- Email: [email protected]
Experimental study on scour profile of pile-supported horizontal axis tidal current turbine
Accepted Manuscript Experimental study on scour profile of pile-supported horizontal axis tidal current turbine Long Chen, Roslan Hashim, Faridah Othman, Motamedi Shervin PII:
S0960-1481(17)30638-9
DOI:
10.1016/j.renene.2017.07.026
Reference:
RENE 9001
To appear in:
Renewable Energy
Received Date: 12 October 2016 Revised Date:
10 May 2017
Accepted Date: 4 July 2017
Please cite this article as: Chen L, Hashim R, Othman F, Shervin M, Experimental study on scour profile of pile-supported horizontal axis tidal current turbine, Renewable Energy (2017), doi: 10.1016/ j.renene.2017.07.026. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Experimental study on scour profile of pile-supported horizontal axis tidal current turbine Long Chen1*, Roslan Hashim1, Faridah Othman1, Motamedi Shervin2 1
Civil Engineering Department, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia 2 Faculty of Engineering, Environment and Computing, School of Energy, Construction and Environment, Coventry University, Coventry, United Kingdom
RI PT
1 2 3 4 5 6 7 8 9
Abstract
13
The study aims to investigate the influence of tip clearance on the scour rate of
14
pile-supported horizontal axis tidal current turbine (TCT) and also attempts to
15
correlate time-dependent scour depth of TCT with the tip clearance. A physical model
16
of TCT was placed in a flume for scour test and the scour rate of the fabricated model
17
was investigated. The results suggest that the decrease in tip clearance increases the
18
scour depth. In addition, the shortest tip clearance results in the fastest and most
19
sediment transport. The maximum scour depth reached 18.5% of rotor diameter.
20
Results indicate that regions susceptible to scour typically persist up to 1.0Dt
21
downstream and up to 0.5Dt to either side of the turbine support centre. The majority
22
of the scour occurred in the first 3.5 hr. The maximum scour depth reaches
23
equilibrium after 24 hr test. An empirical formula to predict the time-dependent scour
24
depth of pile-supported TCT is proposed.
26
M AN U
TE D
EP
AC C
25
SC
10 11 12
Keywords: tidal energy; tidal-current turbine; wake; scour
27 28 29
*Corresponding author Email:[email protected]; Tel: +6017-3106739. 1
ACCEPTED MANUSCRIPT 1.0 Introduction
32
The depletion of fossil fuel is a major problem for mankind. Alternative energy is
33
urgently needed to ensure the continuous development of the world economy (Wu et
34
al., 2013). The ocean has many untapped natural resources, which can be harnessed as
35
renewable energy. Tidal-current power is more easy to predict and quantify than other
36
renewable resources (Watchorn et al., 2000). The predictable tidal cycles are
37
beneficial for control of electrical grid as the generation can be forecast in advance
38
(Clarke et al., 2006). Researchers started intensive studies in the past decade with the
39
deployment of turbines in several locations (OES Annual Report, 2011, 2012, 2013).
40
Tidal-current is a potential option for future energy supplies (Rourker et al., 2010).
M AN U
SC
RI PT
30 31
41
The interaction between Tidal Current Turbine (TCT) with ambient environment is of
43
interest to many researchers. Installation of a TCT accelerates the local flow leading
44
to a change in the surrounding environment (Xia et al., 2010; Shields et al., 2011;
45
Copping et al., 2014). Collision risk, acoustic emission, sediment dynamics and
46
morphodynamics of such device have long been identified (Neill et al., 2012).
47
Continuous assessment of environmental impacts of such a device is acting as a
48
barrier for obtaining permission from relevant authorities. (Hill et al., 2014). Marine
49
Current Turbine Ltd. (MCT) and Verdant Power have had to spend multi-million
50
dollars to monitor environment impacts (Neill et al., 2009). Affordable and
51
environmental friendly tidal energy could strengthen the confidence of interested
52
parties to install tidal turbines in potential sites.
AC C
EP
TE D
42
2
ACCEPTED MANUSCRIPT Neill et al. (2009) developed large grid cell (km-scale) simulations to explore the
54
impacts of array turbine farm on sediment dynamics. It claimed that small amount of
55
energy extracted from a site might affect the erosion and deposition pattern over a long
56
distance from the point of energy extraction. The effects may be present up to 50 km in
57
the case of the Bristol Channel. Vybulkova (2013) modified vorticity transport model
58
to simulate wake and its interactions with local sediment. The results show that the
59
flows downstream of the rotor affect the marine environment and the area of the seabed
60
affected by the horizontal axis tidal turbine is within 1.0 Dt downstream of rotor,
61
where Dt is the turbine diameter. The horizontal axis turbine has a diameter of 10 m in
62
Vybulkova (2013). Hill et al. (2014) conducted a physical modelling on the scour of an
63
axial-flow hydrokinetic turbine under clear water and live-bed conditions. Results
64
indicated that the rotor of turbine increases the local shear stress of sediments around
65
the turbine. The velocity deficit in the wake region leads to the flow acceleration below
66
the rotor. The local scour of the turbine is accelerated and expanded when compared to
67
bridge scour mechanism.
SC
M AN U
TE D
68
RI PT
53
Hong et al. (2013) studied the development of a scour hole with non-cohesive
70
sediments due to the jet induced by a rotating propeller, which focused on the
71
influences of various parameters on the time-dependent maximum scour depth. Hong
72
et al. (2013) claimed that the densimetric Froude number Fo and offset height ratio
73
yo/Dp play the most important role in affecting the scour depth, where yo is offset
74
height (tip clearance), Dp is the diameter of ship propeller. The prediction model
75
proposed by Hong et al. (2013) can monitor the scour depth for a long time. The
76
documentation of ship propeller jets induced scour offers an excellent example of the
77
investigation of TCT induced scour (Hamill, 1987; Hong et al., 2013). The presence
AC C
EP
69
3
ACCEPTED MANUSCRIPT 78
of rotor also adds complexity to the scour process of TCT.
79
No equation is available to predict the scour depth of TCT to date. Chen and Lam
81
(2014) review the equations used to predict the scour depth around the foundation of
82
bridge pier/pile and they suggested that the inclusion of the rotor effects into the scour
83
prediction. Zhang et al. (2015) developed a numerical model based on commercial
84
CFD package FLOW-3D to simulate the current induced scour around the support of
85
tidal turbine.
SC
RI PT
80
86
In addition, the scour process took a long time to reach its equilibrium stage (Melville
88
and Chiew, 1999). Liu (2008) stated that coupled simulations of flow with sediment
89
transport usually take long time. Besides, Harrison et al. (2010) also claimed that a
90
full model of the rotor stands in need of sufficient mesh resolution at the blade surface
91
towards capturing the boundary layer and separation. This necessitates a very large
92
number of mesh elements and computational efforts. Modelling a rotating turbine also
93
requires that the model be unsteady and the blades change position with every time
94
step. It is assumed that the incorporation of the rotor into scour model further
95
increases the computational power. It is more feasible to conduct experimental study
96
at the current stage to understand the scour nature of TCT first.
TE D
EP
AC C
97
M AN U
87
98
The study aims to identify the effects of the tip clearance on the scour rate of TCT. The
99
study documents the temporal scour profiles of TCT. It also attempts to correlate the
100
tip clearance to the time-dependent maximum scour depth. The outcomes of the study
101
provide a fundamental understanding on the scour nature of TCT. It also gives deep
102
insights into the environmental monitoring and impact of TCT. 4
ACCEPTED MANUSCRIPT 2.0 Experiment Set-up
104
A series of clear water scour tests have been performed in the Hydraulic Laboratory of
105
University of Malaya, Malaysia. The flume is 16 m long and 1 m wide, built with
106
transparent glass wall for flow visualization (See Fig.1). The turbine is positioned 5 m
107
downstream of the channel inlet (x=0 m). The incoming flow was conditioned using a
108
porous mattress to break the turbulence structures created from the pump and supply
109
pipes, therefore offering uniform flow before entering the test section. The velocity of
110
the incoming flow is monitored by a 2D mini LDV system. The turbulence intensity
111
of the flow is approximately 2%. The turbine support tower and nacelle were made
112
from PVC pipe. The turbine rotor was made through rapid prototyping using ABS
113
material. The turbine was outfitted with a miniature DC motor for voltage
114
measurements. Reasonable tip speed ratios were maintained by the internal resistance
115
of the device. The flow depth was controlled by the tailgate, while the intake of water
116
was controlled by three pumps.
AC C
EP
TE D
M AN U
SC
RI PT
103
117 118
Fig.1: Tilting flume at Hydraulic Laboratory, University of Malaya 5
ACCEPTED MANUSCRIPT The turbine model used during these experiments was a three bladed axis flow turbine
120
with rotor diameter Dt = 0.2 m. The diameter of the turbine support tower was 0.02 m.
121
A sediment recess 0.1 m deep, 5.0 m long and 1.0 m wide was constructed inside the
122
flume. Before commencement of each experiment, the sediment was first levelled
123
using a sand leveller. The turbine position and test section were selected to be within a
124
domain where the flow and the underlying bed topography were sufficiently far from
125
inlet and outlet boundary conditions. During the tests, the TCT was embedded in the
126
middle of the bed sediment recess. The flume was slowly filled with water from both
127
upstream and downstream pipes at a low rate. This prevents the disturbance of the
128
levelled sand bed. Once the predetermined water depth was reached, the pump was on
129
and the experimental run was started by adjusting the flow rate at the inlet. Two steel
130
plates (slope in Figure 2) were used to fix the sand bed at both the beginning and end
131
of the test section. This set-up was to ensure a smooth flow of water within the test
132
section and did not disturb the sediment. The slope gradient was determined to be 1/4.
133
There is no sediment transport observed in the vicinity of the slopes during the
134
experiment.
AC C
EP
TE D
M AN U
SC
RI PT
119
135 136
Fig.2: Schematic view of scour experiment
6
ACCEPTED MANUSCRIPT The flow depth 0.45 m was maintained by adjusting the tailgate at the end of the
138
flume. The approaching velocity 0.23 m/s were also maintained constantly during the
139
experiments. At 1:70th geometric scaling the experiments in the scour investigation
140
represents a 14 m diameter turbine in water depth of 31.5 m and applying Froude
141
similitude, a mean velocity of 1.92 m/s. The experimental scaling issue of tidal
142
turbine has been highlighted by Harrison et al. (2010) and Myers and Bahaj (2010).
143
High Froude numbers often take place in laboratory that may lead to alteration of
144
water surface. Froude similarity is normally maintained between model system and
145
prototype. Myers and Bahaj (2010) also stated that discrepancy in Reynolds numbers
146
between prototype and model is acceptable for the scaling of hydraulic channels due
147
to two reasons: 1) Froude similarity is maintained; 2) both full-scale and model
148
Reynolds numbers are within the same turbulent region. The experiment condition at
149
Hydraulic Laboratory, University of Malaya are well within the turbulent region.
150
However, the blockage effect may take place in the experiment as the study presented
151
by Schluntz and Willden (2015) and Koh and Ng (2017). The blockage may cause the
152
turbine experience higher velocity.
