Articles | Volume 15, issue 8
https://doi.org/10.5194/gmd-15-3417-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-15-3417-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Development and technical paper
| 
28 Apr 2022
Development and technical paper |
| 28 Apr 2022
| 28 Apr 2022
Conservation laws in a neural network architecture: enforcing the atom balance of a Julia-based photochemical model (v0.2.0)
Air Quality Research Center, University of California, Davis, Davis,
California 95616, USA
Anthony S. Wexler
CORRESPONDING AUTHOR
Air Quality Research Center, University of California, Davis, Davis,
California 95616, USA
Department of Mechanical and Aerospace Engineering, University of
California, Davis, Davis, California 95616, USA
Department of Civil and
Environmental Engineering, University of
California, Davis, Davis, California 95616, USA
Department of Land, Air and Water Resources, University of
California, Davis, Davis, California 95616, USA
Viewed
Total article views: 4,166 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 16 Dec 2021)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 3,002 | 1,077 | 87 | 4,166 | 89 | 109 |
- HTML: 3,002
- PDF: 1,077
- XML: 87
- Total: 4,166
- BibTeX: 89
- EndNote: 109
Total article views: 3,277 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 28 Apr 2022)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 2,431 | 783 | 63 | 3,277 | 76 | 100 |
- HTML: 2,431
- PDF: 783
- XML: 63
- Total: 3,277
- BibTeX: 76
- EndNote: 100
Total article views: 889 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 16 Dec 2021)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 571 | 294 | 24 | 889 | 13 | 9 |
- HTML: 571
- PDF: 294
- XML: 24
- Total: 889
- BibTeX: 13
- EndNote: 9
Viewed (geographical distribution)
Total article views: 4,166 (including HTML, PDF, and XML)
Thereof 4,017 with geography defined
and 149 with unknown origin.
Total article views: 3,277 (including HTML, PDF, and XML)
Thereof 3,175 with geography defined
and 102 with unknown origin.
Total article views: 889 (including HTML, PDF, and XML)
Thereof 842 with geography defined
and 47 with unknown origin.
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
Cited
24 citations as recorded by crossref.
- Case Studies of Dimensionality in Chemical Data A. Blokhuis & R. Pollice 10.1002/ejoc.202400949
- Learning physical models that can respect conservation laws D. Hansen et al. 10.1016/j.physd.2023.133952
- Enabling micro-kinetics based simulation of industrial packed-bed reactors by physics-enhanced neural networks F. Biermann et al. 10.1016/j.cej.2025.163598
- A physics-constrained neural ordinary differential equations approach for robust learning of stiff chemical kinetics T. Kumar et al. 10.1080/13647830.2025.2478266
- Global reaction neural networks with embedded stoichiometry and thermodynamics for learning kinetics from reactor data T. Kircher et al. 10.1016/j.cej.2024.149863
- A Nudge to the Truth: Atom Conservation as a Hard Constraint in Models of Atmospheric Composition Using a Species-Weighted Correction P. Sturm & S. Silva 10.1021/acsestair.4c00220
- Improving air quality assessment using physics-inspired deep graph learning L. Li et al. 10.1038/s41612-023-00475-3
- Big Data in Earth system science and progress towards a digital twin X. Li et al. 10.1038/s43017-023-00409-w
- On the development of steady-state and dynamic mass-constrained neural networks using noisy transient data A. Mukherjee & D. Bhattacharyya 10.1016/j.compchemeng.2024.108722
- Interpretable conservation laws as sparse invariants Z. Liu et al. 10.1103/PhysRevE.109.L023301
- Robust mechanism discovery with atom conserving chemical reaction neural networks F. Döppel & M. Votsmeier 10.1016/j.proci.2024.105507
- Development of Steady-State and Dynamic Mass and Energy Constrained Neural Networks for Distributed Chemical Systems Using Noisy Transient Data A. Mukherjee & D. Bhattacharyya 10.1021/acs.iecr.4c01429
- Exact conservation laws for neural network integrators of dynamical systems E. Müller 10.1016/j.jcp.2023.112234
- Machine learning conservation laws from differential equations Z. Liu et al. 10.1103/PhysRevE.106.045307
- Machine Learning Surrogate Models for Mechanistic Kinetics: Embedding Atom Balance and Positivity T. Kircher & M. Votsmeier 10.1021/acs.jpclett.5c00602
- A numerical compass for experiment design in chemical kinetics and molecular property estimation M. Krüger et al. 10.1186/s13321-024-00825-0
- Phy-ChemNODE: an end-to-end physics-constrained autoencoder-NeuralODE framework for learning stiff chemical kinetics of hydrocarbon fuels T. Kumar et al. 10.3389/fther.2025.1594443
- Physics informed machine learning model for inverse dynamics in robotic manipulators W. Deng et al. 10.1016/j.asoc.2024.111877
- Downscaling atmospheric chemistry simulations with physically consistent deep learning A. Geiss et al. 10.5194/gmd-15-6677-2022
- Accelerating models for multiphase chemical kinetics through machine learning with polynomial chaos expansion and neural networks T. Berkemeier et al. 10.5194/gmd-16-2037-2023
- Advecting Superspecies: Efficiently Modeling Transport of Organic Aerosol With a Mass‐Conserving Dimensionality Reduction Method P. Sturm et al. 10.1029/2022MS003235
- Physics informed deep neural network embedded in a chemical transport model for the Amazon rainforest H. Sharma et al. 10.1038/s41612-023-00353-y
- Parallel hybrid ordinary differential equation for modeling biological phosphorus removal modified for enhanced predictive performance and physical interpretability G. Zhao et al. 10.1016/j.jenvman.2025.124932
- Learning kinetics from non-ideal reactors by implicitly solved finite volumes and global reaction neural networks T. Kircher & M. Votsmeier 10.1016/j.cej.2025.166749
24 citations as recorded by crossref.
