Seismic analysis of pile foundations for bridges

This thesis deals with the seismic response of bridges supported on pile foundations. A substantial portion of thesis is devoted in developing a better understanding of the seismic behavior of the pile foundations and their effects on the overall response of the bridge structure. Current methods...

Full description

Bibliographic Details
Main Author: Thavaraj, Thuraisamy
Format: Others
Language:English
Published: 2009
Online Access:http://hdl.handle.net/2429/10921
id ndltd-UBC-oai-circle.library.ubc.ca-2429-10921
record_format oai_dc
collection NDLTD
language English
format Others
sources NDLTD
description This thesis deals with the seismic response of bridges supported on pile foundations. A substantial portion of thesis is devoted in developing a better understanding of the seismic behavior of the pile foundations and their effects on the overall response of the bridge structure. Current methods for modeling the pile foundations are evaluated and some deficiencies are identified. New methods are developed that model better the seismic behavior of the pile foundations. Case studies based on the recorded response of a two span bridge under ambient vibration and strong earthquake shaking showed that the behaviour of the bridge foundations under strong shaking is highly nonlinear. A new lumped parameter model consisting a set of nonlinear springs and dashpots was developed to reproduce the nonlinear behaviour of the pile foundations during strong shaking. The nonlinear stiffnesses and damping ratios of the pile foundation are determined using an existing Quasi-3D finite element model. The lumped parameter model was incorporated in to a three dimensional stick model of the bridge superstructure to determine the seismic response of the bridge in the time domain. Seismic response analyses using the proposed model showed that the flexibility of the pile foundation and its nonlinear behaviour and the inertial interaction can significantly affect the behaviour of the bridge. A parametric study was conducted to see how different parameters affected the response of the bridge. One finding was that it was the relative stiffness of the foundation and the superstructure that controlled seismic response of the bridge. A new method was developed for determining the dynamic impedances of bridge abutments using simplified models of the abutments. Studies showed that the impedances of the abutments change significantly during strong shaking and that these changes should be taken into account. A finite element method was developed for the fully coupled seismic response analysis of pile foundations and superstructure in the time domain. This method treats the superstructure and foundation as two subsystems during the finite element solution process and hence requires less computer storage. The new method was used to evaluate the uncoupled analyses that are sometimes carried out in practice. Studies showed that the uncoupled analyses failed to give acceptable results. The Winkler model of a single pile is widely used in practice for the analysis of pile foundations. Therefore this method was evaluated in detail. A comprehensive program for the dynamic analysis of a single pile using a Winkler model was developed which included the currently available methods. This method was then used to evaluate the Winkler model and the use of p-y curves recommended by the American Petroleum Institute(API) by simulating a centrifuge test on a single pile under low level and strong shaking. Analyses using the API p-y curves gave poor estimates of the response of the pile under strong shaking. Studies showed that the extrapolation of the group behavior from the behavior of single pile using a group factor should consider both pile-to-pile interaction and the effects of superstructure-foundation interaction. The pile cap condition also needs to be modeled appropriately. An effective stress method was developed for the dynamic analysis of pile foundations in potentially liquefiable soils by incorporating a pore water pressure generation model into the Quasi-3D finite element method. The effective stress method was extensively verified using centrifuge test data on a single pile and (2x2) and (3x3) pile groups under low-level and strong shaking. A number of recommendations are made for future work to explore further some of the issues raised during the research study. === Applied Science, Faculty of === Civil Engineering, Department of === Graduate
author Thavaraj, Thuraisamy
spellingShingle Thavaraj, Thuraisamy
Seismic analysis of pile foundations for bridges
author_facet Thavaraj, Thuraisamy
author_sort Thavaraj, Thuraisamy
title Seismic analysis of pile foundations for bridges
title_short Seismic analysis of pile foundations for bridges
title_full Seismic analysis of pile foundations for bridges
title_fullStr Seismic analysis of pile foundations for bridges
title_full_unstemmed Seismic analysis of pile foundations for bridges
title_sort seismic analysis of pile foundations for bridges
publishDate 2009
url http://hdl.handle.net/2429/10921
work_keys_str_mv AT thavarajthuraisamy seismicanalysisofpilefoundationsforbridges
_version_ 1718588697758138368
spelling ndltd-UBC-oai-circle.library.ubc.ca-2429-109212018-01-05T17:35:36Z Seismic analysis of pile foundations for bridges Thavaraj, Thuraisamy This thesis deals with the seismic response of bridges supported on pile foundations. A substantial portion of thesis is devoted in developing a better understanding of the seismic behavior of the pile foundations and their effects on the overall response of the bridge structure. Current methods for modeling the pile foundations are evaluated and some deficiencies are identified. New methods are developed that model better the seismic behavior of the pile foundations. Case studies based on the recorded response of a two span bridge under ambient vibration and strong earthquake shaking showed that the behaviour of the bridge foundations under strong shaking is highly nonlinear. A new lumped parameter model consisting a set of nonlinear springs and dashpots was developed to reproduce the nonlinear behaviour of the pile foundations during strong shaking. The nonlinear stiffnesses and damping ratios of the pile foundation are determined using an existing Quasi-3D finite element model. The lumped parameter model was incorporated in to a three dimensional stick model of the bridge superstructure to determine the seismic response of the bridge in the time domain. Seismic response analyses using the proposed model showed that the flexibility of the pile foundation and its nonlinear behaviour and the inertial interaction can significantly affect the behaviour of the bridge. A parametric study was conducted to see how different parameters affected the response of the bridge. One finding was that it was the relative stiffness of the foundation and the superstructure that controlled seismic response of the bridge. A new method was developed for determining the dynamic impedances of bridge abutments using simplified models of the abutments. Studies showed that the impedances of the abutments change significantly during strong shaking and that these changes should be taken into account. A finite element method was developed for the fully coupled seismic response analysis of pile foundations and superstructure in the time domain. This method treats the superstructure and foundation as two subsystems during the finite element solution process and hence requires less computer storage. The new method was used to evaluate the uncoupled analyses that are sometimes carried out in practice. Studies showed that the uncoupled analyses failed to give acceptable results. The Winkler model of a single pile is widely used in practice for the analysis of pile foundations. Therefore this method was evaluated in detail. A comprehensive program for the dynamic analysis of a single pile using a Winkler model was developed which included the currently available methods. This method was then used to evaluate the Winkler model and the use of p-y curves recommended by the American Petroleum Institute(API) by simulating a centrifuge test on a single pile under low level and strong shaking. Analyses using the API p-y curves gave poor estimates of the response of the pile under strong shaking. Studies showed that the extrapolation of the group behavior from the behavior of single pile using a group factor should consider both pile-to-pile interaction and the effects of superstructure-foundation interaction. The pile cap condition also needs to be modeled appropriately. An effective stress method was developed for the dynamic analysis of pile foundations in potentially liquefiable soils by incorporating a pore water pressure generation model into the Quasi-3D finite element method. The effective stress method was extensively verified using centrifuge test data on a single pile and (2x2) and (3x3) pile groups under low-level and strong shaking. A number of recommendations are made for future work to explore further some of the issues raised during the research study. Applied Science, Faculty of Civil Engineering, Department of Graduate 2009-07-17T17:07:15Z 2009-07-17T17:07:15Z 2000 2000-05 Text Thesis/Dissertation http://hdl.handle.net/2429/10921 eng For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. 15672866 bytes application/pdf