DOE OSTI · 3673144
HANNA: hard-constraint neural network for consistent activity coefficient prediction
Abstract
We introduce HANNA, the first hybrid neural network model that strictly complies with all thermodynamic consistency criteria for predicting activity coefficients and outperforms current benchmark methods in terms of accuracy and applicability.
Explore related subjects
Keep this discovery
Specht, Thomas (ORCID:0000000204628968), Nagda, Mayank (ORCID:0000000201899121), Fellenz, Sophie (ORCID:0000000253853926), Mandt, Stephan (ORCID:0000000178367839), Hasse, Hans (ORCID:0000000346125995), Jirasek, Fabian (ORCID:0000000225025701). 2024-01-01. HANNA: hard-constraint neural network for consistent activity coefficient prediction. https://doi.org/10.1039/d4sc05115g
Cite the original work for its findings. Save a collection to share your selection of sources.
Discover connections
Connections use source metadata and explicit phrase matches, not verified experimental comparisons.