Renewable Energy 116 (2018) 215e223

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Experimental and theoretical investigation of micro wind turbine for

low wind speed regions
Salih N. Akour a, *, Mohammed Al-Heymari b, Talha Ahmed b, Kamel Ali Khalil b
a
Mechanical Engineering Department, School of Engineering, The University of Jordan, Amman 11942, PO Box 13512, Jordan
b
Department of Sustainable and Renewable Energy Engineering, University of Sharjah, PO Box 27272, Sharjah, United Arab Emirates

a r t i c l e i n f o a b s t r a c t

Article history: Micro wind turbine blades for low average wind speed regions like the Arabian Peninsula, Jordan Desert
Received 12 November 2016 and United Arab Emirates are designed and implemented. Wind profiles for two locations in UAE are
Received in revised form investigated and utilized in the design and the economic analysis. Airfoils BW3, A18 and SG6043 are
17 June 2017
selected and utilized as candidates for designing micro turbine blades. Blade element momentum theory
Accepted 25 September 2017
Available online 26 September 2017
is used to design the blade 3D geometry. A methodology to optimize the blade geometry for average
wind speed 5 m/s based on operational Reynolds number is developed and utilized. To account for the
aerodynamic behavior over the 3D blade geometry, the power coefficient for the blades of each airfoil is
Keywords:
Micro wind turbine
obtained using the simulation software QBlade. Blades developed using airfoil BW3 showed the highest
Arabian Peninsula wind profile performance. A prototype is built using 3D printer and tested in open air environment (natural envi-
Feasibility study ronment) to validate the simulation results. Comparison with existing commercial wind turbines ac-
cording to cost and output power is carried out based on the concept of replacing wind turbines swept
area with the equivalent array of micro wind turbines. The results show that the new design is more cost-
effective and more wind energy is harnessed using equivalent swept area.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction (cut-in speed) and economic feasibility are very important in

optimizing the final design of micro wind turbine blade. The
Wind energy represents one of the most promising sector of starting time optimization carried out by Pourrajabian et al. have
renewable energy. Harvesting energy from low average wind speed shown that the starting time of micro wind turbine can be reduced
profile regions such as the Arabian Peninsula is a challenge. Due to by increasing the number of blades which would also increase the
the low average of wind speeds in Arabian Peninsula, use of large power coefficient, Cp [2]. Increasing the number of blades increases
and medium size conventional wind turbines is not economically the overall energy of the system, but it increases the overall cost of
feasible. These conventional turbines are designed for high rated the system.
wind speeds in range of 10 m/s to 15 m/s. However the average Centimeter-scaled micro wind turbines (2e12 cm diameter) are
wind speed in the Arabian Peninsula is in range of 4 m/s to 7 m/s. very economical to produce, however they have shown low Cp
The intent of this study is to design a micro wind turbine that fits which ranges between 5 and 10%. They usually have 3e12 blades
the low average wind speed regions. United Arab Emirates (UAE) [3e5]. Larger micro wind turbines with 120e420 cm diameter have
located in the Gulf Region of south-eastern part of the Arabian shown greater Cp values of about 40% [6]. The reason for the dif-
Peninsula is chosen as sample site for experimental and theoretical ference in Cp goes back to the fact that the larger blades employ
investigation of micro wind turbine for low wind speed regions. aerodynamic airfoil blade configuration whereas the small ones
The average wind speed in UAE is in range of 5 m/s to 6 m/s [1]. have fan type blade configuration [4].
The key part in designing wind turbines is the blade. Generally, Micro wind turbine performance evaluation and optimizing can
the main goal in designing a wind turbine blade is maximizing the be done using computational fluid dynamics (CFD) as well as Blade
power coefficient. In case of micro wind turbine, the starting time Element Momentum (BEM) theory codes [6e10]. BEM theory codes
have an advantage of having much less computational time over
CFD. Recent studies have shown that there is a good agreement
* Corresponding author. between BEM theory codes results and experimental results of
E-mail address: [email protected] (S.N. Akour). micro wind turbine [6].

https://doi.org/10.1016/j.renene.2017.09.076
0960-1481/© 2017 Elsevier Ltd. All rights reserved.
216 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223

