Page 191 - Fundamentals of Ocean Renewable Energy Generating Electricity From The Sea
P. 191

In Situ and Remote Methods for Resource Characterization Chapter | 7 179


             convertors (WECs), because such devices should be aligned with the direction of
             wave propagation for maximum efficiency. Further offshore, although the wave
             climate will be characterized by a modal wave direction that tends to correspond
             with the predominant wind direction, waves of different periods and heights
             propagate from a wide range of directions, in an environment that may be more
             suited to point absorber devices. However, nearer to the coastline, wave crests
             that may have originated at oblique angles relative to the coastline in deeper
             water tend to align themselves parallel to the coastline in the nearshore due to
             wave refraction (Section 5.2.2). Therefore, a directional wave buoy located in
             deeper water will generally record a much broader range of wave directions,
             compared with a wave buoy that is located in shallower coastal waters.
                Spectral characteristics: Because wave buoys, in contrast to tide gauges
             (Section 7.1.1), sample changes in water surface elevation at HF, they can
             report the spectral properties of waves. Different WECs are tuned to specific
             parts of the wave energy spectrum. For example, the Pelamis device generates
             electricity over a relatively wide range of wave parameters, but is optimal within
             a particular range [19]. To match, or tune, a WEC to the local wave climate,
             it is therefore important to record spectral properties of waves. An example
             of a mean 1D wave spectrum is shown in Fig. 7.16. In this example, the
             spectral peak (over the averaging time period of around 2 weeks) is around
             11 s (f = 0.09 per s), and the wave energy distribution around this spectral
             peak varies as shown. Note that if the spectral shape is assumed, for example,
             by assuming a JONSWAP or P-M spectrum, much of the natural distribution
             will not be captured. This is important in mean wave spectra, but particularly
             for instantaneous wave spectra, especially in cases of bi-modal spectra. In
             addition to 1D spectra, wave buoy postprocessing software also produce 2D
























             FIG. 7.16  Observed 1D wave spectra from the Pentland Firth, Scotland, averaged over the period
             January 17–31, 2012. (Data kindly supplied by Philippe Gleizon, University of the Highlands and
             Islands.)
   186   187   188   189   190   191   192   193   194   195   196