SC
M AN U
TE D
EP
153
RI PT
137
The water was carefully drained off from the flume at the end of each test to prohibit
155
the disturbance to the scour profiles. The scour profiles around the pile-supported
156
TCT are measured by a laser distance meter with an accuracy of 0.1 mm. The laser
157
distance meter measured the bed elevation for each 1 cm in length and 1 cm in width.
158
The detailed results of the experiments are presented in Table 1. The results and
159
discussion in Section 3 highlight the spatial and temporal evolution of localized scour
160
under clear water conditions.
AC C
154
161 7
ACCEPTED MANUSCRIPT The geometric standard deviation was calculated from = (d84/d16)0.5 = 1.3 which is
163
smaller than 1.4. It is indicating that the sand was uniformly graded (Barkdoll et al.
164
2007). The flow depth was also set to ensure that the incoming flow velocity at the
165
approach section (uf) was below the critical incipient velocity (ucr) (uf
166
for the sand bed-material of d50=1.0 mm) based on the Shield curve (Wu and Wang,
167
1999). Fig.2 shows a schematic diagram of the test section, turbine and
168
instrumentation used during experiments. Fig.3 illustrates the photo of turbine model
169
and flume.
M AN U
SC
RI PT
162
EP
172
Fig.3: Photo of turbine model (Dt = 0.2 m) and flume
AC C
171
TE D
170
8
ACCEPTED MANUSCRIPT 173
Table 1: Turbine, flow flume and sediment parameter used during the experiment Variable
Clear Water
0.2
Hydrofoil
NACA 63418
Hub length (m)
0.1
C/Dt
0.25, 0.50, 0.75
TSR
3.67
SC
Dt (m)
RI PT
Turbine parameters
M AN U
Flow, flume and sediment parameters H (m)
0.45
Qw (m3/s)
0.117
b (m)
1.0
δ (%)
RT V∞ (m/s)
AC C
ucr (m/s) d50 (mm) 174
TE D
Rb
EP
Fr
7
0.11 1.04 x 105 4.6 x 104 0.23 0.032 1.0
9
ACCEPTED MANUSCRIPT 3.0 Results and Discussion
176
3.1 Spatial Scour Profile
177
The final bed topography from the clear water scour tests is shown in Fig.4. It aims to
178
investigate the effect of rotor tip clearance on the sediment transport around the
179
support of TCT. During the test, no sediment transport occurred upstream of the TCT,
180
confirming clear water scour conditions and local scour and deposition induced by the
181
TCT. The scour induced by TCT of the three cases has similar pattern. Sediment
182
around the support is eroded and a hole is formed. Sediment eroded from the scour
183
region was deposited behind in a hill shaped dune. The maximum scour depth reached
184
18.5% of rotor diameter (5 cm tip clearance), Dt, or 7% of the flow depth, h. The
185
details of cross section scour profile can be found at Section 3.2. The results show that
186
the scour depth increases with the decrease of tip clearance. This is in line with Zhang
187
et al. (2015). The rotor clearance in Zhang et al. (2015) is lowered from 5.2 m to 1.2
188
m, the maximum depth of scour varies from 1.31 (8% of Dt) to 1.50 m (9.4% of Dt).
189
Results also indicate that regions susceptible to scour typically persist up to 0.5Dt
190
downstream and up to 0.5Dt to either side of the turbine support centre. The peak of
191
the deposition reached elevation of 10% of rotor diameter.
AC C
EP
TE D
M AN U
SC
RI PT
175
10
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
192
a. 5cm tip clearance
AC C
EP
TE D
193
194 195
b. 10cm tip clearance 11
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
196 197
c. 15cm tip clearance
198
Fig.4: 3D surface map of 24-hr scour by SURFER
TE D
199
As discovered in Chen and Lam (2014), the obstruction induced by the spinning rotor
201
results in acceleration of the flow between the bottom rotor tip and bed surface. This
202
acceleration increases the local shear stress and it amplifies the scour mechanism. It
203
therefore the short tip clearance (5 cm) can cause severe sediment transport around
204
the support of TCT. The extent of scour hole increases with the decrease of tip
205
clearance. The highest tip clearance 15 cm results in the least eroded sand from the
206
bed. The scour holes and deposited sand are asymmetric especially in the case of 5 cm
207
tip clearance. This is due to asymmetric shear stress impacted at the bed. This also
208
conforms to the study conducted by Zhang et al. (2015).
AC C
EP
200
209 12
ACCEPTED MANUSCRIPT 3.2 Temporal Scour Profiles
211
Fig. 5, 6 and 7 present the temporal dimensionless scour holes profile at 3.5, 7.0 and
212
24 hr. The turbine rotor places 5 cm, 10 cm and 15 cm from bed, respectively. The
213
investigated cases are in the same manner in terms of scour development. Looking
214
closely at the measured developments of scour profile at various lateral positions, it
215
can be observed that a scour hole is formed around the support tower. The majority of
216
scour occurred at the initial 3.5 hr. The size of scour holes increase with time. The
217
scour depth slightly increases after 3.5 hr scour. A deposition mound is also observed
218
downstream at the initial stage. The depositional wedge also moves further
219
downstream as the test in progress. A secondary sand dune is formed right behind the
220
primary sand dune after 24 hr test.
AC C
EP
TE D
M AN U
SC
RI PT
210
13
ACCEPTED MANUSCRIPT In the case of 5 cm tip clearance, maximum scour occurred at approximately x/Dt=
222
-0.1, y/Dt= -0.1, reached the depth of S/Dt=0.185. The highest depositional wedge
223
reached elevations of approximately 10% of the Dt at location x/Dt=1.0, y/Dt=0.1. The
224
highest secondary sand dune is only 3.5% of Dt at location x/Dt=1.8, y/Dt= 0.2. It is
225
observed that there is no secondary sand dune at the first 7 hr scour test. It is probably
226
due to the primary deposition wedge reaching a certain height where it becomes
227
susceptible to the flow below turbine rotor. The accelerated flow below rotor turbine
228
eroded the deposited sand dune.
M AN U
SC
RI PT
221
229
In the case of 10 cm tip clearance, maximum scour depth is approximately 16% of Dt
231
and took place at x/Dt= -0.05, y/Dt=0 and x/Dt=-0.05, y/Dt= -0.1. The highest primary
232
sand dune has an elevation of 0.075Dt and it happened after 7 hr scour test other than
233
24 hr. This may be due to the primary scour dune deposition is less than its erosion in
234
the subsequent test. The highest primary deposition located at the region of
235
x/Dt=0.6-0.7, y/Dt=0.0-0.1. Secondary dune is also formed but with a lower elevation.
236
The highest secondary sand dune is only 3% of Dt at x/Dt=1.5, y/D=0.0.
EP
AC C
237
TE D
230
238
For the investigated case of 15 cm tip clearance, maximum scour depth occurred at
239
x/Dt= -0.05, y/Dt=0 and x/Dt=-0.05, y/Dt= -0.1 which is same as the case of 10 cm tip
240
clearance. The maximum scour depth is approximate 0.145Dt. The highest primary
241
deposition is 8.5% of Dt located at x/Dt= 0.7, y/Dt=0 and x/Dt=0.7, y/Dt= -0.1. The
242
secondary sand dune is getting lower and lower as the rotor of turbine rotor moves up. 14
ACCEPTED MANUSCRIPT 243
The highest secondary sand dune is only 2% of turbine diameter at locations x/Dt= 1.2,
244
y/Dt=0 and x/Dt=1.2, y/Dt= -0.1.
245
As seen from all the investigated cases, it has been observed that the maximum scour
247
depth increase as the decrease of tip clearance. The maximum scour depth generally
248
occurred in front of the support pile. It is generally at the region x/Dt=-0.05 – -0.1 and
249
y/Dt=0.0 - -0.1. The height of primary and secondary deposited sand dunes also
250
increases as the decreases of tip clearance. The location of peak sand dune is a bit
251
complicated due to the influence of the flow between turbine rotor and bed.
M AN U
SC
RI PT
246
252
Furthermore, the transition point between net erosion and net deposition moves
254
downstream as the test progresses. The furthest transition point located at x/Dt=0.5 in
255
all the investigated cases after 24 hr scour test. The transition points of Hill et al.
256
(2014) are between x/Dt=1.0 to x/Dt =2.0. This region has been noted by Chamorro et
257
al. (2013) and they claimed that the turbine wake coherence starts to break down in
258
this region where the most rapid velocity deficit recovery begins. However, the
259
furthest transition point of the current study located at x/Dt=0.5 in all the cases after
260
24 hr scour test. The difference of the locations of the transition point may be due to
261
the difference of turbine geometry and flow velocity.
AC C
EP
TE D
253
15
ACCEPTED MANUSCRIPT 0.10
0.05
-0.05
RI PT
S/Dt
0.00
-0.10
3.5hr 7.0hr 24hr
-0.20 -1.0
-0.5
0.0
1.0 x/Dt
1.5
2.0
2.5
M AN U
262 263
a.
0.10 0.05
S/Dt
TE D
0.00 -0.05
-0.15
AC C
-0.20
3.5hr
EP
-0.10
-1.0
264 265 266
0.5
SC
-0.15
-0.5
7.0hr 24hr 0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
b.
16
ACCEPTED MANUSCRIPT 0.10 0.05
S/Dt
0.00
RI PT
-0.05 -0.10
3.5hr 7.0hr
-0.15 -0.20 -1.0
-0.5
0.0
0.5
1.0 x/Dt
1.5
2.0
2.5
M AN U
267
SC
24hr
268
c.
0.10
0.05
TE D
S/Dt
0.00
-0.05
3.5hr
EP
-0.10
7.0hr
AC C
-0.15
24hr
-0.20
-1.0
269 270 271
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
d.
17
ACCEPTED MANUSCRIPT 0.10 0.05
-0.05
RI PT
S/Dt
0.00
-0.10
3.5hr 7.0hr
-0.15 -0.20 -1.0
-0.5
0.0
0.5
1.0
273 274 275 276
e.
1.5
2.0
2.5
M AN U
272
x/Dt
SC
24hr
Fig.5: Temporal dimensionless scour holes profile at different times (Tip clearance=5cm); a. y/Dt = 0; b. y/Dt= -0.1; c. y/Dt = 0.1. e. y/Dt = -0.2; f. y/Dt = 0.2.
277 278
TE D
0.10 0.05
EP
-0.05
3.5hr
AC C
S/Dt
0.00
-0.10
7.0hr
-0.15
24hr
-0.20
-1.0
-0.5
0.0
0.5
1.0
279
x/Dt
280 281
a.
1.5
2.0
2.5
18
ACCEPTED MANUSCRIPT 0.10 0.05
S/Dt
0.00
RI PT
-0.05 -0.10
3.5hr
-0.15
7.0hr 24hr
-1.0
-0.5
0.0
0.5
0.05
-0.05
TE D
S/Dt
0.00
-0.10
EP
-0.15 -0.20
AC C
-1.0
285 286
1.5
2.0
2.5
M AN U
b. 0.10
284
1.0 x/Dt
282 283
SC
-0.20
-0.5
0.0
3.5hr 7.0hr 24hr
0.5
1.0
1.5
2.0
2.5
x/Dt
c.
