On the validity of CFD for simulating extreme green water loads on ocean going vessels

The amount of green water and the associated loads that an ocean-going vessel may encounter during its service life are important aspects to consider in the vessel's design and classification. As green water is typically a highly non-linear phenomenon, commonly the maritime industry relies on model tests to predict green water loads and their occurrence. In recent years, however, a lot of progress with Computation Fluid Dynamics (CFD) has been made in predicting non-linear flows and associated loads at a high level of accuracy. Especially in the field of wave impacts on (moored) offshore structures at zero speed, significant progress has been made and documented using CFD. A natural extension of this progress is to expand the obtained confidence in the applicability of CFD for simulating extreme wave events to applications involving vessels at forward speed. To that end, this paper presents a validation study towards the prediction of green water loading on a (typical) container vessel at forward speed by CFD. For validation, two extreme green water events were selected from a model test campaign carried out at MARIN within the context of the CRS (Cooperative Research Ships) working group 'green water dynamics'. In these tests a KRISO Container Ship (KCS) is sailing in head seas when encountering severe green water. As CFD tool, the Cartesian-grid based Volume-of-Fluid CFD solver ComFLOW was selected. Furthermore, a deterministic approach is taken for the validation, by reconstructing the non-linear incoming wave in a high amount of detail and imposing the 6 degrees of motion of the vessel using the wave basin measurements. Time traces of the green water flow on deck and local- and global impact loads on the breakwater are presented and compared against the experimental data. Detailed visualizations of the CFD results are presented to further illustrate the obtained match with the model test results and emphasize the additional value of complementing model tests with deterministic CFD analysis.