5 Wind Energy CH 5

5
Wind Turbine Design
5.1 INTRODUCTION
In wind turbine design, many areas require attention. Table 5.1 shows some of the major
requirements. One of the first tasks is to spell out the basic requirements for the machine to be
built. The design process can then proceed creatively to meet these needs in the least expensive and
most efficient way. It is not possible to satisfy all of them on the first try. As with any design, the
process is iterative.
TABLE 5.1 Design tasks
Rotor: Tip-speed ratio, solidity, number of blades, aerodynamic optimization, static and dynamic
operating loads, parked rotor loads, material selection, manufacturing process, structural
dynamics, fatigue, starting torque vs. friction torque, primary over speed control, secondary
over speed control, blade tower clearance, brake system, yaw control, etc.
Tower: Height, type: pole, truss or tilt up, structural loads, strength, structural dynamics, tower
shadow, erosion protection, etc.
Generator: Type: ac (synchronous, 3f, 1f), or dc (alternator, generator), size, weight, efficiency curves,
speed-torque characteristics, power conditioning, excitation, etc.
Gearbox: Ratio, maximum speed, torque capacity, strength and load deflections, noise, structural
dynamics, lubrication, etc.
(Contd.)
125
126 Wind Energy : Theory and Practice
TABLE 5.1 Design tasks (Contd.)
Control system: Mechanical and/or electrical system, control algorithm, power supply, consequence
of failure, start-up and shut-down transients, wind speed and direction sensors,
reliability, failure analysis, lightning protection, etc.
General: Can it be simplified?, system dynamics, shipping and erection, installation method,
maintenance, aesthetics, safety, sensitivity to vandalism, corrosion protection, specifications
and quality control, etc.
Cost: Design life, development cost, cost per kWh of power produced, cost per kW installed, tax
benefits, rate of return, and pay back period.
The design of a wind turbine is an exciting creative process. Extracting energy from wind is
basically a simple conversion of energy, but the mastery of essential disciplines required to build a
truly efficient, reliable, and competitively priced wind turbine is an immense undertaking. It is
most convenient to have the support of technical experts in all the allied fields, such as aerodynamics,
electrical machinery, structural dynamics, manufacturing processes, and the like.
5.2 ROTOR TORQUE AND POWER

The aerodynamic lift (and drag) forces on the spanwise elements of radius r and length dr of the
several blades of a wind turbine rotor are responsible for the rate of change of axial and angular
momentum of all of the air which passes through the annulus swept by the blade elements. In
addition, the force on the blade elements caused by the drop in pressure associated with the rotational
velocity in the wake must also be provided by the aerodynamic lift and drag.
It is assumed that the forces on a blade element can be calculated by means of two-dimensional
aerofoil characteristics using an angle of attack determined from the incident resultant velocity in
the cross-sectional plane of the element; the velocity component in the span-wise direction is ignored.
Having information about how the aerofoil characteristic coefficients Cd and Cl vary with the angle
of attack, the forces on the blades for given values of a and a¢ can be determined.
Consider a turbine with B blades of tip radius R, each with chord c and set pitch angle b
measured between the aerofoil zero lift line and the plane of the disc. Both the chord length and
pitch angle vary along the blade span. Let the blades be rotating at angular velocity W and let the
wind speed be U¥. The tangential velocity of the wake a¢Wr means that the net tangential flow
velocity experienced by the blade element is (1 + a¢)Wr. Figure 5.1 shows all the velocities and
forces relative to the blade chord line at radius r.
From Figure 5.1, the resultant relative velocity at the blade is
W U 2 (1 a)2 : 2 r 2 (1 a)2 (5.1)

which acts at an angle f to the plane of rotation, such that
U (1 a) : r (1 a)
sin G and cos G (5.2)
W W
The angle of attack a is given by
a=f–b (5.3)
Wind Turbine Design 127
Wra¢
dr
r
–a)
U(1
Wr
r
FIGURE 5.1 A blade element sweeps out an annular ring.
FIGURE 5.2 Blade element velocities and forces.
The lift force on a span-wise length dr of each blade, normal to the direction of W, is therefore
1
EL SW 2 cCl E r (5.4)
2
and the drag force parallel to W is
1
SW 2 c C d r E r
ED (5.5)
2
The component of aerodynamic force on B blade elements resolved in the axial direction is
1
E L cos G E D sin G SW 2 Bc (Cl cos G Cd sin G ) E r (5.6)
2
The rate of change of axial momentum of the air passing through the swept annulus is
S U (1 a) 2Q r E r 2aU 4QS U 2 a (1 a) r E r (5.7)
The drop in wake pressure caused by wake rotation is equal to the increase in dynamic head, which
is
1
S (2 a:r ) 2
2
Therefore, the additional axial force on the annulus is

1
S (2a:r )2 2Q r E r
2
Thus,
1
SW 2 Bc (Cl cos G C d sin G ) E r 4Q P [U 2 a (1 a) (a :r ) 2 ] r E r (5.8)
2
By simplifying, we get
W2 c
B (Cl cos G Cd sin G ) 8Q {a(1 a) ( a MN )2 }N (5.9)
U2
R
r
where parameter N .
R
The element of axial rotor torque caused by aerodynamic forces on the blade element is
1
(E L sin G E D cos G ) r SW 2 Bc (Cl sin G C d cos G ) r dr (5.10)
2
The rate of change of angular momentum of the air passing through the annulus is
SU (1 a) : r 2a r 2Q r E r 4Q SU (:r ) a (1 a) r 2 E r

Equating the two moments, we have
1
SW 2 Bc (Cl sin G Cd cos G ) r E r 4Q S U ( : r ) a (1 a) r 2 E r (5.11)
2
On simplification, we have
W2 c
B (Cl sin G C d cos G ) 8Q M N 2 a (1 a) (5.12)
U 2
R
It is convenient to put
Cl cos f + Cd sin f = Cx
and Cl sin f – Cd cos f = Cy
Solving Eqs. (5.9) and (5.12) to obtain values for the flow induction factor a and a¢ using two-
dimensional aerofoil characteristics requires an iterative process. It gives,
a Tr Ë Tr Û
2 Ì
(C x ) CY2 Ü (5.13)
1 a 4 sin G Í 4 sin 2
G Ý
a T r CY (5.14)
1 a 4 sin G cos G
Blade solidity s is defined as the total blade area divided by the rotor disc area and is a primary
parameter in determining rotor performance chord solidity sr is defined as the total blade chord
length at a given radius divided by the circumferential length at that radius, i.e.
B c N c
Tr (5.15)
2Q r 2Q N R
The calculation of torque and power developed by a rotor requires a knowledge of the flow induction
factors, which are obtained by solving Eqs. (5.13) and (5.14). The iterative procedure is to assume
a and a¢ to be zero initially, determining f, Cl and Cd on that basis, and then to calculate new values
of the flow factors using Eqs. (5.13) and (5.14). The iteration is repeated until convergence is
achieved.
From Eq. (5.13), the torque developed by the blade element of span-wise length dr is
EQ 4QSU (:r ) a (1 a) r 2 E r
If drag, or part of the drag, has been excluded from the determination of the flow induction factors,
then its effects must be introduced when the torque caused by drag is calculated from blade element
forces (see Eq. (5.10)), i.e.
EQ 4QSU (:r ) a (1 a) r 2 E r 1 SW 2 BcCd cos G r E r

2
The complete rotor, therefore, develops a total torque Q:
1 Ë R Î W B (c / R ) Þ Û
Q
2
SU2 Q R3 M Ì
Ì
Í
Ô
0
N 2 Ï8a (1 a) N
Ð U Q
Cd (1 a) ß d N Ü
à ÝÜ
(5.16)
The power developed by the rotor is

P = QW
The power coefficient is
P
CP
1
S U3 Q R2
2
Solving the blade element momentum, Eqs. (5.13) and (5.14) for a given, suitable blade geometry
and aerodynamic design yields a series of values for the power and torque coefficients, which are
function of the tip speed ratio. A typical performance curve for a modern, high-speed wind turbine
is shown in Figure 5.3.
The maximum power coefficient occurs at a tip speed ratio for which the axial flow induction
factor a, which in general varies with radius, approximates most closely to the Betz limit value of
1/3. At lower tip speed ratios, the axis flow induction factor can be much less than 1/3 and aerofoil
angles of attack are high leading to stalled conditions. At high tip speed ratios, a is high, angles of
attack are low, and drag begins to predominate. At both high and low tip speed ratios, therefore,
drag is high and the general level of a is non-optimum so the power coefficient is low. Clearly, it
would be best if a turbine can be operated at all wind speeds at a tip speed ratio close to that which
gives the maximum power coefficient.
0.5
0.4
0.3
Cp
0.2
0.1
0
0 5 10 15
l
FIGURE 5.3 Power coefficient-tip speed performance curve.
5.2.1 Glauert Momentum Vortex Theory

Glauert momentum vortex theory assumes that small
radial sections of the blade can be analysed
independently. Consider an annular section of the rotor
and examine a small section of length Dr of one blade.
The net effect on air flowing this annular section of the
rotor results from the forces and moments on all the
blades. Let B represent the number of blades.
The geometry of a blade element showing the
velocities, forces and moments is depicted in Figure 5.4.
The relative wind at the rotor W varies with blade radius
r and consists of an axial component U and a rotational
component (rWa¢ + rW). The term rW represents the
velocity caused by blade rotation, whereas (rWa¢)
portrays the swirl velocity of the air. The angle of the
blade chord with respect to the plane of rotation is FIGURE 5.4 Velocity diagram.
denoted by q. The angle of attack of the airfoil with
respect to the local relative wind, W, is denoted by a, and the angle of the relative wind with
respect to the rotor plane, by f. The drag force, D, is aligned with the relative wind, W, while the
lift force, L is perpendicular to W.
The force component,
FQ = L sin f – D cos f
in the direction of blade rotation generates useful torque, whereas the component,
FT = L cos f + D sin f
in the downwind direction exerts a thrust load on the rotor. In terms of dimensionless coefficients
Cl and Cd the net force, power and torque caused by B blades, each of local chord c, are as follows:
For torque:
1
'Q SW 2 r [Cl sin G C d cos G ] Bc 'r (5.17)
2
For power:
1
'P :'Q SW 2 r : [Cl sin G Cd cos G ] Bc 'r (5.18)
2
For thrust:
1
'T SW 2 [Cl cos G Cd sin G ] Bc 'r (5.19)
2
where
U r : r :a
W (5.20)
sin G cos G
In the elementary actuator disc theory, we found that the induced velocity was twice as large in the
far wake as in the rotor plane, and this same behaviour is assumed for each streamtube. It is assumed
that the air has acquired half of its final rotational (swirl) velocity when it reaches the rotor.
With these assumptions, the linear momentum loss gives the incremental thrust force DT on
the rotor as
'T Su (2Q r dr ) 2 (V0 U ) (5.21)
and the incremental torque DQ in terms of the rate of change of rotation momentum as
'Q S u (2Q r dr ) 2Wr (5.22)
Equating the expressions for thrust and torque from momentum theory to those from aerodynamic
forces on a blade element leads to
V0 U È Bc Ø È Cl cos G Cd sin G Ø
É Ù (5.23)
U Ê 8Q r Ú É
Ê sin 2 G Ù
Ú
and
W È Bc Ø È Cl sin G Cd cos G Ø
r:a :r
É Ù
Ê 8Q r Ú ÉÊ sin G cos G Ù
Ú
(5.24)
V0 U W
Using dimensionless axial and radial induction factors, a and a and solidity
V0 :r
Bc
T , Eqs. (5.23) and (5.24) become
QR
a È T R Ø È Cl cos G C d sin G Ø
(5.25)
(1 a) ÊÉ 8r ÚÙ ÉÊ sin 2 G Ù
Ú
a È T R Ø È Cl sin G Cd cos G Ø
and É Ù (5.26)
(1 a ) Ê 8r Ú ÉÊ sin G cos G Ù
Ú
U V0 (1 a) 1a
Also, tan G (5.27)
:r r :a :r (1 a) x (1 a )
where x = Wr/V0 is the local speed ratio. At the end of the blade, r becomes R, and the tip speed
ratio (TSR),
R:
X (5.28)
V0
Using X,
È R Ø Ë (1 a) Û
tan G É ÙÌ Ü (5.29)
Ê rX Ú Í (1 a ) Ý
The two-dimensional lift and drag coefficients and a set of relations may be solved iteratively
to find a and a¢ for any pitch angle q. Instead of average solidity, define a symbol called the blade
loading coefficient, l = BcCl/8pr. The parameter l is one-fourth of the average retarding pressure
the blade exerts on the air flowing through the annular streamtube, normalized by the relative
dynamic pressure at the blade element.
From Eq. (5.19),
1
'T SW 2 [Cl cos G Cd sin G ] Bc 'r
2
1 Ë ÈC Ø Û
SW 2Cl Ìcos G É d Ù sin G Ü Bc 'r
2 ÌÍ Ê Cl Ú ÜÝ
Cd
If F
Cl
1
'T SW 2 ( BcCl ) [cos G F sin G ] ' r
2
For small f and small e, cos f @ 1 and e sin f @ 0. Thus,
1
'T # SW 2 (BcCl ) 'r
2
as the retarding force exerted by the blade.
In this, DT is divided by the annular area, 2prDr, to get the average pressure exerted by the
blades on the annulus, and normalized by ½rW2, so
'T BcCl
4M
È 1 2Ø
(2Q r 'r ) É SW Ù
2Q r
Ê2 Ú
BcCl
Here l is blade loading coefficient M
8Q r
The corresponding normalized average local pressure on the blade element would be
4 M (2Q r 'r ) 4M
Cl
Bc 'r Tl
Bc
where sL is local solidity defined as TL
Qr
Thus, increasing l increases the amount of slowing of the air and the pressure on the blade,
whereas when l goes to zero, the blade does not retard the wind. Using l and e and dividing the
right-hand side of Eqs. (5.25) and (5.26) by sin f and cos f, respectively, we obtain
a (cot G F )
M (5.30)
(1 a) sin G
a
and M (tan G F ) / sin G (5.31)
1 a
These equations are simplest possible and are quite insensitive to the value of e for high L/D
airfoils. For e = 0 gives values of a with less than one per cent error for 0 < f < 40 degree when
compared with using the exact equation with e = 0.01. These relationships are shown graphically
in Figure 5.5 plotted for e = 0.01. It is very useful for quick estimates of changes in rotor behaviour,
such as sudden loss of Cl, start-up when stalled, etc.
FIGURE 5.5 Blade element parameters.
The iterative solution procedures, given r/R, c, Cl(a), Cd(a), q and Vo, are as follows:
• Guess values of a and a¢.
• Calculate f from Eq. (5.29).
• Calculate a = f – q and hence Cl and Cd.
• Update a and a¢ from Eqs. (5.25) and (5.26).
The procedure allows for the evaluation of local forces and moments on the rotor, which can
then be integrated to find the total torque and loads. This provides a method of estimating performance
once a candidate blade layout is known. One can iteratively change the blade layout and optimize
the blade.
The thrust and torque coefficients are defined as
T
CT (5.32)
½ ( SV02 ) (Q R 2 )
Q
and CQ (5.33)
S (Q R )(:R)2 R
2
Following Miller et al., these equations can be reduced to the following form for integrating
along the blade length:
ÈrØ È rØ
dCT 8a (1 a) É Ù d É Ù (5.34)
Ê RÚ Ê RÚ
3
4Ø È ÈrØ ÈrØ
and dCQ Ù a (1 a) É Ù d É Ù
É (5.35)
X Ú Ê Ê R Ú Ê RÚ
The Prandtl correction function is conveniently expressed as follows:
È 2Ø
F É Ù cos1 exp ( f ) (5.36)
ÊQÚ
È1 Ø
where f (r ) É R Ù ( R r ) B (1 X 2 ) (5.37)
Ê2 Ú
The function, F(r), approaches unity for 1 – (r/R) < 2/XB but goes rapidly to zero at tips where r = R.
The Prandtl correction for finite number of blades requires that the velocity at the rotor disc be
U = V0(1 – aF)
and the axial and rotational interference factors a and a¢ must now be determined from the following
equations:
p1
a (5.38)
1 p1
p2
a (5.39)
1 p2
where
T (Cl cos G Cd sin G )
p1 (5.40)
8F sin 2 G
T (C l sin G C d cos G )
p2 (5.41)
8F sin G cos G
5.2.2 Optimal Rotors

