Traian Iliescu

Data-Driven Galerkin (d2G) Methods for Turbulence

Our group's research focuses on the development of data-driven Galerkin (d2G) methods for turbulent flows. In a nutshell, our research is about how data has changed (and continues to change) the classical Galerkin framework for turbulent flows.

Examples of d2G methods are : (i) Galerkin reduced order models (e.g., proper orthogonal decomposition (POD) and reduced basis method (RBM)), which employ a data-driven basis. (ii) Data-driven closures and stabilizations, in which data is used to find the optimal model form. (iii) Data assimilation, in which data is used to steer the computational model in the right direction.

Our group is working on modeling (i.e., constructing accurate and efficient d2G methods), numerical analysis (i.e., proving fundamental mathematical properties, e.g., stability, consistency, and convergence), and numerical simulation (e.g., testing the new d2G models in the simulation of turbulent flows in engineering, geophysical, and biomedical flows).

Contact Information:

• iliescu@vt.edu

• 428 McBryde Hall

• Department of Mathematics, Virginia Tech, Blacksburg, VA 24061