Proceedings of the World Congress on Engineering 2010 Vol II

WCE 2010, June 30 - July 2, 2010, London, U.K.

Design Optimization of a Cost-Effective

Micro Wind Turbine
D.Y.C. Leung, Y. Deng, M.K.H. Leung

Abstract- The aim of this paper is to investigate the Various theoretical methods are available to calculate the
performance of a specially designed micro wind turbine for aerodynamic forces on the blades of a horizontal axis wind
urban environment where the wind speed is usually low. turbine (HAWT) [1], such as the Blade Element Momentum
Differing from the traditional wind turbine that can be (BEM) theory [2]-[4], Lifting Line Method (LLM) [5],
connected directly to the grid, the micro wind turbine is linked
to a small generator and mainly used for local applications.
Lifting Surfaces Method (LSM), N-body/particle simulation
The advantage of the micro wind turbine is that, apart from its method, and asymptotic expansion method (Euler special).
low cost, it can be propelled by a wind speed as low as 2 m/s. In Among them, the BEM method is the simplest and most
this paper, the performance of a single micro wind turbine was commonly used. This paper reports the development of a
evaluated through CFD analysis. The CFD simulation shows methodology for evaluating and optimizing the performance
good agreement with experimental results. The validated of a specially-designed low-cost micro wind turbine using a
computer model is then used for optimizing the wind turbine computational fluid dynamics (CFD) technique and verified
blade design with varying blade subtend-angle and blade the result with physical tests conducted in a wind tunnel.
number.

Index Terms - aerodynamics, angular velocity, starting torque, II. METHODOLOGY

wind turbine blade.
A. Wind Turbine Profile

I. INTRODUCTION The wind turbine under investigation is a micro fan-

bladed wind turbine, as shown in Fig 1. Different from
Wind energy is a clean, inexhaustible and sustainable conventional two- or three-blade wind turbines, this micro
energy source. Due to the high market demand, the wind turbine employs a fan-type blade configuration rather
development of wind energy technology has been moving than an aerofoil-type, which has a functional advantage of
very fast in many new dimensions, such as aerodynamics, increased power efficiency [6]. The edgewise view defines
structural mechanics and mechanical engineering. The main the blade thickness distribution over the blade length. Many
trend of wind turbine development is large-scale wind large wind turbines utilize linear taper blades from the root
energy systems (with sizes vary from several hundreds kW to the tip for rigidity [7]. Since the blade of the micro wind
to 10 MW, and rotor diameters over one hundred meters) turbine is not very long, it is designed to be in mono
that are often established along offshore or on vast wind thickness along the blade length. The twist extent of the
sites where annual average wind speed is high. turbine blade is clearly displayed in the transaction view.
On the other hand, a new branch of development in this Most wind turbines use twisted blades to capture a more
field recently emerged. In regions of low wind speed and in efficient torque in different wind conditions, and the micro
crowded urban areas, miniature wind machines or micro wind turbine in this study is no exception. The twist angle of
wind turbines are more suitable. A small wind turbine, with the micro wind turbine is a critical parameter for the present
a rotor diameter as small as a meter or less, can often be set computation and optimization work, and it has a strong
up and stand alone on the roof of houses and buildings for relationship with the blade subtend-angle of the micro wind
electricity generation. This kind of wind energy converter is turbine. Important geometric parameters of a typical micro
normally directly linked to battery rather than connected to wind turbine under investigation are shown in the
the electric grid. As its capacity is small (usually <1kW), its nomenclature.
prime cost is not very high and is affordable for many
household applications. This micro wind turbine has
received attention in recent years and a great deal of
research has been conducted to optimize their performance,
particularly at the low wind speed range.
Manuscript received 2 March 2010. This work is supported by ICEE
and the UDF of Clean Energy of the University of Hong Kong.
D.Y.C. Leung is with the Department of Mechanical Engineering, the
University of Hong Kong, Hong Kong. (Phone: 852 2858 7911, Fax: 852
2858 5415, Email: ycleung@hku.hk).
Y. Deng was with the Department of Mechanical Engineering, the
University of Hong Kong, Hong Kong. He is now with Ove Arup &
Partners Hong Kong Ltd..
M.K.H. Leung is with the Department of Mechanical Engineering, the
University of Hong Kong, Hong Kong. Fig. 1. Micro wind turbine under studying.

