Seismic vulnerability and loss assessment of an isolated simply-supported highway bridge retrofitted with optimized superelastic shape memory alloy cable restrainers

Restrainers, being of relatively low cost and easy to install, are often used to prevent unseating of bridge spans. The potential of using superelastic shape memory alloy (SMA) restrainers in preventing such failure has been discussed in the literature; however, the impact of such smart restrainers with optimized configurations in reducing the failure probability of bridge components and system as well as the long-term economic losses given different earthquake scenarios has not been investigated yet. This study presents a probabilistic seismic fragility and long-term performance assessment on isolated multi-span simply-supported bridges retrofitted with optimized SMA restrainers. First, SMA restrainers are designed following the displacement-based approach and their configuration is optimized. Then, seismic fragility assessment is conducted for the bridge retrofitted with optimized SMA restrainers and compared with those of the original bridge and the bridges with elastic restrainers (steel and CFRP). Finally, long-term seismic loss (both direct and indirect) are evaluated to assess the performance of the retrofitted bridges in a life-cycle context. Results showed that among three considered restrainers, SMA restrainers make the bridge less fragile and help the system lower long-term seismic loss. The design event (DE, 2475-year return period) specified in Canadian Highway Bridge Design Code (CHBDC, CSA S6-14 2014) may underestimate the long-term seismic losses of the highway bridges. Under DE, the damage probability of the bridge retrofitted with optimized SMA restrainers experiencing collapse damage is only 0.7%. Under the same situation, its expected long-term loss is approximate 17.6% of that with respect to the unretrofitted bridge.