INNOVATIVE DESIGN FOR COOLING SYSTEM OF OFF-SHORE
WIND TURBINES
Arturo de Risi, Marco Milanese, Gianpiero Colangelo, Domenico Laforgia
University of Salento, Department of Engineering for Innovation, Via per Arnesano, 73100 Lecce (Italy)
ABSTRACT
In order to reduce the entry of moisture, salt, sand and other external contaminations into the nacelle and
also to reduce the fan noise which reaches the exterior, in this work a study of an innovative cooling system
for off-shore wind turbine has been carried out. The new cooling technique is based on the use of nanofluids
(engineered colloidal suspensions of nanoparticles in a base fluid). Nanofluids allow to increase the thermal
conductivity of fluids and so to reduce the heat exchange surface and the heat transfer fluid flow rate due to
the increased heat capacity. To reduce the amount of nanofluids circulating in the cooling system, the
performance of a two-stage cooling circuit has been investigated. The first circuit takes the heat out of the
generator and of the accessories whereas the second circuit, coupled with the first via an heat exchanger,
dissipates the heat into the ambient. For the second circuit two options have been investigated. In the first
solution the waste heat is dispersed using the tower as dissipator whereas in the second option the waste heat
is exchanged with a titanium heat exchanger using marine water as heat transfer fluid.
Both solutions assure high efficiency of heat exchange, long technical life expectancy and limited
maintenance requirements.
INTRODUCTION
During operation wind turbine generators produce
large amounts of heat and unless it is dissipated, the
generators are unable to operate at maximum
efficiency. Prolonged excessive heat can also lead
to a reduction in the life of the generator. Most
turbines use air as the cooling medium by
encapsulating the generator in a duct where the air
is forced to flow by a large fan. However, some
manufacturers use water cooled generators. Water
cooled generators may be built more compactly,
which also gives some electrical efficiency
advantages over air-cooled ones including high
heat, less drag/friction loss than air-cooled
generators. Windage can account for as much as 3040 percent loss in the efficiency of a generator.
Unless electric generators have adequate cooling,
catastrophic failures and very expensive repair costs
can result specially for off-shore wind power plants.
Improving the operating efficiency and reducing
the maintenance costs of generators is critical in
today's competitive market.
In addition also the power transmission system
needs to be cooled. One of the most used solution to
provide cooling for such a system in the nacelle of a
wind turbine is to establish an air intake on the
upwind side of the nacelle and directing the flow of
air from the intake around the parts of the power
transmission system. Alternatively, one or more parts
of the power transmission system may have a liquid
cooling system comprising a heat exchanger that is
cooled by the flow of air from the air intake.
Such a cooling system has certain disadvantages.
The air intake may be blocked by birds or by ice
formed from rainwater, and for that reason the air
intake may be provided with means for preventing
such blockage, such as a fan for providing a cleaning
counter flow of air or heating means for melting ice.
It is furthermore becoming increasingly difficult to
provide sufficient cooling capacity as the power
output of the new generations of wind turbines
increases. Insufficient cooling may cause mechanical
breakdown of parts of the power transmission system
and lowers the efficiency of the parts.
On the other hand a more efficient cooling system
may provide for some of the parts of the power
transmission system to be constructed more compact,
in particular the electronic parts. The over all heat
loss is typically of the order of 4-8% of the power
output of the wind turbine, the loss in the gearbox
and in the generator being of approximately the same
magnitude, in particular the generator may
advantageously be cooled more efficiently in order to
prevent break-down of the generator and both the
rotor and the stator of the generator may be cooled.
Furthermore, the power control system and the
electrical transformer, e.g. comprising a frequency
converter, may also constitute a part of the power
transmission system and may also be cooled to
obtain better efficiency.
A possible solution for increasing the
performance of water-cooled systems is to use
nanofluids.
Nanofluids are engineered colloidal suspensions
of nanoparticles (1-100 nm) in a base fluid.
