Coupled Sliding Grids and Adaptive Grid Refinement for RANS Simulation of an Oscillating Car Model

Emmanuel Guilmineau (1) , Patrick Queutey (1) , Alban Leroyer (1) , Jeroen Wackers (1)

1 LHEEA - Laboratoire de recherche en Hydrodynamique, Énergétique et Environnement Atmosphérique

Ground vehicles travelling on the road are often subjected to unsteady flows. The yaw angle increases with stronger wind conditions such as gusty cross winds caused by unsteady atmospheric conditions, tunnels, bridges or trucks passing. Separations appear on the leeward side of the body, interact with the base flow and cause unsteady vehicle loading. These aerodynamic elements, added to dynamic properties of the suspension, springing and tires are finally the factors of its dynamic stability and of the safe manoeuvrability appreciated by the driver. With an unsteady separation, it is difficult to have a mesh adapted to the flow. For improved accuracy, automatic adaptive grid refinement is a technique for optimising the grid in the simulation of fluid flow, by adapting the grid to the flow. This is done by locally dividing cells into smaller cells, or if necessary, by merging small cells backs into larger cells in order to undo earlier refinement. The present work aims to explore the flow physics in which the vehicle is exposed to steady wind, but with continuous changes in the yaw angle. For this purpose, the unsteady numerical technique with a sliding grid approach is used in the flow solver ISIS-CFD, developed by the Laboratory of research in Hydrodynamics, Energetics, and Atmospheric Environment, ex-Fluids Mechanics Laboratory of the Ecole Centrale de Nantes. The CFD simulation is carried out with the Explicit Algebraic Reynolds Stress Model (EARSM) turbulence model. The numerical simulations are performed on the square-back Willy model. The Reynolds number is Re = 900,000 and the yaw motion of the model is close to a sinusoidal motion with an amplitude of oscillation Δβ = 10° and a frequency f = 2Hz. The numerical solution is compared with numerical results obtained in static positions. The pressure on the model is also compared with experimental data.

Mots clés: Simulation; Reynolds-averaged Navier–Stokes equations

Domaines: Mécanique des fluides [physics.class-ph]

Emmanuel Guilmineau, Patrick Queutey, Alban Leroyer, Jeroen Wackers. Coupled Sliding Grids and Adaptive Grid Refinement for RANS Simulation of an Oscillating Car Model. ASME 2012, Fluids Engeneering Summer Meeting, FEDSM2012, Jul 2012, Rio Grande, Puerto Rico. ⟨10.1115/FEDSM2012-72081⟩. ⟨hal-01158808⟩