The main objective of the current study is to design an efficient

micro wind turbine blade using BEM theory codes by employing
low Reynolds number airfoils, i.e. low wind speed airfoil. It is
proven by Refs. [11] and [12] that Reynolds number has consider-
able effect on turbine blades performance that employ such airfoils.
The wind profiles of two cities in UAE, Abu Dhabi and Al Ain, are
considered as case studies in examining the performance of the
new blades design. QBlade software is used in evaluating the per-
formance of the micro wind turbine blade since it has been suc-
cessfully validated by researchers against the BEM theory codes and
wind tunnel test data of full scale wind turbines [13].
Fig. 1. Rayleigh and Weibull PDFs for Abu Dhabi.
A 3D printer is used to prototype the blade final geometry. 3D
printed wind turbine blades are utilized by other researchers and
they have shown to have good performance [14]. In this study, the
blades have been built and tested in real life environment to vali-
date the theoretical predictions.

2. Wind data analysis

The wind data of Abu Dhabi and Al Ain is obtained from Iowa
Environmental Mesonet (IEM) that collects environmental data
using the Automated Surface Observing System (ASOS) which is
located at airports. Table 1 shows summary of the data.
Weibull distribution is widely used by many researches for wind Fig. 2. Rayleigh and Weibull PDFs for Al Ain.
data analysis, but since the sites have low wind speeds, it has some
limitations for the low wind speed probabilities [15]. For this
reason, Rayleigh distribution is also used. Graphs for Weibull and 1. Each blade is divided into set of elements.
Rayleigh probability density function (PDF) for Abu Dhabi and Al 2. There is no aerodynamic interaction between elements (thus, no
Ain are shown in Figs. 1 and 2 respectively. radial flow).
It can be seen from Figs. 1 and 2 that wind speeds close to 5 m/s 3. The forces on the blades are determined by the lift and drag
have the highest probability. It is obvious that the average wind characteristics of the airfoil shape of the blades where these are
speeds of both sites are close to 5 m/s. Therefore, this speed is function of Reynolds number.
chosen to be the design rated wind speed for the micro wind tur- 4. The design is optimized for wind speed of 5 m/s. Two main
bine blades. variables are impeded in Reynolds number, the chord length
The wind rose for both cities Abu Dhabi and Al Ain are presented and the wind speed. The design wind speed is fixed at 5 m/s
in Fig. 3. These graphs illustrate that the wind dominant direction is whereas the chord length is varied to reach the optimum ge-
North-West. The wind turbine is equipped with guide vane to allow ometry for the blades that maximizes the power coefficient.
self-adjustment with the wind direction. 5. BEM theory for ideal rotor with wake is used in designing the
blade shape.
6. The performance analysis of the optimum geometry, which is
3. Theoretical modeling obtained from BEM, is performed using Qblade software Pack-
age. Qblade carries the performance analysis by considering the
Theoretical analysis is conducted for selected airfoils of high aerodynamic effects of the neighboring elements that is ignored
performance at low wind speeds (low Reynolds number), i.e. Air- by the BEM theory. This provides performance analysis that is
foils that have high lift coefficient. The blade geometry is optimized very close to the real life performance.
for wind speed of 5 m/s. It is well known that the lift and drag
coefficients are function of angle of attack and Reynolds Number
[11,12]. To reach highest performance, Reynolds number is
3.1. Airfoils
considered as an optimization parameters' along with the other
parameters' considered by the BEM theory. Fig. 3 illustrates the
The airfoils for the micro wind turbine blades are chosen from
developed methodology that considers Reynolds number in
‘Summary of Low-Speed Airfoil Data’ [11,12]. Three airfoils are
designing the blades. This method is developed to reach the opti-
considered for designing the micro wind turbine blades. Simulation
mum geometry of the blades.
is performed to choose the airfoils with highest Cp . Airfoils BW3,
The basic assumptions used in the design are mainly the as-
SG6043 and A18 have been chosen because they are designed to
sumptions of the BEM theory. The assumptions and conditions that
operate at low Reynolds number i.e. to operate at low wind speed.
are implemented in the design process are as the following:
These airfoils are presented in Fig. 4. Careful analysis is done on
these airfoils to investigate their performance at low values of
Table 1 Reynolds number that reflects the operational condition in UAE.
Descriptive statistics for Abu Dhabi and Al Ain.

