The last three articles discussed how wind tunnels and CFD are used as the main predictive tools for aerodynamic performance in motorsport. In today’s article, we will discuss the growing role of data-driven methodologies in this process.
Data-driven processes have been used in motorsport for many decades. For example, teams have acquired data from their race cars, chassis and power units. They have been tweaking computational models based on experimental tests, judging component fatigue based on measured stress cycles, measuring aerodynamic loads in the wind tunnel and converting those into aero maps, and analysing weather forecasts for the race weekends. In one way or another, these processes use data to attempt to predict certain variables in the future.
In other words, data-driven engineering is not new. However, the quantity of data generated has been growing exponentially, and consequently, the usefulness of data-driven processes improved considerably over the last decade. Many fascinating AI technologies developed are now being deployed or researched as part of the aerodynamic development process. Examples include:
- data from experimental measurements being fed into the CFD turbulence modelling, returning increased solution accuracy at a lower simulation cost;
- data from CFD used in the track environment for surface pressure reconstruction;
- Historical CFD data used to generate models predict the flow around new geometries in milliseconds.
These are just a few examples; many more exciting applications are currently being developed to use the vast amounts of data generated. We expect to see substantial changes in the way the race car will be designed aerodynamically over the next decade, with the role of wind tunnels and CFD becoming much more embedded into data-driven processes.
Sabe is actively researching and experimenting with AI-based methodologies applied to aerodynamics. Get in touch if you would like to discuss our work in more detail or would like to discuss your application.
In our next article, we will discuss cooling and how its requirements affect aerodynamic performance.