Assessing the inuence of inow turbulence on noise and

performance of a tidal turbine using large eddy simulations

Thomas P. Lloyd
a, b, *
, Stephen R. Turnock
b
, Victor F. Humphrey
b
a
MARIN Academy, Maritime Research Institute Netherlands, Wageningen, The Netherlands
b
Faculty of Engineering and the Environment, University of Southampton, Southampton, United Kingdom
a r t i c l e i n f o
Article history:
Received 20 February 2014
Accepted 7 June 2014
Available online 17 July 2014
Keywords:
Horizontal axis tidal turbine
Large eddy simulation
Inow turbulence generator
Acoustics
Environmental impact
a b s t r a c t
Large eddy simulations of a model scale tidal turbine encountering inow turbulence have been per-
formed. This has allowed both unsteady blade loading and hydrodynamic noise radiation to be predicted.
The study is motivated by the need to assess environmental impact of tidal devices, in terms of their
acoustic impact on marine species.
Inow turbulence was accounted for using a synthetic turbulence generator, with statistics chosen to
represent the gross features of a typical tidal ow. The turbine is resolved in a fully unsteady manner
using a sliding interface technique within the OpenFOAM

libraries. Acoustic radiation is estimated using

a compact source approximation of the Ffowcs WilliamseHawkings equation.
It is observed that the long streamwise length scale of the inow turbulence results in characteristic
humps in the turbine thrust and torque spectra. This effect is also evident in the far-eld noise spectra.
The acoustic sources on the blades are visualised in terms of sound pressure level and Powell's source
term. These measures show that the dominant sources are concentrated at the blade leading edges
towards the tip. This results from the high loading of the turbine blades, and causes the sound to radiate
more akin to a monopole than a dipole.
The full scale source level, obtained from scaling of the simulation results, is found to be lower than
comparable measured data reported in the literature; this is attributed to additional sources not included
in the present study. Based on the predicted source level, no physical impact on sh is expected.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
The level of acoustic emission from tidal turbines is part of
environmental impact assessment [1]. It is possible that tidal tur-
bine noise will have some impact on marine life, but attempts to
quantify this are limited [2]. We direct the reader to work involving
full-scale turbine noise measurement [3,4], as well as noise esti-
mation for smaller devices [5,6]. Turbine noise sources are
commonly dened in terms of a sound pressure level (SPL)
measured at some far-eld distance. These are often corrected
back to a source level at 1 m from the rotor. A typical level
dened in this way is of the order of 166 dB re 1 mPa
2
at 1 m [3].
Richards et al. [4] expect the dominant noise from horizontal
axis tidal turbines (HATTs) to be due to rotating machinery in a
frequency range z1e100 Hz.
Wang et al. [5] measured the noise of a 0.4 m diameter device,
using a scaling procedure recommended by ITTC [7]. The reported
maximum third-octave bandwidth SPLs (for a freestream velocity
of 2.57 ms
1
) were approximately 115 dB and 125 dB for model and
scaled results respectively.
Numerical studies of tidal turbine noise are less commonly re-
ported. The noise of a vertical axis tidal turbine was estimated using
a discrete vortex method by Li and alis al [6]. These authors found
the peak SPL occurs at 4 Hz, and related their ndings to the
hearing sensitivity of sh, without making direct environmental
impact assessments. No studies of HATT noise have been located;
by contrast, noise simulations of horizontal axis wind turbines are
more commonplace [8e10].
It is important to study the dynamic forces experienced by a
turbine, since they contribute to uid structure interaction effects,
such as blade fatigue [11] or potential improvements in power
capture [12]. Thus studying this behaviour in a dynamic environ-
ment would seem appropriate. The effect of inow turbulence on * Corresponding author.
E-mail address: T.P.Lloyd@soton.ac.uk (T.P. Lloyd).
Contents lists available at ScienceDirect
Renewable Energy
j ournal homepage: www. el sevi er. com/ l ocat e/ renene
http://dx.doi.org/10.1016/j.renene.2014.06.011
0960-1481/ 2014 Elsevier Ltd. All rights reserved.
Renewable Energy 71 (2014) 742e754
turbine wakes will also inuence the layout of arrays in order to
optimise total power capture [13,14].
The effect of inowturbulence on the length of the turbine wake
has been studied using computational uid dynamics [15,14].
Various approaches to simulating unsteady loading have been used.
Churcheld et al. [16] used a library of instantaneous turbulence
realisations as an inlet condition to replicate a turbulent boundary
layer. Alternatively, synthetic turbulence can be generated using a
numerical method to represent the inow turbulence. In Ref. [15],
this approach was used to showthat the inowturbulence reduces
the length of the turbine wake. This effect was also reported by
McNaughton et al. [14], who included the turbine geometry in the
simulation, as opposed to the porous disk representation in Gant
and Stallard [15]. Afgan et al. [17] showed that large eddy simula-
tion (LES) is more capable than an unsteady Reynolds-averaged
NaviereStokes equations solution at predicting the complex ow
features and mean performance of a model scale turbine. LES also
provides better resolution of the inow turbulence and thrust
spectra.
We have previously presented predictions of tidal turbine noise
[18] as well as impact assessment [19], using empirical modelling.
The approach was based on modelling the turbine unsteady thrust
spectrum due to inow turbulence, and predicting the noise
assuming free eld radiation [20]. Inowturbulence is known to be
the dominant noise source due to turbulence in this case [18,21].
The estimated spectral source level (SSL) was z145 dB re
1mPa
2
Hz
1
at 1 m across a frequency range of z10e100 Hz. This
paper presents a development in terms of noise simulation of
HATTs, and is based on the methodology in Lloyd et al. [22].
The paper assumes the following format. x2 outlines the nu-
merical setup for the simulations, including the methods for
generating inowturbulence and predicting sound radiation. In x3,
the test case geometry is described, along with the domain design.
A model scale turbine is used since a detailed blade geometry is
available, as well as mean performance data. This section also in-
cludes grid design considerations and presents an assessment of
the sliding interface technique used in terms of its ability to
interpolate the broadband velocity uctuations present in the
inow. The simulation results are divided into three sections:
inow turbulence statistics (x5); turbine response (x6); and
acoustic emission (x7). The model scale acoustic predictions are
compared to an analytical model, since no experimental validation
data is available. These are then scaled using recommended pro-
cedures in order to provide estimates of full scale turbine noise (x8).
This allows discussion of possible environmental impact. Finally,
conclusions are made in x9.
2. Numerical framework
2.1. Turbulence modelling
In large eddy simulation, the ltered NaviereStokes equations
are resolved in a time-dependent manner. Scales smaller than the
grid are accounted for using a subgrid model. For a detailed
description of large eddy simulation see for example Sagaut [23].
The dynamic mixed Smagorinsky subgrid model [24] is used, since
it has been shown to performwell for complex ows and on coarse
grids [25].
The normalised rst cell height is dened as Dy

