Problem Solutions

B.1 Problem Solving

Most of the problems in this text can be completed without reference to additional material
than is in the text. In some cases, new information is introduced in the problem statement to
extend the readers knowledge.
In some cases data files may be needed. These are available on the web site of the Wind
Energy Center at the University of Massachusetts ([Link]
This site also contains the Wind Engineering MiniCodes that have been developed at the
University of Massachusetts at Amherst. A number of these codes may be useful in solving
problems and, in some cases, may be needed to solve problems. The Wind Engineering
MiniCodes are a set of short computer codes for examining wind energy related issues,
especially in the context of an academic setting.
A number of files are used in the problems. In addition, a number of the problems have
solutions which are illustrated in spreadsheet files. Lists of these files are provided at the
end of this document.
It has been discovered that some problems have errors in the problem statements. This
obviously makes it difficult to solve the problem. A full list of errata and corrected problem
statements can also be found at the same web site mentioned above
([Link] The problem statements included with these
solutions have already been corrected.

B.2 Chapter 2 Problems

2.1 Based on average speed data only, estimate the annual energy production from a
horizontal axis wind turbine with a 12 m diameter operating in a wind regime with an
average wind speed of 8 m/s. Assume that the wind turbine is operating under standard
atmospheric conditions (ρ = 1.225 kg/m3). Assume a turbine efficiency of 0.4.

SOLUTION

The annual energy production (in kWh) can be determined from:

Annual energy production (kWh) = (efficiency) (P) (8760 hrs)

Where P is determined from Equation 2.7, P = πR 2 ρU 3 . For an average wind speed of 8

1
m/s: 2

P = π (6 )2 (1.225)(8)3 = 35.46 kW
1
2
Therefore:
2 Wind Energy Explained

Annual energy production = (0.4)(35.46)(8760) = 124,300 kWh

Note that future consideration of the variability of the wind regime and the characteristics
of the wind machine will greatly change the output of an actual machine operating in a real
wind regime.

2.2 A 40 m diameter, three bladed wind turbine produces 700 kW at a wind speed (hub
height) of 14 m/s. The air density is 1.225 kg/m3. Find:

a) The rotational speed (rpm) of the rotor at a tip-speed ratio of 5.0.

b) What is the tip-speed (m/s)?
c) If the generator turns at 1800 rpm, what gear ratio is needed to match the rotor speed
to the generator speed.
d) What is the efficiency of the wind turbine system (including blades, transmission,
shafts, and generator) under these conditions?

SOLUTION

a) λ = 5
ΩR (Equation 2.76)
λ=
U

Ω=
λU
=
(5.0)(14 m / s ) = 3.5 rad / s
R 20 m

N rotor =
(Ω rad / s )(60 s / min ) = 33.42 rpm
2π rad / rev

b) U tip = ΩR = (20 m )(3.5 rad / s ) = 70 m / s

c) Ngen =1800 rpm; Nrotor = 33.42 rpm

gearbox ratio = N gen / N rotor = 53.86

(d) P = ρπR 2C pηU 3 ; assume Betz limit: CP =16/27

1
2
Problems 3

η= = 0.56
P
ρπR 2C pU 3
1
2

2.3 a) Determine the wind speed at a height of 40 m over surface terrain with a few trees,
if the wind speed at a height of 10 m is known to be 5 m/s. For your estimate use two
different wind speed estimation methods.
b) Using the same methods as part a), determine the wind speed at 40 m if the trees
were all removed from the terrain.

SOLUTION

a) First Method: Assume a simple logarithmic wind profile. The wind speed at a height

U (z ) ln (z / z0 )
of 40 m can be calculated via Equation 2.34

=
U (z r ) ln (z r / z0 )
From Table 2.2 zo = 100 mm = 0.1 m

ln (40 / 0.1)
Thus

U (40) = (5) = 6.51 m / s

ln (10 / 0.1)
Second Method: Assume a power law wind speed profile. The wind speed at a height of
40 m is determined via Equation 2.36

U (z ) ⎛ z
α
⎞
=⎜ ⎟⎟
U (z r ) ⎜⎝ z r ⎠
Let’s assume α = 1/7 = 0.142, then:

⎛ 40 ⎞
U (40) = (5)⎜ ⎟ = 6.09 m / s
0.142

⎝ 10 ⎠

b) First Method: With the trees cut down, assume a rough pasture surface terrain. Using
the data from Table 2.2, zo = 10 mm = 0.01 m. Thus:
⎛ 40 ⎞
ln⎜ ⎟
U ( 40) = (5) ⎝
0.01 ⎠
= 6.0 m / s
⎛ 10 ⎞
ln⎜ ⎟
⎝ 0.01 ⎠
Second Method: If we still assume α = 1/7, the same estimate as part a) holds.
4 Wind Energy Explained

2.4 A 30 m diameter wind turbine is placed on a 50 m tower in terrain with a power law
coefficient (α) of 0.2. Find the ratio of available power in the wind at the highest point the
rotor reaches to its lowest point.

SOLUTION

D = 30 m, hub height = 50 m, α = 0.2

zlow = 50 - 15 = 35 m
zhigh = 50 + 15 = 65 m

( ) = ⎛⎜ zhigh ⎞⎟α = ⎛ 65 ⎞α = 1.13

The ratio of the wind speeds is:

U (zlow ) ⎜ z ⎟ ⎜ ⎟
U z high
⎝ low ⎠ ⎝ 35 ⎠

( ) = ⎡U (z high )⎤ 3 = 1.133 = 1.44

The ratio of the powers is the cube of the ratio of the wind speeds:

P(z low ) ⎣ U (z low ) ⎦

⎢ ⎥
P z high

2.5 Find the size of a wind turbine rotor (diameter in m) that will generate 100 kW of
electrical power in a steady wind (hub height) of 7.5 m/s. Assume that the air density is ρ =
1.225 kg/m3, Cp = 16/27 and = 1.

SOLUTION

a) P = 100 kW; U = 7.5 m/s, ρ = 1.225 kg/m3, CP = 16/27, =1

P= ρπR 2C pηU 3 ; therefore:

1
2

D=2 = 28.8 m
P
ρπC pηU 3
1
2

2.6 From an analysis of wind speed data (hourly interval average, taken over a one year
period), the Weibull parameters are determined to be c = 6 m/s and k = 1.8.
a) What is the average velocity at this site?
b) Estimate the number of hours per year that the wind speed will be between 6.5 and
7.5 m/s during the year.
c) Estimate the number of hours per year that the wind speed is above 16 m/s.
Problems 5

SOLUTION

a) From Equation 2.62

∫
U = U p(U )dU= cΓ ⎜1 + ⎟
∞
⎛ 1⎞
0
⎝ k⎠

Therefore

U = cΓ ⎜1 + ⎟ = (6 )Γ ⎜1 + ⎟ = (6 )(0.8893) =5.34 m/s

⎛ 1⎞ ⎛ 1 ⎞
⎝ k⎠ ⎝ 1.8 ⎠

b) The number of hours that the wind speed will be between 6.5 and 7.5 m/s during the
year is equal to the probability that the wind speed will be between 6.5 and 7.5 m/s times
the number of hours in the year. It is most conveniently found using the cumulative
distribution function, Equation 2.61
⎛ ⎛ 6.5 ⎞ k ⎞ ⎛ ⎛ 7.5 ⎞ k ⎞
p(6.5<U < 7.5) =F (7.5) − F (6.5) = exp⎜ − ⎜ ⎟ ⎟ − exp⎜ − ⎜ ⎟ ⎟ = 0.9198
⎜ ⎝ c ⎠ ⎟ ⎜ ⎝ c ⎠ ⎟
⎝ ⎠ ⎝ ⎠

Thus, the number of hours = 0.09198 x 8760 = 806 hr

c) Similar to part b), the probability that the wind speed will be equal to or greater than 16
m/s is given by:

∫ p(U)dU = 1 − F (16 ) = exp(−⎜ ⎟ ) = 0.002896

∞
⎛ 16 ⎞
p(U ≥16) =
1.8

16
⎝ 6⎠

Thus, the number of hours = 0.002896 x 8760 = 25 hr

2.7 Analysis of time series data for a given site has yielded an average velocity of 6 m/s. It
is determined that a Rayleigh wind speed distribution gives a good fit to the wind data.
a) Based on a Rayleigh wind speed distribution, estimate the number of hours that the
wind speed will be between 9.5 and 10.5 m/s during the year.
b) Using a Rayleigh wind speed distribution, estimate the number of hours per year that
the wind speed is equal to or above 16 m/s.

SOLUTION

Parts a) and b) could be solved as the previous problem assuming a Weibull distribution
with k = 2, which corresponds to the Rayleigh distribution, but the most direct method is to
use the exact equation for the Rayleigh distribution.

Thus, for a)
6 Wind Energy Explained

∫ p(U)dU=F (10.5) − F (9.5)

10 + 0.5
p(9.5 < U < 10.5) =

⎛ π ⎛ 9.5 ⎞ 2 ⎞ ⎛ π ⎛ 10.5 ⎞ 2 ⎞
10-0.5

= exp⎜ − ⎜ ⎟ ⎟ − exp⎜ − ⎜ ⎟ ⎟ =0.0494

⎜ 4⎝ 6 ⎠ ⎟ ⎜ 4⎝ 6 ⎠ ⎟
⎝ ⎠ ⎝ ⎠

Thus, the number of hours = 0.0492 x 8760 = 433 hr

b) Using Equation 2.59

⎛ π ⎛ 16 ⎞ 2 ⎞
p(U ≥ 16 ) = 1- F(16) =exp⎜ − ⎜ ⎟ ⎟= 0.00375
⎜ 4⎝ 6 ⎠ ⎟
⎝ ⎠

Thus, the number of hours = 0.00375 x 8760 = 33 hr

operating at standard atmospheric conditions ( ρ = 1.225 kg/m3) in a 8 m/s average wind

2.8 Estimate the annual production of a 12 m diameter horizontal axis wind turbine

speed regime. You are to assume that the site wind speed probability density is given by the
Rayleigh density distribution.

SOLUTION

The average power for this machine can be found from Equation 2.82:

⎛2 ⎞ 3
P = ρ⎜ D ⎟ U
2

⎝3 ⎠

Thus

⎛2 ⎞
P =1.225⎜ x12 ⎟ 83 = 40.2 kW
2

⎝3 ⎠

The annual energy production would be (40.2) x (8760) kWh = 352,000 kWh

2.9 Assuming a Rayleigh distribution, a researcher (see Masters (2004) Renewable and
Efficient Electric Power Systems. Wiley) has proposed the following simple relationship
for the capacity factor (CF) of a wind turbine:

CF = 0.087 U −
PR
D2

where U is the average wind speed at the hub, PR is the rated power (kW) and D is the
rotor diameter (m). Furthermore, this researcher claims that this equation is accurate to
Problems 7

within 10% for capacity factors between 0.2 to 0.5. Your problem is to check out this
equation for a selection (say 5) commercial wind turbines.

SOLUTION

Based on wind turbine manufacturers’ data from the web, the following graph was obtained
for a number of wind turbines operating at various average hub height wind velocities.
Thus, it looks like this prediction method gives a reasonable first estimate for capacity
factor when a minimum of wind data and turbine data are available.

2.10 A wind turbine with a rotor diameter of 55 m is rated at 1 MW at a hub height wind
speed of 14 m/s. It has a cut-in speed of 4 m/s and a cut-out speed of 25 m/s. Assume that
this machine is located at a site where the mean wind speed is 10 m/s and that a Rayleigh
wind speed distribution can be used. Calculate the following:

a) The number of hours per year that the wind is below the cut-in speed.
b) The number of hours per year that the machine will be shut down due to wind speeds
above the cut-out velocity.
c) The energy production (kWh/year) when the wind turbine is running at rated power.

SOLUTION

a) The probability that the wind speed is smaller than or equal to a given wind speed is
given by Equation 2.59 using the Rayleigh probability density function.
8 Wind Energy Explained

This gives:

⎛ π ⎛U ⎞ ⎞
F (U cut −in ) = 1 − exp⎜ − ⎜ cut −in ⎟ ⎟ = 0.1181
2

⎜ 4⎝ U ⎠ ⎟
⎝ ⎠

Thus, the number of hours per year = (0.118) x (8760 hrs) = 1035 hr/yr

b) The number of hours when the wind is higher than the cut-out velocity is given by:

⎛ 2⎞
[1 − F (U cut −out )](8760) = exp⎜⎜ − π ⎛⎜ U cut −out ⎞⎟ ⎟⎟(8760) = (0.0074)(8760) = 64.7 hr / yr
4⎝ ⎠ ⎠
⎝
U

c) The number of hours that this occurs is the number of hours that the wind speed is
greater than rated speed minus the number of hours that the machine is shut down (part B).
The number of hours when the wind is higher than 14m/s is determined from:
⎛ 2⎞
[1 − F (U rated )](8760) = exp⎜⎜ − π ⎛⎜ U rated ⎞⎟ ⎟⎟(8760) = (0.2145)(8760) = 1879 hr / yr
4⎝ U ⎠ ⎠
⎝
Thus, the number of hr/yr at rated power = 1879 – 64.7 = 1814.3 hr/yr

And, the delivered energy is 1000 x 1814 kWh/yr = 1.814 x 106 kWh/yr

2.11 Based on the spreadsheet ([Link]) which contains one month of data (mph)
from Holoyke, MA, determine:
a) The average wind speed for the month
b) The standard deviation
c) A histogram of the velocity data (via the method of bins–suggested bin width of 2
mph)
d) From the histogram data develop a velocity–duration curve
e) From above develop a power–duration curve for a given 25 kW Turbine at the
Holyoke site.

0 < U ≤ 6 (mph)
For the wind turbine, assume:

P = U3/625 kW 6 < U ≤ 25 (mph)

P = 0 kW

25 < U ≤ 50 (mph)
50 < U (mph)
P = 25 kW
P = 0 kW
f) From the power duration curve, determine the energy that would be produced during
this month in kWh.
Problems 9

SOLUTION:

a) Using given data: Average wind speed = 10.4 mph

b) Using given data: Standard deviation = 5.45 mph

c) Using bin reduced data, the following graph results:

0.16
0.14
Frequency of occurance

0.12
0.10
0.08
0.06
0.04
0.02
0.00
0 10 20 30 40 50
Velocity, mph

d) From the histogram data, the following velocity duration curve results:

40

30
Velocity, mph

20

10

0
0 100 200 300 400 500 600 700
Hours

e) The following power duration curve results:

10 Wind Energy Explained

25

20
Power, kW

15

10

0
0 100 200 300 400 500 600 700
Hours

f) The total energy produced can be determined from integrating the product of the turbine
power and the numbers of hours of operation at that power level, yielding an annual energy
production of 2474 kWh

2.12 Using results from Problem 2.11, carry out the following:
a) Determine Weibull and Rayleigh velocity distribution curves and normalize them
appropriately. Superimpose them on the histogram of Problem 2.11.
b) Determine the Weibull and Rayleigh velocity duration and power duration curves
and superimpose them on the ones obtained from the histogram.
c) Using the Weibull distribution, determine the energy that would be produced by the
25 kW machine at the Holyoke site.
d) Suppose the control system of the 25 kW machine were modified so that it operated
as shown in Figure B.1 (and as detailed in Table B.1) How much less energy would be
produced at Mt. Tom with the modified machine? Find a fourth order polynomial fit to the
power curve. Use the Weibull distribution to calculate the productivity (in any manner you
choose). Plot the power duration curve using the modified power curve and the cubic
power curve from Problem 2.11e).
30
25 Cubic power curve
Power, kW

20
15
10 Modified power curve
5
0

5 10 15 20 25 30 35 40 45 50
Wind speed, mph
Problems 11

Figure B.1 Power curves for Problems 2.11 and 2.12

Table B.1 Power curve below rated power for Problem 2.12d

Wind speed (mph) Power (kW)

11 0
20 12.5
25 20
27.5 22.5
32.5 25

SOLUTION

a) For the Rayleigh distribution, one just needs the average velocity. For the Weibull
distribution, one needs to calculate k and c. The resulting values should be k = 2.016 and c
= 11.74 mph. The distributions are shown below. Because k = 2.016 the Weibull and
Rayleigh distributions are indistinguishable.
Note that the histogram of Problem 2.11 shows the probability that the wind is in a
wind speed bin that is 2 mph wide. The Rayleigh and Weibull probability density
distributions have units of probability per mph. For comparison with the histogram, the
Weibull and Rayleigh distributions in the graphs have been scaled by a factor of 2 to
represent probability per 2 mph.

0.16
0.14
Frequency of occurance

0.12
Data
0.10 Rayleigh/Weibull
0.08
0.06
0.04
0.02
0.00
0 10 20 30 40 50
Velocity, mph

b) The velocity duration and power duration curves are:

12 Wind Energy Explained

Velocity duration curve:

40

30
Velocity, mph

Histogram
20 Rayleigh/Weibull

10

0
0 100 200 300 400 500 600 700
Hours

Power duration curve:

25

20
Power, kW

15 Histogram
Rayleigh/Weibul
10

0
0 100 200 300 400 500 600 700
Hours

c) The total energy produced can be determined from integrating the product of the
turbine power and the numbers of hours of operation at that power level, yielding an annual
energy production of 2474 kWh using the histogram data and 2454 kWh using the Weibull
distribution.
d) Once a polynomial curve is fit to the data, the result look like:
Problems 13

25

Power, kW 20

15 Cubic power curve

Modified power curve
10

0
0 100 200 300 400 500 600 700
Hours

The power output from the modified turbine can be determined a number of ways, the
easiest involves determination of the area under the power duration curve. In all cases, the
resulting answer should be about 1848 kWh.

2.13 Similar to Equation 2.27 in the text, the following empirical expression has been used
to determine the power spectral density (psd) of the wind speed at a wind turbine site with
a hub height of z. The frequency is f (Hz), and n ( n = fz U ) is a non-dimensional
frequency.

