SCEC Award Number 15113 View PDF
Proposal Category Collaborative Proposal (Integration and Theory)
Proposal Title Utilization and validation of CyberShake ground motions for the nonlinear performance-assessment of tall buildings
Investigator(s)
Name Organization
Gregory Deierlein Stanford University Ting Lin Marquette University
Other Participants Nenad Bijelic
SCEC Priorities 6e SCEC Groups GMP, EEII
Report Due Date 03/15/2016 Date Report Submitted 05/04/2016
Project Abstract
 This project focuses on the utilization of earthquake ground motion simulations for the design and response assessments of buildings, with the specific goals to (1) help validate results of earthquake simulations through comparative assessments of building performance, and (2) demonstrate the unique advantages that earthquake simulations offer over conventional approaches for hazard characterization and ground motion selection and scaling. We first provide an update to our previous work related to validation of SCEC Broadband Platform motions for analysis of tall buildings, which draws attention to the significance of correlation between spectral acceleration values. Next we perform comparative analysis of seismic demands for sites in Los Angeles basin by: a) using “conventional” approaches relying on recorded motions coupled with probabilistic seismic hazard assessments; and b) by completely relying on physics-based 3D simulations generated as part of the SCEC CyberShake project. In particular, we consider the deep basin STNI site and contrast the results with our previous analysis done for the LADT site. Unlike for LADT, for the STNI site the “conventional” and “CyberShake” approaches yield very different results. Work is ongoing to understand the sources and implications of these differences, but the results do suggest that physics-based simulations can provide novel insights about seismic risk in situations not well constrained by empirical observations. Finally, we also utilize CyberShake motions to examine efficiency of ground motion intensity measures for prediction of collapse. In particular, we performed nonlinear dynamic analyses of a 20-story building for over 400,000 ground motions generated for LADT site and applied machine learning techniques to investigate relationship between engineering demand parameters and input ground motions. While this work is still ongoing, results generally confirm existing understanding of collapse predictors as gained from scaled recorded motions. However, the statistical interrogations of the large collection of hazard- consistent simulated records also demonstrate the utility of different IMs for collapse predictions in a way that is not possible with recorded motions.
Intellectual Merit N/A
Broader Impacts N/A
Exemplary Figure N/A
Linked Publications

Add missing publication or edit citation shown. Enter the SCEC project ID to link publication.