Downloaded from http://rsta.royalsocietypublishing.

org/ on May 20, 2015

Phil. Trans. R. Soc. A (2012) 370, 345–364

doi:10.1098/rsta.2011.0152

Nearshore oscillating wave surge converters and

the development of Oyster
BY TREVOR WHITTAKER AND MATT FOLLEY*
School of Planning, Architecture and Civil Engineering, Queen’s University
Belfast, Belfast, UK

Oscillating wave surge converters (OWSCs) are a class of wave power technology that
exploits the enhanced horizontal fluid particle movement of waves in the nearshore coastal
zone with water depths of 10–20 m. OWSCs predominantly oscillate horizontally in surge
as opposed to the majority of wave devices, which oscillate vertically in heave and usually
are deployed in deeper water. The characteristics of the nearshore wave resource are
described along with the hydrodynamics of OWSCs. The variables in the OWSC design
space are discussed together with a presentation of some of their effects on capture width,
frequency bandwidth response and power take-off characteristics. There are notable
differences between the different OWSCs under development worldwide, and these are
highlighted. The final section of the paper describes Aquamarine Power’s 315 kW Oyster 1
prototype, which was deployed at the European Marine Energy Centre in August 2009. Its
place in the OWSC design space is described along with the practical experience gained.
This has led to the design of Oyster 2, which was deployed in August 2011. It is concluded
that nearshore OWSCs are serious contenders in the mix of wave power technologies.
The nearshore wave climate has a narrower directional spread than the offshore, the
largest waves are filtered out and the exploitable resource is typically only 10–20% less
in 10 m depth compared with 50 m depth. Regarding the devices, a key conclusion is
that OWSCs such as Oyster primarily respond in the working frequency range to the
horizontal fluid acceleration; Oyster is not a drag device responding to horizontal fluid
velocity. The hydrodynamics of Oyster is dominated by inertia with added inertia being a
very significant contributor. It is unlikely that individual flap modules will exceed 1 MW
in installed capacity owing to wave resource, hydrodynamic and economic constraints.
Generating stations will be made up of line arrays of flaps with communal secondary
power conversion every 5–10 units.
Keywords: wave energy; oscillating wave surge converter; nearshore

1. Introduction

Although patents on wave power conversion technology date back over 200 years,
the bulk of research and development in this area has taken place in the last
*Author for correspondence (m.folley@qub.ac.uk).

One contribution of 18 to a Theo Murphy Meeting Issue ‘The peaks and troughs of wave energy:
the dreams and the reality’.

345 This journal is © 2011 The Royal Society

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

346 T. Whittaker and M. Folley

40 years. The focus of much of this work has been and still is on deep water
offshore systems mainly relying on power extraction from heave [1–5]. Another
significant area of development has been shoreline oscillating water column
devices such as the land installed marine powered energy transformer (LIMPET)
[6] or low-head hydro systems such as the tapered channel (TAPCHAN)
[7]. Until recently, wave power extraction from the surge motion of waves
in nearshore waters has received relatively little attention. Maybe this is
because there is an historic opinion that the nearshore wave power resource is
dramatically less than that offshore. It is true that the nearshore gross wave
power resource in all directions can be less than half that in water depths of
between 50 and 100 m; however, this is very misleading. Folley & Whittaker
[8] proposed that the exploitable wave power resource, which is the resource
available to commercial wave power converters in wave farms with an economic
cap on installed generating capacity, is a more realistic measure. In many
nearshore sites, the exploitable resource is often only 10–20% lower than that
offshore [9].
In 2002, the wave power research group at Queen’s University Belfast turned its
attention towards devices that are predominantly driven by the surge component
of the waves. The team described this family of devices as oscillating wave surge
converters (OWSCs). With funding from the Engineering and Physical Science
Research Council (EPSRC), a 2 year research programme ensued. The study
concluded that a flap hinged to the sea bed at its lower edge with the top edge
penetrating the water surface was the most promising form of this type of device.
In 2005, Aquamarine Power Ltd was formed to commercially develop this system
that they named Oyster [10].
Today, a number of different OWSC concepts have been proposed and
some are under commercial development. In addition to Oyster, these include
WaveRoller [11], bioWAVE [12] and Frond [13]. WaveRoller, unlike Oyster, is a
completely submerged flap. Both bioWAVE and Frond are surface-piercing flaps
in deeper water and connected to a hinge on the sea bed by a stalk. These
four devices occupy different parts of the design space and have remarkably
different hydrodynamic characteristics. There are also floating OWSCs such as
the ‘Farley Triplate’ [14] and the ‘Langlee’ [15] systems, which are designed for
deeper water. Because their performance depends on the dynamics of the support
structure, they occupy a very different part of the design space and they will not
be considered further in this paper. Similarly, hinged devices with significant
change in displacement when they oscillate, which has a large influence on the
hydrodynamics, will not be considered. A notable example in this category is the
‘Edinburgh Duck’ [16].
This paper commences with an overview of the nearshore wave resource,
including the physical attributes of this environment. This is followed by
a description of the hydrodynamics of OWSCs and some of the output
from the extensive research programme during the past 8 years is presented.
The influence of some of the key design parameters on the hydrodynamic
performance of OWSCs is presented. This has led to the development of
the Oyster 1 prototype that was deployed at the European Marine Energy
Centre (EMEC) off Orkney in 2009. Finally, some thoughts are presented on
the future development of this technology and how it might be embodied in
Oyster 2.

