Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
art ic l e i nf o a b s t r a c t
Article history: A new approach has been developed for the optimization of Vertical Axis Wind Turbines (VAWTs). The
Received 16 May 2015 approach is derived from double multiple streamtube theory (DMST) and adopts the concept of
Received in revised form Representative streamtube wherein the entire aerodynamic property of the VAWT is assumed to be
14 March 2016
represented by a single streamtube occupying a specic azimuthal location. Five input parameters
Accepted 14 March 2016
Available online 31 March 2016
namely; power, wind velocity, aspect ratio, air viscosity and air density and six output parameters that are
minimally required for the construction of a straight bladed, constant pitch VAWT are considered in the
Keywords: study. The discrete data of lift and drag coefcients pertaining to specic values of Reynolds number and
Vertical Axis Wind Turbine angle of attack are arranged in the order of decreasing lift to drag ratio and starting with the rst set of
Analysis
coefcients, a check is made if it is eligible to become Representative streamtube for the particular
Renewable energy
problem. The check is done through a tri-directional Demand Factor test that seeks the compatibility of
Double multiple streamtube theory
Optimization Reynolds number and angle of attack values of the aerofoil data under consideration. The rst set which
gains the eligibility to become Representative streamtube marks the solution for the particular problem.
& 2016 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jweia.2016.03.006
0167-6105/& 2016 Elsevier Ltd. All rights reserved.
M. Abdul Akbar, V. Mustafa / J. Wind Eng. Ind. Aerodyn. 153 (2016) 3445 35
seperately and determining the optimum combination based on the are depicted. Additional features such as connecting struts, sub-
results of all permutations and combinations. structure and machine components are not shown for clarity.
The input and output parameters in the optimization of VAWT During operation, the VAWT is subjected to two types of forces
that are considered in this study are listed in Table 2. The input simultaneously. They are:
parameters listed in the table covers most of the factors that are
typically considered in VAWT design and the output parameters 1) Wind force from an arbitrary direction.
include all the parameters that are minimally required to com- 2) The angular rotation of VAWT about its central pole.
pletely manufacture a simple and straight bladed VAWT. Extra
parameters may be required to dene VAWTs with pitched blade Due to the omni-directionality of VAWT rotor, regardless of the
and curved shapes, which are not considered in this study. The wind direction, a symmetric face is always presented to receive
aerofoil shapes are classied into various families like NACA, Selig, the wind without the need for a yawing mechanism which is a
Eppler, Selig/Donovan, Miley, Wortmann, Clark and Gottingen. feature in Horizontal axis wind turbines (HAWTs). Consider a
There are also stand-alone types like Jacobs USNPS4, Bergey BW-3, close-up view of the VAWT geometry shown in Fig. 4. The net
etc. The output parameters listed in the table are depicted in Fig. 2. effect of two forces acting together can be obtained by considering
The optimization process developed through this paper adopts a the resultant of the forces. The angle at which the resultant force
simplistic model of maximization of lift to drag ratio which is a direct hits the blade will also be modied due to the combined action of
measure of increase in the efciency of VAWT. The methodology uses
the forces. Fig. 5 shows a scaled up image of the wind hitting the
the concept of effective lift to drag ratio discussed in (Soraghan et al.,
rotating VAWT wherein, the actual wind velocity, V 1 gets mod-
2013). Through the effective lift to drag ratio, the entire aerodynamic
ied to V a due to the actuator disk effect which further gets
performance of the VAWT is represented by the azimuthal location
modied to V R due to the rotating effect of the VAWT.
of the streamtube that provide the value of torque equal to average
From basic trigonometry using Fig. 5, the angle at which the wind
instantaneous torque for the VAWT. The ratio of lift and drag coef-
hits the rotating VAWT denoted as angle of attack () is given by,
cient corresponding to that particular streamtube is called effective
lift to drag ratio. For convenience, the azimuthal location that pro-
vides the effective lift to drag ratio is referred to as Representative sin V a sin =V R 1
streamtube in this paper. The optimization of the VAWT on the basis Here, V a is the modied wind velocity due to the actuator disk
of Representative streamtube provides a great advantage over the effect of the wind hitting the rotating VAWT. Hence, out of the
regular approach that it obviates the need of carrying out iterations actual wind velocity, V 1 only a portion of the same would be
for all azimuthal locations. effective in generating wind energy. The relationship between
these velocities is denoted by a dimensionless factor termed,
2. Basic formulations
induction factor (a) given by,
Consider the plan view of a 3-bladed VAWT as shown in Fig. 3.
Only parameters that are relevant to the discussion in this section a 1 V a =V 1 2
36 M. Abdul Akbar, V. Mustafa / J. Wind Eng. Ind. Aerodyn. 153 (2016) 3445
Fig. 1. Evolution of stream tube theory: (a) Single streamtube theory, (b) Multiple streamtube theory, and (c) Double multiple streamtube theory.
