This action might not be possible to undo. Are you sure you want to continue?
Prof. John P. Sullivan
Aerodynamic Design of a Large
HorizontalAxis Wind Turbine
Final Project Report
Ted Light & Jeffrey Robinson
December 19
th
, 2003
I. Introduction
This report describes the aerodynamic design project undertaken to analyze and optimize the
performance of a large horizontalaxis wind turbine. The analysis began with the modeling of an
existing wind turbine with documented geometry and performance, and then continued with the
variation of several geometric characteristics in order to improve its performance. This process
enabled an understanding of some of the fundamentals of wind turbine design. By changing
these aerodynamic properties, an improved design was reached.
II. Background
For thousands of years, the wind has provided energy to mankind. The wind has allowed ships
to sail from place to place, and it has turned windmills which allowed the grinding of grain and
the pumping of water. The first
windmills were primitive and
caught as much air as they
could pointing in only one
direction and using simple,
straight, nonaerodynamic
blades; nonetheless, they did
what no other machines had
done before.
However, actual power
generation using windmills did
not occur until the end of the
19
th
century. The advancement
of this technology was slow, especially with the rapid industrialization taking place in Europe
and the United States. As oil prices fell in the 20
th
century, wind energy became a less attractive
energy source. As such a new technology, wind turbines were expensive to develop and build,
and even though the wind itself was free, wind power was still too unreliable to take the place of
fossil fuels.
Not until the Oil Crisis of the 1970s, when gasoline and other petroleum products were very
limited and expensive, did serious interest in wind turbine technology once again emerge. Since
then, wind power has become one of the fastest growing energy sources in the world. Constant
improvements in aerodynamics, structures, and generators have led to more efficient wind
turbines outputting more power than ever.
Their use is not only limited to largescale, commercial energy production. Some small wind
turbines are used in homes, for pumping water, or powering satellite and telecommunications
dishes. Sometimes they are even used in connection with diesel generators, batteries, and solar
power systems. Systems such as these are known as hybrid wind systems and typically are found
in remote locations where outside electric power cannot be utilized.
3
However, it is in these large, commercial utility wind turbines that much technological
improvement has taken place. In addition to individual turbine improvements, changes in the
wind turbine’s size and the total number being used have increased dramatically over recent
years. Today’s large wind turbines are sometimes more than a hundred meters in diameter and
could be only one of tens or hundreds of wind turbines, all being used together to form a “wind
energy power plant.”
Size is only one design consideration for wind turbines. The number of turbine blades varies
between wind machines, along with the blade’s aerodynamic properties, such as taper, twist, and
crosssectional shape. Other advances in wind turbine technology continue today, leading to
more powerful and more efficient wind turbines, which will be used to power more of the world
in coming years.
III. Procedure
The problem undertaken is the aerodynamic design of a large horizontalaxis wind turbine. To
serve as a baseline for comparison purposes, a turbine with established geometry and
performance data was found. This turbine was the Mod2 wind turbine, built by Boeing in the
late 1970s. Table 1 gives the geometry and operational data of the turbine, while Figure 1 is a
photograph of the Boeing Mod2 wind turbine.
Table 1: Mod2 Wind Turbine Information
Diameter [m] 91.4
No. of Blades 2
Airfoil Section NACA 23024
Tip Pitch [deg] 4.0
Root Pitch [deg] 2.55
Tip Chord [m] 1.43
Root Chord [m] 3.45
Wind Speed, V
0
[m/s] 12.5
Rotational Rate [rpm] 17.5
Power Output [MW] 2.5
Figure 1: Photograph of the Boeing Mod2 Wind Turbine
This turbine’s geometry was modeled using CMARC, and then further modified to attain
improved performance. The diameter, rotational rate, wind speed, and number of blades were
kept constant, however, to keep the comparisons similar. The turbines were also considered to
be constantspeed turbines, since our analysis required a constant rotational rate. Since many
turbines in use today are constantspeed turbines, this was not an unreasonable assumption.
The objective of design improvements was therefore to increase the power output by the wind
turbine. The power output of the turbine is given by the simple equation,
) Rate Rotational ( ) axis rotational about Torque ( ) Power ( ⋅ · (1)
The rotational rate was kept constant at 17.5 rpm and the moment about the rotational axis was
calculated by the 3D analysis tool CMARC. Research was conducted to formulate ideas on how
to improve upon the geometry of the wind turbine. These are discussed below.
An expression for the optimal chord length was given by Reference 2. It gives the optimal chord
length based upon the number of turbine blades B, their radius R, and the tip speed ratio ?, which
is the linear speed of the blade tips divided by the wind speed. The expressions for chord length
and tipspeed ratio are given as Equations (2) and (3), respectively.