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 347

2. The nearshore environment

The nearshore environment, defined here as having a water depth of between

10 and 20 m, offers a range of advantages and limitations relative to deeper
water locations in the 50–100 m depth range. Consequently, certain technologies
are better suited to these particular locations and cannot necessarily be
transposed from one water depth to another. For example, the compliant mooring
arrangements of freely floating wave energy converters require a minimum water
depth to work effectively and so are not generally suitable for shallower water.
Conversely, the structural task [17] of wave energy converters that react against
the sea bed increases dramatically with water depth and so are not generally
suitable for deeper water.

(a) Definition of the exploitable wave energy resource

As waves travel from the deep ocean to the shoreline, they transform, as the
fluid particle motions are affected by the presence of the sea bed. The natural
processes of wave shoaling, refraction, diffraction, surf breaking, white capping,
sea bed friction and marine currents all modify the wave properties and the power
available to a wave farm. Unlike wind farms, wave energy converters will be
deployed in lines that might extend tens of kilometres. These lines may follow
depth contours, but, in general, will be approximately orthogonal to the mean
direction of wave propagation to minimize the number of devices that are in the
energy shadow of wave energy converters up-wave. This is different from wind
farms because wave energy converters may extract a significant proportion of
the incident energy, resulting in an extensive energy shadow (wind turbines only
extract a small proportion of the incident energy and so have a small energy
shadow because the wind energy resource extends up into the stratosphere).
Because of the necessary directional orientation of wave farms, the simple measure
of gross annual average wave power density, which includes an un-weighted
sum of wave energy from all directions, is an inappropriate measure of the wave
energy resource.
Using the gross annual average wave power density misrepresents the resource
because it does not discount the contribution of wave energy approaching
perpendicular to the orientation of the wave farm, where the capture of an
individual wave energy converter is reduced because of the energy shadow
of its neighbours. This misrepresentation is compounded where the power
capture is determined by multiplying the individual spectral wave power
densities by a hydrodynamic response factor multiplied by an average power
take-off (PTO) efficiency because the maximum power rating of the wave
energy converter is ignored. An even greater misrepresentation is obtained
when the average wave power density is simply multiplied by an average
device response factor. Unfortunately, it is not uncommon for early stage
reviews of wave energy technologies to use these simplified representations.
The most obvious deficiencies in these methods are that the directional
distribution of the waves and the maximum power plant rating are not
accounted for. These deficiencies are significant because they can result in the
conclusion that wave energy converters deployed in deeper water are substantially

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

348 T. Whittaker and M. Folley

50
gross

wave power density (kW m–1)

net
40 exploitable

average incident
30

20

10

0
offshore 50 m 10 m
water depth

Figure 1. Gross and exploitable wave power at three water depths. (Online version in colour.)

more productive than those in shallower nearshore waters; however, it can

be shown that much of the difference is due to the misrepresentation of
the resource.
In development of a more suitable representation of the wave energy resource,
the configuration of the wave farm and maximum power rating must be
considered. A reasonable measure of the available, or net, wave energy resource
is the energy that crosses the line of the wave farm. That is, the wave resource
must be directionally resolved and taken across the deployment line of the wave
farm. With respect to maximum plant rating and in particular taking proper
account of extreme seas, it must be recognized that all wave energy converters
will have limited installed generating capacity, which is chosen on the basis of
both economics and part load efficiency. The investment in large-capacity PTO
plant, which is used for a few per cent of the year, cannot be economically justified.
Moreover, power train efficiency at low load factors can be substantially less than
those at high load factors. Many of the wave energy converters proposed to date
have load factors of between 25 and 40 per cent and therefore they cannot extract
the energy from the more energetic seas.
Folley & Whittaker [8] proposed the concept of wave resource evaluation using
the ‘average exploitable wave power density’. This is defined as the mean value
of the directionally resolved incident wave power density with the largest seas
capped to four times the mean value. Thus, the fixed directional orientation of
wave farms and limitations of plant rating are recognized.

(b) Nearshore versus offshore

Folley et al. [9] detailed an analysis of the wave resource at EMEC on the
west coast of Orkney, and this is summarized here to illustrate the difference
in wave resource at different depths and using different calculation methods.
The wave resource calculations, shown in figure 1, are for deep offshore 40 km
west of Orkney and at the 50 m and the 10 m deep test sites. The average omni-
directional (gross), the directionally resolved (net) and the exploitable wave power
densities are shown to illustrate the transformation from gross to exploitable wave
energy resource. Although the greatest difference in wave resource, irrespective of

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 349

500

10 m depth contour (kW m–1)

P10 = 0.88 P50– 0.00035 P 2

wave power density at

50
400

300

200

100

0 100 200 300 400 500

wave power density at 50 m depth contour (kW m–1)

Figure 2. Comparison of net incident wave power density at the 10 and 50 m deep wave berths.

calculation method, is from deep offshore to 50 m, the more relevant results are
those from 50 to 10 m deep. Figure 2 shows the directional net wave power density
at 10 m depth plotted against that at 50 m. In the smaller sea states typically less
than around 100 kW m−1 , the reduction at 10 m depth is only 10 per cent less
than at 50 m. However, as the seas increase, the reduction increases owing mainly
to wave breaking. The high occurrence of the smaller seas is clearly shown by the
density of the data points.
Although this analysis of the wave energy resource is indicative, other locations
will be different. Differences will arise owing to factors such as the annual
variability of the wave climate and the horizontal distance between the 50 m
and 10 m depth contours. A highly variable wave climate is likely to have a lower
proportion of exploitable wave energy, while a large distance between the 50
and 10 m depth contours is likely to result in a larger reduction in wave energy
resource. However, it is possible to conclude when comparing the nearshore wave
resource characteristics in the 10–20 m depth range to that at a depth of 50–100 m
that at the nearshore site:

— there is less directional spread of longer and medium period waves;

— the largest waves are limited in height owing to wave breaking; and
— the exploitable wave energy resource is not necessarily significantly
reduced.

