Interpolation of pressure coefficients for low-rise buildings of different plan dimensions and roof slopes using artificial neural networks

Abstract

Database-assisted design (DAD) is emerging as an important tool to design buildings for wind effects. However, there is a need for robust interpolation methods for pressure coefficients to extend the range of conditions beyond those in the aerodynamic database from wind tunnel experiments. An interpolation methodology, using artificial neural networks (ANN), was developed to include variable plan dimensions and roof slopes in the set of parameters considered in earlier interpolation studies. In addition to expanding the capabilities for interpolation, the new models improved predictions in the lee of the ridges for gable-roofed and low-rise buildings.