Conference PaperPDF Available

Spatial Variation of Currents Generated in the FloWave Ocean Energy Research Facility

Authors:

Abstract and Figures

FloWave is a state of the art test facility which can produce combined waves and currents from any direction in a circular tank. Characterisation of this new facility is ongoing, with initial results from the flow generation measurements presented. This is a complex problem, considering different input velocities, 3D spatial variability of the flow in the X, Y, & Z directions, as well as temporal stability of the flow. In a circular tank, production of uniform flow is a non-trivial problem, however this has been achieved across a large test area using precise control of the individual drive units and specially designed turning vanes. This allows the testing of device models and small arrays, in controlled realistic sea conditions, prior to deployment at sea.
Content may be subject to copyright.
Spatial Variation of Currents Generated in the
FloWave Ocean Energy Research Facility
Donald Noble1, Thomas Davey2, Helen Smith3, Panagiotis Kaklis4, Adam Robinson§5, and Tom Bruce§6
FloWave Ocean Energy Research Facility, University of Edinburgh, UK
College of Engineering, Mathematics and Physical Sciences, University of Exeter, Penryn Campus, UK
Naval Architecture, Ocean and Marine Engineering, University of Strathclyde, UK
§Institute for Energy Systems, School of Engineering, University of Edinburgh, UK
1D.Noble@ed.ac.uk 2Tom.Davey@ed.ac.uk 3H.C.M.Smith@exeter.ac.uk
4Panagiotis.Kaklis@strath.ac.uk 5Adam.Robinson@ed.ac.uk 6Tom.Bruce@ed.ac.uk
Abstract—FloWave is a state of the art test facility which can
produce combined waves and currents from any direction in a
circular tank. Characterisation of this new facility is ongoing,
with initial results from the flow generation measurements
presented. This is a complex problem, considering different input
velocities, 3D spatial variability of the flow in the X, Y, &
Z directions, as well as temporal stability of the flow. In a
circular tank, production of uniform flow is a non-trivial problem,
however this has been achieved across a large test area using
precise control of the individual drive units and specially designed
turning vanes. This allows the testing of device models and small
arrays, in controlled realistic sea conditions, prior to deployment
at sea.
Index Terms—Tank testing, tidal current, vertical flow profile,
spatial variation, measurement, characterisation
I. INTRODUCTION
Physical scale model testing is an essential element in
the development of marine renewable technologies and tech-
niques. Laboratory testing provides a repeatable, controlled,
low-risk environment where technological concepts and oper-
ational techniques may be developed [1].
FloWave is a state of the art ocean energy research facility,
designed to provide large scale physical modelling services to
the tidal and wave energy sector. It has the unique ability
to provide complex multi-directional waves combined with
currents from any direction in the 25 m diameter circular tank.
As part of the commissioning and characterisation process
for this new facility it is important to investigate the per-
formance characteristics of the waves and current generation
capability, both individually and in combination. It is also
important to understand the shape and size of the usable test
area. The focus of this paper is on the generation of currents,
and specifically looking at spatial variation thereof.
A. About the facility
FloWave is a circular combined wave and current test
tank. Wavemakers are located around the entire circumference,
with impellers to drive the current recirculation mounted in a
plenum chamber below the test area, as shown in Fig. 1.
The tank is optimised for waves of around 2 s period, and is
capable of generating currents upwards of 1.6 m/s. This offers
A
B
C
D
E
2.0m depth
15m Ø oor
25m Ø tank
A
B
C
D
B
C
Fig. 1. Schematic of FloWave in plan and oblique section showing:
(A) Wavemaker paddles around circumference (168 Nr)
(B) Turning vanes and flow conditioning filters
(C) Current drive impeller units (28 Nr)
(D) Buoyant raisable floor (15 mØ) below test area
(E) Idealised streamlines of flow across tank floor
the ability to model metocean conditions for most renewable
energy devices at a typical scale of between 1:20 and 1:40 [2].
There is a 15 m diameter buoyant floor in the centre of the
tank, which notionally represents the test area. This floor can
be raised above the water level to facilitate model installation
and reconfiguration as required, then submerged to the 2 m
working depth.
Around the circumference of the tank there are 168 active-
absorbing hinged wavemakers. These are able to generate
regular and irregular waves, both long-crested and multi-
directional, as well as complex multi-modal sea states with
waves from multiple directions.
Currents are generated by 28 impeller units mounted in the
plenum chamber below the test floor. Each of these contains
a single 1.7 m diameter low-solidity 5-bladed symmetrical
impeller, driven by a 48 kW motor. Turning vanes mounted
below and in front of the wavemakers direct the current
across the tank [3], as shown in Fig. 1. These turning vanes
incorporate porous screens to provide flow conditioning and
prevent debris ingress to the plenum chamber.
Creating a horizontally uniform current in a circular tank is
a non-trivial matter, requiring precise control of the individual
impellers [4]. In summary, the impeller units on either side
of the required current direction on both the upstream and
downstream side of the tank are driven at varying speeds to
produce the required current corresponding to the desired test
velocity. The control system for the impellers includes the
facility to change the direction of the current during the test,
either to an arbitrary angle or rotating by a set angle every
minute. This capability allows for the simulation of cross-
currents, or a tidal ellipse, without having to reposition the
device model.
B. Device Testing and Scales
The development of new technologies generally follows an
iterative process, refining and developing the initial concept,
towards the goal of producing a viable product. A five-stage
structured development plan has been developed for for wave
energy systems [5], and this can be related to the Technology
Readiness Level (TRL) concept developed by NASA [6]. With
appropriate modifications, this process is also applicable for
other marine renewable energy devices plus the supporting
infrastructure, as captured by the EquiMar Protocols [1]. The
development stages are reproduced in Table I, together with
typical scales for marine renewable energy device testing. It
is important to note that development is not a linear ‘once-
through’ process, and that multiple loops through the different
stages are typical.
Tank testing usually fits into the early development stages,
proving preliminary concepts with small scale models and
refining designs with larger models that are more detailed or
more representative, before moving onto open water testing.
As noted above, the FloWave facility is optimised for models
around 1:40 to 1:20 scale, and so can be used for both concept
and design validation. Depending on the specifics of the device
and the constraints of the tank, both physical as well as the
wave and current generation, it is possible to test at a broader
range of scales.
