Aerodynamic Performance of a Baja SAE Vehicle Using Hybrid RANS-LES Approach.

This study presents a high-fidelity computational investigation of the aerodynamic performance of a Baja SAE off-road vehicle using a hybrid Reynolds-Averaged Navier-Stokes (RANS) and Large Eddy Simulation (LES) turbulence modeling approach. The methodology combines the Spalart-Allmaras RANS model for near-wall flow treatment with Detached Eddy Simulation (DES) for resolving large-scale unsteady turbulent structures in the vehicle's wake. A detailed computational domain and refined meshing strategy were implemented using ANSYS Fluent, including mesh adaptation based on the LES filter size (Δ = 23.5 mm) and mesh independence validation. Simulations were performed under steady (RANS) and unsteady (DES) conditions at 30 km/h, yielding a drag coefficient (Cd) of 1.290 for RANS and 1.249 for DES. While RANS provided stable results with low variance (σ = 0.005), the DES model captured transient phenomena such as vortex shedding and near-wake recirculation with higher accuracy (σ = 0.024), enhancing the prediction of flow separation zones and aerodynamic forces. In this work, the analysis focuses on the drag coefficient as a primary indicator of wake-induced aerodynamic losses, rather than providing a complete aerodynamic force and moment characterization. Pressure and velocity field analyses revealed improved resolution of stagnation zones and vortex dynamics under the DES framework, particularly around the roof, rear, and underbody regions. The Q-criterion visualization showed that DES allowed the identification of both large-scale and fine-scale vortex structures in the near and far wake, offering a comprehensive representation of turbulence intensity and flow instabilities. These findings confirm the suitability of hybrid RANS-LES methods for aerodynamic optimization of complex vehicle geometries, providing enhanced predictive capabilities compared to traditional steady-state models.