B
r R R
c
2
9
) / ( 16
λ
π
· (2)
0
/ V RΩ · λ (3)
Taking five equally spaced control points located at the root, 0.25R, 0.50R, 0.75R, and at the tip,
the optimal chord distribution as defined in Equation (2) was calculated; this distribution is given
in Table 2.
Table 2: Ideal Chord Lengths
r/R Ideal Chord (m)
0.00 15.00
0.25 11.40
0.50 5.70
0.75 3.80
1.00 2.85
Note that the Equation (2) is not valid for the first control point where r/R = 0, so what was
thought to be a reasonable value was substituted.
Reference 1 also gives expressions for optimal wind turbines based upon Glauert’s optimal rotor
theory. Here the optimal blade layout is given product of the chord c and sectional lift
coefficient c
l
.
) 3 1 )( 1 (
) 2 1 (
) 1 4 (
) (
8
0
a a
a
a
c c
V
B
l
− −
]
]
]
−
−
·
,
`
.
 Ω
π
(4)
In Equation (4) above, a is the axial interference factor, which can be approximated by the
following power series as a function of the local speed ratio, x = rΩ/V
0
.
...
59049
418
729
10
81
2
3
1
6 4 2
+
,
`
.

−
,
`
.

+
,
`
.

−
,
`
.

·
x x x
a (5)
These relations were the basis for many of the design changes. The chord and twist values that
were used for the design iterations are tabulated below in Table 3 and Table 4, respectively.
Table 3: Chord Distributions for Design Iterations
r/R (1) (2) (3)
0.00 15.00 14.00 10.00
0.25 11.40 11.40 8.00
0.50 5.70 7.00 6.00
0.75 3.80 3.80 4.00
1.00 2.85 2.85 2.00
Chord Changes (values in meters)
Table 4: Twist Distributions for Design Iterations
r/R (1) (2) (3)
0 6.1 6.0 5.5
0.25 66.7 27.8 83.5
0.5 83.4 49.5 89
0.75 93.8 71.3 90
1 111.6 93.0 92
Twist Changes (values in deg from wind axis)
Our best turbine, which had the highest power output, included the chord distribution listed as
Chord Change (3) above, as well as the twist change listed as Twist Change (1) above. Chord
Change (3) was a linear taper approximately over a range of values given by the optimal chord
length equations above. Twist Change (1) was found by substituting the same chord distribution
into Equation (4) above, and assuming a lift curve slope equal to 2π. One can see that in this
twist distribution, the blade’s tip actually twists back beyond the plane of rotation by more than
20 degrees. It seems suspicious that this distribution resulted in the highest calculated power
output. This is due to the fact that the blade has a pitch greater than 90 degrees, and it will
actually work against the rest of the rotor by lifting in the opposite direction and take power out
of the entire wind turbine system. The surprising result obtained from CMARC raised the
question as to whether CMARC itself was an accurate analysis tool.
IV. Analysis of Results
Beginning with the Mod2 baseline model, several modifications were made. With each change,
the moment about the rotational axis was computed and used to calculate the power output in
kilowatts. This respective power output was then compared to the baseline model in order to
establish if the modification was beneficial or not. Table 5 gives the results of all of the design
iterations. For each model, the model number is given, with a brief description of the
modification, as well as with the power output its relative change from the baseline.
Table 5: Design Iteration Results
Mod2 Baseline Model 5816 
ModA Airfoil Changed to NACA 65
1
012 5516 12%
ModB Airfoil Changed to NACA 65410 6179 15%
ModC Airfoil Changed to Althaus AH 93W145 2551 131%
ModD Chord Change (1) from ModB with no sweep 8914 124%
ModE Chord Change (2) from ModB with no sweep 8442 105%
ModF Chord Change (3) from ModB with no sweep 7872 82%
ModG Chord Change (3) from ModB with sweep 9380 143%
ModH Twist Change (1) from ModG 10367 182%
ModI Twist Change (2) from ModG 7849 81%
ModJ Twist Change (3) from ModG 7735 77%
Mod2 Rating Published Power Rating 2500
Model # Description
Power (kW)
D
from
Baseline
After several modifications to specific elements (i.e., airfoil shape, chord distribution, etc.), the
best design was kept and used in later models. An example of this method is shown when
several different airfoils were tested in the models ModA through ModC. Of these options, the
NACA 65410 was chosen to be the best airfoil section, based on power output values in the
Table 5. This was then chosen to be the airfoil used in subsequent modifications, and ultimately
it was chosen for the final design.
After the airfoil shapes were tested and the best kept for further testing, the chord distribution
was altered in hopes that a more optimal design could be found. As was discussed in the
previous section, the initial chord distribution was obtained using Equation (2). The chord
distribution was modified twice more; the distributions are shown in Table 5. Again, the best
chord distribution was kept and used for later modifications.
