Author Type

Graduate Student

Date of Award

Spring 4-28-2026

Document Type

Thesis

Publication Status

Version of Record

Submission Date

May 2026

Department

Civil, Environmental and Geomatics Engineering

College Granting Degree

College of Engineering and Computer Science

Department Granting Degree

Civil, Environmental and Geomatics Engineering

Degree Name

Master of Science (MS)

Thesis/Dissertation Advisor [Chair]

M. Arockiasamy

Abstract

Bridge fires caused by vehicle accidents, fuel tanker explosions and storage of combustible material beneath highway structures have led to severe structural damage and catastrophic collapse of prestressed concrete bridges. Critical structural collapse due to fire incidents such as I-85 prestressed concrete bridge girder collapse in Atlanta, 2017 and I-95 steel plate girder bridge partial collapse in Philadelphia, 2023 demonstrated the vulnerability of girders when exposed to extreme hydrocarbon fire temperatures exceeding 1000°C within few minutes. These events highlight the urgent need to better understand the thermal and structural behavior of bridge girders and explore design modifications that can enhance fire resistance and structural safety.

This study investigates the fire performance of an AASHTO Type IV prestressed concrete girder subjected to hydrocarbon fire exposure using a coupled thermo-structural finite element modeling using ANSYS Workbench. The study evaluates how increase in web thickness and concrete cover influences heat penetration and structural response during fire exposure. Two girder configurations are analyzed: i) a conventional prestressed concrete girder with 8-inch web thickness and 2-inch concrete cover, and ii) another girder with modification with 10-inch web thickness and 2.5-inch concrete cover. Temperature-dependent material properties for concrete and prestressing steel as per the Eurocode provisions are used in the case studies on the degradation of mechanical properties at elevated temperatures.

The results indicate that increasing both the web thickness and cover thickness significantly reduces the rate of temperature penetration toward the prestressing strands. The modified girder shows a reduction in internal temperature from approximately 783℃ to 342℃ and 56% reduction providing improved protection to prestressing strands and enhancing overall structural stability under fire exposure.

In addition, a machine learning framework using Artificial Neural Network is developed to predict fire-induced responses such as spalling risk, deformation and stress based on geometric and thermal parameters. The model demonstrates reliable predictive with an R2 value in the range of 0.75-0.85 validation of its capability for structural response prediction. This integrated approach supports performance-based design and enhances rapid decision-making for bridge safety under extreme fire conditions.

Download

Share

COinS