w
y
1
u
t
=n,
where y
1
is the rst cell height, u
t
the friction velocity and n the
kinematic viscosity. An average value of Dy

w
40 was achieved
over the blades; due to the surface renement technique employed
by snappyHexMesh, the leading and trailing edges typically
possessed much lower values. Since the viscous sublayer is not fully
resolved (this requires Dy

w
z1), a wall function, based on Spalding's
law of the wall [26] was used.
In order to simulate stochastic loading on the turbine, an inow
turbulence generator is used. This is a numerical method for
generating synthetic turbulence at the simulation inlet. Here we
use the forward stepwise method; for a full description of the
method, see Kim et al. [27]. Evaluations of this method for hydro-
acoustic predictions has previously been made.
2.2. Solution method
Simulations were performed using the OpenFOAM
1
libraries. A
custom solver based on the pimpleDyMFoam application was used.
The main features of the solver are: pressure implicit splitting of
operators (PISO)-type [29] correction of the velocity; outer
corrector loops allowing higher time steps than PISO; grid rotation
via dynamic meshing and an arbitrary mesh interface (AMI); and
velocity uctuations generated by the FSMinserted during the PISO
loop. All discretisation schemes are second-order, apart from
convective acceleration, which uses a hybrid upwind-central dif-
ferencing scheme, giving good accuracy in regions where a central
scheme is less accurate [25,28]. Linear solution was achieved using
the biconjugate gradient method for velocity, and general algebraic
multigrid method for pressure. The solvers exit the iteration loop
when a tolerance of 10
9
(velocity) and 10
6
(pressure) is achieved
within each loop.
The pimpleDyMFoam solver allows the maximum Courant
number Co jujDt=Dx to exceed unity, where juj is a local velocity
magnitude, Dt the time step and Dx the cell dimension; simulations
used a maximum time step Dt

DtU
0
=D 3:5 10
5
, based on
the reference (freestream) velocity U
0
, and turbine diameter D.
where the maximum Courant number was also limited to four. This
time step is the same as that used for other tidal turbine LES [17],
and results in 40 time steps per degree of rotation. A transient
phase of four turbine rotations was assumed (T

TU
0
=Dz2:3),
allowing the inowturbulence to reach the rotor plane. Probe, force
and sound pressure were then sampled at f
sample
n/300, or 100
times per blade passage, for a further T

z6:9, thus ensuring a

complete ow-through of the domain.
2.3. Acoustic analogy
In order to evaluate the acoustic radiation from the blades, we
use a formulation of the Ffowcs William-Hawkings (FW-H) acoustic
analogy [30]. This has been implemented into OpenFOAM

using
only the term relating to uid loading, which is an acoustic dipole.
This is given by
p'x; tz
x
i
4pc
0

2
v
vt
%
S
n
j
p
ij
dhdSy: (1)
In Equation (1), p
0
is acoustic pressure, x and y denote the
receiver and source locations, jrj jx yj, c
0
is the speed of sound,
equal to 1500 ms
1
in water, n
j
is the normal vector to the surface S,
and d is the Dirac delta function. This form of the FW-H equation is
suitable for low Mach number ows where broadband noise is of
interest. It assumes the receiver to be in the acoustic far-eld
(jrj[l) and the source to be compact (Ll, where L is the source
dimension). The integration surface (corresponding to h 0) is
taken to be the solid boundary.
1
www.openfoam.org/.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 743
A receiver distance of jrj 2D was used, since it lies in the
acoustic far-eld for a typical rotor [31], but still within the range
that environmental impact is possible [18]. Sound pressure level
predictions were made at three receiver angles of q 0

, 45

and
90

, where 0

corresponds to the rotor axis, downstream of the

rotor plane. The SPL is dened as
SPLf 10log
10
_
_
p
02
f
p
2
0
_
_
; (2)
where p
0
1 mPa in water. The pressure uctuation p
02
f was
estimated using Welch's algorithm [32], applied to the time traces
obtained from Equation (1).
3. Test case description
3.1. Turbine geometry
The turbine geometry used is a representation of the three
bladed model scale rotor tested by Bahaj et al. [33]. The blades use
NACA 638xx sections with a reduction in chord and thickness
fromroot to tip. Parameters for the chosen case are given in Table 1.
The chosen case represents a high thrust loading. Tip speed ratio is
dened as L UR=U
0
where U is the rotational velocity in radians
per second, R is the turbine tip radius and U
0
is the freestream
velocity.
3.2. Specifying inow turbulence statistics
Characterisation of the inow turbulence was achieved by
aiming to replicate features of full scale environmental turbulence.
The chosen statistics are based on IEC standard 614001, and are
the same as those used by Gant and Stallard [15]. The horizontal
integral length scale is then L
x;z
0:7D, with L
y
L
x;z
=6. A
turbulence intensity of I 10% is used, which is specied as ho-
mogeneous and isotropic. Note the FSM has the ability to generate
fully inhomogeneous, anisotropic mean velocity, length scale and
turbulence intensity proles more akin to realistic tidal ows.
However, as a demonstration of our methodology, only anisotropic
length scales are used here.
4. Grid design
4.1. Domain setup
Since the data presented by Bahaj et al. [33] are corrected for
blockage effects, the turbine is simulated in an open domain. The
blockage ratio (ratio of turbine rotor disc area to domain cross-
section area) is 0.022. Based on corrections made for a similar
blockage ratio, reported by Walker et al. [34], the effect on turbine
performance is expected to be small. The grid consists of two re-
gions, the rotor and stator, with the rotor region fully encompassing
the turbine geometry. Geometrical simplications of the
turbine geometry result in a truncated hub of length and diameter
d
H
/D 0.125. The rotor grid has dimensions of L
I
/D 0.625 and
d
I
/D 1.2, and rotates inside the stator grid at a constant rotational
velocity. The two grid regions are connected using an arbitrary
mesh interface (AMI), which interpolates variables between the
patch faces of each grid region using Galerkin projection. Fig. 1
shows a schematic of the domain layout, with associated bound-
ary conditions given in Table 2.
The background cell size is chosen to resolve z80% of the total
turbulence kinetic energy, with a grid cutoff size of DzL =12 [35].
This ensures that the cutoff lies inside the inertial subrange of the
turbulence spectrum, with an estimated cutoff frequency of
z47 Hz. The maximum resolvable acoustic frequency is expected
to be higher however, due to the convection velocity through the
rotor [36]. Based on L 5:96, an acoustic spectrum up to z280 Hz
is achievable.
The grid was created using the blockMesh and snappyHexMesh
utilities within the OpenFOAM