=
f S(f) 11.40n
( 2.5u ∗ )2 1+192.4n 5 / 3

where

0.4[U (z )]
u∗ =
⎛ z ⎞
ln⎜⎜ ⎟⎟
⎝ z0 ⎠

Determine the power spectral density of the wind at a site where the surface roughness is
0.05 m ( z 0 ) and the hub height is 30 m, and the mean wind speed is 7.5 m/s.

SOLUTION

First solve for the friction velocity, u*: and f S (n)=

41.84
1+ n5 / 3
from:
14 Wind Energy Explained

u∗ = = 0.469 m / s
u( z )(0.4) 7.5(0.4)
=
z 30
ln( ) ln( )
zo 0.05

11.40( fz / U )
Substituting the expression for n as a function of f into the definition of the spectrum:

=
( 2.5u ∗ )2 1+192.4( fz / U )5 / 3
f S(f)

( )
Then S(f) is:

11.40(z / U ) 2.5u ∗
S ( f )=
1+192.4(z / U )5 / 3 f 5 / 3
2

Substituting in values of z = 30 m and U = 7.5 m/s:

S ( f )=
62.69
1+1939.3 f 5 / 3
A graph of the power spectral density (psd) is shown below:

2.14 This problem uses the power spectral density (psd) to examine variance in the wind.
A time series of hourly wind speeds (mph) from Mt. Tom for approximately 1 year is
included in the data [Link]. Routines to perform psd analysis are included
with the UMass Wind Engineering MiniCodes. When the psd is graphed vs. the frequency
it is hard to see features of interest. For this reason it is common to graph f S(f) on the y axis
vs. ln(f) on the x axis. When doing this the area under the curve between any two
frequencies is proportional to the total variance associated with the corresponding range of
frequencies.
Problems 15

a) Use the MiniCodes to calculate the psd for the Mt. Tom wind data. Focus on the
variations in wind over time periods of less than one month by using a segment length of
512.
b) Show from the results that the total variance as given by the integral of the psd (i.e.
S(f) vs. f) is approximately the same as what would be obtained in the normal way.
c) Show by equations that the area under the curve in a plot of fS(f) vs. ln(f) is the same
as it would be for a plot of S(f) vs. f.
d) Plot f S(f) vs. ln(f)
e) Find the amount of variance associated with diurnal fluctuations. Use frequencies
corresponding to cycle times from 22 hours to 27 hours. How much variance is associated
with higher frequency variations and how much with lower frequency variations?

SOLUTION

a) The psd is shown below:

∫ ∑ (S ( f ) + S ( f )[ f i+1 − f i ] .
∞ N −1
b) The variance may be found from σ 2 = S ( f )df ≈ i +1 ) The
i =1
i
0
value is 110.5 (mph)2. This is very close to the actual value, which is 118.2 (mph)2.

∫ ∫ ∫ ∫ ∫
c) The result can be shown by referring to the defining equation for the natural log:
∞ ∞ ∞ ∞
⎛ df ⎞
f S ( f )d (ln( f )) = f S ( f )d ⎜⎜ ⎟⎟ = S ( f )df = S ( f )df =σ 2
f
0 0 ⎝ f ⎠ 0 f
0
d) The graph is shown below.
16 Wind Energy Explained

e) The variance associated with frequencies between 22 and 27 hours (0.045 to 0.037 Hr-1)
can be found from integrating the PSD over the correct range of frequencies. The result is
about 6.6 mph2. Almost all of the variance in this data set of hourly averages is in
frequencies corresponding to time periods of longer than 20 hours.

2.15 A variety of techniques are available for creating data sets that have characteristics
similar to that of real data. The Wind Engineering MiniCodes include a few of these
methods. In the ARMA technique (see Appendix C) the user must input long-term mean,
standard deviation, and autocorrelation at a specified lag. The code will return a time series
with values that are close to the desired values. (Note: a random number generator is used
in the data synthesis routines, so any given time series will not be exactly the same as any
other.)
a) Find the mean, standard deviation, and autocorrelation for the Mt. Tom data:
[Link]. This data, in m/s, is collected at a 25 m height, with a sampling
frequency of 7.4 Hz. Plot a time series of the data. Determine the autocorrelation for a lag
of up to 2000 points.
b) Using the ARMA code with the autocorrelation at a lag of one time step, synthesize
and plot a time series of 10,000 data points with equivalent statistics to those found in part
a. Show a time series graph of the synthesized data.
c) Find the autocorrelation for a lag of up to 2000 points for the synthetic data and plot
the autocorrelations of both the real and the synthesized on the same graph.
d) Comment on any similarities or differences between the two plots.

SOLUTION

a) For the real data: mean = 9.76 m/s, sd. Dev =1.59 m/s, autocorrelation = 0.997 at a lag of
one time step. The data are shown below:
Problems 17

14
Mt. Tom 7.4 Hz wind data

12
Wind speed, m/s

10

0 200 400 600 800 1000 1200 1400 1600

Time, seconds

b) A sample synthetic time series is shown below:

14 Synthesized 7.4 Hz wind data

12
Wind speed, m/s

10

0 200 400 600 800 1000 1200

Time, seconds
c) The autocorrelations are shown below:
1.0
0.8
0.6
Autocorrelation

Mt. Tom data

ARMA data
0.4
0.2
0.0
-0.2
-0.4

0 50 100 150 200 250

Lag, seconds

d) Both autocorrelations look similar, with the wind showing some degree of correlation
over periods of 100 seconds (about 740 lags). The synthesized data is much more
negatively correlated than the real data over time periods longer than 100 seconds. The
synthesized data is also more correlated than the real data over shorter time periods (up to
50 seconds).
18 Wind Energy Explained

B.3 Chapter 3 Problems

3.1 The blades of a wind turbine are ready to be installed on a turbine on top of a ridge.
The horizontal blades are supported at each end by saw horses, when a storm front arrives.
The turbine crew huddles in their truck as the rain starts and the wind picks up, increasing
eventually to 26.82 m/s (60 mph). Realizing that the wind coming up the western slope of
the ridge roughly follows the 10 degree slope, the field engineer performs a quick
calculation and drives his truck upwind of the blades to disrupt the airflow around the
blades, preventing them from being lifted by the wind and damaged.
The blades are 4.57 m (15 feet) long, 0.61 m (2 feet) wide, and have a mass of 45.36 kg
(100 lbm). As the front arrives the temperature drops to 21.2°C (70°F). Assume that the
blades are approximately symmetric airfoils (the engineer remembered that potential flow

approximately: C l = 2π sin α ). Assume that the center of both the lift and the drag force is
theory predicts that, prior to stall, the lift coefficient of a symmetric airfoil is

concentrated over the center of mass of the blade and that the leading edge is facing into
the direction of the wind. Assume the air density is 1.20 kg/m3.
a) Was there a reason to be concerned? At what wind speed will the blades be lifted by
the wind, assuming that there is no drag?
b) If they are lifted by a 26.82 m/s (60 mph) wind, how fast will they be accelerated
horizontally, if the blade's lift to drag ratio, C d C l , is 0.03?

SOLUTION

a) From the geometry of the problem:

Lift
FL FL cos( α)
F
H

FD Drag

α
FD cos( α) FL sin( α) Wind

The planform area of the blades is 2.79 m2 (30 ft2). From the definition of the lift
coefficient:

FL = 1 Cl ρU 2 A = 1 (2π sin α )ρU 2 A

2 2

Assuming that there is no drag, the criteria for the blades being lifted by the wind is:

W = FL cos α = (π sin α )ρU 2 A cos α

Problems 19

where W is the blade weight. Therefore, the wind speed at which the blades will move is:

U=
(π sin α cos α )ρA
W

Here α =10 o = 0.175 radians, ρ = 1.20 kg/m3, so:

U= = 15.76 m/s = 51.7 ft/s = 35.2 mph .

(π (0.173)(0.985))(1.2)(2.79)
45.36

The engineer did the correct thing and got a pat on the back from his boss.

b) Again, from the geometry of the problem:

FH = (− Cl sin α + Cd cos α ) 1 ρU 2 A
2
or

FH = C l ⎛⎜ − sin α + d cos α ⎞⎟ 1 ρU 2 A
⎝ ⎠2
C
Cl

so

FH = (πsinα )⎛⎜ − sin α + d cos α ⎞⎟ ρU 2 A .

⎝ ⎠
C
Cl

If U= 26.82 m/s (60 mph, 88 ft/s), so:

FH = (π 0.174 )(− 0.174 + 0.03(0.985))1.2(26.82 )2 2.79 = −190.2 N = −42.7 lbf

And the horizontal acceleration, a H , of the blades is FH/m:

a H = FH m = − 190.2 45.36 = −4.19 m sec 2 = −13.7 ft sec 2

The blade would have taken off INTO the wind, but for the quick thinking of the field
engineer!

3.2 An inventor proposes to use a rotating cylinder to produce lift in a new wind energy
device. The cylinder will be D = 0.75 m in diameter and will be H = 7.5 m high. It will
rotate with a speed of n = 60 rpm.

a) Recall that circulation around a cylinder is the integral of the tangential velocity about its
perimeter. Show that the circulation is given by:
π 2 D 2n
Γ =
60
20 Wind Energy Explained

Hint: this is easiest done by using polar coordinates.

b) Find an expression for the lift per unit height around a rotating cylinder in terms of the
free stream wind velocity, U (m/s), the rotational speed, n (rpm) and the diameter, D (m),
of the cylinder
c) Find the lift force produced by the cylinder in the inventor’s device in a 10 m/s wind.

SOLUTION
a) The circulation is given by:

⎛ n ⎞⎛ D ⎞ πnD
U tan = (2π )⎜ ⎟⎜ ⎟ =
The air velocity at the perimeter of the cylinder is:

⎝ 60 ⎠⎝ 2 ⎠
m/ s

∫ ∫
60
2π D 2π D ⎛ πDn ⎞
Γ = U tan dθ = ⎜ ⎟dθ
0 2 ⎝ 60 ⎠

∫
0 2

D 2 nπ 2π D 2 nπ
= dθ = (2π )
180 0 180
π 2 D2n
= q.e.d .
60

L = ρ U Γ = ρ U π 2 D 2 n / 60
b) Lift force per unit length is given by:
~

c) Find the lift force produced by the cylinder in the inventor’s device in a 10 m/s wind.
Use: D = 0.75 m, H = 7.5 m high, n = 60 rpm.
L = ρ U Γ = ρ U π 2 D 2 n / 60

( ) ( )( )
~

= 1.225 kg / m3 (10 m / s ) 0.752 m 2 π 2 ⎜⎜

⎛ 60 rev / min ⎞
⎟⎟
⎝ 60 s / min ⎠
= 68 N / m
The total lift force is thus (68 N/m)(7.5 m) = 510 N.

3.3 The operating conditions found at two different points of a blade on a wind turbine are
(Table B.2):

Table B.2

Location Wind velocity at Wind velocity at Chord Chord Angle of attack

r/R blade (m/s) blade (ft/s) (m) (ft) (degrees)
0.15 16.14 52.94 1.41 4.61 4.99
0.95 75.08 246.32 0.35 1.15 7.63

1.33 × 10-5 m2/s. What are the Reynolds numbers found at each blade section?
These conditions were determined at 0°C (32°F), for which the kinematic viscosity is
Problems 21

SOLUTION

The Reynolds number is defined as:

Re = Uc
ν
where: ν in the kinematic viscosity of air, U is the velocity of airflow at airfoil and c is the
chord length. At 0°C (32°F) the kinematic viscosity of air is 1.33 × 10-5 m2/s. The
operating Reynolds numbers are, then:

Location, r/R Re
0.15 1.71e6
0.95 1.98e6

3.4 a) Find θ, θP, θT, and c for one blade section from r/R = 0.45 to r/R = 0.55 (centered on
r/R = 0.50) for an ideal blade (assume C d = 0, a ′ = 0). Assume λ =7, B=3, R=5 m, and
C l =1.0 and the minimum C d C l occurs at α = 7.
b) Assume that C d C l actually equals 0.02 for the above blade section and that the
free stream wind speed, U, equals 10 m/s. Find U rel , dFL1 , dFD1 , dFN 1 , dFT 1 , dQ1 for the

Use a = 1/3, a ′ = 0. Assume the air density is 1.24 kg/m3 (20C).

blade section. Don't forget to consider that the wind velocity is slowed down at the rotor.

c) For the same blade section find C l , α and a using the general strip theory method
(including angular momentum). Also find C l , α and a if the rpm is increased such that λ

curve is C l = 0.1143α + 0.2 ( α in degrees): i.e. C l = 0.2 at α = 0 degrees, C l = 1.0 at α

= 8. Ignore drag and tip loss. Use a graphical approach. Assume that the empirical lift

= 7 degrees.

SOLUTION

a) For r/R = 0.50, the section radius, r, angle of relative wind, ϕ , chord, c, section pitch,
θ P , and twist, θ T can be determined from:

r = (r R )R
λr = (r R )λ
ϕ = tan −1 (2 (3λr ))
c = 8πr sin ϕ (3BCl λr )
θP = ϕ − α
θT = θ P − θ P ,0

where θ P,0 is the blade pitch angle at the tip. The results of the calculations appear below.
22 Wind Energy Explained

r R r, m ϕ , deg θ P , deg θ T , deg c, m

0.50 2.50 10.78 3.78 5.34 .373

b) Using:

U rel = U (1 − a ) sin ϕ (3.64)

dFL = Cl 1
2
ρU rel 2 c dr (3.65)

dF D = C d 1
2
ρU rel 2 c dr (3.66)

dFN = dFL cos ϕ + dFD sin ϕ (3.67)

dFT = dFL sin ϕ − dFD cos ϕ (3.68)

dQ = B r dFT (3.70)
in which dr = 0.10, the various forces and torque at r/R = 0.50 can be determined. If the air
density is assumed to be ρ = 1.145 kg/m3 then:

Relative wind, ϕ dF L dF D dF N dFT dQ

degrees N N N N N
35.63 135.6 2.71 133.7 22.71 170.3

These forces and torques will be proportionally different if a different air density is
assumed.

(cos ϕ − λr sin ϕ )
c) Strip theory requires that, for any given blade geometry:

C l = 4 sin ϕ
σ ' (sin ϕ + λ r cos ϕ )
in which σ ′ = Bc 2πr , λ r = (r R )λ and ϕ = α + θ P .
But the experimentally determined blade performance fixes the lift coefficient as a
function of angle of attack:

C l = 0.1143α + 0.2

The geometry of the blade has been determined in part a). This defines θ P , the section

unknowns: C l and α . These can be solved graphically or numerically. Below is a table

pitch. The tip speed ratio is also a prescribed value. Thus there are two equations with two

of the values of the experimentally determined lift coefficients and those from the strip
Problems 23

theory equation for λ = 7 and for λ = 8. The angle of attack at which the experimentally
determined lift coefficient equals that from strip theory is the angle attack found in turbine
operation.

α
λ =7 λ =8
Cl Cl Cl
degrees experiment
2 0.429 1.014 0.821
2.2 0.451 1.028 0.828
2.4 0.474 1.040 0.834
2.6 0.497 1.051 0.839
2.8 0.520 1.061 0.842
3 0.543 1.070 0.844
3.2 0.566 1.077 0.845
3.4 0.589 1.082 0.844
3.6 0.611 1.086 0.842
3.8 0.634 1.089 0.838
4 0.657 1.091 0.833
4.2 0.680 1.091 0.827
4.4 0.703 1.089 0.819
4.6 0.726 1.087 0.810
4.8 0.749 1.083 0.800
5 0.771 1.077 0.788
5.2 0.794 1.071 0.775
5.4 0.817 1.063 0.761
5.6 0.840 1.053 0.745
5.8 0.863 1.042 0.728
6 0.886 1.030 0.710
6.2 0.909 1.017 0.690
6.4 0.931 1.002 0.669
6.6 0.954 0.986 0.646
6.8 0.977 0.968 0.623
7 1.000 0.949 0.598
7.2 1.023 0.929 0.571
7.4 1.046 0.908 0.543
7.6 1.069 0.885 0.514
7.8 1.091 0.860 0.483
8 1.114 0.835 0.452

The operating point can also be found graphically:

24 Wind Energy Explained

1.4

1.2

1.0
Lift Coefficient

0.8

0.6

0.4
Experimental
0.2 lambda=7
lambda=8
0.0
0 2 4 6 8 10 12 14
Angle of Attack, degrees

For λ = 7, C l = .972, at α = 6.75 degrees.

Either method gives:

For λ = 8, C l = 0.782, and α = 5.10 degrees.

[ ]
The axial induction factor, a, can be determined from:

a = 1 1 + 4 sin 2 ϕ (σ ′C L cos ϕ )

For λ = 7, a = 0.338.
This gives:

For λ = 8, a = 0.366.
Thus, the ideal blade section would have a slightly lower angle of attack than the 7 degrees
used in the ideal analysis, a slightly lower lift coefficient and a higher axial induction
factor.
At the higher tip speed ratio, the angle of attack and lift coefficient decrease compared
to the design conditions and the axial induction factor increases.