Abu Dhabi Al Ain 3.1.1. A18

Average (m/s) 5.36 Mean (m/s) 6.39 The A18 airfoil was not designed for wind turbines initially but
Median (m/s) 5.02 Median (m/s) 5.90 for free flight airplanes by Randy Archer. This airfoil was named
Mode (m/s) 4.21 Mode (m/s) 5.01 after his model that he used in 1993 World championship contest. It
Standard Deviation 3.71 Standard Deviation 3.48
has smooth surface finish, maximum thickness of 7.26% of the
 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223 217

Fig. 3. Illustration of Wind roses for both Abu Dhabi and Al Ain Cities in UAE.

chord length located at 30% chord length measured from the airfoil torque at low wind speeds that will lead to improve energy har-
leading edge and maximum camber of 3.84% of the chord length nessing [11].
located at 45% chord length measured from the airfoil leading edge
[12]. This model was introduced later as a small wind turbine airfoil 3.1.3. SG6043
because it has shown good results for constant speed and stall The SG6043 airfoil is one of the series that are designed for the
regulated wind turbines. In addition, it is suitable for variable speed small wind turbines by Giguere and Selig. This airfoil has a
wind turbines as it has shown good characterization of lift-to-drag maximum Cl of 1.22 at optimum angle of attack, a of 5
. This high
ratio as long as it is clean [12]. lift coefficient makes it a good candidate for the design [11].

3.1.2. BW3 3.2. Methodology for selecting the operational Reynolds number
Bergey Windpower has designed the BW3 for their wind turbine
system. Unlike other airfoil geometries, the BW3 is designed with a BEM theory is coded in MATLAB to obtain the optimum blade
relative thickness of 5% of the airfoil chord length. BW3 airfoil shape for the three airfoils. Reynolds number is function of chord
apparently does not experience the formation of large separation length and wind speed assuming the density and viscosity of air is
bubbles at Reynolds number of 100,000 contrary to other thicker constant. The methodology shown in Fig. 5 is developed and
airfoils. The high lift coefficient, Cl value of this airfoil makes it implemented to obtain the operational Reynolds number and blade
suitable for a small wind turbine blades from an operational point shape. Through the iterations the wind speed is fixed at 5 m/s and
of view because a high Cl provides the turbine with a high startup the chord length is varied along with the other blade geometry
parameters (the element location and twist angle).

3.3. Effect of Reynolds number

Fig. 6 shows how the Reynolds number is varying according to

Fig. 4. Illustration of A18, BW3 and SG6043 Airfoil Profiles in dimensionless format
with respect to the chord length. Fig. 5. Methodology to find the Reynolds number.
218 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223

Table 2
Geometry of the blade made of BW3 airfoil.

r/R Chord Length (m) Twist Angle (Degrees)

0.1 0.077 42.717

0.2 0.117 27.494
0.3 0.105 17.847
0.4 0.087 11.904
0.5 0.073 8.049
0.6 0.062 5.396
0.7 0.054 3.476
0.8 0.047 2.029
0.9 0.041 0.901
1 0.032 0.000

3.5. Simulation results

Fig. 6. Variation of Reynolds number with iterations.