- Case Studies of Dimensionality in Chemical Data A. Blokhuis & R. Pollice 10.1002/ejoc.202400949
- Learning physical models that can respect conservation laws D. Hansen et al. 10.1016/j.physd.2023.133952
- Enabling micro-kinetics based simulation of industrial packed-bed reactors by physics-enhanced neural networks F. Biermann et al. 10.1016/j.cej.2025.163598
- A physics-constrained neural ordinary differential equations approach for robust learning of stiff chemical kinetics T. Kumar et al. 10.1080/13647830.2025.2478266
- Global reaction neural networks with embedded stoichiometry and thermodynamics for learning kinetics from reactor data T. Kircher et al. 10.1016/j.cej.2024.149863
- A Nudge to the Truth: Atom Conservation as a Hard Constraint in Models of Atmospheric Composition Using a Species-Weighted Correction P. Sturm & S. Silva 10.1021/acsestair.4c00220
- Improving air quality assessment using physics-inspired deep graph learning L. Li et al. 10.1038/s41612-023-00475-3
- Big Data in Earth system science and progress towards a digital twin X. Li et al. 10.1038/s43017-023-00409-w
- On the development of steady-state and dynamic mass-constrained neural networks using noisy transient data A. Mukherjee & D. Bhattacharyya 10.1016/j.compchemeng.2024.108722
- Interpretable conservation laws as sparse invariants Z. Liu et al. 10.1103/PhysRevE.109.L023301
- Robust mechanism discovery with atom conserving chemical reaction neural networks F. Döppel & M. Votsmeier 10.1016/j.proci.2024.105507
- Development of Steady-State and Dynamic Mass and Energy Constrained Neural Networks for Distributed Chemical Systems Using Noisy Transient Data A. Mukherjee & D. Bhattacharyya 10.1021/acs.iecr.4c01429
- Exact conservation laws for neural network integrators of dynamical systems E. Müller 10.1016/j.jcp.2023.112234
- Machine learning conservation laws from differential equations Z. Liu et al. 10.1103/PhysRevE.106.045307
- Machine Learning Surrogate Models for Mechanistic Kinetics: Embedding Atom Balance and Positivity T. Kircher & M. Votsmeier 10.1021/acs.jpclett.5c00602
- A numerical compass for experiment design in chemical kinetics and molecular property estimation M. Krüger et al. 10.1186/s13321-024-00825-0
- Phy-ChemNODE: an end-to-end physics-constrained autoencoder-NeuralODE framework for learning stiff chemical kinetics of hydrocarbon fuels T. Kumar et al. 10.3389/fther.2025.1594443
- Physics informed machine learning model for inverse dynamics in robotic manipulators W. Deng et al. 10.1016/j.asoc.2024.111877
- Downscaling atmospheric chemistry simulations with physically consistent deep learning A. Geiss et al. 10.5194/gmd-15-6677-2022
- Accelerating models for multiphase chemical kinetics through machine learning with polynomial chaos expansion and neural networks T. Berkemeier et al. 10.5194/gmd-16-2037-2023
- Advecting Superspecies: Efficiently Modeling Transport of Organic Aerosol With a Mass‐Conserving Dimensionality Reduction Method P. Sturm et al. 10.1029/2022MS003235
- Physics informed deep neural network embedded in a chemical transport model for the Amazon rainforest H. Sharma et al. 10.1038/s41612-023-00353-y
- Parallel hybrid ordinary differential equation for modeling biological phosphorus removal modified for enhanced predictive performance and physical interpretability G. Zhao et al. 10.1016/j.jenvman.2025.124932
- Learning kinetics from non-ideal reactors by implicitly solved finite volumes and global reaction neural networks T. Kircher & M. Votsmeier 10.1016/j.cej.2025.166749
Latest update: 08 Oct 2025
Short summary
Large air quality and climate models require vast amounts of computational power. Machine learning tools like neural networks can be used to make these models more efficient, with the downside that their results might not make physical sense or be easy to interpret. This work develops a physically interpretable neural network that obeys scientific laws like conservation of mass and models atmospheric composition more accurately than a traditional neural network.
Large air quality and climate models require vast amounts of computational power. Machine...