3. The hydrodynamic design of oscillating wave surge converters

OWSCs are single degree of freedom systems, where typically the body motion is
rotation about a hinge aligned orthogonally to the direction of wave propagation.
Power is extracted by resisting the motion of the body induced by the action
of the waves. A significant body of work exists that details the performance of
single degree of freedom wave energy converters, for example, Parks [18], Evans
[19] and Falnes [20]. In particular, theories based on potential flow models have
shown that in the frequency domain, the hydrodynamics can be represented using
frequency-dependent wave force, added inertia and damping terms as shown in
equation (3.1), where the external loads are approximated using a linear spring

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

350 T. Whittaker and M. Folley

and damper,
F = [k − (I + Ia )u2 + j(B + L)u]X , (3.1)
where F is the complex wave force, k is the hydrostatic spring stiffness, I is the
OWSC inertia, Ia is the added inertia, B is the hydrodynamic damping coefficient,
L is the PTO damping coefficient, u is the wave frequency, X √is the complex
OWSC amplitude of motion and j is the imaginary number, j = −1.
The power capture, P, can then be calculated using equation (3.2),
P = 12 Lu2 |X |2 . (3.2)
Assuming for now that the motion is unconstrained and defining the natural
frequency, un , as un2 = k/(I + Ia ), the maximum power capture can be shown to
be given by equation (3.3),
1 |F |2
P=  . (3.3)
4 B + B 2 + (I + Ia )2 (un2 − u2 )2

It can be seen from equation (3.3) that the power capture is influenced by four
parameters. However, it is well known that the wave force and hydrodynamic
damping coefficient are linked by the Haskind relation (e.g. Falnes [20]), leaving
only three independent parameters. The influence of the inertia is of less
interest, but it can be seen that the power capture increases with an increase
in wave force, F , and at wave frequencies close to the natural frequency (un2 ≈
u2 ). Unfortunately, it is not possible to define these parameters arbitrarily to
maximize the power capture because they are both linked through the OWSC
hydrodynamics. To understand how they are linked, each of the parameters is
considered independently before looking at potential OWSC configurations that
involve compromises in these two key parameters.
To understand the hydrodynamics of the OWSC wave force, it is common to
consider the two-dimensional case of the wave force on a vertical wall, which is due
to the rate of change in momentum of the water particles as they are reflected.
In this idealization, the wave force is in-phase with the horizontal velocity of
the wave. However, such a two-dimensional idealization is misleading as the
OWSC would have to be several wavelengths wide, which is generally not the
case. In the three-dimensional case of a vertical wall of finite width, the wave
force is still owing to the change in momentum, but the ability of the water
particles to move around the wall must also be considered. It is well known (e.g.
Newman [21], p. 301) that when a body is small relative to the wavelength (the
long-wave approximation), then the surge wave force will be in-phase with and
proportional to the horizontal fluid particle acceleration of the wave, h̄. The long-
wave approximation also indicates that the wave force, F , is proportional to the
sum of the mass of water displaced by the body, M , plus the hydrodynamic added
mass, Ma . These relationships are summarized by
F ≈ (M + Ma )h̄. (3.4)
Based on this more fundamental understanding of the wave force, it is useful
to consider how the wave force may vary with key design parameters. The first
significant design parameter is the water depth. Folley et al. [22] have shown that

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 351

3.0

surge wave force ratio

2.5

2.0

1.5

1.0

0.5

0 0.5 1.0 1.5 2.0 2.5 3.0

non-dimensional water depth, kh

Figure 3. Change in surge wave force (relative to deep water surge wave force) with water depth
for a floating hemisphere. Solid line, approx.; filled squares, h/R = 2; crosses, h/R = 4; filled circles,
h/R = 8.

40
OWSC wave torque (MNm m–1)

30
wave torque µ flap width2
20

10

0 5 10 15 20 25 30 35
OWSC width (m)

Figure 4. Variation in wave torque with flap width.

as the non-dimensional water depth (defined as the product of the wavenumber

and water depth (kh)) decreases, there is an increase in wave force owing to
shoaling and associated increase in the horizontal acceleration of the wave, as
shown in figure 3.
Another key design parameter is the OWSC width. Figure 4 shows the effect
of OWSC width on the wave force for a surface-piercing flap in 12 m water depth.
The change in wave torque is related to an increase in added inertia, which itself
is associated with the change in accelerations induced on water particles owing
to the presence of the OWSC. Although in the two-dimensional limit, the wave
force will increase linearly with the OWSC width, the wave force for OWSCs of
finite width will increase initially with the square of the width before tending
towards the two-dimensional limit; thus, generally, wave force increases more
rapidly than OWSC width. The final key design parameter considered for the
wave force is the submergence of the OWSC. It has already been noted that
the wave force is associated with the induced wave particle accelerations owing
to the OWSC. In comparison with a surface-piercing OWSC, the induced wave

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

352 T. Whittaker and M. Folley

20

wave torque (MNm m–1)

15

10

0
4 6 8 10 12 14
wave period (s)

Figure 5. Example reduction in wave torque with submergence. Solid line, surface piercing; dashed
line, 0.1 m submerged; dotted line, 0.2 m submerged; dashed-dotted line, 0.5 m submerged.