It is not possible to accurately scale all physical phenomena
by the same factor when undertaking physical model testing
at a scale other than unity [7]. Given that gravitational forces
are likely to be dominant in problems involving a free surface
with waves, Froude scaling is used for most of the testing at
FloWave. This is one of two dimensionless scaling factors
commonly used in tank testing, the Froude and Reynolds
numbers, respectively the ratios between inertia/gravity forces
and inertia/viscous forces. In tank testing, both the small scale
model and full scale prototype are immersed in the same fluid
and are subject to the same gravitational force, therefore it is
not possible to satisfy both relationships simultaneously.
Testing closer to full scale reduces the impact of these
scaling effects, resulting in more accurate and representative
testing. It is also difficult to include power take off and control
systems in small models. Therefore testing physical models at
a larger scale is a valuable stage in the development process
between small scale models and open water testing.
Most marine renewable energy devices are intended to be
installed in arrays, with multiple devices in close proximity.
Therefore modelling inter-array effects is an important part of
the design process, e.g. assessing the impact on power capture.
The physical size of the facilities used to test marine renewable
energy devices will place limits the number of individual units
that can be tested in an array configuration. This is typically
in the region of 2 to 7, but will depend on the shape and size
of both the device and the test facility.
The results from physical model testing can then be used to
validated computer numerical models. These computer models
are typically used to simulate either performance at the level
of an individual component or device, or the interactions
between multiple devices which can be extended to cover large
arrays of devices in a variety of different conditions, subject
to sufficient computational resource.
II. EX PE RI ME NTAL ME TH OD
A series of tests were conducted to characterise the per-
formance of currents generated in the facility. These were
conducted with only the required measurement equipment in
the tank, to avoid potential distortion of flow around a device
model.
TABLE I
FIV E STAGE S OF DE VE LOP ME NT,FO R WAVE AN D TI DAL E NER GY D EVI CE S
Stage TRL Nominal scale Typical infrastructure
1. Concept Validation 1-3 Small scale (c. 1:100-1:25) University laboratory
2. Design Validation 3-5 Larger scale (c. 1:25-1:10) Industrial scale laboratory
3. Systems Validation 5-6 Sub-prototype size (c. 1:4) Benign test site
4. Device Validation 7-8 Approaching full size (c. 1:1) Exposed test site
5. Economics Validation 9 Full size, small arrays Commercial site
TABLE II
SIX DIMENSIONS OF VARIABILITY IN VELOCITY WITHIN THE FACILITY
Dimension of velocity variability Symbol
Reference velocity magnitude U0
Spatial variation across the test area X, Y
Vertical (shear) profile of velocity Z
Different current directions θ
Temporal variations in current t
For effective testing, controlled steady flows need to be
provided in the tank, with a vertical profile representative of
real sites. It is also important to understand any variation in
velocity throughout the test volume, so that the correct velocity
can be specified for any model position.
As part of previous commissioning work, a calibration of
velocity in the tank against primary control motor rpm was
undertaken. This showed a linear relationship, and was used
to set input velocities for these tests.
A. Test Plan
An initial measurement and characterisation program focus-
ing on the performance of generating currents in the tank was
developed, concentrating on the test volume at the tank centre.
Characterisation of the FloWave facility is a complex multi-
dimensional problem, as illustrated in Table II.
The first phase of testing, presented here, considers the
spatial variability of the generated current over a range of
baseline velocities. Measurements were made of the vertical
profiles of velocity, and of the spatial variation of velocity
across the plan area of the tank. These cover the horizontal
components of velocity for the vertical plane (XZ) and
horizontal plane (XY) for different input velocities (U0).
The tank is designed to be rotationally symmetrical, and there-
fore current direction is not discussed here. Measurement of
temporal variation and turbulence is ongoing, but initial results
showing the temporal stability of the facility are presented.
1) Tank Coordinates and Terminology: The tank co-
ordinate system is Cartesian, as shown in Fig. 2, with the
origin at the centre of the tank on the test floor, and Z
positive upwards. Waves and currents are specified as positive
in the direction of the vector, as opposed to the nautical
convention of waves coming from a direction. Currents flow
from upstream to downstream, with left and right assuming a
viewpoint looking downstream in the direction of the current.
B. Test method
All tests were run with a current direction of 0, i.e. flow
in the +Xdirection. At least 10 minutes was allowed for the
current to fully stabilise following changes in velocity before
taking measurements of the steady state condition. The results
from the temporal stability test, Section III-A, demonstrates
that this is sufficient.
For testing with only current, the wavemakers are powered
down and rest on their backstops. This results in the water
level in the tank dropping by approximately 80 mm. This
X
YZ
90° 270°
180°
Gantry
Control
Desk 2
Control Desk 1
Viewing glass
25mØ
overall
15mØ raisable floor
Fig. 2. Tank reference coordinates
configuration was used for all tests, with a water depth in
the test section of 1.93 m. Water temperature during the tests
was approximately 15C.
1) Measurement of vertical profiles: Vertical profiles of
velocity were measured to determine both the variation in
velocity with depth and input velocity (U0Z) and also the
spatial variation in velocity profiles across the tank (XZ).
Tests were undertaken at a range of nominal target veloc-
ities, specified for the centre of the tank 1.5 m above the
floor. These were the tank’s design velocity specification of
0.8 m/s, a typical low end test velocity of 0.2 m/s, and three
additional intermediate velocities of 0.42 m/s, 0.5 m/s, and
0.58 m/s. The preliminary calibration of velocity in the tank
against primary control motor rpm was used to set these
velocities. The variation in the vertical profile with input
velocity was measured at the tank centre throughout the whole
water column.
The development of the vertical profile, from the turning
vanes across the usable test area, was characterised by a
vertical slice along the flow direction. A series of seven
velocity profiles were measured at 2.5 m horizontal spacing
along the direction of current, covering the full diameter of the
raisable floor. Velocity measurements were taken throughout
the whole depth of the water column at each location.
For these tests, a Valeport model 801 single-axis electro-
magnetic (EM) current meter with a flat-type sensor head
for which the sensing volume is a cylinder of approximately
20 mm Ø ×10 mm high. [9]. This was mounted to a height-
adjustable bracket fixed to the gantry across the tank, with the
sensor cable helically wrapped around the supporting pole to
reduce the effects of vortex induced vibration. The raw ASCII
output from the Valeport control display unit was logged
directly to a laptop for further processing.