3.5 a) Find θ , θ p , θ T , and c at all 10 locations (r/R = 0.10, 0.20, …, 1.0) for the Betz
optimum blade. Assume λ = 7, B = 3, R = 5 m, and C l = 1.0 and the minimum C d C l
occurs at α = 7.
b) Sketch the shape (planform) of the blade, assuming that all the quarter chords lie on a
straight line.
c) Illustrate the blade twist by drawing plausible airfoils with properly proportioned
chord lengths, centered at the quarter chord chords for r/R = 0.10, 0.50, 1.0. Be sure to
show where the wind is coming from and what the direction of rotation is.
Problems 25

SOLUTION

a) For r/R = 0.10, 0.20, ............., 1.0, the section radius, r, angle of relative wind, ϕ , chord,
c, section pitch, θP, and twist, θT, can be determined from:

r = (r R )R
λr = (r R )λ
ϕ = tan −1 (2 (3λr ))
c = 8πr sin ϕ (3BCl λr )
θP = ϕ − α
θT = θ P − θ P ,0

where θP,0 is the blade pitch angle at the tip. The results of the calculations appear below.

pitch, θ p twist, θ T
Section Section Angle of Section Section Chord, c

wind, ϕ
radius radius, r relative (m)
r/R (m) degrees (degrees)
(degrees)
0.10 0.50 43.60 36.60 38.16 1.376
0.20 1.00 25.46 18.46 20.02 0.858
0.30 1.50 17.61 10.61 12.17 0.604
0.40 2.00 13.39 6.39 7.95 0.462
0.50 2.50 10.78 3.78 5.34 0.373
0.60 3.00 9.02 2.02 3.58 0.313
0.70 3.50 7.75 0.75 2.31 0.269
0.80 4.00 6.79 -0.21 1.35 0.236
0.90 4.50 6.04 -0.96 0.60 0.210
1.00 5.00 5.44 -1.56 0.00 0.189

b) The resulting plan form, with the quarter chords lying in a straight line appears
below:
26 Wind Energy Explained

0.4
0.2
0.0
Chord, meters
-0.2
-0.4
-0.6
-0.8
-1.0
-1.2
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Radius, meters
b) The blade twist, with the quarter chords lying in a straight line appears below:

rΩ

U (1- a )

Blade Twist Illustration

3.6 Blades for a two-bladed wind turbine with a 24 m diameter have been designed for a
tip speed ratio of 10. The 12-meter blades have the geometric and operational parameters

assuming C l = 1.0, a ′ = 0, no drag, and a = 1/3 using the methods outlined in the text for
listed in Table B.3 for operation at the design tip speed ratio. The rotor was designed

the design of an ideal rotor.

We want to know the rotor power coefficient for two assumed conditions: C d = 0 and
C d = 0.02. Note that the two equations that have been derived do not serve our purpose
here. Equation 3.90 requires a non-zero value for a ′ and Equation 3.91 has also been
derived using relationships between a and a ′ that require non-zero values of a'.

Table B.3

pitch, θ p
Section Section Section Angle of Section Chord, c

wind, ϕ
radius radius relative twist (m)
r/R (m) degrees (degrees)
(degrees)
Problems 27

0.05 0.60 46.13 53.13 49.32 4.02

0.15 1.80 16.96 23.96 20.15 2.04
0.25 3.00 7.93 14.93 11.12 1.30
0.35 4.20 3.78 10.78 6.97 0.94
0.45 5.40 1.43 8.43 4.61 0.74
0.55 6.60 -0.09 6.91 3.10 0.60
0.65 7.80 -1.14 5.86 2.04 0.51
0.75 9.00 -1.92 5.08 1.27 0.45
0.85 10.20 -2.52 4.48 0.67 0.39
0.95 11.40 -2.99 4.01 0.20 0.35

a) Starting with the definitions of the blade forces and the definition of C P :

∫ Ω dQ
R

C P = P Pwind =
ρπR 2U 3
rh
1
2

derive as simple an equation as you can for the power coefficient, C P , of an ideal Betz
limit rotor. The equation should include both lift and drag coefficients and tip speed ratio,
and should assume that a = 1/3. Ignore tip losses
b) Using the above equation find the rotor C P at the design tip speed ratio assuming
that there is no drag ( C d = 0). How does this compare with the Betz limit?
c) For a first approximation of the effect of drag on rotor performance, find the Cp for
the same rotor at the design tip speed ratio assuming the more realistic conditions that Cd/Cl
actually equals 0.02. Assume that the drag has no effect on the aerodynamics and that the
operating conditions assumed for the ideal rotor without drag apply. What effect does drag
have on the rotor C P , compared to the C P assuming that C d = 0?

SOLUTION

a) The power coefficient is defined as:

∫ Ω dQ
R

C P = P Pwind =
ρπR 2U 3
rh
1
2

where: dQ = B r dFT 1

or

dQ = B 1 2 ρU rel 2 (Cl sin ϕ − Cd cos ϕ ) c r dr

(3.71)

Inserting this into the definition for CP and simplifying:

28 Wind Energy Explained

∫U
⎛ ΩB ⎞
CP = ⎜ 2 3 ⎟ (C l sin ϕ − C d cos ϕ ) c r dr
R

⎝ πR U ⎠
2
rel
rh

Given that: U rel = U (1 − a ) sin ϕ (3.64)

and λ = ΩR U

(1 − a )2 (Cl − Cd cot ϕ ) c r dr
∫
⎛λB⎞
CP = ⎜ 3 ⎟
sin ϕ
R

⎝ πR ⎠ rh

(94 )⎛⎜ λ B3 ⎞⎟∫r (C l − C d cot ϕ ) c r dr

Because a is assumed to be 1/3:

CP =
R

⎝ πR ⎠ h

(94 )⎛⎜ λ B3 ⎞⎟∑ (Cl − Cd sin

cot θ i ) ci ri dr
This can be approximated by a summation over i blade sections:

CP =
⎝ πR ⎠ i
ϕ

b) Using the equation derived in part a) with Cd = 0:

Section radius Local tip Angle of Local C P

λr ϕ
r/R speed ratio rel. wind Contribution
dC P
0.05 0.5 53.13 0.006
0.15 1.5 23.96 0.018
0.25 2.5 14.93 0.030
0.35 3.5 10.78 0.041
0.45 4.5 8.43 0.053
0.55 5.5 6.91 0.065
0.65 6.5 5.86 0.077
0.75 7.5 5.08 0.089
0.85 8.5 4.48 0.101
0.95 9.5 4.01 0.113
Total C P : 0.593

Thus, the Cp for this rotor at the design tip speed, if one assumes no drag, is equal to the
Betz limit of 16/27 = 0.593.

b) Using the equation derived in part a) with Cd = 0.02:

Problems 29

Section radius Local tip Angle of Local C P

λr ϕ
r/R speed ratio rel. wind Contribution
dC P
0.05 0.5 53.13 0.006
0.15 1.5 23.96 0.017
0.25 2.5 14.93 0.027
0.35 3.5 10.78 0.037
0.45 4.5 8.43 0.046
0.55 5.5 6.91 0.054
0.65 6.5 5.86 0.062
0.75 7.5 5.08 0.069
0.85 8.5 4.48 0.075
0.95 9.5 4.01 0.081
Total C P : 0.474

For Cd = 0.02 the CP is 0.474, or 80% of the CP determined without drag. Thus, for this
simple approximation, 20% of the rotor power is dissipated in viscous drag with Cd/Cl =
0.02.

3.7 The Better Wind Turbine Company wants to start marketing wind turbines. The plans
call for a 20 meter in diameter, three-bladed, wind turbine. The rotor is to have its peak
power coefficient at a tip speed ratio of 6.5. The airfoil to be used has a lift coefficient of
1.0 and a minimum drag to lift ratio at an angle of attack of 7 degrees.
You, as the new blade designer, are to come up with two blade shapes as a starting
point for the blade design. One shape assumes that there are no losses and that there is no
wake rotation. The second design is based on the optimum rotor shape assuming that there
is wake rotation (but still no losses).
Find the chord length, pitch, and twist at 10 stations of the blade, assuming that the
blade extends right to the center of the rotor. How do the chord lengths and the twists
compare at the tip and at the inner three blade stations?

SOLUTION

a) For r/R = 0.05, 0.15, 0.25, ............., 0.95, the section radius, r, angle of relative wind, ϕ ,
chord, c, section pitch, θ p , and twist, θ T for the Betz limit rotor without wake rotation can
be determined from:
30 Wind Energy Explained

r = (r R )R
λr = (r R )λ
ϕ = tan −1 (2 (3λr ))
c = 8πr sin ϕ (3BCl λr )
θP = ϕ −α
θ T = θ P − θ P ,0

where θ P,0 is the blade pitch angle at the tip. The angle of the relative wind, the chord, and
the pitch can be determined for each blade location. The twist can only be determined once
the pitch is determined at the very end of the blade.

ϕ = (2 3 ) tan −1 (1 λ r )
For the rotor with wake rotation the same formulas apply except that:

8πr
c= (1 − cos ϕ )
BCl

The results of the calculations for the Betz rotor without wake rotation are:

pitch, θ p relative wind, ϕ twist, θ T

Section Section radius Chord Section Angle of Section
radius r c
r/R (m) (m) (degrees) (degrees) (degrees)
0.05 0.5 3.86 57.01 64.01 58.15
0.15 1.5 2.42 27.36 34.36 28.51
0.25 2.5 1.63 15.31 22.31 16.45
0.35 3.5 1.21 9.33 16.33 10.48
0.45 4.5 0.95 5.84 12.84 6.98
0.55 5.5 0.79 3.56 10.56 4.71
0.65 6.5 0.67 1.97 8.97 3.11
0.75 7.5 0.58 0.79 7.79 1.93
0.85 8.5 0.51 -0.12 6.88 1.02
0.95 9.5 0.46 -0.84 6.16 0.31
1 10 0.44 -1.14 5.86 0.00

The results of the calculations for the Betz rotor with wake rotation are:

pitch, θ p relative wind, ϕ twist, θ T

Section Section radius Chord Section Angle of Section
radius r c
r/R (m) (m) (degrees) (degrees) (degrees)
0.05 0.5 1.39 41.00 48.00 42.17
0.15 1.5 1.74 23.48 30.48 24.65
Problems 31

0.25 2.5 1.40 14.07 21.07 15.24

0.35 3.5 1.11 8.82 15.82 9.99
0.45 4.5 0.91 5.58 12.58 6.75
0.55 5.5 0.76 3.42 10.42 4.59
0.65 6.5 0.65 1.88 8.88 3.05
0.75 7.5 0.57 0.73 7.73 1.90
0.85 8.5 0.51 -0.16 6.84 1.01
0.95 9.5 0.46 -0.87 6.13 0.30
1 10 0.43 -1.17 5.83 0.00

At the tip the chord lengths and twist are almost the same. At the inner three blade
sections the twist and chord of the Betz limit blade increase rapidly as the radius decreases.
The blade design that assumes wake rotation has a maximum chord length at r/R = 0.15
that is much less than that of the other blade. It also has increasing twist at the radius
decreases, but the twist is less than the Betz limit blade.

3.8 The Better Wind Turbine Company wants to start marketing wind turbines. Their plans
call for a turbine that produces 100 kW in a 12 m/s wind at a cold site (-22.8°C, -9°F) with
an air density of 1.41 kg/m3. They have decided on a 20 meter in diameter, three-bladed,
wind turbine. The rotor is to have its peak power coefficient at a tip speed ratio of 7 in a 12
m/s wind. The airfoil to be used has a lift coefficient of 1.0 and a minimum drag to lift ratio
at an angle of attack of 7 degrees.
a) You, as the new blade designer, are to come up with the blade shape as a starting
point for the blade design. The design is to be based on the optimum rotor shape assuming
that there is wake rotation (but no drag or tip losses). Find the chord length, pitch, and twist
at 9 stations of the blade (each 1 m long), assuming that the hub occupies the inner tenth of
the rotor.
b) Determine the rotor CP assuming C d = 0. Again determine the power coefficient
assuming that the drag coefficient is 0.02, and that the aerodynamics are the same as the
condition without any drag. How much power is lost due to drag? Which part of the blade
produces the most power?
c) Does it look like the chosen design is adequate to provide the power that the Better
Wind Turbine Company wants?

SOLUTION

a) For r/R = 0.15, 0.25, ............., 0.95, the section radius, r, angle of relative wind, ϕ ,
chord, c, section pitch, θ p , and twist, θ T for the optimum rotor with wake rotation can be
determined from:
32 Wind Energy Explained

r = (r R )R
λr = (r R )λ
ϕ = (2 3 ) tan −1 (1 λr )

c=
8πr
(1 − cosϕ )
BCl
θP = ϕ −α
θ T = θ P − θ P ,0

where θ P ,0 is the blade pitch angle at the tip. The angle of the relative wind, the chord,
and the pitch can be determined for each blade location. The twist can only be determined
once the pitch is determined at the very end of the blade. The results of the calculations
appear below.

pitch, θ p relative twist, θ T

Section radius Section radius Chord Section Angle of Section

wind, ϕ (degrees)
r/R r c
(m) (m) (degrees)
(degrees)
0.15 1.5 1.34 22.07 29.07 23.65
0.25 2.5 1.58 12.83 19.83 14.41
0.35 3.5 1.24 7.80 14.80 9.38
0.45 4.5 0.97 4.74 11.74 6.32
0.55 5.5 0.79 2.71 9.71 4.29
0.65 6.5 0.66 1.26 8.26 2.84
0.75 7.5 0.57 0.19 7.19 1.77
0.85 8.5 0.49 -0.64 6.36 0.94
0.95 9.5 0.44 -1.30 5.70 0.28
1 10 0.39 -1.58 5.42 0.00

( )∫λλ sin
b) The power coefficient can be determined from:

C P = 8 λ2 2
ϕ (cos ϕ − λ r sin ϕ )(sin ϕ + λ r cos ϕ )[1 − (C d C l ) cot ϕ ]λ r 2 dλ r
h

where λ r = Ωr U = (r R )λ
Problems 33

The results for the two assumptions are tabulated below:

r/R CP CP
C d = 0.00 C d = 0.02
0.15 0.015 0.015
0.25 0.028 0.026
0.35 0.040 0.037
0.45 0.052 0.047
0.55 0.064 0.057
0.65 0.076 0.066
0.75 0.088 0.074
0.85 0.100 0.082
0.95 0.112 0.090
Total 0.576 0.493

When wake rotation is included, the power coefficient is slightly less than the Betz limit
when the drag is assumed to be zero. If the drag is assumed to be 0.02, the CP is only
85.6% of the value if the drag is zero. The contributions to the total CP from each section
indicate that the outer part of the blade produces the majority of the power.
c) The expected power can be calculated from:

P = CP 1
2
ρAU 3

At-22.8°C the density of dry air at standard atmospheric pressure is 1.41 kg/m3. The swept
area is 314.2 m3. Assuming that the turbine CP is 0.493 (including realistic drag), the rotor
power would be 188.7 kW. Even with tip losses, losses due to a more easily manufactured
non-ideal shape, and mechanical and electrical losses, the turbine would easily meet the
design criteria.

3.9 A two-bladed wind turbine is designed using one of the LS-1 family of airfoils. The 13
m long blades for the turbine have the following specifications (Table B.4).

Table B.4 LS-1 Airfoil Blade Geometry

r/R Section Radius chord twist

(m) (m) (degrees)
0.05 0.65 1.00 13.000
0.15 1.95 1.00 11.000
0.25 3.25 1.00 9.000
0.35 4.55 1.00 7.000
0.45 5.85 0.87 5.000
34 Wind Energy Explained

0.55 7.15 0.72 3.400

0.65 8.45 0.61 2.200
0.75 9.75 0.54 1.400
0.85 11.05 0.47 0.700

Note: θ p ,0 = -1.97 degrees (pitch at tip)

0.95 12.35 0.42 0.200

(note, α is in degrees):
Assume that the airfoil's aerodynamics characteristics can be approximated as follows

For α < 21 degrees:

C l = 0.42625 + 0.11628 α − 0.00063973 α 2 − 8.712 × 10 −5 α 3 − 4.2576 × 10 −6 α 4
For α > 21: C l = 0.95
C d = 0.011954 + 0.00019972 α + 0.00010332 α 2

ratio of 8: a) angle of attack, α ; b) angle of relative wind, θ ; c) C l and C d ; d) the local

For the midpoint of section 6 (r/R = 0.55) find the following for operation at a tip speed

contributions to C P . Ignore the effects of tip losses.

SOLUTION

a) The angle of attack ( α ) can be determined by equating the lift coefficient from the
measured C l for the blade with the lift coefficient required by strip theory. From
experiment:

For α < 21 degrees:

C l = 0.42625 + 0.11628 α − 0.00063973 α 2 − 8.712 × 10 −5 α 3 − 4.2576 × 10 −6 α 4
For α > 21: C l = 0.95

(cos ϕ i − λ r ,i sin ϕ i )
From strip theory, for each section:

σ i ' (sin ϕ i + λ r ,i cos ϕ i )

C l ,i = 4 sin ϕ i

The pitch, local solidity, and local tip speed ratio for the center of each section are
determined from:

θ P = θ T + θ P ,0

where θ P,0 = blade pitch angle at the tip.

σ ′ = Bc 2πr
Problems 35

λ r = λ (r R )

The simultaneous equations can be solved graphically, iteratively, or numerically. The

graphical solution is illustrated below.
2.0

1.8
1.6
1.4

1.2
1.0

0.8 Strip Theory

Airfoil Data
0.6

0.4
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0

A numerical solution for the lift coefficient gives:

α Strip Theory Lift Curve

1 0.968622 0.541799
1.5 1.11033 0.598915
2 1.23256 0.655486
2.5 1.33541 0.711424
3 1.41894 0.766635
3.5 1.48322 0.821019
4 1.52832 0.874469
4.5 1.5543 0.926871
5 1.56122 0.978106
5.5 1.54914 1.02805
6 1.51811 1.07656
6.5 1.46819 1.12352
7 1.39942 1.16876
7.5 1.31184 1.21214
8 1.20551 1.2535
8.5 1.08044 1.29268
9 0.936697 1.32951
9.5 0.774296 1.3638
10 0.593272 1.39538
10.5 0.393653 1.42406
36 Wind Energy Explained

11 0.175462 1.44963

b) From the blade geometry ϕ = α + θ P and θ P = θ T + θ P ,0 . Thus for section 6 the

By either method, the operating angle of attack is 7.84 degrees at a tip speed ratio of 8.

angle of the relative wind ( ϕ ) is 9.27 degrees.

c) C l and C d : From the equations for the lift and drag coefficients, C l = 1.24 and C d
= .020.
d) The local contribution to C P can be determined from:

⎛ 8Δλ ⎡ ⎛C ⎤
⎟⎟ sin ϕ i (cos ϕ i − λ ri sin ϕ i )(sin ϕ i + λ ri cos ϕ i )⎢1 − ⎜⎜ d
⎞ 2 ⎞
C P ,i = ⎜⎜ 2 r ⎟ cot ϕ i ⎥ λ ri 2
⎟
⎝ λ ⎠ ⎣⎢ ⎝ C l ⎠ ⎦⎥
Plugging in the angle of the relative wind and the local tip speed ratio, the local
contribution to C P is 0.057.

