Behaviour of Cable-Stayed Bridges with High Modulus CFRP Cables

Abstract

A number of cable-stayed bridges have undergone costly cable replacements as a result of cable corrosion. Additionally, steel cables are prone to fatigue due to stress variations under traffic loads. In light of these issues, an interest in alternative cable materials has developed, with carbon fiber-reinforced polymers (CFRP) being the top candidate due to its high strength, corrosion resistance, relatively high stiffness, fatigue resistance, and creep rupture resistance. Aside from cost, the two main differences between steel and CFRP are that steel has a higher elastic modulus and a much higher coefficient of thermal expansion than CFRP. Over the last 3 decades, many papers have been published on the use of CFRP cables for cable-stayed bridges. The literature review revealed a general consensus that the lower elastic modulus of CFRP is somewhat disadvantageous as it results in larger vertical deck displacements under live loads and thus larger deck bending moments. Different studies have used different methods of sizing CFRP cables: some used the equivalent strength principle, while others used the equivalent stiffness principle. The literature review also revealed that there is little research on the behaviour of cable-stayed bridges with CFRP cables subjected to temperature loads. A parametric study is conducted and presented in this thesis to address these findings from the literature review. The parametric study is carried out in two Phases. Phase I involves varying the elastic modulus of the CFRP material. Current literature cites the elastic modulus of CFRP as being lower than that of steel, but high modulus (HM) CFRP materials do exist. The purpose of Phase I is to study the effect of using HM CFRP for the cables of a cable-stayed bridge. Phase II involves varying the cable cross section area while keeping the elastic modulus constant. The purpose of Phase II is to study the effect of the cable sizing method on the behaviour of a cable-stayed bridge. To conduct the parametric study, a series of finite element models were created for a theoretical cable-stayed bridge which has two H-pylons, a semi-fan cable arrangement, a 250 m long main span, and 100 m long side spans. The following results were obtained from the finite element models and discussed: cable forces; vertical deck displacements; longitudinal pylon tip displacements; axial force, shear, and bending moment in the deck and pylons; and the ratio of the 1st torsional frequency (T) to the 1st bending frequency (B). Only dead, live, and temperature loads were applied to the bridge, and only SLS-1, ULS-1, and ULS-2 load combinations from CAN/CSA S6-19 were considered. The parametric study yielded the following findings: • Decreasing the cable area in Phase II of the parametric study has the same effect as decreasing the elastic modulus of the CFRP in Phase I. • Increasing the elastic modulus attracts slightly more live load to the cables, increasing the axial forces and decreasing the shear and bending moment in the deck and pylons. • Temperature loads have a monumental effect on the performance of cable-stayed bridges with CFRP cables due to CFRP having a much lower coefficient of thermal expansion than steel and concrete. • When considering the combined effects of all applied loads, it appears that the use of HM CFRP cables has benefits. • Increasing the elastic modulus or the area of CFRP cables increases T and B but decreases the T:B ratio. Consequently, increasing the elastic modulus of the cables decreases the critical wind speed required to cause flutter. For future work, the parametric study should be extended to include the effects of various bridge geometries, various boundary conditions, wind loads, seismic loads, and ice accretion. A comprehensive analysis of the bridge’s dynamic behaviour should be conducted.