To measure each vertical profile, the sensor was lowered to
the base of the tank (Z = 0.05 m), and data logged at 2 Hz for
60 seconds. It was raised by 0.05 m to the next measurement
position, and the process repeated. In total 38 measurements
were taken for each profile (Z = 0.05 m to Z = 1.90 m). Repeat
measurements confirmed the temporal stability over the time
taken to measure each profile. Vertical position was measured
with a 0.5 mm graduated scale fixed to the height adjustable
bracket. The gantry position was measured with a laser range
finder. A small degree of lateral vibration was observed during
some tests, with the sensor head moving by approximately
±10 mm at around 1-2 Hz, however it is not anticipated that
this will affect the averaged inline velocity.
2) Measurement of spatial variability in plan: To determine
the planar extents of the usable test area, plus any variation
in velocity therein, the velocity was measured on a number of
horizontal transects across the tank, both along and transverse
to the flow direction. This was conducted at three nominal
velocities, 0.2 m/s, 0.5 m/s, and 0.8 m/s.
These tests used two separate 2-axis EM current meters,
Valeport model 802 fitted with a 32 mm discus-type sensor
head, for which the sensing volume is a cylinder of approxi-
mately 32 mm Ø ×16 mm high [10]. The two sensors were
fixed 2 m apart onto a carriage mounted frame that could be
moved along the gantry to set the Y-position in the tank. The
X-position in the tank was adjusted using translation of the
gantry as before. The sensor height was fixed at 1.5m above
the floor for all tests (i.e. 0.43 m below the water surface),
with the supporting frame mostly above the water level.
At each data point the raw ASCII output was recorded at
8 Hz over a 60 s period. The mean uand vhorizontal velocity
components, plus the horizontal velocity vector ~
U, were then
calculated. The measured data points were interpolated to a
regular 0.1 m grid in MATLAB using a triangulation-based
natural neighbour approach.
3) Measurement of long duration temporal variation: As
a first measure of the temporal stability of the tank, the
current ramp-up from rest and the subsequent stable flow was
measured for a period of 20 minutes. The measurement was
taken at the tank centre at 1.5 m above the floor, with a nominal
velocity of 0.48 m/s at 50 rpm.
This longer duration temporal variation test was undertaken
in a similar manner to the spatial plan tests, with a single
Valeport 802 discus-type sensor. At the start of the test the
primary drive motor was increased to 50 rpm in 5 steps over
approximately 1 minute, and then held at this speed throughout
the test.
III. RES ULTS
A. Temporal variability
Conditions in the tank need to be consistent over the
duration of the test and repeatable between tests, in order to
undertake useful model tests. The normalised velocity profiles
in Section III-B show that conditions in the tank are self-
similar and scaleable between different velocities.
As an example of the temporal stability of the tank, Fig. 3
shows a test where the motors were ramped up to 50 rpm and
held at that speed for 20 minutes. During this test, velocity
in the centre of the tank increases asymptotically, to within
10% of target after approximately 2 minutes, and reaches a
stable velocity after 5 or 6 minutes. The flow remains stable
thereafter, with only minor fluctuations.
B. Variation of Vertical Profile with Velocity
Vertical profiles of velocity were measured at the centre of
the tank for five nominal input velocities, shown in Fig. 4. The
drive motors have a linear relationship between rpm and flow,
which is confirmed by these tests, the depth-averaged velocity
increases in a linear manner with nominal velocity, see Fig. 5.
The depth-averaged standard deviation increases as a power
law, with an exponent just below unity.
The shape of the vertical profile of velocity in the tank
is almost independent of average velocity, as shown by the
similarity between the normalised velocity plots in Fig. 6a.
This is most closely described by a 1/15th power law, Fig. 6b,
although it is not dissimilar to other profiles used within the
industry.
Time [minutes]
0 2 4 6 8 10 12 14 16 18 20
Velocity [m/s]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Measured data points (8Hz)
10s moving average
60s moving average
0.48m/s nominal velocity
Fig. 3. Temporal stability test, showing velocity at the centre of the tank 1.5 m above floor, over the 20 minute test duration
Z [m]
0
0.5
1
1.5
2
Velocity [m/s]
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
25 rpm 46 rpm 54 rpm 61 rpm 82 rpm
Fig. 4. Vertical velocity profiles measured at tank centre for different input
drive motor rpm, with error bars showing ±1σdeviation, and water surface
at 1.93 m shown dashed grey
Primary Control Motor RPM
0 10 20 30 40 50 60 70 80 90
Depth Averaged Velocity [m/s]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
V=0.0093×RPM-0.0179
V=0.0011×RPM0.8696
Depth Averaged Standard Deviation
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
Fig. 5. Depth averaged velocity and standard deviation against primary control
motor rpm, showing linear relationship
C. Development of vertical profile along flow direction
The vertical profiles across the 15 m diameter floor were
used to construct a vertical slice through the tank, parallel to
the direction of flow and passing through the centre, Fig. 7.
This shows an increase in flow speed towards the centre of
the tank, a result of the converging nature of the flow in this
region that is required to create uniform flow in a circular tank,
as discussed above.
There is a significant velocity deficit in the lower part of
the water column at the extreme ‘upstream’ edge of the floor
(X = -7.5 m). Above this is a jet of higher velocity flow,
approximately 0.6 m to 1.0 m above the floor. Both of these
features are clearly apparent in the vertical profile Fig. 8a, and
are a result of the current rising at an angle from the turning
vanes.
Close to the centre of the tank, in the middle of test area, is a
relatively uniform section of flow. This covers approximately
X = -2.5 m to X = +5.0 m in the lower part of the water
column, and X = ±10 m in the upper half of the water column.
Normalised Velocity [U/U]
0.7 0.8 0.9 1 1.1
Normalised Depth [z/h]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
25rpm
46rpm
54rpm
61rpm
82rpm
0.7 0.8 0.9 1 1.1
HSE Guidance
1/7th Power
1/10th Power
1/15th Power
Fig. 6. Normalised vertical velocity profiles measured at tank centre, showing
a) different input drive motor rpm, and b) these profiles overlain with various
theoretical models
At the ‘downsteam’ edge of the floor (X = 7.5 m) the velocity
throughout the water column reduces, as a result of the flow
diverging into the tuning vanes around that half of the tank.