The main objective of the simulation in QBlade is to investigate

the performance of the ATT blade where the aerodynamic inter-
the methodology that is presented in Fig. 5. It is obvious from Fig. 6
action among the blade elements is considered. Fig. 8 shows the
that the initial estimate of Reynolds number is very high and it
Cp of the ATT blade with respect to the tip speed ratio (TSR). The
settles down to around 25,000 after three iterations for BW3 and
graph shows a maximum Cp of about 0.38 at TSR of 4.4. It is
A18. However Airfoil SG6043 keeps oscillating close to Reynolds
obvious that the ATT blade can work very well within TSR ranging
number 25,000. The oscillation of SG6043 is damping slowly with
3.75e5.8 since the blade maintains Cp above 0.30 within this TSR
the iterations till it reaches stable level i.e. till it reaches the mean of
range.
the oscillations.
It's important to find the Cp of each blade at the operational
Reynolds number to choose an airfoil with the highest Cp . The 4. Experimental work
higher the Cp , the better the performance of the blade will be
achieved. Figs. 6 and 7 show the sensitivity of the power coeffi- Experiment is conducted to validate the simulation analysis.
cient Cp to Reynolds number for the three different airfoils. It is Prototypes for the ATT blades are constructed and installed on AIR-
obvious from Fig. 7 that BW3 and A18 have the best and stable X turbine i.e. replacing the AIR-X turbine blades with the ATT
performance whereas SG6043 has the least and unstable perfor- blades. The system has been tested in open environment to assure
mance. Among the three blades, BW3 blade has the highest per- realistic results.
formance, therefore BW3 is chosen as the airfoil for the micro
wind turbine blade. From this point forward in this paper, the 4.1. Prototyping
micro wind turbine that utilizes BW3 airfoil blades will be referred
as Akour Team Turbine (ATT). In order to validate the simulation results, a prototype for the
ATT blades was built using the XYZ Da Vinci 1.0 3D printer which
uses fused deposition and modeling (FDM) technique. Due to the
3.4. Blade geometry
limitation in the printing bed size, the 55 cm ATT blade was divided
into four sections that was welded together to form the final shape
The aerodynamic curves (lift coefficient and drag coefficient
of the blade as shown in Fig. 9.
curves) of the BW3 airfoil at the operational Reynolds number have
been used as input for the BEM theory which is coded in MATLAB.
The final blade geometry of ATT obtained from the MATLAB code is 4.2. Experimental setup
shown in Table 2. Thorough details of the BEM theory equations
and sensitivity analysis for the tip speed ratio, number of rotor The ATT blades have been tested using a commercial wind tur-
blades and lift to drag ratio can be found in Ref. [16]. bine AIR-X 12 V. This wind turbine is produced by Southwest Wind
Power Company. The AIR-X is designed to produce 400 W at a rated
wind speed of about 12.5 m/s. It is equipped with a three-phase
brushless permanent magnet alternator. The AIR-X has a built-in
AC-DC rectifier and maximum power point tracking (MPPT)

Fig. 7. Variation of Cp with iterations. Fig. 8. Illustration of Cp at different TSR for ATT Blade.
 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223 219

Fig. 9. Four sections of the ATT blade.

system for its originally designed blades [17]. The tests have been 5.1. Open air environment tests
conducted by replacing the original turbine blades with the ATT
prototype blades on the AIR-X rotor as shown in Fig. 10. A total of 40,000 data points of both current and voltage have
been collected using the DAQ for a wide wind speed range
(2.7e5.9 m/s). Fig. 11 shows that as the wind speed increases the,
4.2.1. Measuring methods
electric current increases. The behavior of the rotor rotational
The electrical power, current and voltage of the wind turbine
speed with respect to the wind speeds is shown in Fig. 12. It is
have been recorded using an NI 6009 USB data acquisition system
obvious that the rotor angular speed is linearly varying with the
(DAQ) module at a 1 kHz sampling rate. The voltage is measured
wind speed.
directly by connecting the two terminal output of the wind turbine
The final power calculated from the wind turbine is not only
to the DAQ whereas the current is measured using a 5 ohm shunt
affected by the Cp of the airfoil but also by the generator, gear box
resistor. The DAQ has been connected to a PC via USB connection.
and all electric circuits used in the wind turbine. Fig. 13 presents the
Each test for ATT has been videotaped using a camera that has
calculated Cp at different TSR in the open air environment tests. The
capability of capturing 30 frames per second (FPS). The recorded
values of the calculated Cp versus TSR obtained from the open air
videos of the tests have been processed using slow-motion video
environment test shown in Fig. 13 closely follow the value of Cp
software VideoMach. The rotational speed of the rotor in terms of
versus TSR obtained from QBlade simulation shown in Fig. 8. Any
revolution per minute (RPM) has been measured through these
differences are due to the continuous variation of the air speed in
recorded videos.
the open air environment.

5. Experimental results and discussion

5.2. MPPT effect

The tests were conducted in open air environment in Al-Dhaid,

The AIR-X has a built in MPPT which controls the RPM of the
Sharjah, UAE (DD coordinates: 25.4178, 55.6821) from December 2,
generator to ensure that the wind turbine operates at its rated TSR
2015 to December 9, 2015. The location is open wide flat field
[18e20]. The rated TSR of AIR-X can be calculated from its rated
without any obstacles in range of 100 m diameter. The mast height
wind speed and RPM which are 12.5 m/s and 700 RPM respectively
of the turbine is 2 m above the ground. The turbine is equipped
[17]. Since the calculated TSR is 3.5, the AIR-X will try to maintain a
with guide vane to adjust for the wind direction.
TSR of 3.5 irrespective of the blades or wind speed. This explains
why all the values of Cp are around TSR of 3.5 in Fig. 13.