ISBN: 978-988-18210-7-2 WCE 2010

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2010 Vol II
WCE 2010, June 30 - July 2, 2010, London, U.K.

The advantage of this micro fan-bladed wind turbine

design is that multiple turbines can be connected together to
add up the power to meet any requirements in a flexible
manner. Such blade and the turbine designs can be produced
by injection molding for mass production; therefore the cost
of the system is only about one third of conventional wind
turbine system designs.

B. Governing Equations

Fig. 2 shows the localized forces acting on a stationary

wind turbine blade. In this figure, Φ, the local angle between
the plane of rotation and the relative wind speed, is defined
as the inflow angle. When the blade is stationary, the inflow
angle Φ and the wind attack angle α reach their maxima and
receive the maximal force F from the wind. The force F on Fig. 2 Aerodynamics on a blade of the static micro wind
the blade caused by the wind can be resolved into Fdrag in the turbine.
axial direction and Ftorque in the tangential direction.
Fig. 3 illustrates the detailed aerodynamic analysis of the
turbine blade. According to the momentum equation, the
force acting on a finite element of the blade dS is:

dF S = dm( U rel sin   U windpass sin  ) (1)

dm = ρ air . U rel . dS (2)
Combining (1) and (2),
2
dF S = ρ air (U rel - U rel ×U windpass ) sinτ . dS (3)

The axial force dFS(drag) and the tangential force dFS(torque),

acting on a finite element of the blade dS, are dFs sinτ and
dFscosτ, respectiverly. By summing all the axial forces
dFS(drag) and tangential forces dFS(torque) of these finite
elements, the total FS(drag) and total FS(torque) acting on the
crossing-section of the blade can be calculated.
F S (drag ) =  dF S
S (drag) (4) Fig. 3. Aerodynamics analysis

F S (torque ) =  dF S (torque) (5)

S
Furthermore, the axial force FB(drag) and the torque TB The wind turbine starts to rotate when the torque T is
acting on a single turbine blade can be worked out by large enough to overcome the static equilibrium,. An
integrating all the axial forces and torques over the rotor aerodynamic analysis of the rotating wind turbine is
blade length as follows: illustrated in Fig. 4. Compared to that of the stationary one
r  R tip in Fig. 2, an additional vector ωR acts on the turbine blade.
This is because the wind turbine is rotating with an angular
F B (drag ) = F
r  R root
S (drag) dr (6)
velocity ω at an opposite direction. As a result, the force
acting on the turbine blade is not from the wind speed U, but
r  Rtip
the relative wind speed Urel that is composed of the wind
TB = F
r  Rroot
s(torque) rdr (7) speed vector and the local speed vector of the blade.

C. CFD Simulation
Finally, the overall drag force Fdrag and torque T acting on
the whole wind turbine can be determined by multiplying The CFD simulation was operated using GAMBIT 2.1
the axial force FB(drag) and the torque TB acting on a single and Fluent 6.1. In the present case, the radius and the width
turbine blade with the number of blades on the wind turbine. of the micro wind turbine are 117 mm and 60 mm,
respectively. A 2500mm × 2500mm × 3000mm volume was
F drag = N B . F B (drag ) (8)
created as the computational domain with its six faces
r  Rtip
treated to be the boundaries of the domain. A uniform wind
= NB .
 (U rel - U rel . U windpass ) sin 2 .dSdr
2
s
air speed profile was introduced at the entrance of the domain.
r  Rroot The micro wind turbine was set at the centre of the domain,
T = NB . T B (9) which, together with the air inside and around it, was named
r  Rtip as the moving-zone and was set as rotational objects in
= NB . boundary conditions. The boundaries between the moving-
  [ (U rel - U rel . U windpass ) sin .cos .dS]rdr
2
air
r  Rroot
s
zone and the rest of the domain are set as the interior.

ISBN: 978-988-18210-7-2 WCE 2010

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2010 Vol II
WCE 2010, June 30 - July 2, 2010, London, U.K.

