How do you communicate and justify your CFD validation decisions to your clients or stakeholders?

1 Define validation

Validation is the process of comparing your CFD results with experimental or analytical data from a reference source, such as a wind tunnel test, a laboratory experiment, or a theoretical solution. Validation helps you assess the accuracy and uncertainty of your CFD model, as well as identify and correct any errors or limitations. Validation is different from verification, which is the process of checking the correctness and quality of your CFD code and numerical methods.

2 Choose appropriate data

One of the challenges of CFD validation is finding appropriate and reliable data to compare with your simulations. You should choose data that are relevant to your problem domain, have sufficient detail and quality, and are publicly available or accessible. You should also consider the sources and methods of data collection, as well as the uncertainties and errors associated with them. You should document and cite the data sources and methods clearly and transparently in your reports and presentations.

3 Perform sensitivity analysis

Another challenge of CFD validation is determining how sensitive your results are to the variations and uncertainties in your input parameters, such as boundary conditions, initial conditions, material properties, and turbulence models. You should perform sensitivity analysis to quantify and rank the effects of these parameters on your output variables, such as lift, drag, pressure, and temperature. Sensitivity analysis can help you justify your parameter choices, optimize your design, and reduce the computational cost and time of your simulations.

4 Use validation metrics

To communicate and justify your validation decisions, you need to use quantitative and qualitative metrics to measure the agreement between your CFD results and the reference data. Coefficient of determination (R2) and normalized root mean square error (NRMSE) are two common metrics. Additionally, scatter plots, histograms, contour plots, and streamline plots are graphical tools that can be used to visualize the distribution, correlation, spatial patterns, and trends of the reference data and CFD data. It's important to choose the most suitable metrics for your problem and objective, and report them along with their confidence intervals and error bars.

5 Explain the discrepancies

No matter how well you validate your CFD model, there will always be some discrepancies between your results and the reference data. These discrepancies can be attributed to model simplifications and assumptions, numerical or discretization errors, experimental or measurement errors, and uncertainties or variabilities in the input parameters and output variables. It is important to explain the sources and causes of these discrepancies, as well as how they affect your conclusions and recommendations. Additionally, you should suggest possible ways to reduce or eliminate these discrepancies, such as refining your mesh, improving your boundary conditions, or conducting more experiments.

6 Address the feedback

The final step of communicating and justifying your validation decisions is to address the feedback and questions from your clients or stakeholders. You should listen to their concerns and expectations, and respond to them in a clear and respectful manner. You should also acknowledge the limitations and uncertainties of your CFD model, and avoid making claims or promises that are not supported by your validation data. You should also provide them with the relevant documentation and references that support your validation decisions.