5.3. Power curve

Prototype blade
Applying the variable speed wind turbine concept, a MPPT can
be programmed to control the generator so that it operates all the
time at the maximum Cp . The maximum Cp obtained experimen-
tally is 0.34 at TSR of 3.5. Figs. 14 and 15 show two different scales of
the power curve for the ATT and AIR-X. Fig. 14 shows that the ATT is

Fig. 10. AIR-X with ATT blades. Fig. 11. Current at different wind speeds.
220 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223

Fig. 12. Angular speed (RPM) at different wind speeds.

Fig. 15. Full power curves of ATT and AIR-X.

Table 3
Characteristics of ATT.

ATT Characteristics

Radius 0.5 m
Area 0.79 m2
Rated Wind Speed 5 m/s
Rotational Speed 415 RPM
Rated Power 25 W

TSR of 3.5.
The variation trend of the current with the wind speed is linear
as it can be seen from Fig. 11 which agrees with the physics of
problem [16]. The flatness in the electric current graph in Fig. 11 can
be referred to the MPPT system that tries to operate the turbine at
Fig. 13. Calculated Cp at different TSR in the open air environment tests. the programed optimum TSR.
The stairway shape of Fig. 12 is due to the time needed for the
harnessing more power than the AIR-X. ATT is producing more turbine to respond to the changes in wind speed and due to the
power than AIR-X within the range of 2 m/s to 11.5 m/s wind speed recording of the average wind speed over half minute (every 30 s).
as shown in Fig. 15. AIR-X has a better performance in the range The linear trend of the variation of the rotor angular speed with
above 11.5 m/s which is beyond the operating wind speed of ATT. respect to the wind speed agrees with physics of problem [16].

5.5. Sub-conclusion
5.4. Discussion
The ATT has rated power of 25 W at the rated wind speed of 5 m/
Figs. 8 and 13 show the power coefficient obtained from the
s. Table 3 shows the final characteristics of ATT.
simulation analysis and experimental tests respectively. The
maximum power coefficient achieved from the simulation and the
experiments is 0.38 at TSR of 4.4 and 0.34 at TSR 3.5 respectively.
The differences in the maximum power coefficient can be referred
to two main reasons: first, the continuous variation of wind gust
speeds where the average wind speed over half minute (30 s) is
considered. Second the MPPT system of AIR-X is tuned for TSR of 3.5
whereas the ATT is design for TSR of 4.4. The second issue also
explains why in the simulation results the power coefficient peak
occurs at TSR of 4.4 whereas in the experimental results occur at

Fig. 14. Power curves of ATT and AIR-X for low wind speeds. Fig. 16. Energy (top) and CO2 footprint (bottom) analysis [21].
 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223 221

Table 4 6.3. Payback period

Cost results [21].

