Poster #232, Earthquake Engineering Implementation Interface (EEII)

Critical assessment of probabilistic seismic demand analysis of ordinary bridge structures using Cybershake simulations

Poster Image: 

Poster Presentation

2020 SCEC Annual Meeting, Poster #232, SCEC Contribution #10286
There is a need for benchmarking and validating simulated ground motions for utilization by the engineering community. The validation method presented herein focuses on bridge engineering applications in Southern California. Catalogs of simulated ground motions representing a 200,000-year history are selected from the CyberShake (ver. 15.12) database for five sites in Southern California (~20,000 unscaled ground motions per site). They are used in Non-Linear Time History Analysis (NLTHA) of four Ordinary Standard Bridge (OSB) structures with first-mode periods between 0.61 and 1.11 sec. For each site, this data is used to obtain simulation-based Engineering Demand Parameter (EDP) hazard, which are then compared against conventional EDP hazard curves constructed using empirical models that are based on recorded ground motions through Incremental Dynamic Analysis (IDA). The simulation-based and conventional EDP hazard curves are compared at various return periods. To further assess differences between simulated and recorded ground motions, comparisons are also made between Intensity Measure (IM) hazard curves and the EDP-IM relations. We observe that, in general, CyberShake (ver. 15.12) simulates acceptable motions and yields similar EDP values to empirical data for shorter return periods. For longer return periods, however, EDPs for shorter-period bridges obtained from the simulation-based analysis tend to be lower than the conventional EDPs obtained from utilizing recorded ground motions. For OSBs with longer periods (e.g., 1.1 sec), EDPs in our simulation-based analysis tend to be higher than those in our conventional EDPs. Further analysis suggests that the difference between EDP-IM relations seen in the simulation-based method and the conventional method is the primary source of EDP hazard curve variations. Thus, it is recommended that validation efforts should go beyond comparisons of IM levels and also include EDP-IM level validation. In this study, we also use regression to develop predictive equations that relate the ratio between the simulation-based EDPs and conventional EDPs to the return period and site characteristics (e.g., Vs30 and Z2.5). Inclusion of site characteristics allows to account for the site-specific differences in the EDP hazard curves. The proposed equations can be used by engineers to scale the EDPs obtained from IDA analysis to the EDPs obtained by conducting NLTHA using the set of CyberShake simulated ground motions.