If Cd is assumed to be zero, dividing Eq. (5.26) by (5.25) gives
a(1 a)
tan 2 G (5.42)
a(1 a)
and comparing this with Eq. (5.27) gives

a(1 a ) V02 1
(5.43)
a(1 a)r x2 :
2 2
At each radius, the right side of Eq. (5.43) is constant, and therefore the left side must remain
constant also, while the power is maximized if the quantity a¢(1 – a) from Eq. (5.42) is maximized.
Performing these operations using Lagrange multipliers yields:
(1 3a)
a (5.44)
(4a 1)
In the wind turbine state, a¢ must be positive for positive output torque. Thus, for small x, a approaches
1/4 and a¢ becomes large, whereas for large x, a approaches 1/3 and a¢ approaches zero.
Substituting a¢ from Eq. (5.44) into Eq. (5.43) yields the required relationship between a and
local speed ratio x as follows:
1a
x (4a 1) (5.45)
1 3a
Miller et al. give a power series in 1/x2 as an approximation for the inverse relationship:
È 1 Ø È 2 Ø È 10 Ø È 418 Ø
a#É Ù É Ù É Ù É Ù ... (5.46)
Ê 3 Ú Ê 81x Ú Ê 729 x Ú Ê 59049 x 2 Ú
2 2
The corresponding relative wind angle f, can be found from Eq. (5.42) as given by,
È1Ø
tan G ÉÊ ÙÚ (1 a)(1 3a) (5.47)
a
The optimum blade layout in terms of the product of chord c and lift coefficient Cl can be found
from Eq. (5.24) with CD = 0
T Cl X T L Cl x È B: Ø Ë (4 a 1) Û
( cC l ) Ì Ü (1 a)(1 3a) (5.48)
Ê 8Q V0 Ú
É Ù
8 4 Í (1 2 a) Ý
As mentioned previously, there is still some design freedom in that both c and Cl may be varied
while their product satisfies Eq. (5.48). Thus, if chord c is held constant, Cl (and hence a and q)
will follow from Eq. (5.48). Likewise, if Cl (and hence a) is held constant, chord c must vary
according to Eq. (5.48), and twist angle q has been specified as a function of r.
Miller et al. also integrate the CP relation along the blade to give closed form and series solution
for CP. Thus, from Eqs. (5.31), (5.44) and (5.45),
2
È 12 Ø Ë (4a 1) (1 a) (1 2a) Û
dCP É ÙÌ Ü da (5.49)
Ê X2 Ú Í (1 3a) Ý
which must be integrated from a = 1/4 at r = 0 to the value of a at the tip, aT, given by
(1 aT )
X (4 aT 1) (5.50)
(1 3aT )
The integration can be done in closed form, resulting in

È 4 Ø ËÈ 4 Ø È 1 Ø È 4363 Ø 2 3 È 65 Ø 5 Û
CP É Ù ÌÉ Ù 12 ln É Ù É Ù 63 z 38 z 124 z 74z 4 É Ùz Ü
Ê 729 X Ú ÍÊ z Ú
2 Ê 4 z Ú Ê 160 Ú Ê 5 Ú Ý
(5.51)
where z = 1 – 3aT. A series expansion valid for tip speed ratio X is
È 8 Ø Ë È 2 Ø È (27 X 2 15) 0.0529 ØÛ
CP É Ù Ì1 É Ù ln É ...Ù Ü (5.52)
Ê 27 Ú Ì Ê 9X 2 Ú Ê 8 X2 ÚÜ
Í Ý
For any tip speed ratio, X, the local speed ratio is x = Xr/R. The parameter a is determined from
Eq. (5.46), and therefore f from Eq. (5.47), the product cCl from Eq. (5.48), and CP from one of the
equations (5.49) through (5.52).
To obtain a single-point optimum including the effects of drag, Stewart begins by deriving a
local power coefficient C P as
SW 2r : Bc (Cl sin G C d cos G ) 'r

1
CP
'P 2 (5.53)
È 1 SV 3 'A Ø Ë 1 SV 3 (2Q 'r ) Û
ÊÉ 2 0 ÚÙ ÌÍ 2 0 ÜÝ
which, by using W 2 V02 (1 a)2 r 2 :2 (1 a)2 and Eq. (5.27) to eliminate x2, reduces to
C p (1 a)2 (1 cot 2 G ) 4 x M (sin G F cos G )
Then, eliminating l using Eq. (5.30) and expanding 1/(cot f + e) in Taylor’s series of two terms,
there results.
CP = 4xa(1 – a)(tan f – e)(1 + e tan f) (5.54)
The last term, (1 + e tan f) is quite small, improving the accuracy less than 1 per cent for e = 0.01
and f = 40 degree. Thus, in most cases it may be neglected, leaving
CP = 4xa(1 – a) (tan f – e) (5.55)
È 27 Ø
By defining a local Froude efficiency, IF ÉÊ ÙÚ CP , we can relate the performance of each
16
blade to the ideal value of unity.
By dividing Eq. (5.31) into Eq. (5.30) and using Eq. (5.27) to eliminate a¢, there results
Ë È tan G F Ø Û
x tan G 1 a Ì1 É ÙÜ (5.56)
Í Ê cot G F Ú Ý
For small e, this is closely given by

x tan f = 1 – a sec2 f (5.57)
If the function, F = a(1 – a)(tan f – e), is then maximised using dF/df = 0, with relation between a
and f given implicity by Eq. (5.56), there results a quadratic equation as follows:
Ë a tan 2 G Û Ë (1 a) Û Ë F sec 2 G Û
Ì ÜÌ Ü 2Ì Ü (5.58)
Í (1 a) Ý Í a Ý Í (tan G F ) Ý
Solving this quadratic equation, the optimal value of a inducing the effect of drag is given by
È 1Ø
É
Ê ÙÚa
2 sec G G G H 2 (5.59)
F sec G tan G
where G and H .
2 (tan G F ) (tan G F )
Thus, a complete single-point optimization of a rotor-blade element, given f and e, proceeds
as follows:
1. Find a from Eq. (5.59).
a sin G
2. Calculate l from M , Eq. (5.30)
(1 a) (cot G F )
a (tan G F )
3. Find a¢ from a , modified Eq. (5.31)
(1 a) (cot G F ) a (tan G F )
(1 a) Ë È tan G F Ø Û
4. Then, x Ì1 Ü , Eq. (5.56)
tan G ] Í ÉÊ cot G F ÙÚ Ý
5. C P 4 xa (1 a) (tan G F ) (1 F tan G )
È 27 Ø
6. IF É Ù C P
Ê 16 Ú
Since hF is the local Froude efficiency of the blade element, the optimization of the power output
from this one blade element is then complete. A plot of hF versus x from various drag/lift ratio, e,
is given in Figure 5.6.
e=0
1.0
0.01
0.8
Local froude efficiency, hF
0.02
0.6 0.0
3
0.0
0.4 4
0.0
5
Cd Drag
0.2 e= =
Cl Lift
0
0 5 10 15 20 25
Local speed ratio, X = r W/V0
FIGURE 5.6 Local Froude efficiency versus local speed ratio.

Miller et al. give an analysis for optimal rotor including the drag effect but neglecting wake
rotation. In their analysis, a drag function, Cd = Cl + C2a2, is given, and the following two parameters
are defined:
16C2 T CL X 3
[ and E (5.60)
T XA2 8
The results of the optimization give
2
aopt
(5.61)
3 [ (3 [ ) 2 8[
È RØ Ë È 8 ØÛ
R opt É Ù (1 aopt ) Ì1 aopt É ÙÜ (5.62)
Ê Xr Ú Í Ê T XA Ú Ý
È RØ
Gopt É (1 aopt ) (5.63)
Ê Xr ÙÚ
È 1 Ø
CP ,max 4 aopt (1 aopt ) É1 [ aopt Ù 2E (5.64)
Ê 2 Ú
5.2.3 Dual Optimum

To find a blade chord c such that a blade element will satisfy the optimal conditions at two speed
ratios, x1 and x2, where x1 > x2. Then, if Dl = l2 – l1 and Df = f2 – f1.
È Bc Ø È Bc Ø
'M É Ù (CL 2 CL1 ) É Ù ao ( 'G U ) (5.65)
Ê 8Q r Ú Ê 8Q r Ú
where ao is the lift-curve slope of the airfoil in appropriate units, t is a possible change in the pitch
angle of the entire blade between the two conditions, and a value of t > 0 implies that l2 > l1.
Since we can calculate the ideal l and f for each speed ratio x, we can calculate Dl and Df.
This permits determination of an optimal value of Bc/8pr, and hence the dual optimum c and the
desired lift coefficients at x1 and x2 may be found.
EXAMPLE 5.1 Suppose at the tip we wish to have a blade chord optimal at both X1 = 6 and
X2 = 4. The calculations are best done in the tabular form shown in Table 5.2(a). Here all angles are
in degrees so that the lift-curve slope is taken to be approximate ao @ 0.1. For convenience, the
values of r/R along the blade are in the ratio 4/5 = 0.8. The angle f for a particular value of x can be
found from Eq. (5.57). Since the inverse of this relation is needed, table of x = x(f) would be
helpful, or a short programme using Newton‘s method might be used. The last two rows of
Table 5.2(a) give the differences, Dl and Df, for the two conditions.
Once these values have been found, blade chords satisfying the dual optimum conditions are
easily calculated, as in Table 5.2(b) for t = 0 and Table 5.2(c) for t = 2 degrees. The Bc/8pR row
shows the variation of chord with blade radius. By allowing a pitch change, a blade with a smaller
chord and a desirable taper can be found, as in Table 5.2(c). The design lift coefficients of this
blade are higher in proportion to the reduction of chord from those of Table 5.2(b).
TABLE 5.2(a) Optimum parameters for X1 = 6, X2 = 4, e = 0
r/R 1 0.8 0.64 0.512 0.4096 0.32768

x1 6 4.8 3.84 3.0720 2.4576 1.9661
l1 0.006055 0.009360 0.014388 0.021928 0.033003 0.048796
f1 6.3082 7.8455 9.7311 12.0207 14.7610 17.9727
x2 4 3.2 2.56 2.048 1.6384 1.3107
l2 0.013307 0.020317 0.030660 0.045500 0.065994 0.92914
f2 9.3574 11.5693 14.2245 17.3502 20.9319 24.8943
Dl 0.007252 0.010957 0.016272 0.023572 0.032991 0.044118
Df 3.0492 3.7238 4.4934 5.3295 6.1709 6.9216
TABLE 5.2(b) Optimum blading for X1 = 6, X2 = 4, e = 0, t = 0
r/R 1 0.8 0.64 0.512 0.4096 0.32768

Bc/8pr 0.02378 0.02942 0.03621 0.04423 0.05346 0.06374
Bc/8 pR 0.02378 0.02354 0.02317 0.02265 0.02190 0.02089
C 11 0.255 0.318 0.397 0.496 0.617 0.766
C12 0.560 0.788 0.847 1.029 1.234 1.458
TABLE 5.2(c) Optimum blading for X1 = 6, X2 = 4, e = 0, t = 2 deg
r/R 1 0.8 0.64 0.512 0.4096 0.32768

Bc/8pr 0.01436 0.01914 0.02506 0.03216 0.04038 0.04945
Bc/8 pR 0.01436 0.01531 0.01604 0.01647 0.01654 0.01620
C 11 0.422 0.489 0.574 0.682 0.817 0.987
C 12 0.927 1.061 1.223 1.415 1.634 1.879
5.3 OPTIMUM DESIGN FOR VARIABLE OPERATION

A turbine operating at variable speed can maintain the constant tip speed ratio required for the
maximum power coefficient to be developed regardless of wind speed. To develop the maximum
possible coefficient requires a suitable blade geometry, the conditions for which will be derived.
For a chosen tip speed ratio l, the torque developed at each blade station is maximum if
d
8QMN 2 a (1 a) 0
da
Equation (5.12), giving
d 1a
a (5.66)
da a
From Eqs. (5.9) and (5.12), a relationship between the flow induction factors can be obtained.
Dividing Eqs. (5.9) and (5.12), we get
Cl
tan G 1
Cd MN a (1 a)
(5.67)
Cl a (1 a ) ( a MN ) 2
tan G
Cd
The flow angle f is given by

1 a
tan G (5.68)
M N (1 a)
Substituting Eq. (5.68) in Eq. (5.67) gives
Cl 1 a
1
C d M N (1 a ) M N a (1 a)
Cl 1 a a (1 a) ( a M N ) 2

