Jan Östh, ÅF Industry
Division of Fluid Dynamics
Due to increasing demands on the performance of electric vehicles (EVs), there is a desire to investigate the impact of tough situations, such as wading, on the vehicle. As most previous wading studies are concerned with conventional internal combustion engine vehicles (ICEVs), automotive manufacturers are striving to expand their knowledge about wading with EVs. Wading is mainly examined through physical tests, but using computational fluid dynamics (CFD) for studying vehicle wading can provide a more rigorous examination as well as reduce cost and time required. One way to further decrease the cost is to use open source software.
In this study, a method for investigating wading with the open source CFD software OpenFOAM is developed. The generic DrivAer model is used, however modified to resemble a model of an EV. The computer aided engineering (CAE) software ANSA is used for geometry cleanup and hexahedral volume meshing. The numerical method uses the volume of fluids (VOF) method for multiphase modeling and large eddy simulation (LES) for turbulence modeling. The study focuses on the water surface topology, water velocity, water flow through the engine compartment and forces acting on the car. Due to stability issues results are only obtained with a simplified vehicle geometry and not for the EV model. Validation is done against a similar case in the commercial software STAR-CCM+. It is found that the results obtained for the water surface and velocity with the OpenFOAM method are similar to those obtained with STAR-CCM+, but that the lift forces on the car are overpredicted by OpenFOAM. To further investigate the effect of wading on EVs, simulations with a more detailed EV geometry are conducted using STAR-CCM+. Results show that splashing is absent at low speed deep water wading, but that engine bay components are still exposed to large amounts of water. It is also found that the water pressure exerted on the battery pack and engine bay components at this velocity is insignificant. It is concluded that the OpenFOAM method shows promising results, but further investigation of the solver settings is required in order to increase the stability when using complex geometries. Regarding wading with EVs, it is concluded that the impact of low speed wading on the vehicle is small however as the water ingress in the engine bay compartment is significant, the components here must be watertight or somehow protected. Keywords: Electric vehicles, Wading, CFD, OpenFOAM, DrivAer, VOF, LES