Forecast of the dynamic behaviour of FRP footbridges

Abstract

Fibre reinforced polymer (FRP) is a new construction material that often is used for pedestrian bridges. It offers a good strength-weight relation, a long-term durability and slender bridge shapes. Slenderness has effect on dynamic behaviour of the construction and vibrations are common for these bridges. Hand calculations are commonly used for the forecast of the dynamic behaviour, but this approach is not very accurate. Royal Haskoning DHV felt the need for a numerical model to be able to forecast the dynamic behaviour of pedestrian bridges of which accelerations, natural frequency and damping ratio are the most important. This thesis aims to fulfil this need. During a literature study relevant calculation methods were found and offered a better insight in the behaviour of bridges and the properties of FRP. FiberCore, manufacturer of FRP bridges, provided information of 15 different bridges all over the Netherlands. These bridges were all examined and vibrations were measured in the field. It was possible to bring 7 of these bridges in vibration using two test methods. To be able to measure accelerations and natural frequency mobile applications were validated in a case study in Puurs Belgium with the University of Leuven. A numerical model has been built for one specific bridge with a finite element method in SOFiSTiK and validated with the information provided by FiberCore. Furthermore, a dynamic load simulating a person walking over the bridge was designed. After having validated the model calculations were made to determine the natural frequency and the accelerations of this bridge as well without as with handrails. The results were compared with hand calculation according to the calculation method of the Joint Research Centre (JRC). Significant differences between the results of the model calculation and the JRC method were found and analysed. The main conclusions of the thesis are regarding the damping factor and the results from the numerical model. The damping factor of 1.4% to 5.4% is measured for 8 FRP bridges spanning 10-25m. The assumption of 1% damping, which is often used in design verifications in engineering practice is conservative. The maximum acceleration from the numerical model, 1.22 m/s2, is higher than the 0.43 m/s2 calculated from the SDOF method from the JRC, which is used in design verifications in engineering practice. The psi factor is a very dominant reduction in the latter calculation. The author concludes with some recommendations for further research.