Seismic risk assessment of reinforced masonry structural wall systems using multivariate data analysis

摘要

In contrast to the single design objective associated with force-based design approaches of different seismic force resisting systems (SFRSs), performance-based seismic design (PBSD) allows the selection of more than one design/performance objective. Each performance objective is linked to a tolerable risk level associated with the considered SFRS experiencing a specific damage at a specified seismic hazard level (presented in the form of fragility curves). Similar to other SFRSs, damage of reinforced masonry structural wall (RMSW) SFRSs can be linked to their lateral displacements/drifts. As such, this study focuses on using a multivariate data analysis techniques, to develop a RMSW load-displacement (backbone) model. The backbone model is calibrated using a database of RMSW experimental test results, and the model is further utilized to investigate the influence of the wall geometrical and mechanical characteristics on altering the displacement of RMSW corresponding to different performance levels. These analysis results are subsequently used to develop RMSW seismic fragility bands to facilitate visualizing the risk (losses) associated with the resulting range of displacement (drift) demands, when the different RMSW ranges of characteristics are considered. The developed fragility bands are compared to the individual fragility curves currently adopted by the Federal Emergency Management Agency, FEMA P-58 pre standards, Seismic Performance Assessment of Buildings. This comparison shows that the developed fragility bands significantly deviate from the individual fragility curves currently adopted by FEMA P-58 at each damage state. The developed backbone model and fragility-bands are expected to not only facilitate adoption of the RMSW SFRSs in the next generation of PBSD codes, but also to equip researchers and designers with a clear understanding of the different aspects governing RMSW systems seismic performance and the associated risk. (C) 2017 Published by Elsevier Ltd.