Table 1
Optimization studies carried out on VAWTs using streamtube theories.
Aspect ratio, solidity, Reynolds number of operation, tip speed ratio (TSR) MST Sullivan, 1979
Optimal performance analysis in the absence of blade drag Newman, 1983
Rotor efciency, solidity, interference coefcients Loth and McCoy, 1983
Optimum value of inow factor Newman, 1986
Symmetric NACA aerofoils DMST Solum et al., 2006
Camber and aerofoil thickness DMST Claessens, 2006
Optimal positioning of the blades to maximize efciency of VGOT (variable geometry oval trajectory) DMST Ponta et al., 2007; Ponta and Lago, 2008
Optimum TSR and number of blades for a Zephyr VAWT MST Ajedegba, 2008
Discusses optimal features with regards to control parameters DMST Farthing, 2009
Optimization model for Darrieus Straight bladed VAWT based on analysis conducted using DMST DMST Ming, 2009
Solidity, TSR, blade pitch angle DMST Asher et al., 2010
Rotor area, strut design, solidity, TSR, number of blades, blade thickness, point of attachment, pitch angle, blade DMST Deglaire, 2010
camber
Blade prole, rotor solidity, Reynolds number, aspect ratio MST Samaraweera et al., 2010
Twisting of blades of an innovative VAWT DMST Sharpe and Proven, 2010
Optimization of pitch angle DMST Okpue, 2011
Optimum positioning of blades, optimum solidity, angle of attack and blade pitch of VGOT DMST Ponta et al., 2011
Aerofoil shape, TSR, radius, chord length, number of blades DMST Castillo, 2011
Camber and pitching angle of wing prole MST Bitsch et al., 2011
Optimization tool for new rotor architectures DMST Castelli et al., 2011
Radius of the rotor, number of blades, chord length, blade height DMST Tang et al., 2011
Tip speed ratio DMST Helms, 2011
Tip speed ratio MST McGowan et al., 2012
Shape optimization of a novel offshore VAWT concept called NOVA DMST Shires, 2013a
Pitch angle DMST Rathi, 2012
Optimization of initial angle of individual blades for a four bladed VAWT SST Diaz and Pinto, 2012
Optimize overall dimensions Islam et al., 2012
Optimal placement in wind farms SST Chen and Agarwal, 2012
Optimal values of Masse coefcient used in calculating dynamic stall based aerodynamic coefcients DMST Shires, 2013b
Solidity, aspect ratio MST Komerath et al., 2013
Effect of variation of ns on a savonius rotor to get the most optimized conguration SST Hasan et al., 2013
Chord optimization along the blade span DMST Bedon et al., 2014
Aspect ratio, optimal strut spacing DMST Clarke et al., 2014
Rotor solidity factor, blade aspect ratio, rotor aspect ratio DMST Ahmadi-Baloutaki et al., 2014
Chord to diameter ratio, height to diameter ratio, solidity, aspect ratio, height, diameter, chord length DMST Bianchini et al., 2015
Aerofoil shapes, number of blades, chord length DMST Bedon et al., 2015
Aerofoil thickness DMST Larrea, 2015
Pitch angle DMST Wille, 2015
From Pythagoras theorem, the relative velocity (V R ) can be the rotor. Substituting Eq. (3) in (1) and using the expressions of
written from Fig. 5 as, Eq. (2), we get Eq. (4) as,
q q
2 2 2 2
VR V a sin V a cos R 3 sin 1 a sin = 1 a sin 1 a cos 4
Here, is the angular velocity of the VAWT which varies Here, is the azimuthal location or the instantaneous position
instantly with the change in wind velocity, and R is the radius of of the streamtube under consideration. Similarly, expressing angle
M. Abdul Akbar, V. Mustafa / J. Wind Eng. Ind. Aerodyn. 153 (2016) 3445 37
Table 2
Input and output parameters in optimization process.
of attack in terms of cosine of the angle, we get Fig. 5. Wind hitting the VAWT.
