The Effect of the Spectral Distribution of Wave Energy on the Performance of a Bottom Hinged Flap Type Wave Energy Converter

Abstract

The power output from a wave energy converter is typically predicted using experimental and/or numerical modelling techniques. In order to yield meaningful results the relevant characteristics of the device, together with those of the wave climate must be modelled with sufficient accuracy. The wave climate is commonly described using a scatter table of sea states defined according to parameters related to wave height and period. These sea states are traditionally modelled with the spectral distribution of energy defined according to some empirical formulation. Since the response of most wave energy converters vary at different frequencies of excitation, their performance in a particular sea state may be expected to depend on the choice of spectral shape employed rather than simply the spectral parameters. Estimates of energy production may therefore be affected if the spectral distribution of wave energy at the deployment site is not well modelled. Furthermore, validation of the model may be affected by differences between the observed full scale spectral energy distribution and the spectrum used to model it. This paper investigates the sensitivity of the performance of a bottom hinged flap type wave energy converter to the spectral energy distribution of the incident waves. This is investigated experimentally using a 1:20 scale model of Aquamarine Power’s Oyster wave energy converter, a bottom hinged flap type device situated at the European Marine Energy Centre (EMEC) in approximately 13m water depth. The performance of the model is tested in sea states defined according to the same wave height and period parameters but adhering to different spectral energy distributions. The results of these tests show that power capture is reduced with increasing spectral bandwidth. This result is explored with consideration of the spectral response of the device in irregular wave conditions. The implications of this result are discussed in the context of validation of the model against particular prototype data sets and estimation of annual energy production.