Date of Award

Fall 12-2021

Level of Access Assigned by Author

Open-Access Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Civil Engineering

Advisor

William Davids

Second Committee Member

Eric Landis

Third Committee Member

Roberto Lopez-Anido

Additional Committee Members

Mohammadali Shirazi

Senthil Vel

Abstract

Bridges represent critical nodes in the United States’ transportation network, and an accurate understanding of their strength and durability is paramount to their continued use to transport people and goods. Traditionally, bridge assessment has been conducted through conventional beam-line analysis of isolated bridge components as specified by the American Association of State Highway and Transportation Officials (AASHTO). Although this method is reasonable for relatively new structures whose behavior are well understood, it can lead to significant over-conservative estimates of live-load capacity for older structures and for new types of structure for which design bases are still being developed. For such structures, investigation by experimental and numerical analysis can lead to much more accurate predictions of live-load capacity, potentially reducing the need for repair and remedial action, or further design optimization. This dissertation presents the results of experimental and numerical analysis of a series of older reinforced concrete (RC) T-beam bridges and a novel fiber reinforced polymer (FRP) composite tub (CT) girder bridge, as well as the development of a novel numerical analysis technique, Proxy Finite Element Analysis (PFEA). RC T-beam bridges are common in the state of Maine and are often much older than their initial design life. These bridges frequently do not rate adequately based on AASHTO procedures, despite continuing to carry modern loading without signs of distress, leading to expensive, possibly unneeded remedial actions. To better assess their live-load capacity, a series of ten such bridges is subjected to non-destructive live-load testing (NDLLT) under high vehicular load. The strain response of each is extracted, allowing a better understanding of its behavior and updated capacity estimates to be determined. Based on the results of these tests, the flexural ratings factors (RFs) of each of these structures could be increased, with six increasing to above 1.0, demonstrating their adequacy for modern loading. The behavior of these bridges is further investigated through numerical analysis of detailed, linear finite element (FE) models. To allow straightforward inclusion of the complex nonlinear constitutive behavior of RC T-beam bridge girders in nonlinear FE analysis for capacity rating, a novel technique, PFEA is developed and later expanded for generality. This technique extracts a girder section’s nonlinear moment-curvature relationship and applies it to a fictitious, “proxy” section for which nonlinear analysis is much less cumbersome. The technique is verified against previous destructive tests of individual girders and a full bridge, and its utility is expanded, demonstrating its generality. Finally, it is used to load rate the previously tested RC T-beam bridges, resulting in significant increases to each structure’s flexural RF. The Hampden Grist Mill Bridge (HGMB) in Hampden, Maine is the first bridge in the world to use the FRP CT girder system developed by the University of Maine and was constructed in 2020. As such, its behavior was relatively unknown. To better characterize the bridge’s behavior, it is subjected to NDLLT and its response measured. It is found that the structure behaves much more rigidly than designed, with more uniform load distribution and significant unintended rotation end fixity. This testing also allows for an updates capacity load rating to be determined, further displaying the structure’s adequate and conservative design. These aspects of the HGMB’s behavior are further investigated through analysis of detailed, linear FE models.

Download

Files over 10MB may be slow to open. For best results, right-click and select "save as..."

Share