Author:

Abstract

Significant improvements of computer resources in recent past years allow to use computational fluid dynamics (CFD) methods as an alternative test method for wind tunnel tests in various wind engineering aspects. However the accuracy of CFD simulation mainly depends on careful setup of three main components of a CFD simulation, which are domain size, adopted mesh, and boundary conditions. This paper presents basic theoretical background of use of CFD for atmospheric boundary layer simulations and proper methods recommended for creating domains and meshes for CFD models. It also demonstrates several empirical methods that can be used as boundary conditions in the absence of more accurate data for simulation. The CFD simulation results of pressure distribution of 112m tall buildings is compared with the wind tunnel test results and found that performance of those empirical methods is satisfactory. The use of CFD simulation for flow visualization around a tall building and evaluating pedestrian level wind velocities are also demonstrated in this paper.

DOI: http://dx.doi.org/10.4038/engineer.v46i3.6784

ENGINEER, Vol. 46, No.03, pp. 43-54 2013

Keywords: CFD, Atmospheric boundary layer, RANS equation, Standard k- ε model