Finally, the twist distribution was modified in a similar fashion to that of the airfoil section and
the chord distribution. Twisting the blade along the length allowed the angle of attack to be
optimized, depending on the relative wind that a given portion of the blade would “see.” Also,
by changing the angle of attack, the lift was maximized, increasing the rotational moment and
ultimately the power output.
Using these modifications, a highly optimized, final wind turbine blade design could be made.
This blade design used the NACA 65410 airfoil section, was approximately 45.7 meters long,
with a chord length of 10 meters at the root, tapering linearly to 2 meters at the tip. The blade
was highly twisted, with the root directed into the natural wind direction and the tip
approximately aligned with the plane of rotation; the blade underwent a major initial change in
pitch—60 degrees in pitch in the first 25% of the blade length.
This design created a moment about the rotational axis of 2.83 x 10
6
Nm. Substituting this
value into Equation (1), along with the given rotation rate of 17.5 rpm (1.833 rad/sec), the power
output was computed to be 10367 kW.
The theoretical limit of the maximum amount of power that can be extracted from a wind turbine
is known as the Betz limit and is approximately 60%. Given that limit, an analysis is possible to
see if the amount of power that can be extracted from the wind has been overpredicted. The
kinetic energy in the wind is given by
2
0
2
4 2
1
V
D
KE
,
`
.

·
π
ρ (6)
The power available can then be shown to be equal to the kinetic energy of the wind multiplied
by its velocity and the Betz limit.
MW
s
m m
m
kg
V
D
V
D
V P
709 . 4 ) 5 . 12 (
4
) 4 . 91 (
225 . 1 3 . 0
4
3 . 0
4 2
1
6 . 0
3
2
3
3
0
2
2
0
2
0
·
,
`
.
 ⋅
⋅ ·
,
`
.

⋅ ·
,
`
.

,
`
.

⋅ ·
π
π
ρ
π
ρ
Therefore, from the above calculation, the maximum theoretical output of a wind turbine, using
the final design’s given size and wind speed is approximately 4.7 megawatts.
(7)
A picture of our final turbine is shown in Figure 2. The changing chord length and twist
distribution along the blade can easily be seen.
Figure 2: Final Wind Turbine Blade Design
Figure 3 is a contour plot the c
p
distribution of the wind turbine blade near the root. The 3D
effects of leakage near the root are evident. The picture also suggests a possible overprediction
of c
p
along the upper surface of the rotor.
Figure 3: Cp Distribution for Final Wind Turbine Blade Design
The results of the animated wake are shown below in Figure 4. Evident is the helix shape as well
as the rollup of the vortices near the tip.
Figure 4: Wake of Final Wind Turbine Blade Design
V. Conclusion
The aerodynamic design of this large horizontalaxis wind turbine was comprised of several
steps, ultimately resulting in a highly optimized final design. The design process began by using
technical data from a realworld wind turbine, the Boeing Mod2, to model a baseline design.
From this design, different aerodynamic parameters were varied in order to achieve a more
optimal design. Not only did this method allow performance improvements, but it provided an
understanding of many of the fundamentals of wind turbine design and qualitative effects of
different aerodynamic changes.
After each design iteration, the power output was computed and compared to both the baseline
model, as well as other previous design iterations. Design changes and modifications continued
in hopes of improving the performance even further. Ultimately, a model with the highest power
output was chosen as the final design.
VI. References
1
Eggleston, David M. and Forrest S. Stoddard. Wind Turbine Engineering Design. New York:
Van Nostrand Reinhold Company, 1987.
2
Piggot, Hugh. Small Wind Turbine Design Notes. Retrived Dec 3, 2003, from
http://users.aber.ac.uk/iri/WIND/TECH/WPcourse.
3
US Department of Energy. Wind and Hydropower Technologies Program: How Wind
Turbines Work. US Department of Energy. Retrived December 18, 2003.