libraries. The resulting unstruc-

tured grid has a total of z4.6 M cells, consisting of approximately
90% hexahedral volumes, with the remaining cells being polyhedra.
Views of the grid are provided in Fig. 2.
4.2. Sliding interface resolution
It is important to consider the effect of the rotating grid on the
resolved turbulence. If eddies do not convect through the interface
accurately, the rotor will not see the correct turbulence spectra.
Fig. 3(a) shows turbulence structures convecting through the AMI
with only limited numerical dissipation, which is consistent with
the ndings of Bensow [37]. The largest interpolation error is seen
at the outer limits of the AMI, which is not expected to have a
large effect on the uctuations experienced by the rotor blades.
Qualitative comparison of the streamwise slices on either side of
the AMI (shown in Fig. 3(b) and (c)) shows small scale turbulence
passing into the rotor region.
Streamwise velocity spectra in Fig. 4 show that the range of
resolved scales does not appear to be affected by the AMI. The grid
cutoff of 2D has been realised, corresponding to the desired
frequency of z47 Hz. An additional measure of the interpolation
quality at the AMI is a Courant number based on the turbine
rotational velocity, i.e.
Co
UR
I
Dt
Dx
; (3)
where R
I
is the radius of the AMI. Based on a cell size Dx 0.015 m,
this results in Co z 0.0069. This is approximately 11 times smaller
than the resolution found to be sufcient for laminar vortex
shedding of a circular cylinder [38], and similar to that used by
Afgan et al. [17] for the same tidal turbine. Thus the resolution used
in the present study is deemed to be suitable for assessing
broadband noise.
5. Inow turbulence statistics
Table 3 summarises turbulence statistics sampled upstream of
the turbine rotor plane. Turbulence intensity is higher than the
desired value of 10% at both centreline probe locations. This is
partially related to the increased value of I specied at the inow
plane in order to account for streamwise decay. However, the
increase in I
x
from probe locations 1 to 2 may be attributed to the
blockage caused by the turbine. The value of L
x
at probe location 1
is also a result of the inow generator, which has been shown to
give larger integral length scales than specied [28]. A reduction in
Table 1
Summary of key tidal turbine test case parameters.
Symbol Meaning Value Unit
D Rotor diameter 0.8 m
B Number of blades 3 e
U
0
Mean freestream velocity 1.4 ms
1
n Rotational velocity 3.29 s
1
L Tip speed ratio 5.96 e
g
r
Blade root twist angle 15 deg
g
T
Blade tip twist angle 0 deg
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 744
length scale is seen at probe location 2, also due to the turbine
blockage effect.
A visualisation of the ow is provided in Fig. 5, showing tur-
bulence structures via two measures: u

x
u
x
D=U
0
is the normal-
ised streamwise vorticity, plotted on a vertical slice; Q is the second
invariant of the velocity gradient tensor, which provides a scalar
eld for vortex identication. The snapshot has been taken at
T
*
1.3, which corresponds to the transient phase of the simula-
tion, in order to show the distinction between the inowand wake
turbulence. On the left of the gure, the large streamwise length
scale of the inow turbulence may be observed; based on the
specied parameters, scales as large as the rotor diameter should
exist. Behind the rotor plane, tip vortices are seen to convect
through the AMI. Vortices also form from the turbine hub. In the
near wake region (x/D < 5) the wake structure appears fairly
coherent, while further downstream, the increase in turbulence
mixing leads to more ne scale structures.
Fig. 6 illustrates this further, using a perspective view and two
different isosurface values for Q: the value Q 0.5 ms
1
allows the
large inowturbulence structures to be seen; while the Q 5 ms
1
contour reveals the turbine wake vortices.
6. Unsteady loading
Performance assessment of the turbine is made in terms of
thrust and power. Fig. 7 depicts time traces of the turbine thrust
and power, while Fig. 8 shows the associated spectra. The instan-
taneous thrust and power coefcients monitored during the
simulation were calculated as
C
T
t
2F
x
t
r
0
AU
2
0
(4a)
and
C
P
t
2UM
x
t
r
0
AU
3
0
; (4b)
where F
x
is the thrust (streamwise force), and M
x
is torque. Time
and frequency have been normalised using the rotation rate n. The
mean values for both numerical and experimental results are
included in Fig. 7, as well as in Table 4. Also included are results for
the same case obtained using blade element momentum theory
(BEMT) by Banks et al. [39], and using LES [17]. An increase in thrust
coefcient between the LES cases with and without inow turbu-
lence is seen. No corresponding increase in power is observed
however. This agrees with similar studies from the literature
[14,17]. In comparison to the BEMT results, the LES value of C
T
is
closer to the experiment, while for C
P
the BEMT shows better
agreement. This is due to the use of wall functions in the LES, which
(in OpenFOAM

) have been shown to under-predict drag compared

to a wall-resolved boundary layer [40]. Since the BEMT includes
empirical blade section lift and drag data, this effect is not as
prominent when using this method. Afgan et al. [17] found that a
grid of 510
6
cells under-predicted thrust by z6% compared to a
Fig. 1. Schematic showing domain layout (not to scale). Patch names relate to boundary conditions in Table 2. Numbered lled circles denote probe locations one to four. Global
origin at turbine rotor plane, on rotor axis.
Table 2
Tidal turbine boundary conditions. Patch locations relative to global origin centred
on turbine rotor plane.
Patch Location Velocity Pressure
Inlet 3D Fixed value Zero gradient
Outlet 7D Convective Fixed value zero
Sides 3D Symmetry Symmetry
Top 3D Symmetry Symmetry
Bottom 3D Symmetry Symmetry
Blades Moving wall Zero gradient
Hub Slip wall Zero gradient
Fig. 2. Views of tidal turbine grid.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 745
21 10
6
cell grid. This may be a further reason for the discrepancy
in the mean thrust prediction.
Root mean square performance coefcients from the present
study are compared to those presented by Afgan et al. [17]. They
reported results for I 1, 10 and 20%, at g 20

and L 6. The
data presented in Table 4 are for I z20%, since this is closest to the
present study. The values for the present study are higher than in
Ref. [17], who simulated 24 turbine rotations, compared to 16
performed here. This suggests that a longer statistical sampling
period would reduce rms coefcients. The magnitudes of the rms
values for the LES case with inowturbulence are approximately 30
times higher than for the case without. Afgan et al. [17] presented
similarly small values for a case with I z 1%.
Two main features of the time traces of thrust and power are
evident in Fig. 7. The slowly varying part is associated with the
passage of the largest length scales; these have a period of just over
1 s, corresponding to a length scale approximately twice the inte-
gral length. The higher frequency uctuations may be attributed to
the blades cutting through long streamwise eddies. This results in
blade-to-blade correlation of the thrust and torque.
The spectra presented in Fig. 8 are characterised by humps
known as haystacks, close to the blade passing frequency (BPF; 0th
harmonic) and rst harmonic (at approximately 10 and 20 Hz
respectively). The decibel difference between the numerical spectra
and representative smooth curve at these frequencies is indicated
Fig. 3. Effect of AMI on resolved turbulence: normalised streamwise velocity u

u=U
0
, upstream of turbine.
Fig. 4. Streamwise velocity spectra at probe locations upstream and downstream of
arbitrary mesh interface: probes 1 and 2 located at x=D 0:25 and 0:375 on the
domain centreline.
Table 3
Summary of velocity probe data for tidal turbine in open domain. Probe locations
shown in Fig. 1.
Probe 1 2 3 4
x=D 0.375 0.25 0.25 0.25
y=D 0.0 0.0 0.35 0.35
u 1.20 0.97 1.11 0.81
Ix=% 13.8 15.7 19.3 24.0
Iy=% 12.7 15.2 17.0 21.0
Iz=% 15.0 20.2 19.7 21.6
Lx=m 0.81 0.43 0.3 0.21
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 746
Fig. 5. Flow visualisation of tidal turbine in open domain at T

