Scientific Machine Learning and Physics-Informed AI for Multiscale and Multiphysics Fluid Mechanics

This Special Issue on Scientific Machine Learning and Physics-Informed AI for Multiscale and Multiphysics Fluid Mechanics will highlight advances at the intersection of fluid physics, computation, data, and AI, especially where conventional modelling is limited by scale separation, multiphysics coupling, sparse data, and computational cost.

Topics include scientific machine learning methods constrained or interpreted through governing equations, conservation laws, and thermodynamic consistency, with applications to turbulence, shocks, interfacial and reacting flows, multiphase systems, fluid–structure interaction, porous and biological media, and electrohydrodynamic and thermo-fluid processes. Contributions are encouraged on reduced-order models, operator learning, physics-informed AI, and AI-enhanced solvers for prediction, control, and design. The issue also emphasizes trustworthy AI, including uncertainty quantification, verification and validation, interpretability, reproducibility, benchmark design, and comparisons with theory, simulations, and experiments, particularly for strongly nonlinear and multiscale flows.