Database building and statistical methods to predict sailing yacht hydrodynamics

The aim of the paper is to build a database of numerical simulation results and a parametrical model able to describe the evolution of the hydrodynamic behaviour depending on the hull shape through variables describing the hull geometry. This type of approach using meta-models can be used in various fields to approximate a complex behaviour in a very short computational time. This paper presents the parametrical tool used to produce the hulls and a specific hydrostatic routine featuring new coefficients and measurements. An emphasis is made on the statistical methodology developed to produce the database and then select the relevant predictive variables to build generalized linear regressions in the context of yachts hydrodynamics. The regressions allowing the prediction of the bare hull forces, moments and running attitude of racing monohulls are detailed and compared with numerical simulations. Finally, the implementation of these formulations in a six degrees of freedom VPP including appendage and aerodynamic models is presented. The increase in accuracy allowed by these new formulations is discussed, especially the central role of the formulations approximating the running attitude of the bare hull.