Performance-based topology optimization for wind-excited tall buildings: A framework

Topology optimization methods were originally developed in a deterministic setting notwithstanding the inherently uncertain environment in which the final structural systems must exist. Indeed, most design applications are affected by uncertainties in material properties, model idealization etc. as well as loads that are inherently aleatory in nature. In addition, the responses of the systems of interest to this study are generally affected by a significant amount of dynamic amplification therefore complicating the governing equations. While the performance-based assessment and optimization of fixed-topology structures set in the aforementioned environment has been the focus of a number of studies, the possibility of performing topology optimization within this setting has yet to be fully investigated. aper presents a performance-based topology optimization framework developed for wind-sensitive tall buildings that rigorously accounts for the time-variant stochastic nature of the aerodynamic loads while considering additional time-invariant uncertainties describing the state/knowledge of the system parameters defining the mechanical properties of the structure. The framework is based on decoupling the performance-based assessment from the topology optimization problem through the definition of a number of approximate sub-problems. The successive resolution of these results in a final design that is not only optimum but also satisfies the user-defined performance targets.