Bridges built today are larger, but also lighter, more slender, and more efficient structures than they were a century ago. As the free span of future suspension bridges increases, so does the need for reducing dead loads. Using Fiber Reinforced Polymer (FRP) deck for suspension bridges is one way to achieve significantly lighter dead loads. Although the cost of FRP materials is more than double the cost of conventional concrete and steel deck, the hypothesis of this research is that the savings in the anchoring system and foundation and the reduction in weight of the main cables and suspenders will result in an overall cost reduction. It is also the hypothesis of this research that the use of FRP deck will impair the aerodynamic stability of suspension bridges. Significant studies have been performed on the use of FRP materials in bridge structures. The Federal Highway Administration initiated research on FRP composite bridge decks in the early 1980s, primarily focused on deck strength and stiffness. In addition, several research projects have been conducted for health monitoring and to assess the long-term performance of FRP materials in bridge construction. Overall, the results suggest that long-term structural response was consistent and well within acceptable strength and serviceability design limits. For the research described in this dissertation, a parametric study was performed considering several bridges of different spans, materials, soil conditions, and material unit prices to study the economic and aerodynamic implications of using FRP deck in suspension bridges . Two groups of suspension bridges with 200 m, 400 m, and 600 m free spans were designed, one group using a reinforced concrete deck and the other group using the much lighter FRP deck. Since soil conditions affect the design of the anchorage and the overall cost of the bridges, three different soil types were considered in this research. The three soil conditions that were considered in this research were sound rock, medium sand, and stiff clay. Then, the aerodynamic stability was examined for all of the bridges using Selberg’s approach. Three-dimensional finite element analyses was performed for each bridge to obtain the values for the torsional moment of inertia and the vertical and torsional frequencies. These values were used in Selberg’s equation to determine the flutter speed of each bridge. A linear elastic analysis was performed to validate the three-dimensional finite element analysis results. The predicted flutter speeds obtained from the linear elastic approach and the finite element approach were within 9% for all the spans and deck materials. The use of FRP deck reduced the predicted flutter speed of the 200 m span bridge, 400 m span bridge, and 600 m span bridges by 35%, 36%, and 37%, respectively. Sensitivity cost analysis was performed of the 200 m, 400 m and 600 m span bridges founded in three different soil types. The three soil types considered were sound rock, medium sand, and stiff clay. The maximum savings was realized in the case of the weakest soil with the least resistance to the main cables tension force: stiff clay. Consistent with the research hypothesis, the cost of the FRP deck is more than twice the cost of the concrete deck, yet the overall cost savings for using an FRP deck were 30% to 42% of the cost using a concrete deck depending on span length and soil conditions. While earlier studies have demonstrated that the life cycle cost analysis could be advantageous in the long term because it requires less maintenance, the findings of the research described in this dissertation showed that the use of FRP deck could result in a 30% to 42% reduction in initial construction cost.
| Identifer | oai:union.ndltd.org:cmu.edu/oai:repository.cmu.edu:dissertations-1471 |
| Date | 01 December 2014 |
| Creators | Ahmed, Elfatih |
| Publisher | Research Showcase @ CMU |
| Source Sets | Carnegie Mellon University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Dissertations |
Page generated in 0.0018 seconds