Cd M N (1 a )
By simplifying, we get
Cl
(1 a) M N (1 a)
Cd M N a (1 a)
C a (1 a) ( a M N )2
M N (1 a) l (1 a)
Cd
Î Cl Þ
Ï (1 a) M N (1 a) ß {a(1 a) ( a MN )2 }
Ð Cd à
Î Cl Þ
ÏMN (1 a) (1 a) ß MN a (1 a) (5.69)
Ð Cd à
At this stage, the process is made easier to follow if drag is ignored. Equation (5.69) then reduces to
a(1 – a) – l2m 2a¢ = 0 (5.70)
Differentiating Eq. (5.70) with respect to a¢ gives
d
(1 2 a) a M2 N2 0 (5.71)
da
and substituting Eq. (5.66) into Eq. (5.71) gives
(1 – 2a)(1 – a) – l2m2a¢ = 0 (5.72)
Equations (5.70) and (5.72), together, give the flow induction factors for optimized operation,
1 a (1 a)
a and a (5.73)
3 M2 N2
which agree exactly with the momentum theory prediction because no losses, such as aerodynamic
drag, have been induced and the number of blades is assumed to be large; every fluid particle
which passes through the rotor disc interacts with a blade resulting in a uniform axial velocity over
the area of the disc.
To achieve the optimum conditions, the blade design has to be specific and can be determined
from either of the fundamental Eqs. (5.9) and (5.11). Choosing Eq. (5.11), because it is simpler,
and ignoring the drag, the torque developed in optimized operation is
U3
EQ 4QSU (:r ) a (1 a) r 2 E r 4QS a (1 a) 2 r E r
:
The component of lift per unit span in the tangential direction is therefore
U 3
L sin G 4QS a (1 a)2
:
By the Kutta–Joukowski theorem, the lift per unit span is
L = rWT
where G is the sum of the individual blade circulations.
Consequently,
U 3
SWT sin G S*U (1 a) 4QS a (1 a) 2 (5.74)
:
U 2
So, * 4Q a (1 a) (5.75)
:
The circulation is therefore uniform along the blade span and this is a condition for optimized
operation.
To determine the blade geometry, that is, how should the chord size vary along the blade and
what pitch angle b distribution is necessary, (see Eq. (5.12)).
W2 c
B C l sin G 8QN 2 M a (1 a)
2
U R
Substituting for sin f gives
W c
B Cl (1 a) 8QMN 2 a(1 a) (5.76)
U R
From which is derived
B c 4MN 2 a
2Q R W
Cl
U
The only unknown on the right-hand side of the above equation is the value of the lift coefficient Cl
and so it is common to include it on the left-side of the equation with the chord solidity as a blade
geometry parameter. The lift coefficient can be chosen as that value which corresponds to the
maximum lift/drag ratio Cl/Cd as this will minimize drag losses; even though drag has been ignored
in the determination of the optimum flow induction factors and blade geometry, it cannot be ignored
in the calculation of torque and power. Blade geometry also depends upon the tip speed ratio l, so
it is also included in the blade geometry parameter. Hence
N c 4 M 2 N 2 a
T r M Cl M Cl (5.77)
2Q R (1 a) 2 {MN (1 a )}2
Introducing the optimum conditions of Eq. (5.73),
8/9
T r MC l 2
(5.77a)
È 1Ø
2
Î 2 Þ
ÉÊ 1 Ù M 2 N 2 Ï1 2 2 ß
3 Ú Ñ 9( M N ) à
Ð Ñ
The parameter lm is called the local speed ratio and is equal to the tip speed ratio when m = 1. Over
the outboard, half of the blade, which produces the bulk of the power, the local speed ratio lm, will
normally be large enough to enable the denominator to be approximated as lm, giving
Bc( N ) 8
T r M Cl MCl (5.77b)
2Q R 9 MN
where B is the number of blades. After rearrangement, this gives
2
È :R Ø 16Q R
c( N ) ÉU Ù (5.77c)
Ê Ú 9C l B N
Hence it can be seen that, for a family of designs optimized for different rotational speeds at
the same wind speed, the blade chord at a particular radius is inversely proportional to the square of
the rotational speed, assuming that N and R are fixed and the lift coefficient is maintained at a
constant value by altering the local blade pitch to maintain a constant angle of attack.
If for a given design Cl is held constant, then Figure 5.7 shows the blade plan form for increasing
tip speed ratio. A high design tip speed ratio would require a long, slender blade (high aspect ratio),
while a low design tip speed ratio would need a short, fat blade. The design tip speed ratio is that at
which optimum performance is achieved. Operating a rotor at other than the design tip speed ratio
gives a less than optimum performance even in ideal drag free conditions.
FIGURE 5.7 Variation of blade geometry parameter with local speed ratio.
In off-optimum operation, the axial inflow factor is not uniformly equal to 1/3. In fact, it is not
uniform at all.
The local inflow angle f at each blade station also varies along the blade span as shown in
Eq. (5.78) and Figure 5.8.
1 a
tan G (5.78)
MN (1 a )
FIGURE 5.8 Variation of inflow angle with local speed ratio.
which, for optimum operation, is

1
1
tan G 3 (5.79)
È 2 Ø
MN É1 2 2 Ù
Ê 3M N Ú
EXAMPLE 5.2 Take NACA 4412 blade aerofoil, popular for hand-built wind turbines because
the bottom (high pressure) side of the profile is almost flat, which facilitates manufacture. At a
Reynolds number of about 5 × 105, the maximum lift/drag ratio occurs at a lift coefficient of about
0.7 and an angle of attack of about 3°. Assuming both Cl and a are to be held constant along each
blade and there are to be three blades operating at a tip speed ratio of 6, then the blade design in
plan-form and pitch (twist) variation are as shown in Figure 5.9(a) and (b), respectively.
0.4
c /R
0.2
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
r /R
(a)
Twist angle in degrees
30
20
10
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
r /R
(b)
FIGURE 5.9 Optimum blade design with l = 6.
The blade design of Figure 5.9 is efficient but complex to build. A straight line drawn through
the 70 per cent and 90 per cent span points as shown in Figure 5.10 not only simplifies the plan-
form but removes a lot of materials close to the root.
0.4
c /R
0.2
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
r /R
FIGURE 5.10 Uniform taper blade design for optimal operation.
cU 8 È MN Ø 2Q
ÉÊ 2 (5.80)
R 9M 0.8 M 0.8 ÙÚ Cl M B
The 0.8 in Eq. (5.80) refers to the 80 per cent point, midway between the target points.
Equations (5.77) and (5.80) can be combined to give the required span-wise variation of Cl for
optimal operation.
8 1
Cl
9 2
BcU M È 2 2 Î Þ
2
1Ø 2
2Q ÉÊ 1 3 ÙÚ M N Ï1 2 2 ß
Ñ 9(M N ) à
Ð Ñ
Close to the blade root, the lift coefficient approaches the stalled condition and drag is high,
but the penalty is small because the adverse torque is small in that region.
Assuming that stall does not occur and that for the aerofoil in question, which has 4 per cent
camber the lift coefficient is given approximately by,
Cl = 0.1 (a + 4 deg)
where a is in degrees, so
Cl
B 4 deg
0.1
The blade twist distribution can now be determined from Eqs. (5.79) and (5.80). The twist angle
close to the root is still high but lower than for the constant CL blade. The span-wise distribution of
the lift coefficient required for the linear taper blade is shown in Figure 5.11 and distribution of the
twist of blade is shown in Figure 5.12.
2
Lift coefficient
0
0.2 0.4 0.6 0.8 1.0
r /R
FIGURE 5.11 Span-wise distribution of the lift coefficient required for the linear taper blade.
Twist angle degrees

30
20
10
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
r /R
FIGURE 5.12 Span-wise distribution of the twist required for the linear taper blade.
For grid connected wind turbines, the rotational speed is maintained at constant values. Due to
variation in wind speed, the tip speed ratio continuously changes. In view of this, a blade optimized
for fixed tip speed ratio would not be appropriate. No simple solution is available for optimum
design of a blade or rotor running at constant rotational speed. With known wind speed distribution
at a particular site, a non-linear programming method could be applied for maximizing energy
capture for specified duration of time, say, in a year. Alternatively, a design tip speed ratio could be
chosen which corresponds to the representative wind speed at a site or average wind speed at a site.
More practical approach would be adjustment of pitch angle of blade for different ranges of wind
speeds to maximize energy capture. The variation of maximum coefficient of power would turbine
rotor with design tip speed ratio for a range of lift to drag ratio is plotted in Figure 5.13.
FIGURE 5.13 The variation of power coefficient with tip-speed ratio for different lift to drag ratios.
EXAMPLE 5.3 The turbine blade is designed for optimum performance at a tip speed ratio of 7,
with an angle of attack as 8° as an illustrated example. Assuming the maximum lift to drag ratio for
these conditions for NACA 632xx aerofoil. Further, the turbine is operating at constant rotational
speed with fixed pitch angle. The geometrical characteristics of aerofoil are given in Table 5.3 and
aerodynamic characteristics are shown in Figure 5.14. The power performance characteristics of
the variation of coefficient Cp with blade tip ratio l is shown in Figure 5.3.
TABLE 5.3 Blade Design of a 8.5 m radius rotor
Radius, r r/R Chord, c Pitch, b Thickness/chord

(mm) (mm) (degree) (%)
1700 0.20 1085 15.0 24.6
2150 0.30 1005 9.5 20.7
3400 0.40 925 6.1 18.7
4250 0.50 845 3.9 17.6
5100 0.60 765 2.4 16.6
5950 0.70 685 1.5 15.6
6800 0.80 605 0.9 14.6
7225 0.90 525 0.4 13.6
8500 1.00 445 0.0 12.6
FIGURE 5.14 Aerodynamic characteristics of NACA Aerofoil.
At lower tip speed ratio, the entire

blade is stalled as shown in Figure 5.15.
For a rotational speed of 60 r.p.m., the
corresponding wind speed is 25 m/s,
which is cut-out speed of turbine. For
the highest tip speed ratio, the
Corresponding wind speed is 4.2 m/s,
which is cut-in speed of turbine.
Maximum power is developed at a tip
speed ratio of 5 at a wind speed of
14 m/s. The variation of flow factors is
shown in Figure 5.16. The axial flow
factor increases with tip speed ratio,
while the tangential flow factor
decreases with tip speed ratio. FIGURE 5.15 Angle of attack distribution for a range of tip speed ratio.
FIGURE 5.16 Distribution of axial and tangential flow induction factor for a range of tip speed ratio.
At high tip speed ratio of l = 12, the effect of

drag reduces torque as the square of the local speed
ratio and also causes significant loss of power. The
dramatic effect of stall is illustrated for the
differences in torque distribution for the tip speed
ratio of 2 and 4 in Figure 5.17
It must be noted that the actual thrust force
acting on the blade will increase with wind speed as
illustrated by Figure 5.18(a) and axial force
coefficient with increasing tip speed ratio in
Figure 5.18(b).
FIGURE 5.17 Difference in torque distribution

for tip ratio of 2 and 4.
FIGURE 5.18 Axial force variation with wind speed and axial force coefficient variation with tip speed ratio.
5.4 INFLUENCE OF REYNOLDS NUMBER

The lift coefficient increases linearly with increasing angle of attack below the stall condition.
With the increase of Reynolds number the stall angle will increase and at the same time the maximum
value of the lift coefficient will also rises. The variation of Reynolds number determines the flow
around an aerofoil and significantly affects the values of lift and drag coefficients. The value of
drag coefficient will rise with decreasing value of Reynolds number. Below a critical Reynolds
number of about 200000, the boundary layer remains laminar and it causes rapid increase in the
drag coefficient. The influence of Reynolds number is illustrated in Figure 5.19 and Figure 5.20.
FIGURE 5.19 Variation of drag coefficient with increasing Reynolds number at low angle of attack for NACA aerofoil.
FIGURE 5.20(a) Variation of drag coefficient with Reynold FIGURE 5.20(b) Variation of lift coefficient with Reynolds
number in stall region for NACA aerofoil. number in stall region for NACA aerofoil.
5.5 CAMBERED AEROFOILS

Cambered aerofoils are asymmetrical about central chord line, which is a curved line to allow for
producing lift even at zero angle of attack. Cambered aerofoils have higher lift to drag ratio as
compared to symmetrical aerofoils for

positive angle of attack, and therefore are
used in wind turbine blades. A typical
cambered aerofoil NACA 4412 is shown in
Figure 5.21. The four-digit designation used FIGURE 5.21 Profile of NACA 4412 aerofoil.
for NACA aerofoil from left to right is as
follows: the first digit represents the amount of camber as a percentage of chord length, the second
digit the percentage chord position in units of 10 per cent at which the maximum camber occurs
and the last two digits are the maximum thickness to chord ratio, as a percentage of the chord
length. The curved camber line is made of two parabolic arcs which join smoothly at maximum
camber point. A straight line joining the ends of the camber line is called chord line and angle of
attack a is measured from it.
For cambered aerofoils, zero lift occurs at a small negative angle of attack of about 4°. The
characteristic behaviour of NACA aerofoil is shown in Figure 5.22.
Camber line
A°
A% chord
30% Chord
Zero lift line
10×B% Chord
FIGURE 5.22 Designation of NACA xxxx aerofoil.
For asymmetrical aerofoil, the centre of pressure lies at 1/4 of chord from the leading edge or
nose. For cambered aerofoils, the centre of pressure lies aft of 1/4 chord position and with increasing
angle of attack, further moves towards the trailing edge. The resultant force acting through the
centre of pressure point gives rise to pitching moment. When the pitching moment has the tendency
of moving the nose up, it is assumed to be positive by convention. A pitching moment coefficient
is defined as
Pitching/Unit span
Cm = (5.81)
1
SU c2
2
There will be a position, called the aerodynamic centre, for which
dC m
=0
dC l
For all practical purposes, the aerodynamic centre lies at the 1/4 chord position. The value of Cm
depends upon the degree of camber that is asymmetry and its value is 0.1 for NACA 4412. Above
the stall condition, the pitching moment becomes independent of a and, therefore, pre-stall position
continues to be used to determine the pitching moment coefficient.
FIGURE 5.23 Typical characteristics of NACA aerofoil.
5.5.1 Rotor Sizing

