Floating bridges can be a promising solution for crossing deep fjords, where traditional bridge concepts become costly or technically challenging. In my project, I worked on modeling a floating bridge in OrcaFlex, aiming to eventually perform fatigue calculations at relevant locations along the bridge. The bridge concept is based on research from Seoul National University.
The project started with learning basic hydrodynamics, focusing on hydrostatics, wave theory, rigid body dynamics and potential coefficients. This understanding was essential for interpreting results from the software.
A large part of my research has focused on understanding the behavior of the floating pontoons. I first modeled the pontoon geometry in GeniE in order to mesh it and export it to OrcaWave. In OrcaWave, the hydrodynamic coefficients were obtained by running a diffraction analysis, which produced RAOs describing the pontoon’s response. To verify the pontoon model, I performed a free-decay test in the time domain in OrcaFlex. This gave me the information needed to calculate the pontoon’s natural frequency, which I then compared against the frequency-domain results.
I then moved on to modeling a beam in OrcaFlex that will represent the structural component of the bridge deck. OrcaFlex does not have dedicated beam elements, so a beam must instead be modeled as a «Line» object. The Line object is built as a lumped-mass model and handles axial and torsional stiffness directly. By further defining bending properties, it is possible to obtain results consistent with Euler-Bernoulli beam theory. Since the Line object is also represented as a finite-element model, the number of segments along its length is important. To achieve results consistent with Euler-Bernoulli beam theory, I ran a convergence test to determine the necessary number of segments.
Unfortunately, I did not manage to finish modeling the full bridge or carry out the fatigue calculations. Still, the project gave me the opportunity to learn basic hydrodynamics and become familiar with several softwares. I found the project both rewarding and challenging. Finally, a big thank you to my supervisors for their guidance throughout the project: Soomin Kim, PhD candidate at Seoul National University, and Professor Zhiyu Jiang.