Optimal Design of Nonlinear Viscous Dampers for Protection of Isolated Bridges

Introduction

Applications of seismic isolation techniques to bridges (Figure 1) have gained significant attention over the last decade (Jangid, 2008; Jónsson, Bessason, & Haflidason, 2010; Makris & Zhang, 2004; Perros & Papadimitriou, 2009; Tsopelas, Constantinou, Okamoto, Fujii, & Ozaki, 1996; Wang, Fang, & Zou, 2010). Lead-rubber bearings or friction pendulum systems are selected for this purpose in order to isolate the bridge deck from its support, at the abutments and potentially at the locations of intermediate piers. This configuration provides enhanced capabilities for energy dissipation during earthquake events while also accommodating thermal movements during the life-cycle of operation of the bridge. It is associated though with large displacement for the bridge deck relative to its supports, especially under near fault earthquake ground motions (Dicleli, 2006; Dimitrakopoulos, Makris, & Kappos, 2010; Shen, Tsai, Chang, & Lee, 2004). These motions include frequently a strong, longer period component (pulse) that has important implications for flexible, isolated structures (Bray & Rodriguez-Marek, 2004; Mavroeidis & Papageorgiou, 2003). For base-isolated bridge systems, such large displacements under strong ground motions may lead to (i) large inelastic deformations and formation of plastic hinges at the piers and abutments and (ii) pounding of the deck between adjacent spans or to intermediate seismic stoppers or to the abutments supporting the ends of the bridge (Dimitrakopoulos, Makris, & Kappos, 2009; Dimitrakopoulos et al., 2010; Ruangrassamee & Kawashima, 2003). Such pounding will then lead to high impact stresses and increased shear forces for both the bridge deck and its supports (abutments and piers).

Figure 1.

Two-span base-isolated bridge

This overall behavior associated with excessive vibrations will ultimately lead to significant damages that affect not only the serviceability but also the structural integrity of the bridge system. For controlling such vibrations, application of seismic dampers has been proposed and applied to isolated bridges (Hwang & Tseng, 2005; Makris & Zhang, 2004; Ruangrassamee & Kawashima, 2003). The explicit consideration of the hysteretic behavior of the isolators and the highly nonlinear behavior of the dampers in the design process, as well as the proper modeling of soil-structure interaction at the foundations of the bridge are some of the main challenges encountered in the design of such dampers. Linearization techniques are frequently adopted for modeling the bridge system (Hwang & Tseng, 2005); this simplifies the analysis, but there is great doubt if it can accurately predict the combined effect of the non-linear viscous damping, provided by the dampers, along with the non-linear hysteretic damping, provided by the isolators, while appropriately addressing the soil-structure interaction characteristics and the nonlinearities introduced by pounding effects. Another challenge is the explicit consideration of the variability of future excitations and of the properties of the structural system since a significant degree of sensitivity has been reported between these characteristics and the overall system performance (Dimitrakopoulos et al., 2009; Perros & Papadimitriou, 2009).