In English

Accurate Prediction of the Flow Around a Truck and the Asymmetry of a Notchback Ahmed Body

Joacim Mattsson
Göteborg : Chalmers tekniska högskola, 2016. Diploma work - Department of Applied Mechanics, Chalmers University of Technology, Göteborg, Sweden, ISSN 1652-8557; 2016:69, 2016.
[Examensarbete på avancerad nivå]

Partially Averaged Navier Stokes is a turbulence model purported to seamlessly shift between RANS and LES solution to achieve an accurate solution in a cost effective manner. The simulations were carried out in the commercial software AVL FIRE. The purpose of this thesis was to show that PANS can be used to achieve results in good comparison to the references. The flow around a generic truck, consisting of to boxes connected by two cylinders, was simulated at a Reynolds number of 510 000 based on the height of the trailer of the truck, using Partially Averaged Navier Stokes. The simulated flow turned steady after some simulated physical time. This was not expected to happen for such a blunt body, neither did it happen in the reference Large Eddy Simulation of the same body. Different numerical schemes, sizes of meshes, boundary conditions, and even workstations were used but all configuration gave these perplexing results. Also the flow around a fully detailed half scale truck from VOLVO AB in a wind tunnel were meant to be simulated at a Reynolds number of 4:5  106. However due to some computational difficulties no meaningful results could be found. These difficulties included problems saving data files, which is theorized to be due to memory running out on the computational cluster. Lastly a notchback Ahmed Body was simulated to try to reproduce the asymmetry of the wake that earlier has been seen in experiments. The flow was simulated at a Reynolds number of 1:9  106 based on the length of the body. No asymmetry of the flow was seen in the current work which contradicts the reference.

Nyckelord: CFD, PANS, Turbulence, Truck, Ahmed, Aerodynamics, VOLVO, AVL

Publikationen registrerades 2016-12-14.

CPL ID: 246127

Detta är en tjänst från Chalmers bibliotek