Characteristics of a highly efficient propeller type small wind...
Transcript of Characteristics of a highly efficient propeller type small wind...
DTD 5 ARTICLE IN PRESS
Characteristics of a highly efficient propeller type
small wind turbine with a diffuser
Toshio Matsushima*, Shinya Takagi, Seiichi Muroyama
Research and Development Headquarters, NTT FACILITIES, INC. 2-13-1, Kita-otsuka,
Toshima-ku, Tokyo 170-0004, Japan
Received 17 December 2004; accepted 24 July 2005
Abstract
We studied the improved effects a diffuser had on the output power of small wind turbine
systems, aiming to introduce these systems to radio relay stations as an independent power supply
system. A frustum-shaped diffuser was chosen from an economical standpoint and wind speed
distribution. The effect the diffuser’s shape had on the wind speed was analyzed by simulation and
showed that the wind speed in the diffuser was greatly influenced by the length and expansion angle
of the diffuser, and maximum wind speed increased 1.7 times with the selection of the appropriate
diffuser shape. The wind speed in the diffuser was fastest near the diffuser’s entrance. We
conducted field tests using a real examination device with a diffuser and confirmed that the output
power of the wind power generator increased by up to 2.4 times compared to that of a conventional
turbine. Moreover, it was confirmed that the diffuser was especially useful where the wind direction
was constant.
q 2005 Elsevier B.V.. All rights reserved.
Keywords: Wind turbine; Diffuser; Wind speed; Output power; Energy production
1. Introduction
In recent years, the need to protect the global environment has seen the use of clean
energy systems being extended into the telecommunications [1]. A good example of this is
Renewable Energy xx (2005) 1–12
www.elsevier.com/locate/renene
0960-1481/$ - see front matter q 2005 Elsevier B.V.. All rights reserved.
doi:10.1016/j.renene.2005.07.008
* Corresponding author. Tel.: C81 3 5907 6421; fax: C81 3 5961 6424.
E-mail addresses: [email protected] (T. Matsushima), [email protected] (S. Takagi), muroya22@ntt-
f.co.jp (S. Muroyama).
T. Matsushima et al. / Renewable Energy xx (2005) 1–122
DTD 5 ARTICLE IN PRESS
the introduction of stand-alone power supply systems for sites such as radio relay stations
in mountainous areas where commercial power cannot be supplied [2]. Wind-solar hybrid
systems are desirable as stand-alone power supply systems in such applications, in terms
of steady power generation and stable power supply. The introduction of wind power
devices to the hybrid systems is beneficial for obtaining more power, because wind power
devices can generate power continuously throughout the day, so long as they receive wind
energy. So, an improvement in the output power generation of wind-solar hybrid systems
is desirable for this particular application.
The energy (P) generated by a wind turbine is proportional to the swept area (A) of the
turbine and the third power of the wind speed (n), as follows [3].
PZ1
2rAn3ðr : specific gravity of airÞ (1)
Therefore, enlarging the swept area (A) or increasing the wind speed (n) can effectively
increase the output power. In particular, since, the output is proportional to the third power
of the wind speed, increased output will be obtained even with a slight increase in wind
speed. One idea for increasing wind speed is the attachment of a diffuser to a wind turbine.
This idea was proposed in the middle of 1900 [4–6]. Recently, there have been reports on
the construction of large-scale wind turbine prototype systems with diffusers in New
Zealand [7]. The application of this kind of diffuser to small-scale wind turbines has also
been tried [7–9]. Grassmann [9] has analyzed the pressure distribution around the
propeller of a small wind turbine with wing-profiled ring diffusers and reported the test
results of an increased output voltage on that turbine. However, he did not analyze the
wind speed distribution around the propeller, the wind speed in the diffuser nor describe
the relationship between the increase in wind speed and the diffuser’s shape. This
relationship is important and must be analyzed because the diffuser’s shape is directly
related to the improvement in output power generation and must be designed accordingly.
Moreover, the actual increase in power generation by a real wind turbine has not been
measured and reported.
In this study, a diffuser with a simple external frustum view was chosen from the
economical and ease of processing standpoints, and the relation between the diffuser’s size
and wind speed was clarified. We constructed a real examination device by fitting a
diffuser to a propeller-type wind turbine and examined the effects it had on output power
generation, in outside field tests. We report the results and field test data obtained from
these tests.
2. Wind speed simulation in the diffuser
Fig. 1 shows the propeller-type wind turbine surrounded by the diffuser that we studied
in this report. In this examination, the diffuser shape was selected as a simple frustum,
taking a cost suppression in the manufacturing process at practical use.
Diffuser
Wind flow
Low atmospheric pressure
Wind turbine
Fig. 1. Schematic cross sectional view of a diffuser and wind speed increase mechanism.
