The roof geometry of a leeward sawtooth roof building can have a large influence on the cross-ventilation flow. In this paper, five different leeward sawtooth roof geometries are evaluated using Computational Fluid Dynamics (CFD). The 3D CFD simulations are performed using the steady Reynolds-Averaged Navier-Stokes approach with the SST k-¿ turbulence model to provide closure to the governing equations. The computational grid is based on a grid-sensitivity analysis and the computational model is successfully validated using PIV measurements for a generic isolated building from literature. The five different roof geometries that are studied include one straight and four curved roofs. The curved roofs can be subdivided in one concave, one hybrid (convex-concave) and two convex roof geometries. It is shown that a straight or convex roof geometry can maximize the underpressure in the wake of the building, where the outlet opening is located, which results in enhanced wind-driven cross-ventilation flow. Analysis of the results shows that for a normal wind incidence angle (0°) the straight and convex leeward sawtooth roof geometries can result in an increase of the volume flow rate by 13.0%, 12.5% and 12.3% respectively compared to a concave roof geometry. Furthermore, the increase of the indoor air velocity can be as high as 90% in the upper part of the occupied zone (at h = 1.7 m above ground level) for convex versus concave roofs.
|Number of pages||13|
|Journal||Journal of Wind Engineering and Industrial Aerodynamics|
|Publication status||Published - 2015|