| Summary: | This paper presents the results of an extensive development and validation process to produce a highly accurate and reproducible external aerodynamics simulation methodology. The approach makes use of a specialized volume meshing methodology with controllable cell quality metrics and an incompressible Delayed Detached Eddy Simulation (DDES) flow solver to consistently predict the aerodynamic forces on a range of different vehicle shapes. In addition a wizard-like graphical user interface (GUI) has been developed that streamlines the set-up process and applies vehicle dependent best practices derived during the validation process. Several measures have been taken to improve the robustness, accuracy and speed of the flow solver. These include solution adaptive discretization, improved DDES turbulence modelling and various mesh generation developments to enhance the fidelity of the computational grid. Significantly improved solution times have also resulted from a combination of variable time-step sizes and extensive optimization of solver settings for the types of grids employed. Over 100 vehicle types and permutations have been used for the validation and best-practice development. In each case wind tunnel data is compared with CFD results in an iterative and recursive fashion, as best-practices evolve. Vehicle shapes used in the validation include: sedans, hatchbacks, estates and SUVs, motorsport applications, as well as both light and heavy duty trucks. In addition to different vehicle shapes a wide range of experimental configurations were evaluated including different ground simulation techniques, yaw angles, ride heights and test speeds. Although the focus was on improving absolute accuracy, variations in drag for different swappable parts and configurations were also checked for directionality and magnitude.
|
|---|