q
cos 1 a cos = 1 a sin 2 1 a cos 2 5 The relative velocity ratio can be written by dividing Eq. (3) by
actual wind velocity, V 1 and using the expression of Eq. (2) as,
Dividing Eq. (4) by Eq. (5), we get a simpler expression for angle q
2 2
of attack as, V R =V 1 1 a sin 1 a cos 8
tan 1 a sin = 1 a cos 6 Betz thrust coefcient, C T for an ideal wind turbine is given by,
In Eqs. (4)(6), is the Tip speed ratio (TSR) and is dened as C T 4a1 a 9
Eq. (10). instantaneous torque for the VAWT. Therefore, for Representative
2 streamtube, Eq. (13) can be re-written without the summation
4a1 a NC= R V R =V 1 C t cos = sin C n 10
(independent of the azimuthal location) as,
Eq. (10) which is an implicit equation is solved through trial 2
C Q NC=2R V R =V 1 C t 20
and error procedure for each of the streamtubes to determine its
induction factor. The normal coefcient C n and tangential coef- The coefcient of power of Representative streamtube can be
cient C t can be calculated from lift (C L and drag coefcients written using Eqs. (14) and (19) as,
C D based on the geometry of Fig. 4 as, 2
C P NC=2R V R =V 1 C t 21
C n C L cos C D sin 11
Considering Eq. (10) for Representative streamtube and divid-
C t C L sin C D cos 12 ing it by Eq. (21) and re-arranging the terms using Eqs. (11) and
The torque coefcient C Q for the VAWT is obtained after (12), we get
!
summation of the contributions of all the streamtubes (Eq. (13)) 2C P C L =C D cos sin
and coefcient of power C P ) is obtained after the multiplication of 4a1 a cos = sin 22
C L =C D sin cos
torque coefcient by TSR (Eq. (14)),
" # For a 40.4, Eq. (22) can be modied using a correction to Eq.
XNs VR 2
C Q NC=2R 1=N s Ct 13 (9) originally proposed by Glauert in 1926 and widely used
i1 V1
i thereafter (Pratumnopharat and Leung, 2011). The resulting
expression is given by Eq. (23).
CP CQ 14
0:889 0:444a 1:556a2
The coefcient of power (C P ) is a direct measure of the ef- !
2C P C L =C D cos sin
ciency of VAWT in producing power. The total power generated is cos = sin 23
given by Eq. (15). C L =C D sin cos
P C P RLV 31 15 Considering Eq. (21), it can be seen that using the Eq. (18) for
Sol, and using the expression for C t presented in Eq. (12) and
Reynolds number (Re ), which is necessary for the determina-
further using the expression for relative velocity ratio of Eq. (8);
tion of the lift and drag coefcients is calculated for the stream-
the entire expression of coefcient of power (C P for Representa-
tube under consideration using Eq. (16) which uses the value of
tive streamtube can be expressed as,
relative velocity, V R obtained through Eq. (3).
h 2 i 2
Re V R C= 16 C P Sol 1 a sin 1 a cos C L sin C D cos =2
Upper bound: - A maximum coefcient of power of 0.47 was C pmax CpL LD LDL CpU CpL =LDU LDL 26
attained for an effective lift to drag ratio of 140 at an optimal tip
speed ratio of 4.9. The key aspect of the approach developed in this paper is to
determine the highest lift to drag ratio (along with its associated
Based on these values, a linear proportionality diagram was properties) that is eligible to become Representative streamtube
created for calculating the optimal tip speed ratio and maximum for the problem under consideration. The maximization of effec-
coefcient of power for a given effective lift to drag ratio (Fig. 6). tive lift to drag ratio is the optimization criteria as it is a direct
Here, opt optimum tip speed ratio for the given effective lift measure of efciency of the VAWT (Eq. (24)). Through this paper, it
to drag ratio (LD). is attempted to t a lift to drag ratio (along with its associated
properties) to be Representative streamtube for the given problem,
U uper bound of tip speed ratio4.9. an approach that is not seen presented so far by works of
L lower bound of tip speed ratio 3.85. literature.
CPmax maximum coefcient of power for the given effective lift It is known that a set of lift and drag values is dependent on
to drag ratio (LD). three factors viz: aerofoil shape, Reynolds number and angle of
CPU upper bound of maximum coefcient of power 0.47. attack. The compatibility of a given lift to drag ratio for a particular
CPL lower bound of maximum coefcient of power0.15. problem should satisfy the local compatibility of these three
LD effective lift to drag ratio at which calculation is performed. parameters. Based on logical reasoning, it is concluded that a
LDU upper bound of effective lift to drag ratio 140. factor based approach for ensuring compatibility should fulll the
LDL lower bound of effective lift to drag ratio10. following basic criteria's:
The optimal value of solidity is constant throughout. For the 1) The factor should provide for a tri-directional testing as there
purposes of this paper, the intermediate optimal values of tip are three criteria's that needs to be compatible (for a given set of
speed ratio and solidity is assumed to be obtained through linear lift and drag values) viz: the aerofoil shape, Reynolds number
interpolation of the effective lift to drag ratio at the specic point and angle of attack.
at which analysis is carried out (Eqs. (25) and (26)). 2) The equivalence of the factor written in three forms should
include at-least one factor in terms of the Reynolds number and
opt L LD LDL U L =LDU LDL 25 at-least one factor in terms of angle of attack as these variables
Linear
interpolation of
coefficient of
opt LD CPmax power based on
intermediate
values
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9
diagram ( = 0.9 to 1.5). diagram ( = 1.6 to 2.2). diagram ( = 2.3 to 2.9).
