Shape parameters optimisation of a quayside heaving rectangular wave energy converter.

In order to increase the competitiveness of wave energy industry, device's efficiency and cost have to be both optimised. This study is a contribution to this approach by presenting the optimisation of a quayside rectangular float. This configuration has been chosen since quayside systems present the advantage of greatly reducing maintenance and grid connection costs. Moreover, rectangular shapes enable us to validate the optimisation algorithm with a linear potential model. Thus, the evolutionary algorithm CMA-ES (Covariance Matrix Adaptation-Evolutionary Strategy) is used to perform Capture Width (CW) optimisations in regular and irregular waves. Results with and without a vertical wall are compared. In an open sea configuration, at a given frequency, the shape that maximises efficiency is a horizontal flat rectangle with a minimum draft and a breadth equals to a half wavelength. The presence of the wall doubles the CW compared to the case in open sea. Moreover, for a quayside case, one optimal shape for a given draft has a breadth twice smaller than the open sea case. In addition, the resonances of the water column in the gap between the float and the dike, that occur in regular waves, lead to various possible optimised shapes. However, only the shape corresponding to a minimum draft and a quarter of the wavelength wide is also recovered in irregular waves. This shape also appears to be less dependent to the clearance distance between the float and the vertical wall. Finally, bi-objective optimisation results are realised in irregular waves for a rectangular quayside float in order to maximise the CW and minimise the WEC (Wave Energy Converter) cost. A significant cost reduction is possible by reducing the float immersed surface and volume. It is shown that a reduction of the float's breadth of 50% from the optimal value only reduces the CW of about 7%. • Optimal shape of a rectangular float in front of a vertical wall is found. • Compared to open sea cases, the absorbed efficiency is doubled and the width reduced. • Effect of gap resonances on the maximum absorbed power is highlighted. • CMA-ES algorithm is performed using mono and multi-objective optimisation. • Pareto fronts based on shape and cost, are obtained for a quayside float. [ABSTRACT FROM AUTHOR]