<http://www.eere.energy.gov/windandhydro/wind_how.html>
VII. Appendix
A.1 – CMARC Input File for Best Wind Turbine (ModH)
MODIFIED TURBINE H  modH.in > TWIST CHANGE (1)
Airfoil = NACA 65410
Cbar = 6.0 [m]
R = 45.7 [m]
Alpha at Root = 6.1 [deg]
Alpha at Tip = 111.6 [deg]
> Twist is nonlinear
Phi dot = 105 [deg/sec]
Vinf = 12.5 [m/s]
&BINP2 LSTINP=2, LSTOUT=1, LSTFRQ=1, LENRUN=0, &END
&BINP3 LSTGEO=1, LSTNAB=0, LSTWAK=3, LSTCPV=0, LSTJET=0,
&END
&BINP4 MAXIT=600, SOLRES=0.005, &END
&BINP5 NTSTPS=60, DTSTEP=0.08, &END
&BINP6 RSYM=1.0, RGPR=0.0, RFF=5.0, RCORE=0.05, &END
&BINP7 VINF=12.5, VSOUND=340, UNIT=1, COMPOP=0.0, &END
&BINP8 ALDEG=0.0, YAWDEG=0.0, THEDOT=0.0, PSIDOT=0.0, PHIDOT=105, &END
&BINP8A PHIMAX=0, THEMAX=0.0, PSIMAX=0.0,
WRX=0.0, WRY=0.0 WRZ=0.0 &END
&BINP8B DXMAX=0, DYMAX=0.0, DZMAX=0.0,
WTX=0, WTY=0.0 WTZ=0.0 &END
&BINP9 CBAR=6.0, SREF=274.2, SSPAN=45.7,
RMPX=0.00, RMPY=0.00, RMPZ=0.00, &END
&BINP10 NORSET=0, NBCHGE=0, NCZONE=0,
NCZPAN=0, CZDUB=0.0, VREF=0.0, &END
&BINP11 NORPCH=0, NORF=0, NORL=0,
NOCF=0, NOCL=0.0, VNORM=0.0, &END
&BINP12 KPAN=0, KSIDE=0, NEWNAB=0, NEWSID=0, &END
&BINP13 NBLIT=0, &END
&ASEM1 ASEMX=0.00, ASEMY=0.00, ASEMZ=0.00,
ASCAL=1.00, ATHET=0.00, NODEA=5, &END
&COMP1 COMPX=0.0000, COMPY=0.0000, COMPZ= 0.0000,
CSCAL=1.000, CTHET=0.0, NODEC=5, &END
&PATCH1 IREV=0, IDPAT=1, MAKE=0, KCOMP=1, KASS=1, &END
   RIGHT BLADE   
&SECT1 STX=0, STY=0.0, STZ=0, SCALE=10.0, ALF=6.1,
THETA=0.0, INMODE=4, TNODS=0, TNPS=0, TINTS=0,
&END
   NACA 65410 AIRFOIL   
1.00000 0 0.00000
0.94971 0 0.00327
0.89943 0 0.00226
0.84924 0 0.00037
0.79912 0 0.00393
0.74910 0 0.00792
0.69915 0 0.01211
0.64927 0 0.01621
0.59947 0 0.02004
0.54971 0 0.02340
0.50000 0 0.02606
0.45032 0 0.02773
0.40064 0 0.02854
0.35097 0 0.02863
0.30128 0 0.02814
0.25157 0 0.02710
0.20183 0 0.02547
0.15202 0 0.02314
0.10212 0 0.01999
0.07711 0 0.01791
0.05203 0 0.01536
0.02682 0 0.01191
0.01411 0 0.00944
0.00893 0 0.00781
0.00628 0 0.00661
0.00000 0 0.00000
0.00372 0 0.00861
0.00607 0 0.01061
0.01089 0 0.01372
0.02318 0 0.01935
0.04797 0 0.02800
0.07289 0 0.03487
0.09788 0 0.04067
0.14798 0 0.05006
0.19817 0 0.05731
0.24843 0 0.06290
0.29872 0 0.06702
0.34903 0 0.06983
0.39936 0 0.07138
0.44968 0 0.07153
0.50000 0 0.07018
0.55029 0 0.06720
0.60053 0 0.06288
0.65073 0 0.05741
0.70085 0 0.05099
0.75090 0 0.04372
0.80088 0 0.03577
0.85076 0 0.02729
0.90057 0 0.01842
0.95029 0 0.00937
1.00000 0 0.00000
&BPNODE TNODE=3, TNPC=0, TINTC=0, &END
&SECT1 STX=0, STY=11.4, STZ=1.0, SCALE=8.0,
ALF=66.7, THETA=0.0, INMODE=0, TNODS=0,
TNPS=14, TINTS=3, &END
&SECT1 STX=0, STY=22.9, STZ=2.0, SCALE=6.0,
ALF=83.4, THETA=0.0,
INMODE=0, TNODS=0, TNPS=14, TINTS=3, &END
&SECT1 STX=0, STY=34.3, STZ=3.0, SCALE=4.0,
ALF=93.8, THETA=0.0,
INMODE=0, TNODS=0, TNPS=14, TINTS=3, &END
&SECT1 STX=0, STY=45.7, STZ=4.0, SCALE=2.0,
ALF=111.6, THETA=0.0,
INMODE=0, TNODS=3, TNPS=14, TINTS=3, &END
&PATCH1 IREV=0, IDPAT=1, MAKE=1, KCOMP=1, KASS=1, &END
   RIGHT TIP   
&PATCH2 ITYP=1, TNODS=5, TNPS=3, TINTS=3, NPTTIP=0, &END
&WAKE1 IDWAK=1, IFLXW=0, &END
   RIGHT WAKE   
&WAKE2 KWPACH=1, KWSIDE=2, KWLINE=0, KWPAN1=0, KWPAN2=0, NODEW=5,
INITIAL=0, &END
&ONSTRM NONSL=0 KPSL=0 &END
&BLPARAM RN =1640000, VISC =0.00016, NSLBL = 1, &END
&VS1 NVOLR=0, NVOLC=0, &END
&SLIN1 NSTLIN=0, &END
nonetheless. wind energy became a less attractive energy source. an improved design was reached. straight. By changing these aerodynamic properties. As oil prices fell in the 20th century. the wind has provided energy to mankind. and then continued with the variation of several geometric characteristics in order to improve its performance. The advancement of this technology was slow. wind turbines were expensive to develop and build. nonaerodynamic blades. However. especially with the rapid industrialization taking place in Europe and the United States. . The analysis began with the modeling of an existing wind turbine with documented geometry and performance. they did what no other machines had done before. As such a new technology. and it has turned windmills which allowed the grinding of grain and the pumping of water. actual power generation using windmills did not occur until the end of the 19th century. II. Introduction This report describes the aerodynamic design project undertaken to analyze and optimize the performance of a large horizontalaxis wind turbine.I. The wind has allowed ships to sail from place to place. The first windmills were primitive and caught as much air as they could pointing in only one direction and using simple. This process enabled an understanding of some of the fundamentals of wind turbine design. Background For thousands of years.
