AERODYNAMIC INTERFERENCE IN TALL BUILDING

Abstract

With the increase in population and land cost in large cities, the construction of tall buildings is becoming increasingly a necessity. Use of higher strength materials, improved capabilities in analysis and design as also in construction technology have lead to the construction of tall slender buildings of lighter density, thus making their aerodynamic stability an important concern. A tall building which is interfered by another in its vicinity will behave differently under wind action as compared to when it is in isolation, due to the modified wind flow around it. In this thesis the response of a typical tall rectangular building having 6:1:1.2 aspect ratio has been studied both experimentally as well as using theoretical approaches. An aeroelastic model of this building was designed and fabricated and its along-wind as well as across-wind responses have been measured. The building model has been tested in a 2 m x 2 m boundary layer wind tunnel for two orientations of the model, i.e., shorter and longer side facing the wind. Extensive measurements were made to determine the wake boundary behind the building model as well as to observe the change in turbulence intensity on the upstream and downstream sides of the model. For the determination of the extent of interference, initially, two rigid models of the same size and height were placed in a simulated open terrain in the wind tunnel and forces as well as moments were measured for the buildings arranged in tandem and offset positions in plan. Next an aeroelastic model of the building was tested in the wind tunnel with a rigid interfering model placed at a number of locations upstream as well as downstream of the aeroelastic building model. Testing was carried (i) out in a simulated flow expected over a built-up approach terrain. Responses of the aerolastic model were recorded in the two principal directions, i.e., along-wind and across-wind by a set of strain gauge transducers developed to measure the displacements at the top of the model. It was observed that due to the presence of an interfering building, the mean response of the building generally reduced on account of shielding, while the dynamic response of the building usually increased. The effect of interference is more pronounced when an interfering building is located on the upstream side than if it is situated on the downstream side. For theoretical analysis, Davenport's approach [29] was adopted to estimate the along-wind response of the isolated building. The results computed by using this method agree closely with those predicted from experimental observations. The across-wind response of the building has been computed using force spectra obtained by Saunders [108] for rectangular [1:1.5 and 1:2 aspect ratio in plan ] and square buildings as well as Kareem [64] for square buildings. According to the study made in this thesis, Saunders force spectra for square building may be used for estimation of across-wind response of buildings with 1:1.2 aspect ratio in plan. Kareem's spectra yield results not far from those of Saunders. (