Testing of a power take-off system for an OWC spar-buoy wave energy converter

J. C. C. Henriques , R. P. F. Gomes , L. M. C. Gato , A. F. O. Falco , J. C. C. Portillo , E. Robles and S. Ceballos

1

LAETA, IDMEC, Instituto Superior Tcnico, Universidade de Lisboa, Av. Rovisco Pais,1049-001 Lisboa, Portugal
2
Energy and Environmental Division, Tecnalia Research and Innovation, Derio 48160, Spain

System description and objectives

Generator control

NAREC large scale testing

Time series

I The OWC spar-buoy is an axisymmetric device consisting basically of a

submerged vertical tail tube open at both ends, fixed to a floater that
moves essentially in heave. The oscillating motion of the internal free
surface relative to the buoy, produced by the incident waves, makes the air
flow through a new type of self-rectifying air turbine, the biradial turbine.
I The work concerns the development of the power take-off (PTO) control of
an OWC oscillating-water-column (OWC) spar-buoy wave energy converter.
a)

I The objective is to increase the PTO system efficiency and reduce the

b)

OWC spar-buoy at NAREC at 1/16th scale.

overall costs, an electrical generator was adopted with a rated power twice

Tecnalia testing

the maximum expected average power output of the buoy.

Compared efficiency t of the biradial, Wells and axial flow impulse

(fixed guide vanes) turbines versus flow rate coefficient /|max ,
where |max is the flow rate coefficient at the point of maximum
efficiency for each turbine.

I This level of generator rated power poses great challenges for the PTO
control due to the irregular characteristics of the sea waves.
I The tests were performed using an hardware-in-the-loop configuration. The
hydrodynamics of the OWC spar-buoy and the aerodynamics of the air

a)

Results obtained for tests 55 and 23, both for a significant wave height of Hs = 6 m
and an energy period of Te = 12 s. Valve strategy for both tests was T1. Tests 55
and 23 use control laws A5 and A1, respectively.

b)

turbine were numerically simulated in real time and coupled to the physical

Conclusions

model of the turbine/electrical generator set. The instantaneous air turbine

I It was shown that it is quite challenging to simultaneously control the

torque is emulated through the use of the electrical motor.

Hardware-in-the-loop simulation and data logging

I The control of this set was the object of tests performed at Tecnalia

TECNALIA test rig

c)

d)

power of twice the expected average annual power extraction.

PTO test configuration.

Electrical PTO Laboratory, in Bilbao, Spain.

I In the reported implementation, only irregular wave conditions were
considered.
I The experimental results allowed the dynamic behaviour of the PTO to be
characterized, in order to ensure the practical applicability of the proposed
control algorithms and provide a basis for the validation of the numerical

2x
2x
d
d
2
1

+ F dr
(m1 + A
)
+
A
=

gS
x
+
p
S
p
w 1 1
at 2
12
12
11
dt2
dt2
2x
2x
d
d
1
2
+ F dr
)
(m
A
+
=

gS
x

p
S
p
+
A
w 2 2
at 2
2
21
22
21
dt2
dt2

d
dp
(p + 1) (x1 x2) + (h0 + x1 x2)
=
dt
dt
1
Qt
(p + 1)

S2

models.

d 1 2
I = Pt Pg
dt 2

b)

(1)

In the hardware-in-the-loop tests, the rotational speed of Eq. (1)

I Through the use of a relief valve in parallel with the turbine, it is possible to
control the turbine/generator set rotational speed and the pneumatic power
Generator power control laws. a) The basic control law, PA1 =
a b, obtained from exponential regression of the maximum power
extraction computed for a set of sea-states characteristic of the
wave climate off the Portuguese west coast. b) Basic control law
combined with two hysteresis loops where the maximum value is
rated. c) The curve used in a) with
the generator rated power, Pgen
the maximum power clipped to the generator rated power. d)
Modification of curve plotted in c), where the generator power
increases smoothly from zero to the basic control law between
500 rpm and 800 rpm.

is replaced by the experimental values. The floater is body 1 and

the oscillating water column (OWC) is body 2, and the main vari-

a)

OWC spar-buoy geometry and biradial turbine representation.

I The use of a high-speed stop valve was not entirely successful since it
resulted in large pneumatic power peaks.
I The delay imposed by the control of the generator should be taken into
account for highly energetic sea conditions, where fast transients may occur.
The value of the delay was found to be 150 ms.
I Although not apparent from the shown results, simulations of systems at

Results

level of friction torque of this particular test rig, since power losses would
not be scaled as net power.

A
ij - added mass hydrodynamic coefficient
D - diameter of the buoy at the free surface
F dr - excitation force plus radiation force
g - acceleration of gravity
p pat - relative pressure between air chamber and atmosphere
Pg - generator power
Pt - turbine power
Si - representative cross section of body i
t - time
Vc and h0 - air chamber volume, and air chamber height
xi, vi, ai and mi - position, velocity, acceleration and mass of
body i
at - air density at atmospheric conditions
w - water density
and I - rotational speed and turbine/generator set inertia

ICOE 2014 - International Conference on Ocean Energy - Halifax, Nova Scotia, Canada

available to the turbine.

smaller scales would be expected to reveal self-start problems due to the

ables are:

c)

rotational speed and the instantaneous power of the generator with a rated

I Future work should be focused on the use of the high-speed stop valve to
perform latching control, as well as on the application of Model Predictive
Control to determine the valve opening and closing instants, while taking
into account limitations of the generator.

Acknowlegements

Comparison of the results for tests 60, 62, 61, 63, 54 and 55,
performed using control law A5 and valve control strategy T1, for
three different significant wave heights, Hs = 6, 4 and 2 m, and
two energy periods, Te = 8 and 12 s.

The research was partially funded by the European Communitys Seventh Framework Programme under MARINET initiative. This work was also funded by the
Portuguese Foundation for Science and Technology (FCT) through IDMEC, under LAETA Pest-OE/EME/LA0022 and contracts PTDC/EME-MFE/103524/2008
and PTDC/EME-MFE/111763/2009. The second author was supported by postdoctoral fellowship SFRH/BPD/93209/2013 from the FCT.

luis.gato@tecnico.ulisboa.pt // http://www.facebook.com/OWCsparbuoy // Report at: http://waves.tecnico.ulisboa.pt/SPOWCON2013/MARINET-SPOWCON.pdf