In this study, the influence of different roof pitches on the mean pressure distributions around isolated buildings subject to different wind directions were investigated with wind tunnel tests, 3D steady Reynolds-averaged Navier-Stokes (RANS) simulation and large eddy simulation (LES). Wind tunnel experiments were conducted to obtain mean pressure distributions around buildings having three commonly used roof pitches, namely, 1:5, 2:5 and 3:5. The critical high suction areas created by conical vortices were identified, and the influence of roof pitch on these critical areas with high localised pressures was investigated under various wind directions. In addition, computational fluid dynamics (CFD) analysis was performed, and the performance of RANS and LES were evaluated and compared with the results measured from the wind tunnel tests by considering both the accuracy of results and computational cost. The assessment on RANS and LES was used to determine the best guidelines for the flow problems involved in this study.Results from the RANS show good agreement with the experimental results on pressure distribution when the building is subject to perpendicular wind directions. However, a significant improvement was found using LES over RANS in the prediction of near-building flow field and localised pressure under oblique wind directions but with an increased computational cost by a factor of more than 80. Furthermore, the high suction pressures are observed to be more critical on buildings with a lower roof pitch under both perpendicular and oblique wind directions. This indicates that a low roof pitch should be applied with caution especially in windstorm-prone areas.
Read full abstract