T. Matsushima et al. / Renewable Energy xx (2005) 1–12 3
DTD 5 ARTICLE IN PRESS
2.1. Simulation method
Wind speed in the diffuser was simulated, varying the external dimensions of the
diffuser. Fig. 2 shows the shape of the diffuser, and Fig. 3 shows the 20 m!10 m analysis
space for the simulations. In these simulations, we used the thermo-hydrodynamic analysis
software program, I-DEAS [10]. I-DEAS is a three-dimensional design aid system
developed by SDRC Co. in the United States and mainly used in the manufacturing of
items such as automobiles, aircraft, and home electric appliances.
Fig. 2. External view of the diffuser.
Fig. 3. Analysis space.
T. Matsushima et al. / Renewable Energy xx (2005) 1–124
DTD 5 ARTICLE IN PRESS
In the simulations, after the diffuser was set in the space, a uniform amount of wind was
sent from the inflow inlet toward the outlet. We simulated the speed of the wind passing
through the diffuser, varying the diffuser’s main body length (L), entrance diameter (D), its
expansion angle (q) and flange length (T). Simulations were conducted on the diffuser
without a wind turbine in it. The diameter (D) of the entrance was selected to be 1 m,
taking the rotating diameter of a small propeller-type wind turbine into consideration.
Simulation parameters were as follows;
D: 1 m
L: 2–4 m
T: 0.1–0.5 m
q: 0–128
2.2. Simulation results
Fig. 4 shows some examples of analysis of wind speed distribution when wind speed (n)
is 5 msK1. From these analyses, we found that a diffuser can influence wind speed and that
the wind speed is highest at the entrance of the diffuser and lowest at the rear of the diffuser
outlet. Analysis also showed that longer the main body, the higher the wind speed, and that
maximum wind speed can be obtained along the inside of the diffuser near the entrance,
regardless of the length of the diffuser.
As the wind speed shows the maximum value at a point inside the entrance of the
diffuser, the wind speed at this point was selected and its dependency on each of the
parameters was analyzed.
Fig. 5 shows the wind speed ratio when the main body length L was changed and when
qZ48, TZ0.1 m and nZ5 msK1. From this figure, we find that an increase in the L initially
raised the wind speed ratio, but that as the L became larger the wind speed ratio gradually
approached a constant value.
Fig. 6 shows how the expansion angle q affected the wind speed ratio whenDZ1, LZ2,
TZ0.1 m and nZ5 msK1. The wind speed ratio increased more steeply in angles of less
than 48, reaching a maximum when expansion angle q was 68, and decreased at angles of
more than 68. We analyzed the relationship between flange length T and the wind speed
ratio, when DZ1, LZ2 m, qZ48 and nZ5 msK1. Fig. 7 shows the results. The initial 1.4
wind speed ratio without flange increased to 1.7 when a flange was attached. However, it
was found that flange length T had little effect on the increase of the wind speed ratio: for
all values of T greater than 0.1 m, the wind speed ratio remained essentially constant.
3. Characteristics of field trial device
3.1. Field trial device
A field trial device was constructed on the basis of the above results. Fig. 8
shows the dimensions of the diffuser that was used in the field tests: main body
4.5
4.26.9
7
8
5.4.555
5.0
5.1
5.0
5.0
5.1
4.5
3.93.6
3.3
3.3
3.94.24.85.7
7.84.
4.5
4.5
4.23.93.63.34.24.85.7
7.8
5.4
4.5
5.45.0
5.05.0 5.4
8.4
4.5
4.5 5.04.23.94.8
7.2
6.95.4
Unit : ms–1
Unit : ms–1
Unit : ms–1
diffuser
diffuser
diffuser
(b) L=2 m
(c) L=3 m
(a) L=1 m
Fig. 4. Example of analysis results.Simulation was done at a wind speed n of 5 msK1. Dimensions of the diffuser
were DZ1 m, qZ48, and TZ0.1 m.
T. Matsushima et al. / Renewable Energy xx (2005) 1–12 5
DTD 5 ARTICLE IN PRESS
length LZ2 m, entrance diameter DZ1 m, expansion angle qZ48, and flange length
TZ0.1.
A five-blade propeller type wind turbine was installed in the diffuser to make a field
trial device (rotor diameter: 950 mm, rated power: 62 W at 8 msK1). Diffuser was made of
aluminum frame and a 0.5 mm thick polyester sheet on it to lighten the weight. In addition,
a tail unit 0.4 m high and 1.0 m long was installed at the top and bottom at the rear of the
diffuser to make the diffuser follow the wind direction.
1.0
1.2
1.4
1.6
1.8
2.0
Main body length L (m)
Win
d sp
eed
ratio
s
D=1, =4 degrees, T=0.1 m
0 1 2 3
Fig. 5. Relation between main body length L and wind speed ratio. Wind speed ratios are calculated based on an
outside speed of 5 msK1.