19
ACCEPTED MANUSCRIPT 0.10
0.05
RI PT
S/Dt
0.00
-0.05
3.5 hr
-0.10
7.0 hr
-0.15 -1.0
-0.5
0.0
0.5
287 288
0.05
TE D
S/Dt
0.00
-0.05
EP
-0.10
-0.15
1.5
2.0
2.5
M AN U
d. 0.10
-0.5
0.0
3.5 hr 7.0hr 24 hr
0.5
1.0
1.5
2.0
2.5
x/Dt
AC C
-1.0
289 290 291 292 293
1.0 x/Dt
SC
24 hr
e. Fig.6: Temporal dimensionless scour holes profile at different times (Tip clearance=10 cm); a. y/Dt = 0; b. y/Dt = -0.1; c. y/Dt = 0.1. e. y/Dt = -0.2; f. y/Dt = 0.2.
20
ACCEPTED MANUSCRIPT 0.10
0.05
-0.05
RI PT
S/Dt
0.00
-0.10
3.5hr
-0.15
7.0hr
-0.20 -1.0
-0.5
294 295
0.0
0.5
x/Dt
0.10 0.05
TE D
0.00 S/Dt
1.5
2.0
2.5
M AN U
a.
1.0
SC
24hr
-0.05 -0.10
EP
-0.15
3.5hr 7.0hr 24hr
-0.20
297 298
AC C
-1.0
296
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
b.
21
ACCEPTED MANUSCRIPT 0.10
0.05
RI PT
S/Dt
0.00
-0.05
3.5hr
-0.10
7.0hr 24hr
SC
-0.15
-0.20 -0.5
0.0
0.5
1.0
1.5
2.0
2.5
M AN U
-1.0
x/Dt
299 300
c.
0.10
TE D
0.05
S/Dt
0.00
3.5hr
EP
-0.05
AC C
-0.10
7hr 24hr
-0.15
-1.0
301 302
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
d.
22
ACCEPTED MANUSCRIPT 0.10
0.05
RI PT
S/Dt
0.00
-0.05
3.5hr
-0.10
7hr
-0.15 -1.0
-0.5
0.0
0.5 x/Dt
303
1.5
2.0
2.5
EP
TE D
M AN U
e. Fig.7: Temporal dimensionless scour holes profile at different times (Tip clearance=15cm); a. y/Dt = 0; b. y/Dt = -0.1; c. y/Dt = 0.1. e. y/Dt = -0.2; f. y/Dt = 0.2.
AC C
304 305 306
1.0
SC
24 hr
23
ACCEPTED MANUSCRIPT 3.3 Temporal Variation of Maximum Scour Depth
308
Fig.8 illustrates the temporal scour depth under different tip clearances. It has been
309
observed that the majority of scour of all three cases occurred in the first 3.5 hours.
310
After 7 hr of scour test, all three cases reach near-equilibrium conditions. The depth of
311
scour increased more quickly in the case with 5 cm tip clearance, while the slowest
312
scour occurred in the case of 15 cm tip clearance. The maximum scour also occurred
313
in the case of 5 cm tip clearance, which is approximately 18.5% of rotor diameter Dt
314
(8% of flow depth). The least scour occurred in the case of 15 cm tip clearance, which
315
is approximately 14% of rotor diameter Dt (6% of flow depth).
M AN U
SC
RI PT
307
0.20 0.18 0.16
TE D
0.14
St/Dt
0.12 0.10
0.06
AC C
0.04
EP
0.08
Clearance 5 cm Clearance 10cm Clearance 15cm
0.02 0.00
0
316 317
200
400
600 800 t (mins)
1000
1200
1400
1600
Fig.8: Time-dependent maximum scour depth
24
ACCEPTED MANUSCRIPT 318
Research to provide methods for predicting scour depth around the pile-supported
319
structure of tidal turbine is limited. The maximum scour depth at any time t, St, is
320
assumed to be a function of 10 independent variables as follows:
322
= ( , , , , , , , , , )
323
RI PT
321
Equation 1
In the works of ship propeller jet induced scour (Rajaratnam, 1981; Hamill, 1999;
325
Hong et al, 2013), the fluid viscosity could be neglected as the Reynolds number of
326
the jet is greater than 10,000. The Reynolds number of the wake induced by the
327
physical model of the turbine is around 29,100. The viscosity term is ignored in the
328
following analysis. The water depth of the study is reasonable high, the Froude
329
number is less than 0.1 and its effects on scour are not examined in the study. The
330
efflux velocity is calculated from
331
(2015). By referencing the work of Hong et al. (2013), the Buckingham π theorem
332
has been applied and choosing , , as fundamental variables. Equation (2) has
333
been obtained as the follows:
335 336
M AN U
TE D
= 1 − , as introduced by Lam et al.
EP
AC C
334
SC
324
" !
= # (
$ %&! '! " !
,
" !
,
" !
, ((
(
) *()
, ) +
Equation 2
337
Based on experience of ship propeller jet induced scour and re-arrange the above
338
equation, the maximum depth of scour hole could be written as:
339 25
ACCEPTED MANUSCRIPT
340
$
= # (,- ,
'
) $ %&! $ ⁄ +
,
'
,
Equation 3
341 () *(
) and ⁄ - , which is denoted by tR, is the
342
In Equation (3), Fo= - 0/(
343
proposed time scale. A nonlinear regression analysis has been performed and an
344
empirical equation for the estimation of time-dependent scour depth of pile-supported
345
TCT is proposed as:
RI PT
(
347
where
349
Equation 4
'
'
1 = 0.01 ($ )* .< (% ) . = , .<<>
350
&!
.@A . = * . = ) ( ) ,
TE D
351
1# = 1.9 (
352
= 1 [34 5 $+ 6 − 1# ]89
M AN U
348
$
SC
346
EP
1> = 2.18 (
. = * . < * . ) ( ) ,
Equation 4 is in the same form of equation proposed by Hong et al. (2013), which is
354
developed to predict the scour induced by ship propeller jets. Fig. 9 presents the
355
overall comparison between the computed and observed temporal maximum scour
356
depth. The R-squared value ranges from 0.7747 to 0.9737. The best fitting of the
357
experimental data with predicted value is in the case C=10 cm. The lowest R-squared
358
value in the case of C=15 cm is 0.7747 which is still consider very good based on
359
Henriksen et al. (2003). Equation 4 has good agreement with present experimental
AC C
353
26
ACCEPTED MANUSCRIPT 360
data under the conditions of 0.25 ≤ C⁄ ≤ 0.75 , 50 ≤ C⁄ ≤ 150 and
361
, = 1.8965.
RI PT
0.20
SC
0.10
M AN U
Predicted St/Dt
0.15
C=5cm, R2=0.9692
0.05
C=10cm, R2=0.9737 C=15cm, R2=0.7747
0.00 0.00
0.10
0.15
0.20
Measured St/Dt
TE D
362
0.05
Fig.9: Comparison between measured and predicted maximum time-dependent scour
364
depth
AC C
EP
363
27
ACCEPTED MANUSCRIPT 3.4 Comparison of Scour Evolution
366
A comparison between the authors’ experimental results and Hill et al. (2014) of
367
temporal scour depth is presented in Fig. 10. The time has been normalized by the
368
respective velocity at hub height. It shows that the large scale experiment did not
369
approach
370
near-equilibrium. Hill et al. (2014) also indicated that the sour hole evolves faster
371
when the rotor is on the upstream side of the support tower. The temporal evolution of
372
scour depth of small-scale experiments indicated similar pattern with the author’s
373
results (C=5 cm), particularly the case of rotor upstream. The authors’ experiments are
374
conducted with rotor upstream. The results of Hill et al. (2014) in Fig.10 are taken
375
from x/Dt=0.66 and y/Dt=0 for all cases. The authors’ results are taken from x/Dt=-0.1
376
and y/Dt=-0.1. These locations are on the downstream side of the rotors. With the
377
rotor on the downstream of the support tower, the flow acceleration between rotor and
378
seabed may take place later. The scour amplification occurs potentially further
379
downstream from the rotor.
380
The temporal evolution of scour hole for large-scale experiment is relatively faster
381
compare to small-scale experiments. It should be noted that the velocity at hub height
382
is 0.66 m/s. The tidal turbine employed in the study of Zhang et al. (2015) is a full
383
scale turbine with a 16 m diameter rotor, and the scour is quickly formed in the
384
vicinity of support structure. It only takes 30 mins the scour depth reached
385
approximately 1.31 m with current speed of 3.7 m/s. There are five different types of
386
components of gravel formed in the seabed of Zhang et al. (2015). The numerical
stage.
The
small
scale
experiments
approach
the
AC C
EP
TE D
M AN U
SC
equilibrium
RI PT
365
28
ACCEPTED MANUSCRIPT 387
setup of Zhang et al. (2015) is not identical with the current study. Therefore, the
388
influence of rotor size and sediment formation on the tidal turbine induced scour
389
cannot be seen from such comparison. Small scale: rotor downstream (Hill et al., 2014) Small scale: rotor upstream (Hill et al., 2014) Large scale: rotor downstream (Hill et al., 2014) C=5.0cm C=10cm C=15cm
0.30
0.25
SC
0.15
0.10
0.05
0.00
391 392
1.0
1.5 tVhub/Dt
2.0
2.5
3.0 x104
Fig.10: Comparison of temporal scour depth
EP
390
0.5
TE D
0.0
M AN U
St/Dt
0.20
RI PT
0.35
3. 5 Comparison with Conventional Pile Scour
394
The presence of rotor adds complexity to the scour induced by the pile-supported TCT.
395
The scour mechanism of TCT is different from the conventional pile scour. The main
396
differences come from the generated wake. The flow below rotor experience
397
acceleration and it amplifies the sediment transport at seabed. In order to take into
398
account the effect of rotor on the scour of TCT, the proposed model in Section 3.3 is
399
based on the reference guide of ship propeller jets induced scour. The attempts by
400
referencing ship propeller jet induced scour predictor is promising and the model can
AC C
393
29
ACCEPTED MANUSCRIPT be used as a starting point for future model development.
402
There are many existing scour predictors as documented by Chen and Lam (2014). To
403
demonstrate the applicability of the existing models for the prediction of TCT induced
404
scour, two equations (Breuser et al., 1977；Richardson and Davis, 2001) are used to
405
calculate the scour depth for the experimental condition of the current study. These
406
two equations have been selected by Harris et al. (2010) to develop STEP to predict
407
the development of scour evolution of offshore structures. The two equations are
408
shown as the follows:
409
$
410
M AN U
SC
RI PT
401
P
= 1.5I IJ IK I% tanh( $ ) IK = 0 ,
411
412
TE D
RS IK = 2 V W − 1, RST IK = 1,
Q
Equation 5
RS < 0.5 RST
Q 0.5 ≪
Q
RS <1 RST
RS ≫1 RST 0.35
= 2.0IZ I[ I\ I I] ()
,_ 0.43
Equation 6
and
414
IK is the correction factor for bed condition, I% is the correction factor for size of
415
bed material, IJ is the correction factor for flow angle of attack, I is the correction
416
factor of pier shape, Uc is the depth-averaged current speed and Ucr is the threshold
417
depth-averaged current speed. Ia is the enhance correction factor for pier width/pile
418
diameter. Table 2 tabulates the results of the calculation. It shows that the use of
419
equations proposed by Breuser et al. (1977) and Richardson and Davis (2001) are not
420
applicable for the experiment of current study. These two equations are under- or
421
over- predicted for most of the current experimental cases. It is noticed that the scour
AC C
EP
413
30
ACCEPTED MANUSCRIPT depth in the case C=15 cm is very close to the predicted value by Breuser et al.
