INVESTIGATION ON REASONS FOR POSSIBLE DIFFERENCE BETWEEN VIM RESPONSE IN THE FIELD AND IN MODEL TESTS

摘要

Floating offshore structures, such as production semi-submersibles and spars, can exhibit significant in-line and transverse oscillatory motions under current conditions. When caused by vortex shedding from the floater, such motions are generally called Vortex-Induced Motions (VIM). For semi-submersibles these motions could have a strong impact on the fatigue life of mooring and riser systems. Some field development studies indicate that the VIM induced fatigue damage for larger diameter Steel Catenary Risers (SCRs) can have a magnitude equal to or larger than the wave-induced fatigue damage. The VIM phenomenon for multi-column floaters is characterized by complex interactions between the flow and the motions of the floater. Presently, model tests are the preferred method to predict the VIM response of a multi-column floater. However, several studies indicate that the observed VIM response in the field is less than what is observed in model test campaigns: typical model test results are very conservative. Using such test results in the development of mooring and riser design can easily result in very conservative designs which can have a significant impact on mooring and riser cost, or even affect SCR selection and/or feasibility. The primary objective of the VIM JIP was to increase the physical insight into the VIM phenomenon. This knowledge is then used to address possible areas that could explain the differences between the results from model tests and field observations. To address these objectives, the JIP focused on model testing and CFD studies. A key segment of the JIP was the use of identical semi-submersible hull geometries for the numerical and experimental studies thereby facilitating the interpretation of the various response comparisons. The JIP identified that a CFD model, at model-scale Reynolds number, can reasonably well reproduce the VIM response observed in model tests. However, to have confidence in the CFD results extensive numerical verification studies have to be carried out. The effect of external damping was investigated in model tests and in CFD calculations. Both the numerical and experimental results show that external damping significantly reduces the VIM response. Comparisons between CFD results at model- and full-scale Reynolds number indicate that Froude scaling is applicable, with minor scale effects identified on the amplitudes of the VIM motions. Changing the mass ratio of the floater has a small influence on the VIM response. Experimentally it was found that VIM response under inline or transverse waves is slightly smaller than without the presence of waves and is wave heading and wave height dependent. The presence of waves does not explain the observed differences between model test results and field observations. The effect of unsteady current on the VIM response is minimal. Based on the results from the JIP it is concluded that increased external damping reduces the VIM response. The questions that remain are if the increased external damping is actually present in full-scale conditions and if the mooring and riser systems provide the required damping to reduce the VIM amplitudes.