GEOMETRICAL OPTIMIZATION OF U-OSCILLATING WATER COLUMNS IN RANDOM WAVES

This paper deals with the problem of designing an optimal U-Oscillating Water Column (U-OWC) device equipped with a Wells turbine. Specifically, the paper proposes the implementation of a genetic algorithm for designing a U-OWC exposed to the typical sea states available in the Mediterranean Sea. The first challenge encountered in this problem is the efficient calculation of the U-OWC hydrodynamic parameters. The second challenge relates to the fact that the U-OWC dynamics is governed by two coupled nonlinear ordinary differential equations with no closed-form solution. For reducing the computational cost, the genetic algorithm is combined with a semi-analytical approach used for determining the U-OWC hydrodynamic parameters and with a statistical linearization based approximate solution of the equations governing the U-OWC dynamics. Such a procedure allows estimating efficiently, albeit approximately, the power output of the system. Numerical results compare a design based on a conventional "design sea state" vis-a-vis a design based on a "design wave climate". For this purpose, the case study of the Roccella Jonica marina (Reggio Calabria, Italy) is considered, as relevant wave data are available to characterize the most energetic seas as well as depicting the global wave climate available at that location. The numerical results highlight the fact that an optimization conducted on the basis of a design sea state does not lead to an optimal design in a wave climate.