WIND PRESSURE COEFFICIENTS PREDICTION ON DIFFERENT SPAN TO HEIGHT RATIOS DOMES USING ARTIFICIAL NEURAL NETWORKS

Large spans always fascinated architects and engineers. Domes provide an easy and economic method of roofing large areas with minimum material in all forms of space structures. Wind loads have significant proportion of the total load to act on such structures that’s why the magnitude and distribution of the resultant pressures must be considered. To overcome this problem, the concept of Artificial Neural Network is adopted to find wind induced pressure coefficients for spherical domes of different span/height ratio. This paper aims to use this neural network application in steel space structures. Here, pressure measurements had been made on a large dome roof model with Computational Fluid Dynamics (CFD) technique and the data generated were used as the training sets to develop artificial neural network models to recognize the input–output patterns.