| Fragility functions are one of the key technical ingredients in seismic risk assessment. The derivation of fragility functions has been extensively studied in the past; however, large uncertainties still exist, mainly due to limited collaboration between the interdependent components involved in the course of fragility estimation. This research aims to develop a systematic Bayesian-based framework to estimate high-fidelity fragility functions by integrating monitoring, modeling, and hybrid simulation, with the final goal of improving the accuracy of seismic risk assessment to support both pre- and post-disaster decision-making. In particular, this research addresses the following five aspects of the problem: (1) monitoring with wireless smart sensor networks to facilitate efficient and accurate pre- and post-disaster data collection, (2) new modeling techniques including innovative system identification strategies and model updating to enable accurate structural modeling, (3) hybrid simulation as an advanced numericalexperimental simulation tool to generate highly realistic and accurate response data for structures subject to earthquakes, (4) Bayesian-updating as a systematic way of incorporating hybrid simulation data to generate composite fragility functions with higher fidelity, and 5) the implementation of an integrated fragility analysis approach as a part of a seismic risk assessment framework. This research not only delivers an extensible and scalable framework for highfidelity fragility analysis and reliable seismic risk assessment, but also provides advances in wireless smart sensor networks, system identification, and pseudo-dynamic testing in civil engineering applications. |