423
(1977). This may due to the rotor is further to the sand bed so that the amplification to
424
the bed stress is lesser compared to other cases. The equation proposed by Breuser et
425
al. (1977) could be used to predict the equilibrium scour depth around TCT when the
426
rotor is far away from the bed.
RI PT
422
Table 2: Factors used for scour depth prediction analysis Tip Clearance 5cm 10cm 15cm
Measured (cm) 3.5 3.1 2.9
Predicted by Breuser et al., Predicted by Richardson and Davis, 2001 (4.58 cm) 1977 (2.7 cm） 22.9%, 30.9%, 12.9% 47.7%, 6.9% 57.9%
M AN U
428
SC
427
4.0 Conclusion
430
The presence of rotor amplifies the scour process around the support of TCT. The
431
decrease of tip clearance increases the scour depth. The maximum scour depth
432
reached 18.5% of the rotor diameter. Results indicate that regions susceptible to scour
433
typically persist up to 1.0Dt downstream and up to 0.5Dt to either side of the turbine
434
support centre. The tip clearance plays the most important role in affecting the scour
435
depth and scour rate. The case of 5 cm tip clearance results in the fastest scour and
436
most sediment transport, while the case of 15 cm tip clearance has the slowest and
437
least sediment transport. Majority of the scour occurred in the first 3.5 hr in all the
438
cases. The scour depths of all the cases reach near-equilibrium state after 7 hr test.
439
Based on the experimental data, the / has been found to be a logarithmic
440
function of time. An empirical formula (Equation 4) has been proposed to estimate the
441
time-dependent maximum scour depth, where the efflux velocity = 1 − is
AC C
EP
TE D
429
31
ACCEPTED MANUSCRIPT an initial input for the proposed scour prediction. In future research works, it is
443
suggested to investigate the influence of turbine geometry, rotor size and sediment
444
formation on the scour issue of TCT.
445
Acknowledgement
446
The authors wish to extend their gratitude to Ministry of Higher Education Malaysia
447
and University of Malaya for the financial support received under the UM/MOHE
448
High
449
UM.C/HIR/MOHE/ENG/47. The authors also wish to thank Professor. Dr. Lam Wei
450
Haur. His efforts to initiate the research project are highly appreciated.
451
List of Symbols
452
yo
propeller tip clearance
453
C
turbine tip clearance
454
CT
thrust coefficient
455
Dt
diameter of the turbine disc
456
Dp
diameter of propeller
457
Vo
efflux velocity in m/s
458
V∞
free stream velocity in m/s
459
Vx,r
lateral distribution velocity in m/s
460
x
longitudinal (streamwise) direction
461
y
lateral (spanwise) direction
462
n
rotational speed in rev/s
463
r
radial distance
UM.C/HIR/MOHE/ENG/34
SC
Grant
and
M AN U
Research
AC C
EP
TE D
Impact
RI PT
442
32
ACCEPTED MANUSCRIPT R
radius of turbine rotor
465
ReD
Reynolds number
466
S
Scour Depth
467
St
Scour depth at time t
468
ucr
threshold depth averaged current velocity
469
Uc
depth averaged current velocity
470
Uuc
undisturbed current velocity
471
Rb
bulk Reynolds number (Rb=hV∞/ )
472
RT
turbine rotor Reynolds number
473
U
velocity at the edge of the bed boundary layer,
474
Qw
flow discharge
475
Kinematic viscosity of water
476
Fr
Froude number of incoming flow
477
Fo
densimetric Froude number
478
H
flow depth
479
hp
pile height
480
δ
481
d50
median sediment grain size
482
g
acceleration due to gravity
483
∆
difference between the mass density of the sediment and the fluid
484
density of fluid
485
a
density of water
AC C
EP
TE D
M AN U
SC
RI PT
464
blockage ratio
33
ACCEPTED MANUSCRIPT
density of sand
487
d
turbulent viscosity
488
geometric standard deviation of sediment particles
489
IK
correction factor for bed condition,
490
I%
correction factor for size of bed material,
491
IJ
correction factor for flow angle of attack,
492
I
correction factor of pier shape
493
Ia
enhance correction factor for pier width/pile diameter.
AC C
EP
TE D
M AN U
SC
RI PT
486
34
ACCEPTED MANUSCRIPT Reference Barkdoll, B. D., Ettema, R., & Melville, B. W. (2007). Countermeasures to protect bridge abutments from scour (Vol. 587). Transportation Research Board.
RI PT
Breusers, H. N. C., Nicollet, G., & Shen, H. W. (1977). Local scour around cylindrical piers. Journal of Hydraulic Research, 15(3), 211-252. Chamorro, L. P., Hill, C., Morton, S., Ellis, C., Arndt, R. E. A., & Sotiropoulos, F. (2013). On the interaction between a turbulent open channel flow and an axial-flow turbine. Journal of Fluid Mechanics, 716, 658-670.
SC
Chen, L., Lam, W.H. (2014) Methods for predicting seabed scour around marine current turbine. Renewable and Sustainable Energy Reviews, 29,683-692
M AN U
Clarke, J.A., Connor, G., Grant, A.D., Johnstone, C.M. (2006) Regulating the output characteristics of tidal current power stations to facilitate better base load matching over lunar cycle. Renewable Energy, 31 (2), 173-180. Copping, A., Battey, H., Brown-Saracino, J., Massaua, M., & Smith, C. (2014). An international assessment of the environmental effects of marine energy development. Ocean & Coastal Management, 99, 3-13.
TE D
Hamill, G. A. (1987). Characteristics of the screw wash of a manoeuvring ship and the resulting bed scour (Doctoral dissertation, Queen's University of Belfast).
EP
Harris, J., Whitehouse, R. J. S., & Benson, T. (2010). The time evolution of scour around offshore structures. ICE-Maritime Engineering, 163(1), 3-17. Harrison, M. E., Batten, W. M. J., Myers, L. E., & Bahaj, A. S. (2010). Comparison between CFD simulations and experiments for predicting the far wake of horizontal axis tidal turbines. IET Renewable Power Generation, 4(6), 613-627.
AC C
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537
Henriksen, H. J., Troldborg, L., Nyegaard, P., Sonnenborg, T. O., Refsgaard, J. C., & Madsen, B. (2003). Methodology for construction, calibration and validation of a national hydrological model for Denmark. Journal of Hydrology, 280(1), 52-71. Hill, C., Musa, M., Chamorro, L. P., Ellis, C., & Guala, M. (2014). Local scour around a model hydrokinetic turbine in an erodible channel. Journal of Hydraulic Engineering, 140(8), 04014037. Hong, J. H., Chiew, Y. M., & Cheng, N. S. (2013). Scour caused by a propeller jet. Journal of Hydraulic Engineering, 139(9), 1003-1012. 35
ACCEPTED MANUSCRIPT 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558
Koh, W. X. M., & Ng, E. Y. K. (2017). A CFD study on the performance of a tidal turbine under various flow and blockage conditions. Renewable Energy, 107, 124-137.
559 560 561 562
Neill, S. P., Jordan, J. R., & Couch, S. J. (2012). Impact of tidal energy converter (TEC) arrays on the dynamics of headland sand banks. Renewable Energy, 37(1), 387-397.
563 564 565 566 567 568 569 570 571
OES-IA Annual Report 2010 (2011). Implementing Agreement on Ocean Energy Systems. Lisbon, Portugal.
572 573 574
Rajaratnam, N. (1981). Erosion by plane turbulent jets. Journal of hydraulic Research, 19(4), 339-358.
575
Richardson, E.V., & Davis, S.R. (2001). Evaluating scour at bridges (4th ed.). Washington, D.C.: Hydraulic Engineering Circular No.18 (HEC-18), Federal Highway Administration.
576 577 578 579 580 581
Lam W.H., Chen, L., Hashim, R. (2015) Analytical wake model of tidal current turbine. Energy, 512-521.
RI PT
Liu, X. (2008). Numerical models for scour and liquefaction around object under currents and waves. University of Illinois at Urbana-Champaign.
SC
Myers, L. E., & Bahaj, A. S. (2010). Experimental analysis of the flow field around horizontal axis tidal turbines by use of scale mesh disk rotor simulators. Ocean Engineering, 37(2), 218-227.
M AN U
Melville, B. W., & Chiew, Y. M. (1999). Time scale for local scour at bridge piers. Journal of Hydraulic Engineering, 125(1), 59-65.
TE D
Neill, S. P., Litt, E. J., Couch, S. J., & Davies, A. G. (2009). The impact of tidal stream turbines on large-scale sediment dynamics. Renewable Energy, 34(12), 2803-2812.
EP
OES-IA Annual Report 2011 (2012). Implementing Agreement on Ocean Energy System. Lisbon, Portugal.
AC C
OES-IA Annual Report 2012 (2013). Implementing Agreement on Ocean Energy System. Lisbon, Portugal.
Rourke, F.O., Boyle, F., & Reynolds, A. (2010). Marine current energy devices: Current status and possible future applications in Ireland. Renewable and Sustainable Energy Reviews, 14(3), 1026-36. 36
ACCEPTED MANUSCRIPT 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599
Schluntz, J., & Willden, R. H. J. (2015). The effect of blockage on tidal turbine rotor design and performance. Renewable Energy, 81, 432-441.
600 601 602 603 604 605 606 607
Vybulkova, L. (2013). A study of the wake of an isolated tidal turbine with application to its effects on local sediment transport (Doctoral dissertation, University of Glasgow).
608 609 610 611
Zhang, J., Gao, P., Zheng, J., Wu, X., Peng, Y., & Zhang, T. (2015). Current-induced seabed scour around a pile-supported horizontal-axis tidal stream turbine. Journal of Marine Science and Technology, 23(6), 929-936.
RI PT
Shields, M.A., Woolf, D.K., Grist, E.P.M., Kerr, S.A., Jackson, A.C., Harris R.E., … Jonathan, S. (2011). Marine renewable energy: the ecological implications of altering the hydrodynamics of the marine environment. Ocean and Coastal Management, 54, 2-9. Watchorn, M., Trapp, T., Sayigh, A.A.M. (2000). Tidal stream renewable offshore power generation. World Renewable Energy Congress VI. Oxford: Pergamon. 2664-667.
SC
Wu, W., & Wang, S. S. (1999). Movable bed roughness in alluvial rivers. Journal of Hydraulic Engineering, 125(12), 1309-1312.
M AN U
Wu, B., Zhang, X., Chen, J., Xu, M., Li, S., & Li, G. (2013). Design of high-efficient and universally applicable blades of tidal stream turbine. Energy, 60:187-94.
AC C
EP
TE D
Xia, J.Q., Falconer, R.A., & Lin, B. (2010). Impact of different tidal renewable energy projects on the hydrodynamic processes in the Severn Estuary, UK. Ocean Modelling, 32, 86-104.
