Hydrodynamic Interaction Among the Pontoons of a Floating Bridge: Effect of Global Responses

The National Public Road Administration has made a plan to establish a ferry-free road connection between Kristiansand and Trondheim. Bjørnafjorden is one of these fjords that have to be crossed, and several solutions are proposed for crossing. The design is developed in a cooperation between COWI, Aas Jakobsen, Johs Holte As and Global Maritime as a part of The Norwegian Public Roads Administrations (NPRA). The purpose of this master thesis is to examine the effects of hydrodynamic interaction on the dynamic response in regular and irregular waves. The results shows that large oscillations for multibody configuration begins for frequency between 1-2 rad/s. The design chosen in this thesis is a curved floating bridge, with a cable-stayed section in the south end that allows ship traffic to pass under the bridge. It is free floating without mooring lines, and the shear forces are carried through membrane stresses with the curved design. The bridge girder has a total distance from south to north of 5435 meters. In the south end, a navigation channel is placed with a span length of 525 meters. The low bridge has a span length of 100 meters, and the main girder is 16.2 meters above sea-level. First part of this project was a literature study regarding floating bridge concepts for Bjørnafjorden, dynamic loads on floating bridges and hydrodynamic interactions between rigid bodies. The pontoon model was created in GeniE with a reasonable mesh. The second part was to do a first order potential flow analysis of the different pontoon size in HydroD and Wadam. The curved bridge model was created in SIMA where hydrodynamic interaction between the pontoons was studied. Static analysis and eigenmode analysis was also carried out to verify the model is modeled correctly. The static analysis mainly focuses on bending moment, shear stress and static displacement of the bridge girder and compared to the reference model by The Norwegian Public Roads Administrations. An eigenvalue analysis was conducted, and large period deflection modes were observed for horizontal bending of the bridge girder. The maximum eigenvalue was found to be 65.4 seconds. The results of the eigenvalue analysis were compared with the reference analysis and were found to correspond well. This gave confidence for the model being able to represent the structural response of the bridge reasonably well. A simplified floating bridge was established to do further analysis of the effect of hydrodynamic interaction in three different wave directions. The wave heading from the north-west is most critical regarding moments and displacements. That may be because of distribution of all six load components, while waves from the west only have three components In a RIFLEX model the hydrodynamic couplings matrix for radiation data is not included. The interaction problem is therefore based on first order wave force transfer function and radiation data in the diagonal matrix. In an early stage, I realized how complicated a floating bridge concept is and cover all the aspects are impossible. The complete floating cable-stayed bridge with a total length of more than 5 kilometers turned out to be too large to analysis the hydrodynamic interaction effects. The primary focus was put on studying the response of simple bridge caused by wave loads from different headings