D. Spatial variability of velocity across the plan area
The spatial variation in flow across the central section of the
tank is shown in Figs. 9 and 10 for three nominal velocities.
This shows the velocity magnitude to be broadly symmetrical
about the current flow direction. It is also relatively consistent
(around ±10% or ±0.05 m/s) across a test area approximately
8 to 10 m wide and 6 m long, which is offset about 1 m
downstream of the tank centre.
The measurements show a slight asymmetry in the velocity
magnitude, with marginally faster flows (<5%) on the right
hand side of the flow. There is also marginally slower flow
(5%) along the centreline of the current near the middle
of the tank. The velocity vector plots in Fig. 11 show that
an acceptable horizontally uniform flow can successfully be
created in the circular tank, with only a slight directional bias
around the outside of the raisable floor.
IV. DISCUSSION
Physical model testing is a well-established practice in
the development of new technologies. Experiments in a tank
facility offer more control over test parameters than conducting
open water testing, however this requires the facility to be well
calibrated. Dedicated test facilities are also able to produce the
desired conditions on demand, rather than being dependent on
the vagaries of the weather.
A. Velocity Calibration and Stability
Measured velocity at a reference point in the tank needs
to be calibrated against the control input, the primary drive
motor rpm. Together with a transfer function based on the
velocity variations measured throughout the tank, this allows
a prescribed velocity to be produced at any particular location
in 3D space above the test area floor.
The depth averaged velocity in the tank has been shown to
vary linearly with drive motor rpm. This allows for accurate
control of the reference velocity in the tank. Initial test results
demonstrate that the tank can also produce a stable current
over time. There are minor fluctuations around this stable
current that may be due to large scale turbulent structures,
and further work is required to investigate this.
B. Velocity Shear Profile and Comparison with Theory
The 1/7th power law is frequently used to describe the
vertical distribution of velocities fluid flow. This was originally
developed to model boundary layer effects in turbulent pipe
flow, however it is often applied to tidal flow in coastal regions.
Other power law profiles, such as 1/10th, are also commonly
used. Guidance by the UK Health and Safety Executive [11]
suggests the following discontinuous function for the velocity
profile for coastal seas around the UK with a uniform velocity
in the upper half of the water column.
U=(z
0.32h
1
7¯
U, 0z0.5h
1.07 ¯
U, 0.5hzh(1)
where ¯
Uis the mean volumetric flow velocity, hthe water
depth, and zthe position in the water column.
In reality, flow in highly energetic tidal flows is far more
complex than a simple vertical shear profile, with significant
effects from large scale turbulent eddies and bathymetry
effects. Measurements undertaken at the European Marine
Energy Centre (EMEC) tidal test site at the Fall of Warness
as part of the ReDAPT project show complex shear profiles,
with the velocity at the surface significantly less than at points
lower in the water column for some states of the tide [12], [13].
Fig. 7. Vertical section across the tank in line with flow, for nominal 0.8 m/s input velocity. Flow direction left to right.
Z [m]
0
0.5
1
1.5
2
0 0.4 0.8
X= -7.5m
0.536m/s
0 0.4 0.8
X= -5.0m
0.640m/s
0 0.4 0.8
X= -2.5m
0.706m/s
Velocity [m/s]
0 0.4 0.8
X= 0.0m
0.740m/s
0 0.4 0.8
X= 2.5m
0.720m/s
0 0.4 0.8
X= 5.0m
0.659m/s
0 0.4 0.8
X= 7.5m
0.537m/s
Fig. 8. Vertical velocity profiles across the tank in line with flow, at X coordinates noted. Nominal 0.8 m/s input velocity, with depth averaged velocity given
for each profile location. Error bars show ±1σ.
Fig. 9. Variation in velocity across test area for horizontal plane 1.5 m above floor. Three different nominal input velocities: a) 0.2 m/s b) 0.5 m/s c) 0.8 m/s,
with flow direction from left to right. Measurement points indicated by + marker, 15 m diameter raisable floor shown as a grey circle
Fig. 10. Variation in velocity, relative to nominal input, across test area for horizontal plane 1.5 m above floor. Three different nominal input velocities:
a) 0.2 m/s b) 0.5 m/s c) 0.8 m/s, with flow direction from left to right. Nominal test area shown by black dashed rectangle, 15 m diameter raisable floor
shown as a grey circle.
-8 -4 0 4 8
Tank Y coordinate [m]
-8
-4
0
4
8
Tank X coordinate [m]
-8 -4 0 4 8
-8
-4
0
4
8
-8 -4 0 4 8
-8
-4
0
4
8
Fig. 11. Velocity vectors across test area for three different nominal input velocities: a) 0.2 m/s b) 0.5 m/s c) 0.8 m/s. Vector length is proportional to velocity
at measurement point relative to input velocity. Flow direction from left to right. Nominal test area shown by grey dashed rectangle, 15 m diameter raisable
floor shown as a grey circle.
The flow profile at FloWave was found to be close to a
1/15th power law, which is reasonably similar to observed
and theoretical profiles. By increasing the roughness on the
test floor, and/or using different spatial locations in the tank,
it should be possible to model a range of different flow profiles.
This offers the ability to model differential loading at varying
water depths, something that is of concern for developers of
tidal turbines.
The development of the shear profile across the tank, shown
in Fig. 7, follows the trend of previous work on inlet design
for combined wave and current test facilities [3]. This work
was undertaken in a flume channel, with a range of inlet vane
angles, so is not directly comparable to FloWave. However
the jet of water above mid depth with a region of slower
flow below was clearly present in those tests as well as the
accompanying CFD model.
C. Spatial Variability and Usable Test Area
The usable test area in the tank is at least 50 m2, which is
large enough for testing small arrays of devices. Whilst there
is some variation in velocity over this area, it is only around
10% in plan and in depth. Knowing this baseline variation
allows a reference velocity to be calculated at any point in
the tank, or device in an array. Theoretical forces can then
be predicted with a computer model and compared to those
measured, for example. In addition, velocity measurements are
typically made at a point close to the model during testing.
