Estimation of Coefficient of Pressure in High Rise Buildings Using Artificial Neural Network

Tendency to build more slender and more flexible tall buildings have made these structures susceptible to action of wind. Therefore, wind force is one of the prime considerations in design of tall buildings. The prediction of wind-induced pressure coefficients on the surface of the buildings is of considerable practical importance. Wind tunnel testing is one of the main methods for wind load determination on structures. But this being, time consuming and costly wind tunnel tests can only cover a limited number of cases. The present work focuses on the application of artificial neural networks (ANNs) to estimate pressure coefficients on surface of tall buildings. In the present study, two cases of training data set (consisting of geometrical coordinates of pressure points and angle at which wind strikes at the face of the building as the input to the network) has been used to predict the windinduced `pressure coefficients Cp (mean) (output of the network) for the previously any wind incident angle. The performance of the network is assessed in terms of Root Mean Square Error (RMSE) and Correlation Coefficient R. From the present study, it is concluded that the value of Cp (Mean) goes on decreasing with increase in Wind Incidence Angle for the same pressure point. Also, suction effect is noticed near the corners of the building.