Phase Cost (USD) The payback period can be calculated using the following
Material 1.51
equation:
Manufacture 0.47
Transport 0 Total cost
Payback ¼ (1)
Use 0 Annual energy cost savings
Disposal 0
Total 1.98 According to Equation (1), the annual energy cost savings is
equal to the annual energy production of ATT in UAE multiplied by
the UAE green tariff rate of AED 0.23/kWh (USD 0.06/kWh). Abu
6. Economic feasibility study Dhabi is considered as sample location to calculate the annual ATT
energy production in UAE. Rayleigh PDF shown in Fig. 1 is used to
The economic feasibility study has been conducted for the ATT calculate the annual number of hours for different wind speeds as
blades as per the selected material and the selected mass produc- shown in Table 5. Annual energy production is calculated by sum-
tion technique. The software CES EduPack's eco audit tool is used to ming up the product of annual number of hours of different wind
generate the life cycle of ATT blades showing the cost, energy and speeds and power produced at those wind speeds. Power produced
CO2 footprint. by ATT at different wind speeds is determined by the data of the
The chosen material is thermoplastics because it is light in power curve shown in Fig. 15.
weight, durable, eco-friendly and can be recycled. The injection The annual energy cost saving is USD 27.36 calculated by
molding is chosen as manufacturing technique since it is cost- multiplying the total ATT annual energy production 456 kWh by
effective and suitable for thermoplastic material [21]. The relative the green tariff rate USD 0.06/kWh. The total price of single ATT
cost index for the injection molding is ranging from USD 16.44 to including all components is estimated to be USD 50. This gives
USD 111.63 per unit. This value is based on material cost of USD 8/ payback period for ATT of 1.83 years.
kg, blade mass of 1 kg and batch size of 1000 unit [21].
6.4. Comparison of ATT with other wind turbines
6.1. Energy and CO2 footprint analysis
To demonstrate the effectiveness of ATT, it is compared with
Fig. 16 shows comparison between the energy consumption and some wind turbines available in market. The comparison is in terms
CO2 footprint. It can be seen from Fig. 16 that the material pro- of swept area, power, cost and annual energy production. Table 6
duction ranks first in terms of the energy consumption whereas presents the comparison analysis between ATT and other wind
CO2 footprint and the manufacturing process rank second and third turbines available in market. All the values are obtained for wind
respectively. The material stage consumes 85% of the energy speed of 5 m/s. The price of a single ATT is estimated to be USD 50 as
whereas the manufacturing is 14% of the total energy. The material mentioned in previous section. All prices shown in the table are the
phase represents 81% of CO2 footprint whereas the manufacturing prices of the wind turbine themselves excluding the installation
represent 17.5% of CO2 footprint. Due to recycling process, there is cost which would vary depending on the location.
recovery of 36.5 MJ of energy and 2.11 kg of CO2 footprint. SY-1kw is so much cheaper compared to Britwind R9000 and
Bergey Excel 10 because it's Chinese product i.e. the labor cost is
6.2. Cost analysis cheap compared to USA and Europe. ATT has very low cost
compared to Britwind R9000 and Bergey Excel 10 and at the same
Cost analysis showed that the three ATT blades cost USD 1.98. As time it is producing more power. For SY-1kw turbine, the cost per
seen in Table 4, the material accounts for 76.3% of the cost and the Watt is similar to ATT but it has higher Cost compared to Equivalent
manufacturing makes up 23.6% of the cost. ATT in Terms of Swept Area.
Table 7 shows the annual energy production for the commercial
wind turbines SY-1kw, Bergey Excel 10 and Britwind R9000 based
Table 5
on the data of wind speeds and their corresponding annual hours
Annual energy production of ATT in UAE.
that are presented in Table 5. Power curves available at the website
Wind Speeds (m/s) No. of Hours Annual Energy of each commercial wind turbine are used to determine the power
Annually (h) Production (Wh)
produced at different wind speeds by these turbines. A concept of
1 467 0 replacing these commercial wind turbines with equivalent number
2 860 1222
of micro wind turbines (ATT in this case) in terms of swept area is
3 1125 5397
4 1238 14,084 presented.
5 1210 26,876 Table 7 demonstrates that more energy is produced from
6 1074 41,238 equivalent array of micro wind turbines in terms of swept area
7 878 53,521 compared to one wind turbine. This energy would also be produced
8 665 60,554
9 470 60,901
at a much lower cost as illustrated in Table 6. From the analysis, it
10 310 55,175 can be seen that the concept of replacing one wind turbine with
11 192 45,460 equivalent array of micro wind turbines in terms of swept area is
12 112 34,302 more cost-effective and harnesses more energy from equivalent
13 61 23,829
swept area. Other advantages would include producing power at
14 31 15,305
15 15 9117 very low wind speeds and the micro wind turbine system can also
16 7 5051 be installed in urban areas because unlike the large wind turbines,
17 3 2608 micro wind turbines can harness wind energy in turbulent air flow
18 1 1257 since the uniform air mass needed as small as the size of the ATT
Total Annual Energy Production (kWh) 456
rotor. For example, according to the last two rows in Table 7 the
222 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223

Table 6
Comparison of ATT with wind turbines available in market at wind speed of 5 m/s.

Data at 5 m/s

Turbine ATT Britwind R9000 Bergey Excel 10 kW SY-1kw

Diameter (m) 1 5.5 7 9.2

Swept Area (m2) 0.79 23.76 38.48 66.48
Power (W) 25 576 848 2500
No. of ATT Required for Equivalent Swept Area e 30 49 85
Total Power from ATT Turbine (W) e 750 1225 2125
Power Ratio e 1.30 1.44 0.85
Cost (USD) 50 33,214 32,720 5000
Cost of Equivalent ATT in Terms of Swept Area (USD) e 1500 2450 4250
Cost Ratio e 0.05 0.07 0.85
Cost per W (USD) 2.00 57.66 38.58 2.00
Cost per W Equivalent ATT in Terms of Swept Area (USD) e 2 2 2
References e [22] [23] [24]

Table 7
Comparison of annual energy production of ATT compared to wind turbines available in market.

