failed Edge Comparison ❌Original model has 89 species.
Observables Test Case: SO2 Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! sulfur Passed Observable Testing ✅Regression test superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:00:39 superminimal Passed Core Comparison ✅Original model has 13 species. superminimal Passed Edge Comparison ✅Original model has 18 species. Regression test RMS_constantVIdealGasReactor_superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:02:25 RMS_constantVIdealGasReactor_superminimal Passed Core Comparison ✅Original model has 13 species. RMS_constantVIdealGasReactor_superminimal Passed Edge Comparison ✅Original model has 13 species.
Observables Test Case: RMS_constantVIdealGasReactor_superminimal Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_constantVIdealGasReactor_superminimal Passed Observable Testing ✅Regression test RMS_CSTR_liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:06:01 RMS_CSTR_liquid_oxidation Passed Core Comparison ✅Original model has 37 species. RMS_CSTR_liquid_oxidation Passed Edge Comparison ✅Original model has 206 species.
Observables Test Case: RMS_CSTR_liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_CSTR_liquid_oxidation Passed Observable Testing ✅beep boop this comment was written by a bot 🤖 |
|
Thanks @ChrisBNEU! Everything looks good to me now. @12Chao can you take another quick look. |
| entry( | ||
| index = 8, | ||
| label = "C-C;VacantSite", | ||
| kinetics = SurfaceArrheniusBEP( | ||
| A = (2.09e17, 'm^2/(mol*s)'), |
There was a problem hiding this comment.
I wonder about this one, because the group is symmetrical it probably will match both ways around and end up being doubled. So perhaps it should be halved here.
A lot of the other groups (eg. C-OH, and C-H) have a forbidden structures defined to stop it from matching in both directions.
There was a problem hiding this comment.
Good catch @rwest. I checked the a generated mechanism and it is indeed being doubled. I'll divide the pre-exponential coefficient by 2 and add a note.
bjkreitz
force-pushed
the
clean_kinetic_trees
branch
from
f1901c3
to
a6aa53a
Compare
Regression Testing ResultsWARNING:root:Initial mole fractions do not sum to one; normalizing. Detailed regression test results.Regression test aromatics:Reference: Execution time (DD:HH:MM:SS): 00:00:01:09 aromatics Passed Core Comparison ✅Original model has 15 species. aromatics Passed Edge Comparison ✅Original model has 106 species.
Observables Test Case: Aromatics Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! aromatics Passed Observable Testing ✅Regression test liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:15 liquid_oxidation Failed Core Comparison ❌Original model has 37 species. liquid_oxidation Failed Edge Comparison ❌Original model has 202 species. Non-identical kinetics! ❌
kinetics:
Observables Test Case: liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! liquid_oxidation Passed Observable Testing ✅Regression test nitrogen:Reference: Execution time (DD:HH:MM:SS): 00:00:01:25 nitrogen Failed Core Comparison ❌Original model has 41 species. nitrogen Failed Edge Comparison ❌Original model has 132 species.
Observables Test Case: NC Comparison
✅ All Observables varied by less than 0.200 on average between old model and new model in all conditions! nitrogen Passed Observable Testing ✅Regression test oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:31 oxidation Passed Core Comparison ✅Original model has 59 species. oxidation Passed Edge Comparison ✅Original model has 230 species.
Observables Test Case: Oxidation Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! oxidation Passed Observable Testing ✅Regression test sulfur:Reference: Execution time (DD:HH:MM:SS): 00:00:00:56 sulfur Passed Core Comparison ✅Original model has 27 species. sulfur Failed Edge Comparison ❌Original model has 89 species.
Observables Test Case: SO2 Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! sulfur Passed Observable Testing ✅Regression test superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:00:41 superminimal Passed Core Comparison ✅Original model has 13 species. superminimal Passed Edge Comparison ✅Original model has 18 species. Regression test RMS_constantVIdealGasReactor_superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:02:26 RMS_constantVIdealGasReactor_superminimal Passed Core Comparison ✅Original model has 13 species. RMS_constantVIdealGasReactor_superminimal Passed Edge Comparison ✅Original model has 13 species.
Observables Test Case: RMS_constantVIdealGasReactor_superminimal Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_constantVIdealGasReactor_superminimal Passed Observable Testing ✅Regression test RMS_CSTR_liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:06:01 RMS_CSTR_liquid_oxidation Failed Core Comparison ❌Original model has 37 species. RMS_CSTR_liquid_oxidation Failed Edge Comparison ❌Original model has 206 species. Non-identical kinetics! ❌
kinetics:
Observables Test Case: RMS_CSTR_liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_CSTR_liquid_oxidation Passed Observable Testing ✅beep boop this comment was written by a bot 🤖 |
Regression Testing ResultsWARNING:root:Initial mole fractions do not sum to one; normalizing. Detailed regression test results.Regression test aromatics:Reference: Execution time (DD:HH:MM:SS): 00:00:01:05 aromatics Passed Core Comparison ✅Original model has 15 species. aromatics Passed Edge Comparison ✅Original model has 106 species.