1:3: isosurface of Q 10 s
1
and x-y plane slice of u

x
at domain centreline. Inlet on left.
Fig. 6. Flow visualisation of tidal turbine in open domain at T

1:3: isosurfaces of Q 0:5 s

1
(inow turbulence) and Q 10 s
1
(turbine wake) and x-y plane slice of u

at
z=D 2. Inlet on left.
Fig. 7. Time trace of thrust and power coefcient for cases with and without inow
turbulence. Time non-dimensionalised as t

tn. Dashed and dotted lines show nu-

merical and experimental mean values, with averaging duration indicated by length of
line.
Fig. 8. Power spectral density of time traces. Frequency normalised using the blade
passing frequency (Bn). Dashed line indicates equivalent smooth spectrum (L xP);
dotted line denotes blade passing frequency and 1st harmonic.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 747
in the gure. A magnitude of 10log
10
(B) (or z4.8 dB) is also pre-
dicted by Blake [20, chap. 10] using Equations A.1a and A.1b. This is
due to the shape of the admittance function (Equation (A.5)), which
applies only when L
x
[P=B.
An estimate of the turbine hydrodynamic pitch can be made
using the local resultant ow velocity seen by a blade section. A
radius r 0.7R has been used, with streamwise and tangential
inow factors of a
x
0.32 and a
q
0.025 estimated using BEMT
(taken from Ref. [13]). The pitch is given by P/D ptan(4), where
4 a g is the hydrodynamic pitch angle and a is the local angle of
attack. This results in P/B z 0.134 m, which conrms that the
magnitude of the haystacks seen in Fig. 8 is due to the streamwise
integral length scale exceeding the rotor pitch. Jiang et al. [41] used
a similar analytical model to analyse propeller broadband forces.
They emphasised the presence of only two prominent haystacks,
whose peaks are skewed to slightly higher frequencies than the BPF
and rst harmonic. This effect can be seen in Fig. 8.
The haystacking observed in the thrust and power spectra is
elucidated by comparing total rotor and single blade thrust, pre-
sented in Fig. 9. The combined thrust from all three blades is also
shown; this is equivalent to increasing the single blade thrust by
10log
10
(3). Note that the dashed line is comparable to the dashed
line in Fig. 8, which corresponds to a response spectrum where
L
x
< P. Summing the blade thrust spectra removes phase infor-
mation relating to blade-to-blade correlation, and hence the
spectral humps are not captured.
7. Model scale noise predictions
The acoustic sources on the blades are viewed as the sound
pressure level on the wall (SPL
w
), which is dened as
SPL
w
10log
10
_
_
p
02
w
p
2
0
_
_
: (5)
Contours of SPL
w
on the suction sides of the blades for cases with
and without inowturbulence are given in Fig. 10. Avalue range has
been chosen to highlight the difference between the two cases;
hence some of the detail of the source distribution in Fig. 10(a) is not
shown. Fig. 10(a) shows a higher source level across most of the
blade span, with the highest source amplitude located on the outer
part of the blade. The exact spanwise and chordwise location of the
highest SPL
w
is unclear however, due to much of the source level
exceeding the maximumvalue of 130 dB. This is addressed in Fig. 11.
In order to locate the dominant acoustic source more accurately,
the SPL
w
has been re-scaled and displayed close to the tip
(r/R 0.81.0). Fig. 11 reveals that the acoustic source is centred in
the outer region of the blade, but not at the tip (r/R z 0.9). This
corresponds to the location of peak spanwise loading for a typical
turbine blade [13], and is similar to that expected for blade trailing
edge noise [42]. In addition, the source is concentrated at the blade
leading edge. This was expected based on the source distributions
exhibited for typical wind turbine aerofoils [43]. Certain locations
in Fig. 11(a) do showa higher source level towards the trailing edge
however. This is also revealed in Fig. 12, and may be explained by
ow separation. A nal point to note is the higher SPL
w
on the
suction side of the blade, which indicates that the far-eld noise
may be higher downstream of the device.
Acoustic sources can also be visualised in terms of vorticity. This
follows from the fact that surface pressure uctuations result from
the passage of eddies close to the surface. Here we use Powell's
source term [44], which is a volume dipole source dened as
P V,u u; (6)
where u is the vorticity vector. This measure has been used pre-
viously to examine the effect of wind turbine blade tip shape on
noise [8]. The difference between the source distributions is clear
when comparing Fig. 12(a) and (c). A localised region of the iso-
surface towards the blade trailing edge (see Fig. 12(a) insert) is
associated with blade separation, which results in an additional
noise source. The separation is caused by a region of high velocity
turbulence impinging onto blade 1, as shown in Fig. 12(d). This
shifts the local relative velocity, increasing the angle of attack,
leading to ow separation. When the blade is not experiencing an
increased inow velocity, the size of this separated region is
reduced, as shown in Fig. 12(b).
It is known that the highest overall SPL for inow turbulence
noise occurs at q 0

, based on the dipole assumption; this data is

presented in Fig. 13, with comparison to the analytical model of
Blake [20], chap. 10, the formulation of which is included in
Appendix A. The numerical result compares favourably with the
analytical model. The maximum discrepancy between numerical
and analytical SPL is 5 dB, which occurs at the BPF. Differences
between the two analytical scenarios presented are also clear,
especially at low frequencies. Haystacks at the BPF and associated
harmonics are well captured by the simulation. The intermediate
humps are however not as clearly visible in the numerical spectrum
as the analytical; this is partly related to the bandwidth of the fast
Fourier transform and is most evident at low frequencies. This part
of the spectrum would be more accurately predicted if the total
simulation duration were increased.
Noise directivity is examined in Fig. 14. Fig. 14(a) compares the
simulation result presented in Fig. 13 to predictions made at q 45

and 90

. It is evident that the expected behaviour of a pure acoustic

dipole, assumed in the analytical model, is not fully realised in the
simulation. At the BPF, a difference of approximately 20 dB is seen
between receiver angles of 0