In a high-wind speed area, a relatively small rotor can develop a large amount of power. The
power loading per unit area of the rotor in watt/square metre is a useful index in comparing wind
turbines. Preliminary rotor sizing can be done with the elementary actuator disc momentum theory.
The power output of an ideal rotor is given by
1
P S A CPV03 (5.82)
2
where
r = air density
V0 = wind speed
A = rotor swept area
CP = power coefficient
Air density r varies with altitude and temperature. In SI units, this is as follows:
È 0.297 h Ø
S S0 exp É (5.83)
Ê 3048 ÙÚ
where r = 1.22496 kg/m3 and h in metres.
The temperature defined for the standard atmosphere is a linearly decreasing function of
altitude given by
È h Ø
T 15 1.983 É Ù C (5.84)
Ê 304.8 Ú
where h is in metres.
The maximum rotor power coefficient is CP = 16/27, whereas real rotors achieve considerably
lower power. With very precise, smooth airfoil blades and tip coefficient speed ratios above 10,
rotor power coefficients as high as 0.45 have been reported. For most machines, a power coefficients
of 0.3 to 0.35 would be possible with good design. With a drive train efficiency, hd, and a generator
efficiency, hg, the actual power output would be
1
Pout S ACP Id Ig V03 (5.85)
2
when this expression is solved for the rotor-swept area, A, we have
2 Pout
A (5.86)
S CP Id Ig V03
EXAMPLE 5.4 Find the area required for a rotor to generate 4.5 kW in a 5 m/s. Assume a drive
train efficiency of 0.85, a generator efficiency of 0.75, and an altitude of 92 m.
Solution: The standard atmospheric air density at 92 m would be
Ë ( 0.297)(92) Û
S 1.225 exp Ì Ü
Í 3048 Ý
= 1.214 kg/m3
and the standard atmosphere temperature would be
È 92 Ø
T 15 1.983 É Ù 14.4C
Ê 304.8 Ú
The required area would be
2(4500)
A
(1.214)(0.35)(0.85)(0.75)(53 )
= 265 m2 = pR2
D = 2R = 18.4 m.
5.6 LOAD CALCULATION

The simplified method for load calculation is from the Danish Standard DS 472. The dimensioning
loads of the wind turbine in the operational mode, production operation, may be calculated according
to a simplified method as examined in Danish standard, provided that the following requirements
are met. The calculated load range distributions are used to assess the fatigue limit states of the
wind turbine. In addition, an ultimate load limit state must be examined, whereby the average
value of the parameters studied is combined with the corresponding half maximum load range
found in the load range distribution of the parameter.
The following assumptions define the type of wind turbine, which may be dealt with by
means of simplified calculations:
1. The wind turbine shall have 3 blades, generate electricity and be stall-regulated/pitch
regulated with fixed blade angle, with no hub hinge.
2. The wind turbine shall produce electricity by means of an induction generator, connected
to the power network at a fixed frequency.
3. The wind turbine rotor shall be placed on the windward side (upwind rotor) with a maximum
10o tilt, radius R < 15, 5 m, solidity between 5 and 15 per cent and a minimum distance
from the blade to the tower during production operation of half of the local tower diameter
for the outer half of the blade.
4. Defined maximum power of the wind turbine Pmax may exceed the maximum long-term
average power (nominal power) Pnom (maximum point on the power curve between Vmin
and Vmax) by 15 per cent at the most (Figure 5.24).
FIGURE 5.24 Power curve, Pmax, Pnom and Vnom.
5. The nominal generator power of the electric generator of the wind turbine Pnom, g must be
at least Pnom.
6. The maximum operating frequency of rotation of the wind turbine nr, max shall be limited,
so that nr, max < 1.05 nr, syn , where n r, syn is the operating frequency of rotation
corresponding to idling (synchronous frequency of rotation) on the primary generator.
7. The yawing speed of the wind turbine shall not exceed wk = 1°/second.
8. At an arbitrary given wind speed, V the 10 minute yaw-error in the wind turbine must be
distributed with a maximum 10° average yaw-error and a maximum 10° standard deviation.
9. The stop wind velocity of the wind turbine Vmax shall not exceed 25 m/s.
10. The lowest blade and tower natural frequencies no during operation shall differ from the
operating frequency of rotation nr by at least 10 per cent.
11. The lowest natural frequency of the tower for bending shall be less than 2.5 nr.
12. The Weibull k parameter must be between 1.85 and 2.00.
13. For wind turbines in wind farms or groups, the distance between wind turbines must be >
5 × D.
The characteristic aerodynamic load value, Po shall be calculated as a basis for the load
calculations. The characteristic load is defined as the y¢ component of the aerodynamic load per
unit of length of the blade acting at the 2/3-point of the blade (i.e. where r = 2/3R and at right angles
to the rotor plane, see definition of the rotor coordinates system in Table 5.6 and blade geometry in
Figure 5.25(a) and 5.25(b)).
FIGURE 5.25(a) Definition of rotor radius at the 2/3-point and characteristic chord line length.
FIGURE 5.25(b) Coordinate system for blade loads, positions and deflections.
Po is calculated from
1
Po S W 2 cC l (5.87)
2
where
Cl is the lift coefficient at the 2/3-point
W is the resulting wind velocity

r is air density
c is the characteristic chord line length at 2/3 R (see Figure 5.25(b))
As lift coefficient, the maximum for the blade profile concerned shall be used, but it must be
at least Cl = 1.5. The resulting wind velocity W shall be determined for blade loads by means of,
2
È2 Ø
W2 É R * 2Q nr Ù 2
V0 (5.88)
Ê3 Ú
where nr is the rotor frequency of rotation in Hz with the primary generator connected, and R is the
rotor radius.
V0 is the nominal stall wind velocity defined as the lower of the two following wind velocities:
nominal wind velocity Vnom, or 10-min average wind velocity, where just the whole blade is stalled
with airflow parallel with the rotor shaft. The wind velocity is defined at hub height. The wind
velocity corresponding to the 10-minute average, at which the stall just extends to the whole blade,
can be determined by means of
V0,stall 2Q nr , Rtg (B stall,tip R b,tip ) (5.89)
where astall,tip is angle of attack for the tip profile at max Cl and qb,tip is blade angle for tip.
As a rule, a linear line load distribution outwards along blade of por/R (triangular load) can be
assumed for the y¢ direction. It may be advantageous to replace this with an alternative distribution:
P (r ) const * {(2Q nr r )2 V02 } c(r ) Cl (r ) (5.90)
where the local relative wind velocity and the exact chord line distribution along the blade is
included. The constant in the expression for p(r) is adjusted so that the moment in the rotor centre
has the same value as for the distribution p0r/R, namely P0R2/3. In the case of fatigue calculations,
loads are generally described by means of a constant average value overlaid with an accumulated
load range distribution FD(Nv) (see Figure 5.27).
An accumulated load range distribution is described in terms of a distribution function
FD(Nv). This function is defined in such a way that FD(Nv) is the load range which is exceeded
Nv times in the service life of the wind turbine. The model used for this presupposes that the
blade line loads are divided into two types. The first type (called deterministic) describes the
effect of the gravitational force on the rotating blade. The corresponding load range distribution
includes only load ranges of one constant size to a number, Nf, which is equal to the total number of
rotor revolutions during production operation in the design service life of the wind turbine, tf. (see
Figure 5.27, curve b).
The second type (called stochastic) includes all other loads, e.g. variable aerodynamic loads.
This type of distribution is described with the help of a standardized load range distribution, with a
load distribution FD(Nv), of this type being expressed by means of a dimensioning constant multiplied
by the standardized distribution ^FD(Nv) (see Figure 5.27, curve a).
The standardized load range distribution is defined by means of,
^FD(Nv) = bÿlog (Nf) – log (Nv)) + 0.18 until FD(Nv) = 2.0kb (5.91)
where
b = 0.11kb (I + 0.1) (A + 4.4) (5.92)
I = the turbulence intensity at hub height in the direction of the wind at the actual site
kb = a correction factor for the slope of the distribution
Nf = number of widths during the lifetime of the turbine corresponding to the characteristic
load frequency, given as
Î kÞ Î V kÞ
Ic t f ÑÏexp ÈÉ min ØÙ Ñß exp ÑÏ ÈÉ max ØÙ Ñß
V
Nf (5.93)
Ñ Ê A Ú Ñ Ñ Ê A Ú Ñ
Ð à Ð à
where
A, k = scale parameter and shape parameter, respectively, for the Weibull distribution
nc = the characteristic load frequency
tf = the design service life of the wind turbine.
The correction factor kb for the bias in the distribution is usually 1. Only for calculating the
rotor pressure Fy (see Table 5.7) shall a standard range distribution with kb = 2.5 be used. In
connection with wind farms or groups, it is necessary to consider how far the park configuration
affects the average wind velocity (and therefore the A parameter) and the turbulence intensity I.
The formula for Nf, takes into account how often wind velocity is between Vmin and Vmax.
The standardized load range distribution is shown in Figure 5.26. The distributions, which are
defined for different loads by means of this standardized load range distribution, are described as a
constant times the standard distribution. There are three parameters which affect a given standard
distribution: Nf is the total number of ranges in the
service life of the wind turbine. A change in Nf will
involve parallel displacement of the distribution along
the frequency axis, b changes the slope of the load 2.00
distribution, but the greatest loads remain unchanged,
and kb also changes the bias. In this case, it also
^FD(Nv)
increases the greatest loads. The dimensioning

constant, which is multiplied by the standard
distribution, is independent of Nv, but does depend
0.18
on wind turbine geometry, climate etc. This means
that if, for example, it is necessary to find range log (1) log (Nv) log (Nf)
distributions as the integral along the blade of load FIGURE 5.26 Standardized stochastic load
range distributions, the integration variable is range distribution.
contained in the dimensioning constant, while the
standard distribution ^FD(Nv) acts as a constant as far as the integration is concerned.
When load range distributions are calculated for sums of loads of the two types, stochastic and
deterministic, the following shall apply. Deterministic loads are compounded first, with the following
dependency of time t being assumed (the blade vertically upwards, when t = 0):
P d (t, P d 'c, P d 's, nc ) P 1/ 2 P d 'c * cos (2Q nc t ) 1/ 2 P d 's * sin (2Q nc t )
(5.94)
When stochastic loads are compounded with deterministic or other stochastic loads, it is necessary
to be conservative in the calculations, as the phase relationships for stochastic loads are not known.
Load ranges from each of two distributions are added directly one at a time. The largest from each
distribution is taken first, and then the smaller

ones are taken in succession according to size. c (sum)
Each load range is included only once.
Therefore, it is assumed that each load range in
b (deterministic)
the first distribution occurs the same time as the
largest still unused range from the second
FD(Nv)
distribution.
Figure 5.27 shows the accumulated load a (stochastic)
distribution for px. Curve (a) is the stochastic a and c concurrently

range distribution, the number of ranges
Na = 3Nr. Curve (b) is the deterministic (gravity
induced) distribution with constant range 2 mg log (1) log (Nb) log (Na)
and number of ranges Nb = Nr. The solid-line log (Nv)
curve (c) is the sum distribution, as it can be FIGURE 5.27 Compounding of stochastic and
shown that the conservative summation method deterministic loads.
specified above can be achieved by direct
addition of the two accumulated distribution functions (a) and (b) for frequencies < Nr. (This is
only correct for accumulated distributions.) For frequencies > Nr, the sum is equal to the stochastic
one, because the deterministic ranges are ‘used up’.
5.6.1 Blade Loads

Table 5.5 sums up the blade line loads used according to blade coordinates (Table 5.4). However,
the deterministic contribution is divided into a cosine and a sine element, which must be used when
compounding several deterministic loads. The last series in the table indicates the characteristic
load frequency for the load type in question.
TABLE 5.4 Description of blade coordinates
Direction Description
x¢ Rotation direction and perpendicular to the main shaft
y¢ Parallel to main shaft in wind direction
z¢ Following the blade length perpendicular to the main shaft
m = m(r) = blade mass per unit of length at distance r

g = gravitational acceleration
Mnom = Pnom/2p nrh0 = nominal driving torque corresponding to nominal power
h0 = nominal efficiency (£ 0.9)
FD(Nv) = standardized distribution, see above
Dc, Ds = indices for load ranges for cosine and sine elements, respectively.
TABLE 5.5 Blade line load
Average value Load range distributions

Direction Deterministic Stochastic
P P dDc P dDs PsD
x¢ 2Mnom/(3R 2) 0 +2 mg 0.3FD(Nv)Po
y¢ 1.5Por/R* 0 0 FD(Nv)Por/R
z¢ (2p nr) mr
2
–2 mg 0 0
Frequency n c nr nr 3n r
*P Po for r 2 / 3 R
Hub loads
When the internal loads on the rotor hub are calculated, allowance should be made for blade loads
with reciprocal phase lag of 120°. For the purposes of the calculation, the hub is assumed to be
rigidly fixed at the shaft attachment, where the rotor loads are crucial. It should be pointed out that
the rotor loads are defined in the fixed x, y, z system (Table 5.6) and therefore must be converted to
the hub system, with allowance being made also for dead weight.
TABLE 5.6 Description of blade coordinates
Direction Description
x Horizontal direction and perpendicular to the main shaft
y Parallel to main shaft in wind direction
z ‘Vertical’ and perpendicular to the main shaft
5.6.2 Rotor Loads

The rotor loads are defined on the basis of the characteristic aerodynamic line load Po, the
standardized load range distribution form FD(Nv) as well as the nominal driving torque Mnom defined
in Table 5.5.
The average value, amplitude and frequency of rotor loads are given in Table 5.7, with the
individual load components being separately compounded as
1
F F F' cos (2Q nc t ) (5.95)
2
where t is time in seconds.
In Table 5.7 and Figure 5.28,
M = rotor mass
kR = a correction factor which for Mx and Mz takes into account the position of relevant
resonance frequencies in relation to the frequency content of the external influences.
The kR factor is read from Figure 5.28 as a function of the resonance frequency
no = lowest resonance frequency for the oscillation form in question
nR = lowest resonance frequency for the collective asymmetric rotor oscillation at standstill,
where one blade oscillates out of phase to the two others. The oscillation will link
with nacelle/tower, for which reason its frequency in yawing and tilting, respectively,
may be different
nT = lowest resonance frequency for tower in bending (< 2.5nr).
TABLE 5.7 Rotor loads
Load Average Range Frequency Resonance frequency