Observables Test Case: Aromatics Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! aromatics Passed Observable Testing ✅Regression test liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:09 liquid_oxidation Failed Core Comparison ❌Original model has 37 species. Non-identical kinetics! ❌
kinetics: liquid_oxidation Failed Edge Comparison ❌Original model has 202 species. Non-identical kinetics! ❌
kinetics:
Observables Test Case: liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! liquid_oxidation Passed Observable Testing ✅Regression test nitrogen:Reference: Execution time (DD:HH:MM:SS): 00:00:01:24 nitrogen Passed Core Comparison ✅Original model has 41 species. nitrogen Failed Edge Comparison ❌Original model has 133 species. Non-identical thermo! ❌
thermo: Thermo group additivity estimation: group(O2s-CdN3d) + group(N3d-OCd) + group(Cd-HN3dO) + ring(oxirene) + radical(CdJ-NdO) Non-identical kinetics! ❌
kinetics:
Observables Test Case: NC Comparison
✅ All Observables varied by less than 0.200 on average between old model and new model in all conditions! nitrogen Passed Observable Testing ✅Regression test oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:02:27 oxidation Passed Core Comparison ✅Original model has 59 species. oxidation Passed Edge Comparison ✅Original model has 230 species.
Observables Test Case: Oxidation Comparison
✅ All Observables varied by less than 0.500 on average between old model and new model in all conditions! oxidation Passed Observable Testing ✅Regression test sulfur:Reference: Execution time (DD:HH:MM:SS): 00:00:00:54 sulfur Passed Core Comparison ✅Original model has 27 species. sulfur Failed Edge Comparison ❌Original model has 89 species.
Observables Test Case: SO2 Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! sulfur Passed Observable Testing ✅Regression test superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:00:35 superminimal Passed Core Comparison ✅Original model has 13 species. superminimal Passed Edge Comparison ✅Original model has 18 species. Regression test RMS_constantVIdealGasReactor_superminimal:Reference: Execution time (DD:HH:MM:SS): 00:00:02:24 RMS_constantVIdealGasReactor_superminimal Passed Core Comparison ✅Original model has 13 species. RMS_constantVIdealGasReactor_superminimal Passed Edge Comparison ✅Original model has 13 species.
Observables Test Case: RMS_constantVIdealGasReactor_superminimal Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_constantVIdealGasReactor_superminimal Passed Observable Testing ✅Regression test RMS_CSTR_liquid_oxidation:Reference: Execution time (DD:HH:MM:SS): 00:00:05:55 RMS_CSTR_liquid_oxidation Failed Core Comparison ❌Original model has 37 species. RMS_CSTR_liquid_oxidation Failed Edge Comparison ❌Original model has 206 species.
Observables Test Case: RMS_CSTR_liquid_oxidation Comparison
✅ All Observables varied by less than 0.100 on average between old model and new model in all conditions! RMS_CSTR_liquid_oxidation Passed Observable Testing ✅beep boop this comment was written by a bot 🤖 |
There was a problem hiding this comment.
Looks good,
I ran a mechanism with NH3, CO, and O2. I saw that on the main branch 36 species and 122 reactions were found in the core, and 119 species and 533 reactions were found in the edge. On the new branch, I saw 37 species and 149 reactions were found in the core, and 122 species and 534 reactions were found in the edge.
Most of the kinetics were identical, only a small handful of reactions were effected. I think it is important to note that on the main branch, if H was dissociated from *NH2, the A from the top level lode was used and degeneracy was applied, but if Hydrogen from CH2 was dissociated, the A from a rule at a lower level was used where it was divided by 2. This lead to inconsistencies in when A was divided by degeneracy or not in the main branch.
This PR seems to allow RMG to now apply degeneracy in a consistent way, so I approve.
Description
In our last RMG-Cat meeting, we extensively discussed the concept of reaction path degeneracy and whether we should consider this for surface reactions, too. In the general catalysis community, this effect is not considered in the pre-exponential coefficients. Since RMG applies a reaction path degeneracy by default, it was necessary to create extra nodes in the trees of the reaction families. These extra nodes contained rate rules with pre-exponential coefficients that were divided by the reaction path degeneracy that is later applied by RMG (typically 2 or 3). Based on our discussion, we reached the conclusion that the reaction path degeneracy is necessary to accurately estimate the kinetics of surface reactions. Thus, I removed all the extra nodes in the trees along with their rate rules. This makes the trees less bulky and gets rid of some confusion.
I removed nodes and rules from the following families:
The only family that still contains pre-exponential coefficients adjusted by the degeneracy is
Surface_Adsorption_Dissociative. Sticking coefficients are typically based on empirical estimates and are in the range between 0-1. We decided to turn off degeneracy for sticking coefficients, but I haven't looked into this yet. Once I turn it off in the RMG-Py, I'll go back and remove the extra nodes in this family, too.Testing
I generated a mechanism with and without the degeneracy-corrected nodes for the oxidation of C2H6 over Pt(111). I obtained identical mechanisms for both cases. This result was anticipated since reaction path degeneracy doesn't have a huge impact on the rates.
Review
I suggest generating a mechanism with this branch and with the regular branch. The results should be almost the same.
Otherwise, please double-check that I haven't accidentally deleted important nodes from the trees. The review should be fairly quick because I didn't make major changes.