Considering that the target of this research was the Table 1 Computation cases and important parameters.
performance of the wind turbine, very fine grids were used
to mesh the region around the wind turbine, where accurate Blade Blade Blade Blade Solidity
results were demanded. Tetrahedral grid elements were Subtend- Pitch Twist No. Σ (%)
introduced to discretize the domain and define the spatial Angle Angle θ Angle NB
resolution of the numerical solution. Triangular cells were α (tip~root) Β
also applied to mesh the boundary layers and inner faces. 30° 46.22°~ 21.16° 12~3 78.4~
The total number of grids was about 1.5 million. The Semi- 67.38° 19.6
Implicit Method for Pressure-Link Equations (SIMPLE) 40° 39.07°~ 22.76° 9~3 78.4~
algorithm and the standard k   turbulence model were 61.82° 26.1
used to simulate the incompressible, steady-state turbulent 45° 36.42°~ 23.07° 8~3 78.4~
flow. 59.49° 29.4
The numerical simulations were conducted to model the 60° 31.07°~ 23.12° 6~3 78.4~
performance of the micro wind turbine under different wind 54.18° 39.2
speeds with the aerodynamics equations stated in Section 72° 28.75°~ 22.85° 5~3 78.4~
IIB. 51.60° 47.0
80° 27.91°~ 22.71° 4~3 69.6~
D. Model Validation 50.63° 52.2
90° 27.55°~ 22.64° 4~3 78.4~
An experiment was conducted in a wind tunnel to 50.19° 58.8
validate the CFD model used. A photo-sensor and a torque- 100° 27.91°~ 22.71° 3 65.3
meter were used to determine the angular velocity of the 50.63°
wind turbine and the torque acting it as shown in Fig. 4. 110° 29.04°~ 22.90° 3 71.8
51.94°
120° 31.07°~ 23.12° 3 78.4
54.18°

III. RESULTS

A. Modeling and Experimental Results

Fig. 5 shows the comparison of the CFD and

experimental results on the maximal angular velocity of
the wind turbine at different wind speeds. Without any
loading, the micro wind turbine rotates freely and the
angular velocity of it increased linearly with increasing
wind speed in both cases. As illustrated in this figure, the
Fig. 4. Torque measurement. experimental results match well with the modeling results.

E. Optimization

The optimization study was conducted by varying the

blade subtend-angle of the micro wind turbine and the
number of turbine blades. Table 1 shows all the simulation
cases and some important parameters in the optimization
work. These cases were simulated by CFD to obtain the
following parameters:
- maximal angular velocity at different wind speeds;
- drag force and torque when stationary and rotating at a
certain angular velocity under different wind speeds;
- mechanical energy captured by the micro wind turbine at
different wind speeds.

For simplicity only a few cases will be presented here and

discussed in the following sections. More information and
simulation results can be found elsewhere [8].

Fig. 5. Maximal angular velocity Vs wind speed.

ISBN: 978-988-18210-7-2 WCE 2010

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2010 Vol II
WCE 2010, June 30 - July 2, 2010, London, U.K.

(a)

Fig 7. Cp- λ characteristic of a micro wind turbine.

(b)
Fig. 7 shows the computed relationship between the
power coefficient (Cp) and the tip speed ratio (λ) of the
micro wind turbine. It is recognized that small-scale multi-
bladed wind turbines normally operate at a tip speed ratio
between 0 to 2 while the large-scale one with two or three
blades operates at a tip speed ratio higher than 4 [9]. As
indicated, the tip speed ratio of the present micro wind
turbine is between 0 and 1, which meets closely with the
traditional, small, multi-bladed wind turbine. Besides, the
maximal power coefficient of the micro wind turbine
indicates that the efficiency of the transformation from
kinetic wind energy to mechanical energy is about 12%.