F F FD nc (oscillation from), no
Fx 0 0 0
Fy 1.5PoR 0.5PoRFD(Nv) Stator: nT nT (tower, bending)
Rotor: 3nR
Fz Mg 0
Mx 0 1/3kRPoR2FD(Nv) 3nr nR (rotor, tilt, asymmetry)
My** 0.45PoR2FD(Nv) 3nr
Mz 1/3kRPoR2FD(Nv) 3nr nR (rotor, yaw, asymmetry)
** It may be assumed that My is always between 0 and 1.3 Mnom.
1.7 Inadmissible
frequency range
KR (no /nr)
(*)
1.0
0.8
1.1
0.9
1 2 3 4 5
no /nr
FIGURE 5.28 Response amplification.
It should be pointed out that, as far as gearbox calculations are concerned, it is necessary to use
a duration curve for shaft moment. For example, it may be conservatively assumed here that the
moment is constant at the upper limit indicated. Note that for Fy, it is necessary to use a modified
standard distribution with the bias b multiplied by kb = 2.5 as mentioned in the definition of the
standard distribution above. The two frequencies which are indicated for Fy are used for Fy
distributions in, respectively, the stationary (nc = nT) and the rotating part of the wind turbine
(nc = 3nr), thus taking into account that the oscillations are damped when it is transmitted from the
rotating part to the stationary part. When the load range distributions are determined, it is necessary
to be conservative in calculations when adding the individual components and distributions, as

discussed in connection with Figure 5.28. In the calculations shown, the wind load on the tower
can be assumed to have been included.
5.6.3 Codes and Standards

The IEC 61400-1 wind turbine generator systems–Part 1 safety requirements identifies four different
classes of wind turbine to suit differing site wind conditions, with increasing class designation
number corresponding to reducing wind speed. The wind speed parameters for each class are given
in Table 5.8.
TABLE 5.8 Wind speed parameters for wind turbine classes
Parameters Class I Class II Class III Class IV
Reference wind speed, Uref (m/s) 50 42.5 37.5 30

Annual average wind speed, Uave (m/s) 10 8.5 7.5 6
50-year return gust speed, 1.4 Uref (m/s) 70 59.5 52.5 42
1 year return gust speed, 1.05 Uref (m/s) 52.5 44.6 39.4 31.5
The reference wind speed is defined as the 10 min mean wind speed at hub-height with a
50-year return period. To allow for other sites where conditions do not conform to any of these
classes, a fifth class is provided for in which the basic wind parameters are to be specified by the
manufacturer. The normal value of air density is specified as 1.225 kg/m3. A crucial parameter for
wind turbine design is the turbulence intensity, which is defined as the ratio of the standard deviation
of wind speed fluctuations to the mean. The standard specifies two levels of turbulence intensity,
designated category A (higher) and category B (lower), which are independent of the wind speed
classes above. In each case, the turbulence varies with hub height mean wind speed, U , according
to the formula,
È
15 Ø
É a
U ÙÚ
Ê
IU I15 (5.96)
( a 1)
where I15 is the turbulence intensity at a mean wind speed of 15 m/s, defined as 18 per cent for
category A and 16 per cent for category B. The constant a takes the values 2 and 3 for categories A
and B, respectively.
Germanischer Lloyd’s Regulation for the classification of Wind Energy Conversion systems,
commonly referred to as GL rules, specifies a single value of hub-height turbulence intensity of
20 per cent.
Danish Standard DS 472 bases the derivation of design-extreme wind speeds on four terrain
classes, ranging from very smooth (expanse of water) to the very rough (e.g. built-up areas). The
base wind velocity is taken to be the same all over Denmark, so the result is four alternative
profiles of wind speed variation with height.
5.6.4 Design Loads

The sources of loading to be taken into account may be categorized as follows:
• Aerodynamic loads
• Gravitational loads
• Inertia loads (including centrifugal and gyroscopic effects)
• Operational loads arising from actions of the control system (e.g. braking, yawing, blade-
pitch control, generator disconnection).
The load cases selected for ultimate load design must cover realistic combinations of a wide
range of external wind conditions and machine states. It is a common practice to distinguish between
normal and extreme wind conditions on the one hand, and between normal machine states and fault
states, on the other. The load cases for design are then chosen from:
• Normal wind conditions in combination with normal machine states
• Normal wind conditions in combination with machine fault states
• Extreme wind conditions in combination with normal machine states.
Extreme and normal wind conditions are generally defined in terms of the worst condition
occurring with a 50-year and 1 year return period, respectively. It is assumed that machine fault
states arise only rarely and are uncorrelated with extreme wind conditions, so that the occurrence
of a machine fault in combination with an extreme wind condition is an event with such a high
return period that it need not be considered as a load case. However, IEC 61400-1 wisely stipulates
that if there is some correlation between an extreme external condition and a fault state, then the
combination should be considered as a design case.
A typical wind turbine is subjected to a severe fatigue loading regime. The rotor of a typical
wind turbine will rotate some 2 × 108 times during a 20-year life, with each revolution causing a
complete gravity stress reversal in the low speed shaft and in each blade, together with a cycle of
blade out-of-plane loading due to the combined effects of wind shear, yaw error, shaft tilt, tower
shadow and turbulence. It is therefore hardly surprising that the design of many wind turbine
components is often governed by fatigue rather than by ultimate load.
The partial safety factors for ultimate loads stipulated by IEC 61400-1 the GL rules and DS
472 are given in Table 5.9.
TABLE 5.9 Partial safety factors for loads
Sources of loading Unfavourable loads Favourable loads

Types of loading
Normal and extreme Abnormal
IEC GL DS IEC GL and DS IEC GL DS
Normal Extreme
Aerodynamic 1.35 1.2 1.5 1.3 1.1 1.0 0.9 – –
Operational 1.35 1.35 1.2 1.3 1.1 1.0 0.9 – –
Gravity 1.1* 1.1* 1.1* 1.0 1.1 1.0 0.9 1.0 –
Inertia 1.25 1.1* 1.1* 1.0 1.1 1.0 0.9 1.0 –
*Factor increased to 1.35 if masses are not determined by weighing.
Normally, the critical operating parameters are:

• Turbine operational speed
• Power output
• Vibration level
• Twist of pendant cables running up into nacelle.
The fluctuation of the wind speed about the short-term mean, or turbulence, naturally has a
major impact on the design loadings, as it is the source of both the extreme gust loading and a large
part of the blade fatigue loading. The latter is exacerbated by the gust slicing effect in which a
blade slices through a localized gust repeatedly in the course of several revolutions. Within a wind
farm, turbines operating in the wake of other turbines experience increased turbulence and reduced
mean velocities. In general, a downwind turbine will lie off-centre with respect to the wake of the
turbine immediately upwind, leading to horizontal wind shear. The consensus is not emerged for
the models which are describing velocity deficit and the increase in turbulence intensity due to
turbine wake are yet to be used in design calculations.
A non-operational machine state is defined as one in which the machine is neither generating
power, nor starting up, nor shutting down. It may be stationary, i.e. parked or idling. The design
wind speed for this load case is commonly taken as the gust speed with a return period of 50 years.
The magnitude of the 50-year return gust depends on the gust duration chosen, which, in turn,
should be based on the size of the loaded area. IEC 61400-1 and the GL rules specify the use of
gust durations of 3s and 5s, respectively, regardless of the turbine size.
5.7 COST MODELLING

The sensitivity of the cost of energy to changes in the values of parameters governing turbine
design can be examined with the aid of a model of the way component costs vary in response. It is
usually sufficiently accurate to represent the relationship between component cost and mass as a
linear one with a fixed component:
È m( x ) Ø
C( x) CB É N (1 N )Ù (5.97)
Ê mB Ú
where C(x) and m(x) are the cost and mass of the component, respectively, when the design parameter
takes the value x, and CB and mB are the baseline values; m is the proportion of the cost that varies
with mass, which will obviously differ for different baseline machine sizes.
5.7.1 Simplified Cost Model

The baseline machine design is taken as a 60 m diameter, 1.5 MW turbine, with the costs of the
various components given as a percentage of the total. Machine designs for the other diameters are
obtained by scalling all dimensions of all components in the same proportion, except in the case of
gear box, generator, grid connection and controller. Rotational speed is kept inversely proportional
to rotor diameter to maintain constant tip speed, and hence constant tip speed ratio at a given wind
speed. As a result, all machine designs reach rated power at the same wind speed, so that rated
power is proportional to diameter squared. Consequently, the low-speed shaft torque increases as
diameter cubed, which is the basis for assuming that the gear box mass increases as the cube of
rotor diameter, even though the gear box ratio changes. Hence, if a blanket value of m of 0.9 is
adopted for simplicity, the cost of all components, apart from generator, controller and grid
connection, for a machine of diameter D, is given by:
ÎÑ È D Ø 3 ÞÑ
C1 (D) 0.8CT (60) Ï0.9 É Ù 0.1ß (5.98)
Ê 60 Ú
ÐÑ àÑ
where CT(60) is the total cost of the base line machine. The rating of the generator and the grid
connection is proportional. Only to the diameter squared. It is assumed that Eq. (5.97) applies to
the cost of these components, but with mass replaced by rating. Thus if m is taken as 0.9 once more,
the costs of generator and grid connection are given by:
Î
Ñ ÈDØ
2 Þ
Ñ
C2 ( D) 0.158 CT (60) Ï0.9 É Ù 0.1ß (5.99)
Ñ Ê 60 Ú Ñ
Ð à
The controller cost is assumed to be fixed. Hence, the resulting turbine cost as a function of diameter
is:
Ë Î DØ
3 Þ Î 2 Þ Û
Ñ È Ñ Ñ È DØ Ñ
CT ( D) CT (60) Ì0.8 Ï0.9 É Ù 0.1ß 0.158 Ï0.9 É Ù 0.1ß 0.042 Ü
Ì Ê 60 Ú Ê 60 Ú Ü
Í Ñ
Ð Ñ
à Ñ
Ð Ñ
à Ý
ÎÑ È DØ
3
ÈDØ
2 Þ
Ñ
CT (60) Ï0.72 É Ù 0.1422 É Ù 0.1378ß (5.100)
Ê 60 Ú Ê 60 Ú
ÐÑ Ñ
à
As the tower height, along with all other dimensions is assumed to increase in proportion to rotor
diameter, the annual mean wind speed at hub height will increase with rotor diameter because of
wind shear. The energy yield should thus be calculated taking this effect into account. The cost of
energy (excluding operation and maintenance costs) can then be calculated in dollar/kWh/annum
by dividing the turbine cost by the annual energy yield. The variation of energy cost with diameter,
calculated according to the assumptions described above, is plotted in Figure 5.29 for two levels of
140
Wind shear corresponding to
Zo = 0.05 m (agriculture land)
130
Wind shear corresponding
120 to Zo = 0.001 m (open sea)
Energy cost index
110
Zero wind shear
100
90
80
70
0 10 20 30 40 50 60 70 80
Rotor diameter (m)
FIGURE 5.29 Variation of optimum turbine size with wind shear (assuming constant hub height of diameter ratio).
wind shear, corresponding to roughness lengths, Zo of 0.001 m and 0.05 m, the hub-height mean
wind speed being scaled according to the relation
Z Ø
È
U ( Z ) B ln ÉÙ
Ê Zo Ú
i.e. the wind shear is then given by a logarithmic wind profile, where Z is the height above ground
and Zo is the surface roughness length (tabulated in Table 5.10).
TABLE 5.10 Typical surface roughness lengths
Type of terrain Roughness length Zo(m)

Cities, forests 0.7
Suburbs, wooded countryside 0.3
Villages, countryside with trees and hedges 0.1
Open farmland, few trees and buildings 0.03
Flat grassy plains 0.01
Flat desert, rough sea 0.001
In Figure 5.29, it also included a plot for the case of zero wind shear. It is apparent that the
level of wind shear has a noticeable effect on the optimum machine diameter, which varies from
44 m for zero wind shear to 52 m for the wind shear corresponding to a surface roughness length of
0.05 m, which is applicable to the farmland with boundary hedges and occasional buildings. Strictly,
the impact of the increased annual mean wind speed with hub height on the fatigue design of the
rotor and other components should also be taken into account, which would reduce the optimum
machine size slightly.
It should be emphasized that the optimum sizes derived above depend critically on the value of
m adopted. For example, if m were taken as 0.8 instead of 0.9, the optimum diameter in the absence
of wind shear would increase to 54 m, although the minimum cost of energy would alter by only
0.3 per cent. A more sophisticated approach would allocate different values of m to different
components.
The cost model outlined above provides a straight forward means of investigating scale effects
on machine economics for a chosen machine design. In practice, the use of different materials or
different machine configurations may prove more economical at different machine sizes and will
yield a series of alternative cost versus diameter curves.
5.8 POWER CONTROL
5.8.1 Passive Stall Control

The simplest form of power control is passive stall control, which makes use of the post-stall
reduction in lift coefficient and associated increase in drag coefficient to place a ceiling on output
power as wind speed increases, without the need for any changes in blade geometry. The fixed-
blade pitch is chosen so that the turbine reaches its maximum or rated power at the desired wind
speed. Stall-regulated machines suffer from the disadvantage of uncertainties of aerodynamic
behaviour post-stall, which can result in inaccurate prediction of power levels and blade loadings
at rated wind speed and above.
5.8.2 Active Pitch Control

Active pitch control achieves power limitations
above rated wind speed by rotating all parts of each
blade about its axis in the direction which reduces
the angle of attack and hence the lift coefficient—a
process known as blade feathering shown in Figure
5.30. The main benefits of active pitch control are
increased energy capture, the aerodynamic braking
facility it provides, and the reduced extreme loads
on the turbine when shut-down.
Although full-span pitch control is the option
favoured by the overwhelming majority of FIGURE 5.30 Blade pitching system using a separate
manufacturers, power control can still be fully electric motor for each blade.
effective even if only the outer 15 per cent of the
blade is pitched. The principal benefits are that the duty of the pitch actuator is significantly reduced,
and that the inboard portion of the blade remains in stall, significantly reducing the blade load
fluctuations. On the other hand, partial–span pitch control has several disadvantages as follows:
• The introduction of extra weight near the tip
• The difficulty of physically accommodating the actuator within the blade profile
• The high bending moment to be carried by the tip-blade shaft
• The need to design the equipment for the high centrifugal loadings found at large radii
• The difficulty of access for maintenance.
5.8.3 Passive Pitch Control

An attractive alternative to active control of the blade pitch to limit power is to design the blade
and/or its hub mountings to twist under the action of loads. On the blades, in order to achieve the
desired pitch changes at higher wind speeds. Unfortunately, the principle is easy to state but
difficult to achieve in practice because of the required variation in blade load. In the case of
stand-alone wind turbines, the optimization of energy yield is not the key objective, so passive
pitch control is sometimes adopted, but the concept has not been utilized as yet for many grid
connected machines.
5.8.4 Active Stall Control

