Shipping water on a deck is analyzed by a particle method in three dimensions. The calculation result shows good agreement with experimental data which was obtained by Tanizawa et al. A numerical model for a rigid body is developed for involving interaction between ship motion and large waves in the particle method.
Shipping water load may damage hatch covers and equipment on decks. Shipping water behavior and deck load have been studied mainly by experiments (Tasaki, 1961; Suhara et al., 1973; Goda and Miyamoto, 1976; Mizoguchi, 1988; Takagi and Naito, 1993; Ogawa et al., 1997; Greco et al., 2004). Prediction methods, such as a dam break model (Mizoguchi, 1988) and a flood wave model (Ogawa et al., 1997; Ogawa et al., 1999), were proposed. However, these methods were analytical and simplifications were necessary. For example, three-dimensional water behavior could not be predicted.
Recently, numerical analysis of shipping water was carried out by Hu and Kashiwagi (2004) using CIP (Constrained Interpolation Profile) method of Yabe et al. (2003), and Nielsen and Mayer (2004) using an advanced VOF method of Ubbink (1997).
Numerical analysis is preferable for the wide application. However, we need to analyze separation and merging of water as well as large deformation of free surfaces in shipping water. In the usual numerical methods, such as the finite volume and the finite element methods, separation and merging of water are very difficult to calculate. If the grid is dynamically fitted to the free surface motion, the calculation will fail when the grid is extremely distorted. If the grid is fixed and a transport equation of the water fraction is solved, numerical diffusion derived from the convection terms will make the free surfaces unclear. The essential reason of these troubles is the existence of the grids.