B. Optimization Results

To compare the performance of the micro wind turbine

with different blade profiles, they were categorized into
several series according to their blade subtend-angle. For
Fig. 6 Torque and power Vs angular velocity of the wind each series, the blade number varied from three to a number
turbine at a wind speed of 7 m/s. (a) Torque; (b) Power. which the blade plane of the turbine is fully-occupied.
For the blades with 30-degree subtend-angle, the blade
Fig. 6(a) presents the comparison of the CFD and number varied from three to twelve. No significant
experimental results on the relationship between the torque difference in the maximal power output can be observed for
and the angular velocity of the wind turbine at a medium the micro wind turbines with eight or more blades as shown
wind speed (7 m/s). It can be clearly seen that the in Fig. 8, while their maximal power outputs are much
experimental data match quite well again with the numerical higher than those turbines with fewer blades. The optimal
analysis and they both demonstrated that the torque power coefficient of the turbines with a 30-degree blade
decreased with increasing rotational velocity of the wind subtend-angle is about 12.5% while the optimal tip ratio is
turbine. The relationship between the torque and maximal 0.5 to 0.6 for the 8-bladed to 12-bladed profile. In addition,
angular velocity is not strictly linear. Other important the maximal tip speed ratio of the micro wind turbines with
parameters affecting the performance are the power output different blade numbers is about the same.
and the power coefficient of the wind turbine. The power To determine which of these wind turbine blade designs is
output can be worked out by multiplying the angular optimal, the starting torque of the wind turbine needs to be
velocity of the turbine with the torque captured by the considered. The larger the torque developed, the easier the
turbine when it was rotating at that angular velocity. Fig. 6(b) wind turbine to overcome its static equilibrium. It is found
shows the relationship between the power output and the from the experimental result that no obvious difference in
angular velocity of the wind turbine at the same wind speed. the torque of the micro wind turbines with eight or more
The curves indicate that the mechanical power output of the blades. On the other hand, a micro wind turbine rotor with
turbine increased with increasing angular velocity, reaching fewer blades captures a smaller torque, which is not
its maximum at the optimal angular velocity (optimal favorable for energy conversion.
velocity), and then decreased thereafter.

ISBN: 978-988-18210-7-2 WCE 2010

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2010 Vol II
WCE 2010, June 30 - July 2, 2010, London, U.K.

Fig. Fig. 9. Relationship between maximal power coefficient and

Fig. 8 Power coefficient of the micro wind turbine with solidity of wind turbine.
different number of blades with 30-degree blade subtend-
angle. Fig. 9 shows the relationship between the maximal power
coefficient and the solidity of the micro wind turbine. The
The results of other degree subtend-angle can be found figure illustrates that the maximal power coefficient of the
elsewhere [8]. In general, for a given blade subtend-angle, turbine rises with a rising solidity in low-solidity range, and
more blades yield a better performance. However, a fully- then becomes fairly stable at around 0.2, after the solidity is
occupied rotor plane is not beneficial to both power output higher than 0.5. In other words, it is better to select a micro
and starting torque of a micro wind turbine. Moreover, a wind turbine profile whose solidity is higher than 0.5 to
blade with a subtend-angle larger than 90-degrees is not receive a higher power output.
recommended design due to its poor starting performance. According to the results listed in Table 2, the 4-bladed
micro wind turbine with 80-degree blade subtend-angle has
the highest power coefficient. However, this is not the
IV. DISCUSSION optimal profile for the micro wind turbine due to its
comparatively weak starting effect. Fig. 10 shows the torque
Table 2 lists the solidity of the micro wind turbines and produced by those high-efficiency wind turbine profiles
their maximal power coefficient in the optimization analysis. under stationary condition. Among these rotors whose
These data indicate that turbines with high-solidity have, in power coefficients are higher than 0.18, the 5-bladed rotor
general, higher power coefficients than that of low-solidity. with 60-degree blade subtend-angle is considered to be the
For the low-solidity turbines, the power coefficient increases optimal micro wind turbine profile. Compared with that of
with increasing blade subtend-angle for certain solidity. For original micro wind turbine profile (30-degree blade
high-solidity rotors, however, the maximal power coefficient subtend-angle, 8-bladed), the maximal power coefficient of
occurs at a certain blade subtend-angle. the 5-bladed rotor with 60-degree blade subtend-angle raises
from 12.5% to 19.3%.
Table 2. Solidity of the micro wind turbines and their
maximal power coefficient.

Solidity Blade Blade Maximal Power

Subtend-Angle No. Coefficient
19.6% 30° 3 0.073
26.1% 40° 3 0.103
29.4% 45° 3 0.120
32.6% 30° 5 0.105
34.8% 40° 4 0.131
39.2% 45° 4 0.149
43.5% 40° 5 0.145
47.0% 72° 3 0.170
49.0% 45° 5 0.163
52.2% 80° 3 0.188
58.8% 90° 3 0.186
62.7% 72° 4 0.189
65.3% 60° 5 0.193
69.6% 80° 4 0.203
71.8% 110° 3 0.185 Fig. 10. Starting torque of several high-power
78.4% 72° 5 0.191 coefficient micro wind turbines.

ISBN: 978-988-18210-7-2 WCE 2010

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2010 Vol II
WCE 2010, June 30 - July 2, 2010, London, U.K.