particle accelerations owing to a fully submerged OWSC will be smaller, resulting

in a reduction in the wave force. Figure 5 shows the rapid reduction in wave torque
on an 18 m wide OWSC in 12 m water depth owing to submergence.
The above analysis uses the same assumptions as virtually all publications
about wave forces on wave energy converters; however, another source of wave
force is the drag force owing to the relative motion of the wave and OWSC.
Typically, drag forces are assumed to be proportional to the relative velocity
squared and so could dominate at large velocities or where other wave forces
are small. Difficulties with determining the drag coefficient, which defines
the relationship between the force and velocity squared, together with the
nonlinearity of the system, means that calculating the power capture of OWSCs
driven by drag forces is difficult. Unlike potential flow models, the wave force will
reduce as the body moves owing to a reduction in the relative velocity.
Now consider the natural frequency of an OWSC. In general, this is dependent
on the ratio of the OWSC pitch stiffness and the total OWSC rotational inertia,
including the added inertia. Assuming that the water plane area has a negligible
influence, the OWSC pitch stiffness is dependent on the excess buoyancy and
its distance to the OWSC hinge. For OWSCs with the same relative vertical
excess buoyancy distribution, this means that the pitch stiffness will increase
linearly with the OWSC width and quadratically with the OWSC height. The
total OWSC rotational inertia will, in general, increase with OWSC height and
width at a rate greater than that of the OWSC pitch stiffness. For the OWSC
height, rotational inertia increases with the cube of the distance from the hinge,
while for the OWSC width, it is associated with the increase in added inertia. The
added inertia increases more rapidly than the OWSC width for the same reason
that the OWSC wave force also increases more rapidly: because of the increase
in induced water particle accelerations. Thus, in general, the natural frequency
of an OWSC will decrease with increasing OWSC height and width as shown in
figure 6 for a 2 m thick surface-piercing OWSC.
Having identified the key hydrodynamic parameters of wave force and natural
frequency, and related these parameters to some key design parameters, it is
possible to consider the potential of a number of different regions of the design
space. While the design space is a continuum, it is possible that a number of

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 353

20
24
18

20
16
water depth (m)
12
16
24
14 20

16
12

12
20
10 16
8 12
8
16
6
5 10 15 20 25 30
width of OWSC (m)

Figure 6. Variation in natural pitching period (contour lines) with OWSC height (water depth)
and width for a surface-piercing OWSC.

10 water surface
OWSC height (m)

0
5 10 15 20
OWSC width (m)

Figure 7. Variation of OWSC height with width to obtain a natural period of 10 s.

regions with local optima exist. These are typically associated with adoption of
a particular design principle, either consciously or otherwise. Recognizing that
OWSCs can be designed based on different design principles can help explain
why different designs exist.
First consider a region of the design space where the OWSC is tuned. For this,
initially consider an OWSC that has a natural pitching period of 10 s so that
it is well tuned to the incident waves for the North Atlantic. Figure 7 shows the
relationship required between OWSC width and height to obtain a natural period
of 10 s for a 2 m thick rectangular OWSC, with a constant density of 300 kg m−3
located at a water depth of 12 m with the hinge axis 3 m above the sea bed. Some
examples of these OWSCs are shown in figure 8. It can be seen that the surface-
piercing OWSC (flap A) has a width of only 4.4 m and that a wider OWSC
(flap B) must be submerged to obtain the specified natural frequency. However,
these OWSC dimensions required to obtain an appropriate natural frequency
mean that the wave force will be small. An associated problem with the small wave
force is that the optimum amplitude of motion at resonance is large. Consequently,
when motion constraints are included, the increase in power capture owing to

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

354 T. Whittaker and M. Folley

A
B C

Figure 8. Three examples of tuned flaps with very different geometries. (Online version in colour.)
maximum power capture (kW)

1400
1200
1000
800
600
400
200
0
5 10 15 20
wave period (s)

Figure 9. Constrained and unconstrained power capture of a submerged 12 m wide flap tuned to
be resonant at 10 s. (Wave amplitude is 1 m, maximum rotation is 0.5 radians (30◦ ).) Dashed line,
unconstrained; solid line, constrained.

tuning is often not evident, as discussed by Folley et al. [22]. This is illustrated
in figure 9, where it can be seen that constraining the maximum amplitude of
motion to a reasonable value completely eliminates the peak in maximum power
capture at the OWSC’s natural frequency.
Thus, in this area of the design space, the requirements for tuning and wave
force appear incompatible. That is, to obtain the required natural frequency,
the wave force becomes small, this then means that the optimum amplitude of
motions are large and so the increase in power capture owing to tuning cannot
be exploited owing to motion constraints.
To avoid the problems with motion constraints, the water depth can be
increased together with the buoyancy force to maintain the OWSC’s natural
frequency (flap C in figure 8). While this would improve the power capture and
enable the benefits associated with tuning to be exploited, the modifications
come with a set of associated problems. In particular, the increased water
depth would increase the structural task significantly, both because of the
larger torques and the larger distances over which they must be transmitted.
In addition, the increased buoyancy required to obtain the required natural
frequency would increase the loads on the anchor significantly. However, it
appears to be the only viable solution that exploits the benefits of a tuned wave
energy converter.

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 355

100

reactive force coefficient

80

(MNm rad–1)
60

40

20

0
4 6 8 10 12 14
wave period (s)

Figure 10. Reactive force coefficients versus wave period (50 m wide flap) shows multiple resonances
as the inertial force equals the spring force. Solid line, inertia; dashed line, stiffness.

