An Innovative Application of FRP Composites in Bridge Strengthening and Repair

Abstract

Abstract With the gradual increase of vehicle loads in recent years, growing numbers of the existing bridge stock are becoming structurally obsolete. To tackle this issue, a bridge strengthening project, named SUREBridge, was initiated in 2015 at Chalmers University of Technology, with the support of several other partners. This project proposed an innovative strengthening method; the utilization of pretensioned carbon fiber reinforced polymer (CFRP) laminates on the base of bridge girders, in combination with glass fiber reinforced polymer (GFRP) sandwich panels on the top, to increase the yield and ultimate capacities of existing bridge decks. From experimental tests, 90 and 160% increases in the bending stiffness and ultimate bending capacity were found. This master thesis aims to replicate the experimental results from the four SUREBridge beams which were tested in the laboratory, through the finite element software ABAQUS and DIANA. Verification is made through comparisons of the loaddeflection curves, maximum crack width sizes and crack patterns. A secondary purpose is to document all approaches and modelling techniques when developing the finite element models. To achieve the objective, an extensive literature study was performed to determine appropriate approaches to include bond-slip between the steel reinforcement and the concrete beam, as well as the CFRP and the concrete beam. Several sensitivity analyses were also performed to determine the most appropriate model and material parameters, to find the best fit to the experimental results. From the analyses, it was concluded that the inclusion of the full bond-slip curves between both the concrete and steel, and concrete and CFRP are vital to obtain accurate results. However, the delamination load can accurately be predicted with a deviation of less than 2.5% by modelling the steel reinforcement as embedded. Hence it is concluded that in design, embedded steel can be utilized without significant losses in accuracy, and with considerable savings in the amount of time spent modelling and analyzing.