Optimal distribution of friction dampers to improve the seismic performance of steel moment resisting frames

Abstract

This study aims to develop a low computational cost framework to optimize the seismic performance of the steel moment-resisting frames (SMRFs) equipped with friction dampers at different performance levels. To achieve the optimal design of dampers in a structure, a novel approach called adaptive optimisation technique (AOT) is adopted. The basis of this method is to achieve a uniform distribution of damage (UDD) in the structure to exploit the maximum energy dissipation capacity of the dampers. The optimisation objective is to obtain the best position of friction dampers and minimize their slip-threshold force to satisfy a predefined inter-story drift (i.e. selected performance target). The efficiency of the proposed method is demonstrated through optimum design of 4, 8, and 16-story SMRFs subjected to a set of strong natural earthquake records. The unique features of AOT are examined including the high convergence rate, the independence of the optimal solution from an arbitrary starting point, and also the ability to perform multi-level performance optimisation. Finally, the dependency of the seismic optimisation methods to the selected design acceleration record, as a challenge in the practical seismic design process, is addressed by proposing a modification in the proposed adaptive formula. In addition, the results of the proposed method are compared with iterative, non-iterative, and some metaheuristic optimisation methods such as genetic algorithm (GA), particle swarm optimisation algorithm (PSO), and simulated annealing algorithm (SA). It is shown that the AOT can lead to optimum design solutions with significantly less computational costs (up to 98%) compared to the conventional techniques.