Annual Energy Production (Wh)

Wind Speeds (m/s) No. of Hours Annually (h) SY-1kW Bergey Excel 10 kW Britwind R9000 ATT

1 467 0 0 933 0
2 860 0 0 859 1222
3 1125 0 114,720 15,745 5397
4 1238 1,238,080 493,994 259,996 14,084
5 1210 3,024,058 1,025,760 696,743 26,876
6 1074 3,759,333 1,621,883 1,185,801 41,238
7 878 4,389,259 2,109,478 1,565,210 53,521
8 665 4,990,297 2,396,673 1,691,378 60,554
9 470 5,874,892 2,383,326 1,574,001 60,901
10 310 4,966,607 2,128,191 1,265,553 55,175
11 192 4,035,194 1,703,044 889,280 45,460
12 112 2,791,944 1,215,612 548,449 34,302
13 61 1,525,534 733,416 309,134 23,829
14 31 784,477 388,943 161,319 15,305
15 15 379,959 189,903 78,150 9117
16 7 173,452 87,045 35,793 5051
17 3 74,669 37,343 15,582 2608
18 1 30,326 15,092 6322 1257
Total Annual Energy Production (kWh) 38,038 16,644 10,300 456
Total Annual Energy Production from Equivalent No. of ATT 38,752 22,339 13,677
in Terms of Swept Area (kWh)

annual energy produced by Bergey Excel, if it is installed in Abu parameters, the operational Reynolds number is considered since
Dhabi, is 16,644 kwh compared to 22,339 kwh produced by ATT the lift and drag coefficients of the airfoils are function of it. Rey-
array that covers the same area swept by Bergey Excel rotor. Also nolds number shows considerable effect on the turbine blades
the annual energy produced by Britwind is 10,300 kwh compared performance. Reynolds number sensitivity analysis shows that the
13,677 kwh produced by equivalent array of ATT. The energy pro- ATT turbine blade airfoil BW3 has the best performance compared
duced by equivalent arrays of 1.34 times of the energy produced by to A18 and SG6043 airfoils. Reynolds number is controlled by the
Bergey Excel and 1.32 times of the energy produced by Britwind. blade average chord length whereas the wind speed of 5 m/s is kept
For SY, the equivalent array can produce 1.02 times he energy constant as major objective for micro turbine blade design.
produced by SY. Wind profile for two sites in UAE: Abu Dhabi and Al Ain are
analyzed as sample sites. Based on the analysis of the wind data of
7. Conclusion Abu Dhabi and Al Ain cities, the rated wind speed of the micro wind
turbine blade is set to be 5 m/s. The accumulated hours of each
New efficient design of micro wind turbine for low wind speed wind speed range for Abu Dhabi site based on Rayleigh PDF are
regions such as Arabian Peninsula, Jordan Desert and UAE is ach- utilized to conduct performance and economic comparison with
ieved. The performance analysis for the theoretical design is vali- available commercial micro wind turbines. The economic feasibility
dated experimentally and both results show very good agreement study illustrates that the payback period for ATT is 1.83 years. The
with each other. The power coefficient for theoretical and experi- concept of replacing commercial wind turbines available in the
mental results are 0.38 and 0.34 respectively. In order to validate market with equivalent array of ATT micro wind turbines in terms
the simulation, a prototype is manufactured using 3D printer. The of swept area is presented. The economic feasibility study shows
ATT prototype blades are tested in open air environment to have that the concept is cost-effective and more energy is produced by
more realistic results. exploiting the same swept area. The results show that ATT micro
Methodology to design the ATT blades that considers Reynolds wind turbine is efficient and economically feasible of producing
Number as one of the design parameter is developed and utilized. energy in low wind speed regions such as Arabian Peninsula and
Along with the Blade element momentum (BEM) theory design Jordan Desert compared to available commercial wind turbines.
 S.N. Akour et al. / Renewable Energy 116 (2018) 215e223 223

