Framework to Explore the Design Space for Design of Tall Buildings

Abstract

Design of tall buildings is undergoing a resurgence that is driven by a variety of factors – economical growth, scarcity of land in urban areas, high land costs, increased population density, technological advancements and man’s desire to build taller structures. Considerable research work has been done in the last two decades to meet this demand. Computer-based tools that help design engineers explore design alternatives are indispensable in tackling this complex problem. In addition, a framework that finds the near optimal design, adds value to this exploratory work. In this paper, we develop a general framework for the design optimization of buildings using sizing, shape, and topology design variables. Sizing optimization can be carried out using discrete design variables (from a database of available sections) or continuous design variables (cross-sectional dimensions of custom wide flange sections). Similarly, shape optimization can be carried out using either discrete or continuous design variables. And finally, topology optimization can be carried out using boolean design variables. Allowable stress design guidelines are used as constraints along with displacement, inter-story drift, total structural weight, and frequency constraints. The finite element model is made of three-dimensional beam elements. A typical function evaluation involves a linear, static analysis with multiple load cases, a linear, modal analysis to extract the lowest few eigenpairs, and a linear, buckling analysis to find the buckling capacity. An optimization toolbox that contains gradient-based and population-based optimizers, is a part of the framework. Numerical results how that the framework is capable of producing efficient designs effectively