Title
Detached Eddy Simulation applied to three-dimensional aerodynamic full car simulations in motorsport
Author
de Jong, Daniëlle (TU Delft Aerospace Engineering)
Contributor
van Zuijlen, Alexander (mentor)
Dutzler, Gerhard (mentor)
Degree granting institution
Delft University of Technology
Programme
Aerospace Engineering | Aerodynamics
Date
2019-11-15
Abstract
Detached Eddy Simulation is investigated as a potential method to improve the design process in the Aerodynamic Prototype Development Department of Porsche Motorsport. Detached Eddy Simulation is a CFD method that combines the scale resolving capabilities of Large Eddy Simulation with the low cost modelling of the Reynolds-Averaged Navier-Stokes (RANS) equations. The method promises to capture unsteady flow phenomena better than RANS with only a limited increase in costs. Various DES variants exist, which use different methods to switch between the RANS and the LES regions in the flow domain. The main method used is the Improved Delayed Detached Eddy Simulation (IDDES) method. This is a combination of the Delayed Detached Eddy Simulation (DDES) method and a Wall-Modelled LES (WMLES) method. The DDES method shields the full attached boundary layer from LES and models it with RANS. The WMLES method does the opposite, allowing the use of LES within the boundary layer. IDDES chooses the most suitable one of these methods locally, to ensure a high accuracy and a robust method. The performance of the method when simulating limited span airfoil models and complex full car geometries is assessed and improved. A number of parameter variations are done on the airfoil models to improve understanding of the method through less costly simulations. The main application domain however, is motorsport. Three different aerodynamically complex car geometries are simulated with the IDDES method. The performance of the method is assessed by comparison with various RANS results and wind tunnel data in the form of axial force data, pressure taps and PIV data. Some parameter variations on the car are also tested to improve the results and obtain a reliable design process. The IDDES method is able to visualize more flow phenomena and characteristics. Additionally, drag and the impact of ride height changes are predicted well. Downforce is unfortunately generally overpredicted. There are indications that the results might be improved by varying the inflow conditions and by testing with finer time steps and/or meshes. Because of the costs involved and the desire to investigate the method as part of an extensive design process, this was deemed outside the scope to the master thesis. It is however, a recommended step for further exploration of the topic.
Subject
CFD
Detached Eddy Simulation
Motorsport
Full Car Simulations
DES
IDDES
To reference this document use:
http://resolver.tudelft.nl/uuid:1b6bdccb-5275-4d2c-baeb-d7a5d423b1e0
Embargo date
2024-11-15
Part of collection
Student theses
Document type
master thesis
Rights
© 2019 Daniëlle de Jong