In this paper, the wave transmission, reflection, and energy dissipation characteristics as well as the hydrodynamic forces of two partially submerged breakwaters are compared. The breakwaters under consideration are designed to be installed in coastal areas where the water depth may be important (typically 50 m), making traditional bottom seated configurations unsuitable. They are composed of a free-surface piercing caisson mounted on vertical piles. The first breakwater considered in this study, as a matter of reference for comparison, is a rectangular caisson. The second one is a new profile proposed by Colmard and called BYBOP by the writer. The seaside face, fronting the ocean, is an elliptic profile while the harbor side is a plane inclined 55 to the vertical. A numerical simulation on both these shapes, at model scale, was carried out in two numerical wave tanks. The first one, CANAL, is based on fully nonlinear potential flow theory while the second one, ICARE, is a Reynolds averaged Navier-Stokes equations solver simulating viscous free-surface flow around structures. The problem is considered as a two dimensional one and solved in the time domain. Reflection, transmission, and dissipation coefficients are computed, together with hydrodynamic forces, on the two different caisson shapes. The new design is shown to have far better efficiency, at less material expenses, for the same width.
The breakwater under consideration is intended to be installed in coastal areas where water depth may be large (typically fifty meters) compared to usual operating water depth in civil coastal engineering. For that reason, it is constituted of a free-surface piercing solid caisson mounted on concrete piles instead of being bottom standing as usual. This configuration makes it partly permeable to the incoming ocean waves which can be transmitted beyond the dike by propagating below. The final goal of the study is the hydrodynamic optimization of this breakwater with regard to the transmission and reflection coefficient, together with forces and moment minimization. The end effects being disregarded at the moment, the problem is regarded as a 2D one and it is investigated both experimentally, in a physical wave tank, and numerically using two Numerical Wave Tanks (NWT): the first one based on a fully non-linear potential flow theory, the second one, solving RANSE equations, simulating a viscous flow around the structure. Comparisons between the results of these three different approaches are given.