wind power was still too unreliable to take the place of fossil fuels. along with the blade’s aerodynamic properties. Today’s large wind turbines are sometimes more than a hundred meters in diameter and could be only one of tens or hundreds of wind turbines. such as taper. structures. commercial energy production. Since then.” Size is only one design consideration for wind turbines. wind power has become one of the fastest growing energy sources in the world. changes in the wind turbine’s size and the total number being used have increased dramatically over recent years. and crosssectional shape. batteries. twist. Not until the Oil Crisis of the 1970s. all being used together to form a “wind energy power plant. when gasoline and other petroleum products were very limited and expensive. or powering satellite and telecommunications dishes.3 However. and solar power systems. commercial utility wind turbines that much technological improvement has taken place. Other advances in wind turbine technology continue today. which will be used to power more of the world in coming years.and even though the wind itself was free. leading to more powerful and more efficient wind turbines. for pumping water. Sometimes they are even used in connection with diesel generators. The number of turbine blades varies between wind machines. Constant improvements in aerodynamics. did serious interest in wind turbine technology once again emerge. and generators have led to more efficient wind turbines outputting more power than ever. In addition to individual turbine improvements. it is in these large. Their use is not only limited to largescale. Some small wind turbines are used in homes. Systems such as these are known as hybrid wind systems and typically are found in remote locations where outside electric power cannot be utilized. .
5 Figure 1: Photograph of the Boeing Mod2 Wind Turbine .III. Procedure The problem undertaken is the aerodynamic design of a large horizontalaxis wind turbine. To serve as a baseline for comparison purposes. Table 1: Mod2 Wind Turbine Information Diameter [m] No.55 1. built by Boeing in the late 1970s.5 2.5 17. This turbine was the Mod2 wind turbine.45 12.4 2 NACA 23024 4. V0 [m/s] Rotational Rate [rpm] Power Output [MW] 91. while Figure 1 is a photograph of the Boeing Mod2 wind turbine. a turbine with established geometry and performance data was found. Table 1 gives the geometry and operational data of the turbine. of Blades Airfoil Section Tip Pitch [deg] Root Pitch [deg] Tip Chord [m] Root Chord [m] Wind Speed.43 3.0 2.
Since many turbines in use today are constantspeed turbines. respectively. The diameter. and number of blades were kept constant. rotational rate. this distribution is given in Table 2. this was not an unreasonable assumption. the optimal chord distribution as defined in Equation (2) was calculated. 16 π R( R / r ) 9 λ2 B c= (2) (3) λ = RΩ /V0 Taking five equally spaced control points located at the root.25R. wind speed. (Power ) = (Torque about rotational axis ) ⋅ (Rotational Rate) (1) The rotational rate was kept constant at 17. The expressions for chord length and tipspeed ratio are given as Equations (2) and (3). The objective of design improvements was therefore to increase the power output by the wind turbine. 0.5 rpm and the moment about the rotational axis was calculated by the 3D analysis tool CMARC. and at the tip. 0. Research was conducted to formulate ideas on how to improve upon the geometry of the wind turbine. . The power output of the turbine is given by the simple equation. 0. to keep the comparisons similar.75R. It gives the optimal chord length based upon the number of turbine blades B.50R. however. their radius R. and then further modified to attain improved performance. which is the linear speed of the blade tips divided by the wind speed.This turbine’s geometry was modeled using CMARC. and the tip speed ratio ?. These are discussed below. An expression for the optimal chord length was given by Reference 2. The turbines were also considered to be constantspeed turbines. since our analysis required a constant rotational rate.