A second region of the design space can be defined by ensuring that the OWSC
wave force is large. The wave force can be maximized by ensuring that the
OWSC is deployed in relatively shallow water, that the OWSC is surface piercing
and that it is relatively wide. A possible configuration would be a 2 m thick, 18 m
wide rectangular OWSC deployed in 12 m of water depth. With a constant
density of 300 kg m−3 , this configuration has a natural period of around 20 s, well
outside the incident wave frequencies. However, although the theoretical
maximum power capture cannot be achieved in this configuration because
it is not optimally tuned, the capture factor remains high. Both numerical
and physical models of this configuration have obtained capture factors over
0.7 in irregular waves, with an average capture factor for a wave climate of
around 0.5. Thus, although the configuration is sub-optimal, the power capture
remains significant.
An additional, interesting, region of the design space emerges if the OWSC
width is increased to about 50 m and the thickness to about 5 m. In this region,
it is possible to obtain multiple natural frequencies of the OWSC covering the
full range of incident wave frequencies, as illustrated in figure 10. These multiple
natural frequencies occur because the added inertia is frequency dependent,
allowing multiple solutions for the natural frequency. Physically, the variation
in added inertia occurs because of displacement of the local water surface
with flap rotation. This effect can also be deduced from the oscillatory nature
of the impulse response function, which at frequencies close to the frequency
of the impulse response function’s oscillation results in very low torques in-
phase with flap rotational acceleration. Numerical models of this configuration
indicate that it should be possible to capture very nearly (greater than 99%) the
maximum theoretical power capture for this device width (which is greater than
for an idealized surging point absorber owing to the finite device width) across
the whole frequency range without the need for any reactive energy control.
In addition, the large wave forces for such a wide OWSC mean that motion
constraints are typically not an issue. However, although the maximum theoretical
power capture is nearly possible, the capture factor is less than for the un-
tuned OWSC with half the width. This occurs because the maximum capture

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

356 T. Whittaker and M. Folley

factor of the wider OWSC is beginning to be affected by the two-dimensional

limit of 0.5, while the maximum capture factor of a narrower OWSC is more
influenced by the point absorber effect, which enables the capture factor to be
greater than 1.
The final region of the design space to be considered is defined by the wave force
being dominated by drag. To maximize this force, the OWSC should be deployed
in relatively shallow water to increase the horizontal motions of the waves. An
OWSC driven by drag will have relatively small motions because they will always
be less than the horizontal motions of the waves. Although it is possible for drag
forces to dominate in large bodies, it is likely that OWSCs in this region will be
relatively small with relatively small power captures.
These design space regions are not meant to be exclusive or exhaustive and
clearly alternatives exist. However, they do illustrate that different locally optimal
designs may exist and depend on how the OWSC is envisioned to operate. It is
possible to see how the designs of the current OWSC concepts being commercially
developed can be placed in different regions of the design space, although it is
generally unclear how conscious the developers are of the choices that they have
made and how they have optimized their designs.

4. The development of Oyster

(a) Design philosophy

When designing a wave energy convertor, in addition to understanding the
different hydrodynamic regions of the design space, there are a range of
fundamental principles that should be adhered to if cost-effective reliable
machines are to evolve. The biggest challenge is how to extract energy
economically from a highly variable resource, which has extremes several orders of
magnitude greater than the mean and without resorting to substantial structures
that are redundant for most of the time. This design philosophy was originally
proposed by Whittaker & Folley [23]; the ideal characteristics identified are
summarized below:

(1) a wide bandwidth response so that capture width is maintained over the
working frequency range instead of being highly tuned to a particular
frequency;
(2) a progressive decoupling from the incident wave excitation as the wave
height increases so that in extreme seas, the capture efficiency is very
low, thus, a relatively modest installed capacity can cope with the power
conversion;
(3) high structural efficiency ensured by the shortest load paths from the wave
excitation force through the PTO system into the reaction platform;
(4) a machine in which all elements are essential to the power conversion
process with minimal redundancy, except where duplication is deemed
necessary to maintain reliability; and
(5) a ‘plug and play’ approach to PTO sub-assemblies, which are most prone
to failure.