3.10 This problem is based on the blades used for the UMass wind machine WF-1. Refer
to Table B.5 for the blade geometry at specific locations along the blade. There is no airfoil

In addition, note that for the NACA 4415 airfoil: for α < 12 degrees: C l = 0.368 +
below r/R = 0.10

0.0942 α , C d = 0.00994 + 0.000259 α + 0.0001055 α 2 (note, α is in degrees). Radius:

mph). The pitch at the tip is:θP,0 = -2 degrees

4.953 m (16.25 ft); No. of blades: 3; Tip speed ratio: 7; Rated wind speed: 11.62 m/s (26

1/10). For the midpoint of each section find the following: i) angle of attack, α ; ii) angle of
a) Divide the blade into 10 sections (but assume that the hub occupies the innermost

relative wind, ϕ ; iii) C l and C d ; iv) the local contributions to C P and thrust. Include the
effects of tip losses.
b) Find the overall power coefficient. How much power would the blades produce at
11.62 m/s (26 mph)? Include drag and tip losses. Assume an air density of 1.23 kg/m3.

Table B.5 WF-1 Airfoil Geometry

r/R Radius Radius Chord Chord Twist

(ft) (m) (ft) (m) (degrees)
0.10 1.63 0.495 1.35 0.411 45.0
0.20 3.25 0.991 1.46 0.455 25.6
0.30 4.88 1.486 1.26 0.384 15.7
0.40 6.50 1.981 1.02 0.311 10.4
0.50 8.13 2.477 0.85 0.259 7.4
0.60 9.75 2.972 0.73 0.223 4.5
0.70 11.38 3.467 0.63 0.186 2.7
0.80 13.00 3.962 0.55 0.167 1.4
0.90 14.63 4.458 0.45 0.137 0.40
Problems 37

1.00 16.25 4.953 0.35 0.107 0.00

SOLUTION

blade radius (r), chord (c), twist ( θ T ), pitch ( θ p ), local solidity ( σ ′ ), and local tip speed
a) The physical measurements for the blade is given for each 1/10 of the blade radius. The

ratio ( λ r ) must all be determined for the center of each blade section to be used in the
calculation. Interpolation from the given data gives:
38 Wind Energy Explained

Section Section Section Section Section Section Section

r (m) θ T (deg) c (ft)

Radius, radius radius twist Chord Chord
r/R r (ft) c (m)
1 0.05 0.81 0.248 hub hub hub
2 0.15 2.44 0.743 35.30 1.405 0.428
3 0.25 4.06 1.239 20.65 1.360 0.415
4 0.35 5.69 1.734 13.05 1.140 0.347
5 0.45 7.31 2.229 8.90 0.935 0.285
6 0.55 8.94 2.724 5.95 0.790 0.241
7 0.65 10.56 3.219 3.60 0.680 0.207
8 0.75 12.19 3.715 2.05 0.590 0.180
9 0.85 13.81 4.210 0.90 0.500 0.152
10 0.95 15.44 4.705 0.20 0.400 0.122
Tip 1 16.25 4.953 0.00 0.350 0.107

The pitch, local solidity, and local tip speed ratio for the center of each section are
determined from:

θ P = θ T + θ P ,0

σ ′ = Bc 2πr

λ r = λ (r R )

The results of the calculations are:

Section Section Local Local tip Section

σ′ ratio, λ r θ p , deg
radius solidity speed pitch
r/R
1 0.05 hub hub hub
2 0.15 0.275 1.05 33.30
3 0.25 0.160 1.75 18.65
4 0.35 0.096 2.45 11.05
5 0.45 0.061 3.15 6.90
6 0.55 0.042 3.85 3.95
7 0.65 0.031 4.55 1.60
8 0.75 0.023 5.25 0.05
9 0.85 0.017 5.95 -1.10
10 0.95 0.012 6.65 -1.80
Tip 1 7.00 -2.00
Problems 39

The angle of attack ( α ) can be determined by equating lift coefficient from the measured
C l for the blade with the lift coefficient required by strip theory. From experiment:

C l = 0.368 + 0.0942 α

(cos ϕ i − λ r ,i sin ϕ i )
From strip theory:

σ i ' (sin ϕ i + λ r ,i cos ϕ i )

C l ,i = 4 Fi sin ϕ i

⎡ ⎛ ⎧ (B 2 )[1 − (ri R )]⎫ ⎞⎤

Fi = (2 /π ) cos −1 ⎢exp ⎜ − ⎨ ⎬ ⎟⎟⎥
⎝ ⎩ (ri R ) sin ϕ i ⎭ ⎠⎥⎦
⎜
where
⎢⎣

where ϕ = θ P +α
The simultaneous equations can be solved graphically, iteratively, or numerically. The
graphical solution is illustrated below. The curves for i =8, 9, and 10 cross the airfoil lift
curve in two places. The right-hand intersection points 1) provide a solution in which the
angles of attack vary smoothly along the airfoil, and 2) yield axial induction factors that are
less than 1/2, a requirement of momentum theory. A calculation shows that left hand
intersection points yield solutions with axial induction factors much greater than 1/2.

1.8

1.6

1.4

1.2
Lift coefficient

1.0

0.8

0.6

0.4
i=2 i=6
i=3 i=7
0.2 i=4 i=8
i=5 i=9
i=10
0.0
1 2 3 4 5 6 7 8 9 10 11
Angle of attack, degrees
40 Wind Energy Explained

A detailed look at the graph follows:

1.4

1.3

1.2

1.1
Lift coefficient

1.0

0.9

0.8
i=2 i=6
i=3 i=7
0.7 i=4 i=8
i=5 i=9
i=10
0.6
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0
Angle of attack, degrees

Numerical solution gives:

Section r/R α , deg Cl Cd F ϕ , deg

1 0.05 hub hub hub hub hub
2 0.15 4.99 0.838 0.0139 1.000 38.29
3 0.25 5.99 0.932 0.0153 1.000 24.64
4 0.35 6.67 0.997 0.0164 1.000 17.72
5 0.45 6.98 1.025 0.0169 1.000 13.88
6 0.55 7.34 1.059 0.0175 0.999 11.29
7 0.65 7.81 1.104 0.0184 0.995 9.41
8 0.75 8.04 1.125 0.0188 0.982 8.09
9 0.85 8.17 1.138 0.0191 0.926 7.07
10 0.95 7.63 1.087 0.0181 0.696 5.83

The local contribution to C P and thrust can be determined from:

⎛ 8Δλ ⎡ ⎛C ⎤
⎟⎟ sin ϕ i (cos ϕ i − λ ri sin ϕ i )(sin ϕ i + λ ri cos ϕ i )⎢1 − ⎜⎜ d
⎞ 2 ⎞
C P ,i = ⎜⎜ 2 r ⎟ cot ϕ i ⎥ λ ri 2
⎟
⎝ λ ⎠ ⎢⎣ ⎝ C l ⎠ ⎥⎦
Problems 41

and dFN = B 12 ρU rel 2 (C l cos ϕ + C d sin ϕ ) cΔr

U (1 − a )
U rel = =
sin ϕ (σ ′C l / 4 F ) cot ϕ + sin ϕ
where U

To determine the thrust an air density must be assumed. Assuming an air density of ρ =
1.23 kg/m3 (0.0765 lbm/ft3) the following values can be determined.

Section Local Cp Relative Relative Thrust Thrust Torque Torque

m/s ft/s N lbf Nm ft-lb
1 hub hub hub hub hub hub hub
2 0.0092 16.78 55.06 73.1 16.43 41.42 30.55
3 0.0211 23.34 76.56 175.3 39.4 95.27 70.27
4 0.0329 30.66 100.6 283.7 63.77 148.30 109.38
5 0.0432 38.33 125.76 380.7 85.59 194.90 143.75
6 0.0536 46.16 151.44 486.4 109.35 241.77 178.32
7 0.0639 54.07 177.41 602.1 135.35 288.20 212.57
8 0.0725 62.05 203.58 703.4 158.13 326.86 241.08
9 0.0769 70.06 229.85 769.9 173.07 346.88 255.85
10 0.0652 78.08 256.17 732.2 164.6 293.96 216.82
Total 0.4385 4206.8 945.69 1977.53 1458.6

b) The overall power coefficient can be determined by summing the individual

contributions from each blade section. The result, from the table above is 0.439. This
calculation uses the equation above, which includes drag and tip losses. The total power
produced by the blades is:

P = C P Pwind = C P 1
2
ρπR 2U 3

where U is the free stream wind velocity, 11.62 m/s (26 mph). The total power produced
by the blades is 32.51 kW.
c) The total thrust and torque at 11.62 m/s (26 mph) can also be determined by
summing the individual contributions from each blade section. The total thrust on three
blades is 4.21 kN and the total torque from the three blades is 1.98 kNm.

3.11 A two-bladed wind turbine is designed using one of the LS-1 family of airfoils. The
13 m long blades for the turbine have the specifications listed previously in Table B.4.

(note, α is in degrees):
Assume that the airfoil's aerodynamics characteristics can be approximated as follows

For α < 21 degrees:

42 Wind Energy Explained

C l = 0.42625 + 0.11628 α − 0.00063973 α 2 − 8.712 × 10 −5 α 3 − 4.2576 × 10 −6 α 4

For α > 21: C l = 0.95
C d = 0.011954 + 0.00019972 α + 0.00010332 α 2

for operation at a tip speed ratio of 8: a) angle of attack, α , with and without tip losses; b)
For the midpoint of the outermost section of the blade (r/R = 0.95) find the following

angle of relative wind, θ ; c) C l and C d ; d) the local contributions to C P ; e) the tip loss
factor, F, and the axial induction factor, a, with and without tip losses. How do tip losses
affect aerodynamic operation and the local contribution to C P at this outermost section?

SOLUTION

a) The angle of attack ( α ) can be determined by equating the lift coefficient from the
measured C l for the blade with the lift coefficient required by strip theory. From

For α < 21 degrees:

experiment:

C l = 0.42625 + 0.11628 α − 0.00063973 α 2 − 8.712 × 10 −5 α 3 − 4.2576 × 10 −6 α 4

For α > 21: C l = 0.95

(cos ϕ i − λ r ,i sin ϕ i )
From strip theory, for each section:

σ i ' (sin ϕ i + λ r ,i cos ϕ i )

C l ,i = 4 Fi sin ϕ i

⎡ ⎛ ⎧ (B 2 )[1 − (ri R )]⎫ ⎞⎤

Fi = (2 /π ) cos −1 ⎢exp ⎜ − ⎨ ⎬ ⎟⎟⎥
⎝ ⎩ (ri R ) sin ϕ i ⎭ ⎠⎥⎦
⎜
where
⎢⎣

and ϕ = θ P +α
The pitch, local solidity, and local tip speed ratio for the center of each section are
determined from:

θ P = θ T + θ P ,0

σ ′ = Bc 2πr

λ r = λ (r R )

The simultaneous equations can be solved graphically, iteratively, or numerically. The

graphical solution is illustrated below. It can be seen that the curve of the angle of attack
calculated from strip theory intersects the airfoil lift curve at two points. The axial
induction factor as a function of angle of attack is the same for each case and is also shown
Problems 43

on the graph below. Strip Theory requires that a <= 0.5. Thus the right intersection points
are the correct solutions.
2.0

1.6
Lift coefficient

1.2

0.8

0.4 Airfoil Lift Curve a

Strip Theory - no tip losses
Strip Theory - tip losses
0.0
2 3 4 5 6 7 8 9 10
Angle of attack, degrees
A numerical solution gives:

Angle of Airfoil Strip Theory Strip Theory Chord

α
Attack lift without with c
curve tip losses tip losses

2 0.655 0.189 0.189 0.991

2.5 0.711 0.559 0.553 0.922
3 0.767 0.871 0.823 0.818
3.5 0.821 1.126 1.000 0.709
4 0.874 1.325 1.105 0.610
4.5 0.927 1.467 1.152 0.525
5 0.978 1.553 1.154 0.454
5.5 1.028 1.582 1.117 0.396
6 1.077 1.555 1.048 0.348
6.5 1.124 1.472 0.951 0.308
7 1.169 1.333 0.827 0.275
7.5 1.212 1.137 0.680 0.247
8 1.254 0.886 0.512 0.223
8.5 1.293 0.579 0.324 0.202
9 1.330 0.216 0.117 0.184

By either method, without including tip losses, the operating angle of attack is 7.35
degrees at a tip speed ratio of 8. When tip losses are included in the calculation the angle

b) From the blade geometry ϕ = α + θ P and θ P = θ T + θ P ,0 . C l and C d can be

of attack is determined to be 5.89 degrees.

calculated from the equations for the lift and drag coefficients using the angle of attack.
The local contribution to Cp can be determined from:
44 Wind Energy Explained

⎛ 8Δλ ⎞ ⎡ ⎛C ⎤
C P ,i = ⎜⎜ 2 r ⎟⎟ sin 2 ϕ i (cos ϕ i − λ ri sin ϕ i )(sin ϕ i + λ ri cos ϕ i )⎢1 − ⎜⎜ d
⎞
⎟ cot ϕ i ⎥ λ ri 2
⎟
⎝ λ ⎠ ⎣⎢ ⎝ C l ⎠ ⎦⎥

⎡ ⎛ ⎧ (B 2 )[1 − (ri R )]⎫ ⎞⎤

Fi = (2 /π ) cos −1 ⎢exp ⎜ − ⎨ ⎬ ⎟⎟⎥
⎜ ( ) ϕ
where
⎢⎣ ⎝ ⎩ ri R sin i ⎭ ⎠⎥⎦

[ ]
And the axial induction factor, a, can be determined from:

a = 1 1 + 4 F sin 2 ϕ (σ ′C l cos ϕ )

The results of these calculations are tabulated below:

α ϕ Cl C d Local C P F a
(degrees) (degrees)
No Tip Losses 7.35 5.58 1.199 0.019 0.0898 1.00 0.52
Tip Losses 5.89 4.12 1.066 0.017 0.0548 0.68 0.45

The tip losses result in a reduced angle of attack and resulting reductions in lift
coefficient and angle of the relative wind. In the outermost blade section of this example
the contribution to the power produced by the rotor is reduced by 39% due to the effect of
tip losses.

3.12 A two-bladed wind turbine is operated at two different tip speed ratios. At 8.94 m/s
(20 mph) (tip speed ratio = 9) one of the blade sections has an angle of attack of 7.19
degrees. At 16.09 m/s (36 mph) (tip speed ratio = 5) the same 1.22 m (4 ft) section of the
blade is starting to stall, with an angle of attack of 20.96 degrees. Given the following
operating conditions and geometric data, determine the relative wind velocities, the lift and
drag forces, and the tangential and normal forces developed by the blade section at the two
different tip speed ratios. Determine, also, the relative contribution (the fraction of the
total) of the lift and drag forces to the tangential and normal forces developed by the blade
section at the two different tip speed ratios.
How do the relative velocities and the lift and drag forces compare? How do the
tangential and normal forces compare? How do the effects of lift and drag change between
the two operating conditions?
Operating Conditions are listed in Table B.6:
Table B.6 Operating conditions

λ α ϕ a
5 20.96 22.21 .070
9 7.19 8.44 .390

This particular blade section is 1.22 m (4 ft) long, has a chord length of 0.811 m (2.66

valid for α < 21 degrees, where α is in degrees:

ft), and has a center radius of 5.49 m (18 ft). The following lift and drag coefficients are
Problems 45

C l = 0.42625 + 0.11628 α − 0.00063973 α 2 − 8.712 × 10 −5 α 3 − 4.2576 × 10 −6 α 4

C d = 0.011954 + 0.00019972 α + 0.00010332 α 2

SOLUTION

The information given in the problem statement can be used to solve for the relative wind

U rel = U (1 − a ) sin ϕ
velocity at the blades using:

The results are:

λ U U rel
m/s m/s
5 16.09 39.61
9 8.94 37.13

The airfoil characteristics can be used to determine the lift and drag coefficients at the

C l = 0.42625 + 0.11628 α − 0.00063973 α 2 − 8.712 × 10 −5 α 3 − 4.2576 × 10 −6 α 4

prescribed angles of attack using:

C d = 0.011954 + 0.00019972 α + 0.00010332 α 2

The lift and drag curves are shown below.

1.6
1.4
1.2
Lift Coefficient

1.0
0.8
0.6
0.4
0.2
0.0
0 2 4 6 8 10 12 14 16 18 20
Angle of Attack, degrees
46 Wind Energy Explained

-3
60x10

50

Drag Coefficient 40

30

20

10

0
0 2 4 6 8 10 12 14 16 18 20
Angle of Attack, degrees

For lambda=5: Cl = 0.959 and Cd = 0.062.

For lambda=9: Cl =1.186 and Cd = 0.019.

To determine the forces an air density must be assumed. Assuming an air density of ρ =
1.23 kg/m3, the forces can be determined using:

dFl = C l 1
2 ρU rel 2 c dr

dFd = C d 1
2 ρU rel 2 c dr

dFN = dFl cos ϕ + dFd sin ϕ

dFT = dFl sin ϕ − dFd cos ϕ

The fraction of the normal force caused by lift is:

dFl cos ϕ
dFNl =
dFN

The fraction of the normal force caused by drag is:

dFd sin ϕ
dFNd =
dFN

The fraction of the tangential force caused by lift is:

dFl sin ϕ
dFTl =
dFT

The fraction of the tangential force caused by drag is:

Problems 47

−dFd cos ϕ
dFTd =
dFT

The results of these calculations appear below:

λ 5 9 units
U rel 39.61 37.13 m/s
Angle of Attack 20.96 7.19 degrees

Lift Force 915.55 994.93 N

Drag Force 59.19 15.94 N

Normal Force 870.00 986.49 N

Contribution from lift 0.97 1.00
Contribution from drag 0.03 0.00

Tangential Force 291.28 130.26 N

Contribution from lift 1.19 1.12
Contribution from drag -0.19 -0.12

In English units:

λ 5 9
U rel 129.95 121.82 ft/s
Angle of Attack 20.96 7.19 degrees

Lift Force 204.69 222.45 lbf

Drag Force 13.23 3.56 lbf

Normal Force 194.5 220.56 lbf

Contribution from lift 0.97 1.00
Contribution from drag 0.03 0.00

Tangential Force 65.12 29.13 lbf

Contribution from lift 1.19 1.12
Contribution from drag -0.19 -0.12
48 Wind Energy Explained

An inspection of the table shows that:

1. The relative velocities and lift forces for the two conditions are very similar.

2. The drag force is greater at the lower tip speed ratio (stalled) condition and drag
forces at both conditions are much less than the lift forces.