40 8.00 0..85 2.40 11.80 3.00 0. which can be approximated by the following power series as a function of the local speed ratio.00 0. x = rΩ/V0 .Table 2: Ideal Chord Lengths r/R 0.50 0. Here the optimal blade layout is given product of the chord c and sectional lift coefficient c l.00 10. 1 2 10 + a = − 2 4 3 81 x 729 x 418 − 59049 x 6 + .40 5. Reference 1 also gives expressions for optimal wind turbines based upon Glauert’s optimal rotor theory.25 0.00 0.00 .70 7. a is the axial interference factor..00 Ideal Chord (m) 15. so what was thought to be a reasonable value was substituted.00 14.50 5. BΩ ( 4a − 1) 8πV ( c cl ) = (1 − 2a) (1 − a )(1 − 3a) 0 (4) In Equation (4) above.00 15.00 6.75 1.85 Note that the Equation (2) is not valid for the first control point where r/R = 0.00 2.00 1.75 3. The chord and twist values that were used for the design iterations are tabulated below in Table 3 and Table 4.70 3. respectively. (5) These relations were the basis for many of the design changes.00 11. Table 3: Chord Distributions for Design Iterations Chord Changes (values in meters) r/R (1) (2) (3) 0.85 2.25 11.80 4.80 2.
. several modifications were made.0 92 Our best turbine. the blade’s tip actually twists back beyond the plane of rotation by more than 20 degrees. included the chord distribution listed as Chord Change (3) above. with a brief description of the modification. One can see that in this twist distribution. Analysis of Results Beginning with the Mod2 baseline model.5 83. which had the highest power output. as well as with the power output its relative change from the baseline.1 6. It seems suspicious that this distribution resulted in the highest calculated power output. as well as the twist change listed as Twist Change (1) above. This respective power output was then compared to the baseline model in order to establish if the modification was beneficial or not.0 5. the moment about the rotational axis was computed and used to calculate the power output in kilowatts.7 27.8 83. With each change. For each model.3 90 1 111. This is due to the fact that the blade has a pitch greater than 90 degrees.5 89 0. the model number is given. and assuming a lift curve slope equal to 2π.5 0. IV. The surprising result obtained from CMARC raised the question as to whether CMARC itself was an accurate analysis tool. Twist Change (1) was found by substituting the same chord distribution into Equation (4) above.75 93. and it will actually work against the rest of the rotor by lifting in the opposite direction and take power out of the entire wind turbine system.Table 4: Twist Distributions for Design Iterations Twist Changes (values in deg from wind axis) r/R (1) (2) (3) 0 6.5 0.4 49. Chord Change (3) was a linear taper approximately over a range of values given by the optimal chord length equations above. Table 5 gives the results of all of the design iterations.25 66.6 93.8 71.
by changing the angle of attack. the distributions are shown in Table 5. The chord distribution was modified twice more. As was discussed in the previous section.Table 5: Design Iteration Results Model # Mod2 ModA ModB ModC ModD ModE ModF ModG ModH ModI ModJ Mod2 Rating Description Baseline Model Airfoil Changed to NACA 651012 Airfoil Changed to NACA 65410 Airfoil Changed to Althaus AH 93W145 Chord Change (1) from ModB with no sweep Chord Change (2) from ModB with no sweep Chord Change (3) from ModB with no sweep Chord Change (3) from ModB with sweep Twist Change (1) from ModG Twist Change (2) from ModG Twist Change (3) from ModG Published Power Rating Power (kW) 5816 5516 6179 2551 8914 8442 7872 9380 10367 7849 7735 2500 D from Baseline 12% 15% 131% 124% 105% 82% 143% 182% 81% 77% After several modifications to specific elements (i. increasing the rotational moment and ultimately the power output. based on power output values in the Table 5.. the twist distribution was modified in a similar fashion to that of the airfoil section and the chord distribution. the best chord distribution was kept and used for later modifications. the NACA 65410 was chosen to be the best airfoil section. An example of this method is shown when several different airfoils were tested in the models ModA through ModC. Twisting the blade along the length allowed the angle of attack to be optimized. final wind turbine blade design could be made. This blade design used the NACA 65410 airfoil section. This was then chosen to be the airfoil used in subsequent modifications. the lift was maximized. airfoil shape.). Using these modifications. the best design was kept and used in later models. and ultimately it was chosen for the final design. a highly optimized. Again. .” Also. the initial chord distribution was obtained using Equation (2). etc. chord distribution. depending on the relative wind that a given portion of the blade would “see. was approximately 45. Finally. Of these options. After the airfoil shapes were tested and the best kept for further testing. the chord distribution was altered in hopes that a more optimal design could be found.7 meters long.e.
