Research reports/preprints
2024/07
On the feasibility of foundational models for the simulation of physical phenomena
Alicia Tierz, Mikel M. Iparraguirre, Icíar Alfaro, David González, Francisco Chinesta, Elías Cueto.
Submitted, 2024
2024/06
Graph neural networks informed locally by thermodynamics.
Alicia Tierz, Icíar Alfaro, David González, Francisco Chinesta, Elías Cueto.
Submitted, 2024.
2024/05
An indirect training approach for implicit constitutive modelling using recurrent neural networks and the virtual fields method.
R. Lourenço, P. Georgieva, E. Cueto, A. Andrade-Campos.
Computer Methods in Applied Mechanics and Engineering, 425, 116961, 2024.
2024/04
A comparison of Single and Double Generator Formalisms for Thermodynamics-Informed Neural Networks.
P. Urdeitx, I. Alfaro, D. Gonzalez, F. Chinesta, E. Cueto.
Comput Mech (2024).
https://doi.org/10.1007/s00466-024-02564-3
2024/03
A Neural Network Architecture for Physically-consistent Haptic Rendering.
Q. Hernandez, P. Martins, L. Tesan, I. Alfaro, D. Gonzalez, F. Chinesta, E. Cueto.
Submitted, 2024.
2024/02
Thermodynamics-informed super-resolution of scarce temporal dynamics data.
C. Bermejo-Barbanoj, B. Moya, A. Badías, F. Chinesta, E. Cueto.
Computer Methods in Applied Mechanics and Engineering, in press, 2024.
2024/01
Structure-preserving formulations for data-driven analysis of coupled multi-physics systems.
Alba Muixí, David González, Francisco Chinesta, Elías Cueto. Computational Mechanics, in press, 2024.
Documentary videos
Congresses and conferences
- Thermodynamics of learning physical phenomena. E. Cueto, semi-plenary conference. 9th European Congress on Computational Methods in Applied Sciences and Engineering ECCOMAS Congress 2024. 3-–7 June 2024, Lisbon, Portugal.
- Physics informed Graph Neural Networks – towards real-time applications for haptic devices. Pedro Martins, Beatriz Moya, Francisco Chinesta and Elías Cueto. 9th European Congress on Computational Methods in Applied Sciences and Engineering ECCOMAS Congress 2024. 3-–7 June 2024, Lisbon, Portugal.
- Physics-Informed Machine Learning for Characterizing Multistable Stochastic Dynamics. Beatriz Moya, Elias Cueto, Eleni Chatzi, and Francisco Chinesta. 9th European Congress on Computational Methods in Applied Sciences and Engineering ECCOMAS Congress 2024. 3-–7 June 2024, Lisbon, Portugal.
- On the use of thermodynamic biases for learning physical phenomena. E. Cueto. Groupement de Recherche I-GAIA, CNRS, Paris, March 2024.
- Thermodynamics-informed Graph Neural Networks for Lagrangian fluid simulation. Alicia Tierz,Beatriz Moya,Icíar Alfaro,David González,Francisco Chinesta and ElíasCueto. MORTech 2023-6th International Workshop on Model Reduction Techniques 22-24 November 2023, Paris-Saclay, France.
- Super-Resolution of Fluid-Dynamics Problems by Thermodynamics-Informed Deep Learning. Carlos Bermejo-Barbanoj, Beatriz Moya, Alberto Badías, Francisco Chinesta and Elías Cueto. MORTech 2023-6th International Workshop on Model Reduction Techniques 22-24 November 2023, Paris-Saclay, France.
- Recent advances in thermodynamics-informed neural networks for the prediction of physical phenomena. Quercus Hernandez, Beatriz Moya, Carlos Bermejo, Francisco Chinesta, Elias Cueto. MORTech 2023-6th International Workshop on Model Reduction Techniques 22-24 November 2023, Paris-Saclay, France.
- A Graph Neural Network (GNN) application to model thin shell deformations. Pedro Martins, Beatriz Moya, Alberto Badías, Francisco Chinesta and Elías Cueto. MORTech 2023-6th International Workshop on Model Reduction Techniques 22-24 November 2023, Paris-Saclay, France.