References FAST for performing design optimization of offshore wind turbines: FAS-
TLognoter, Renew. Energy 55 (2013) 69e78.
[11] C. Lyon, A. Broeren, P. Giguere, A. Gopalarathnam, M. Selig, Summary of Low-
[1] T.B.M.J. Ouarda, C. Charron, J.-Y. Shin, P.R. Marpu, A.H. Al-Mandoos,
speed Airfoil Data, vol. 3, 1997.
M.H. Al-Tamimi, H. Ghedira, T.N. Al Hosary, Probability distributions of wind
[12] M. Selig, J. Guglielmo, A. Broeren, P. Giguere, Summary of Low-speed Airfoil
speed in the UAE, Energy Convers. Manag. 93 (2015) 414e434.
Data, vol. 1, 1995.
[2] A. Pourrajabian, R. Ebrahimi, M. Mirzaei, Applying micro scales of horizontal
[13] D. Marten, J. Wendler, G. Pechlivanoglou, C. Nayeri, C. Paschereit, QBlade: an
Axis wind turbines for operation in low wind speed regions, Energy Convers.
open source tool for design and simulation of horizontal and vertical Axis
Manag. 35 (2014) 23e30.
wind turbine, Int. J. Emerg. Technol. Adv. Eng. 3 (3) (2013) 264e269.
[3] M.Y. Zakaria, D.A. Pereira, M.R. Hajj, Experimental investigation and perfor-
[14] K. Bassett, R. Carriveau, D.-K. Ting, 3D printed wind turbines Part 1: design
mance modeling of centimeter-scale micro-wind turbine energy harvesters,
considerations and rapid manufacture potential, Sustain. Energy Technol.
J. Wind Eng. Ind. Aerodynamics 147 (2015) 58e65.
Assessments 11 (2015) 186e193.
[4] D. Leung, Y. Deng and M. Leung, “Design optimization of a cost-effective micro
[15] M.R. Islam, R. Saidur, N.A. Rahim, Assessment of wind energy potentiality at
wind turbine,” in World Congress on Engineering, London, 2010.
Kudat and Labuan, Malaysia using Weibull distribution function, Energy 36 (2)
[5] F. Xu, F. Yuan, L. Liu, J. Hu, Y. Qiu, Performance prediction and demonstration
(2011) 985e992.
of a miniature horizontal Axis wind turbine, J. Energy Eng. 139 (3) (2013)
[16] J. Manwell, J. Mcgowan, A. Rogers, Wind Energy Explained: Theory, Design
143e152.
, Wind tunnel testing of a and Application, 2009.
[6] J.P. Monteiro, M.R. Silvestre, H. Piggott, J.C. Andre
[17] AIR-X Land Manual, Arizona, 2002, p. 4.
horizontal Axis wind turbine rotor and comparison with simulations from two
[18] Y. Zhu, M. Cheng, W. Hua, W. Wang, A novel maximum power point tracking
blade element momentum codes, J. Wind Eng. Ind. Aerodynamics 123 (2013)
control for permanent magnet direct drive wind energy conversion systems,
99e106.
Energies 5 (2012) 1398e1412.
[7] M. A. Abrar, A. I. Mahbub and M. Mamun, “Design optimization of a horizontal
[19] H.G. Jeong, R.H. Seung, K.B. Lee, An improved maximum power point tracking
Axis micro wind turbine through development of CFD model and experi-
method for wind power systems, Energies 5 (2012) 1339e1354.
mentation,” in 10th International Conference on Mechanical Engineering,
[20] R. Carriveau (Ed.), Fundamental and Advanced Topics in Wind Power, InTech,
2014.
2011.
[8] Y.-F. Wang, M.-S. Zhan, 3-Dimensional CFD simulation and analysis on per-
[21] CES EduPack, Granta Design Limited, 2015.
formance of a micro-wind turbine resembling Lotus in shape, Energy Build. 65
[22] Evance R9000 UK MCS Certification Summary, Evance Wind.
(2013) 66e74.
[23] SWCC Summary Report.
[9] L. White, S. Wakes, Permitting best use of wind resource for small wind-
[24] 27 June 2016. [Online]. Available from: http://www.alibaba.com/product-
turbines in rural New Zealand: a micro-scale CFD examination, Energy Sus-
detail/horizotal-axis-renewable-energy-1kw-homemade_60432711624.html.
tain. Dev. 21 (2014) 1e6.
[10] J.E. Gutierrez, B. Zamora, J. García, M.R. Peyrau, Tool development based on