SAE Expert Insights: Computational aspects of Automotive Aerodynamics: Current Trends and Future Challenges
Advances in high performance computing (HPC) continue to transform computational aerodynamics from a specialist tool into a routine part of the automotive design workflow. With aerodynamic drag accounting for a significant portion of an electric vehicle's energy consumption at highway speeds, it is imperative that computational fluid dynamics (CFD) simulations provide rapid and accurate guidance, helping shape design decisions early. This session surveys various practical aspects of aerodynamic CFD tools in industrial use today — from everyday RANS solvers through scale-resolving DES and LES approaches — examining what each method gets right, where it fails, and at what computational cost. The disruptive role played by computational accelerators like GPUs and machine learning will also be examined, along with trends that might shape future developments. Nevertheless, stringent accuracy and consistency requirements continue to pose significant physics modeling challenges, as illustrated by the gap between simulation results and experiments; ongoing efforts to close this gap will be discussed. Beyond passenger vehicles, distinct difficulties associated with heavy commercial and racing vehicle simulations will also be highlighted. Finally, we will close with some emerging and unresolved challenges, and chart a possible roadmap for the coming decade.
presented by Parvez Sukheswalla