How do you validate and verify your CFD results against experimental data or other sources of uncertainty?

1 Validation vs verification

Validation and verification are two related but distinct concepts in CFD. Validation is the process of comparing your CFD results with experimental data or other reliable references, such as analytical solutions, wind tunnel tests, or flight data. Verification is the process of checking the accuracy and quality of your CFD model, such as the mesh, the boundary conditions, the numerical schemes, and the convergence criteria. Both validation and verification are essential to ensure your CFD results are credible and useful for aircraft design.

2 Sources of uncertainty

There are many sources of uncertainty that can affect your CFD results, such as the model assumptions, the physical parameters, the numerical errors, the experimental errors, and the human errors. You need to identify and quantify these sources of uncertainty and evaluate their impact on your CFD results. You can use various methods to estimate the uncertainty, such as sensitivity analysis, error propagation, uncertainty quantification, and uncertainty reduction.

3 Validation methods

Validation of CFD results can be done in different ways, depending on the type and access to experimental data or other references. Dimensionless analysis, for instance, involves using dimensionless numbers such as Reynolds number, Mach number, lift coefficient, drag coefficient, etc., to compare CFD results with similar flow conditions and geometry from experiments or other sources. Graphical analysis is another method which involves plotting CFD results and experimental data or other references on the same graph and using error bars, confidence intervals, or statistical measures like correlation coefficient, mean absolute error, or root mean square error to quantify the difference. Lastly, qualitative analysis involves comparing the flow features such as vortices, shocks, separation etc. between CFD results and experiments or other references by looking at contour plots, streamlines, pressure distributions etc., as well as using flow visualization techniques such as smoke, oil or dye to observe the flow patterns in experiments.

4 Verification methods

Verifying your CFD model depends on the type and complexity of the problem, and there are several methods to do so. For example, you can check the correctness and consistency of your CFD code by performing simple test cases and comparing results with analytical solutions or other codes. Additionally, you can assess the sensitivity and convergence of your CFD results to the mesh size and quality by conducting a grid convergence study. Lastly, you can check the accuracy and stability of your CFD results to numerical schemes and parameters by performing a solution verification study. This involves using different numerical schemes and parameters, such as upwind, central, or hybrid, and comparing the CFD results using verification indices like discretization error or round-off error.