1.00 121082 4 6
1.2
1.4
1.6
1.8
2.0
Expansion angle (degree)
Win
d sp
eed
rat
ios
D=1 m, L=2 m, T=0.1 m
Fig. 6. Relation between expansion angle q and wind speed ratio. Wind speed ratios are calculated based on an
outside speed of 5 msK1.
1.0
1.2
1.4
1.6
1.8
2.0
0 0.1 0.2 0.3 0.4 0.5
Flange length T(m)
Win
d sp
eed
ratio
s
D=1 m, L=2m, =4 degrees
Fig. 7. Relation between flange length T and wind speed ratios. Wind speed ratios are calculated based on an
outside speed of 5 msK1.
T. Matsushima et al. / Renewable Energy xx (2005) 1–126
DTD 5 ARTICLE IN PRESS
Fig. 8. Dimensions of diffuser applied for the field trial device.
T. Matsushima et al. / Renewable Energy xx (2005) 1–12 7
DTD 5 ARTICLE IN PRESS
Fig. 9 shows an external view of the device. A conventional five-blade propeller type
wind turbine of the same type as that used for the device was set up at the same test site to
compare their characteristics. Fig. 10 shows the makeup of the experimental apparatus.
The output power from the two wind turbines is stored in a lead-acid battery and the excess
energy is consumed by a dummy load. A propeller type anemometer was used for the
measurement of wind speed and direction. Wind speed, wind direction, and output power
were measured by a data logger at 1-s intervals.
3.2. Experimental results and discussion
Fig. 11 shows typical data obtained for the wind speed and output power on the field
trial device and the conventional wind turbine. Both devices had largely the same output
Conventionalwind turbine
Field trial device
Fig. 9. External view of field trial device installed at test site.
Shuntresistor
Shuntresistor
200AhVRLA
Data loggerPropeller typeanemometer
Field trial device
Dummyload
wind speedwind direction
Controller
Controller
voltage
current
Conventionalwind turbine
Fig. 10. Experimental apparatus.
T. Matsushima et al. / Renewable Energy xx (2005) 1–128
DTD 5 ARTICLE IN PRESS
power up until 2:00 pm, but from that time the field trial device delivered larger output
power than the conventional wind turbine.
Fig. 12 shows changes in energy production and the energy production ratio over time.
Until 12:00 pm the energy production of the conventional wind turbine was generally
00 2 4 6 8 10 12 2 4 6 8 10
5
10
15
Timea.m. p.m.
0
5
10
15
0
2.5
5.0
Win
d Sp
eed
(ms–1
)O
utpu
tpow
er (
W)
Out
putp
ower
(W
) Conventional wind turbine
Field trial device
12
(1) Wind speed
(2) Output power
Fig. 11. Changes in wind speed and output power characteristics over time (2002/10/19).
Time
Ene
rgy
prod
uctio
n (W
h)
0
1
2
3
4
Rat
io o
f th
e en
ergy
prod
uctio
n
a.m. p.m.
0
1
2
3
4
Field trial device
Conventional wind turbine
Ratio of energy production
0 2 4 6 8 10 12 2 4 6 8 10
Fig. 12. Changes in energy production characteristics and ratios over time (2002/10/19).
T. Matsushima et al. / Renewable Energy xx (2005) 1–12 9
DTD 5 ARTICLE IN PRESS
larger than or equal to that of the field trial device. However, the proportion of energy that
was generated by the test device increased to reach a maximum of 1.75 times at 5:00 pm.
Total energy production of the field trial device for the entire day was 1.16 times that of the
conventional wind turbine. A larger energy production than this, however, had been
expected on the basis of the simulation results obtained.
We next focused attention on how the wind-following performance affected output
power. Both wind turbines were fixed facing the direction where frequency distribution of
the wind speed was high, and their output power was measured. The results obtained
(Fig. 13) show the power from the field trial device was larger than that from the
conventional wind turbine for the entire day. Energy production and the energy production
ratios are shown in Fig. 14. Fixing both wind turbines increased the superiority of the
energy production ratio of the field trial device to over one and raised the energy ratio to a
maximum of 2.44 times, and in terms of total energy production for the entire day, the
output of the field trial device was 1.65 times than that of the conventional wind turbine.
These results indicate wind-following performance is a problem affecting the energy
production characteristics of the field trial device. That is, when the wind direction
changes frequently over a short period, it is difficult for the device to quickly and correctly
adjust itself to the new direction of the wind. In these weather conditions, therefore, the
device may not make effective use of wind energy. From a visual evaluation of both
turbines, when the wind direction changed frequently, the conventional turbine adjusted
itself more than did the field trial device, suggesting a relationship between this and the
above-mentioned energy production ratio.
