Nonlinear seismic soil-pile interaction analysis for bridges founded in Quebec soil

Abstract

This doctorate program focuses on the nonlinear behavior of seismic soil-structure interaction of deep foundations with particular emphasis in Quebec soil. Soil-structure interaction (SSI) plays a crucial role when analyzing and designing important or essential structures, such as skyscrapers, nuclear reactor facilities and highway bridges. For the latter, deep foundations are commonly used to support the bridge superstructure. Historically, the analysis of the interaction between the piles and the embedment soil under lateral loadings used the beam on Winkler theory which is implemented through the p-y curves general method. The commonly-used springs are driven originally from full-scale tests on piles with static or slow cyclic lateral load application. Evidences, from disturbed or damaged structures after earthquake, indicate that SSI follows different curve from the static p-y curve used in the design. The main objective of this research study is to contribute bridge the gap by carrying comprehensive nonlinear seismic analyses on soil-pile interaction for Quebec soils. Extensive numerical investigations have been conducted on 600 parametrical models in order to evaluate these variations and gaps. Results from this research study shows that the following parameters have a direct impact on the seismic soil-pile interaction: the length of the pile, the mass of the structure, as well as the intensity of the seismic record. The seismic soil-pile interaction (SPSI) is normally investigated by dynamic time history analysis on continuum soil-structure models. However, this process is long and time consuming particularly in engineering practice. A new simplified method is proposed in this study to scale the static p-y curves in order to simulate the dynamic behavior of SPSI. The results indicate a reasonable matching between the results from time-history and static analysis. Scaling the current p-y curves in order to simulate the SPSI, would results in a more accurate estimation for the seismic demands on the bridge. Furthermore it will reduce the computation cost by benefiting from currently-used and fast to implement p-y curve method. Field experiments on several types of soil are recommended in order to normalize these findings for engineering practice.

Date	15 Dec 2015
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Omar Chaallal (Supervisor)