3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8
diagram ( = 3.0 to 3.6). diagram ( = 3.7 to 4.2). diagram ( = 4.3 to 4.8).
were listed in increments of one (e.g. 10, 11, 12, 13, .). They were induction factor and were discarded. Those combinations which
then combined with azimuthal location of 90 to 270 in yielded an unrealistic value of induction factor (values greater
increments of one degree (e.g. 90, 89, 88, 87, .). The than one) were also discarded. The signicance of this exercise
resulting set of 47,160 combinations (131 multiplied by 360) of was to nd out whether the particular combination of lift to drag
azimuthal location and lift to drag ratio were substituted in Eqs. ratio and azimuthal location is eligible to become Representative
(22) or (23) (depending upon the value of induction factor) to see
streamtube. If it is a possibility, then the value of induction factor
if there is a realistic value of induction factor that could sustain
obtained is the one that will sustain the particular value of lift to
each of those combinations. In those equations, the values of and
drag ratio at that particular azimuthal location. On the contrary,
CPmax used are those obtained using Eqs. (25) and (26) for the
given value of lift to drag ratio. the combinations of lift to drag ratio and azimuthal location for
The combinations of lift to drag ratio and azimuthal locations which a converged/realistic value of induction factor was not
which yielded a converged and realistic value of induction factor obtained means that the particular value of lift to drag ratio could
were retained separately. Some combinations of lift to drag ratio never be Representative streamtube at that particular azimuthal
and azimuthal location did not yield a converged value of location.
42 M. Abdul Akbar, V. Mustafa / J. Wind Eng. Ind. Aerodyn. 153 (2016) 3445
Aero-Induction demand factor given by Eq. (28) was calculated DF A value closest to the rounded off value was retained and others
for the combinations of lift to drag ratio and azimuthal locations were discarded.
which yielded a converged and realistic value of induction factor. Table 3 depicts this situation where it can be seen that a total of
The value of DF A was rounded off to the nearest 10th of a decimal. nine different azimuthal locations provide a DF A of 2.3 (after
From the resulting set, it was found out that for certain lift to drag rounding off to the nearest 10th of a decimal) for a lift to drag ratio
of 127. In the table, intermediate calculations are not shown for
ratios, multiple azimuthal locations gave the same value of
clarity of viewing. Out of these cases, the difference between the
demand factor. Out of those combinations, the case that led to a
actual DF A and rounded off DF A was least (0.0037) for the azi-
muthal location of 208 (highlighted in grey shade) and was
selected out of all the nine combinations. Similar exercise was
performed for all combinations of lift to drag ratio and DF A which
140 had multiple azimuthal locations. After this process of screening, a
No. of compatible combinations
120 total of 3309 values remained and they are plotted in Fig. 9(a)(g).
100 This is a polar plot with the radial direction representing the lift to
drag ratio (10140) and the angular location representing the
80
actual azimuthal location ( 90 to 270) plotted on plan (view) of
60
the VAWT. Each set of plot corresponds to a particular value of
40 DF A . A legend for interpreting the plots is given in Fig. 8.
20 As evident from the gures, the values follow a denite pattern
0 as they increase from the minimum value to the maximum value.
0.8 1.8 2.8 3.8 4.8 Fig. 10 shows the plot of compatible combinations versus Aero-
Aero-Induction demand factor Induction demand factor for each combination of azimuthal loca-
Fig. 10. Plot of number of compatible combinations for each value of DF A . tion and lift to drag ratio (subjected to a maximum value of 131,
which is obtained as the number of increments in one from 10 to
140). DF A less than 0.9 and greater than 5.4 did not provide
Table 4 compatible combinations of lift to drag ratio and azimuthal loca-
Permissible lift to drag ratios for different values of DF A . tion for any of the combinations (and thus were not included in
0.9 3.1 1011 128140 15 4.4 5
the gure). For DF A values between 1.4 and 2.8, all the lift to drag
1014 1020 1531 3.7 18 1832 5961 ratio values between 10 and 140 have a chance of becoming
1 2346 36 10 2327 3941 8488 Representative streamtube.