Active stall control achieves power limitation above the rated wind speed by pitching the blades
initially into stall, i.e. in the opposite direction to that employed for active pitch control, and is thus
sometimes known as negative pitch control. At higher wind speeds, however, it is usually necessary
to pitch the blades back towards feather in order to maintain power output at rated wind speed.
A significant advantage of active stall control is that the blade remains essentially stalled
above the rated wind speed, so that gust slicing results in much smaller cyclic fluctuations in
blade loads and power output. It is found that only small changes of pitch angle are required to
maintain the power output at rated, so pitch rates do not need to be as large as for positive pitch
control.
5.8.5 Yaw Control

As most horizontal-axis wind turbines employ a yaw drive mechanism to keep the turbine headed
into the wind, the use of the same mechanism to yaw the turbine out of wind to limit power output
is obviously an attractive alternative. However, there are two factors which militate against the
rapid response of such a system to limit power : first, the large amount of inertia of the nacelle and
rotor about the yaw axis and second, the cosine relationship between the component of wind speed
perpendicular to the rotor disc and the yaw angle. The latter factor means that, at small initial yaw
angles, yaw changes of, say, 10°C only bring about reductions in power of a few per cent, whereas
blade pitch changes of this magnitude can easily halve the power output. Thus, active yaw control
is only practicable for variable speed machines where the extra energy of a wind gust can be stored
a rotor kinetic energy until the yaw drive has made the necessary yaw correction.
5.8.6 Teetering
Two-bladed rotors are often mounted on a teeter hinge—with hinge axis perpendicular to the shaft
axis, but not necessarily perpendicular to the longitudinal axis of the blades—in order to prevent
differential blade root out-of-plane bending moments arising during operation. Instead, differential
aerodynamic loads on the two-blades result in rotor angular acceleration about the teeter axis,
with large teeter excursions being prevented by the restoring moment generated by centrifugal
forces.
5.9 BRAKING SYSTEMS

The normal practice is to provide both aerodynamic and mechanical braking. However, if
independent aerodynamic braking systems are provided on each blade, and each has the capacity
to decelerate the rotor after the worst-case grid loss, then the mechanical brake will not necessarily
be designed to do this. The function of the mechanical brake in this case is solely to bring the rotor
to rest, i.e. to park it, as aerodynamic braking is unable to do this.
5.9.1 Active Pitch Control Braking

Blade pitching to feather (i.e. to align the blade chord with the wind direction) provides a highly
effective means of aerodynamic braking. Blade pitch rates of 10° per second are generally found
adequate, and this is of the same order as the pitch rate required for power control. The utilization
of the blade pitch system for start up and power control means that it is regularly exercised with the
result that the existence of a dormant fault is highly unlikely. In machines relying solely on blade
pitching for emergency braking, independent actuation of each blade is required, together with
fail-safe operation should power or hydraulic supplies passing through a hallow low-speed shaft
from the nacelle be interrupted. In the case of hydraulic actuators, oil at pressure is commonly
stored in accumulators in the hub for this purpose.
5.9.2 Braking by Pitching Blade Tips

Blade tips which pitch to feather have become the standard form of aerodynamic braking for stall-
regulated turbines. Typically, the blade tip is mounted on a tip shaft, as illustrated in Figure 5.31,
and held in against centrifugal force during normal operation by a hydraulic cylinder.
Rotary bearing
Tip shaft Blade pitch
Screw
angle
Flow
Compressed spring
FIGURE 5.31 Passive control for tip blade, using screw on tip shaft and spring.
On release of the hydraulic pressure (which is triggered by a control system, or directly by an

overspeed sensor, the tip blade flies outward under the action of centrifugal force, pitching to
feather simultaneously on the shaft screw. The length of the tip blade is commonly some 15 per
cent of the tip radius.
The penalty is that the low-speed shaft needs to be hollow to accommodate the feed to the
hydraulic cylinder.
5.10 TURBINE BLADE DESIGN

A successful blade design must satisfy a wide range of objectives, some of which are in conflict.
These objectives are:
1. Maximise annual energy yield for the specified wind speed distribution.
2. Limit maximum power output (in the case of stall regulated machines.
3. Resist extreme and fatigue loads.
4. Restrict tip deflections to avoid blade/tower collisions (in the case of upwind machines).
5. Avoid resonances.
6. Minimise weight and cost.
The design process can be divided into two stages: the aerodynamic design, in which objectives
(1) and (2) are satisfied, and the structural design. The aerodynamic design addresses the selection
of optimum geometry of the blade external surface—normally simply referred to as the blade
geometry—which is defined by the aerofoil family and the chord, twist and thickness distributions.
The structural design consists of blade material selection and the determination of a structural
cross section or spar within the external envelope that meets objectives (3) to (6). Inevitably, there
is interaction between the two stages, as the blade thickness needs to be large enough to accommodate
a spar which is structurally efficient.
The aerodynamic design encompasses the selection of aerofoil family and optimization of the
chord and twist distribution. The variation of thickness to chord ratio along the blade also has to be
considered, but this ratio is usually set at the minimum value permitted by structural design
considerations, as this minimizes drag losses.
The process for optimizing the blade design of machines operating at a fixed tip speed ratio is
described with analytical expressions for the blade geometry parameter,
Bc ( N )
T r MCl MC l
2Q R
and the local inflow angle, f, are derived as a function of the local tip speed ratio, lm = lr/R. If
lm >> 1, the expressions can be approximated by
Bc ( N ) 8
T r MCl MCl
2Q R 9 MN
2
and G
2MN
If it is decided to maintain the angle of attack, a, and hence the lift coefficient, CL, constant along
the blade, then these relations translate to
16 Q R
c( N )
9C l B M 2 N
2
and C B
3MN
so that both the chord and twist are inversely proportional to radius.
The blade root area is normally circular in cross section in order to match up with the pitch
bearing in the case of pitchable blades, or to allow pitch angle adjustment at the bolted flange (to
compensate for non-standard air density) in the case of stall regulated blades.
5.10.1 Form of Blade Structure

A hollow shell corresponding to the defined blade envelope clearly provides a simple, efficient
structure to resist flexural and torsional loads and some blade manufacturers adopt this form
of construction (see Figures 5.32, 5.33 and 5.34). However, in the case of small and medium size
machines, where the out-of-plane loads dominate, there is greater benefit in concentrating skin
material in the forward half of the blade, where the blade thickness is maximum, so that it acts
more efficiently in resisting out-of-plane bending moments. The weekened areas of the shell
Glass/Epoxy
25 mm wood veneers
Epoxy glue
Aluminium screen for
lightening protection (unusual)
Glass/Epoxy
Polyurethane point
FIGURE 5.32 Wood/Epoxy blade construction utilizing full blade shell.
Filler
Glass/Epoxy web
Glass/Epoxy
Glass/Epoxy
4 mm wood Styrofoam
veneers
Epoxy glue
Glass/Epoxy
Glass/Epoxy Gel coat
Gel coat
FIGURE 5.33 Wood/Epoxy blade construction utilizing forward half of blade shell.
CSM = Continuous Strand Mat
CSM skins
CSM
Filler
Moulded GRP PVC foam
shear webs UD Glass fibre/ CSM
polyester
Gel coat
FIGURE 5.34 Glass-fibre construction using blade skins in forward portion of blade cross section and linking shear webs.
towards the trailing edge are then typically stiffened by means of sandwich construction utilizing
a PVC foam filling.
5.10.2 Blade Materials

The ideal material for blade construction will combine the necessary properties—namely high
strength to weight ratio, fatigue life and stiffness—with low cost and the ability to be formed into
desired aerofoil shape.
Table 5.11 lists the structrual properties of the materials in general use for blade manufacture
and those of some other candidate materials. For comparative purposes values are also presented
of:
• Compressive strength-to-weight ratio
• Fatigue strength as a percentage of compressive strength
• Stiffness-to-weight ratio
• A panel stability parameter, E/(UCS)2 (UCS is ultimate compressive strength).
It is evident that glass and carbon-fibre composites (GFRP and CFRP) have a substantially
higher compressive strength-to-weight ratio compared with the other materials. Figure 5.35 shows
a typical GFRP blade. However, this apparent advantage is not as decisive as it appears, for two
reasons. First of all, the fibres of some of the plies making up the laminated blade shell have to be
TABLE 5.11 Structural properties of materials for wind-turbine blades
Material Ultimate Ultimate Specific Compressive Mean fatigue Mean fatigue Young’s Stiffness to Panel stability
(NB: UD denotes tensile compressive gravity strength to strength at 107 strength as modulus, weight ratio parameter
unidirectional fibres – strength strength (s.g.) weight ratio cycles percentage E (GPa) E/s.g. E/(UCS)2
i.e., all fibres running (UTS) (UCS) UCS/s.g. (amplitude) of UCS (GPa) (MPa)–1
longitudinally) (MPa) (MPa) (MPa)
(Mean for composites,
minimum for metals)
1. Glass/polyester ply 860–900 ~720 [1] 1.85 390 140 [3] 19% 38 [12] 20.5 0.07
with 50 per cent fibre [1] [2]
volume fraction and
UD lay-up
2. Glass/epoxy ply with Properties are generally very close to those for GRP given above
50 per cent fibre
volume fraction and
UD lay-up
3. Glass/polyester 690–720 ~580 1.85 310 120 21% 33.5 18 0.1
laminate with
50 per cent fibre
volume fraction and
80 per cent of fibres
running longitudinally
4. Carbon fibre/epoxy 1830 [4] 1100 [4] 1.58 700 350 [5] 32% 142 [4] 90 0.12
ply with 60 per cent
fibre volume fraction
and UD lay-up
5. Khaya ivorensis/ 82 [6] 50 [6] 0.55 90 15 [7] 30% 10 [8] 18 4
epoxy laminate
6. Birch/epoxy laminate 117 [9] 81 [10] 0.67 121 16.5 [7] 20% 15 [10] 22.5 2.3
7. High yield steel 510 510 7.85 65 50 [11] 10% 210 27 0.81
(Grade Fe 510)
8. Weldable aluminium 295 [12] 295 [12] 2.71 109 17 [13] 6% 69 [12] 25.5 0.79
Wind Turbine Design
alloy AA6082
(formerly H30)
169
CSM = Continuous Strand Mat
CSM/polyester
Adhesive PVC foam
TFT wound spar
CSM/polyester
Gel coat
FIGURE 5.35 Glass-fibre blade construction using compact spar wound with transverse filament type (TFT) on mandrel.
aligned off-axis (typically at ± 45°) to resist shear loads, giving reduced strengths in the axial
direction. Secondly, the relatively low Young’s modulus of these composites means that resistance
to buckling of the thin skins governs the design rather than simple compression yielding. The
likelihood given in the last column of the blade, so that materials with high values, such as wood
composites will be least sensitive to buckling. As a result, wood composite blades are generally
lighter than equivalent glass-fibre composite blades. Table 5.12 lists different wind turbine blade
materials with design strength-to-stiffness ratios.
TABLE 5.12 Design strength-to-stiffness ratios for different wind turbine blade materials
Material Ultimate Partial safety Compression Young’s Strength-to-

compression factor for design strength, modulus, E stiffness ratio,
strength, scu strength, gmu scd (GPa) (scd/E) × 103
(MPa) (MPa)
Glass/polyester laminate with 580 2.94 197 (ignoring) 32.5 6.1

50 per cent fibre volume (buckling)
fraction and 80 per cent UD
Carbon fibre/epoxy ply with 1100 2.94 374 142 2.6
60 per cent fibre volume
fraction and UD lay-up
Khaya/epoxy laminate 50 1.5 33 10 3.3
Birch/epoxy laminate 81 1.5 54 15 3.6
Yield strength,
sy gmy
High-yield steel
(Grade Fe 510) 355 1.1 323 210 1.54
Weldable aluminium alloy
AA6082 240 1.1 218 69 3.2
The properties of glass/polyester and glass/epoxy plies with the same fibre volume fraction
and lay-up are generally very similar, i.e. the influence of the matrix is slight. They will therefore
be treated as the same material except in relation to fatigue, where some differences have been
noted. The glass used in blade construction is E-glass, which has good structure properties in
relation to its cost.
The plate elements forming the spar of a GFRP blade are normally laminates consisting of
several plies, with fibres in different orientations to resist the design loads. Within a ply (typically
0.5–1.0 mm in thickness), the fibres may all be arranged in the same direction, i.e. UD (unidirectional)
or they may run in two directions at right angles in a wide variety of woven or non-woven fabrics.
Although the strength and stiffness properties of the fibres and matrix are well-defined, only some
of the properties of a ply can be derived from them using simple rules. Thus, for a ply reinforced by
UD fibres, the longitudinal stiffness modulus E1, can be derived accurately from the rule of mixtures
formula
E1 = EfVf + Em (1 – Vf) (5.101)
where Ef is the fibre modulus (72.3 GPa for E-Glass), Em is the matrix modulus (in the range
2.7–3.4 GPa), and Vf is the fibre volume fraction. On the other hand, the inverse form of this
formula, e.g.,
1 (1 V f ) V f
(5.102)
E2 Em Ef
significantly underestimates the transverse modulus, E2 and the in-plane shear modulus, G12. The
longitudinal tensile strength of a ply reinforced by UD fibres, slt, can be estimated from:
Ë Em Û
T lt T fu ÌV f (1 V f ) Ü (5.103)
Ì
Í
Ef Ü
Ý
where sfu is the ultimate tensile strength of the fibres. However, the tensile strengths of E-glass
single fibres (3.45 GPa) cannot be realized in a composite, where fibre strength reductions of upto
50 per cent have been measured. Accordingly, a value of sfu of 1750 MPa should be used in
Eq. (5.103).
5.10.3 SERI Blade Sections

The design requirements for airfoil sections of a wind machine are different from those of
airplanes.
In wind machines, fouling is a major problem because once the machine goes on line, it has to
run round the day and round the year preferably without any break. Also, the blades of the wind
machine are inaccessible once installed and cannot be easily cleaned.
Further, dirt deposited on the blades affects the performance of the blade and, in turn, that of
the machine significantly as shown in Figure 5.36.
Another consideration is the thickness of the blade at root. Typical NACA sections, common
for airplanes, used for wind machine are shown in Figure 5.37. They have high CL and high (CL/CD)
ratio. But they are comparatively thinner. Structurally, a thick blade at root will always be easier to
design to take loads.
Based on these considerations, blades have been designed by National Renewable Energy
Laboratory (NREL) in USA. A set of these blades used for tip, middle span and root is shown in
Figure 5.38.
The comparative performance of wind machine with NACA and SERI blades when they are
clean and fouled (dirty) is shown in Figure 5.39 and Figure 5.40.
80
Clean blades
70
Power output (kW)
Fouled blades
60
50
0 5 10 15 20 25
Wind speed at hub (m/s)
FIGURE 5.36 Effect of dirt on blade and the performance of the machine.
FIGURE 5.37 NACA airfoils. FIGURE 5.38 SERI airfoils.
700 SERI Airfoils
600 Rated power

