Wall Models for Separated Turbulent Flows
Making large-eddy simulation affordable for flows that separate, using pressure-gradient-aware wall models.
Wall-resolved large-eddy simulation (LES) is too expensive for most engineering flows at realistic Reynolds numbers. This project develops wall models that remain accurate when the flow separates from a surface — the regime where classical equilibrium assumptions break down.
Goals
- Build a sensor that detects non-equilibrium regions from resolved LES quantities.
- Couple equilibrium and non-equilibrium wall models without spurious log-layer mismatch.
- Validate on periodic hills, a backward-facing step and a swept wing section.
Outcomes so far
Our first journal paper shows a ~30% improvement in reattachment-length prediction at about 2% of the cost of the wall-resolved reference. Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.