MULTIVARIATE DATA ANALYSIS FOR FORCE-, DISPLACEMENT-, AND PERFORMANCE-BASED SEISMIC DESIGN OF REINFORCED MASONRY SHEAR WALLS
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Over the past decades, seismic analysis and design have been thoroughly addressed
by the majority of international building codes and standards. Although seismic
damage of structural components typically correlate better to displacement, rather
than forces, current seismic codes adopt force-based seismic design (FBSD)
approaches. However, in the past two decades, there has been a gradual shift from
“strength” to “performance” through the introduction of performance-based
seismic deign (PBSD) of structural components that can be implemented with
relative ease through displacement-based seismic design (DBSD) approaches. For
reinforcement masonry shear wall (RMSW), several gaps still exist in
implementing FBSD, DBSD and PBSD. Some of these gaps include: quantifying
the reliability of the ductility-related parameters when a FBSD approach is adopted;
quantifying a reliable maximum lateral load displacement when a DBSD approach
is implemented; developing a backbone curve model so that different damage states
can be identified when implementing PBSD approach. As such, this dissertation
attempts to tackle these issues within the different seismic design approaches.
For FBSD, this study assesses the reliability of eight published plastic hinge models
using a large database of experimentally tested RMSW. In addition to assessing
model reliability in terms of displacement predictability, this dissertation also
specifies calibration factors to further improve the evaluated models. For DBSD,
multivariate data analysis technique through the principal component analysis
(PCA) technique and partial least square (PLS) analysis technique is utilized to
develop an experimentally calibrated RMSW displacement prediction expressions
for both the maximum lateral load limit state and the ultimate displacement limit
state. These expressions are further utilized to evaluate the parameters influencing
RMSW displacement capacities. For PBSD, complete backbone curve model for
RMSW under lateral loads is developed in this dissertation and subsequently
utilized to generate an analytical RMSW performance database. This large database
is further utilized to generate seismic fragility bands that are compared to the FEMA
P-58 fragility curves in an effort to represent a more realistic means of quantifying
RMSW performance and damage states under different seismic demand levels.