Nonlinear air-water interface problems through a BEM-Level set domain decomposition

Several water flow problems of practical interest are characterized by fluid regions where the flow evolution is efficiently de- scribed by the potential theory, and other regions where this model is not valid. For instance, confined fluid areas can experience large air-water interface deformations followed by wave breaking and fragmentation phenomena. Limited water portions can be characterized by substantial vorticity generation due to water-water or water-structure interaction. In these cases the surrounding fluid domains can be slightly affected by such events. The ship hydrodynamic field is full of similar circumstances. The water-on-deck problem represents an example. In this case, compact masses of water enter the ship deck and the subsequent motion can result in important loads for the deck superstructures. On a long time scale, water breaking, air entrainement and vorticity generation are expected to occur. The later water-off-deck phase will cause the re-entering of water in the sea surrounding the vessel. As a result, near the vessel the free surface cannot be modeled as a smooth surface. Both the water shipping event and the final water-entry phase can involve substantial induced water loads on the vessel and large movements of the ship. Therefore related phenomena are of great interest for ship hydrodynamics, both from the operability and safety points of view. Due to the large free-surface deformations involved, a nonlinear analysis is needed. Before breaking and/or vortex shedding events, potential flow theory can capture accurately and with computational efficiency the involved flow evolution and predict connected loads and motions. After that, in the water regions where such phenomena occur and develop, this model has to be substituted by more general methods suitable to track the free surface deformations after the breaking, to handle the flow vorticity introduced in the fluid domain and to model the entrapped air. The present research activity is aimed to develop a numerical method able to simulate such ship flows and to adapt itself to the specific analyzed problem for an efficient and suitable solution. This has been done by considering a domain-decomposition strategy (see i.e. Quarteroni and Valli 1999, Campana and Iafrati 2001)