An automated structural health monitoring procedure for post-earthquake damage evaluations considering building-level displacement ductility response

Structural health monitoring (SHM) systems are increasingly used in recent years to: (i) provide real-time monitoring of a building’s condition, and (ii) evaluate building safety following major natural disasters such as earthquakes. While tracking changes in dynamic properties is performed in most applications, it is difficult to accurately relate such changes with the extent of building damage. As an alternative, methods to track the hysteretic response of buildings using floor acceleration data have been developed to explicitly monitor building-level displacement ductility response. However, most methods require the selection of hysteretic and damping models and their initial properties, which is often based on subjective judgement and can influence the method’s accuracy. Recently, a model-free SHM framework to extract the hysteretic backbone curve for multistory buildings using wavelet transform methods to track building-level ductility response has been proposed. This paper details the concepts of this new framework, recent modifications made to allow automation of the framework for easier adoption in engineering practice, and ongoing research to further refine the framework. Validation of this framework using shake-table tests of large-scale reinforced concrete buildings performed at the E-Defense facility in Japan is also presented.