Generated power (kW)
500 NACA 4415–4424 Airfoils
400
300
200
100
0 5 10 15 20 25
Wind speed at hub elevation (m/s)
FIGURE 5.39 Comparative performance of wind machine with clean NACA and SERI blades.
If the aerofoil is sensitive to roughness, good performance is lost if the wings are contaminated
by dust, rain particles or insects. On a wind turbine, this could alter the performance with time if,
for example, turbine is sited in an area with many insects. If a wind turbine is situated near the
coast, salt might built-up on the blades, if the winds comes from the sea and, if the aerofoil used are
sensitive to roughness, the power output from the turbine will become dependent on the direction
of the wind.
700
600 Rated power

Generated power (kW)
SERI Airfoils
500
400 NACA 4415–4424 Airfoils

300
200
100
0 5 10 15 20 25
Wind speed at hub elevation (m/s)
FIGURE 5.40 Comparative performance of wind machine with fouled NACA and SERI blades.
5.11 ROTOR HUB

The relatively complex three-dimensional geometry of rotor hubs favours the use of casting in
their manufacture, with spheroidal graphite iron being the material generally chosen. Two distinct
shapes of hub for three-bladed machines can be identified: tri-cylindrical or spherical The former
consists of three cylindrical shells concentric with the blade axes, which flare into each other
where they meet, while the latter consists simply of a spherical shell with cut-outs at the three-
blade mounting positions. Diagrams of both types are shown in Figure 5.41, while an actual spherical
hub is illustrated in Figure 5.42.
The structural actions of the hub in resisting three loadings are as follows:
1. Symmetrical rotor thrust loading: The blade root bending moments due to symmetric
rotor thrust loading put the front of the hub in bi-axial tension near the rotor axis and the
rear in bi-axial compression, while the thrust itself generates out-of-plane bending stresses
in the hub shell adjacent to low speed shaft flange connection.
2. Thrust loading in single blade: This generates out-of-plane bending stresses in the hub
shell at the rear, and in-plane tensile stresses around a curved load path between the upwind
side of the blade bearing and the portion of the low-speed shaft flange connection remote
from the blade. The resultant lateral loads will result in out-of-plane bending.
3. Blade gravity moments: On the tri-cylindrical hub, equal and opposite blade gravity
moments are communicated via the cylindrical shells to areas near the rotor axis at front
MY
Side view Front view
(a) (b)
FIGURE 5.41 (a) Tri-cylindrical hub, (b) Spherical hub.
FIGURE 5.42 Rotor hub, view of spherical-shaped rotor hub.
and rear where they cancel each other out. It is less straight forward to visualize the
corresponding load paths on the spherical hub, as out-of-plane bending is likely to be
mobilized.
The complexity of the stress states arising from the latter two types of loading renders finite-
element analysis of rotor hubs more or less mandatory. At the most, six load cases need to be
analyzed, corresponding to the separate application of moments about the three axes and forces
along the three axes at a single hub/blade interface. Then the distribution of hub stresses due to
combination of loadings on different blades can be obtained by superposition. Similarly, the
fluctuation of hub stresses over time can be derived by inputting the time histories of the blade
loads obtained from a wind simulation.
The critical stresses for hub design are the in-plane stresses at the inner or outer surface, where
they reach a maximum because of shell bending. For any one location on the hub, these are defined
by three quantities at each surface: the in-plane direct stresses in two directions at right angles, and
the in-plane shear stress. In general, these stresses will not vary in-phase with each other over time,
so the principal stress directions will change, complicating the fatigue assessment. The following
methods cater for one or more series of repeated stress cycles rather than the random stress
fluctuations resulting from turbulent loading:
1. Maximum shear method: The fatigue evaluation is based on the maximum shear stress
ranges, calculated from the (s1 – s2)/2, s1/2 or s2/2 time histories. The effect of mean
stress is allowed for using the Goodman relationship:
Ua U 1
m (5.104)
SSN SSU H
where ta is the alternating shear stress, tm is the mean shear stress, SSN is the alternating
shear stress for N loading cycles from the material S–N curve, SSU is the ultimate shear
strength, and g is the safety factor. Having used Eq. (5.104) to determine SSN, the permitted
number of cycles for this loading range can be derived from the S–N curve, enabling the
corresponding fatigue damage to be calculated.
2. ASME boiler and pressure vessel code method: This is similar to the maximum shear
method, but the shear stress ranges are based on national principal stresses calculated from
the changes in the values of sx, sy, sz, txy, tyz and tzx from datum values occurring at one of
the extremes of the stress cycle. Mean stress effects are not included.
3. Distortion energy method: The fatigue evaluation is based on the fluctuations of the
effective or Von Mises stress. In the case of hub shell, the stress perpendicular to the hub
surface (and hence the third principal stress) is zero, so the effective stress is given by:
(T 1 T 2 ) 2 T 12 T 22
T (5.105)
2
As the effective stress is based on the distortion energy, it is a scalar quantity, so it needs to be
assigned a sign corresponding to that of the dominant principal stress. The effect of mean stress is
allowed for in the same way as for the maximum shear method, except that the stresses in Eq. (5.104)
are now direct stresses instead of shear stresses.
REVIEW QUESTIONS
1. A wind machine with 60 m diameter rotor, cut-in wind speed, Vc, of 3 m/s and furling wind
speed, VF of 25 m/s is available with a wind machine manufacturing organization. The
machine is to be marketed in the region with wind speed probability distribution at 20 m
height given by Weibull parameters k = 2.2 and c = 5 m/s. The machine hub height is 60 m.
The rated wind speed, VR is to be optimized for this region and accordingly, the capacity of
the generator is to be selected. The rotational speed of the wind machine is designed as 15
rpm and gear ratio of 100 resulting into typical efficiencies, hmech and helect as 0.94 and 0.96,
respectively. Assume air density of 1.2 kg/cu.m. The Cp–l curve for the machine is given as,
Cp = 0.003l3 + 0.015l2 + 0.075l
(a) Find k and c for 60 m height, if
È Z1 Ø
n 1 0.0881 ln É Ù
È c2 Ø È Z2 Ø È k2 Ø Ê 10 Ú
Éc Ù ÉZ Ù and Ék Ù
Ê 1Ú Ê 1Ú Ê 1Ú È Z2 Ø
1 0.0881 ln É Ù
Ê 10 Ú
where
0.37 0.0881 ln (c1 )
n
ÈZ Ø
1 0.0881 ln É 1 Ù
Ê 10 Ú
(b) Find an expression for CplRated as function of VR.
(c) Calculate capacity factor CF for different values of VR using the following equation:
exp ( (Vc / c)k ) exp ( (VR / c) k )
CF exp ( (VF /c) k )
(VR / c) k (Vc / c)k
(d) Calculate PN for different values of VR using the above values. Find VR for maximum PN.
(e) Find rated electrical power output capacity, PeR of the generator to be installed.
(f) Find its Annual Energy Output (AEO).
(g) Find the number of hours it will run in a year.
(h) Calculate energy content of the wind at 60 m height and compare it with AEO.
2. A wind machine has 3 linearly tapering blades. The radius at blade root is 0.75 m and chord
at this point is 1.2 m. Radius and chord at tip of the blade are 15 m and 0.4 m, respectively.
(a) Calculate the solidity of the rotor.
(b) If each blade weighs 1.5 tons, then find linear density, rbl of the blade. Assume uniform
distribution of the mass along the length of the blade.
(c) Derive an expression for centrifugal force of the blade element if the rotational speed of
rotor is w. Integrate it to yield an equation for centrifugal force acting at the root.
(d) Calculate the centrifugal force acting at the blade root if the machine rotates at constant
rotational speed of 42 rpm. Assume that the coning angle is zero.
(e) If the machine develops 225 kW, calculate the driving torque on the rotor. Assume
hmech = 0.95 and helect = 0.93.
(f) Draw velocity diagram at the blade cross section at 75 per cent radius and show, a, b
and f. Calculate f.
(g) Calculate the thrust on each blade if Vf = 25 m/s, Cl = 0.5 and Cd = 0.01 at furling all
along the blade and that the aerodynamic configuration at 75 per cent radius represents
the blade. Annual average air density = 1.155 kg/cu.m.
(h) If the blades were free coning, what would be the coning angle at furling? Neglect
changes in the values of thrust and centrifugal force due to change in angles.
(i) If the wind machine is mounted upwind on a cylindrical steel tower of 30 m height and
0.5 m diameter midway, then what should be the overhang shaft length measured from
the centreline of the tower if the clearance of 0.2 m has to be maintained between the
blade tip and tower at furling?
3. lr, a and a¢ (using the usual notations) can be calculated from application of wake analysis.
These values are given in Table 5.13. If Cl = 2pa with a in radians for the blade airfoil and
if the blade is to be designed for Cl = 0.7 for a 3-bladed propeller machine with rated wind
speed of 8 m/s and rotor speed of 38.18 rpm, then based on blade element theory, calculate:
(a) chord at radius of 4 m, 8 m and 12 m
(b) blade angle at radius of 4 m, 8 m and 12 m.
Take air density = 1.165 kg/m3 and Cd/Cl = 0.01.

TABLE 5.13 Values of lr, a and a¢
lr a a¢
0.5 0.298 0.543
1.0 0.317 0.183
2.0 0.328 0.052
3.0 0.331 0.024
4.0 0.332 0.014
5.0 0.332 0.009
6.0 0.333 0.006
7.0 0.333 0.004
8.0 0.333 0.004
9.0 0.333 0.003
10.0 0.333 0.002
11.0 0.333 0.002
12.0 0.333 0.002
4. A horizontal axis wind machine with 30 m diameter rotor has 3 blades with blade-root at
0.5 m radius from the axis. At root, its chord is 0.75 m and tip the chord is 0.5 m. The lift
coefficient, Cl for the blade airfoil can be taken as, Cl = 2pa, where a is the angle of attack
on the blade airfoil (radians) and e for the blade airfoil can be taken as e = Cd/Cl = 1/200. The
rotor is designed to rotate at constant speed of 47 rpm. At the tip, the blade makes 4° w.r.t.
the rotor plane.
(a) Define different parameters of the rotor and write down the governing equations if the
rotor is to be analysed based on the ‘Blade Element Theory’.
(b) Write down the equation for Cp.
(c) Write down the procedure to solve the equations.
(d) Is the given data sufficient to solve the equations? If not, then what additional information
do you require?
5. Darrieus Machine of diameter, D = 17 m and height, H = 17 m has troposkien blades. If the
swept area for this machine is given by As = 2D2/3, then find power of the machine, if
undisturbed wind speed = 10 mis, air density = 1.165 kg/m3 and Cp = 0.3.
6. The aerodynamic performance of this machine can be predicted on the basis of blade
element momentum method. The machine is designed to have 60 m diameter and 13.37 rpm
rotor speed. The root diameter is 2 m, number of blades is 3, and air density at the site is
1.16 kg/m3.
(a) Explain the procedure followed in BEM method to calculate a, a¢ and f.
(b) Find (1 – a), a¢ and f at 65.5 per cent of the blade length at rated wind speed. The design
strategy is that the product of chord and lift coefficient is maintained at 0.32 m all along
the blade using the following approximation for l > 2.
For values of l > 2, the relations between a, a¢ and lr can be given by

(1 AMr ) 2
(1 a)
(1 AMr ) 2 AMr
A
a
Mr A
with A = (No. of blades × Chord × Lift coefficient × Rotor speed)/(8p wind speed). f can
be calculated using trigonometric relation for velocity triangle at blade aerofoil section.
(c) Find the torque and thrust developed on the above rotor at rated wind speed assuming
that the conditions at 65.5 per cent blade length represent the aerodynamic loading on
the blade. The drag-to-lift ratio for the blade cross section is 0.007.
(d) The mechanical and electrical efficiencies are 0.95 and 0.97 at rated wind speed. Find
the power developed by the above machine.
(e) The weight of each blade is 3 tons. Its weight per unit length is inversely proportional to
its radius. Further, its weight per unit length at root is 4 times that at tip. Calculate the
centrifugal force acting on the blades discussed in the above questions. Calculate the
point of action of this force on the blade.
(f) Find the coning angle of the blade if moments due to thrust load and moments due to
centrifugal force should balance each other at the rated wind speed.
7. A wind machine rotating at 17 rpm has 2 linearly tapering blades. Radius at root is 1.5 m and
at tip of the blade is 47.3. The blade solidity is 0.01.
(a) If each blade weighs 21 tons, then calculate the centrifugal force acting at the blade root.
Assume that the coning angle is zero.
(b) If the machine develops 1.2 MW, calculate the driving torque on the rotor. Assume
hmech = 0.95 and helect = 0.93.
(c) Draw velocity diagram at the blade cross section at 66 per cent radius and show a, b and
f. Calculate f.
(d) Calculate the thrust on each blade if Vfurling = 42 m/s, Cl = 0.7 and Cd = 0.01 at furling all
along the blade and that the aerodynamic configuration at 66 per cent radius represents
8. lr, a and a¢ (using the usual notations) can be calculated from application of wake analysis.
These values are given in Table 5.14 for a propeller machine with rated wind speed of 8 m/s
and rotor speed of 38.18 rpm. Then, based on wake analysis for actuator disc, with air
density = 1.165 kg/m3, machine diameter = 30 m, hub diameter = 2 m and taking 2 m wide
strips of the rotor disc from hub to tip for numerical calculations, calculate:
(a) the thrust acting on the rotor at rated wind speed.
(b) the driving torque on the rotor at rated speed.
(c) The power developed by the rotor at rated wind speed.
TABLE 5.14 Values of lr, a and a¢
lr a a¢
0.5 0.298 0.543