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 357

(b) Oyster 1
The conclusion from the EPSRC-funded research programme [24] was that
the simplest type of device that meets these criteria is a flap structure with its
bottom edge hinged to the sea bed. As it rotates to larger angles, the vertical
projected frontal area reduces, giving progressive decoupling as the wave force
increases (characteristic 2). The sea bed provides the frame of reference and
avoids the need for a dynamic reaction, which adds complexity to the system
(characteristic 4). Although attaching to the sea bed is the most difficult part of
the operation, it results in a stream of advantages such as the avoidance of flexible
power transmission connections and allows tight geometric spacing of units within
the wave farm, thus maximizing the extraction of the wave resource over a given
sea bed footprint. The solid framework means that a ‘plug and play’ approach is
more easily achieved (characteristic 5).
Refining the optimum OWSC design further, then the design space of
most promise is that where the wave force is maximized. It has been shown
above that this region of the design space has a broad bandwidth response
(characteristic 1) and relatively short load paths (characteristic 3). Conversely,
the design space for tuned OWSCs will almost certainly, by definition, have
a narrower bandwidth (contravening characteristic 1) and/or longer load paths
(contravening characteristic 3), which were shown to be necessary to exploit the
benefits of the tuning. An OWSC driven by drag forces will experience larger
forces as the waves get bigger owing to the quadratic relationship of the force with
water velocity (contravening characteristic 2), although wave force decoupling
at large angles may reduce this effect. The final region of the design space
identified above is that containing the multi-resonant OWSCs. While this region
also satisfies characteristics 1 and 3, the capture factor is typically smaller. Thus,
although the design space should not be discarded, it is sensible to first focus on
the region of the design space where wave force is maximized.
Having identified a region of the design space, the first consideration is water
depth. Being a surging device, it is essential to tap the amplified horizontal fluid
particle motion as a wave climate shoals in shallow water. However, figure 11
shows that below around 10 m depth, an ocean wave with an average period of
10 s and a significant wave height of 4 m will lose a significant proportion of its
power owing to breaking. A significant wave height of 4 m or less represents the
vast majority of sea states that occur along the western seaboard of the North
Atlantic. Taking a target tidal range of 4 m or less, a nominal mean water depth of
12 m was chosen for Oyster 1 and continues to be the preferred depth for Oyster 2.
An extensive research programme comprising a combination of physical and
mathematical modelling has been undertaken at Queen’s University to determine
the relationship between the physical dimensions of the flap, its dynamic
characteristics, the PTO damping characteristics and power conversion in a
wide range of seas [25]. Figure 12 shows an example of effect of flap width
on capture factor in a range of sea states with different energy periods (Te)
and power flux (P). In all but the shortest period waves, the capture factor
increases with increasing flap width, shown here from 9 to 36 m. The reduction
in capture factor for the 36 m wide flap at Te of 7 s is a result of the high
width to wavelength ratio that results in the flap starting to behave as a two-
dimensional device, which ultimately limits the capture factor to 0.5. This also

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

358 T. Whittaker and M. Folley

50

net wave power density (kW m–1)

45

40

35

30

25
50 40 30 20 10
water depth (m)

Figure 11. Reduction in net wave power with water depth over various bed slopes. Solid line, slope
1 : 50; dashed line, slope 1 : 100; dashed-dotted line, slope 1 : 200; dotted line, slope 1 : 500.

0.8
width = 9 m
0.7 width = 12 m
width = 18 m
0.6 width = 24 m
capture factor

0.5

0.4

0.3

02

0.1

0
Te = 7 s Te = 9 s Te = 9 s Te = 9 s Te = 11 s Te = 11 s
P = 5 kW m–1 P = 7 kW m–1 P = 41 kW m–1 P = 70 kW m–1 P = 16 kW m–1 P = 61 kW m–1

Figure 12. Variation of capture factor with sea state and flap width. (Online version colour.)

indicates that there will be a limit to the performance enhancement of flap

width, with even wider flaps only gaining in the longer period seas and losing
in the shorter ones. A second observation is that performance reduces as the sea
power increases for a given energy period, demonstrating the OWSC’s ability
to progressively decouple. It should be noted that the capture factors presented
in this figure are indicative of what can be achieved, but can be increased or
decreased depending on factors such as flap thickness, freeboard, pitch stiffness
and damping torque characteristics.
However, the cost of power cannot be minimized by optimizing hydrodynamic
performance alone. Another key factor for sea bed reacting devices is foundation
loads. Unfortunately, increases in power output per flap are accompanied
by increases in both surge and heave loads on the sea bed connections.

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 359

24 m wide
HPS
16 m depth

max surge load

24 m wide 28 m wide
HPS HPS
13 m depth 13 m depth
18 m wide 28 m wide
MPS MPS
13 m depth 13 m depth
24 m wide
MPS
24 m wide 13 m depth
LPS
13 m depth

average power output of flap

Figure 13. Variation in maximum surge load and average power capture for a range of OWSC
configurations.

24 m wide
max total heave load

HPS
24 m wide
16 m depth
HPS
13 m depth 28 m wide
HPS
24 m wide 13 m depth
18 m wide MPS
MPS 13 m depth 28 m wide
13 m depth MPS
13 m depth

24 m wide
LPS
13 m depth

average power output of flap

Figure 14. Variation in maximum heave load and average power capture for a range of OWSC
configurations.

Figures 13 and 14 show how surge and heave loads, together with average
hydraulic power in a weighted range of seas can be significantly changed by flap
width, pitch stiffness (high, HPS; medium, MPS; low, LPS) and water depth.
Again, these results are only indicative and can be varied significantly by flap
shape and damping characteristic.
The above presentation gives a very small sample of the interplay between
the many variables that govern the performance and cost of OWSC devices.
Sea bed hinged flaps might seem to be the simplest of wave power converters;
however, their design optimization, even from a purely hydrodynamic viewpoint
is challenging. Then, it must be remembered that the hydrodynamics is only
one part of the optimization process that includes structure, secondary power
conversion and transmission, installation and finally maintenance.

(c) The deployment of Oyster 1

The general basis of design for Oyster 1 was fixed in 2007 and much of the work
reported above has been conducted since then. Aquamarine Power successfully

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

360 T. Whittaker and M. Folley

Oyster wave
energy converter hydroelectric
power conversion plant

high-pressure
flow line
water piston

Figure 15. General schematic of the Oyster concept. (Online version in colour.)

