Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems
This work addresses the problem of accurately modeling thermodynamics in complex physical systems for researchers in physics and machine learning, though it appears incremental as it builds on existing port-Hamiltonian methods.
The authors tackled the challenge of learning complex physical systems by developing port-metriplectic neural networks that enforce thermodynamic principles like energy conservation and non-negative entropy production, enabling predictions at the full system scale while reducing experimental and learning burdens.
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.