37
ACCEPTED MANUSCRIPT Highlights:
AC C
EP
TE D
M AN U
SC
RI PT
The influence of tip clearance on the scour profile of pile-supported tidal turbine is investigated The scour profile of tidal turbine is presented A correlation between tip clearance and time-dependent scour rate of tidal turbine is proposed
S0960-1481(17)30638-9
DOI:
10.1016/j.renene.2017.07.026
Reference:
RENE 9001
To appear in:
Renewable Energy
Received Date: 12 October 2016 Revised Date:
10 May 2017
Accepted Date: 4 July 2017
Please cite this article as: Chen L, Hashim R, Othman F, Shervin M, Experimental study on scour profile of pile-supported horizontal axis tidal current turbine, Renewable Energy (2017), doi: 10.1016/ j.renene.2017.07.026. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Experimental study on scour profile of pile-supported horizontal axis tidal current turbine Long Chen1*, Roslan Hashim1, Faridah Othman1, Motamedi Shervin2 1
Civil Engineering Department, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia 2 Faculty of Engineering, Environment and Computing, School of Energy, Construction and Environment, Coventry University, Coventry, United Kingdom
RI PT
1 2 3 4 5 6 7 8 9
Abstract
13
The study aims to investigate the influence of tip clearance on the scour rate of
14
pile-supported horizontal axis tidal current turbine (TCT) and also attempts to
15
correlate time-dependent scour depth of TCT with the tip clearance. A physical model
16
of TCT was placed in a flume for scour test and the scour rate of the fabricated model
17
was investigated. The results suggest that the decrease in tip clearance increases the
18
scour depth. In addition, the shortest tip clearance results in the fastest and most
19
sediment transport. The maximum scour depth reached 18.5% of rotor diameter.
20
Results indicate that regions susceptible to scour typically persist up to 1.0Dt
21
downstream and up to 0.5Dt to either side of the turbine support centre. The majority
22
of the scour occurred in the first 3.5 hr. The maximum scour depth reaches
23
equilibrium after 24 hr test. An empirical formula to predict the time-dependent scour
24
depth of pile-supported TCT is proposed.
26
M AN U
TE D
EP
AC C
25
SC
10 11 12
Keywords: tidal energy; tidal-current turbine; wake; scour
27 28 29
*Corresponding author Email:[email protected]; Tel: +6017-3106739. 1
ACCEPTED MANUSCRIPT 1.0 Introduction
32
The depletion of fossil fuel is a major problem for mankind. Alternative energy is
33
urgently needed to ensure the continuous development of the world economy (Wu et
34
al., 2013). The ocean has many untapped natural resources, which can be harnessed as
35
renewable energy. Tidal-current power is more easy to predict and quantify than other
36
renewable resources (Watchorn et al., 2000). The predictable tidal cycles are
37
beneficial for control of electrical grid as the generation can be forecast in advance
38
(Clarke et al., 2006). Researchers started intensive studies in the past decade with the
39
deployment of turbines in several locations (OES Annual Report, 2011, 2012, 2013).
40
Tidal-current is a potential option for future energy supplies (Rourker et al., 2010).
M AN U
SC
RI PT
30 31
41
The interaction between Tidal Current Turbine (TCT) with ambient environment is of
43
interest to many researchers. Installation of a TCT accelerates the local flow leading
44
to a change in the surrounding environment (Xia et al., 2010; Shields et al., 2011;
45
Copping et al., 2014). Collision risk, acoustic emission, sediment dynamics and
46
morphodynamics of such device have long been identified (Neill et al., 2012).
47
Continuous assessment of environmental impacts of such a device is acting as a
48
barrier for obtaining permission from relevant authorities. (Hill et al., 2014). Marine
49
Current Turbine Ltd. (MCT) and Verdant Power have had to spend multi-million
50
dollars to monitor environment impacts (Neill et al., 2009). Affordable and
51
environmental friendly tidal energy could strengthen the confidence of interested
52
parties to install tidal turbines in potential sites.
AC C
EP
TE D
42
2
ACCEPTED MANUSCRIPT Neill et al. (2009) developed large grid cell (km-scale) simulations to explore the
54
impacts of array turbine farm on sediment dynamics. It claimed that small amount of
55
energy extracted from a site might affect the erosion and deposition pattern over a long
56
distance from the point of energy extraction. The effects may be present up to 50 km in
57
the case of the Bristol Channel. Vybulkova (2013) modified vorticity transport model
58
to simulate wake and its interactions with local sediment. The results show that the
59
flows downstream of the rotor affect the marine environment and the area of the seabed
60
affected by the horizontal axis tidal turbine is within 1.0 Dt downstream of rotor,
61
where Dt is the turbine diameter. The horizontal axis turbine has a diameter of 10 m in
62
Vybulkova (2013). Hill et al. (2014) conducted a physical modelling on the scour of an
63
axial-flow hydrokinetic turbine under clear water and live-bed conditions. Results
64
indicated that the rotor of turbine increases the local shear stress of sediments around
65
the turbine. The velocity deficit in the wake region leads to the flow acceleration below
66
the rotor. The local scour of the turbine is accelerated and expanded when compared to
67
bridge scour mechanism.
SC
M AN U
TE D
68
RI PT
53
Hong et al. (2013) studied the development of a scour hole with non-cohesive
70
sediments due to the jet induced by a rotating propeller, which focused on the
71
influences of various parameters on the time-dependent maximum scour depth. Hong
72
et al. (2013) claimed that the densimetric Froude number Fo and offset height ratio
73
yo/Dp play the most important role in affecting the scour depth, where yo is offset
74
height (tip clearance), Dp is the diameter of ship propeller. The prediction model
75
proposed by Hong et al. (2013) can monitor the scour depth for a long time. The
76
documentation of ship propeller jets induced scour offers an excellent example of the
77
investigation of TCT induced scour (Hamill, 1987; Hong et al., 2013). The presence
AC C
EP
69
3
ACCEPTED MANUSCRIPT 78
of rotor also adds complexity to the scour process of TCT.
79
No equation is available to predict the scour depth of TCT to date. Chen and Lam
81
(2014) review the equations used to predict the scour depth around the foundation of
82
bridge pier/pile and they suggested that the inclusion of the rotor effects into the scour
83
prediction. Zhang et al. (2015) developed a numerical model based on commercial
84
CFD package FLOW-3D to simulate the current induced scour around the support of
85
tidal turbine.
SC
RI PT
80
86
In addition, the scour process took a long time to reach its equilibrium stage (Melville
88
and Chiew, 1999). Liu (2008) stated that coupled simulations of flow with sediment
89
transport usually take long time. Besides, Harrison et al. (2010) also claimed that a
90
full model of the rotor stands in need of sufficient mesh resolution at the blade surface
91
towards capturing the boundary layer and separation. This necessitates a very large
92
number of mesh elements and computational efforts. Modelling a rotating turbine also
93
requires that the model be unsteady and the blades change position with every time
94
step. It is assumed that the incorporation of the rotor into scour model further
95
increases the computational power. It is more feasible to conduct experimental study
96
at the current stage to understand the scour nature of TCT first.
TE D
EP
AC C
97
M AN U
87
98
The study aims to identify the effects of the tip clearance on the scour rate of TCT. The
99
study documents the temporal scour profiles of TCT. It also attempts to correlate the
100
tip clearance to the time-dependent maximum scour depth. The outcomes of the study
101
provide a fundamental understanding on the scour nature of TCT. It also gives deep
102
insights into the environmental monitoring and impact of TCT. 4
ACCEPTED MANUSCRIPT 2.0 Experiment Set-up
104
A series of clear water scour tests have been performed in the Hydraulic Laboratory of
105
University of Malaya, Malaysia. The flume is 16 m long and 1 m wide, built with
106
transparent glass wall for flow visualization (See Fig.1). The turbine is positioned 5 m
107
downstream of the channel inlet (x=0 m). The incoming flow was conditioned using a
108
porous mattress to break the turbulence structures created from the pump and supply
109
pipes, therefore offering uniform flow before entering the test section. The velocity of
110
the incoming flow is monitored by a 2D mini LDV system. The turbulence intensity
111
of the flow is approximately 2%. The turbine support tower and nacelle were made
112
from PVC pipe. The turbine rotor was made through rapid prototyping using ABS
113
material. The turbine was outfitted with a miniature DC motor for voltage
114
measurements. Reasonable tip speed ratios were maintained by the internal resistance
115
of the device. The flow depth was controlled by the tailgate, while the intake of water
116
was controlled by three pumps.
AC C
EP
TE D
M AN U
SC
RI PT
103
117 118
Fig.1: Tilting flume at Hydraulic Laboratory, University of Malaya 5
ACCEPTED MANUSCRIPT The turbine model used during these experiments was a three bladed axis flow turbine
120
with rotor diameter Dt = 0.2 m. The diameter of the turbine support tower was 0.02 m.
121
A sediment recess 0.1 m deep, 5.0 m long and 1.0 m wide was constructed inside the
122
flume. Before commencement of each experiment, the sediment was first levelled
123
using a sand leveller. The turbine position and test section were selected to be within a
124
domain where the flow and the underlying bed topography were sufficiently far from
125
inlet and outlet boundary conditions. During the tests, the TCT was embedded in the
126
middle of the bed sediment recess. The flume was slowly filled with water from both
127
upstream and downstream pipes at a low rate. This prevents the disturbance of the
128
levelled sand bed. Once the predetermined water depth was reached, the pump was on
129
and the experimental run was started by adjusting the flow rate at the inlet. Two steel
130
plates (slope in Figure 2) were used to fix the sand bed at both the beginning and end
131
of the test section. This set-up was to ensure a smooth flow of water within the test
132
section and did not disturb the sediment. The slope gradient was determined to be 1/4.
133
There is no sediment transport observed in the vicinity of the slopes during the
134
experiment.
AC C
EP
TE D
M AN U
SC
RI PT
119
135 136
Fig.2: Schematic view of scour experiment
6
ACCEPTED MANUSCRIPT The flow depth 0.45 m was maintained by adjusting the tailgate at the end of the
138
flume. The approaching velocity 0.23 m/s were also maintained constantly during the
139
experiments. At 1:70th geometric scaling the experiments in the scour investigation
140
represents a 14 m diameter turbine in water depth of 31.5 m and applying Froude
141
similitude, a mean velocity of 1.92 m/s. The experimental scaling issue of tidal
142
turbine has been highlighted by Harrison et al. (2010) and Myers and Bahaj (2010).
143
High Froude numbers often take place in laboratory that may lead to alteration of
144
water surface. Froude similarity is normally maintained between model system and
145
prototype. Myers and Bahaj (2010) also stated that discrepancy in Reynolds numbers
146
between prototype and model is acceptable for the scaling of hydraulic channels due
147
to two reasons: 1) Froude similarity is maintained; 2) both full-scale and model
148
Reynolds numbers are within the same turbulent region. The experiment condition at
149
Hydraulic Laboratory, University of Malaya are well within the turbulent region.
150
However, the blockage effect may take place in the experiment as the study presented
151
by Schluntz and Willden (2015) and Koh and Ng (2017). The blockage may cause the
152
turbine experience higher velocity.
SC
M AN U
TE D
EP
153
RI PT
137
The water was carefully drained off from the flume at the end of each test to prohibit
155
the disturbance to the scour profiles. The scour profiles around the pile-supported
156
TCT are measured by a laser distance meter with an accuracy of 0.1 mm. The laser
157
distance meter measured the bed elevation for each 1 cm in length and 1 cm in width.