3. The normal forces are also very similar for the two conditions but drag contributes a
slightly greater proportion of the force in the stalled condition.

4. The tangential forces at the blade section are very different for the two operating
conditions. Due to the much greater angle of attack at the higher wind speed, the tangential
force is much greater than at the lower wind speed even though the lift coefficient is
somewhat less and the drag coefficient is greater at the higher wind speed. The drag
decreases the torque produced by the rotor and the relative negative contribution of the
drag to the tangential force is greater in the stalled condition. Due to the much higher
tangential force at the higher wind speed, the absolute contribution of both the lift and the
drag forces to the tangential force is greater at the higher wind speed.

3.13 Assume a VAWT operates with an axial unduction factor of 0.3 and with a tip speed
ratio of 12. Equations 3.147 and 3.149 can be used to determine the ratio of Urel to U and
the angle of attack. Equations 3.148 and 3.150 can be used estimate those same quantities
for high tip speed ratios. What is the maximum percentage deviation of the estimates of
these quanties from the exact values as the blade of the VAWT makes one full rotation? If
the blades are replaced and now the turbine operates at a tip speed ratio of 20 and the same
axial induction factor?

SOLUTION
At a tip speed ratio of 12, the error in the approximation of Urel/U is only 0.2% but the error
of the approximation of the axial induction factor is up to 5.8%. At tip speed ratios of 20,
these errors are reduced to less than 0.1% and less than or equal to 3.5%. The details of the
solution can be found in VAWT_High Tipspeed Approximations_Solution.xls.

3.14 Determine the operating axial induction factor of a vertical axis wind turbine with
straight blades using Equation 3.156. Guess an “input” axial induction factor and solve the
right hand side of the equation. Using the quadratic equation solve the left side for the
value of the axial induction factor that is less than 0.5. If the two do not agree, adjust the
input value until it does. The resulting axial induction factor is the solution. Assume the
following input conditions and that the lift coefficient as a function of the angle of attack is
the same as that for a flat plate (Equation 3.51):

Number of blades B 3
Radius R 6
Chord c 0.1
Tip speed ratio lambda 15
Problems 49

SOLUTION
The axial induction factor for these conditions is 0.295. The details of the solution are
contained in VAWT_Lambda_a_Solution.xls.

B.4 Chapter 4 Problems

4.1 A wind turbine rotor turning at 60 rpm is brought to a stop by a mechanical brake. The
rotor inertia is 13,558 kg m2.
a) What is the kinetic energy in the rotor before it is stopped? How much energy does
the brake absorb during the stop?
b) Suppose that all the energy is absorbed in a steel brake disc with a mass of 27 kg.
Ignoring losses, how much does the temperature of the steel brake disc rise during the
stop? Assume a specific heat for steel of 0.46 kJ/kg C.

SOLUTION

a) A rotating body contains kinetic energy given by:

E= JΩ 2
1
2
Here the angular speed = 6.283 rad/s. J = 13 558 kgm2 so the total energy that must be
absorbed by the brake is 267.6 kJ.

Therefore the temperature increase, ΔT , is:

b) The brake plate has a mass, m, of 27 kg. Steel has a specific heat, cp, of 0.46 kJ/kg-C.

ΔT =
Q
mc p

where Q is the energy absorbed by the brake. Therefore:

ΔT = (267.6 kJ) / [ (0.46)(27) ] = 21.5 C = 38.8 F

4.2 A wind turbine has a rotor with polar mass moment of inertia, J, of 4.2 x 106 kg m2. It
is operating in steady winds at a power of 1,500 kW and a rotational speed of 20 rpm.
Suddenly the connection to the electrical network is lost and the brakes fail to apply.
Assuming that there are no changes in the aerodynamic forces, how long does it take for
the operating speed to double?

SOLUTION

Find the rotational speed in rad/s:

50 Wind Energy Explained

Ω = N * π / 30 = 2.09 rad/s

Find the torque on the rotor

Q = P / Ω = (1,500 kNm / s ) / 2.09 rad / s = 716 kNm

Find angular acceleration, α

α = Q / J = 716 kNm / 4.2 E 6 kg m 2 = 0.171 rad/s2

The time to double speed is:

2Ω 0 − Ω 0
t= = = 12.3 sec
α
2.09
0.171

Note: in a real situation the aerodynamic torque would not remain constant over this speed
range, so this simple approach to a solution would not be sufficient.

4.3 A cantilevered 2 m long main shaft of a wind turbine holds a 1500 kg hub and rotor at
its end. At rated power the turbine develops 275 kW and rotates at 60 rpm. The shaft is a
0.15 m in diameter cylindrical steel shaft.
a) How much does the shaft bend down at its end as a result of the load of the rotor and
hub?
b) How much does the shaft twist when the turbine is operating at rated power? What is
the maximum shear stress in the shaft?

SOLUTION
a) The shaft can be treated as a beam with a point load at the end. The area moment of
inertia for a shaft is:

πd4
I=
64

Using d = 0.15 m, the area moment of inertia comes out to be I = 2.48 10-5 m4.
The maximum deflection of a beam with a load at the end is:

ymax =
WL3
3EI

Here L =2 m and the modulus of elasticity for steel is: E = 210 GPa.
The load on the shaft is W = mg = (1500 kg)(9.81 m/s2) = 14.715 kN. Using the
equation for the maximum beam deflection, the shaft bends:
Problems 51

ymax = 0.0075 m = 7.5 mm.

b) At rated load the torque, Q, is:

Q = (Power/Angular Speed) = 275 kW/ (6.283 rad/s) = 42.77 kNm.

For a solid shaft the polar moment of inertia, J, is given by:

πd4 π r4
J= =
32 2

d = 0.15 m so J = 4.97 * 10-5 m4

The angular deflection is:

ϕ=
QL
JG

If we estimate G from E and Poisson's ratio ( μ ):

G=
2(1 + μ )
E

Assuming μ = 0.3, E = 210 GPa, then G = 80.8 GPa. The length L = 2 m. Therefore the
deflection at the end of the shaft is 0.0213 radians = 1.22 degrees.
The shear stress can be determined from

ϕ Gr
τ=
L

Thus the maximum shear stress at the outside of the shaft is 64.5 MPa.

4.4 A wind turbine on a 24.38 m (80 ft) tower is subject to a thrust load of 26.69 kN (6000
lbf) during operation at 250 kW, the rated power of the turbine. In a 44.7 m/s (100 mph)
hurricane the thrust load on the stopped turbine is expected to be 71.62 kN (16,100 lbf).
a) If the tower is a steel tube 1.22 m (4 ft) in outer diameter (O.D.) with a 0.0254 m (1
in) thick wall, how much will the top of the tower move during rated operation and in the
hurricane force winds?
b) Suppose the tower were a three-legged lattice tower with the specifications given in
Figure B.2, how much will the top of the tower move during rated operation and in the
hurricane force winds? Ignore any effect of cross bracing on the tower.
52 Wind Energy Explained

1.83 m = 6 ft

1.83 m = 6 ft

Legs: Thrust
0.152 m = 6" OD
0.127 m = 5" ID
Steel Pipe

1.83 m = 6 ft

Figure B.2 Lattice tower cross-section; ID, inner diameter; OD, outer diameter
SOLUTION
a) The tower can be treated as a beam with a point load at the top. The area moment of

( )
inertia, I, for an annular ring is:

π d 04 − d i4
I=
64

Using 1.22 m for the O.D. ( d 0 ) and 1.17 m for the I.D. ( d i ), the area moment of
inertia comes out to be I = 0.0168 m4. Using 48 in. for the O.D. and 46 in. for the I.D., the
area moment of inertia is I = 40790 in4 = 1.97 ft4.
The maximum deflection of a beam with a load at the end is:

y max =
WL3
3EI

The modulus of elasticity for steel is usually assumed to be E = 210 GPa or E = 30 *106
psi = 4.32*109 lbf/ft2.
The tower length is L = 24.38 m = 80 ft. and the load at the top of the tower, W, is 26.69
kN = 6000 lbf. Using the equation for the maximum beam deflection, the tower moves:

y max = 0.0365 m = 0.120 ft = 1.44 in.

In a hurricane the tower moves:

Problems 53

y max = 0.098 m = 0.322 ft = 3.86 in.

b) Once again the tower can be treated as a beam with a point load at the top, but in this
case the tower is a composite of 3 pipes. First the centroid of the tower structure, the area
of the pipes, and the area moment of inertia of the pipe need to be determined. Then the
area moment of inertia of the composite tower needs to be determined using the parallel
axis theorem.

( )
The area moment of inertia for the pipe is the same as for the annular ring:

π d 04 − di4
I=
64

The area moment of inertia is I = 1.34 * 10-5 m4 = 32.9 in4 = 0.0016 ft4. The area of the
pipe cross section is 0.0055 m2 = 0.060 ft4.

∑y A
The centroid of the structure can be determined from:

∑A
y=
i i

where Aiis the area of each pipe and yi is the distance from a convenient axis. Using the
axes illustrated below, it can be determined that the upper pipe is 1.58 m (5.196 ft) above
the two lower ones and that the centroid is located 0.534 m (1.752 ft) above the two lower
ones. The centroid is located on the y axis due to symmetry.
54 Wind Energy Explained

1.83 m (6 ft)

Centroid
1.58 m
(5.196 ft)

0.534 m
(1.752 ft)

1.83 m (6 ft)

The area moment of inertia about an axis through the centroid of the tower can be
determined using:

IC = ∑I i + Ai y i 2

where:
I C = moment of inertia about the centroid of each pipe
y i = Distance between the centroid of each pipe and the centroid of the tower.
Ai = Cross section area of each pipe.

Thus:

I C = 1.085 ft 4 = 0.0092 m 4

The maximum deflection of a tower, treated as a beam with a load at the end is:

y max =
WL3
3EI C
Problems 55

The modulus of elasticity for steel is usually assumed to be E = 210 GPa or E = 30 *106
psi = 4.32*109 lbf/ft2.
Using the equation for the maximum beam deflection, the tower moves:

y max = 0.067 m = 0.218 ft under operation

y max = 0.179 m = 0.586 ft in a hurricane

4.5 The wind turbine rotor shown in Figure B.3 has a rotor rotation velocity, Ω , of 1 Hz
(60 rpm) and is yawing at an angular velocity, ω , of 10 degrees per second. The polar
mass moment of inertia of the rotor is 13,558 kg m2 (10,000 slug ft2). The rotor weighs
1459 kg (100 slugs) and is 3.05 m (10 feet) from the center of the bed plate bearing
support. Centered over the bed plate bearing support are the bearings holding the main
shaft. These bearings are 0.91 m (3 ft) apart. The directions of positive moments and
rotation are indicated in the figure.

a) What are the bearing loads when the turbine is not yawing?
b) What are the bearing loads when the turbine is yawing?

ω
Precession
(yaw)
velocity +

Bearings

Ω+
Rotor
rotation Rotor
velocity
Bedplate
+

3.05 m = 10 ft

Figure B.3
SOLUTION

a) The sum of the moments about the tower top and the sum of the forces must both be
equal to zero. If W is the weight of the rotor, F1 is the downward force of the left bearing
on the shaft and F2 is the upward force of the right bearing on the shaft then the sum of the
moments about the tower top requires that:

(W) (3.05 m) – F1(0.455 m) – F2(0.455 m) = 0

The balance of the forces requires:

56 Wind Energy Explained

W – F1+ F2 = 0

W = (1459 kg) (9.81 m/s2). Solving these two equations gives:

F1 = 55,128 N =12 393 lbf

F2 = 40,815 N = 9175 lbf

The resulting forces at the bearings are shown below.

40 815 N

14 313 N 55 128 N
b) The moment caused by the yaw rotation is:

M = ω × IΩ
r r r

Thus, the magnitude of M is:

M = ΩωI

The yaw rate, ω , is 10 degrees/s or 0.174 rad/s. The rotor angular velocity is, Ω , 2 π
rad/s and the rotor polar moment of inertia, I, is 13,558 kgm2. Thus, the moment caused by
the rotation rate is + 14,823 Nm = 10,932 ft-lbf, adding to the static moment caused by the
blade weight. The resulting gyroscopic force at the bearings is:

FG = 14,823 Nm / 0.91 m = 16 289 N = 3,662 lbf

Adding the two forces together, the resulting forces are:

F1 = 69,951 N =15,724 lbf

F2 = 55,638 N = 12,507 lbf

Problems 57

These are shown in the figure below.

55 638 N

16 289 N

14 313 N 69 951 N
4.6 A 3 bladed wind turbine has a rotor diameter of 38.1 m. The blade chord is constant
and equal to 1.0 m. Each blade has a non-rotating natural frequency of 1.67 Hz. In
operation, rotating at 50 rpm, each blade has a natural frequency of 1.92 Hz. The pitch
angle of the blade is 1.8 degrees. (The hub is sufficiently small that it can be ignored.) The
turbine is operating in a wind of 8 m/s. The airfoil lift curve slope is equal to 2π. Air
density is 1.225 kg/m3. If the blade has a mass of 898 kg, what (a) hinge–spring stiffness,
(b) moment of inertia and (c) offset would be used to model the blade dynamics with the
hinge spring blade model?

SOLUTION

2(Z − 1)
As described in the text, the non-dimensional hinge offset is given by:

e=
3 + 2(Z − 1)
where

ω R2 − ω NR
Z=
2

Ω 2

The mass moment of inertia of the hinged blade can be approximated by:

Ib = M b
R2
[1 − e]3
3
and the flapping spring constant is:

K β = ω NR
2
Ib

In this case the non-rotating natural frequency is 1.67 Hz (= 10.49 rad/s), the rotating
natural frequency is 1.92 Hz (= 12.06 rad/s), and the rotor speed is 50 rpm (= 5.236 rad/s).
58 Wind Energy Explained

Thus Z is 1.2902 and

e = 0.162.
(The MiniCodes confirm this)

Then the moment of inertia of the equivalent blade is

I b = 63,900 kg m2
(The Minicode say 63,903 kg m2)

The spring constant (stiffness) is:

K β = 7,007 kNm/rad.
(The MiniCodes say 7,001 kNm/rad)

4.7 Consider the wind turbine rotor used in Problem 4.6. Using the hinge spring blade
model, find the flapping angle during these conditions. How far is the tip of the blade from
the plane of rotation?

Hint: In solving this, there are a number of intermediate calculations that need to be made.
It may be helpful to confirm your results with the mini-codes. In any case, check the units
of all parameters carefully.

γA
SOLUTION
This is solved by the following: β 0 =
2K
It is first necessary to find γ, K and A.

Find γ:

γ = ρ C Lα c R 4 / I b = 15.85

Find K from:

K = 1+ ε + K β / I bΩ 2

Use this for IB:

I b = mB ( R 2 / 3)(1 − e )3 = 63,900 kg m 2 (from Problem 4.6)

3+e
ε= = 0.29 , using e = 0.164, from Problem 4.6
2(1 − e )

Then, using Kβ = 7,007 from Problem 4.6:

Problems 59

K = 1 + ε + K β / I b Ω 2 = 1 + 0.29 + = 5.29
(63,900)(12.47 )
7,007,000

( )
Find the axisymmetric flow term, A, using:

A = (Λ / 3) − θ p / 4 = 0.0097

Assume, λ = ΩR / U = (5.236)(19.04 ) / 18 = 12.46

First find non-dimensional inflow, assuming a = 1/3:

Λ = U (1 − a ) / Ω R = 0.0535

(Note that this definition is slightly different than Eggleston and Stoddard’s original, which
is used in the MiniCodes. The answers are almost the same, however.)

Therefore:

( )
A = (Λ / 3) − θ p / 4 =
0.535 (1.8)(π / 180)
3
−
4
= 0.0097

γA
Find flapping angle, β0, from β 0 =

γA (15.85)(0.0097 )
2K

β0 = = = 0.0145 rad = 0.83 deg

2K (2)(5.29)
(Compare: The MiniCodes say 0.79 deg. The difference is due to method used to calculate
non-dimensional inflow, Λ)

The deflection = sin(β)*radius = 0.277 m

The output from the MiniCodes is shown below.

