Shear wall layout optimization strategy for high-rise buildings based on conceptual design and data-driven tabu search

Abstract

The scope of this paper is to present a novel design methodology that relies on the tabu search (TS) algorithm and aims to optimize the shear wall layout of high-rise buildings. By discretizing the distributable area of a shear wall into wall elements, different possible shear wall layouts can be described by binary code where ‘0’ and ‘1’ indicate the absence or presence of a wall element respectively. Furthermore, the conceptual design is combined with the optimization algorithm to play a guiding role in the optimization, and the surrogate model is built through support vector machine (SVM) to drive the optimization process with historical data and reduce the cost of the real model’s evaluation. Then, the structural weight can be automatically minimized with constraints on the story drift and period ratio and an optimization program with an independent interface is compiled in this paper. This method was used to optimize the layouts of different shear wall structures, which proved the feasibility and effectiveness.