158
The detailed results of the experiments are presented in Table 1. The results and
159
discussion in Section 3 highlight the spatial and temporal evolution of localized scour
160
under clear water conditions.
AC C
154
161 7
ACCEPTED MANUSCRIPT The geometric standard deviation was calculated from = (d84/d16)0.5 = 1.3 which is
163
smaller than 1.4. It is indicating that the sand was uniformly graded (Barkdoll et al.
164
2007). The flow depth was also set to ensure that the incoming flow velocity at the
165
approach section (uf) was below the critical incipient velocity (ucr) (uf
166
for the sand bed-material of d50=1.0 mm) based on the Shield curve (Wu and Wang,
167
1999). Fig.2 shows a schematic diagram of the test section, turbine and
168
instrumentation used during experiments. Fig.3 illustrates the photo of turbine model
169
and flume.
M AN U
SC
RI PT
162
EP
172
Fig.3: Photo of turbine model (Dt = 0.2 m) and flume
AC C
171
TE D
170
8
ACCEPTED MANUSCRIPT 173
Table 1: Turbine, flow flume and sediment parameter used during the experiment Variable
Clear Water
0.2
Hydrofoil
NACA 63418
Hub length (m)
0.1
C/Dt
0.25, 0.50, 0.75
TSR
3.67
SC
Dt (m)
RI PT
Turbine parameters
M AN U
Flow, flume and sediment parameters H (m)
0.45
Qw (m3/s)
0.117
b (m)
1.0
δ (%)
RT V∞ (m/s)
AC C
ucr (m/s) d50 (mm) 174
TE D
Rb
EP
Fr
7
0.11 1.04 x 105 4.6 x 104 0.23 0.032 1.0
9
ACCEPTED MANUSCRIPT 3.0 Results and Discussion
176
3.1 Spatial Scour Profile
177
The final bed topography from the clear water scour tests is shown in Fig.4. It aims to
178
investigate the effect of rotor tip clearance on the sediment transport around the
179
support of TCT. During the test, no sediment transport occurred upstream of the TCT,
180
confirming clear water scour conditions and local scour and deposition induced by the
181
TCT. The scour induced by TCT of the three cases has similar pattern. Sediment
182
around the support is eroded and a hole is formed. Sediment eroded from the scour
183
region was deposited behind in a hill shaped dune. The maximum scour depth reached
184
18.5% of rotor diameter (5 cm tip clearance), Dt, or 7% of the flow depth, h. The
185
details of cross section scour profile can be found at Section 3.2. The results show that
186
the scour depth increases with the decrease of tip clearance. This is in line with Zhang
187
et al. (2015). The rotor clearance in Zhang et al. (2015) is lowered from 5.2 m to 1.2
188
m, the maximum depth of scour varies from 1.31 (8% of Dt) to 1.50 m (9.4% of Dt).
189
Results also indicate that regions susceptible to scour typically persist up to 0.5Dt
190
downstream and up to 0.5Dt to either side of the turbine support centre. The peak of
191
the deposition reached elevation of 10% of rotor diameter.
AC C
EP
TE D
M AN U
SC
RI PT
175
10
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
192
a. 5cm tip clearance
AC C
EP
TE D
193
194 195
b. 10cm tip clearance 11
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
196 197
c. 15cm tip clearance
198
Fig.4: 3D surface map of 24-hr scour by SURFER
TE D
199
As discovered in Chen and Lam (2014), the obstruction induced by the spinning rotor
201
results in acceleration of the flow between the bottom rotor tip and bed surface. This
202
acceleration increases the local shear stress and it amplifies the scour mechanism. It
203
therefore the short tip clearance (5 cm) can cause severe sediment transport around
204
the support of TCT. The extent of scour hole increases with the decrease of tip
205
clearance. The highest tip clearance 15 cm results in the least eroded sand from the
206
bed. The scour holes and deposited sand are asymmetric especially in the case of 5 cm
207
tip clearance. This is due to asymmetric shear stress impacted at the bed. This also
208
conforms to the study conducted by Zhang et al. (2015).
AC C
EP
200
209 12
ACCEPTED MANUSCRIPT 3.2 Temporal Scour Profiles
211
Fig. 5, 6 and 7 present the temporal dimensionless scour holes profile at 3.5, 7.0 and
212
24 hr. The turbine rotor places 5 cm, 10 cm and 15 cm from bed, respectively. The
213
investigated cases are in the same manner in terms of scour development. Looking
214
closely at the measured developments of scour profile at various lateral positions, it
215
can be observed that a scour hole is formed around the support tower. The majority of
216
scour occurred at the initial 3.5 hr. The size of scour holes increase with time. The
217
scour depth slightly increases after 3.5 hr scour. A deposition mound is also observed
218
downstream at the initial stage. The depositional wedge also moves further
219
downstream as the test in progress. A secondary sand dune is formed right behind the
220
primary sand dune after 24 hr test.
AC C
EP
TE D
M AN U
SC
RI PT
210
13
ACCEPTED MANUSCRIPT In the case of 5 cm tip clearance, maximum scour occurred at approximately x/Dt=
222
-0.1, y/Dt= -0.1, reached the depth of S/Dt=0.185. The highest depositional wedge
223
reached elevations of approximately 10% of the Dt at location x/Dt=1.0, y/Dt=0.1. The
224
highest secondary sand dune is only 3.5% of Dt at location x/Dt=1.8, y/Dt= 0.2. It is
225
observed that there is no secondary sand dune at the first 7 hr scour test. It is probably
226
due to the primary deposition wedge reaching a certain height where it becomes
227
susceptible to the flow below turbine rotor. The accelerated flow below rotor turbine
228
eroded the deposited sand dune.
M AN U
SC
RI PT
221
229
In the case of 10 cm tip clearance, maximum scour depth is approximately 16% of Dt
231
and took place at x/Dt= -0.05, y/Dt=0 and x/Dt=-0.05, y/Dt= -0.1. The highest primary
232
sand dune has an elevation of 0.075Dt and it happened after 7 hr scour test other than
233
24 hr. This may be due to the primary scour dune deposition is less than its erosion in
234
the subsequent test. The highest primary deposition located at the region of
235
x/Dt=0.6-0.7, y/Dt=0.0-0.1. Secondary dune is also formed but with a lower elevation.
236
The highest secondary sand dune is only 3% of Dt at x/Dt=1.5, y/D=0.0.
EP
AC C
237
TE D
230
238
For the investigated case of 15 cm tip clearance, maximum scour depth occurred at
239
x/Dt= -0.05, y/Dt=0 and x/Dt=-0.05, y/Dt= -0.1 which is same as the case of 10 cm tip
240
clearance. The maximum scour depth is approximate 0.145Dt. The highest primary
241
deposition is 8.5% of Dt located at x/Dt= 0.7, y/Dt=0 and x/Dt=0.7, y/Dt= -0.1. The
242
secondary sand dune is getting lower and lower as the rotor of turbine rotor moves up. 14
ACCEPTED MANUSCRIPT 243
The highest secondary sand dune is only 2% of turbine diameter at locations x/Dt= 1.2,
244
y/Dt=0 and x/Dt=1.2, y/Dt= -0.1.
245
As seen from all the investigated cases, it has been observed that the maximum scour
247
depth increase as the decrease of tip clearance. The maximum scour depth generally
248
occurred in front of the support pile. It is generally at the region x/Dt=-0.05 – -0.1 and
249
y/Dt=0.0 - -0.1. The height of primary and secondary deposited sand dunes also
250
increases as the decreases of tip clearance. The location of peak sand dune is a bit
251
complicated due to the influence of the flow between turbine rotor and bed.
M AN U
SC
RI PT
246
252
Furthermore, the transition point between net erosion and net deposition moves
254
downstream as the test progresses. The furthest transition point located at x/Dt=0.5 in
255
all the investigated cases after 24 hr scour test. The transition points of Hill et al.
256
(2014) are between x/Dt=1.0 to x/Dt =2.0. This region has been noted by Chamorro et
257
al. (2013) and they claimed that the turbine wake coherence starts to break down in
258
this region where the most rapid velocity deficit recovery begins. However, the
259
furthest transition point of the current study located at x/Dt=0.5 in all the cases after
260
24 hr scour test. The difference of the locations of the transition point may be due to
261
the difference of turbine geometry and flow velocity.
AC C
EP
TE D
253
15
ACCEPTED MANUSCRIPT 0.10
0.05
-0.05
RI PT
S/Dt
0.00
-0.10
3.5hr 7.0hr 24hr
-0.20 -1.0
-0.5
0.0
1.0 x/Dt
1.5
2.0
2.5
M AN U
262 263
a.
0.10 0.05
S/Dt
TE D
0.00 -0.05
-0.15
AC C
-0.20
3.5hr
EP
-0.10
-1.0
264 265 266
0.5
SC
-0.15
-0.5
7.0hr 24hr 0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
b.
16
ACCEPTED MANUSCRIPT 0.10 0.05
S/Dt
0.00
RI PT
-0.05 -0.10
3.5hr 7.0hr
-0.15 -0.20 -1.0
-0.5
0.0
0.5
1.0 x/Dt
1.5
2.0
2.5
M AN U
267
SC
24hr
268
c.
0.10
0.05
TE D
S/Dt
0.00
-0.05
3.5hr
EP
-0.10
7.0hr
AC C
-0.15
24hr
-0.20
-1.0
269 270 271
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
d.
17
ACCEPTED MANUSCRIPT 0.10 0.05
-0.05
RI PT
S/Dt
0.00
-0.10
3.5hr 7.0hr
-0.15 -0.20 -1.0
-0.5
0.0
0.5
1.0
273 274 275 276
e.
1.5
2.0
2.5
M AN U
272
x/Dt
SC
24hr
Fig.5: Temporal dimensionless scour holes profile at different times (Tip clearance=5cm); a. y/Dt = 0; b. y/Dt= -0.1; c. y/Dt = 0.1. e. y/Dt = -0.2; f. y/Dt = 0.2.
277 278
TE D
0.10 0.05
EP
-0.05
3.5hr
AC C
S/Dt
0.00
-0.10
7.0hr
-0.15
24hr
-0.20
-1.0
-0.5
0.0
0.5
1.0
279
x/Dt
280 281
a.
1.5
2.0
2.5
18
ACCEPTED MANUSCRIPT 0.10 0.05
S/Dt
0.00
RI PT
-0.05 -0.10
3.5hr
-0.15
7.0hr 24hr
-1.0
-0.5
0.0
0.5
0.05
-0.05
TE D
S/Dt
0.00
-0.10
EP
-0.15 -0.20
AC C
-1.0
285 286
1.5
2.0
2.5
M AN U
b. 0.10
284
1.0 x/Dt
282 283
SC
-0.20
-0.5
0.0
3.5hr 7.0hr 24hr
0.5
1.0
1.5
2.0
2.5
x/Dt
c.