Common base fluids include water, organic or metal
liquids. Nanoparticles are typically made of
chemically stable metals, metal oxides or carbon in
various forms. The size of the nanoparticles imparts
some unique characteristics to these fluids,
including greatly enhanced energy, momentum and
mass transfer, as well as reduced tendency for
sedimentation and erosion of the containing
surfaces. Solid particles have a higher conductive
heat transfer coefficient than liquids. Nanoparticles
are added because they stay suspended much longer
than larger particles and their surface area (referred
to the volume) is 1000 times larger than that of
microparticles. The smaller the particles’ size is the
higher their capacity of heat transfer enhancing.
Nanofluids are being investigated for numerous
applications, including cooling, manufacturing,
chemical and pharmaceutical processes, medical
treatments, cosmetics, etc.
The idea of increasing heat conductivity in fluids
by suspending high conductive particles in a liquid
was first proposed by Maxwell in 1881 in his “A
Treatise on Electricity and Magnetism” [1].
In the first studies millimeter or micrometer
particles sized were used, but, although revealed
some enhancement, their dimensions caused quick
sedimentations, abrasions and clogging [3, 4].
The first studies revealed an increase in thermal
conductivity of 20% for 4%vol CuO nanoparticles
with average diameter 35 nm dispersed in ethylene
glycol. A similar behavior has been observed with
Al2O3 nanoparticles and better results might be
obtained by using Cu nanoparticles or Carbon
Nanotube (CNT) [5, 6].
The reduced dimensions of the particles used in
nanofluids have many advantages against bigger
particles. Such enhancement depends on many
factors such as the particles shape, dimensions,
volume fractions and the materials thermal
properties. Each solid particle is surrounded with a
liquid layer having better conductive characteristics
than remaining base liquid [7]. Besides, there is a
nanoconvection
motion
phenomenon
near
nanoparticle [8].
Therefore, in the present investigation the
possibility to use nanofluids based cooling system
combined with innovative solutions for dissipating
the waste heat has been studied.
NANOFLUID CHARACTERIZATION
In order to acquire significant data on the thermal
properties of nanofluids mixtures of base fluid with
different commercially available nanoparticles have
been tested. The used nanoparticles are reported in
Table 1.
Table 1: Powders used in the nanofluids samples
Material
Codex
CuO
E
Effective
density
(kg/m3)
6.50
Al2O3
B
3.97
ZnO
M
5.60
Mean size
(nm)
19.29
(SMD)
22.91
(SMD)
70 (leq)
Thermal
Conductivity
(W/m K)
20.0
25.1
29.0
Liquid phases used in the nanofluids samples is
demineralised water with non ionic dispersant, to
stabilize the suspension.
The nanofluids thermal conductivity was measured
through an instrument based on hot-wire technology
which is a standard method to measure thermal
conductivity of non-metallic liquids according to the
standard ASTM D 2717 – 95 [9].
The experimental apparatus is made by a short
platinum wire welded on the support and immersed
in a cylindrical cell where the nanofluid sample is
placed. Besides, a thermocouple measures the
average temperature in the measuring cell. Border
effects on the wire are negligible and thermal
conductivity value is not dependent on eccentricity
between wire and internal surface cell, as
demonstrated by equation (1).
d T
q

d ln t 4k
(1)
Where T [°C] is the difference between fluid
temperature and a reference temperature, k [W/m K]
is fluid thermal conductivity, q [W/m] is heat flux
and t [s] is the time.
In the apparatus used in this work, a platinum wire
(with diameter of 0.1 mm and length of 35 mm) and
a thermocouples are connected with an electronic
device that allows the simultaneous determination of
the thermal conductivity and the average
temperature. Reproducibility is ±1%.
To assure a perfect mixing between the
nanoparticles and the base fluid both liquid and
solid phases, were mixed for 60 minutes with
magnetic homogenizator. The suspension then was
shacked in an ultrasonic vibrator to break clusters of
nanoparticles.
The solid volume fractions in these samples were
1.0%, 2.0% and 3.0% and the only temperature
investigated was 20°C. H2O TS identifies thermal
conductivity of the base fluid (water with no
powders inside). Figure 1, Figure 2 and Figure 3
show experimental data.
