Multi-Objective Optimization of Viscous Damper Placement for Building Structures Subjected to Ground Motion

Abstract

Viscous dampers mitigate the interstory drift ratios (IDRs) and peak floor accelerations (PFAs) of buildings subjected to earthquakes. This leads to a multi-objective optimization problem (MOOP) for a viscous damper placement along the building height to minimize IDRs and PFAs simultaneously. This paper proposes innovative methods to address the MOOP and compares those proposed methods to non-dominated sorting algorithm II (NSGA-II) through case studies. Subsequently, the meaning of solutions on the Pareto optimal front in future earthquake events is investigated. The case studies apply each method to a two-dimensional ten-story shear building and adopt four measures to evaluate the performance of searched solutions in multiple aspects. The results show that the proposed methods, by executing fewer number of time history analyses and with convergence comparable to that of NSGA-II, successfully offer improvement against NSGA-II in the aspect of productivity and diversity. As for understanding solutions on the Pareto front in future earthquake events, the knee point solution’s design, which proposed methods can arrive at or approach, successfully reduces both peaks IDR and PFA under 20 ground motions.