CFD Post-Processing, Validation & Verification Flashcards

What is the purpose of computing the 'pressure coefficient' (Cp) distribution in external aerodynamics CFD post-processing?

Answer: To non-dimensionalize pressure for comparison across different flow speeds and geometries

The pressure coefficient Cp = (p - p∞) / (0.5ρU²∞) normalizes local pressure against freestream dynamic pressure, enabling comparison of pressure distributions across different scales, speeds, and geometries.

In a CFD report for engineering design, which information is essential to include for the simulation to be reproducible?

Answer: Solver settings, mesh statistics, boundary conditions, turbulence model, and convergence criteria

Reproducibility requires complete documentation of solver type, discretization schemes, mesh details, all boundary conditions, physics models (e.g., turbulence), and convergence criteria so another engineer can replicate the simulation.

Which approach is used to extract a boundary layer velocity profile from a CFD solution for comparison with the Blasius or log-law theoretical profiles?

Answer: Sampling velocity data along a line normal to the wall at a specific streamwise location

A line probe (or 'rake') is placed perpendicular to the wall at the desired streamwise station to extract the velocity magnitude versus wall-normal distance, which is then compared with analytical profiles.

The 'effective viscosity ratio' (μt/μ) in post-processing of RANS simulations indicates:

Answer: The local importance of turbulent mixing relative to molecular viscosity

The turbulent-to-laminar viscosity ratio μt/μ shows where turbulence dominates momentum transport; very high values (>1000) may indicate unrealistic over-prediction of turbulent viscosity in some RANS models.

Which numerical artifact can appear in CFD results when using a first-order upwind scheme on a coarse mesh, and is revealed during post-processing?

Answer: Numerical diffusion — excessive smearing of sharp gradients and shocks

First-order upwind schemes introduce significant numerical diffusion (false diffusion), which smears sharp features like boundary layers, shocks, and mixing layers, producing overly smooth solutions on coarse meshes.

In post-processing of combustion CFD results, the 'mixture fraction' variable is used to:

Answer: Track the local composition between pure fuel and pure oxidizer streams

Mixture fraction is a conserved scalar ranging from 0 (pure oxidizer) to 1 (pure fuel) that tracks local composition in non-premixed combustion, allowing species and temperature to be mapped from a flame library.

When performing uncertainty quantification (UQ) in CFD using Monte Carlo sampling, the primary reason for the large number of required simulations is:

Answer: Statistical convergence of output statistics requires many independent input samples

Monte Carlo UQ propagates input uncertainties by running many independent CFD simulations with sampled input parameters; statistical convergence of mean and variance estimates requires a large ensemble (often thousands of samples).