60 Wind Energy Explained

4.8 Blade bending moments are being measured on a research wind turbine on a day with
mean hub height winds of 9.14 m/s (30 ft/s), but a wind shear results in winds of 12.19 m/s
(40 ft/s) at the top of the blade tip path and 6.09 m/s (20 ft/s) at the bottom of the blade tip

second in the + X ′ direction (see Figure 4.17 in the text). Meanwhile a wind direction
path. A gust of wind has started the wind turbine yawing at a steady rate of 0.1 radians per

starts with a flap hinge angle, β , of 0.05 radians and, at the moment that measurements of
change results in a crosswind of +0.61 m/s (2.0 ft/s). The 24.38 m (80 ft) diameter turbine

are being made, the rate of change of the flap hinge angle is 0.01 rad/s. The rotor is 3.05
m/s (10 ft) from the yaw axis on this fixed speed turbine that rotates at a speed of 1 Hz. The
very efficient turbine is operating with an axial induction factor of 1/3.
Ignoring tower shadow and transient effects, if these operating conditions were to
persist for one revolution of the rotor, what would the perpendicular and tangential wind as
a function of azimuth angle be half way out on the blades? What would the magnitude of
the contributions to the perpendicular wind be from yaw rate, shear, crosswind, and blade

changes? Assume that the blade pitch angle, θ p , is 0.05 radians (2.86 degrees).
flapping? How would the angle of attack vary at this part of the blade as the azimuth

SOLUTION

U P = U (1 − a) − r β& − (V0 β + q r ) sin (ψ ) − U (r / R )K vs cos(ψ )

The perpendicular and tangential components of the wind at the blade are:

( )
U T = Ω r − V 0 + q d yaw cos(ψ )

where:
U = Free stream wind velocity
a = Axial induction factor
Problems 61

V0 = Cross wind velocity (due to yaw error)

q = Yaw rate (constant)
K sh = Wind shear coefficient (linear)
d yaw = Distance from rotor plane to yaw axis

The component of the perpendicular wind velocity due to the mean wind and axial
induction factor is:

U PU = U (1 − a ) = (9.14 )⎜ ⎟ = 6.09 m/s = 20 ft/s

⎛2⎞
⎝3⎠

U Pβ& = − r β& = −6.1(0.01) = −0.061 m/s = −0.2 ft/s

The component of the perpendicular wind velocity due to the flapping motion is:

U P ,crs = −V0 β sin (ψ ) = −(0.61)(0.05)sin (ψ ) = −0.0305sin (ψ ) m/s = −0.1sin (ψ ) ft/s

The component of the perpendicular wind velocity due to the crosswind is:

U P , yaw = − q r sin (ψ ) = −(0.1)(6.1)sin (ψ ) = −0.61sin (ψ ) m/s = −2sin (ψ ) ft/s

The component of the perpendicular wind velocity due to the yaw rate is:

The component of the perpendicular wind velocity due to the wind shear is:

U P ,vs = −U (r / R ) K vs cos(ψ ) = −9.14⎜ ⎟⎜ ⎟cos(ψ ) = −1.52cos(ψ ) m/s = −5cos(ψ ) ft/s

⎛ 1 ⎞⎛ 1 ⎞
⎝ 2 ⎠⎝ 3 ⎠

U P = 6.04 − 0.64sin (ψ ) − 1.52cos(ψ ) m/s = 19.8 − 2.1sin (ψ ) − 5cos(ψ ) ft/s

Taking all of the terms into account, the perpendicular wind velocity is:

U T = Ω r = 2π (6.1) = 12.2 m/s = 125.7 ft/s

The component of the tangential wind velocity due to the rotor rotation is:

U T ,crs = −V 0 cos(ψ ) = −0.61 cos(ψ ) m/s = −2 cos(ψ ) ft/s

The component of the tangential wind velocity due to the crosswind is:

U T , yaw = − q d yaw cos(ψ ) = −(0.1)(3.05) cos(ψ ) = −0.31 cos(ψ ) ft/s = −1 cos(ψ ) ft/s
The component of the tangential wind velocity due to the yaw rate is:

U T = 38.3 − 0.915 cos(ψ ) m/s = 125.7 − 3 cos(ψ ) ft/s

Taking all of the terms into account, the tangential wind velocity is:

α = φ − θ p = tan −1 (U P / U T ) − θ p ≈ U P / U T − θ p
The angle of attack is:
62 Wind Energy Explained

6.04 − 0.64sin (ψ ) − 1.52cos(ψ )

or

α≈ − 0.05
38.3 − 0.915 cos(ψ )
Graphs of these terms appear below.

7.5

7.0
Velocity, m/s

6.5
Total perpindicular wind
6.0 r/R = 0.5

5.5

5.0

4.5

0 1 2 3 4 5 6
Azimuth, radians

1.5
Perpindicular wind
components
1.0 r/R = 0.5

0.5
Velocity, m/s

0.0

-0.5 Blade flapping

Cross wind
-1.0 Yaw rate
Vertical shear
-1.5
0 1 2 3 4 5 6
Azimuth, radians
Problems 63

39.0
Velocity, m/s

38.5

Total tangential wind

38.0 r/R = 0.5

37.5

0 1 2 3 4 5 6
Azimuth, radians

0.6
Tangential wind
0.4 components
r/R = 0.5
0.2
Velocity, m/s

0.0

-0.2
Cross wind
-0.4 Yaw rate

-0.6
0 1 2 3 4 5 6
Azimuth, radians

8
Angle of attack
Angle of attack, degrees

r/R = 0.5
7

0 1 2 3 4 5 6
Azimuth, radians
64 Wind Energy Explained

calculation of the flap angle, the only non-zero term is the steady state flap angle, β 0 .
4.9 When only blade rotation and the hinge–spring and offset are included in the

a) What is the derivative of the steady state flap angle with respect to wind velocity?
Assume that the axial induction factor is also a function of wind velocity.
b) Assuming that the Lock number is positive, what would the effect of a negative
value of da/dU be on the steady state flap angle?
c) Suppose a = 1/3 and da/dU = 0. In this case, what is the expression for the derivative
of the steady state flap angle with respect to wind velocity?
d) What is the derivative of the steady state flap angle with respect to blade pitch?

SOLUTION

The flap equations of motion, including only rotation and the spring and offset can be
expressed as:
⎡ ⎤
⎢K ⎡γ ⎤
0 ⎥ ⎡ β 0 ⎤ ⎢ A⎥
⎢ ⎥
γ ⎢ ⎥ ⎢2 ⎥
0
⎢0 K −1 ⎥ β 1c = 0
⎢ 8 ⎥⎢ ⎥ ⎢ ⎥
⎢0 γ ⎥ ⎢⎣ β 1s ⎥⎦ ⎢ 0 ⎥
⎢⎣ − K − 1⎥ ⎢⎣ ⎥⎦
8 ⎦
where

K = 1+ ε + K β / I bΩ 2
K = Flapping inertial natural frequency (includes rotation, offset, hinge spring),

(
A = Axisymmetric flow term, = (Λ / 3) − θ p / 4 )
Λ = Non-dimensional inflow, Λ = U (1 − a) / Ω R
γ = Lock number, γ = ρ C Lα c R 4 / I b

The solution is:

1 γ
β0 = A
K 2

The coning angle results from a balance between the aerodynamic moments on the one
hand and the centrifugal force and hinge spring moments opposing them.
a) The derivative of the steady state flap angle with respect to wind velocity is:

dβ 0 γ ⎛ dA ⎞ γ ⎛ (1 − a ) U ⎛ da ⎞⎞
= ⎜ ⎟= ⎜ − ⎜ ⎟ ⎟⎟
dU 2 K ⎝ dU ⎠ 2 K ⎜⎝ 3ΩR 3ΩR ⎝ dU ⎠⎠

b) If the Lock number is positive and da/dU is negative then, from the equation above,
the steady state coning angle increases as the wind speed increases.
c) If a=1/3 and da/dU=0 then the derivative of the steady state flap angle with respect to
wind velocity reduces to:
Problems 65

dβ 0 γ ⎛ (1 − a ) U ⎛ da ⎞⎞ 1 ⎛ γ ⎞
= ⎜⎜ − ⎜ ⎟ ⎟⎟ = ⎜ ⎟
dU 2K ⎝ 3ΩR 3ΩR ⎝ dU ⎠ ⎠ 9 ⎝ KΩR ⎠

d) The derivative of the steady state flap angle with respect to blade pitch is:

dβ 0 γ ⎞ γ ⎛ 1 ⎡U (1 − a ) ⎤ θ p ⎞
⎟=− γ
⎛ dA
= ⎜⎜ ⎟⎟ = ⎜
dθ P 2 K ⎜ 3 ⎢⎣ ΩR ⎥⎦ − 4 ⎟
⎝ dθ P ⎠ 2 K dθ P
d
⎝ ⎠ 8K

4.10 Just as the flap angle can be represented by a the sum of a constant term, a sine term,
and a cosine term, the lead-lag angle in the simplified dynamics model can be represented
by the sum of a constant term, a sine term, and a cosine term:

ζ ≈ ζ 0 + ζ 1c cos(ψ ) + ζ 1s sin (ψ )
The following matrix equation for the solution of the lead–lag motion can be derived by
substituting this solution into the lead–lag equations of motion (note that flap coupling

( )
terms have been omitted for simplicity):
⎡ γ ⎡ θPΛ ⎤⎤
⎢− ⎢ K vsU A4 − 2 V0 + q d ⎥ ⎥
⎡ 2B (K 2 − 1) 0 ⎤⎡ζ 0 ⎤ ⎢ 2 ⎣ ⎦⎥
γ
⎢ 0
⎢ 0 (K 2 − 1)⎥⎥ ⎢⎢ζ 1c ⎥⎥ = ⎢⎢ − 2 B − q A4
⎥
⎥
⎢⎣ K 2 0 ⎥⎦ ⎢⎣ζ 1s ⎥⎦ ⎢ γ ⎥
2
⎢ ΛA2 ⎥
B
⎢⎣ 2 ⎥⎦

where:
⎛ ωζ ⎞
K 2 = inertial natural frequency (includes offset, hinge–spring) = ε 2 + = ⎜⎜ ⎟
Kζ
2

⎟
IbΩ 2 ⎝ Ω ⎠
Λ θp
A2 = axisymmetric flow term (includes tip speed ratio and pitch angle) = +

θ
2 3
A4 = axisymmetric flow term (includes tip speed ratio and pitch angle) = 2 Λ − p
3 4
The other terms are all defined in the text.

a) Write the matrix equation for lead–lag motion, including the terms for the steady
mean wind and the hinge spring model, but assuming that gravity is zero and that other
aerodynamic forcing functions are zero.
b) Solve the equations and find the expression for the steady state lead–lag angle as a
function of the Lock number, the non-dimensional wind speed, the lead–lag natural
frequency, and the blade pitch.
c) Suppose the pitch were 5.7 degrees, the rotor speed were 50 rpm and the diameter
were 9.14 m (30 ft). By what factor would the steady state lead–lag angle increase if the
66 Wind Energy Explained

wind speed increased from 7.62 m/s (25 ft/s) to 15.24 m/s (50 ft/s)? Assume that the axial
induction factor decreases from 0.30 to 0.25 as the speed increases to 15.24 m/s (50 ft/s).

SOLUTION

a) The equation of motion, assuming zero yaw rate, shear, gravity, and crosswind, is:

⎡ 0 (K 2 − 1) 0 ⎤⎡ζ 0 ⎤ ⎢ 0 ⎥
⎡ ⎤
⎢ 0
⎢ (K 2 − 1)⎥ ⎢ζ 1c ⎥ = ⎢ 0 ⎥
⎥ ⎢ ⎥
⎢γ ⎥
0
⎢⎣ K 2 0 ⎥⎦ ⎢⎣ζ 1s ⎥⎦ ⎢ ΛA2 ⎥
⎣2 ⎦
0

b) The steady state lead lag angle can be found using Cramer's Rule:

(K 2 − 1)
(K 2 − 1)
0 0

γ
0 0
ΛA2
ζ0 =
0 0

(K 2 − 1)
2

(K 2 − 1)
0 0
0 0
K2 0 0

The determinant of the top matrix is:

(K 2 − 1) (K 2 − 1)
(K 2 − 1) = −(K 2 − 1) γ ΛA
0 0 0

γ
0 0
ΛA2
2 0
0 0 2
2

⎛γ ⎛γ
= −(K 2 − 1)⎢0 − (K 2 − 1)⎜ ΛA2 ⎟⎥ = (K 2 − 1)2 ⎜ ΛA2 ⎟
⎡ ⎞⎤ ⎞
⎣ ⎝2 ⎠⎦ ⎝2 ⎠

(K 2 − 1)
The determinant of the bottom matrix is:

(K 2 − 1)
(K 2 − 1) = −(K 2 − 1)
0 0
0
0 0
K2 0
K2 0 0

= −(K 2 − 1)[0 − (K 2 − 1)(K 2 )] = K 2 (K 2 − 1)2

Thus:
Problems 67

(K 2 − 1)2 ⎛⎜ γΛA2 ⎟
⎞
⎝2 ⎠ γΛA2
ζ0 = =
K 2 (K 2 − 1) 2 2K 2

or

γΛ ⎛⎜ Λ θ p ⎞⎟
ζ0 = +
2 K 2 ⎜⎝ 2 3 ⎟⎠

U − u i U (1 − a )
c) The non-dimensional inflow can be written as:

Λ= =
ΩR ΩR
Thus, the equation can be written as:

γ ⎛ U (1 − a ) ⎞⎛ 1 ⎛ U (1 − a ) ⎞ θ p ⎞ γ (1 − a ) ⎛ U 2 (1 − a ) Uθ p ⎞
ζ0 = ⎜ ⎟⎜ ⎜ ⎟− ⎟= ⎜ ⎟
2 K 2 ⎜⎝ Ω R ⎟⎠⎜⎝ 2 ⎜⎝ Ω R ⎟⎠ 3 ⎟ 2 K Ω R ⎜ 2Ω R − 3 ⎟
⎠ 2 ⎝ ⎠
The ratio of the angles at 15.24 m/s (50 ft/s), designated with a subscript of f, to the lead lag
angle at 7.62 m/s (25 ft/s), designated with a subscript of i, can be written, after some

(1 − a f )(3U f 2 (1 − a f )− U f (2Ω Rθ p ))
manipulation, as:

ζ o, f

(1 − a i )(3U i 2 (1 − a i ) − U i (2Ω Rθ p ))
=
ζ o ,i

The pitch is 5.7 degrees = 0.1 radians.

The radial velocity is 50 rpm = 5.24 rad/s.
The axial induction factor at 7.62 m/s (25 ft/s), ai , is 0.30.
The axial induction factor at 15.24 m/s (50 ft/s), a f , is 0.25.
The radius is 4.57 m = 15 feet.

(.75)(3(15.24 2 (.75))− 15.24(2(5.24)(4.57 )(0.1))) = 5.64

Substituting, one gets:
ζ o, f

(.7 )(3(7.62 2 (.7 ))− 7.62(2(5.24)(4.57 )(0.1)))

=
ζ o ,i

Thus, this two fold increase in wind speed increases the steady state lead lag angle 5.6
times.

4.11 Using the lead–lag equations described in Problem 4.10:

a) Write the matrix equation for lead–lag motion if all of the terms are ignored except
those that have to do with steady winds and vertical wind shear.
68 Wind Energy Explained

b) Solve the equations and find the expression for the lead–lag angle with steady winds
and vertical shear. What is the effect of vertical wind shear on each term of the lead–lag
angle in the absence of yaw rate, gravity, and crosswind?

SOLUTION
a) The equation for the lead lag motion, assuming zero yaw rate, gravity, and
crosswind, is:

⎡ γ ⎤
⎡ 0 (K 2 − 1) ⎤ ⎡ ζ 0 ⎤ ⎢− K vsU A4 ⎥
⎢ 0 (K 2 − 1)⎥⎥ ⎢⎢ζ 1c ⎥⎥ = ⎢⎢ ⎥
0
2
⎢ 0
γ
0
⎥
⎢⎣ K 2 0 ⎥⎦ ⎢⎣ζ 1s ⎥⎦ ⎢ ΛA2 ⎥
⎣⎢ 2 ⎦⎥
0

Constant Term, ζ0
b)

The steady state lead lag angle can be found using Cramer's Rule:
γ
− (K 2 − 1)
(K 2 − 1)
K vsU A4 0
2

γ
0 0
ΛA2
ζ0 =
0 0

(K 2 − 1)
2

(K 2 − 1)
0 0
0 0
K2 0 0

The determinant of the top matrix is:

γ
− (K 2 − 1) −
γ
(K 2 − 1)
(K 2 − 1) = −(K 2 − 1)
K vsU A4 0
2 K vsU A4
γ
2
γ ΛA2
0 0
ΛA2
0
0 0 2
2

⎛γ ⎛γ
= −(K 2 − 1)⎢0 − (K 2 − 1)⎜ ΛA2 ⎟⎥ = (K 2 − 1)2 ⎜ ΛA2 ⎟
⎡ ⎞⎤ ⎞
⎣ ⎝ 2 ⎠ ⎦ ⎝ 2 ⎠

(K 2 − 1)
The determinant of the bottom matrix is:

(K 2 − 1)
(K 2 − 1) = −(K 2 − 1)
0 0
0
0 0
K2 0
K2 0 0
Problems 69

= −(K 2 − 1)[0 − (K 2 − 1)(K 2 )] = K 2 (K 2 − 1)2

Thus:

(K 2 − 1)2 ⎛⎜ γ ΛA2 ⎟
⎞
⎝2 ⎠ γΛA2
ζ0 = =
K 2 (K 2 − 1)2 2K 2

or

γΛ ⎛⎜ 2 θp ⎞
ζ0 = Λ− ⎟
⎜
2K 2 ⎝ 3 4 ⎟⎠

This constant blade bending in the lead lag direction is the result of steady wind and
shows no effect of wind shear in the absence of yaw rate, gravity, and crosswind.

Cosine Term, ζ 1c
The coefficient of the cosine term can also be found using Cramer's Rule:
γ
−
(K 2 − 1)
0 K vsU A4 0
2

γ
0 0
ΛA2
ζ 1c =
K2 0

(K 2 − 1)
2

(K 2 − 1)
0 0
0 0
K2 0 0

The determinant of the top matrix is:

γ
− γ
−
(K 2 − 1) = −(K 2 − 1)
0 K vsU A4 0
2 0 K vsU A4
γ
2
γ ΛA2
0 0
ΛA2
K2
K2 0 2
2

⎡ ⎛ γ ⎛γ
= −(K 2 − 1)⎢0 − ⎜ − K vsU A4 ⎟ K 2 ⎥ = − K vs (K 2 − 1)⎜ K 2U A4 ⎟
⎞ ⎤ ⎞
⎣ ⎝ 2 ⎠ ⎦ ⎝2 ⎠

The determinant of the bottom matrix is the same as before:

70 Wind Energy Explained

(K 2 − 1)
(K 2 − 1) = K 2 (K 2 − 1)2
0 0
0 0
K2 0 0

Thus:

⎛γ
− K 2 (K 2 − 1)⎜ K vs U A4 ⎟ − γ K vs U A4
⎞
⎝ ⎠
ζ 1c = = 2
K 2 (K 2 − 1)2 (K 2 − 1)
2

or
γ
− ⎛2 θ ⎞
ζ 1c = ⎜ Λ− p ⎟
(K 2 − 1)
K vsU
⎜3 ⎟
2
⎝ 4 ⎠

Thus, the perturbation introduced on top of the steady state term, in the absence of yaw
rate, gravity, and crosswind, is negative when the cosine of the azimuth is positive (in the
lower half of the rotor disk) where there is less wind. The lead lag bending angle is
increased in the upper half of the rotor disk.