1031 4856 3946 1617 2939 6163 5.1 Table 4 shows the values of lift to drag ratios that are admis-
1.1 59140 5155 2124 5356 4.5 6264
sible for various values of DF A . In this table, the DF A values are
1054 3.2 5960 2638 4.1 2846 8990
1.2 1033 7281 4852 16 6465 5.2 marked in bold. As evident from the table, they contain the entire
1083 3544 118140 57 19 4.6 6566 range of DF A values from 0.9 to 5.4. Under each value of DF A , the
1.3 4776 3.5 3.8 2834 3644 91 values of lift to drag ratios that provided a converged solution out
10122 92140 1018 1018 4048 4758 5.3
of the range considered (10140) are noted down. For instance,
1.42.8 3.3 2026 2330 5758 4.7 6768
All 1023 3239 3946 4.2 4449 5.4 under the DF A value of 3.1, the following ranges have been listed:
2.9 2538 4752 5356 3336 5969 69 1020, 2346, 4856 and 59140. It means that lift to drag ratios
1027 40 5658 120140 4955 4.8 of 21, 22, 47, 57 and 58 which are not part of the listing are not
29140 4450 82127 3.9 59 50 54 admissible for the DF A of 3.1. The table should be read as con-
3 5458 3.6 1331 4.3 7077
1024 62 1428 4752 1019 4.9
tinuous columns. For instance, the second column ends with the
2765 77140 4047 57 3638 5558 DF A value of 3.4 and the permissible lift to drag ratio under it
68140 3.4 5356 4 5660 7883 should be read starting from the third column (e.g. 1011, 1531,
etc.). Such an arrangement was required to ensure that the entire
Table 5
Aerofoil data for problem 1.
Re NACA CL CD C L =C D DF C L DF C U
table ts in one page, without affecting the readability of the 6. Limitations of the approach
values. Since the DF A values from 1.4 to 2.8 provided converged
solution for the entire range of lift to drag ratios (also evident from The proposed approach is based on the assumption that
Fig. 10), they are clubbed together as a range and mentioned All Representative streamtube is representative of the entire VAWT.
in the table. The lower and upper bounds of the variables as depicted in Fig. 6 is
only valid for NACA0012. For the application of this approach to
other aerofoils, the enveloping bounds of the optimum char-
acteristics should be obtained through calculations.
5. Steps involved in problem solving The approach is based on a constant external wind velocity.
However, for Representative streamtube located in the down-
The optimization process of a particular problem is carried out stream half of the rotor, there is an additional change in the
through the following steps: velocity reaching the streamtube owing to its passage through the
upstream streamtube (as depicted through Eq. (19)). This change
1) The set of aerofoil data with lift and drag values (for different
in velocity is not accounted in the calculation of DF C using Eq. (30).
values of Reynolds number and angle of attack) that are parti-
The resulting inaccuracy can be mitigated through either one of
cipating in the optimization process is arranged in decreasing
the following ways:
order of their lift to drag ratio's.
2) Starting with the rst aerofoil, the range of compatibility
1) Neglect the cases that correspond to Representative streamtube
demand factor, DF C is calculated as given in Eq. (30). For this
located in the downstream half of the rotor. This can be brought
calculation, the values of Reynolds number and angle of attack
about by screening such cases during the elimination stage
pertaining to the aerofoil data is used. The value of power,
discussed through Table 3.
velocity, density and viscosity of wind pertaining to the opti-
2) Adopt a suitable correction factor.
mization problem are used. From the values of desired aspect
ratio range and a realistic minimum and maximum value for
number of blades, the permissible range for DF C is obtained.
3) From Table 4, the permissible values of DF A for the lift to drag
7. Conclusion
ratio under consideration is referenced. Those DF A values which
comes under the range of DF C calculated in step 2 is retained
An approach for optimizing Vertical Axis Wind Turbines
separately.
4) For these common values (with DF A DF C ), DF H is calculated (VAWT) has been presented that uses the theory of double mul-
using Eq. (29) and all those cases which satisfy Eq. (31) tiple streamtube method but eliminates the need of iterations. A
identically becomes the set of solutions which are eligible to single streamtube termed Representative streamtube is assumed
become Representative streamtube for the given problem. to represent the property of the entire VAWT. All the participating
Since the aerofoils are arranged in the decreasing order of their aerofoils are arranged in descending order of their lift to drag
lift to drag ratios, the rst case which satises this condition ratios (for various combinations of Reynolds number and angle of
becomes the optimized solution. attack) and starting with the rst set of lift and drag values, a
5) Once the azimuthal location of Representative streamtube check is made to determine if it is eligible to become Repre-
corresponding to the optimized solution is found out, the ve sentative streamtube. The check is carried out through a factor
variables namely chord length, length of blade, number of
that was developed through systematic modication of existing
blades, radius of rotor and angular velocity at rated speed can be
formulation in a way that it satises the intended objective. The
calculated using Eqs. (16), (15), (18), (17) & (7) respectively.
rst of the lift to drags ratios (along with its associated properties)
Two demonstrative problems based on this approach have that satises the requirement of becoming Representative
been included in Appendix. streamtube is taken as the optimized solution for the problem.
