JuliaHub Case Study
Williams Grand Prix Technologies and JuliaHub
Published
01 Oct 2024
Est Reading Time
3 min
In the high-stakes world of Formula One racing, every advantage counts. Teams rely on cutting-edge technologies to push the boundaries of performance both on and off the track. One of the most critical tools in this pursuit is the use of simulation to understand the performance of the car and make strategic decisions during races. JuliaSim is a cloud platform purpose-built for fast simulations, and it has become a valuable tool for the Williams Racing Team.
The JuliaHub partnership with Williams opens up the opportunity for Williams Grand Prix Technologies clients to tap into the joint capability offering, to help solve engineering problems across industries.
As a case study example, here is what the JuliaHub partnership delivers to the race team.
Hardware Accelerated Aeromap Modeling
Formula One teams benefit from capturing pressure measurements all around their cars to understand aerodynamics at different attitudes and positions around the track. Williams Racing uses predictive modeling methods based on these pressure measurements. By utilizing advanced simulation techniques offered by Julia’s high-level GPU programming features, Williams Racing was able to significantly improve the performance of its predictive models. The new model runs over 169 times faster and delivers 7% more accuracy, giving the team a substantial edge in analysing aerodynamic data.
This enhanced model is deployed on JuliaHub and is integrated with in-house tools.
Model Comparison for Speed and Accuracy with JuliaSim and MATLAB | JuliaHub
Digital Twin Replaces Physical Sensor
Williams Racing also employed JuliaSim to create a digital twin for a physical sensor, which provides valuable in-lap insights without the negative impact of additional weight and reduced aerodynamics that come with running a race with the physical sensor. In the past, Williams Racing tackled this problem using classic machine learning techniques. However, JuliaSim reimagined and improved the approach by implementing Scientific Machine Learning (SciML) techniques. The JuliaSim neural network uses a special architecture[1] with two major advantages: it captures high-frequency features commonly found in vehicle control inputs, and it incorporates known physical relationships that model the vehicle's motion. This architecture introduces an ability to learn relevant mathematical relationships while maintaining its data- driven nature of learning the missing relationships not known a priori. The resulting digital twin produces faster and more accurate results than the pure Machine Learning (ML) model. JuliaSim deployed the model as an FMU (Functional Mock-up Unit) for integration with standard modelling tools.
Faster Simulations on More Complex Geometry
Tires in Formula One are the great equalizer. Every team uses the same tires and knowledge around them is protected in order to maintain a fair playing field. Races are often won or lost based on a team's decision to change their vehicle's tires. This is a strategic decision that depends heavily on weather and track conditions. For this reason, Williams Racing requires fast, reliable simulations to model tire deformation under various conditions. Using JuliaSim’s capabilities, the team achieved over 1000x speed improvements in their simulations for the quasi-static partial differential equations (PDE) and 8x speedup for the dynamic PDE. Both results were achieved on a geometry that was 2.3x higher fidelity than the mesh previously possible.
Conclusion
Williams Racing was able to improve three areas of its engineering process: aeromap modeling, speed over ground sensor, and tire deformation. JuliaHub was able to deliver these improvements thanks to its cloud-native capabilities and JuliaSim's advanced offerings for Scientific Machine Learning.
If you feel that these capabilities could be used to help solve your engineering challenges, please do get in touch with the team here at Williams Grand Prix Technologies