Comparative Analysis of Turbulence Models for Automotive Aerodynamic Simulation and Design

Abstract

The reduction of energy consumption and environmental impacts of road vehicles is always a major design objective. Many investigations suggest that reducing aerodynamic drags such as pressure drag and skin friction drag is a more efficient method than engine modification in achieving the design objective. Turbulent flow around a bluff body is notoriously difficult to simulate accurately due to the complexity of the flow conditions around the body, such as complex flow separation and laminar to turbulent flow transition. This paper investigates the flow over a benchmark model called the Ahmed body with a slant angle of 25°, which is considered a challenging problem for RANS approach with two-equation turbulence models (Menter, 1994; Serre et al., 2013). Three popular turbulence models, such as the k-ε, k-ω and SST models are evaluated by benchmarking their predictions against experimental data and those of the latest LES solvers. The main purpose of the study is to compare the performances of these models in simulating such a category of flows. In addition, the accuracy and factors that determine the success of such simulation are discussed. Our findings, for the first time, show that with a skillfully generated mesh and proper discretization schemes, the RANS approach with the above two-equation turbulence models are capable of capturing the salient features of the highly complex flow over the Ahmed body with a slant angle of 25°. The performances are as good as the LES solvers as reported in Serre et al. (2013), if not better for time-averaged flow simulations. The SST model produces the best results among the three models studied. This study could assist designers in the automotive industry in the applications of these cost-effective tools to improve their design productivity. Future study will focus on the performances of the models in simulating time-dependent flows over the Ahmed body.