19
ACCEPTED MANUSCRIPT 0.10
0.05
RI PT
S/Dt
0.00
-0.05
3.5 hr
-0.10
7.0 hr
-0.15 -1.0
-0.5
0.0
0.5
287 288
0.05
TE D
S/Dt
0.00
-0.05
EP
-0.10
-0.15
1.5
2.0
2.5
M AN U
d. 0.10
-0.5
0.0
3.5 hr 7.0hr 24 hr
0.5
1.0
1.5
2.0
2.5
x/Dt
AC C
-1.0
289 290 291 292 293
1.0 x/Dt
SC
24 hr
e. Fig.6: Temporal dimensionless scour holes profile at different times (Tip clearance=10 cm); a. y/Dt = 0; b. y/Dt = -0.1; c. y/Dt = 0.1. e. y/Dt = -0.2; f. y/Dt = 0.2.
20
ACCEPTED MANUSCRIPT 0.10
0.05
-0.05
RI PT
S/Dt
0.00
-0.10
3.5hr
-0.15
7.0hr
-0.20 -1.0
-0.5
294 295
0.0
0.5
x/Dt
0.10 0.05
TE D
0.00 S/Dt
1.5
2.0
2.5
M AN U
a.
1.0
SC
24hr
-0.05 -0.10
EP
-0.15
3.5hr 7.0hr 24hr
-0.20
297 298
AC C
-1.0
296
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
b.
21
ACCEPTED MANUSCRIPT 0.10
0.05
RI PT
S/Dt
0.00
-0.05
3.5hr
-0.10
7.0hr 24hr
SC
-0.15
-0.20 -0.5
0.0
0.5
1.0
1.5
2.0
2.5
M AN U
-1.0
x/Dt
299 300
c.
0.10
TE D
0.05
S/Dt
0.00
3.5hr
EP
-0.05
AC C
-0.10
7hr 24hr
-0.15
-1.0
301 302
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
x/Dt
d.
22
ACCEPTED MANUSCRIPT 0.10
0.05
RI PT
S/Dt
0.00
-0.05
3.5hr
-0.10
7hr
-0.15 -1.0
-0.5
0.0
0.5 x/Dt
303
1.5
2.0
2.5
EP
TE D
M AN U
e. Fig.7: Temporal dimensionless scour holes profile at different times (Tip clearance=15cm); a. y/Dt = 0; b. y/Dt = -0.1; c. y/Dt = 0.1. e. y/Dt = -0.2; f. y/Dt = 0.2.
AC C
304 305 306
1.0
SC
24 hr
23
ACCEPTED MANUSCRIPT 3.3 Temporal Variation of Maximum Scour Depth
308
Fig.8 illustrates the temporal scour depth under different tip clearances. It has been
309
observed that the majority of scour of all three cases occurred in the first 3.5 hours.
310
After 7 hr of scour test, all three cases reach near-equilibrium conditions. The depth of
311
scour increased more quickly in the case with 5 cm tip clearance, while the slowest
312
scour occurred in the case of 15 cm tip clearance. The maximum scour also occurred
313
in the case of 5 cm tip clearance, which is approximately 18.5% of rotor diameter Dt
314
(8% of flow depth). The least scour occurred in the case of 15 cm tip clearance, which
315
is approximately 14% of rotor diameter Dt (6% of flow depth).
M AN U
SC
RI PT
307
0.20 0.18 0.16
TE D
0.14
St/Dt
0.12 0.10
0.06
AC C
0.04
EP
0.08
Clearance 5 cm Clearance 10cm Clearance 15cm
0.02 0.00
0
316 317
200
400
600 800 t (mins)
1000
1200
1400
1600
Fig.8: Time-dependent maximum scour depth
24
ACCEPTED MANUSCRIPT 318
Research to provide methods for predicting scour depth around the pile-supported
319
structure of tidal turbine is limited. The maximum scour depth at any time t, St, is
320
assumed to be a function of 10 independent variables as follows:
322
= ( , , , , , , , , , )
323
RI PT
321
Equation 1
In the works of ship propeller jet induced scour (Rajaratnam, 1981; Hamill, 1999;
325
Hong et al, 2013), the fluid viscosity could be neglected as the Reynolds number of
326
the jet is greater than 10,000. The Reynolds number of the wake induced by the
327
physical model of the turbine is around 29,100. The viscosity term is ignored in the
328
following analysis. The water depth of the study is reasonable high, the Froude
329
number is less than 0.1 and its effects on scour are not examined in the study. The
330
efflux velocity is calculated from
331
(2015). By referencing the work of Hong et al. (2013), the Buckingham π theorem
332
has been applied and choosing , , as fundamental variables. Equation (2) has
333
been obtained as the follows:
335 336
M AN U
TE D
= 1 − , as introduced by Lam et al.
EP
AC C
334
SC
324
" !
= # (
$ %&! '! " !
,
" !
,
" !
, ((
(
) *()
, ) +
Equation 2
337
Based on experience of ship propeller jet induced scour and re-arrange the above
338
equation, the maximum depth of scour hole could be written as:
339 25
ACCEPTED MANUSCRIPT
340
$
= # (,- ,
'
) $ %&! $ ⁄ +
,
'
,
Equation 3
341 () *(
) and ⁄ - , which is denoted by tR, is the
342
In Equation (3), Fo= - 0/(
343
proposed time scale. A nonlinear regression analysis has been performed and an
344
empirical equation for the estimation of time-dependent scour depth of pile-supported
345
TCT is proposed as:
RI PT
(
347
where
349
Equation 4
'
'
1 = 0.01 ($ )* .< (% ) . = , .<<>
350
&!
.@A . = * . = ) ( ) ,
TE D
351
1# = 1.9 (
352
= 1 [34 5 $+ 6 − 1# ]89
M AN U
348
$
SC
346
EP
1> = 2.18 (
. = * . < * . ) ( ) ,
Equation 4 is in the same form of equation proposed by Hong et al. (2013), which is
354
developed to predict the scour induced by ship propeller jets. Fig. 9 presents the
355
overall comparison between the computed and observed temporal maximum scour
356
depth. The R-squared value ranges from 0.7747 to 0.9737. The best fitting of the
357
experimental data with predicted value is in the case C=10 cm. The lowest R-squared
358
value in the case of C=15 cm is 0.7747 which is still consider very good based on
359
Henriksen et al. (2003). Equation 4 has good agreement with present experimental
AC C
353
26
ACCEPTED MANUSCRIPT 360
data under the conditions of 0.25 ≤ C⁄ ≤ 0.75 , 50 ≤ C⁄ ≤ 150 and
361
, = 1.8965.
RI PT
0.20
SC
0.10
M AN U
Predicted St/Dt
0.15
C=5cm, R2=0.9692
0.05
C=10cm, R2=0.9737 C=15cm, R2=0.7747
0.00 0.00
0.10
0.15
0.20
Measured St/Dt
TE D
362
0.05
Fig.9: Comparison between measured and predicted maximum time-dependent scour
364
depth
AC C
EP
363
27
ACCEPTED MANUSCRIPT 3.4 Comparison of Scour Evolution
366
A comparison between the authors’ experimental results and Hill et al. (2014) of
367
temporal scour depth is presented in Fig. 10. The time has been normalized by the
368
respective velocity at hub height. It shows that the large scale experiment did not
369
approach
370
near-equilibrium. Hill et al. (2014) also indicated that the sour hole evolves faster
371
when the rotor is on the upstream side of the support tower. The temporal evolution of
372
scour depth of small-scale experiments indicated similar pattern with the author’s
373
results (C=5 cm), particularly the case of rotor upstream. The authors’ experiments are
374
conducted with rotor upstream. The results of Hill et al. (2014) in Fig.10 are taken
375
from x/Dt=0.66 and y/Dt=0 for all cases. The authors’ results are taken from x/Dt=-0.1
376
and y/Dt=-0.1. These locations are on the downstream side of the rotors. With the
377
rotor on the downstream of the support tower, the flow acceleration between rotor and
378
seabed may take place later. The scour amplification occurs potentially further
379
downstream from the rotor.
380
The temporal evolution of scour hole for large-scale experiment is relatively faster
381
compare to small-scale experiments. It should be noted that the velocity at hub height
382
is 0.66 m/s. The tidal turbine employed in the study of Zhang et al. (2015) is a full
383
scale turbine with a 16 m diameter rotor, and the scour is quickly formed in the
384
vicinity of support structure. It only takes 30 mins the scour depth reached
385
approximately 1.31 m with current speed of 3.7 m/s. There are five different types of
386
components of gravel formed in the seabed of Zhang et al. (2015). The numerical
stage.
The
small
scale
experiments
approach
the
AC C
EP
TE D
M AN U
SC
equilibrium
RI PT
365
28
ACCEPTED MANUSCRIPT 387
setup of Zhang et al. (2015) is not identical with the current study. Therefore, the
388
influence of rotor size and sediment formation on the tidal turbine induced scour
389
cannot be seen from such comparison. Small scale: rotor downstream (Hill et al., 2014) Small scale: rotor upstream (Hill et al., 2014) Large scale: rotor downstream (Hill et al., 2014) C=5.0cm C=10cm C=15cm
0.30
0.25
SC
0.15
0.10
0.05
0.00
391 392
1.0
1.5 tVhub/Dt
2.0
2.5
3.0 x104
Fig.10: Comparison of temporal scour depth
EP
390
0.5
TE D
0.0
M AN U
St/Dt
0.20
RI PT
0.35
3. 5 Comparison with Conventional Pile Scour
394
The presence of rotor adds complexity to the scour induced by the pile-supported TCT.
395
The scour mechanism of TCT is different from the conventional pile scour. The main
396
differences come from the generated wake. The flow below rotor experience
397
acceleration and it amplifies the sediment transport at seabed. In order to take into
398
account the effect of rotor on the scour of TCT, the proposed model in Section 3.3 is
399
based on the reference guide of ship propeller jets induced scour. The attempts by
400
referencing ship propeller jet induced scour predictor is promising and the model can
AC C
393
29
ACCEPTED MANUSCRIPT be used as a starting point for future model development.
402
There are many existing scour predictors as documented by Chen and Lam (2014). To
403
demonstrate the applicability of the existing models for the prediction of TCT induced
404
scour, two equations (Breuser et al., 1977；Richardson and Davis, 2001) are used to
405
calculate the scour depth for the experimental condition of the current study. These
406
two equations have been selected by Harris et al. (2010) to develop STEP to predict
407
the development of scour evolution of offshore structures. The two equations are
408
shown as the follows:
409
$
410
M AN U
SC
RI PT
401
P
= 1.5I IJ IK I% tanh( $ ) IK = 0 ,
411
412
TE D
RS IK = 2 V W − 1, RST IK = 1,
Q
Equation 5
RS < 0.5 RST
Q 0.5 ≪
Q
RS <1 RST
RS ≫1 RST 0.35
= 2.0IZ I[ I\ I I] ()
,_ 0.43
Equation 6
and
414
IK is the correction factor for bed condition, I% is the correction factor for size of
415
bed material, IJ is the correction factor for flow angle of attack, I is the correction
416
factor of pier shape, Uc is the depth-averaged current speed and Ucr is the threshold
417
depth-averaged current speed. Ia is the enhance correction factor for pier width/pile
418
diameter. Table 2 tabulates the results of the calculation. It shows that the use of
419
equations proposed by Breuser et al. (1977) and Richardson and Davis (2001) are not
420
applicable for the experiment of current study. These two equations are under- or
421
over- predicted for most of the current experimental cases. It is noticed that the scour
AC C
EP
413
30
ACCEPTED MANUSCRIPT depth in the case C=15 cm is very close to the predicted value by Breuser et al.