1.0 0.317 0.183
2.0 0.328 0.052
3.0 0.331 0.024
4.0 0.332 0.014
5.0 0.332 0.009
6.0 0.333 0.006
7.0 0.333 0.004
8.0 0.333 0.004
9.0 0.333 0.003
10.0 0.333 0.002
11.0 0.333 0.002
12.0 0.333 0.002
9. A wind machine with 60 m diameter and 1 m hub radius has 3 blades weighing 10 tons
each. The machine rotates at 17 rpm The blade mass per unit length is given as, rb = A/r,
where r is the radius and A is a constant. The blades are coned at an angle of 3°.
(a) Find the value of A.
(b) Calculate the centrifugal force acting on the blades as a function of r and on the hub.
(c) Calculate the bending moment due to the centrifugal force acting on the hub.
(d) If the coning effectively nullifies bending moment at the blade root due to thrust at
rated speed, then find the thrust acting on the rotor. Assume that the thrust may be
considered to be acting at 20 m radius at rated speed.
(e) Calculate maximum driving torque acting on the rotor, if PeR = 1.2 MW, hm = 0.95 and
he = 0.97.
10. A wind machine rotating at 17 rpm has 2 linearly tapering blades with the ratio of blade
weight per unit length at root to that at tip as 5 : 1. Radius at blade root is 0.85 m and at tip of
the blade is 47.3 m.
(a) If each blade weighs 12 tons, then calculate the centrifugal force acting at the blade root.
Assume that the coning angle is zero.
(b) If the machine has Cp = 0.378 at 11.5 m/s, calculate driving torque on the rotor and the
power developed at generator. Assume hmech = 0.95 and helect = 0.93.
(c) Draw velocity diagram at the blade cross-section at 66 per cent radius and show a, b
and f. Calculate f with ideal values of a and a¢.
(d) Calculate thrust on each blade if Vfurling = 42 m/s, Cl = 0.7 and Cd = 0.01 at furling at
66 per cent radius and that the aerodynamic configuration at 66 per cent radius represents
(e) Calculate moment at the blade root due to thrust at furling speed if the thrust force can
be considered to be acting at 66 per cent radius.
11. Define power coefficient, CP and tip speed ratio l, for a horizontal axis wind rotor of area, A
in a wind speed of U.
(a) At an element of the blade of the rotor, the relative velocity of the wind to the blade
undergoes an axial velocity reduction by a factor a and a transverse velocity increase by
a factor a¢. Explain the origins of these factors and state the conditions under which CP
attains its maximum value.
(b) With reference to a large wind turbine suitable for use in a wind farm, explain why
aerofoil sections having very high lift/drag ratio are necessary.
(c) Describe, with the use of sketches, the general arrangement of gear box, electrical
machinery and controls in the nacelle of a large wind turbine.
12. What are cut-in, rated and cut-out wind speeds with reference to power performance
characteristics of a wind turbine? How it is chosen for optimal design of wind turbine?
13. How the strip theory is used for design of rotor blades of HAWT?
14. List all the design tasks involved in the design of a modern wind turbine.
15. For one blade 25 kW, 10 m diameter, pitch 0° and blade is divided into ten stations with
wind speed 10 m/s and tip speed ratio X = 6.11 following sample output table is given as
design example:
Blade element data for delta beta = 0.00
X = 6.11, yaw = 0.00
Refer Table 5.15, Figures 5.43, 5.44, 5.45, and 5.46.
TABLE 5.15 Design parameters of blade
Element 1 2 3 4 5 6 7 8 9 10
Theta 180 180 180 180 180 180 180 180 180 180
Vel 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
A 0.296 0.140 0.188 0.204 0.230 0.213 0.195 0.206 0.231 0.39
AP 0.073 0.021 0.016 0.012 0.010 0.008 0.006 0.006 0.006 0.39
Cl 0.813 1.005 1.160 1.206 1.334 1.311 1.168 1.037 0.918 0.770
Cd 0.014 0.098 0.053 0.043 0.020 0.019 0.016 0.014 0.013 0.010
Phi 49.92 42.48 27.54 20.14 15.45 13.03 11.35 9.74 8.34 6.71
ANG 7.92 19.18 15.74 14.84 13.35 12.93 11.35 9.74 8.34 6.71
TC 0.384 0.526 0.622 0.656 0.707 0.665 0.609 0.610 0.609 0.54
QC 0.040 0.059 0.073 0.075 0.083 0.079 0.074 0.073 0.0699 0.03
PC 0.243 0.363 0.443 0.459 0.508 0.485 0.453 0.443 0.421 0.34
TD, lb/ft 2.64 6.03 11.90 17.57 24.37 28.01 30.31 35.04 329.6 41.6
QD,
ft-lb/ft 4.38 10.92 22.21 32.26 45.86 53.47 59.02 66.73 71.85 65.5
PD, kW 0.024 0.298 0.606 0.880 1.251 1.458 1.610 1.820 1.959 1.78
Rey, *106 0.92 0.862 0.922 0.931 0.910 0.868 0.890 1.004 1.132 1.13
Rotor pitch X TC QC PC VO TD MD QD PD
2 blades m/s lb ft-lb ft-lb kW
0.0 6.1 0.614 0.070 0.427 10.0 752 3984 1372 23.3
0 1 2 3 4 5 6 7 8 9 10
FIGURE 5.43 Variation of blade geometry along the length.
7 m/s 9 m/s 11 m/s 13 m/s
4
Power, kW
0
1 2 3 4 5 6 7 8 9 10
Blade station
FIGURE 5.44 Variation of blade elemental power output along the length.
29
28
Power, kW
27
26
25
24
–5 –4 –3 –2 –1 0 1 2 3 4 5
Pitch angle, degrees
FIGURE 5.45 Variation of power output vs. pitch angle.
45
40
35
30
Power, kW
25
20
15
10
0
3 4 5 6 7 8 9 10 11 12 13 14 15 16
Wind speed, m/s
FIGURE 5.46 Power performance characteristics with varying wind speed.
Twist and plan form for 25 kW wind turbine blade. Blade is divided into ten sections for
analysis, and the station is at the mid-point of the section.
Prediction of power output for one blade station for four wind speeds, tip speed ratio = 6.1.
Prediction of rotor power output for different pitch angles at 10 m/s. The turbine is a fixed
pitch, constant-rpm machine.
Theoretical power curve for 25 kW wind turbine, tip speed ratio = 6.1.
16. A blade is 12 m long, weight is 500 kg, and the centre of mass is at 5 m. What is the torque
if the force is 320 Nm?
17. Find the power loss for three struts on a HAWT. Struts are 4 m long, 2.5 cm in diameter.
Rotor speed is 180 revolutions per minute. Use numerical approximation by dividing strut
into 1 m sections and calculate at midpoint of section. Then add the value for each section,
Cd = 1.
18. Calculate the power loss for the struts on VAWT. Centre tube, torque tube, diameter
= 0.5 m. Struts are at top and bottom, 2 m long from torque tube to blades, diameter = 5 cm,
rotor speed is 80 rpm. Cd = 1.
19. Find the value of u (speed of drag device) that produces the maximum CP for a drag device.
Use equation F/A = 0.5rV2CD, wind speed is V and V0 is the wind speed at infinity.
20. Aerodynamic efficiency can be maintained for different solidities of the rotor. If solidity
increases, will you increase or decrease the tip speed ratio?
21. Explain the difference in performance of a wind turbine if it
(a) Operates at a constant tip speed ratio
(b) Operates at constant rpm.
22. What is the maximum theoretical efficiency for a wind turbine? What general principles
were used to calculate this number?
23. If the solidity of the rotor is very small, for example, a one-bladed rotor, what is the value of
the rpm for maximum CP compared to the same size rotor with higher solidity?
24. Calculate the value of axial interference factor for which CP is maximum for a lift device.
Then show that this gives a maximum CP = 59 per cent. Find the value of a that produces the
maximum CP by plotting the curve {P/A = 0.5rV02 4a(1 – a)2} for different values of a.
25. A rotor reaches maximum CP at a tip speed ratio of 7. Calculate rotor rpm for four different
wind turbines (diameters 5, 10, 50 and 100 m) at wind speeds of 10, 20 and 30 m/s.
26. A wind turbine that operates at constant rpm will reach maximum efficiency at only one
wind speed. What wind speed should be chosen?
For Questions 27–33, specifications for a wind turbine are induction generator (rpm = 65),
fixed-pitch, rated power = 300 kW, hub height = 50 m, rated wind speed = 18 m/s, tower
head weight = 3,091 kg; rotor : three blades, mass of one blade = 500 kg, hub radius = 1.5 m,
rotor radius = 12 m.
27. How fast is the tip of the blade moving?
28. How fast is the blade root (at hub radius) moving?
29. Put the mass at the midpoint and calculate the kinetic energy for one blade. Assume the mass
of the blade is distributed evenly over ten sections. What is the kinetic energy for one blade?
30. At rated wind speed, calculate the torque.
31. At 10 m/s, what is the thrust (force) on the rotor trying to tip the unit over? Calculate for that
wind speed over whole swept area.
32. If the unit produced 800,000 kWh/year, calculate output per rotor swept area.
33. Calculate the annual output per weight on top of tower, kWh/kg.
For Questions 34–41, specifications for a wind turbine are induction generator (rpm = 21),
variable-pitch, rated power = 1,000 kW, hub height = 60 m, rated wind speed = 13 m/s,
tower head weight = 20,000 kg, rotor: three blades, mass of one blade = 3,000 kg, hub
radius = 1.5 m, rotor radius = 30 m.
34. How fast is the tip of the blade moving?
35. How fast is the blade root (at hub radius) moving?
36. Place the mass at the midpoint of the blade and calculate the kinetic energy for one blade.
Assume the mass of the blade is distributed evenly over ten sections. Now, what is the
kinetic energy for one blade?
37. Calculate the torque at the rated wind speed.
38. At 15 m/s, what is the thrust (force) on the rotor trying to tip the unit over? Calculate for that
wind speed over the whole swept area.
39. If the unit produces 2,800,000 kWh/year, calculate the specific output, annual kWh/rotor
area.
40. Calculate the annual output weight on top of tower, kWh/kg.
41. For a 12 m blade, centre of mass at 5 m, weight = 500 kg, calculate the angular momentum
if the rotor is operating at 60 rpm.
42. For the blade in Question 41, the angular momentum is around 8 × 104 kg m2/s. Calculate
the torque on the blade at that point if the angular momentum of the rotor is stopped in 5 s.
Use T = DL/Dt, where T = torque, DL = change of angular momentum and Dt = change in
time. Then estimate the force trying to bend the blade.
43. Why are the blades for large wind turbines made from fibreglass-reinforced plastics?
44. Why are yaw rates limited on large wind turbines or yaw dampers installed on small wind
turbines?
45. How does furling work on small wind turbines?
46. For loss of load on small wind turbines connected to the utility grid, how long can it take for
overspeed shutdown?
47. For megawatt-size wind turbines, what is the most common configuration?
48. From internet search for blade design, make a paper model of the blade to see the principle
for passive control.
49. List two methods with brief description of non-destructive testing of wind turbine blades.
50. By approximately what per cent will bugs on blades reduce the power?

5 Wind Energy CH 5

Uploaded by

Copyright:

Available Formats

5 Wind Energy CH 5

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

5 Wind Energy CH 5

Uploaded by

Copyright:

Available Formats

5

Wind Turbine Design

TABLE 5.1 Design tasks (Contd.)

5.2 ROTOR TORQUE AND POWER

W U 2 (1  a)2  : 2 r 2 (1  a)2 (5.1)

FIGURE 5.1 A blade element sweeps out an annular ring.

FIGURE 5.2 Blade element velocities and forces.

Therefore, the additional axial force on the annulus is

SU (1  a) : r 2a r 2Q r E r 4Q SU (:r ) a (1  a) r 2 E r

EQ 4QSU  (:r ) a (1  a) r 2 E r  1 SW 2 BcCd cos G r E r

The power developed by the rotor is

5.2.1 Glauert Momentum Vortex Theory

FIGURE 5.5 Blade element parameters.

5.2.2 Optimal Rotors

and comparing this with Eq. (5.27) gives

The integration can be done in closed form, resulting in

SW 2r : Bc (Cl sin G  C d cos G ) 'r

For small e, this is closely given by

FIGURE 5.6 Local Froude efficiency versus local speed ratio.

5.2.3 Dual Optimum

TABLE 5.2(a) Optimum parameters for X1 = 6, X2 = 4, e = 0

r/R 1 0.8 0.64 0.512 0.4096 0.32768

TABLE 5.2(b) Optimum blading for X1 = 6, X2 = 4, e = 0, t = 0

r/R 1 0.8 0.64 0.512 0.4096 0.32768

TABLE 5.2(c) Optimum blading for X1 = 6, X2 = 4, e = 0, t = 2 deg

r/R 1 0.8 0.64 0.512 0.4096 0.32768

5.3 OPTIMUM DESIGN FOR VARIABLE OPERATION

The flow angle f is given by

FIGURE 5.8 Variation of inflow angle with local speed ratio.

which, for optimum operation, is

Twist angle degrees

TABLE 5.3 Blade Design of a 8.5 m radius rotor

Radius, r r/R Chord, c Pitch, b Thickness/chord

FIGURE 5.14 Aerodynamic characteristics of NACA Aerofoil.

At lower tip speed ratio, the entire

At high tip speed ratio of l = 12, the effect of

FIGURE 5.17 Difference in torque distribution

5.4 INFLUENCE OF REYNOLDS NUMBER

5.5 CAMBERED AEROFOILS

compared to symmetrical aerofoils for

Zero lift line

FIGURE 5.22 Designation of NACA xxxx aerofoil.

FIGURE 5.23 Typical characteristics of NACA aerofoil.

5.5.1 Rotor Sizing

5.6 LOAD CALCULATION

FIGURE 5.24 Power curve, Pmax, Pnom and Vnom.

W is the resulting wind velocity

increases the greatest loads. The dimensioning

distribution is taken first, and then the smaller

distribution for px. Curve (a) is the stochastic a and c concurrently

5.6.1 Blade Loads

TABLE 5.4 Description of blade coordinates

m = m(r) = blade mass per unit of length at distance r

TABLE 5.5 Blade line load

Average value Load range distributions

TABLE 5.6 Description of blade coordinates

5.6.2 Rotor Loads

Load Average Range Frequency Resonance frequency

** It may be assumed that My is always between 0 and 1.3 Mnom.

W U 2 (1 a)2 : 2 r 2 (1 a)2 (5.1)

SU (1 a) : r 2a r 2Q r E r 4Q SU (:r ) a (1 a) r 2 E r

EQ 4QSU (:r ) a (1 a) r 2 E r 1 SW 2 BcCd cos G r E r

SW 2r : Bc (Cl sin G C d cos G ) 'r