Background
These studies have been conducted in link with the WEAMEC project RC-PLus: Resource Characterization to Reduce the Cost of Energy through Coordinated Data Enterprise.
Abstract
Power production estimates are essential in the development of wave energy converter (WEC) concepts.
They rely on two key elements: a characterisation of the wave resource and a model of the WEC. In this study, we model an omnidirectional device using a state-of-theart time domain boundary element method code and we investigate three methods to compute power production estimates. The most accurate approach is based on the hindcast HOMERE database which provides frequency spectrum of the seastates on an hourly basis, which we use
as an input to the WEC numerical model. From the hindcast database we derive two lower fidelity representations of the wave resource: the significant wave height and the energy period for each seastate, on an hourly basis and
a discrete scatter diagram of these integral parameters. For the former, we then generate a JONSWAP spectrum for every seastate based on those integral parameters and we use them as inputs to the WEC numerical model. For the latter, we combine the scatter diagram with the power matrix of the WEC. We then present a comparison of those two methods with the approach based directly on the hindcast database spectra as the reference. At the site considered, the two lower fidelity methods overestimate the mean annual power production by about 6% compared with the reference approach.
Index Terms—wave energy conversion, energy production estimates, wave resource characterisation, numerical modelling.