and 90

. This is also shown in terms

of the overall sound pressure level (OASPL), plotted in Fig. 14(b).
The OASPL is dened as
Table 4
Comparison of turbine performance coefcients: mean (top) and rms (bottom).
Values from BEMT reported in Banks et al. [39]. Large eddy simulation cases with (
y
)
and without (
z
) inow turbulence.
Case Exp. LES
y
LES
z
BEMT
C
T
1.00 0.98 0.90 0.86
C
P
0.36 0.43 0.43 0.37
Case LES
y
LES [17] LES
z
C
0
T;rms
0.0535 0.0039 0.0015
C
0
P;rms
0.0510 0.0024 0.0017
Fig. 9. Comparison of thrust spectra for single blade, all blades and rotor. Spectral level
for all blades is 10log3 higher than for single blade. Dashed line equivalent to dashed
line in Fig. 8.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 748
OASPL 10log
10
_
_
f2
f1
p
02
df
p
2
0
_
; (7)
which is the decibel level of the normalised acoustic energy across
the frequency range f
1
f
2
. The reduction in OASPL between 0

and
90

is z16 dB.
Morton et al. [45] attributed this monopole-like behaviour to
increased tip loading at low advance coefcients. The turbine
advance coefcient (J p=L 0:53) is slightly lower than that
used by Morton et al. [45], where J 0.7. This may explain the larger
increase in level observed, compared to Morton et al. [45], who saw
a 10 dB difference between the same angles. Hence there is po-
tential for further investigation of the effect of turbine operating
condition on noise directivity. Despite this, Fig. 14(a) provides evi-
dence of the improved predictive capabilities of the simulation
compared to the analytical model used here.
Cases with and without inow turbulence are now compared.
The noise directivity without inow turbulence is shown in Fig. 15,
along with the data for the case with inow turbulence at q 0

(already presented in Fig. 13). Since a 40 dB difference in SPL exists

at all frequencies between the two data sets at q 0

, it may be
concluded that the loading noise in the turbine axis is negligible
when the inow is steady. This would be expected, since there are
no incoming velocity uctuations to generate unsteady thrust on
the rotor. However, at q 45 and 90

, there is clear evidence of a

tonal noise component. This dominates at the BPF, but also appears
to have some frequency content at f n, and is known as steady
loading, which is known to be negligible for subsonic rotors [[46],
chap. 3]. This is shown by the z20 dB difference between the
maximum steady loading noise (at q 90

) and the maximum

unsteady loading noise (at q 0

).
8. Full scale noise
8.1. Acoustic scaling
In order to carry out environmental impact studies, full scale
turbine source levels are required. This data may be used to
investigate animal response experimentally [1] or make noise
impact predictions at the turbine design stage [18]. Full scale tur-
bine noise has been assessed using two approaches: scaling of the
model scale simulation results; and use of Blake's model for a full
scale turbine, as previously presented in Lloyd et al. [18]. The
scaling procedure was rst presented in Ref. [22], and is similar to
that recommended by ITTC [7], and by Wang et al. [5] for tidal
turbine measured noise.
Rudimentary scaling is based on the Strouhal number
St
fD

U
2
0
U
2
R
2
_ (8)
and the acoustic intensity
I
p
02
r
0
c
0
: (9)
Denoting model and full scale values by the subscripts 'M' and 'F',
the frequency scales as
f
F
f
M
n
F
n
M
: (10)
The acoustic intensity can be assumed to scale as
I
r
0
u
5

2
c
2
0
jrj
2
; (11)
Fig. 10. Acoustic source on suction (downstream) side turbine blades for open domain, visualised as surface sound pressure level SPLw p
0
1m Pa.
Fig. 11. Acoustic source on turbine blade tip, visualised as surface sound pressure level
SPLw p
0
1m Pa.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 749
Fig. 12. Distribution of Powell's sound source close to turbine blades: isosurfaces of P V,u u 5 10
6
s
1
(black) and rotor plane slices of normalised instantaneous streamwise velocity u

.
T
.
P
.
L
l
o
y
d
e
t
a
l
.
/
R
e
n
e
w
a
b
l
e
E
n
e
r
g
y
7
1
(
2
0
1
4
)
7
4
2
e
7
5
4
7
5
0
following Howe [[47], chap. 3]. Hence, taking u U
T
(the tip
velocity),
p
02
F
p
02
M
U
F
U
M
_ _
5