Given that limit.5 rpm (1.225 3 m 4 s (7) Therefore.3 ⋅1. Substituting this value into Equation (1). from the above calculation. along with the given rotation rate of 17. the maximum theoretical output of a wind turbine.3 ⋅ ρ 4 V0 2 4 kg π ⋅ (91.709 MW = 0. The theoretical limit of the maximum amount of power that can be extracted from a wind turbine is known as the Betz limit and is approximately 60%. the power output was computed to be 10367 kW. This design created a moment about the rotational axis of 2.833 rad/sec). with the root directed into the natural wind direction and the tip approximately aligned with the plane of rotation. The kinetic energy in the wind is given by 1 π D 2 2 V KE = ρ 2 4 0 (6) The power available can then be shown to be equal to the kinetic energy of the wind multiplied by its velocity and the Betz limit. using the final design’s given size and wind speed is approximately 4.4 m) 2 m (12. tapering linearly to 2 meters at the tip. an analysis is possible to see if the amount of power that can be extracted from the wind has been overpredicted.5 ) 3 = 4.6 ⋅V0 ρ V0 = 0. 1 π D2 2 π D2 3 P = 0.83 x 106 Nm. The blade was highly twisted.with a chord length of 10 meters at the root. the blade underwent a major initial change in pitch—60 degrees in pitch in the first 25% of the blade length.7 megawatts. .
Figure 2: Final Wind Turbine Blade Design . The changing chord length and twist distribution along the blade can easily be seen.A picture of our final turbine is shown in Figure 2.
Figure 3 is a contour plot the cp distribution of the wind turbine blade near the root. Figure 3: Cp Distribution for Final Wind Turbine Blade Design . The picture also suggests a possible overprediction of cp along the upper surface of the rotor. The 3D effects of leakage near the root are evident.
Figure 4: Wake of Final Wind Turbine Blade Design . Evident is the helix shape as well as the rollup of the vortices near the tip.The results of the animated wake are shown below in Figure 4.
. The design process began by using technical data from a realworld wind turbine. to model a baseline design. Ultimately. ultimately resulting in a highly optimized final design. After each design iteration. Not only did this method allow performance improvements. as well as other previous design iterations. but it provided an understanding of many of the fundamentals of wind turbine design and qualitative effects of different aerodynamic changes. the power output was computed and compared to both the baseline model. a model with the highest power output was chosen as the final design.V. Conclusion The aerodynamic design of this large horizontalaxis wind turbine was comprised of several steps. Design changes and modifications continued in hopes of improving the performance even further. the Boeing Mod2. From this design. different aerodynamic parameters were varied in order to achieve a more optimal design.
Small Wind Turbine Design Notes. Stoddard. Retrived Dec 3.aber. 2003. and Forrest S.uk/iri/WIND/TECH/WPcourse. References 1 Eggleston.gov/windandhydro/wind_how. Retrived December 18. US Department of Energy.energy. from http://users. New York: Van Nostrand Reinhold Company. Wind and Hydropower Technologies Program: How Wind Turbines Work. David M. Hugh.eere.html> . 1987. 2003.VI.ac. 2 Piggot. Wind Turbine Engineering Design. 3 US Department of Energy. <http://www.