Sine Term, ζ 1s
The coefficient of the sine term can also be found using Cramer's Rule:
γ
0 (K 2 − 1) − K vsU A4
2
γ
0 0 0
K2 0 ΛA2
ζ 1s =
(K 2 − 1)
2

(K 2 − 1)
0 0
0 0
K2 0 0

The determinant of the top matrix is:

γ
(K 2 − 1) −
γ
(K 2 − 1) −
0 K vsU A4
= K2 =0
2
γ
K vsU A4
0 0 0
ΛA2
2
K2 0 0 0
2

So:
ζ 1s = 0
Problems 71

Thus, in the absence of yaw rate, gravity, and crosswind, wind shear only affects the
cosine term in the lead lag angle equation.

4.12 Consider the data file, [Link]. This data represents 10 minutes of blade bending
moment data taken from an operating wind turbine. The turbine has a nominal operating
speed of 60 rpm. The data was sampled at 40 Hz. Find the psd of the data. What are the
predominant frequencies? What do you suspect that these frequencies might be caused by?

SOLUTION

Using the MiniCodes, the psd the range of interest is shown below. The peak frequencies
are 1.07 Hz and 4.307 Hz. The spike at 1.07 is presumably due to rotation. The other one
may be due to a blade vibration mode.
psd

200
180
160
140
120
100
psd
80
60
40
20
0
-20 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10

4.13 Consider the following turbine. The tower is 50 m high, has diameter of 2 m at the
top and 3.36 m at the bottom. The mass of the nacelle and rotor is 20,400 kg. The wall
thickness is 0.01384 m. Assume that the density of steel is 7,700 kg/m3 and that its
modulus of elasticity is 160 GPa. The problem considers a number of approaches
(successively improving) to estimate the natural frequency (1st mode) of the turbine.

a) Assume that the tower is not tapered, but has a constant diameter of 2 m. What is the
natural frequency using the simple equation, but ignoring the mass of the rotor nacelle
assembly (RNA) on the top of the tower.

b) Repeat (a) using the Euler method. Do this directly or with the MiniCodes.

c) Repeat (a) using the Myklestad method. Use the MiniCodes.

d) Repeat (a), taking into account the weight of the RNA the top, using the simple
equation.
72 Wind Energy Explained

e) Repeat (d), using the Myklestad method. This will require some approximations to
account for the weight of the RNA.

f) Repeat (e), but take into account the actual taper of the tower.

SOLUTION

Parts of this problem can be solved with the help of the MiniCodes. An Excel spreadsheet
is also useful for solving the simpler parts and for preparing the input files for the
MiniCodes. See, for example, [Link].

a) f = 1 3(160 E 9)(0.0426)
= = 0.736 Hz
2π (0.23 mtower + mnacelle )L 2π ((0.23)(33,248))503
3EI 1
0 3

b) See MiniCodes screenshot: ω = 4.51rad / s , fo = 0.72 Hz

c) This is for straight tower with no weight, using the Myklestad method. The answer is
almost the same. Ten identical sections works well enough:

Input: tower straight_2.csv

Use:
Problems 73

0 0.086 0.043
0.1 0.086 0.043
0.2 0.086 0.043
0.3 0.086 0.043
0.4 0.086 0.043
0.5 0.086 0.043
0.6 0.086 0.043
0.7 0.086 0.043
0.8 0.086 0.043
0.9 0.086 0.043
1 0.086 0.043

d) f = 1 3(160 E 9 )(0.0426)
= = 0.384 Hz
2π (0.23 mtower + mnacelle )L π ((0.23)(33,248) + 20,400)503
3EI 1
0 3
2

e) We used 20 sections. The top one was heavier to include effect of weight. Note that
method uses averages, so we needed to introduce an extra section.
Use:
74 Wind Energy Explained

0 1.146 0.0426
0.05 1.146 0.0426
0.0501 0.0864 0.0426
0.1 0.0864 0.0426
0.15 0.0864 0.0426
0.2 0.0864 0.0426
0.25 0.0864 0.0426
0.3 0.0864 0.0426
0.35 0.0864 0.0426
0.4 0.0864 0.0426
0.45 0.0864 0.0426
0.5 0.0864 0.0426
0.55 0.0864 0.0426
0.6 0.0864 0.0426
0.65 0.0864 0.0426
0.7 0.0864 0.0426
0.75 0.0864 0.0426
0.8 0.0864 0.0426
0.85 0.0864 0.0426
0.9 0.0864 0.0426
0.95 0.0864 0.0426
1 0.0864 0.0426

f) This is whole thing, much wider at the base, much more steel. The RNA is only
approximated here; could have used average area of top 2 sections to find factor to include
effect of weight:

For inputs we used:

Problems 75

0 1.1461 0.0426
0.05 1.1461 0.0519
0.0501 0.0896 0.0519
0.1 0.0928 0.0612
0.15 0.0961 0.0704
0.2 0.0993 0.0797
0.25 0.1026 0.089
0.3 0.1058 0.0983
0.35 0.1091 0.1076
0.4 0.1123 0.1169
0.45 0.1156 0.1262
0.5 0.1188 0.1354
0.55 0.122 0.1447
0.6 0.1253 0.154
0.65 0.1285 0.1633
0.7 0.1318 0.1726
0.75 0.135 0.1819
0.8 0.1383 0.1912
0.85 0.1415 0.2004
0.9 0.1448 0.2097
0.95 0.148 0.219
1 0.1512 0.2283

Summary
Tower RNA Method Frequency, Hz
Straight No Simple 0.736
Straight No Euler 0.72
Straight No Myklestad 0.74
Straight Yes Simple 0.38
Straight Yes Myklestad 0.40
Tapered Yes Myklestad 0.78

Conclusion:
Adding weight to top reduces stiffness; making the tower wider at the base increases
stiffness.

B.5 Chapter 5 Problems

5.1 A 48 V DC wind turbine is hooked up to charge a battery bank consisting of four 12 V
batteries in series. The wind is not blowing, but the batteries must supply 2 loads. One is a
light bulb, rated at 175 W at 48 V; the other is a heater, rated at 1000 W at 48 V. The loads
are in parallel. The batteries may be considered to be constant 12 V voltage sources, with
an internal series resistance of 0.05 Ohm each. How much power is actually supplied to the
76 Wind Energy Explained

loads? How does that power compare to that which is expected? (Ignore the effect of
temperature on resistance of filament and heating elements.)

SOLUTION

Find the resistance of each of the loads, based on the nominal ratings.

Rlight = Vrated
2
/ Prated = 48 2 / 175 = 13.17 Ω

Rheater = Vrated
2
/ Pheater ,rated = 48 2 / 1000 = 2.304 Ω

The series resistance in the battery bank is:

Rbank = 4 x 0.05 = 0.2 Ω

The total resistance is:

R = Rbank + 1 /(1 / Rheater + 1 / Rlight ) = 2.161 Ω

The current is:

I = V/R = 48/2.161 = 22.21 A

The voltage across the battery bank is:

Vbank = E − Rbank I = 48 − (0.2 )(22.21) = 43.56V

where E = the internal voltage of the bank.

The actual power dissipated in the light is

Plight = Vbank
2
/ Rlight = 43.56 2 / 13.17 = 144 W

The actual power dissipated in the heater is:

Pheater = Vbank
2
/ Rheater = 43.56 2 / 2.304 = 824 W

The net effect is to reduce the power provided to the loads by nearly 18%

5.2 An electromagnet is used to hold a wind turbine's aerodynamics brakes in place during
operation. The magnet supplies 30 lbs (133.4 N) of force to do this. The magnet is supplied
by a 60 V DC source. The coil of the electromagnet draws 0.1 A. The core of the
Problems 77

electromagnet is assumed to have a relative permeability of 105. The diameter of the core is
3 in (7.62 cm).
Find the number of turns in the coil and the wire size. Assume that the relation between
force, magnetic flux, and area of the core is given by F = 397 840 B 2 Ac :
where: F= force (N), B = magnetic flux (Wb/m2), Ac = area of core (m2). Also, assume that

ρ = 1.72 × 10 −6 Ω cm .
the length of each turn is equal to the circumference of the core. The resistivity of copper is

SOLUTION

Find the area of core from the given diameter

π (0.0762 ) 2
Ac = = 0.00456 m 2
4

Find magnetic flux, B, from the relation given in the problem statement

B= = = 0.272Wb / m 2
F 134.4
(397,840) ( Ac ) (397,840)(.00456)

Find the magnetic field strength, H, from Equation 5.31

H= = = 68 A turns
μ0 μr
B 0.272
(10 )(4 x10 −7 )
4

The total number of turns is

Nt = = = 680 turns
H 68
I 0.1

Find the resistance

R= = = 600 Ω
V 60
I 0.1

Find length of each turn:

Lt = πD = π (0.0762) = 23.9 cm

The total length of wire is

L = Lt N t = 162.5 m
78 Wind Energy Explained

The cross sectional area is

ac = ρ = (1.72 x10 −6 Ω cm) = 0.0000466 cm 2

600 Ω
Lt 16252 cm
R

The diameter of the wire is

⎛a ⎞ ⎛ 0.0000466 cm 2 ⎞
d w = (2) ⎜ c ⎟ = (2) ⎜ ⎟ = 0.0077 cm = 0.003"
⎝ π ⎠ ⎜ π ⎟
⎝ ⎠

The wire size in this case is very close to 40 AWG. (For a table on wire gauge sizes, see,
for example, [Link] .)

ˆ = 10 + j 14 , Y
ˆ = −4 + j 5 . Find the following, and
ˆ +Y ˆ −Y
5.3 Consider the following phasors: X
express in both rectangular and polar form: X ˆ, X ˆ, X
ˆYˆ, Xˆ /Yˆ .

SOLUTION

Using the rules given in the chapter or the wind engineering mini-codes
ˆ = 10 + j 14 = 17.2∠54.5 deg
X

ˆ = −4 + j 5 = 6.40∠128.7 deg
Y

ˆ +Y
X ˆ = 6 + j19 = 19.9∠72.5 deg

ˆ −Y
X ˆ = 14 + j9 = 16.6∠32.7 deg

ˆ ⋅Y
X ˆ = -110 − j 6 = 110.2∠183.1deg

ˆ = 0.732 − j 2.585 = 2.69∠ − 74.2 deg

ˆ /Y
X

5.4 Show that the magnitude of the line-to-neutral voltage in a balanced, Y connected
three phase system ( VLN ) is equal to the line-to-line voltage ( VLL ) divided by the square
root of 3, i.e.

VLN = VLL 3.

The line feeding a Y-connected three-phase generator has a line-to-line voltage of 480
V. What is the line-to-neutral voltage?
Problems 79

SOLUTION

Consider three phases a, b, c with line to line voltages Vˆaa , Vˆbb , Vˆcc . The relation between
any two of the line to neutral voltages and the line to line voltage is given by the vector
sum: Vˆ =V ˆ −V ˆ . Suppose V ˆ =V ˆ ∠0 and V ˆ = V ˆ ∠120 . Then
ab an bn an an bn an

ˆ ∠0 − Vˆ ∠120
1∠0 − 1∠120
= = = 3
ˆ
V V
1∠0
an an

Van ∠0
ab
ˆ
Van ˆ

If the line to line voltage is 480 V, then the line to neutral voltage is
Vˆ = Vˆ / 3 = 480 / 3 = 277 V
ln ll

5.5 A circuit has a resistance and inductance in series. The applied AC voltage is
240∠0 o V, 60 Hz. The resistance is 8 Ω and the induction is 10 Ω. Find the current in the
circuit.

SOLUTION

Find the impedance: Z = R + jXL = 8+10j. The current is:

V 240∠0o
I= = = 15.6∠ − 51.3° A
Z 8 + 10 j

5.6 An AC circuit has a resistor, capacitor and inductor in series with a 120 V, 60 Hz
voltage source. The resistance of the resistor is 2 Ohms, the inductance of the inductor is
0.01 Henry; the capacitance of the capacitor is 0.0005 Farads. Find the following: reactance
of the capacitor and inductor, current, apparent power, real power, reactive power, power
factor angle and power factor.

SOLUTION

The reactance of the inductor is:

X L = 2 π f L = (2)(π )(60)(0.01) = 3.77 Ω

The reactance of the capacitor is:

XC = = = 5.30 Ω
2 π f C (2)(π )(60)(0.0005)
1 1
80 Wind Energy Explained

The total impedance is:

Z = R + j X L − j X C = 2 + j 3.77 − j 5.30 = 2.52 ∠ − 37.5 Ω

The current is

120∠0 V
I= = = 47.6∠37.5 A
Z 2.52 ∠ − 37.5 Ω
V

The apparent power is

S = I V = (47.6 A) (120V ) = 5712W

The power factor angle is

θ = ∠V − ∠I = 0 − 37.5 = −37.5

The real power is

P = I V cos(θ ) = (47.6) (120) cos(−37.5) = 4532 W

The reactive power is

Q = I V sin(θ ) = (47.6) (120) sin(−37.5) = −3478 VAr

The power factor is

pf = cos(θ ) = 0.793 , leading since the power factor angle < 0.

5.7 A circuit as a resistor and an inductor in parallel. In addition a second resistor is in

series with a capacitor in parallel and the two of them are in parallel with the inductor and
first resistor. The resistance of the first resistor is 8 Ω and that of the second resistor is 12
Ω. The impedance of the inductor is j16 Ω and that of the capacitor is –j22 Ω. The applied
voltage is 200 V. Find the apparent power, the power factor and the real power.

SOLUTION

First find the impedance

Z= =
1 / 8 + 1 / j16 + 1 / (12 − j 22) ) 0.125 − j 0.0625 + 0.0191 + j 0.035
1 1
Problems 81

Z= = 6.82∠10.8° Ω
1
0.1441 − j 0.0275

The current is

I= = = 29.33∠ = 10.8° A
V 200
Z 6.82∠10.8°

The apparent power is:

S = I V = (200 )(29.33) = 5866 W

The power factor is the cosine of the angle

pf = cos(10.8) = 0.982 , lagging

The real power is

P = (S )( pf ) = (5866 )(0.982 ) = 5762 W

5.8 A transformer rated at 120 V/480 V, 10 kVA has an equivalent circuit as shown in
Figure B.4. The low voltage side is connected to a heater, rated at 5 kW, 120 V. The high
voltage side is connected to a 480 V, 60 Hz, single-phase power line. Find the actual power
transferred, the magnitude of the measured voltage across the heater, and the efficiency
(power out/power in) of the transformer.
1.15 j 1.99
4 :1

480 V 3072 j 486 heater

E2 E1

Figure B.4 Transformer equivalent circuit

SOLUTION

Find resistance of the heater:

82 Wind Energy Explained

Rheater = V 2 / Pheater = 120 2 / 5000 = 2.88 Ω

Refer the heater's resistance to the high voltage side of the transformer:

⎛N ⎞ ⎛ 480 ⎞
= Rheater ⎜⎜ 1 ⎟⎟ = 2.88⎜ ⎟ = 46.08 Ω
2 2

⎝ N2 ⎠ ⎝ 120 ⎠
'
Rheater

The total parallel resistance is then:

R p = 1 /(1 / Rheater
'
+ 1 / 3072) = 1 /(1 / 46.08 + 1 / 2072) = 45.4 Ω

The total impedance is:

Z = 1.15 + j 1.99 + 1 /(1 / 45.4 + 1 / j 486) = 46.21 + j 6.209 Ω

The current is:

I = V / Z = 480 /(46.21 + j 6.209) = 10.295∠ − 7.653 A

The parallel impedance is

Z p = 1 /(1 / 45.4 + 1 / j 486) = 45.06 + j 4.219 Ω

The voltage across the load (referred to high side)

VL = I Z p = (10.295∠ − 7.653 A)(45.06 + j 4.219) = 465.92∠ − 2.304 = 465.55 − j 18.73V

The current in the heater is

I h = VL / R = (465.55 − j 18.73) / 46.08 = 10.111∠ − 2.304 A

The actual power in the heater is

Ph,actual = I h2 R = (10.1112 )(46.08) = 4710 W

The total power entering the transformer is

Ptotal = V I cos(−7.653) = ( 480)(10.295)∠ − 7.653 = 4898 W

The efficiency of the transformer is:

Problems 83

η = Ph,actual / Ptotal = 4710 / 4898 = 96.2%

The magnitude of the voltage across the heater, referred to the low side is

Vh,low = Vh / 4 = 116.4V

5.9 A small wind turbine generator (single phase) produces a 60 Hz voltage at 120 V rms.
The output of the generator is connected to a diode bridge full-wave rectifier, which
produces a fluctuating DC voltage. What is the average DC voltage? A silicon-controlled
rectifier (SCR) is substituted for the diode rectifier. Under one condition the SCRs are
turned on at 60 deg after the beginning of each half cycle. What is the average DC voltage
in that case?

SOLUTION

The maximum voltage is

Vmax = Vrms 2 = 169.7 V

The average DC voltage is:

π∫
π π
VDC = Vmax sin(θ ) = − V cos(θ ) = Vmax = 169.7 = 108 V
π π π
1 1 2 2
max
0 0

Using the SCR rectifier the average DC voltage is

π ∫π
π π
VDC = Vmax sin(θ ) = − Vmax cos(θ ) = Vmax = 169.7 = 81 V
π π π
1 1 1.5 1.5
/3 π /3

5.10 The parameters of an induction machine can be estimated from test data taken under a
few specified conditions. The two key tests are the no-load test and blocked-rotor test.
Under the no-load test, the rated voltage is applied to the machine and it is allowed to run at
no-load (i.e. with nothing connected to the shaft). In the blocked-rotor test the rotor is
prevented from turning and a reduced voltage is applied to the terminals of the machine. In
both cases the voltage, current and power are measured. . Under the no-load test, the slip is
essentially equal to 0, and the loop with the rotor parameters may be ignored. The
magnetizing reactance accounts for most of the impedance and can be found from the test
data. Under blocked-rotor conditions, slip is equal to 1.0 so the magnetizing reactance can
be ignored, and the impedance of the leakage parameters can be found. A third test can be
used to estimate the windage and friction losses. In this test the machine being tested is
driven by a second machine but it is not connected to the power system. The power of the
84 Wind Energy Explained

second machine is measured, and from that value is subtracted the latter's no-load power.
The difference is approximately equal to the test machine's windage and friction.
In a simplified version of the analysis, all of the leakage terms may be assumed to be on
the same side of the mutual inductance, as shown in Figure B.5. In addition, the resistance
in parallel with the magnetizing reactance is assumed to be infinite, the stator and rotor
resistances, R S and R′R , are assumed equal to each other and the leakage inductances are
also assumed to be equal to each other.
I X LS RS X'LR R'R

IM

IR (1- s )
R'R
V s
XM

Figure B.5 Simplified induction machine equivalent circuit

No-load test: V0 = 480 V, I 0 = 46 A, P0 = 3500 W

The following test data are available for a wye-connected, three-phase induction machine:

Blocked rotor test: V B = 70 V, I B = 109 A, PB = 5600 W

Windage and friction losses equal 3000 W.
Find the parameters for the induction generator model.

