Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems
DOI10.1007/s00466-023-02296-warXiv2211.01873MaRDI QIDQ6164288
Francisco Chinesta, Quercus Hernández, Alberto Badías, Elias Cueto
Publication date: 27 July 2023
Published in: Computational Mechanics (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/2211.01873
port-Hamiltonian formalismscientific machine learningdouble thermoelastic pendulumthermodynamics principleviscoelastic beam interaction
Artificial neural networks and deep learning (68T07) Learning and adaptive systems in artificial intelligence (68T05) Rods (beams, columns, shafts, arches, rings, etc.) (74K10) Thermodynamics in solid mechanics (74A15) Thermal effects in solid mechanics (74F05) Numerical and other methods in solid mechanics (74S99)
Related Items
Cites Work
- A paradigm for joined Hamiltonian and dissipative systems
- Robust port-Hamiltonian representations of passive systems
- Deep learning of thermodynamics-aware reduced-order models from data
- Sympnets: intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems
- Structure-preserving neural networks
- Learning non-Markovian physics from data
- A proposal on machine learning via dynamical systems
- Pseudo-Hamiltonian neural networks with state-dependent external forces
- Thermodynamics-informed neural networks for physically realistic mixed reality
- Multiscale Thermo-Dynamics
- Particle and bracket formulations of kinetic equations
- Thermodynamically consistent time-stepping algorithms for non-linear thermomechanical systems
- Gradient and GENERIC time evolution towards reduced dynamics
- Port-Hamiltonian Systems Theory: An Introductory Overview
- Twenty years of distributed port-Hamiltonian systems: a literature review
- Machine Learning and Computational Mathematics
- A thermodynamics-informed active learning approach to perception and reasoning about fluids