The results obtained with these samples were in
accordance with those presented in literature [1, 2,
8]. As expected thermal conductivity is directly
proportional to volume fraction and best results
have been obtained with ZnO nanoparticles.
Figure 3: Experimental data for nanofluids with water
and dispersant and ZnO
COOLING SYSTEM CONFIGURATION
In the present investigation two cooling system
options have been considered and each solution is
characterized by two steps, as Figure 4 and Figure 5
show.
Figure 1: Experimental data for nanofluids with water
and dispersant and Al2O3
Figure 4: Cooling system configuration (option 1)
Figure 2: Experimental data for nanofluids with water
and dispersant and CuO
wind turbine and typical dimensions are reported in
Table 2.
Table 2: Wind turbine characteristics
Maximum tower diameter
4,15 m
Minimum tower diameter
2,3 m
Tower height
80 m
In the option 1, to simulate the overall
performance of the cooling system a 0D model has
been built. Heat transfer from the tower to the
ambient has been modelled according to the theory of
external flow normal to the axis of a circular
cylinder. According to this theory the local Nusselt
number varies with the nature of the boundary layer
development on the surface and therefore is a
function of the angular position , defined as in
Figure 4.
Figure 5: Cooling system configuration (option 2)
The first heat exchanger placed inside the nacelle
of the wind turbine is the same for both the two
configurations described. It is conceived to work
with nanofluid as heat transfer fluid. This solution
can assure good heat extraction from the electric
and electronic part preventing overheating in a
better way than traditional fluids because nanofluids
are capable to enhance the heat transfer per unit of
area, that is the weak point of the electric devices
that produce a big amount of thermal power in a
small area. The enhancement of heat transfer from
the electric and electronic devices can be a very
important goal to assure long technical life and less
maintenance costs.
In the first option (Figure 4) the waste heat is
dissipated through the tower. To obtain this result
the internal side of the tower should be welded to a
spiral pipe with a step of 10 cm and an internal
diameter of 2 inch. In this way the wind tower can
be used as a heat dissipater.
The second option considers the possibility to
dissipate the waste heat directly in the sea, using a
titanium heat exchanger in order to minimize the
maintenance costs.
Both options were referred to a typical to 2/3 MW
Figure 4: Boundary layer formation and separation on a
circular cylinder in cross-flow.
Correlations may be obtained for the local Nusselt
number, and at the forward stagnation point for
Pr>0.6, boundary layer analysis yields an expression
of the form:
NuD   0  1.15  Re1D/ 2  Pr1/ 3
(2)
However from the standpoint of the engineering
calculation carried out in the present investigation the
empirical correlation due to Hilpert and reported next
as equation (3) was used.
Nu 
hD
 C  Re mD  Pr1/ 3
k
(3)
Where C and m are constant that for ReD number
greater than 106 are equal to 0.076 and 0.7,
respectively.
It is important to notice that the above correlation is
reasonable only for high values of the Reynolds
number (ReD >106).
To simplify the calculations each spire of the
tower heat exchanger has been considered
separately with its boundary conditions linked to the
previous and following one. In particular the inlet
water temperature is the outlet water temperature of
the previous spire. In this way it is possible to
approximate the heat flux along the spire surface as
a constant and to have a good approximation in the
results.
Three cases have been investigated to evaluate
the portion of the tower that must be employed to
assure a proper heat dissipation at different external
air temperatures, as Table 3 shows.
Table 3: Case under investigation
Case 1 Case 2 Case 3
Air velocity (m/s)
14
14
14
Air temperature (°C)
20
30
40
Calculations have been carried out considering that
the typical waste thermal power for a 2 MW wind
turbine is around 100 kW and thus the heat
exchanger has been designed to dissipate this
thermal power.