423
(1977). This may due to the rotor is further to the sand bed so that the amplification to
424
the bed stress is lesser compared to other cases. The equation proposed by Breuser et
425
al. (1977) could be used to predict the equilibrium scour depth around TCT when the
426
rotor is far away from the bed.
RI PT
422
Table 2: Factors used for scour depth prediction analysis Tip Clearance 5cm 10cm 15cm
Measured (cm) 3.5 3.1 2.9
Predicted by Breuser et al., Predicted by Richardson and Davis, 2001 (4.58 cm) 1977 (2.7 cm） 22.9%, 30.9%, 12.9% 47.7%, 6.9% 57.9%
M AN U
428
SC
427
4.0 Conclusion
430
The presence of rotor amplifies the scour process around the support of TCT. The
431
decrease of tip clearance increases the scour depth. The maximum scour depth
432
reached 18.5% of the rotor diameter. Results indicate that regions susceptible to scour
433
typically persist up to 1.0Dt downstream and up to 0.5Dt to either side of the turbine
434
support centre. The tip clearance plays the most important role in affecting the scour
435
depth and scour rate. The case of 5 cm tip clearance results in the fastest scour and
436
most sediment transport, while the case of 15 cm tip clearance has the slowest and
437
least sediment transport. Majority of the scour occurred in the first 3.5 hr in all the
438
cases. The scour depths of all the cases reach near-equilibrium state after 7 hr test.
439
Based on the experimental data, the / has been found to be a logarithmic
440
function of time. An empirical formula (Equation 4) has been proposed to estimate the
441
time-dependent maximum scour depth, where the efflux velocity = 1 − is
AC C
EP
TE D
429
31
ACCEPTED MANUSCRIPT an initial input for the proposed scour prediction. In future research works, it is
443
suggested to investigate the influence of turbine geometry, rotor size and sediment
444
formation on the scour issue of TCT.
445
Acknowledgement
446
The authors wish to extend their gratitude to Ministry of Higher Education Malaysia
447
and University of Malaya for the financial support received under the UM/MOHE
448
High
449
UM.C/HIR/MOHE/ENG/47. The authors also wish to thank Professor. Dr. Lam Wei
450
Haur. His efforts to initiate the research project are highly appreciated.
451
List of Symbols
452
yo
propeller tip clearance
453
C
turbine tip clearance
454
CT
thrust coefficient
455
Dt
diameter of the turbine disc
456
Dp
diameter of propeller
457
Vo
efflux velocity in m/s
458
V∞
free stream velocity in m/s
459
Vx,r
lateral distribution velocity in m/s
460
x
longitudinal (streamwise) direction
461
y
lateral (spanwise) direction
462
n
rotational speed in rev/s
463
r
radial distance
UM.C/HIR/MOHE/ENG/34
SC
Grant
and
M AN U
Research
AC C
EP
TE D
Impact
RI PT
442
32
ACCEPTED MANUSCRIPT R
radius of turbine rotor
465
ReD
Reynolds number
466
S
Scour Depth
467
St
Scour depth at time t
468
ucr
threshold depth averaged current velocity
469
Uc
depth averaged current velocity
470
Uuc
undisturbed current velocity
471
Rb
bulk Reynolds number (Rb=hV∞/ )
472
RT
turbine rotor Reynolds number
473
U
velocity at the edge of the bed boundary layer,
474
Qw
flow discharge
475
Kinematic viscosity of water
476
Fr
Froude number of incoming flow
477
Fo
densimetric Froude number
478
H
flow depth
479
hp
pile height
480
δ
481
d50
median sediment grain size
482
g
acceleration due to gravity
483
∆
difference between the mass density of the sediment and the fluid
484
density of fluid
485
a
density of water
AC C
EP
TE D
M AN U
SC
RI PT
464
blockage ratio
33
ACCEPTED MANUSCRIPT
density of sand
487
d
turbulent viscosity
488
geometric standard deviation of sediment particles
489
IK
correction factor for bed condition,
490
I%
correction factor for size of bed material,
491
IJ
correction factor for flow angle of attack,
492
I
correction factor of pier shape
493
Ia
enhance correction factor for pier width/pile diameter.
AC C
EP
TE D
M AN U
SC
RI PT
486
34
ACCEPTED MANUSCRIPT Reference Barkdoll, B. D., Ettema, R., & Melville, B. W. (2007). Countermeasures to protect bridge abutments from scour (Vol. 587). Transportation Research Board.
RI PT
Breusers, H. N. C., Nicollet, G., & Shen, H. W. (1977). Local scour around cylindrical piers. Journal of Hydraulic Research, 15(3), 211-252. Chamorro, L. P., Hill, C., Morton, S., Ellis, C., Arndt, R. E. A., & Sotiropoulos, F. (2013). On the interaction between a turbulent open channel flow and an axial-flow turbine. Journal of Fluid Mechanics, 716, 658-670.
SC
Chen, L., Lam, W.H. (2014) Methods for predicting seabed scour around marine current turbine. Renewable and Sustainable Energy Reviews, 29,683-692
M AN U
Clarke, J.A., Connor, G., Grant, A.D., Johnstone, C.M. (2006) Regulating the output characteristics of tidal current power stations to facilitate better base load matching over lunar cycle. Renewable Energy, 31 (2), 173-180. Copping, A., Battey, H., Brown-Saracino, J., Massaua, M., & Smith, C. (2014). An international assessment of the environmental effects of marine energy development. Ocean & Coastal Management, 99, 3-13.
TE D
Hamill, G. A. (1987). Characteristics of the screw wash of a manoeuvring ship and the resulting bed scour (Doctoral dissertation, Queen's University of Belfast).
EP
Harris, J., Whitehouse, R. J. S., & Benson, T. (2010). The time evolution of scour around offshore structures. ICE-Maritime Engineering, 163(1), 3-17. Harrison, M. E., Batten, W. M. J., Myers, L. E., & Bahaj, A. S. (2010). Comparison between CFD simulations and experiments for predicting the far wake of horizontal axis tidal turbines. IET Renewable Power Generation, 4(6), 613-627.
AC C
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537
Henriksen, H. J., Troldborg, L., Nyegaard, P., Sonnenborg, T. O., Refsgaard, J. C., & Madsen, B. (2003). Methodology for construction, calibration and validation of a national hydrological model for Denmark. Journal of Hydrology, 280(1), 52-71. Hill, C., Musa, M., Chamorro, L. P., Ellis, C., & Guala, M. (2014). Local scour around a model hydrokinetic turbine in an erodible channel. Journal of Hydraulic Engineering, 140(8), 04014037. Hong, J. H., Chiew, Y. M., & Cheng, N. S. (2013). Scour caused by a propeller jet. Journal of Hydraulic Engineering, 139(9), 1003-1012. 35
ACCEPTED MANUSCRIPT 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558
Koh, W. X. M., & Ng, E. Y. K. (2017). A CFD study on the performance of a tidal turbine under various flow and blockage conditions. Renewable Energy, 107, 124-137.
559 560 561 562
Neill, S. P., Jordan, J. R., & Couch, S. J. (2012). Impact of tidal energy converter (TEC) arrays on the dynamics of headland sand banks. Renewable Energy, 37(1), 387-397.
563 564 565 566 567 568 569 570 571
OES-IA Annual Report 2010 (2011). Implementing Agreement on Ocean Energy Systems. Lisbon, Portugal.
572 573 574
Rajaratnam, N. (1981). Erosion by plane turbulent jets. Journal of hydraulic Research, 19(4), 339-358.
575
Richardson, E.V., & Davis, S.R. (2001). Evaluating scour at bridges (4th ed.). Washington, D.C.: Hydraulic Engineering Circular No.18 (HEC-18), Federal Highway Administration.
576 577 578 579 580 581
Lam W.H., Chen, L., Hashim, R. (2015) Analytical wake model of tidal current turbine. Energy, 512-521.
RI PT
Liu, X. (2008). Numerical models for scour and liquefaction around object under currents and waves. University of Illinois at Urbana-Champaign.
SC
Myers, L. E., & Bahaj, A. S. (2010). Experimental analysis of the flow field around horizontal axis tidal turbines by use of scale mesh disk rotor simulators. Ocean Engineering, 37(2), 218-227.
M AN U
Melville, B. W., & Chiew, Y. M. (1999). Time scale for local scour at bridge piers. Journal of Hydraulic Engineering, 125(1), 59-65.
TE D
Neill, S. P., Litt, E. J., Couch, S. J., & Davies, A. G. (2009). The impact of tidal stream turbines on large-scale sediment dynamics. Renewable Energy, 34(12), 2803-2812.
EP
OES-IA Annual Report 2011 (2012). Implementing Agreement on Ocean Energy System. Lisbon, Portugal.
AC C
OES-IA Annual Report 2012 (2013). Implementing Agreement on Ocean Energy System. Lisbon, Portugal.
Rourke, F.O., Boyle, F., & Reynolds, A. (2010). Marine current energy devices: Current status and possible future applications in Ireland. Renewable and Sustainable Energy Reviews, 14(3), 1026-36. 36
ACCEPTED MANUSCRIPT 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599
Schluntz, J., & Willden, R. H. J. (2015). The effect of blockage on tidal turbine rotor design and performance. Renewable Energy, 81, 432-441.
600 601 602 603 604 605 606 607
Vybulkova, L. (2013). A study of the wake of an isolated tidal turbine with application to its effects on local sediment transport (Doctoral dissertation, University of Glasgow).
608 609 610 611
Zhang, J., Gao, P., Zheng, J., Wu, X., Peng, Y., & Zhang, T. (2015). Current-induced seabed scour around a pile-supported horizontal-axis tidal stream turbine. Journal of Marine Science and Technology, 23(6), 929-936.
RI PT
Shields, M.A., Woolf, D.K., Grist, E.P.M., Kerr, S.A., Jackson, A.C., Harris R.E., … Jonathan, S. (2011). Marine renewable energy: the ecological implications of altering the hydrodynamics of the marine environment. Ocean and Coastal Management, 54, 2-9. Watchorn, M., Trapp, T., Sayigh, A.A.M. (2000). Tidal stream renewable offshore power generation. World Renewable Energy Congress VI. Oxford: Pergamon. 2664-667.
SC
Wu, W., & Wang, S. S. (1999). Movable bed roughness in alluvial rivers. Journal of Hydraulic Engineering, 125(12), 1309-1312.
M AN U
Wu, B., Zhang, X., Chen, J., Xu, M., Li, S., & Li, G. (2013). Design of high-efficient and universally applicable blades of tidal stream turbine. Energy, 60:187-94.
AC C
EP
TE D
Xia, J.Q., Falconer, R.A., & Lin, B. (2010). Impact of different tidal renewable energy projects on the hydrodynamic processes in the Severn Estuary, UK. Ocean Modelling, 32, 86-104.
37
ACCEPTED MANUSCRIPT Highlights:
AC C
EP
TE D
M AN U
SC
RI PT
The influence of tip clearance on the scour profile of pile-supported tidal turbine is investigated The scour profile of tidal turbine is presented A correlation between tip clearance and time-dependent scour rate of tidal turbine is proposed