Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Computational Mechanics Année : 2023

Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems

Résumé

We develop inductive biases for the machine learning of complex physical systems based on the port-Hamiltonian formalism. To satisfy by construction the principles of thermodynamics in the learned physics (conservation of energy, non-negative entropy production), we modify accordingly the port-Hamiltonian formalism so as to achieve a port-metriplectic one. We show that the constructed networks are able to learn the physics of complex systems by parts, thus alleviating the burden associated to the experimental characterization and posterior learning process of this kind of systems. Predictions can be done, however, at the scale of the complete system. Examples are shown on the performance of the proposed technique.
Fichier principal
Vignette du fichier
s00466-023-02296-w.pdf (520.89 Ko) Télécharger le fichier
Origine : Publication financée par une institution
Licence : CC BY - Paternité

Dates et versions

hal-04163352 , version 1 (17-07-2023)

Licence

Paternité

Identifiants

Citer

Quercus Hernández, Alberto Badías, Francisco Chinesta, Elías Cueto. Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems. Computational Mechanics, 2023, 72, pp.553-561. ⟨10.1007/s00466-023-02296-w⟩. ⟨hal-04163352⟩
58 Consultations
17 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More