M
_ _
2
jrj
M
jrj
F
_ _
2
: (12)
It has been assumed that the speed of sound is constant be-
tween model and full scale. The result of applying this scaling
procedure to the model scale data is shown in Fig. 16. Blake's model
using full scale parameters is also included. The full scale turbine is
assumed geometrically similar to the model scale device, with a
rotor diameter of 22 m; this is reasonable for installed turbines [48],
and has been used for both empirical [19] and numerical [22]
studies. The tidal velocity is taken to be 2.5 ms
1
, with the same
turbulence characteristics as described in Section 2.
Fig. 16 is evidence that the scaling procedure is reasonably ac-
curate for this simple case. The cutoff frequency, as a result of the
scaling procedure, reduces to z14 Hz. In order to allowcomparison
between model and full scale data, and to published source levels
(which may use a different bandwidth), indicative full scale sound
levels are provided in Table 5.
The 1 Hz bandwidth source level of 1441 mPa
2
m
2
Hz
1
is z6 dB
lower than the peak value predicted from measurement data by
Wang et al. [5], using a similar scaling method. This discrepancy is
likely due to cavitation noise, which was observed in the experi-
ments, and has not been simulated. Suction side leading edge
cavitation was seen by Molland et al. [49], who tested the NACA 63-
815 section used to design the turbine blade simulated here.
A further noise source is expected to be mechanical noise, which
results from the drive train components of the turbine. Account for
this noise source can be made using an empirical model by Lloyd
et al. [19]. In this case, the resulting third-octave turbine sound
level is z160 dB re 1 mPa
2
Hz
1
at 1 m. In Lloyd et al. [19], the source
level of the mechanical and hydrodynamic noise sources was
estimated to be approximately equal. Assuming the two sources are
incoherent, their sound pressure levels may be combined as
SPL
AB
10log
10
_
_
10
SPL
A
10
10
SPL
B
10
_
_
; (13)
where A and B are two different incoherent sources. This results
in a z3 dB increase in combined source level.
8.2. Environmental impact and mitigation
The acoustic predictions presented in Fig. 16 may be used to
perform environmental impact assessments. Here a simplied
assessment is made based on the source level presented in Table 5.
The species most likely to be affected are sh, due to their hearing
sensitivity being higher at low frequencies [1]. Li and alis al [6]
found that vertical axis tidal turbine noise peaked at 4 Hz, but
did not present an impact assessment. The peak of inow turbu-
lence noise simulated here occurs at <1 Hz. Since this is below the
lowest frequency available from hearing threshold data for marine
species [4], it is hard to assess its environmental impact. The cutoff
frequency of the spectrum does however lie within the range of
hearing threshold data. At 10 Hz, typical values for sh hearing
threshold and ocean background noise are 80 dB re 1 mPa
2
[4] and
Fig. 13. Sound pressure level for model scale turbine in open domain at q 0
+
and
jrj 2D, predicted using FW-H equation and Blake's analytical model. Df 1 Hz.
Dashed line indicates equivalent smooth analytical spectrum L xP; dotted line
denotes analytical spectrum where L x[P=B.
Fig. 14. Noise directivity for model scale turbine at jrj 2D.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 751
75 dB re 1 mPa
2
Hz
1
[[50], chap. 7] respectively. The latter value is
typical of shipping noise. Although the simulation does not resolve
spectra up to the highest frequency of interest in terms of envi-
ronmental impact (z100 Hz), the maximum amplitude of the hy-
drodynamic noise is captured.
Based on the SPL of 108 dB re 1 mPa
2
Hz
1
quoted in Table 5, no
hearing threshold shift would be expected. This requires the SPL to
exceed the species' hearing threshold by at least 75 dB for 8 hour
within a 24 hour period [4]. Hearing threshold shift has been pre-
dicted by Ref. [19] however; accounting for mechanical noise as
well as an array of three turbines, a third-octave SPL of 140 dB re
1 mPa
2
Hz
1
was estimated at a frequency of 160 Hz. This agrees
with full scale measurements of tidal turbine source levels used to
carry out environmental impact assessments [1,3], and suggests
that mechanical noise may be more important at higher fre-
quencies than inow turbulence noise.
Since the unsteady inow conditions experienced by installed
turbines are difcult to avoid, noise reduction of devices would
ideally focus on reducing tip speed, due to the fth power de-
pendency of acoustic pressure on velocity. However, as turbines
typically have an optimum tip speed ratio of 5e6 [33], this strategy
would lead to a reduction in turbine efciency. Thus we recom-
mend to reduce the turbine diameter, while maintaining a constant
tip speed ratio. Due to the fact that turbine power is proportional to
D
2
u
3
, the energy generation of smaller turbines will reduce
signicantly. This may be counteracted by the installation of mul-
tiple devices [51]. A move towards utilising numerous smaller
turbines would reduce overall noise radiation per unit of power
generated. Assuming that all turbines have the same source level,
the noise due to an array of turbines would only increase by
10log
10
(N
D
) compared to a single device, where N
D
is number of
devices. Large scale arrays may cause masking however, affecting
animal communications [2].
9. Conclusions
A simulation approach for predicting tidal turbine hydrody-
namic noise has been developed and evaluated. The simulations
utilise three key components not used together before in the cur-
rent application. These are: numerically generated inow turbu-
lence; fully resolved turbine geometry using large eddy simulation
and an arbitrary mesh interface; and acoustic predictions via the
Ffowcs WilliamseHawkings equation.
The spatial and temporal quality of the interpolation at the
interface was assessed and found to be sufcient to allow the un-
steady inow to convect onto the rotor. The inow turbulence
characteristics used (anisotropic length scales and isotropic tur-
bulence intensity) captured the gross features of the turbine
response. However, the simulations could be developed to include
inhomogeneous turbulence statistics more similar to a tidal
channel.
Acoustic predictions were shown to be in good agreement with
an analytical model, and exhibited characteristic haystacks caused
by blade-to-blade correlation of the thrust response. The dominant
acoustic sources have been shown to be concentrated at towards
the blade tips, due to the high loading condition of the turbine. This
also causes the noise to radiate more akin to a monopole than a
dipole source. Estimates of full scale turbine noise were derived
using rudimentary scaling procedures, which were shown to agree
with analytical estimates. The derived source level of 144 dB re
1 mPa
2
Hz
1
at 1 m is not expected to cause physical impact to sh.
The reported simulations may also be used to assess dynamic
loads for blade design, such as root bending moment, although this
was not the aim of the present study. It is expected that this type of
simulation may be extended to allow uid structure interaction
analyses in the future.
Acknowledgements
The authors wish to thank Dr. Yusik Kimfor providing the inow
turbulence generator code. Computations were performed using
the IRIDIS 4 cluster at the University of Southampton. The nancial
support of dstl, QinetiQ and the University of Southampton is
gratefully acknowledged.
Nomenclature
Latin
A rotor disc area [m
2
]
B number of blades []
Fig. 15. Noise directivity for model scale turbine: case without inow turbulence, at
jrj 2D. Steady loading (Gutin sound) indicated at rotation rate (n) and blade passing
frequency (BPF). Data for case with inow turbulence at q 0
+
included for compar-
ison (turb).
Fig. 16. Sound pressure level for full scale turbine in open domain at q 0
+
and
jrj 2D, predicted using scaling method and Blake's analytical model. Df 0:01 Hz.
For SPL comparable to model scale data presented in Fig. 13, see Table 5.
Table 5
Full scale turbine noise levels at blade passing frequency, chosen to be representa-
tive of peak spectral level. Sound pressure level from scaled data.
Quantity Df Correction Value Unit
SPL 0.01 e 88 dB re 1mPa
2
Hz
1
SPL 1 e 108
SSL 0.01 20log
10
jrj 1 124 dB re 1m Pa
2
Hz
1
at 1 m
SSL 1 10log
10
Df 144
third octave SSL 0.4 10log
10
Df 140
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 752
Co Courant number []
C
P
power coefcient []
C
T
thrust coefcient []
c
0
sound speed [ms
1
]
D rotor diameter [m]
f frequency [s
1
]
I acoustic intensity [kgs
3
]
I turbulence intensity [%]
J advance ratio []
L length [m]
L integral length scale [m]
n rotation rate [s
1
]
P blade pitch [m]
P Powell's source term [s
2
]
p pressure [kg m
1
s
2
]
p
0
reference pressure [kg m
1
s
2
]
Q second-invariant of velocity gradient tensor [s
1
]
R turbine tip radius [m]
r radius [m]
jrj source-receiver distance [m]
T time [s]
Dt time step [s]
U
0
reference velocity [ms
1
]
u (streamwise) velocity [ms
1
]
x streamwise position [m]
Dx cell dimension [m]
Dy

w
non-dimensional rst grid cell height []
Greek
g blade twist angle [deg]
q receiver angle [deg]
L tip speed ratio []
l wavelength [m]
r
0
uid density [kg m
3
]
U angular velocity [rad s
1
]
u angular frequency [rad s
1
]
u vorticity vector [s
1
]
Subscripts
0 reference value
F full scale
I interface value
M model scale
rms root mean square
w wall value
Superscripts
* normalised
0
uctuation
-
mean
Acronyms
AMI arbitrary mesh interface
BEMT blade element momentum theory
BPF blade passing frequency
HATT horizontal axis tidal turbine
ITTC international towing tank conference
LES large eddy simulation
OASPL overall sound pressure level
PISO pressure implicit splitting of operators
SPL sound pressure level
SSL spectral source level
Appendix A. Analytical thrust loading model
As no experimental noise data is available for comparison, an
analytical model for thrust loading noise has been used [20, chap.
10]. It assumes one of two forms, depending on the ratio of the
integral length scale to the rotor pitch. It is also a free-eld model,
i.e the effects of solid boundaries on the owand acoustic radiation
are not taken into account. The mean square thrust for a bandwidth
Df is approximated as
T
2
f ; Df
_
_
_
16p
3
L