Table 6
Comparison of optimized shape parameters of the VAWT. Table 8
Comparison of optimized shape parameters of the VAWT.
1
N NACA C (m) L (m) R (m) (s )
N NACA C (m) L (m) R (m) (rpm)
This approach 2 0012 0.50 9.55 7.76 0.81
SAWT Energy, 5 Not specied Not specied 9 6.5 Not specied This approach 2 0012 2.26 71.96 34.72 17.86
2015 Marsh and Peace, 2005 3 Not specied 5 55 38.5 28
Table 7
Aerofoil data for problem 2.
Re NACA CL CD C L =C D DF C L DF C U
Appendix References
To demonstrate the working of this approach, two sample Ahmadi-Baloutaki, M., Carriveau, R., S-K Ting, D., 2014. Straight-bladed vertical axis
VAWT proportioning problems have been included as under. wind turbine rotor design guide based on aerodynamic performance and
loading analysis. J. Power Energy 228 (7), 742759.
Ajedegba, J.O., 2008. Effects of Blade Conguration on Flow Distribution and Power
Output of a Zephyr Vertical Axis Wind Turbine (Master of Applied Science
Problem 1 thesis). University of Ontario Institute of Technology, Ontario.
Asher, I.M., Drela, M., Peraire, J., 2010. A Low Order Model for Vertical Axis Wind
The proposed VAWT needs to produce a rated power of 50 kW Turbines. In: Proceedings of the 28th AIAA Applied Aerodynamics Conference,
Chicago.
for a rated wind speed of 13 m/s. The architectural requirements Bedon, G., Paulsen, U.S., Madsen, H.A., Bellonia, F., Castelli, M.R., Benini, E., 2015.
and space constraints of the location restrict the aspect ratio to the Aerodynamic benchmarking of the deepwind design. Energy Procedia 75,
range of 0.3 to 1.0. Viscosity and density of air for the site can be 677682.
Bedon, G., Paulsen, U.S., Madsen, H.A., Belloni, F., Castelli, M.R., Benini, E., 2014.
taken as 1.78E 05 kg/m-s and 1.08 kg/m3 respectively. Aerodynamic Benchmarking of the Deepwind Design. The European Wind
Energy Association, Barcelona.
Beri, H., Yao, Y., 2011. Double multiple stream tube model and numerical analysis of
Solution vertical axis wind turbine. Energy Power Eng. 3, 262270.
Bianchini, A., Ferrara, G., Ferrari, L., 2015. Design guidelines for H-Darrieus wind
turbines: optimization of the annual energy yield. Energy Convers. Manag. 89,
For this problem, the lift and drag coefcients pertaining to 690707.
NACA0012 is taken from experiments conducted by (Sheldahl and Bitsch, J., Darlow, R., Nielsen, J.G., Qumneur, C., Savino, S., 2011. Modeling of a
Klimas, 1981) under the aegis of Sandia National Laboratories. Vertical Axis Wind Turbine with Respect to Pitching Angle and Wing Proles
(Bachelor Project). Aalborg University, Denmark.
Implementing the steps discussed in Section 5 leads to the Blusseau, P., Patel, M.H., 2012. Gyroscopic effects on a large vertical axis wind
following: turbine mounted on a oating structure. Renew. Energy 46, 3142. http://dx.
Step 1) The aerofoil data that are participating in the optimi- doi.org/10.1016/j.renene.2012.02.023.
Bogeanu, R., Dobrescu, B., Nil, I., 2010. Aerodynamic performance prediction of
zation process is arranged in decreasing order of their lift to drag Darrieus-type wind turbines. INCAS Bull. 2 (2), 2632.
ratio and the rst 17 of them are shown in Table 5. Borg, M., Collu, M., Brennan, F.P., 2013. Use of a wave energy converter as a motion
Step 2) The range of DF C is calculated and shown in Table 5. suppression device for oating wind turbines. Energy Procedia 35, 223233.
http://dx.doi.org/10.1016/j.egypro.2013.07.175.
The lower limit is given as DF C L and the upper limit as Castelli, M.R., Englaro, A., Benini, E., 2011. The Darrieus wind turbine: proposal for a
DF C U. For this calculation, the range of minimum and max- new performance prediction model based on CFD. Energy 36, 49194934.
imum number of blades is taken as 2 and 5 respectively. Castillo, J., 2011. Small-Scale Vertical Axis Wind Turbine Design (Bachelors Thesis).
Tampere University of Applied Sciences, Finland.