Fig. 15 shows the output power measured for both wind turbines after fixing them in the
direction where the frequency distribution of the wind was high. At each wind speed, the
actual output power of the field trial device was from 3 to 4 times larger than that of
0.0
2.5
5.0
Win
d sp
eed
(ms–1
)
Timea.m. p.m.
Conventional wind turbine
Field trial device
0 2 4 6 8 10 12 2 4 6 8 10
0
5
10
15
20
Out
put p
ower
(W
)
0
5
10
15
20
Out
put p
ower
(W
)
(1) Wind speed
(2) Output power
Fig. 13. Changes in wind speed and output power characteristics over time (2002/11/16). Both wind turbines were
fixed in the same direction.
T. Matsushima et al. / Renewable Energy xx (2005) 1–1210
DTD 5 ARTICLE IN PRESS
the conventional turbine, in the wind speed conditions over 3 m/s. Consequently, we are
convinced that the application of the diffuser is useful for the improvement of the output
power from a wind turbine, when wind speed and direction are stable.
From our numerical simulation, the maximum wind speed ratio and output power
increase in the trial device are calculated to be 1.7 and 5, respectively. Therefore, the
measured value was somewhat lower than expected. One possible reason is that the
simulation was done on a diffuser without a wind turbine in it, and in the field tests, wind
flow into the diffuser may have been affected by the presence of the wind turbine
propellers. Another possible reason is the frequent and rapid changes in wind speed and
direction. In these conditions, the generator may not give an optimal performance.
4. Conclusions
We evaluated a wind turbine fitted with a diffuser with the aim of improving the
turbine’s output power characteristics. We used thermohydrodynamic analysis software to
simulate the effect of the diffuser parameters on the wind speed, and evaluated the turbine
00 2 4 6 8 10 12 2 4 6 8 10
1
2
3
Time
Ene
rgy
prod
uctio
n (W
h)
0
1
2
3
4
5
6
Ene
rgy
prod
uctio
n ra
tios
a.m. p.m.
Field trial device
Conventional wind turbine
Ratio of energy production
Fig. 14. Changes in energy production characteristics and ratio over time (2002/11/16). Both wind turbines were
fixed in the same wind direction.
00 1 2 3 4 5 6 7
20
40
60
80
Wind speed (ms–1)
Out
put p
ower
(W
)
Conventional wind turbineField trial device
Fig. 15. Power curve measured in the field tests.
T. Matsushima et al. / Renewable Energy xx (2005) 1–12 11
DTD 5 ARTICLE IN PRESS
characteristics using a field trial device. The following results were obtained for a wind
turbine with a diffuser.
1 We ascertained the effect on wind speed for each of the diffuser parameters (main body
length L, entrance diameter D, expansion angle q and flange length T). Results showed
that the parameters were able to increase the maximum wind speed in the vicinity of
the diffuser entrance by around 1.7 times.
T. Matsushima et al. / Renewable Energy xx (2005) 1–1212
DTD 5 ARTICLE IN PRESS
2 The fitting of a diffuser improved the power curve and increased the energy production
of the wind turbine. A maximum energy production ratio of around 2.4 times was
obtained by collecting wind energy in the turbine.
3 The diffuser is useful at sites where the wind direction is comparatively steady, by
setting the turbine in the direction of the wind.
References
[1] Ueno H, Serada T. Wind turbine generation system & photovoltaic power generation system. NTT Power
Build Facil J 2002;39(231):42–5.
[2] Tanezaki S, Kuramoto M, Yamanaka T. Construction of stand-alone power system using photovoltaic
power generation. NTT Power Build Facil J 2000;37(220):18–21.
[3] Solar Energy Utilization Handbook, p.55, Japan Solar Energy Society.
[4] Glauert H. The elements of airfoil and airsrew theory. Cambridge: Cambridge University Press; 1959.
[5] A. Kogan, A. Seginer. Shrouded Aerogenerator Design Study II, Axisymmetrical shroud performance,
Department of Aeronautical Engineering, Technion, T.A.E. report 32, (1963).
[6] Oman RA, Foreman KM. Cost effective diffuser augmentation of wind turbine power generators Grumman
aerospace corporation, Bethpage, New york 11714 Second workshop on wind energy conversion systems,
Washington, D.C., June 9–11 1975.
[7] Bet F, Grassmann H. Upgrading conventional wind turbines. Renew Energy 2003;28:71.
[8] Ohya Y, Toritani T, Sakurai A, Inoue M. Part 2— Development of high-performance wind turbine system
by wind-lens effect The 24th wind energy symposium 2002.
[9] Grassmann H, Bet F, Cabras G, Ceschia M, Cobai D, DelPapa C. A partially static turbine—first
experimental results. Renewable Energy 2003;28:1779.
[10] Russell R, Louie J. Recent advances in thermal/flow simulation: integrating thermal analysis into the
mechanical design process, 11th IEEE Semi-Therm, 1995.