R
3R
Df
n
_
qc
J
_
2
u
02
U
2
0

S
_
u

c
R
_

2
F
_
u

q
_

Au

2
; L
x
>P (A.1a)
T
2
f ; Df
16p
3
BL

R
3R
Df
n
_
qc
J
_
2
u
02
U
2
0

S
_
u

c
R
_

2
F
_
u

q
_
; L
x
P
_
B
(A.1b)
where
u

u
_
U; L

R
L
R
_
R and L

q
L
q
_
R (A.2)
are the normalised angular frequency, and normalised radial and
circumferential integral length scales;
J U
0
_
nD; n U
_
2p and q 1
_
2r
0
U
2
0
(A.3)
are the advance coefcient, the rotational frequency and the dy-
namic pressure; c and u
02
are the blade tip chord and mean square
velocity uctuation; and P and B are the pitch and number of blades
respectively. The Sears and admittance functions are given by

S
_
u

c
R
_

2
z
1
1 puc=U
R
(A.4)
and

Au

sinpu

expfiB 1g
_
pu

B
_
sin
_
pu

B
_ : (A.5)
The length scale function is
F u

q
_ _

L

q
1 u

q
_ _
2
: (A.6)
Integral length scales in Equation A.1a and A.1b are approxi-
mated as a circumferential average of the Cartesian length scales
L
0
used in the simulations, giving L
R
L
q
z0:35 m.
The far eld acoustic mean square pressure is derived from T
2
using
p
02
jrj; f ; Df
kcosq
4pjrj
_ _
2
T
2
f ; Df ; (A.7)
where jrj is the receiver distance and k 2p/l the wavenumber.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 753
References
[1] Halvorsen M, Carlson T, Copping A. Effects of tidal turbine noise on sh
hearing and tissues. Tech. Rep. PNNL-20786. Sequim: Pacic Northwest Na-
tional Laboratory; 2011.
[2] Frid C, Andonegi E, Depestele J, Judd A, Rihan D, Rogers SI, et al. The envi-
ronmental interactions of tidal and wave energy generation devices. Environ
Impact Assess Rev 2012;32(1):133e9. http://dx.doi.org/10.1016/
j.eiar.2011.06.002.
[3] Parvin S, Workman R, Bourke P, Nedwell J. Assessment of tidal current turbine
noise at the Lynmouth site and predicted impact of underwater noise at
Strangford Lough. Tech. Rep. Subacoustech Ltd.; 2005
[4] Richards S, Harland E, Jones S. Underwater noise study supporting scottish
executive strategic environmental assessment for marine renewables. Tech.
Rep. QinetiQ Ltd; 2007
[5] Wang D, Atlar M, Sampson R. An experimental investigation on cavitation,
noise, and slipstream characteristics of ocean stream turbines. Proc Inst Mech
Eng Part J Power Energy 2007;221(2):219e31. http://dx.doi.org/10.1243/
09576509JPE310.
[6] Li Y, alis al SM. Numerical analysis of the characteristics of vertical axis tidal
current turbines. Renew Energy 2010;35(2):435e42. http://dx.doi.org/
10.1016/j.renene.2009.05.024.
[7] ITTC. Cavitation committee report. In: 18th International Towing Tank Con-
ference. 18the24th October, Kobe; 1987.
[8] Fleig O, Iida M, Arakawa C. Wind turbine blade tip ow and noise prediction
by large-eddy simulation. J Sol Energy Eng 2004;126(4):1017e24. http://
dx.doi.org/10.1115/1.1800551.
[9] Sezer-Uzol N, Long LN. 3-D time-accurate CFD simulations of wind turbine
rotor owelds. In: Proceedings of the 44th AIAA Aerospace Sciences Meeting
and Exhibit. 9th-12th January, Reno; 2006. pp. 1e23.
[10] Tadamasa A, Zangeneh M. Numerical prediction of wind turbine noise. Renew
Energy 2011;36(7):1902e12. http://dx.doi.org/10.1016/j.renene.2010.11.036.
[11] McCann GN. Tidal current turbine fatigue loading sensitivity to waves and
turbulence a parametric study. In: Proceedings of the 7th European Wave and
Tidal Energy Conference. 11the13th September, Porto; 2007.
[12] Nicholls-Lee R, Turnock S, Boyd S. Application of bend-twist coupled blades
for horizontal axis tidal turbines. Renew Energy 2013;50:541e50. http://
dx.doi.org/10.1016/j.renene.2012.06.043.
[13] Turnock SR, Phillips AB, Banks J, Nicholls-Lee R. Modelling tidal current tur-
bine wakes using a coupled RANS-BEMT approach as a tool for analysing
power capture of arrays of turbines. Ocean Eng 2011;38(11e12):1300e7.
http://dx.doi.org/10.1016/j.oceaneng.2011.05.018.
[14] McNaughton J, Rolfo S, Apsley DD, Stallard T, Stansby PK. CFD power and load
prediction on a 1MW tidal stream turbine with typical velocity proles from
the EMEC test site. In: Proceedings of the 10th European Wave and Tidal
Energy Conference. 2nde5th September, Aalborg; 2013.
[15] Gant S, Stallard T. Modelling a tidal turbine in unsteady ow. In: Proceedings
of the 18th International Ocean and Polar Engineering Conference. Vancou-
ver; 2008. pp. 473e9.
[16] Churcheld MJ, Li Y, Moriarty PJ. A large-eddy simulation study of wake
propagation and power production in an array of tidal-current turbines.
PhilosTrans Ser Math Phys Eng Sci 2013;371(20120421):1e15. http://
dx.doi.org/10.1098/rsta.2012.0421.
[17] Afgan I, McNaughton J, Rolfo S, Apsley D, Stallard T, Stansby P. Turbulent ow
and loading on a tidal stream turbine by LES and RANS. Int J Heat Fluid Flow
2013;43:96e108. http://dx.doi.org/10.1016/j.ijheatuidow.2013.03.010.
[18] Lloyd T, Turnock S, Humphrey V. Modelling techniques for underwater noise
generated by tidal turbines in shallow waters. In: Proceedings of the 30th
International Conference on Ocean, Offshore and Arctic Engineering.
19the24th June, Rotterdam; 2011.
[19] Lloyd T, Humphrey V, Turnock S. Noise modelling of tidal turbine arrays for
environmental impact assessment. In: Proceedings of the 9th European Wave
and Tidal Energy Conference. 5th-9th September, Southampton, UK; 2011.
[20] Blake W. Aero-hydroacoustics for ships. Tech. Rep.vol. II. Bethesda: David W.
Taylor Naval Ship Research and Development Center; 1984
[21] Moreau S, Roger M. Competing broadband noise mechanisms in low-speed
axial fans. AIAA J 2007;45(1):48e57. http://dx.doi.org/10.2514/1.14583.