Step 3) With the range of DF C values known, the values of DF A Chen, X., Agarwal, R., 2012. Optimal placement of horizontal - and vertical-axis
that are acceptable for the particular value of lift to drag ratio wind turbines in a wind farm for maximum power generation using a genetic
are obtained from Table 4. The rst case which has some values algorithm. Int. J. Energy Environ. 3 (6), 927938.
Claessens, M.C., 2006. The Design and Testing of Airfoils for Application in Small
in the acceptable range is marked with grey shading in Table 5. Vertical Axis Wind Turbines (Master of Science thesis). Delft University of
The acceptable values are 1.3, 1.4 & 1.5. Technology, Netherlands.
Step 4) Out of the acceptable values of 1.3, 1.4 & 1.5, the DF H Clarke, J., Hancox, L., MacKenzie, D., Whelan, M., 2014. Design of a Vertical-Axis
Wind Turbine Phase II. Group 11 - MUN VAWT Design.
gives the same value only for 1.5.
Deglaire, P., 2010. Analytical Aerodynamic Simulation Tools for Vertical Axis Wind
Step 5) The VAWT output parameters are calculated using Eqs. Turbines (Doctor of Philosophy dissertation). Acta Universitatis, Upsaliensis,
(15)(18) and (7). The deign parameters of the optimized VAWT Uppsala, Sweden.
Diaz, D.A.G., Pinto, F.A.N.C., 2012. Vertical Wind Turbine with Variable Blade
is summarized in Table 6. It is compared with a VAWT designed
Angular Position. In: Proceedings of ABCM Symposium Series in Mechatronics.
for a similar condition in (SAWT Energy, 2015). vol. 5, pp. 474483.
Farthing, S.P., 2009. Vertical axis wind turbine induced velocity vector theory. Proc.
IMechE Part A: J. Power Energy 223, 103114.
Problem 2 Hasan, O.D.S., Hantoro, R., Nugroho, G., 2013. Studi eksperimental vertical axis wind
turbine tipe savonius dengan variasi jumlah n pada sudu. J. Tek. Pomits 2 (2),
B-350B-355.
Redesign the VAWT mentioned in problem 1 for a rated power Helms, S.R., 2011. Simulations of Vertical Axis Wind Turbine Blades (Master of
of 3.0 MW and for a rated wind speed of 14 m/s. Science thesis). Drexel University, Philadelphia.
Homicz, G.F., 1991. Numerical Simulation of VAWT Stochastic Aerodynamic Loads
Produced by Atmospheric Turbulence: VAWT-SAL Code, Sandia report SAND91-
Solution 1124. Sandia National Laboratories, Albuquerque.
Islam, M., Carriveau, R., Fartaj, A., 2012. Performance analyses of a xed-pitch
straight bladed VAWT with selected low Reynolds number airfoils. Int. J.
Assumptions on minimum and maximum number of blades Environ. Stud. 69 (2), 289298.
and other values are retained from Problem 1. Islam, M., Ting, S.-K., Fartaj, A., D., 2008. Aerodynamic models for Darrieus-type
straight-bladed vertical axis wind turbines. Renew. Sustain. Energy Rev. 12,
Step 1) The list of rst four aerofoil data that are participating 10871109.
in the optimization process (in decreasing order of lift to drag Komerath, N.M., Pendharkar, A.M., Raghav, V., 2013. The low-cost vertical axis wind
ratio) are shown in Table 7. turbine project: an exercise in learning across disciplines. In: Proceedings of the
120th ASEE Annual Conference and Exposition, Atlanta.
Step 2) The lower limit (DF C L) and upper limit (DF C U) of Larrea, L.R., 2015. Vibration Analysis of a VAWT (Master thesis). Technical Uni-
DF C are shown in Table 7. versity of Soa, Bulgaria.
Step 3) The rst row (marked in grey shade) of Table 7 has Loth, J.L., McCoy, H., 1983. Optimization of Darrieus turbines with an upwind and
downwind momentum model. J. Energy 7 (4), 313318.
values in the acceptable range. The acceptable values are 1.3, 1.4,
Maheri, A., Noroozi, S., Toomer, C., Vinney, J., 2006. Damping the Fluctuating
1.5, 1.6 & 1.7. Behaviour and Improving the Convergence Rate of the Axial Induction Factor in
Step 4) Out of those values, Eq. (31) is identically satised for the BEMT-Based Rotor Aerodynamic Codes. European Wind Energy Conference,
Athens.
values of 1.4, 1.5 & 1.6.
Marsh, G., Peace, S., 2005. Tilting at windmills: utility-scale VAWTs: towards
Step 5) Taking 1.4 as the solution and based on the azimuthal 10 MW and beyond? Refocus 6 (5), 3742. http://dx.doi.org/10.1016/
location of that particular value, the output parameters of VAWT S1471-0846(05)70457-1.
have been found. The deign parameters of the optimized VAWT McGowan, R., Morillas, K., Pendharkar. A., Pinder, M., Komerath, N., 2012. Optimi-
zation of a Vertical Axis Micro Wind Turbine for Low Tip Speed Ratio Operation.
is summarized in Table 8. It is compared with a VAWT designed In: Proceedings of the 10th International Energy Conversion Engineering
for a similar condition in (Marsh and Peace, 2005). Conference, Atlanta.
