Seismic Risk–Based Stochastic Optimal Control of Structures Using Magnetorheological Dampers

Abstract

Magnetorheological (MR) dampers are regarded as among the most promising control devices owing to their perfect dynamic damping behaviors. The operating efficiency of MR dampers, however, upon randomly excited structural systems remains a challenge because the conventional schemes employing linear quadratic Gaussian (LQG) control lack a logical treatment of randomness inherent in external excitations. A scheme of physically based stochastic optimal control designed to bypass the dilemma was proposed in recent years. To this end, in the present paper, a design and optimization procedure for the semi-active control of randomly base-excited structures with MR dampers is developed. Stochastic modeling of seismic ground motions as a result of the source properties and propagation path is carried out. The control efficiency of MR damped structures with respect to seismic risk and variation is investigated. Numerical results reveal that MR damping control can strengthen the seismic safety of structures significantly whether in the case of low or high seismic risk. The MR damping control, meanwhile, has proven robust in accommodating sample variations. In addition, the appropriately designed semi-active controller can achieve almost the same effect as an active controller in a probabilistic sense. Additionally, the MR damper gains a satisfactory performance which behaves as a type-like Bouc-Wen model with strength deterioration, stiffness degradation, and a pinch effect.