V. CONCLUSION z : width of blade area (present design =60mm)

ρair : air density
This study investigates the variation of the performance N B  As
of micro wind turbine with different design parameters. The σ : solidity of the wind turbine (= )
results showed that the performances of high-solidity wind Rturbine 2
rotors are better than those of low-solidity ones. However, (present design =52.2%)
for turbines with identical blade subtend-angles, one with a ω : angular velocity of the wind turbine
fully-occupied rotor may not be the best profile, since its λ : tip speed ratio of the wind turbine
blades block the wind acting on neighboring blades. The α : blade subtend-angle (present design =30o)
optimization results also show that the preferable solidity of α' : attack angle
the micro wind turbine is higher than 50%. β : twist angle of the turbine blade (present design =21.2o)
From the optimization analysis, it is known that rotors Φ : inflow angle
with larger number of blade are able to produce higherτ τ : the local angle between the plane of rotation and the
torque when they are stationary. As a result, a multi-blade resultant force F S
approach is preferable for a micro scale wind turbine system.
Considering the power coefficient and the starting effect, theθ :  : local angle between the chord line and the rotation
5-bladed micro wind turbine with 60-degree blade subtend- plane
angle is the optimal turbine profile. Its maximal power
coefficient is much higher than that of the preliminary
turbine design (8-bladed rotor with 30-degree blade subtend- REFERENCES
angle) and its higher power coefficient range is much wider.
[1] F. Wang, L. Bai, J. Fletcher, J. Whiteford, D. Cullen,
“The methodology for aerodynamic study on a small
APPENDIX: NOMENCLATURE domestic wind turbine with scoop.” J. Wind Eng. Ind.
Aerodyn., doi: 10.1016/j. jweia. 2007. 03. 004.
As : area of turbine blade projected on the cross-section [2] R. Lanzafame, M. Messina, “Fluid dynamics wind
Cp : power coefficient of wind turbine turbine design: Critical analysis, optimization and
dS : a finite element of the turbine blade application of BEM theory.” Renewable Energy 32,
D : diameter of the belt pulley 2007, 2291-2305.
F :differential frictional force between the two readings of [3] A. Varol, C. İlkılıç, Y. Varol, “Increasing the efficiency
the spring balances of wind turbines.” Journal of Wind Engineering and
FB(drag) : axial force acting on a single turbine blade Industrial Aerodynamics 89, 2001, 809-815.
Fdrag : drag force acting on the whole wind turbine [4] M. Jureczko, M. Pawlak, A. Mężyk, “Optimization of
FS(drag) : axial force acting on the finite element dS wind turbine blades.” Journal of Materials Processing
FS(torque) : tangential force (contributes to torque) acting on Technology 167, 2005, 463-471.
the finite element dS [5] M.M. Duquette, K.D. Visser, “Numerical Implications
 of Solidity and Blade Number on Rotor Performance of
m : mass flow rate (kg/s) Horizontal-Axis Wind Turbines.” J. Solar Energy
N B : number of turbine blades (present design=8) Engineering, 125, 2003, 425-432.
[6] H. Hirahara, M.Z. Hossain, M. Kawahashi,Y. Nonomura,
Poutput : extracted power output of the wind turbine “Testing basic performance of a very small wind turbine
r : local radius designed for multi-purposes.” Renewable Energy 30,
Rroot : radius of root circle of the turbine blade (present 2005, 1279-1297.
design=50mm) [7] S.M. Habali, I.A. Saleh, “Local design, testing and
Rtip : radius of root circle of the turbine blade (present manufacturing of small mixed airfoil wind turbine
design =115mm)) blades of glass fiber reinforced plastics Part I: Design of
Rturbine: radius of the original micro wind turbine (present the blade and root.” Energy Conversion & Management
design =117mm) 41, 2000, 249-280.
T : torque acting on the whole turbine [8] Y. Deng , Design Optimization of a Micro Wind Turbine
TB : tangential force acting on a single turbine blade using Computational Fluid Dynamics. MPhil Thesis,
T ω : torque acting on the belt pulley The University of Hong Kong, 2008;150pp.
U : upstream undisturbed wind speed [9] G.L. Johnson, Wind Energy Systems, Prentice-Hall,
U rel : upstream relative wind speed Englewood Cliffs, New Jersey 1985.
Uwindpass : downstream wind speed

ISBN: 978-988-18210-7-2 WCE 2010

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)