M. Abdul Akbar, V. Mustafa / J. Wind Eng. Ind. Aerodyn. 153 (2016) 3445 45
Ming, W.F., 2009. Investigation on Straight-bladed Wind Turbines (Master's thesis). Sheldahl R.E., Klimas, P.C., 1981. Aerodynamic Characteristics of Seven Symmetrical
Lanzhou University of Technology, Lanzhou. Airfoil Sections Through 180-Degree Angle of Attack for Use in Aerodynamic
Newman, B.G., 1983. Actuator-disc theory for vertical-axis wind turbines. J. Wind Analysis of Vertical Axis Wind Turbines, SAND80-2114. Sandia National
Eng. Ind. Aerodyn. 15 (13), 347355. Laboratories energy report, Albuquerque.
Newman, B.G., 1986. Multiple actuator-disc theory for wind turbines. J. Wind Eng. Solum, A., Deglaire, P., Eriksson, S., Stlberg, M., Leijon, M., Bernhoff, H., 2006.
Ind. Aerodyn. 24, 215225. Design of a 12 kW vertical axis wind turbine equipped with a direct driven PM
Okpue, A.S., 2011. Aerodynamic Analysis Of Vertical And Horizontal Axis Wind synchronous generator. In: Proceedings of European Wind Energy Conference &
Turbines (Master of Science thesis). Michigan State University, East Lansing. Exhibition, Athens.
Paraschivoiu, I., 1981. Double-multiple streamtube model for Darrieus wind tur- Soraghan, C.E., Jamieson, P., Leithead, W.E., 2013. Inuence of lift to drag ratio on
bines. In: Second DOE/NASA Wind Turbines Dynamics Workshop, pp. 1925. optimal aerodynamic performance of straight blade vertical axis wind turbines.
Pratumnopharat, P., Leung, P.S., 2011. Validation of various windmill brake state In: Proceedings of European Wind Energy Association Annual Conference,
models used by blade element momentum calculation. Renew. Energy 36, Vienna.
32223227. Sharpe, T., Proven, G., 2010. Crossex: concept and early development of a true
Ponta, F.L., Lago, L.I., 2008. Analysing the suspension system of variable-geometry building integrated wind turbine. Energy Build. 12, 23652375.
oval-trajectory (VGOT) Darrieus wind turbines. Energy Sustain. Dev. 12 (2), Shires, A., 2013a. Design optimisation of an offshore vertical axis wind turbine.
516. Proc. Instit. Civil Eng. Energy 166 (EN1), 718.
Ponta, F., Otero, A., Lago, L., 2011. Innovative concepts in wind-power generation: Shires, A., 2013b. Development and evaluation of an aerodynamic model for a novel
the VGOT Darrieus. Wind Turbines, 137162. vertical axis wind turbine concept. Energies 6, 25012520.
Ponta, F.L., Seminara, J.J., Oteroa, A.D., 2007. On the aerodynamics of variable- Sullivan, W.N., 1979. Economic Analysis of Darrieus Vertical Axis Wind Turbine
geometry oval-trajectory Darrieus wind turbines. Renew. Energy 32, 3556. Systems for the Generation of Utility Grid Electrical Power, Volume II The
Rathi, D., 2012. Performance Prediction and Dynamic Model Analysis of Vertical Economic Optimization Model, SAND78-0962. Sandia Laboratories energy
Axis Wind Turbine Blades with Aerodynamically Varied Blade Pitch (Master of report, Albuquerque.
Science thesis). North Carolina State University, Raleigh. Tang, S., Xu, Y., Zhu, G., Fang, W., 2011. Orthogonal experiment optimum design in
Samaraweera, K.K.M.N.P., Pathirathna, K.A.B., De Silva, H.E.D., Sugathapala, A.G.T., the power of the rotor on straight-bladed VAWT. Energy Procedia 12, 856861.
2010. Development of Darrieus-Type Vertical Axis Wind Turbine for Stand- Wille, T., 2015. Development of a Calculation method for the Interpretation of pitch
Alone Applications. In: Proceedings of International Conference on Sustainable regulations operated in Vertical Wind turbines by Open Source criteria,
Built Environment, Kandy, pp. 421430. Bachelor thesis. Beuth University of Applied Sciences, Berlin.
SAWT Energy, http://www.sawtenergy.com/pk50.html, 1st May 2015.