SREF=274. THEMAX=0.0. NORF=0.5.00393 0. RMPZ=0.in > TWIST CHANGE (1) Airfoil = NACA 65410 Cbar = 6.84924 0 0. THETA=0.0.0. RCORE=0.0. INMODE=4.0. LSTOUT=1.0 &END &BINP9 CBAR=6.0.. &END &END &END &END &END MAXIT=600. CTHET=0.00.0. Appendix A.5 [m/s] &BINP2 &BINP3 &END &BINP4 &BINP5 &BINP6 &BINP7 &BINP8 &BINP8A LSTINP=2. ALDEG=0. NODEA=5. ATHET=0.0. SSPAN=45.&SECT1 STX=0.0 [m] R = 45.00. ALF=6. &COMP1 COMPX=0. YAWDEG=0.1.0. NEWSID=0.01211 0.1.79912 0 0. KSIDE=0. LENRUN=0. NORL=0. COMPY=0. NTSTPS=60. NCZPAN=0. RSYM=1.00.00327 0. PHIDOT=105.0. COMPOP=0.0. STZ=0.0.02004 0. &END . LSTGEO=1. UNIT=1.08.VII. KCOMP=1. CZDUB=0.0. WRY=0.0.00.7. SOLRES=0.94971 0 0.00. SCALE=10.RIGHT BLADE .0.64927 0 0. PHIMAX=0. ASCAL=1.0000.0 WRZ=0. &BINP11 NORPCH=0. VSOUND=340. STY=0.00000 0 0. WTY=0..00.01621 0. ASEMY=0.00226 0. RMPY=0. VREF=0.. CSCAL=1.74910 0 0. LSTCPV=0.2.00000 0. &BINP12 KPAN=0.1 [deg] Alpha at Tip = 111.0. TNODS=0. COMPZ= 0.0. &BINP13 NBLIT=0.0000. TNPS=0.00.0.6 [deg] > Twist is nonlinear Phi dot = 105 [deg/sec] Vinf = 12.NACA 65410 AIRFOIL .00037 0. &END .0 WTZ=0.. PSIDOT=0. LSTNAB=0.0. &BINP10 NORSET=0.7 [m] Alpha at Root = 6. VNORM=0. DTSTEP=0. NBCHGE=0. RMPX=0. RGPR=0. WRX=0. &END &END &END &END &END &END &END &PATCH1 IREV=0.0 &END &BINP8B DXMAX=0.00.00792 0.. NOCF=0.modH. MAKE=0. NEWNAB=0. VINF=12. DZMAX=0.005.54971 0 0..89943 0 0. ASEMZ=0. NODEC=5.02340 .0000.0.000.0.1 – CMARC Input File for Best Wind Turbine (ModH) MODIFIED TURBINE H . DYMAX=0. WTX=0. LSTFRQ=1. NCZONE=0. TINTS=0. NOCL=0.0. KASS=1.69915 0 0.59947 0 0.05. RFF=5. PSIMAX=0. IDPAT=1. THEDOT=0. LSTWAK=3. &END LSTJET=0. &ASEM1 ASEMX=0.
02318 0.24843 0.0.06983 0.00861 0.05099 0.05741 0. &SECT1 STX=0. SCALE=2.8.01372 0.03487 0. TINTS=3. KASS=1.04372 0.0.35097 0. &PATCH1 IREV=0.34903 0.70085 0. IDPAT=1.6.4. INMODE=0. TINTS=3. TINTS=3. THETA=0.02710 0.03577 0.7. TNPS=14.0.0.07138 0.0.02854 0.25157 0.7.09788 0. STZ=3. TNODS=0.01411 0.07289 0.07711 0.01842 0. SCALE=8.07153 0.05006 0.85076 0. STY=11. SCALE=4.06702 0.0.01999 0.00781 0.06720 0.3.07018 0.95029 1.39936 0.02547 0. STZ=1.0. TNPC=0.06290 0.0. ALF=66.02863 0.55029 0.00893 0.00937 0. TINTS=3. INMODE=0.45032 0. INMODE=0. STY=22.44968 0.65073 0.04797 0. ALF=83. &END &SECT1 STX=0.04067 0.01791 0.01191 0. STY=45.0.4.06288 0.00372 0. THETA=0.19817 0.00000 0.00000 0.05203 0. STZ=4.14798 0.00661 0. TNODS=0.02814 0.40064 0.0.0.75090 0. STZ=2.00944 0. &END &SECT1 STX=0. TNPS=14.29872 0. THETA=0.20183 0.02800 0. ALF=93.02729 0. ALF=111.60053 0.50000 0.15202 0.02682 0. TNODS=0.90057 0. STY=34.02606 0. &END &END &END &END .9. MAKE=1.05731 0.02314 0.00000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.50000 0.00000 &BPNODE TNODE=3. THETA=0. TNPS=14.80088 0.01061 0. &SECT1 STX=0. INMODE=0. KCOMP=1.02773 0.00628 0. TNODS=3.01536 0. TINTC=0.0.0.01935 0. SCALE=6. TNPS=14.00607 0.10212 0.30128 0.01089 0.
KWPAN1=0. ... NSTLIN=0. NODEW=5. TINTS=3. &END &END &WAKE1 IDWAK=1.RIGHT TIP . KWLINE=0. &END &END &END &END . &ONSTRM &BLPARAM &VS1 &SLIN1 NONSL=0 KPSL=0 RN =1640000. INITIAL=0..&WAKE2 KWPACH=1.. IFLXW=0. VISC =0.RIGHT WAKE . TNODS=5. KWSIDE=2. NVOLR=0.00016... TNPS=3. NVOLC=0. NSLBL = 1. NPTTIP=0.&PATCH2 ITYP=1. &END KWPAN2=0..