Figure 16. The offshore Oyster 1 device. (Online version in colour.)

installed the Oyster 1 full-scale proof-of-concept device at EMEC in Orkney in

the summer of 2009 and used it to inform the design of Oyster 2, which was
installed in the summer of 2011.
Oyster 1 is basically a wave-powered pump driving water at high pressure
through pipelines to a high-head hydroelectric plant onshore. The general layout
of Oyster 1 is shown in the schematic of figure 15.
The flap structure and general arrangement of the offshore PTO components
are shown in figure 16. The body of the flap was constructed from five 1.8 m
diameter steel tubes. The 18 m wide, 11 m high flap was secured to the sea bed
by drilled and grouted piles in a connector frame. A general view of the onshore
hydroelectric plant is shown in figure 17. The container with open doors houses
the Pelton wheel, flywheel and generator. The container to the left houses the
electrical power converters and computer monitoring systems.

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 361

Figure 17. Oyster 1 onshore power station. (Online version in colour.)

(d) Offshore power take-off components

The wave-induced movement of the flap is converted into pressurized water
hydraulic energy using two double-acting hydraulic cylinders mounted between
the flap itself and the base frame structure. The flap’s resistance to motion,
or damping torque, is determined by the internal area of the two cylinders,
the lever arm that governs stroke and the water pressure. The device does not
use any continuous ‘wave-by-wave’ control of the damping torque, but energy
capture can be optimized by slowly adjusting the system target pressure at the
onshore hydroelectric plant, which changes the damping torque. High-pressure
water is transmitted to the onshore hydroelectric plant through conventional
directionally drilled pipelines. A closed loop is used, which has been found to
be more economically and technically attractive than pumping sea water because
it avoids the challenges of offshore filtration, corrosion, bio-fouling and location
of the discharge piping.

(e) Onshore hydroelectric plant

The onshore hydroelectric plant comprises largely standard components
combined and controlled in a novel manner. The power plant uses a variable
speed induction generator coupled with a Pelton wheel turbine and flywheel. The
flywheel is the primary source of energy storage in the Oyster power train and acts
to smooth out the delivered power over a wave cycle and significantly reduces the
required generator capacity. A relatively simple control system operates the plant
efficiently and safely. The system pressure is regulated by the spear valves that
control the flow of high-pressure water onto the Pelton wheel. They are adjusted
continuously to keep the average operating pressure in the system as close as
possible to the optimum target pressure for the sea state, while simultaneously

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

362 T. Whittaker and M. Folley

keeping the ratio of the spear valve nozzle velocity and the Pelton wheel bucket
speed close to its optimal value. The response time of the spear valves must be
sufficiently fast to limit pressure fluctuations during each wave cycle. Changes in
target pressure occur over much longer time frames according to changes in the
incident wave climate. The final regulation of the variable speed generator output
is by a power convertor/electronics that provide the necessary full rectification
and inversion before a step-up transformer supplies power to the grid.
Following commissioning, the device produced first power in October 2009
and provided invaluable operational data regarding device performance and
loading, which is being used for numerical and wave-tank model calibrations
and refinements. It was decommissioned in 2011 when Oyster 2 was installed.
Aquamarine Power has gained a phenomenal amount of knowledge regarding
the design, construction, installation, maintenance and operation of this type of
system. The sea trials to date have shown that power conversion and delivery
to the grid using an OWSC is technically possible. As can be expected with any
new technology, even with one using mainly ‘off the shelf’ components, there are
technical challenges with some components in the PTO system. Parts have had
to be redesigned, manufactured and retrofitted. With Oyster 1, it was decided
early in the design process to undertake component maintenance in situ rather
than detach the flap from its sea bed frame and take it to harbour.
This has been possible because of one of the most important features of Oyster:
the ability to water ballast the flap and sink it onto its base frame where it is
latched. For health and safety reasons, this is an essential requirement for diver
maintenance and inspection. In its first year of sea trials, Oyster 1 has survived
very severe wave conditions in both the operational and locked down positions. It
is the only wave power device in the world that has been deliberately sunk several
times in a fully controlled manner for maintenance. Component refinement and
replacement was ongoing during the operational period, and was an essential part
of this sea trial.

5. Conclusions

The deployment of Oyster 1 at EMEC is a world first in the sense that previously
nobody had ever fixed a mobile machine of this size to the sea bed in shallow
coastal waters. Its first year of operation at sea has shown that it is technically
possible to convert ocean wave power into electrical energy delivered to a national
grid. Consequently, OWSCs located in the shallow nearshore region with depths
as low as 10 m are serious contenders in the challenge to produce economic,
reliable power from ocean waves. This conclusion is supported by the following:

— the exploitable nearshore wave resource at a nominal depth of 10 m is only

10–20% lower than that at 50 m for sites with a 1 : 50 bed slope;
— for surging devices located in the nearshore region, this is offset by
exploitation of the amplified horizontal wave-induced fluid motion owing
to shoaling, the reduced directional spread in the seas and filtering out of
the extreme seas owing to breaking;
— devices such as Oyster are point absorbers that primarily respond to the
horizontal fluid acceleration in the working frequency range, Oyster is not
a drag device responding to horizontal fluid velocity;

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

Nearshore OWSCs and Oyster development 363

— it had been found that resonant OWSCs do not realize their theoretical
potential owing to the motion constraints of ‘real’ devices and consequently
do not meet the set of optimal design criteria that have been proposed;
— Oyster is a full-depth flap from above the water surface to the sea bed. It
is not resonant in the working frequency range and maximizes the wave
excitation force, while maintaining an acceptable angular velocity to limit
vortex losses at the edges;
— OWSCs have a natural survival characteristic in that they progressively
decouple from waves as they become larger and the flap oscillation angles
from the vertical increase;
— one of the most important features of Oyster has proved to be the ability
to water ballast the flap to sink it onto its base frame, where it is locked
to enable safe diver access for maintenance; and
— OWSCs cannot be optimized on hydrodynamics alone, structure,
secondary power conversion and transmission, installation and finally
maintenance over decades must be considered to develop cost-effective
reliable wave power converters.