D. Flow Field in a Circular Tank
The tests undertaken show that across the test area, the flow
is acceptably straight and horizontally uniform. This allows
consistent testing over a large area of the tank, for example, to
investigate impacts between small arrays of devices. However
the configuration of a circular tank, with circumferential wave-
makers and impeller drive units below the floor, by necessity
leads to non-uniformities in the current flow field around
the turning vanes. The extent of non-uniform flow is limited
through careful design of the turning vanes and drive motor
control.
Full characterisation of the variation in flow field allows
for a greater variety of flow conditions to be modelled in the
tank, making use of the velocity gradients that exist in specific
locations away from the central test area. A circular tank also
has the significant advantage of being able to create complex
multi-directional and multi-modal sea states, as discussed in
Section I.
V. CONCLUSIONS
The FloWave facility offers the ability to test a wide variety
of realistic sea conditions prior to deployment of prototype
devices at sea, provided that the generation of these conditions
in the tank is well understood. Results of the initial character-
isation programme show that steady currents can be generated
in the tank, with a large uniform test volume in the centre of
the tank.
The vertical flow profile in the tank is not particularly
dependent on the input velocity, and can be approximated
by a 1/15th power law. This is a more uniform flow than
the 1/7th and 1/10th profiles commonly used to represent
tidal flows.
The flow over the test area is uniform within 10% or
0.05 m/s over an area greater than 50 m2around the
centre of the tank. The shape and size of this test area
is close to that predicted by the CFD modelling and
experimental tests carried out during the design phase.
The velocity is broadly symmetrical on both sides of the
tank about the flow direction, although it varies along the
direction of flow from ‘upstream’ to ‘downstream’. The
variation in velocity across the tank is consistent between
the different input velocities tested.
Steady flows can be produced in the facility, following
a ramp up period of about 5 minutes. However further
work is required to better understand the turbulent nature
of the flow.
ACKNOWLEDGMENT
The authors would like to thank the Energy Technology In-
stitute and RCUK Energy programme for funding this research
as part of the IDCORE programme (EP/J500847/1), and the
UK Engineering and Physical Science Research Council for
funding the FloWave facility (EP/I02932X/1).
REFERENCES
[1] EquiMar, “EquiMar Protocol II.A Tank Testing,” in Protocols for the
Equitable Assessment of Marine Energy Convertors, D. M. Ingram,
G. H. Smith, C. Bittencourt Ferreira, and H. C. M. Smith, Eds.,
Edinburgh, 2011, ch. 3.3.
[2] D. M. Ingram, A. R. Wallace, A. Robinson, and I. G. Bryden, “The
design and commissioning of the first, circular, combined current and
wave test basin.” in Proceedings of Oceans 2014 MTS/IEEE, Taipei,
Taiwan, 2014.
[3] A. Robinson, D. Ingram, I. Bryden, and T. Bruce, “The effect of inlet
design on the flow within a combined waves and current flumes, test
tank and basins,” Coastal Engineering, vol. 95, pp. 117–129, Jan. 2015.
[4] ——, “The generation of 3D flows in a combined current and wave
tank,” Ocean Engineering, vol. 93, Jan. 2015.
[5] B. Holmes and K. Nielsen, “Guidelines for the development & testing
of wave energy systems,” Hydraulics Maritime Research Centre, UCC,
Cork, Ireland, Tech. Rep. June, 2010.
[6] J. C. Mankins, “Technology readiness levels,” White Paper, April, pp.
4–8, 1995.
[7] V. Heller, “Scale effects in physical hydraulic engineering models,”
Journal of Hydraulic Research, vol. 49, no. 3, pp. 293–306, Jun. 2011.
[8] T. Vyzikas, D. Greaves, D. Simmonds, C. Maisondieu, H. C. M. Smith,
and L. Radford, “Application of numerical models and codes Task
3.4.4 of WP3 from the MERiFIC Project A,” University of Plymouth,
MERiFIC, Plymouth, UK, Tech. Rep., 2014.
[9] Valeport, “Model 801 Electromagnetic Flow Meter Datasheet,” Totnes,
UK, Tech. Rep., 2011. [Online]. Available: http://www.valeport.co.uk/
Portals/0/Docs/Datasheets/Valeport Model801 v2a.pdf
[10] ——, “Model 802 Electromagnetic Flow Meter Datasheet,” Totnes,
UK, 2011. [Online]. Available: http://www.valeport.co.uk/Portals/0/
Docs/Datasheets/Valeport Model802 v2a.pdf
[11] HSE, “Environmental considerations,” Health & Safety Executive, Lon-
don, Tech. Rep., 2002.
[12] K. Gunn and C. Stock-Williams, “On validating numerical hydrody-
namic models of complex tidal flow,International Journal of Marine
Energy, vol. 3-4, pp. e82–e97, Dec. 2013.
[13] ——, “Falls of Warness 3D Model Validation Report,” E.ON
Technologies, Tech. Rep., 2014. [Online]. Available: http://www.eti.co.
uk/redapt-full- report-falls-of-warness-3d-model-validation/
... As shown in Fig. 1, beneath the floor, in the plenum chamber along the entire tank circumference, are 28 impeller units that create a re-circulating flow system [26]. This arrangement of impellers allows the generation of a predominantly straight flow in any direction across the central test area of the tank [27]. ...
... Due to the nature of flow generated in a circular tank, there is some spatial variation of the mean flow and turbulence intensity across the tank area, which have been characterised in Refs. [27,29]. [27] shows a 50 m 2 region of relatively straight uniform flow (±10%) in the tank centre. ...
... [27,29]. [27] shows a 50 m 2 region of relatively straight uniform flow (±10%) in the tank centre. The turbines, used in this test campaign, were installed in this region of the tank. ...
Article
Full-text available
Coupled blade element momentum-computational fluid dynamic (BEM-CFD) approaches have been extensively used to study tidal stream turbine performance and wake development. These approaches have shown to be accurate when compared to tests conducted in tow-tanks or in regulated flumes with uniform flows across the turbine. Whilst such studies can be very useful, it is questionable as to what extent the results would differ in a larger scale environment where the flow is more representative of real-world conditions, being either unsteady or non-uniform. In this work, the effectiveness of a generalised actuator disk-computational fluid dynamics (GAD-CFD) approach in accurately capturing fluid-machine interaction for single and multiple tidal energy converters models is further assessed. A unique large-scale experimental facility, FloWave, has been used to conduct physical testing of three instrumented model tidal energy converters of rotor diameter 1.2 m under differing turbine layouts and realistic scaled environmental conditions. These large-scale tests provide a unique dataset against which this work's numerical simulations have been extensively validated. Comparisons between the tank and GAD-CFD approach show good agreement, particularly when comparing modelled to measured thrust, and enabled an evaluation of the effects of turbine spacing and arrangement on turbine performance and flow-field response.