[22] Lloyd TP, Turnock SR, Humphrey VF. Computation of inow turbulence noise
of a tidal turbine. In: Proceedings of the 10th European Wave and Tidal Energy
Conference. 2nde5th September, Aalborg; 2013.
[23] Sagaut P. Large eddy simulation for incompressible ows. Berlin: Springer-
Verlag; 2006, ISBN 978-3540263449.
[24] Zang Y, Street R, Koseff J. A dynamic mixed subgrid-scale model and its
application to turbulent recirculating ows. Phys Fluids Fluid Dyn 1993;5(12):
3186e96. http://dx.doi.org/10.1063/1.858675.
[25] James M, Lloyd T. Large eddy simulations of circular cylinders at a range of
Reynolds numbers. In: Proceedings of ITTC Workshop of Wave Run-up and
Vortex Shedding. 17the18th October, Nantes; 2013.
[26] Spalding DB. A single formula for the Law of the Wall. J Appl Mech
1961;28(3):455e8. http://dx.doi.org/10.1115/1.3641728.
[27] Kim Y, Castro IP, Xie ZT. Divergence-free turbulence inow conditions for
large-eddy simulations with incompressible ow solvers. Comput Fluids
2013;84:56e68. http://dx.doi.org/10.1016/j.compuid.2013.06.001.
[28] Lloyd TP. Large eddy simulations of inow turbulence noise: application to
tidal turbines. Ph.D. thesis. University of Southampton; 2013.
[29] Issa R. Solution of the implicitly discretised uid ow equations by operator-
splitting. J Comput Phys 1986;62(1):40e65. http://dx.doi.org/10.1016/0021-
9991(86)90099-9.
[30] Ffowcs Williams J, Hawkings D. Sound generation by turbulence and surfaces
in arbitrary motion. Philos Trans Royal Soc Math Phys Eng Sci
1969;264(1151):321e42. http://dx.doi.org/10.1098/rsta.1969.0031.
[31] Makarewicz R. Is a wind turbine a point source? J Acoust Soc Am 2011;129(2):
579e81. http://dx.doi.org/10.1121/1.3514426.
[32] Welch P. The use of fast Fourier transform for the estimation of power spectra.
IEEE Trans Audio Electroacoust 1967;15(2):70e3. http://dx.doi.org/10.1109/
TAU.1967.1161901.
[33] Bahaj AS, Molland AF, Chaplin JR, Batten W. Power and thrust measurements
of marine current turbines under various hydrodynamic ow conditions in a
cavitation tunnel and a towing tank. Renew Energy 2007;32(3):407e26.
http://dx.doi.org/10.1016/j.renene.2006.01.012.
[34] Walker JM, Flack KA, Lust EE, Schultz MP, Luznik L. Experimental and nu-
merical studies of blade roughness and fouling on marine current turbine
performance. Renew Energy 2014;66:257e67. http://dx.doi.org/10.1016/
j.renene.2013.12.012.
[35] Pope SB. Turbulent ows. Cambridge: Cambridge University Press; 2000, ISBN
978-0521598866.
[36] Carolus TH, Schneider M, Reese H. Axial ow fan broad-band noise and pre-
diction. J Sound Vib 2007;300(1e2):50e70. http://dx.doi.org/10.1016/
j.jsv.2006.07.025.
[37] Bensow RE. Simulation of unsteady propeller loads using OpenFOAM. In:
Proceedings of the 16th Numerical Towing Tank Symposium. 2nd-4th
September, Mlheim; 2013.
[38] McNaughton J, Afgan I, Apsley DD, Rolfo S, Stallard T, Stansby PK. A simple
sliding-mesh interface procedure and its application to the CFD simulation of
a tidal-stream turbine. Int J Numer Methods Fluids 2013. http://dx.doi.org/
10.1002/d.3849.
[39] Banks J, Bercin K, Lloyd TP, Turnock SR. Fluid structure interaction analyses of
tidal turbines. In: Proceedings of the 16th Numerical Towing Tank Sympo-
sium. 2nde4th September, Mlheim; 2013.
[40] Banks J. Modelling the propelled resistance of a freestyle swimmer using
Computational Fluid Dynamics. Ph.D. thesis. University of Southampton;
2013.
[41] Jiang CW, Chang M, Liu Y. The effect of turbulence ingestion on propeller
broadband forces. In: Proceedings of 19th Symposium on Naval Hydrody-
namics. 23rde28th August, Seoul; 1994. pp. 751e69.
[42] Oerlemans S, Sijtsma P, Mendezlopez B. Location and quantication of noise
sources on a wind turbine. J Sound Vib 2007;299(4e5):869e83. http://
dx.doi.org/10.1016/j.jsv.2006.07.032.
[43] Migliore P, Oerlemans S. Wind tunnel aeroacoustic tests of six airfoils for use
on small wind turbines. J Sol Energy Eng 2004;126(4):974e85. http://
dx.doi.org/10.1115/1.1790535.
[44] Powell A. Theory of vortex sound. J Acoust Soc Am 1964;36(1):177e95. http://
dx.doi.org/10.1121/1.1918931.
[45] Morton M, Devenport W, Alexander WN, Glegg SAL, Borgoltz A. Rotor inow
noise caused by a boundary layer: inow measurements and noise pre-
dictions. In: Proceedings of the 18th AIAA/CEAS Aeroacoustics Conference.
4th-6th June, Colorado Springs; 2012. http://dx.doi.org/10.2514/6.2012-2120.
[46] Goldstein ME. Aeroacoustics. New York: McGraw-Hill; 1976, ISBN 0-07-
023685-2.
[47] Howe MS. Acoustics of uid-structure interactions. Cambridge: Cambridge
University Press; 1998, ISBN 0-521-63320-6.
[48] Betschart M. Andritz Hydro Hammerfest. In: Ocean Renewable Energy Group
Annual Conference. 13th-14th September, Halifax; 2012. http://www.
marinerenewables.ca/wp-content/uploads/2012/09/Michael-Betschart-
OREG-2012.pdf.
[49] Molland AF, Bahaj AS, Chaplin JR, Batten WMJ. Measurements and predictions
of forces, pressures and cavitation on 2-D sections suitable for marine current
turbines. Proc Inst Mech Eng Part J Eng Marit Environ 2004;218(2):127e38.
http://dx.doi.org/10.1243/1475090041651412.
[50] Urick R. Principles of underwater sound for engineers. 3rd ed. New York:
McGraw-Hill; 1996, ISBN 978-0932146625.
[51] StarzmannR, Baldus M, GrohE, HirschN, Lange NA, Scholl S. Full-scale testing of
a tidal energy converter using a tug boat. In: Proceedings of the 10th European
Wave and Tidal Energy Conference. 2nd-5th September, Aalborg; 2013.
T.P. Lloyd et al. / Renewable Energy 71 (2014) 742e754 754