Parametric Excitation of Hangers at the Hardanger Bridge

There has been observed large vibrations on some of the longest hangers at the Hardanger Bridge. Parametric excitation has been suggested as the cause of these vibrations. The main aim of this thesis has therefore been to verify if this is the case. As a preface to this inquiry, parametric excitation of cables has been looked into, and a search to find a numerical convergence criterion has been carried out. Though, without any concluding results. The question of parametric excitation of the hangers at the Hardanger Bridge was first investigated by looking at numerical models of the hangers and the bridge. This was done to see if vibrations of the girder could excite the hangers. In addition to the question of parametric excitation, the dynamic properties of the hangers were sought. Experiments at the bridge were executed, rendering measurements to be utilized in modal analyses, and retrieving information to asses if parametric excitation could be a reasonable conclusion to the vibrations. Executed at the bridge were hammer tests and recordings of ambient vibration. Information of passing traffic was also retrieved to see if the traffic could excite the hangers. The dynamic properties were estimated using three modal analysis methods, i.e. circle fit, line fit and least-square complex exponential. Based on the measurements from the Hardanger Bridge, circle fit and line fit could estimate reliable results for the natural frequencies. While the methods showed poor results for these measurements considering damping and mode shapes. Least-square complex exponential gave varying results for all the dynamic properties, and those results are therefore not presented. Results from the numerical models indicated that parametric excitation due to girder vibrations was unlikely. While influence from traffic could be excluded after examining the ambient vibrations together with the traffic information. Concluding that parametric excitation is not probable, the ambient vibration data was inspected considering vortex shedding as the cause of the vibrations. Brief investigations indicated that this could be the case