Multi-objective optimization of inter-story isolated buildings using metaheuristic and derivative-free algorithms: Fid-Bau Portal

Multi-objective optimization of inter-story isolated buildings using metaheuristic and derivative-free algorithms

Skandalos, Konstantinos / Afshari, Hamid / Hare, Warren / Tesfamariam, Solomon

Abstract Inter-story isolation (ISI) is a passive control technique that involves the placement of seismic isolation devices between stories. Multiple isolation layers installed at different story levels can reduce the response of the isolation devices, while maintaining control of the primary building. A multi-objective optimization (MOO) study is conducted in this paper to select the optimum stories of installation and the effective stiffness and damping properties of specified number of isolation layers. The maximum standard deviations of inter-story drift ratio (ISDR) and isolation drift, derived from random vibration analysis (RVA) of a shear-building model are used as the objective functions of the optimization problem, thus accounting for the seismic responses of both the primary building and the isolation devices. To solve the resulting mixed-integer problem, the combinations of integer variables are fully enumerated and a continuous-variable sub-problem is solved for each combination. Four metaheuristic and one derivative-free algorithms are comparatively assessed for solving the continuous-variable multi-objective sub-problems corresponding to the combinations of the integer variables. Along with the optimal trade-offs between the response objectives, multi-objective optimization reveals trends in response with varying isolation configuration as well as the enhancement of the ISI building performance with increasing number of isolation layers.

Highlights • Multi-objective optimization (MOO) of inter-story isolated buildings. • Consideration of multiple isolation layers at different stories of multi-story buildings as means to reduce isolation drift. • Comparative assessment of four metaheuristic and one derivative-free MOO algorithms Identification of optimal stories of installation and for specified number of isolation layers. • Performance enhancement assessment with increasing number of isolation layers for the case study building model.