STC Seminar Hall, Dept. of Aerospace Engineering

Data-driven approaches have generated tremendous excitement in turbulence modeling, but enthusiasm has often outpaced scientific rigor. Many current AI/ML turbulence models lack physical interpretability, exhibit limited generalizability across flow regimes, and do not reflect the true dynamical nature of turbulence. A new strategy is needed—one that leverages AI while remaining fully compliant with the physics […]