A thermodynamics-informed active learning approach to perception and reasoning about fluids

DOI10.1007/S00466-023-02279-XMaRDI QIDQ6164293zbMATH OpenOpenAlexFDO

Authors

Publication date 27 July 2023

Published in Computational Mechanics (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2203.05775

zbMATH Keywords

hybrid twindynamic data-driven applicationliquid sloshing simulationlow-data regimephysics-informed incremental learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Jets and cavities, cavitation, free-streamline theory, water-entry problems, airfoil and hydrofoil theory, sloshing (76B10) Foundations, constitutive equations, rheology, hydrodynamical models of non-fluid phenomena (76A99) Basic methods in fluid mechanics (76M99)

Abstract: Learning and reasoning about physical phenomena is still a challenge in robotics development, and computational sciences play a capital role in the search for accurate methods able to provide explanations for past events and rigorous forecasts of future situations. We propose a thermodynamics-informed active learning strategy for fluid perception and reasoning from observations. As a model problem, we take the sloshing phenomena of different fluids contained in a glass. Starting from full-field and high-resolution synthetic data for a particular fluid, we develop a method for the tracking (perception) and analysis (reasoning) of any previously unseen liquid whose free surface is observed with a commodity camera. This approach demonstrates the importance of physics and knowledge not only in data-driven (grey box) modeling but also in the correction for real physics adaptation in low data regimes and partial observations of the dynamics. The method presented is extensible to other domains such as the development of cognitive digital twins, able to learn from observation of phenomena for which they have not been trained explicitly.

Recommendations

• Learning Physics from Data: A Thermodynamic Interpretation
• scientific article; zbMATH DE number 1096541
• Machine learning active-nematic hydrodynamics
• Active learning of constitutive relation from mesoscopic dynamics for macroscopic modeling of non-Newtonian flows
• Theoretical prerequisites for physically justified machine learning and its applications to fluid dynamics
• Deep learning of thermodynamics-aware reduced-order models from data
• Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations
• Learning by neural networks under physical constraints for simulation in fluid mechanics
• A multiscale computational framework using active learning to model complex suspension flows
• Deep learning observables in computational fluid dynamics

Cites work

• scientific article; zbMATH DE number 1229396 (Why is no real title available?)
• scientific article; zbMATH DE number 2161894 (Why is no real title available?)
• scientific article; zbMATH DE number 3261539 (Why is no real title available?)
• Deep learning
• Deep learning of thermodynamics-aware reduced-order models from data
• Digital twins that learn and correct themselves
• End-to-end training of deep visuomotor policies
• Metriplectic integrators for dissipative fluids
• Structure-preserving neural networks
• Thermodynamically consistent time-stepping algorithms for non-linear thermomechanical systems

Cited in

(2)

• NN-mCRE: a modified constitutive relation error framework for unsupervised learning of nonlinear state laws with physics-augmented neural networks
• Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems

This page was built for publication: A thermodynamics-informed active learning approach to perception and reasoning about fluids

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6164293)