... Currents are generated in FloWave using 28 impeller units mounted in a plenum chamber beneath the tank floor (see e.g. Noble et al. (2015)). The turbulence intensity is between 5 and 11% and the vertical shear profile in the centre of the tank follows approximately a 1/15 power law . ...
... Slightly higher values are measured for the flexible line, which was unexpected due to the increased submergence for the inextensible line, and hence expected increase in the drag force. It is speculated that this discrepancy is a result of velocity variations along the axis as observed in Noble et al. (2015) coupled with the large changes in mean position observed (ADV measurements were made at a single location as defined in Table 1). The statics model presented in Section 2.2.1 (solid green line) provides a reasonable agreement with the experimental results, with errors likely arising due to the unaccounted for spatial variation of current velocity. ...
Article
Full-text available
Wave measurement buoys provide characterisation of wave climates that forms the basis for the design of offshore systems. These buoys are commonly subjected to currents which affect the resulting wave measurements, and if not accounted for will result in errors in the estimated sea state parameters. The present work provides results and observations from experiments aimed at assessing the impact that currents have on wave buoy measurements, thereby informing processing techniques to more accurately include this effect. Through scaled testing (circa 1:15) in a combined wave–current test tank, buoy motions (diameter, D = 0.24 m) are recorded in current only, waves only, and combined wave–current including oblique conditions. From these, the wave-induced motions are extracted and compared against three prediction methods based on established transfer function approaches as well as a frequency-domain hydrodynamic coefficient (HC) model based on potential flow. The scaled buoy was observed to have large, complex, irregular oscillatory vortex-induced motions (VIM) exceeding the buoy diameter. Both the magnitude and frequency of these oscillations was found to be significantly altered by the mooring stiffness and configuration whilst the addition of collinear waves was found not to affect the magnitude of VIM. Furthermore, due to the lack of VIM heave response and a large difference between the frequencies of the vortex-induced and wave induced horizontal motions, it was found that the VIM did not significantly alter the interpretation of the wave climate for the tested conditions. The HC model was found to accurately capture the observed modified hydrodynamics for opposing wave–current conditions, where larger horizontal motions than (typically) predicted are observed for all frequencies. This behaviour is concluded to result from increased excitation forces owing to the higher wavenumbers. The experiments highlight the potential effects of VIM on wave measurement performance of wave buoys, along with the complex and mooring-dependent nature of the response. Altered dynamics in the presence of currents are described which must be accounted for to avoid errors and the presented prediction methods provide a mechanism to account for these effects in wave processing methodologies which can subsequently reduce uncertainty in our understanding of the offshore environment.
... All investigated flow speeds were based on a previous calibration (Noble et al., 2015) and they have proven to be very accurate, highly reproducible and consistent. The tank produces a realistic turbulence level in the main testing area. ...
... Schematic diagram of the FloWave circular tank in plan and oblique section showing (A) the wavemakers, (B) the flow turning vanes, (C) the impeller units, (D) the buoyant raisable floor and (E) idealised streamlines of water flow across tank floor(Noble et al., 2015). ...
Article
Remotely Operated (underwater) Vehicles (ROV) have a wide range of maritime applications, including repair and maintenance. Quantifying hydrodynamic loads is important for the design and control of these ROVs. A novel approach with eight tethers was used to restrain a commercially available ROV, namely the BlueROV2 (Blue Robotics, Torrance, USA), in the mid depth of the FloWave wave and current test tank. This experimental set-up allowed the measurement of the forces under realistic flow around the ROV without introducing significant interference. The paper presents the analysis of the load cell data as forces and moments in relation to the observed motion and rotation of the ROV. In addition to active propelled cases, a variation of current speed (up to 1 m/s) coming out of the four directions as well as different regular waves were tested. Three different distances of a cylindrical obstacle provided a quantification of the effect of flow shadowing from a structure in front of the ROV. The results can also be used as a validation experiment to expand the application of ROVs and the influence of obstacles based on numerical simulations.
... In a different circumstance, a similar jet can be observed such as a bursting bubble, 11 breaking waves, 12 and wave interaction. 13 Such hollows on a liquid surface commonly retract due to capillarity, gravity, or inertia raising the central jet, similar to the above explained cavity reversal. ...
Article
Full-text available
We investigate the impact of a vertically falling droplet onto a non-uniform liquid depth having a linear slope of the bottom substrate. Here, we report that the resulting jet direction is inclined to the shallow liquid depth after the droplet impact, which is found to be markedly distinct from a vertically falling droplet onto a uniform liquid bath. From experimental and numerical results, we observe that initially the cavity grows almost axisymmetrically, and then, when it retracts, asymmetric capillary waves exhibit. The asymmetric cavity reversal leads to the inclined jet ejection that is related to pressure distribution and velocity of the interface. For the systematic study, we explore the jet dynamics by varying the surface tension, the droplet size, the droplet impact speed, the inclination angle of the bottom substrate, and the depth of the liquid bath. Finally, we provide a simple scaling model to predict the inclination angle of the resulting jet after the drop impact on the inclined liquid pool.
... This capability enables the conditions defined in Section 3.2 to be produced. More information on the facility can be found in Ref. [9,26,32] where diagrammatic representations of the facility can be found alongside baseline measurements of flow variability and turbulence. For the flow velocities used in these experiments the turbulence intensity is around 7%. ...
