Informing Code-Compliant Site-Specific Infrastructure Seismic Evaluations With Physics-Based Simulated Ground Motions
Loading...
Authors
Matinrad, Pezhman
Issue Date
2024
Type
Dissertation
Language
Keywords
Alternative Title
Abstract
Site-specific code-compliant approaches for seismic design require the use of a suite of ground motions that are selected and scaled to a target spectrum for performing nonlinear time-history analyses. Suitable sets of ground-motion records shall be selected by appropriately considering the earthquake parameters controlling the hazard at the site, including magnitude, fault distance, tectonic regime, and impulsive character. When tasked with the design at sites residing in the vicinity of a major active fault with site conditions not well represented in the existing catalogs of records, engineers face the challenge of not having a sufficient number of motions. This has led design codes, including ASCE/SEI 7, to contemplate the possibility of supplementing the existing database of records with simulated ground motions. Physics-based ground-motion simulations that incorporate the characteristics of fault rupture, geological structure, and topography are therefore becoming a fundamental resource to support structural design and advance understanding of seismic risk. However, technical guidance on their correct use in engineering applications is yet to be defined. This research provides the technical basis to inform the utilization of simulated motions in code-compliant structural design procedures, with a focus on two aspects: the capability of simulations to enable 'true' site-specific structural assessments as compared to approaches relying on catalogs of historical records, and the implications of different methods for modeling soft sediments on the predicted structural responses (simulation-based vs semi-empirical). Results show that utilizing site-specific simulated ground motions that incorporate path, fault geometry, and site-condition effects as opposed to historical records in code-compliant approaches may lead to differences in the structural demands above a factor of 1.5. Such differences are highly spatially variable and difficult to predict. It is also demonstrated that the utilization of hybrid methods combining simulations and empirical factors may lead to significant misestimates of structural responses, requiring the implementation of processing methods specific to the geological characteristics of the domain of interest. Finally, evidence from these analyses is collectively utilized to develop a method for the selection of simulated motions targeting component-specific spectral amplitudes and variability at the site of interest. The analyses and findings of this work are demonstrated utilizing two and three-dimensional archetypal reinforced concrete buildings of different heights.