This has been a phenomenal journey from the initial thoughts discussed at coffee in 2001 through
the initial EPSRC grant that funded the initial research, the formation of Aquamarine Power Ltd
by Allan Thompson in 2005 to develop the Oyster concept, the deployment of the 315 kW Oyster 1
at EMEC in 2009 and now the deployment of Oyster 2. Hundreds of people have been involved.
However, none of this would have been possible without very substantial funding from both the
public, but mainly the private sectors. The final acknowledgement must go the late Prof. Alan Wells
FRS, who founded the wave power group at Queen’s University in the mid 1970s and devoted much
of the last 30 years of his life to the development of wave power technology.

References
1 Budal, K. & Falnes, J. 1978 Wave-power conversion by point absorbers. Norwegian Marit. Res.
6, 2–11.
2 Korde, U. A. 2000 A note on the hydrodynamics of a tail tube buoy. Ocean Eng. 27, 1473–1484.
(doi:10.1016/S0029-8018(99)00053-0)
3 Evans, D. V. 1981 Maximum wave-power absorption under motion constraints. Appl. Ocean
Res. 3, 200–203. (doi:10.1016/0141-1187(81)90063-8)
4 Falcão, A. F. de O. 2008 Phase control through load control of oscillating-body wave energy
converters with hydraulic PTO system. Ocean Eng. 35, 358–366. (doi:10.1016/j.oceaneng.
2007.10.005)
5 Cruz, J., Sykes, R., Siddorn, P. & Taylor, R. E. 2009 Wave farm design: preliminary studies
on the influences of wave climate, array layout and farm control. In 8th European Wave and
Tidal Energy Conf., Uppsala, Sweden, September 2009. EWTEC.
6 Whittaker, T., Beattie, W. C., Folley, M., Boake, C., Wright, M. & Osterried, M. 2003
Performance of the LIMPET wave power plant: prediction, measurement and potential. In
5th European Wave Energy Conf., Cork, Ireland, September 2003.
7 Tjugen, K. J. 1993 TAPCHAN ocean energy project. In Proc. European Wave Energy Symp.,
Edinburgh, UK, July 1993. European Commission EUR 15571: 265–276.
8 Folley, M. & Whittaker, T. J. T. 2009 Analysis of the nearshore wave energy resource. Renew.
Energy 34, 1709–1715. (doi:10.1016/j.renene.2009.01.003)
9 Folley, M., Elsaesser, B. & Whittaker, T. 2009 Analysis of the wave energy resource at
the European Marine Energy Centre. In Coasts, Marine Structures and Breakwaters Conf.,
Edinburgh, UK, September 2009. ICE.

Phil. Trans. R. Soc. A (2012)

 Downloaded from http://rsta.royalsocietypublishing.org/ on May 20, 2015

364 T. Whittaker and M. Folley

10 Aquamarine Power Ltd. 2010 Oyster. See http://www.aquamarinepower.com.

11 AW-Energy Oy. 2010 WaveRoller. See http://www.aw-energy.com.
12 BioPower Systems Pty. Ltd. 2010 bioWAVE. See http://www.biopowersystems.com.
13 Engineering Business Ltd. 2005 EB Frond wave energy converter – phase 2. DTI Report
URN05/865.
14 Farley, F. J. M. 1982 Wave energy conversion by flexible resonant rafts. Appl. Ocean Res. 4,
57–63. (doi:10.1016/S0141-1187(82)80022-9)
15 Langlee Wave Power. 2010 Langlee wave power converter. See http://www.langlee.no.
16 Salter, S. 1974 Wave power. Nature 249, 720–724. (doi:10.1038/249720a0)
17 French, M. J. 1985 Conceptual design for engineers, 2nd edn. London, UK: Springer.
18 Parks, P. C. 1978 Some simple mathematical models of wave power machines. In UK
Department of Energy workshop on wave energy, London, UK, November 1978. ETSU.
19 Evans, D. V. 1985 The hydrodynamic efficiency of wave-energy devices. In Hydrodynamics of
ocean wave-energy utilisation, Lisbon, December 1985, pp. 1–34. Springer-Verlag.
20 Falnes, J. 2002 Ocean waves and oscillating systems. Cambridge, UK: Cambridge University
Press.
21 Newman, J. N. 1977 Marine hydrodynamics. Cambridge, MA: MIT Press.
22 Folley, M., Whittaker, T. J. T. & Henry, A. 2007 The effect of water depth on the performance
of a small surging wave energy converter. Ocean Eng. 34, 1265–1274. (doi:10.1016/j.oceaneng.
2006.05.015)
23 Whittaker, T. & Folley, M. 2005 Optimisation of wave power devices towards economic wave
power systems. In World Renewable Energy Congress, Aberdeen, UK, May 2005, pp. 927–932.
Elsevier.
24 Whittaker, T., Folley, M., Causon, D. M., Ingram, D. M. & Mingham, C. G. 2005 An
experimental and numerical study of oscillating wave surge converters. EPSRC report
GR/S12326/01.
25 Henry, A., Doherty, K., Cameron, L., Whittaker, T. & Doherty, R. 2010 Advances in the design
of the Oyster wave energy converter. In Marine and Offshore Renewable Energy Conf., London,
UK, April 2010. Royal Institute of Naval Architects.

Phil. Trans. R. Soc. A (2012)