Modelling wind-induced interference effects using backpropagation neural networks

The modification of wind loads on a building due to adjacent buildings known as interference, can be quite significant. Interference effects depend mainly on the geometry and arrangement of adjacent buildings, their orientation with respect to the direction of flow and the upstream terrain conditions. Data collected through the literature on wind-induced interference effects on buildings has been analysed and compared in order to identify common points of agreement and areas of concern. Given the limited and sometimes inconsistent available data coupled with a large number of variables involved, empirical generalizations are difficult to obtain. In order to overcome these limitations the paper suggests a neural network approach for the assessment of wind-induced interference effects on design loads for buildings.