Article
The presence of waves exposes tidal stream turbines to large and cyclic hydrodynamic loads which significantly influence the design requirements for tidal turbine blades. Here we describe a loading phenomenon not previously considered in literature caused as blades rotationally sample an oscillating and vertically decaying wave-induced velocity field. Although implicitly incorporated into numerical models, the dominant causes and relative influence have not previously been considered. In this article this effect is described through theoretical analysis and validated through scaled experiments; including irregular waves at angles to the rotor and current field. The associated loads are found to be strongly correlated to the wavenumber. The nature of the rotational-sampling-effect is confirmed through analysis of the experimental results, where characteristic sidebands are effectively predicted in the blade root bending moment spectra. It is estimated to account for between 8% and 16% of the fatigue damage and between 7% and 13% of the peak root bending moment for the conditions tested. A key finding is that two bilaterally-symmetrical oblique wave conditions do not produce equivalent loading patterns: one produces higher frequency oscillations. Additionally, it is found that the frequency of these loads reduces linearly with rotational speed; highlighting another consideration for tidal stream turbine operation.
... Beneath the floor, along the entire tank circumference, are 28 impeller units that create a re-circulating flow system [13]. This arrangement of impellers allows the generation of a predominantly straight flow in any direction across the central test area of the tank [14]. The turbulence intensity is around 7% for the flow velocities used in these experiments. ...
Conference Paper
Full-text available
Hydrokinetic tidal turbines are a promising alternative for the generation of clean electrical energy. They are still far behind, with respect to their technological development, in comparison to offshore wind turbines, which are currently in the stage of commercial energy production. Thus, more studies and analyses of the behaviour of tidal devices and their interaction with the surrounding ocean space are required. How this interaction is interrelated to the power production system is also necessary to be further examined. In this paper, the development of a whole system, fully-coupled model of a laboratory-scale hydrokinetic tidal turbine, along with its interactions with the ocean environment and its electrical control system is described. The model was developed in fastFlume (SOWFA, NREL) coupled with an external torque control system. The control system is devel- oped from the optimal torque speed curve based Maximum Power Point Tracking (MPPT) algorithm. The optimal torque speed curve of the turbine used in the model was obtained from experimental work in a test tank. The hydrokinetic tidal turbine and the control system models were implemented independently. They were coupled in order to reach an energy balance between the surrounding flow, the tidal turbine, and the control system. Three flow stream velocities were imposed in the inlet of the model do- main, starting the rotor from zero rotational speed. After the optimal rotational speed is attained, the electrical power generated and the loads experienced by the turbine rotor were studied. In the simulations, the tidal device is controlled to keep the optimal power production for any flow stream velocity. The results of the modelling work were compared with experimental measurements taken from 1:15th scaled testing of a fully-instrumented and controllable tidal device at the Flowave Ocean Energy Research Facility, The University of Edinburgh, a combined wave and current test facility. The results show time series of turbine and generator variables like mechanical and electrical torque and power, as well as thrust and the optimal rotational speed for each of the tested cases. The validation shows good agreement between the numerical and experimental results which encourages futures studies using the coupled model, including the turbine working in more complex flow conditions and controlled by more complex control schemes.
... Due to the circular tank the flow speed varies over the full area of the tank but based on the well-established control strategy a realistic, straight, velocity distribution can be provided in the main testing area around the centre of the tank by forcing a big eddy on each side of the tank to cover the remaining water body outside of the main testing area. The presented current speed is a mean velocity at the centre of the tank in mid water depth [29]. For the wave cases three different time windows can be distinguished. ...
Article
Full-text available
Hydrodynamic forces are an important input value for the design, navigation and station keeping of underwater Remotely Operated Vehicles (ROVs). The experiment investigated the forces imparted by currents (with representative real world turbulence) and waves on a commercially available ROV, namely the BlueROV2 (Blue Robotics, Torrance, USA). Three different distances of a simplified cylindrical obstacle (shading effects) were investigated in addition to the free stream cases. Eight tethers held the ROV in the middle of the 2 m water depth to minimise the influence of the support structure without completely restricting the degrees of freedom (DoF). Each tether was equipped with a load cell and small motions and rotations were documented with an underwater video motion capture system. The paper describes the experimental set-up, input values (current speed and wave definitions) and initial processing of the data. In addition to the raw data, a processed dataset is provided, which includes forces in all three main coordinate directions for each mounting point synchronised with the 6DoF results and the free surface elevations. The provided dataset can be used as a validation experiment as well as for testing and development of an algorithm for position control of comparable ROVs.
Article
The performance benefits of deploying tidal turbines in close side-by-side proximity to exploit constructive interference effects are demonstrated experimentally using two 1.2 m diameter turbines. The turbines are arrayed side-by-side at 1/4 diameter tip-to-tip spacing, and their performance compared with that of a single rotor. Tests were completed in the 25 m diameter, 2 m deep wave and current FloWave Ocean Energy Research facility. A detailed assessment of inflow conditions at different control points is used to understand the impact that rotors, designed for high blockage conditions, have on the approach flow. After accounting for global blockage, a 10.8 % uplift in the twin-turbine-averaged power coefficient, relative to that for a single turbine, is found for the turbine design speed, at the expense of a 5.2 % increase in thrust coefficient and 3.1 % increase in tip-speed-ratio. Flowfield mapping demonstrated flow effects at array and device scale including array bypass flows and jetting between turbines. Azimuthal variation of blade root flapwise and edgewise bending moments show that the turbines interact in a beneficial manner, with additional and sustained loading peaks as the blades pass in close proximity to the neighbouring rotor. Peak performance for the twin turbines occurred at a higher tip-speed-ratio than for the single turbine, which is consistent with the twin turbines exerting a higher thrust on the flow to achieve maximum power. The twin turbine performance variation with tip-speed-ratio is found to be more gradual than for the single turbine. Using differential rotor speed control we observe that array performance is robust to small differences in neighbouring rotor operating point. Through these experiments we demonstrate that there is a substantial, achievable performance benefit from closely arraying turbines for side-by-side operation and designing them for constructive interference.
Article
Tidal energy has the potential to significantly contribute to energy security by providing predictable renewable energy. New technology is needed to decrease the levelised cost of energy and to make this energy sector competitive in the energy market. A key area where technology can contribute to decrease costs is mitigating the hydrodynamic load fluctuations, and thus increasing the fatigue life of the turbine. Here, we formulate a passive morphing blade concept that aims to mitigate the unsteady thrust without affecting the mean torque and thus the harvested power. We show that a blade with a trailing edge that deflects perfectly elastically can suppress virtually all fluctuations without varying the mean loads. The effect of the hydrodynamic and blade's inertia, the material damping, and the radial shear stress, decrease the performances. Using a low-order model of the blade, we show that when a gust occurs, the angle of attack experienced by a rigid blade increases, whilst that experienced by a well-designed morphing blade decreases. This counter-intuitive mechanism is what makes morphing blades highly effective. While blades that could passively twist have previously been developed, this theoretical study suggests that chordwise flexibility is a suitable alternative that should be further explored.
