Abstract

Most wave energy concepts will have to be deployed in large arrays to achieve a power production sufficiently large to motivate the installation and operational costs. The individual wave energy converters will interact by scattered and radiated waves, which affects their dynamics and energy absorption. Much effort has been made to model arrays of wave energy converters and optimize their layout and other parameters, using both analytical, numerical, and experimental tools. The complexity grows fast with the number of interacting bodies, and there has been a lack of methods available to model parks of hundred devices or more. For small point-absorbers such as the one developed at Uppsala University, commercial parks will consist of several hundred devices, which motivated this work. In the talk, I will discuss several approaches that we have applied to model and optimize large wave energy parks. These include a nearest neighbour approach, a multiple clustering scattering method, and also a back-of-an-envelope computation for quick estimations of the park interactions. These methods share the common feature that they do not require assumptions in terms of layout periodicity, expansion in different length scales, or neglected degrees of freedom. In a realistic wave energy park, these assumptions do not hold, and previously developed approaches are not applicable. Special emphasis in the talk will be put on the clustering approach, in which the hydrodynamic interaction is computed exactly within clusters exactly, but iteratively between them.  The method constitutes a bridge between the iterative and non-iterative multiple scattering methods, and enables a faster modelling of parks of many floating bodies, to a retained accuracy.