Figure 6 shows the results obtained for the 3 cases
considered: the minimum height of the tower heat
exchanger has been calculated in function of the
inlet temperature of the heat transfer fluid, from a
minimum of 60 °C to a maximum of 90 °C. In
particular in the figure each curve is referred to a
different external air temperature.
Height of the exchanging tower [m]
45
40
External air
temperature 20 °C
35
External air
temperature 30 °C
30
External air
temperature 40 °C
25
20
tower should be equal to 15 m, instead with 40 °C for
external temperature and 50 °C for heat transfer fluid
this value rises up to 40 m. Both these results are
compatible with the wind tower height, that is 80 m.
In the second option (Figure 6) the tower heat
exchanger is substituted by a closed hydraulic circuit
that exchanges thermal power with the sea through
an additional titanium heat exchanger. This solution
is more simple than the first one, but can yield more
problems because of the use of sea water as heat
transfer fluid.
CONCLUSIONS
In this work different cooling solutions for offshore wind turbines have been analyzed. In particular
the proposed options use nanofluids to enhance the
performance of the heat exchanging devices, to
assure long technical life and low maintenance cost
for the electric and electronic parts of the turbine.
The first results show that both the solutions are
very promising and need of further investigations to
setup the final configuration and to start numerical
and experimental campaign to test the proposed
solutions.
REFERENCES
[1] Q. Xue, Wen – Mei Xu, A model of thermal
conductivity of nanoflujids with interfacial shell,
Materials Chemistry and Physics 90 (2005), pp. 298
– 301.
[2] Yimin Xuan, Qiang Li, Heat transfer
enhancement of nanofluids, International Journal of
Heat and Fluid Flow 21 (2000), pp. 58 – 64.
[3] Yurong He, Yi Jin, Haisheng Chen, Yulong Ding,
Daqiang Cang, Hulin Lu, Heat Transfer and flow
behaviour of aqueous suspensions of TiO2
nanoparticles (nanofluids) flowing upward through a
vertical pipe, International Journal of Heat and Mass
Transfer 50 (2007), pp. 2272 – 2281.
15
[4] A.G. Agwu Nnanna, Experimental Model of
Temperature – Driven Nanofluid, Journal Heat
Transfer – June 2007 – Volume 129, pp. 697 – 704.
10
5
0
50
60
70
80
90
Temperature of heat transfer fluid [°C]
Figure 6: Height of the exchanging tower for 3
cases.
Considering the highest temperatures: 40 °C for
external air and 90 °C for heat transfer fluid it is
evident that the minimum height of the exchanging
[5] Qing – Zhong Xue, Model for effective thermal
conductivity of nanofluids, Physics Letter A 307
(2003), pp. 313 – 317.
[6] Y. Hwang, J.K. Lee, C.H. Lee, Y.M. Jung, S.I.
Cheong, C.G. Lee, B.C. Ku, S.P. Jang, Stability and
Thermal conductivity characteristic of nanofluids,
Thermochimica Acta 455 (2007), pp. 70 – 74.
[7] P. Keblinski, S.R. Phillpot, S.U.S. Choi, J.A.
Eastman, Mechanism of heat flow in suspensions of
nano-sized particles (nanofluids), International
Journal of Heat and Mass Transfer 45 (2002), pp.
855 – 863.
[8] Seok Pil Jang, Stephen U.S. Choi, Effect of
parameters on nanofluids thermal conductivity,
Journal Heat Transfer – May 2007 – Volume 129,
pp. 617 – 623.
[9] ASTM D 2717 – 95, Standard Test Method for
Thermal Conductivity of Liquids.
[10] Alessandro Franco, An apparatus for the
routine measurement of thermal conductivity of
materials for building application based on a hot –
wire method, Applied Thermal Energy 27 (2007)
pp. 2495 – 2504.
[11] Kau – Fui Vincent Wong and Tarun Kurma,
Tranport properties of alumina nanofluids,
Nanotechnology 19 (2008) 345702, (8 pp.).
[12] J. Buongiorno, Convective transport in
nanofluids, Journal Heat Transfer – March 2006 –
Volume 128, issue3, 240 (11 pp.).