Open Collections

Abstract

This thesis presents a comprehensive framework for the seismic analysis and design of low damage post-tensioned rocking steel bridge piers. Conventional bridge systems are prone to residual drifts and strength degradation after major earthquakes, leading to costly repairs and extended service interruptions. To address these challenges, the research combines detailed finite element modelling, a calibrated multi-spring macro model, an optimization-based calibration procedure, predictive equations, multi-platform hybrid simulation, and a computational tool within an integrated methodology. High fidelity continuum analyses were first conducted to capture key nonlinear mechanisms such as uplift, post-tensioning effects, and local buckling. These simulations provided benchmark responses for developing and validating the proposed multi-spring model (MSM). The MSM reproduced global and local seismic responses with high accuracy while achieving substantial computational efficiency. A genetic algorithm-based optimization framework was developed and implemented to automatically calibrate spring parameters, ensuring consistent and reliable performance across different pier configurations. Using a large database of nonlinear analyses, predictive equations were formulated to estimate the pier pushover response directly from design parameters. These equations were validated through statistical and machine learning evaluations, confirming their reliability and physical interpretability. The framework was further applied to extensive parametric studies on single piers and bridge systems to quantify the effects of geometry, post-tensioning level, dissipator strength, and ground motion orientation on overall performance. A multi-platform hybrid simulation was also conducted to link detailed local modelling with efficient global analysis, demonstrating the ability to capture both continuum and system-level responses within one framework. To support practical implementation, a C# based software tool was developed as a pre and post processor that automates model generation, calibration, and result visualization. Overall, this research establishes a validated and scalable framework that bridges advanced numerical modelling and practical seismic design, equipping engineers with reliable tools to design resilient bridge capable of sustaining strong earthquakes while ensuring rapid restoration of functionality.