Multiobjective Aerodynamic Optimization of Tall Building Openings for Wind-Induced Load Reduction

Abstract

The dependency of wind forces and responses on the outer shape of tall buildings provides a unique opportunity to reduce wind effects through the aerodynamic optimization of the outer shape. The current study presents a multiobjective aerodynamic optimization procedure (AOP) by introducing horizontal openings to reduce wind load. The adopted method joins an optimization algorithm, large eddy simulation (LES), and an artificial neural network (ANN) surrogate model. The base building geometry is the Commonwealth Advisory Aeronautical Research Council (CAARC) building. The present optimization process is carried out by introducing three openings along the height of the building. However, the adopted method is useful for any shape modifications. In all cases, opening volumes are maintained at 1% of the total building volume to enable a simple comparative performance assessment. The objective of the optimization process is to identify the optimal aspect ratio and distances between the openings that produce the best aerodynamic performance. Three optimization problems are presented. The first two follow single-objective optimization to reduce the peak base moment coefficients about the x- and y-directions, respectively. Reductions of 47% and 42% are achieved for the moments about the x- and y-directions, respectively. The third optimization is a multiobjective optimization problem, which aims to simultaneously reduce the two base moments.