Article
Tidal stream turbines (TST) are subject to large unsteady hydrodynamic loads which will result in fatigue of components and potentially failure. Control algorithms, if used effectively, have the ability to reduce these unsteady loads without affecting overall performance. To examine the effect of control strategy on performance and loads, experimental tests were carried out using a 1:15 scale TST in a combined wave–current facility. Loose and stiff speed controllers were developed and implemented, in addition to torque control, and tested for a range of tip-speed-ratios in a fixed inflow velocity. The speed controllers are additionally tested under regular, irregular and focused wave conditions. Through time and frequency-domain analysis, it is demonstrated that looser controllers, through larger speed variations, induce larger variations in the streamwise forces (rotor thrust and root bending moment), but smaller variations in torque. Similarly, larger extremes are recorded for streamwise forces under looser controllers. Mean values and hence overall turbine performance is found to be unchanged. It is noted that for the wave cases tested the controllers do not significantly affect the streamwise forces; likely a result of the large wave-induced velocities dominating the resulting blade angles-of-attack. Additionally, for some rotational speed values significant amplification of mechanical vibrations are observed. These results highlight the complexities associated with controller choice; considering the trade-offs in torque and thrust variations, potential mechanical resonance, and varying performance depending on the flow conditions. Proper appreciation of these considerations will be vital as we move towards commercial arrays of devices which must be controlled optimally to maximise performance whilst minimising loading and associated O&M costs.
Technical Report
Full-text available
This report indicates the importance of numerical modelling in the modelling process, gradually builds the essential background theory in the fields of fluid mechanics, wave mechanics and numerical modelling, discusses a list of commonly used software and finally recommends which models are more suitable for different engineering applications in a marine renewable energy project.
Conference Paper
Full-text available
In 2010 Bryden, Ingram and Wallace applied to the UK Engineering and Physical Sciences Research Council (EPSRC) for funding to construct the \emph{all waters combined current and wave test facility}. Funding was awarded and over the next three years the FloWave TT facility was constructed. At its heart is a 25m diameter, circular basin, equipped with 168 force feedback wave makers, 28 bidirectional impellers and a liftable floor. The 2m deep test section is designed to generate currents (at 0.8m/s) and 700mm high, 2s period, waves from any relative directions. Allowing a model to be subjected to scale tests over the full tidal ellipse simultaneously with multi-directional waves. Construction of the facility was completed in November 2013, and calibration is currently in progress. Initial work has shown that maximum flows of 2m/s can be achieved across the test section, while the circular wave maker array allows very large focused waves to be created. This paper describes the FloWave facility, its construction and commissioning and presents some preliminary results.
Article
The motion of the sea, through waves and currents, represents a large source of clean and safe energy. However, any structure built to operate in the sea will experience large varying forces and a difficult environment. It is therefore crucial to develop realistic and repeatable sea-like conditions in a laboratory in order to lower the cost and risk of developing off-shore structures. Building on previous efforts, an experimentally validated numerical model is used to predict the current-only flow in flumes capable of combining waves and current. This model is then used to simulate the flows within common flume configurations and within a new concept known as the “isolating inlet flume”. The results of these simulations are then analysed to assess the performance of each flume type and to understand the fluid dynamics that govern each type. Flume performance is found to be largely determined by the creation and dissipation of shear layers. The tests proved that a flume using the isolating inlet requires significantly less downstream length to achieve a developed flow and acceptable turbulence levels than the previous flume configurations. The isolating inlet has the additional benefit of creating a still zone where a conventional wave-maker might be used. Further simulations are used to investigate the design of the isolating inlet flume and demonstrate how it works. This paper should be of use to scientists and engineers seeking to design flumes, test tanks and basins that create sea-like test conditions, thus improving the scope and range of laboratory testing.
Article
Scale effects arise due to force ratios which are not identical between a model and its real-world prototype and result in deviations between the up-scaled model and prototype observations. This review article considers mechanical, Froude and Reynolds model–prototype similarities, describes scale effects for typical hydraulic flow phenomena and discusses how scale effects are avoided, compensated or corrected. Four approaches are addressed to obtain model–prototype similarity, to quantify scale effects and to define limiting criteria under which they can be neglected. These are inspectional analysis, dimensional analysis, calibration and scale series, which are applied to landslide generated impulse waves. Tables include both limiting criteria to avoid significant scale effects and typical scales of physical hydraulic engineering models for a wide variety of hydraulic flow phenomena. The article further shows why it is challenging to model sediment transport and distensible structures in a physical hydraulic model without significant scale effects. Possible future research directions are finally suggested.
The generation of 3D flows in a combined current and wave tank
--, "The generation of 3D flows in a combined current and wave tank," Ocean Engineering, vol. 93, Jan. 2015.
Guidelines for the development & testing of wave energy systems
  • B Holmes
  • K Nielsen
B. Holmes and K. Nielsen, "Guidelines for the development & testing of wave energy systems," Hydraulics Maritime Research Centre, UCC, Cork, Ireland, Tech. Rep. June, 2010.
Application of numerical models and codes Task 3.4.4 of WP3 from the MERiFIC Project A
  • T Vyzikas
  • D Greaves
  • D Simmonds
  • C Maisondieu
  • H C M Smith
  • L Radford
T. Vyzikas, D. Greaves, D. Simmonds, C. Maisondieu, H. C. M. Smith, and L. Radford, "Application of numerical models and codes Task 3.4.4 of WP3 from the MERiFIC Project A," University of Plymouth, MERiFIC, Plymouth, UK, Tech. Rep., 2014.
Model 801 Electromagnetic Flow Meter Datasheet
  • Valeport
Valeport, "Model 801 Electromagnetic Flow Meter Datasheet," Totnes, UK, Tech. Rep., 2011. [Online]. Available: http://www.valeport.co.uk/ Portals/0/Docs/Datasheets/Valeport Model801 v2a.pdf