Multivariate adaptive regression spline approach to the assessment of surface mean pressure coecient on surfaces of C-shaped building

Proper assessment of wind load ensures a durable design of structures under varying wind load conditions. Accurate prediction of surface mean pressure coecient (Cp) on any irregular plan shaped buildings is essential to assessing wind loads and structural design. The main objective of this study is to present an equation in the line of Multivariate Adaptive Regression Spline (MARS) approach using experimental data of Cp. This developed equation can be used satisfactorily for accurate prediction of Cp values on the surfaces of C-shaped buildings. Extensive experimentation was carried out to determine the coecient of pressure over the surfaces of C-shaped building models under varying sizes, corner curvatures, and angles of incidence in a sub-sonic wind tunnel. The predicted values of pressure coecient of dierent C-shaped buildings using the developed model were compared with the equations developed by Swami and Chandra (S&C) and Muehleisen and Patrizi (M&P). The comparison indicated that the proposed MARS model could predict Cp values more accurately than S&C and M&P models on frontal and side surfaces. Further, the model was used to validate the results using the actual building, and Tokyo Polytechnic University (TPU) data were employed to show the applicability of the proposed equation.