Applied Turbulence Modeling - Lecture 2

21 October 2026, Sebastian Spring

Kolmogorov's hypotheses

argumentation for why turbulence models are needed: complete resolving at kolmogorov scales (smallest scales of turbulence) is extremly expensive (can be calculated using rule of thumb equation found on slide ??)

Local Isotropy: At sufficiently high $Re$, ther small-scale motions $\ell << L$ are statiistically isotropic
First similarity hypothesis: The statistics of the small scales are uniquely determined by $\nu$ and $\varepsilon$
Second similarity hypothesis: Ilias was down so i couldn't check the slides

Modeling Strategies

• DNS: Full energy spectrum is resolved
• LES: large eddie simulation - more conservative, older approach, where large eddies are resolved, rest is modeled
• RANS: everything is modeled

Governing Equations and Reynolds Averaging

pretty theoretical, had to focus