Back to Search Start Over

Effects of liquefiable soil and bridge modelling parameters on the seismic reliability of critical structural components.

Authors :
Padgett, Jamie E.
Ghosh, Jayadipta
Dueñas-Osorio, Leonardo
Source :
Structure & Infrastructure Engineering: Maintenance, Management, Life-Cycle Design & Performance. Jan2013, Vol. 9 Issue 1, p59-77. 19p. 1 Diagram, 9 Charts, 2 Graphs.
Publication Year :
2013

Abstract

This study investigates the sensitivity of seismic fragility estimates for bridge components to variation in structural and liquefiable soil modelling parameters. A rigorous sensitivity analysis is conducted to evaluate the relative importance of 13 random variables that reflect uncertainty in the seismic performance assessment of bridges in regions with liquefiable soils. The results indicate that the fixed and expansion bearings and bent piles tend to be sensitive to the greatest number of modelling parameters for the case study system, while the abutments are less sensitive. The most significant modelling parameters affecting the seismic fragility include such parameters as undrained shear strength of soil, structural damping ratio, soil shear modulus, gap between deck and abutment, ultimate capacity of soil and fixed and expansion bearing coefficients of friction. The 5% and 95% confidence intervals reveal wide bounds on the seismic fragility curves, particularly for more vulnerable bridge components such as the piles or expansion bearings. The results offer insights to improve seismic reliability assessment in liquefaction susceptible regions and provide a basis for efficient bridge network reliability analyses. The findings guide future uncertainty treatment, management of computational resources and investment in refined modelling parameter estimates through field testing or other means. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
15732479
Volume :
9
Issue :
1
Database :
Academic Search Index
Journal :
Structure & Infrastructure Engineering: Maintenance, Management, Life-Cycle Design & Performance
Publication Type :
Academic Journal
Accession number :
82153367
Full Text :
https://doi.org/10.1080/15732479.2010.524654