PSI - Issue 70

Edavalath Nadeem et al. / Procedia Structural Integrity 70 (2025) 19–26

21

27 GPa and 200 GPa. The circular column has a diameter of 1.83 meters and is reinforced with longitudinal steel corresponding to 2.22% of the gross cross-sectional area. Transverse reinforcement is provided using lateral stirrups, with a volumetric ratio of 0.42% per meter of column height. At the ends of the girder, seat-type abutments equipped with elastomeric bearings are incorporated. Longitudinal resistance is provided by the abutment backwalls, while transverse forces are resisted by shear keys. The elastomeric bearings are modelled using an elasto-plastic material formulation (Steel01) calibrated against the cyclic response model proposed by Mander and Basoz (1999). Shear keys are represented by zero-length spring elements that simulate their allowable shear strength, gap opening behaviour, and ultimate deformation capacity, in accordance with guidelines provided by Megally et al ., (2002). The abutment response is modelled by considering the combined effect of pile stiffness and passive soil resistance. The longitudinal abutment behaviour incorporates both backfill and pile contributions modelled in parallel, while transverse behaviour is governed by independent pile action. Pile-soil interaction is modelled using a tri-linear force-deformation relationship as recommended by earlier studies (Caltrans, 2019; Choi, 2002). The hyperbolic-gap backfill model proposed by Shamsabadi and Yan (2012) is employed to simulate the nonlinear soil response in OpenSees using the Hyperbolic-Gap Material. This model captures essential cohesive soil properties, including maximum displacement capacity, peak force, and average stiffness characteristics. To represent pounding interactions between the bridge deck and abutments under seismic excitation, a bi-linear force-deformation model with gap initiation is implemented at both span ends, based on the formulation by Muthukumar and DesRoches (2006). Figure 1 illustrate the structural layout of the RC bridge and the adopted nonlinear modelling approach in detail. Further technical details regarding model development, material calibration, and validation procedures can be found in previous studies (Padgett, 2007; Ramanathan, 2012; Dukes, 2013; Mangalathu, 2017; Soleimani, 2017). It is worth noting that the current investigation assumes idealized foundation conditions with negligible soil – structure interaction effects. However, this simplification may need to be revisited in future studies to obtain a more refined and realistic assessment of the bridge’s seismic response.

Fig. 1. Nonlinear numerical details of various components of the RC bridge model (Mangalathu, 2017). A comprehensive suite of ground motion time histories is selected to represent a wide range of intensity measures pertinent to the seismic hazard in the region of interest. These ground motions, informed by probabilistic seismic hazard analysis, are instrumental in developing reliable seismic demand models and constructing robust fragility curves. Additionally, variations in critical hazard parameters such as earthquake magnitude and epicentral distance are explicitly considered to account for their influence on structural response. For this study, a set of 160 ground motion records compiled by Baker et al ., (2011) under the PEER Transportation Research Program is utilized. This dataset comprises 40 near-field pulse-like records and 120 moderate-to-strong ground motion records recorded at varying distances from seismic fault sources. The selected suite captures a wide range of seismic characteristics, with peak ground accelerations (PGA) ranging from 0.02 g to 1.01 g and moment magnitudes spanning from 5.3 to 7.9, where g denotes the acceleration due to gravity. The durations of these records range from 15.58 seconds to 100.0 seconds, encompassing both short- and long-duration seismic events. Source-to-site distances vary from 1 km to 80