SOLUTION

The voltage for each phase is

V1,0 = = 277 V
480
3

The power per phase is:

P0 − Pwindage 3500 − 3000

P1,0 = = = 166 W
3 3

The power factor is

⎛ ⎞
φ = cos −1 ⎜⎜ ⎟ = 89.3 deg
⎟
P1,0
⎝ ( I 0 )(V1, 0 ) ⎠
Problems 85

The inductive reactance is the imaginary part of the impedance

277∠0
Zˆ = = = 0.074 − j 6.02 Ω
Vˆ
ˆI 46∠89.3

To a good approximation, then: X m = 6.02 Ω

The sum of the two leakage resistances can be found from:

2 rl = = = 0.157 Ω
P1φ , B 5600 / 3
I B2 109 2

The power factor under blocked rotor conditions is:

( )
⎛ P1φ , B ⎞ ⎛ ⎞
φ = cos −1 ⎜
( )
⎟ = cos −1⎜ 5600 / 3 ⎟ = 64.9 deg
⎜ V1φ , B (I B ) ⎟
⎝ 70 / 3 (109 ) ⎠
⎜ ⎟
⎝ ⎠

The sum of the resistances and sum of inductances may be found by phasor division of the

( )
voltage by the current.

70 / 3 ∠0
Zˆ l = = = 0.157 − j 0.336
Vˆ1φ , B
ˆI
B 109 ∠64.9

Therefore:

rls = rlr = 0.157 / 2 = 0.0785 Ω

xls = xlr = 0.336 / 2 = 0.168 Ω

parameters: X LS = X LR = 0.15 Ω , RS = 0.014 Ω, RR = 0.0136 Ω, X M = 5 Ω .

5.11 A four pole induction generator is rated at 300 kVA and 480 V. It has the following

How much power does it produce at a slip of -0.025? How fast is it turning at that time?
Also, find the torque, power factor and efficiency. (Ignore mechanical losses.)
Suppose the generator is used in a wind turbine, and the torque due to the wind is
increased to a value of 2100 Nm. What happens?

SOLUTION

The rotor speed of a 4 pole generator at slip of -0.025 is:

N = N sync − s N sync = 1800 − (−0.025)1800 = 1845 rpm

86 Wind Energy Explained

Voltage in each phase is Vln = Vll / 3 = 277.13∠0 V

Find the total impedance:

Z = rls + j xls + =
1
+
1 1
jxm rlr / s + j xlr

0.014 + j 0.145 + = −0.49311 + j 0.3492 Ω

1
+
1 1
j 5 0.0136 /(−0.025) + j 0.15

The current in each phase is:

277.13∠0
Ip = = = 458.64∠215.3 deg
Vln
Z − 0.49311 + j 0.3492

The voltage drop across the rotor is:

Vr = Vln − I p (rls + j xls ) = 277.13 − 458.64∠215.3(0.014 + j 0.15) = 242.89∠13.86 V

The current through the rotor is:

242.89∠13.86
I r = Vr / Z r = = 442.83∠209.27 A
1
0.0136 /(−0.025) + j 0.15

The power in (that which is converted to electricity) in all 3 phases is:

rlr (1 − s ) 0.0136(1 − (−0.025))

Pin = 3 I r2 = (3)(442.832 ) = 328.0 kW
s − 0.025

The power lost in the rotor resistance is:

Ploss , rotor = 3 I r2 rlr = (3)(442.832 )(0.0136) = 8.00 kW

The power loss in the stator resistance is:

Ploss , stator = 3 I 2 rls = (3)(458.6 2 )(0.014) = 8.83 kW

The power out of the terminals is:

Pout = Pin − Ploss, rotor − Ploss , stator = 328.3 − 8.00 − 8.83 = 311.2 kW
Problems 87

The power factor is:

cos −1 (θ ) = = = 0.816
Pout 311,200
3V I p (3)(277.13)(458.6)

The efficiency (not including windage and friction) is:

η= = = 0.949
Pout 311,200
Pin 328,000

The speed is:

Ω = (1.025)(1800 rpm )(1 / 60 s / min )(2π rad / rev ) = 193.2 rad / s

Torque is then:

Q = P / Ω = 328000 / 193.8 = 1698 Nm

The above calculations can be repeated for increasing values of slip to find the largest
value of torque the generator can supply. It is convenient to use the Wind Engineering
Mini-Codes for this. It can be seen that a torque of 2100 Nm would just exceed the
maximum possible torque (breakaway torque.) The rotor would thus begin to accelerate
without limit.

5.12 A resistive load of 500 kW at 480 V is supplied by an eight-pole synchronous

generator, connected to a diesel engine, and a wind turbine with the same induction
generator as that in Problem 5.11. The synchronous machine is rated to produce 1000 kVA.
Its synchronous reactance is 0.4 Ohms. A voltage regulator maintains a constant voltage of
480 V across the load and a governor on the diesel engine maintains a fixed speed. The
speed is such that the grid frequency is 60 Hz. The power system is three-phase, and the
electrical machines are both Y-connected. The power from the wind is such that the
induction generator is operating at a slip of -0.01. At this slip, the power 153.5 kW and the
power factor is 0.888.
a) Find the synchronous generator's speed, power factor, and power angle.
b) Confirm that the power and power factor are as stated.

SOLUTION

a) Since the grid frequency is fixed at 60 Hz, and the synchronous generator has 8 poles,
the speed is 900 rpm.
According to the problem statement, the electrical output power of the induction
generator is 153.5 kW and the power factor is 0.888.
88 Wind Energy Explained

The apparent power is therefore:

S = P / cos(θ ) = 153.5 / .888 = 173.2 kVA

The power factor angle of the induction machine is

θ = cos −1 (0.888) = 27.38 deg

The synchronous machine must supply the additional required real power and the
reactive power. Since the load is 500 kW, the synchronous machine supplies 500-153.5 =
346.5 kW.
The reactive power required is:

Q = S sin(θ ) = 173.2 sin(27.38) = 79.7 kVA , lagging

The apparent power from the synchronous machine is:

S SM = QSM
2
+ PSM
2
= 79.7 2 + 346.52 = 355.5 kVA

The power factor of the synchronous machine is:

pf SM = = = .975
PSM 346.5
S SM 355.5

The power factor angle for the synchronous machine is:

θ = cos −1 ( pf SM ) = cos −1 (0.957 ) = 12.8 deg , lagging

The per phase current is:

IP = =
( )
= 427.6∠ − 12.8 A
S SM 1 355,500 W
VP 3 480 / 3 V

The internal voltage, E, is:

E = V + I p Z SM = 277.13 + (427.6∠ − 12.8)( j 0.4) = 356.4∠27.9 deg

Since the power angle, δ , is, by definition, the angle between the field voltage and the
terminal voltage and since the voltage is assumed to have a reference angle of zero, the
power angle is immediately 27.9 deg.
Problems 89

b) Using a slip of –0.01 and solving the circuit for the induction generator as in problem
5.11, or using the MiniCodes, one can verify that the power and power factor are as stated
for this induction generator.

5.13 A diesel engine driven synchronous generator and a wind turbine with a wound rotor
induction generator (WRIG) are used to supply a resistive load as shown in Figure B.6.
The power system is three-phase, and the electrical machines are both Y-connected.

RS XS I XLS RS X’LR R’R/s

IM

IR RX/s
E VV
Rload
XM

Figure B.6 System with synchronous generator and WRIG

The WRIG has four poles and is rated at 350 kVA and 480 V. It has the following
parameters: X LS = X LR = 0.15 Ω , RS = 0.014 Ω, RR' = 0.0136 Ω, X M = 5 Ω . The external
resistor in the rotor circuit may be varied.

The resistive load consumes 800 kW at 480 V.

The synchronous generator is the same one as in Problem 5.12. It has eight-poles and is
rated to produce 1000 kVA. Its synchronous reactance is 0.4 Ohms. A voltage regulator
maintains a constant voltage of 480 V across the load and a governor on the diesel engine
maintains a fixed speed. The speed is such that the grid frequency is 60 Hz.

All mechanical losses may be ignored

The external resistance has been bypassed. The wind conditions are such that the WRIG is
operating at a slip of -0.02
a) How fast (rpm) is the generator turning at this time?
b) How much mechanical power is coming into the generator rotor from the wind
turbine’s rotor?
c) What is the power factor of the WRIG?
d) What is (i) the power from the synchronous generator; (ii) what is its power factor,
(iii) what is the power angle; (iv) what is the generator speed (rpm) and what is the internal
voltage, E?

SOLUTION

a) The speed is 1.02 x 1800 = 1836 rpm

90 Wind Energy Explained

b) The input power is 283.1 kW (see screen shot)

c) The power factor is 0.852, lagging (see screen shot)

d) What is (i) the power from the synchronous generator; (ii) what is its power factor, (iii)
what is the power angle; (iv) what is its speed (rpm) and what is the internal voltage, E?
The power from synchronous generator is remaining power required by load, = 800 - 271.4
= 528.6 kW, reactive power supplied = that required by the IG, = kVA sin(power factor
angle); kVa = Pout/pf = 271.4/0.852 = 318.5; Reactive power = 318.5 sin(31.57) = 166.7
kVAr; the required kVa is:
kVAreq = 528.6 2 + 166.7 2 = 554.3
The power factor angle is: θ Pf = cos −1 (528.6 / 554.3) = 17.5 deg
The power factor is: pf = kW/kVaapparent =528.6/554.3 = 0.954

With the MiniCodes:

Problems 91

Another way to do this is via impedances. One can show that the impedance of the
induction generator at a slip of -0.02 is -0.615982 + 0.378812j. The resistance of the load
to give 800 kW at 480 V is found from R=800,000/480 = 0.288 Ω. The overall impedance
is then:
Z= = = 0.3964 + .1252 j
1 1
+ +
1 1 1 1
Z load Z IG 0.288 − 0.615982 + 0.378812

The answer is almost the same!

5.14 This problem involves the same generator and load as in Problem 5.13, except now
the external resistance of the WRIG is included and it may be varied as required. The wind
speed has increased and the control system has responded such that the slip is now -0.42
but the stator currents are unchanged.
a. What is the value (Ω) of the external resistance?
b. What is the speed (rpm) of the generator?
c. How much mechanical power from the turbine’s rotor is now coming into the
generator’s rotor?
d. How much electrical power is coming out of the stator windings?
e. How much power is dissipated in the external resistance?

SOLUTION

a) Find k:
RR' + R X
= k so k = R = = −0.68
R' 0.0136
s s − .02

R X = ks − RR' = (− 0.68)(− 0.42 ) − 0.0136 = 0.272 Ω

Thus:

b) Speed = 1.42 x 1800 = 2556 rpm

c) The power in = 394.1 kW (see screen shot for slip = -0.42)

92 Wind Energy Explained

d) Power out of stator = 271.4 kW (same as total power above)

e) To find the stator loss and rotor loss, use values from slip = -0.02
Total loss = 283.1 – 271.4 = 11.7 kW
Stator loss = (3)(I2R)=(3)(383.22)(0.014)/(1000 W/kW)=6.17 kW
Rotor loss = Total loss- Stator loss= 11.7 – 6.17 = 5.53 kW
Total loss with Rx is 394.1-271.4 = 122.7 kW
Loss in external resistance = 122.7 – 11.7 = 111 kW

5.15 This problem involves the same situation as in Problem 5.14. Instead of an external
resistor in the WRIG’s rotor circuit, however, an ideal inverter has been inserted. The
inverter feeds the same power as was previously dissipated in the rotor resistance into the
local network.
a. How much total useful power is produced by the WRIG?
b. What is the power and power factor of the synchronous generator?

SOLUTION

a) See solution to Problem 5.14. The useful power = stator power + rotor power = 271.4 +
111 = 382.4 kW

b) The synchronous generator real power is 800-382.4=417.6 kW. The reactive power
requirement is the same as before, so:
kVAreq = 417.6 2 + 166.7 2 = 449.6
The power factor is given by: pf = kW/kVA= =417.6/449.6 = 0.929
See also the screen shot from the MiniCodes:
Problems 93

5.16 A six-pulse inverter has a staircase voltage, two cycles of which are shown in Figure
B.7. The staircase rises from 88.85 V to 177.7 V, etc. The frequency is 60 Hz.
200

100
Volts

-100

-200
0 10 20 30 40 50
Time, ms
Figure B.7 Inverter staircase voltage
a) Show that the voltage can be expressed as the following Fourier series:

⎡ ⎤
Vinv (t ) = A⎢sin( ω t ) + sin(5 ω t ) + sin(7 ω t ) + sin(11 ω t ) + ...⎥
1 1 1
⎣ 5 7 11 ⎦

where A = (3 π )(177.7) and ω = 2 π f

b) A series parallel resonance filter is connected across the terminals. The reactance of
each component is the same, so that:

(
⎩
)
f ( n) = n 2 ⎧⎨ n 2 − 1 − n 2 ⎫⎬ .
2

⎭
94 Wind Energy Explained

Show how the filter decreases the harmonics. Find the reduction to the filter of the
fundamental and at least the first three non-zero harmonics. Illustrate the filtered and
unfiltered voltage waveform.
Hint: in general, a Fourier series for a pulse train (a square wave) can be expressed as:

⎡ ⎤
V (t ) = H ⎢sin(ω t ) + sin(3 ω t ) + sin(5 ω t ) + sin(7 ω t ) + sin(9 ω t ) + ...⎥
π
4 1 1 1 1
⎣ 3 5 7 9 ⎦

where ω = 2πf and H = the height of the square wave above zero.

SOLUTION

a) The voltage can be considered to be the difference of two pulse trains, one with a
fundamental at 60 Hz, the other with a fundamental at 3 x 60 Hz = 180 Hz and a magnitude
one third of the first one. Call them V p ,1 ,V p , 2 ,Vinv .

Assuming a magnitude A = H and an angular frequency ω = 2 π f the first pulse train is

π
4

⎡ ⎤
V p ,1 (t ) = A⎢sin(ω t ) + sin(3ω t ) + sin(5ω t ) + sin(7 ω t ) + sin(9 ω t ) + ....⎥
1 1 1 1
⎣ 3 5 7 9 ⎦

The second pulse train is

1 ⎡ ⎤
V p , 2 (t ) = A⎢sin(3ω t ) + sin(9ω t ) + sin(15ω t ) + sin(21ω t ) + sin(27 ω t ) + ....⎥
1 1 1 1
3 ⎣ 3 5 7 9 ⎦

The difference is

Vinv (t ) = V p ,1 (t ) − V p , 2 (t )
⎡ ⎤
= A⎢sin(ω t ) + sin(3ω t ) + sin(5ω t ) + sin(7 ω t ) + sin(9ω t ) + ....⎥
1 1 1 1
⎣ 3 5 7 9 ⎦
1 ⎡ ⎤
− A⎢sin(3ω t ) + sin(9ω t ) + sin(15ω t ) + sin( 21ω t ) + sin( 27 ω t ) + ....⎥
1 1 1 1
3 ⎣ 3 5 7 9 ⎦
⎡ ⎛1 1⎞ ⎛1 1⎞ ⎤
= A⎢sin(ω t ) + ⎜ − ⎟ sin(3ω t ) + sin(5ω t ) + sin(7 ω t ) + ⎜ − ⎟ sin(9ω t ) + ....⎥
1 1
⎣ ⎝3 3⎠ 5 7 ⎝9 9⎠ ⎦

Simplifying yields:
Problems 95

⎡ ⎤
Vinv (t ) = A⎢sin( ω t ) + sin(5ω t ) + sin(7 ω t ) + sin(11 ω t ) + ....⎥
1 1 1
⎣ 5 7 11 ⎦

( )
b) As stated:

f (n) = n 2 / ⎧⎨ n 2 − 1 − n 2 ⎫⎬
⎩ ⎭
2

Therefore, for the non-zero terms:

f (1) = 1, f (5) = = 0.045, f (7) = = 0.021, f (11) = = 0.00847,....

25 49 121
576 − 25 2304 − 49 14400 − 121

A comparison of the filtered and non-filtered voltages is shown below.

B.6 Chapter 6 Problems

6.1 A wind turbine has a hollow steel shaft, 5 m long. The outer radius is 1 m and the
inner radius is 0.8 m. Find the area moment of inertia, I, and the polar mass moment of
inertia, J. The moments are to be taken about the long axis of the shaft. Assume that the
density of steel, ρs, is 7800 kg/m3.

SOLUTION

I=
(
π 14 − 0.8 4 ) = 0.9274 m4
2
96 Wind Energy Explained

( )
Find area density:

ρ area = (5 m ) 7800 kg / m 3 = 39,000 kg/m2

( )( )
J = ρ area I = 39,000 kg / m 2 0.9274 m 4 = 36,170 kg m2

6.2 The low speed shaft of a wind turbine is 10 m long and has a diameter of 0.5 m. It is
made of steel with a shear modulus of elasticity of G = 80 GPa. It is rotating at 12.1 rpm
and the turbine is generating 5 MW. (a) Find the applied rotor torque, assuming an overall
drive train efficiency of 90%. Find (b) the angle of deflection, (c) the energy stored in the
shaft and (d) the maximum stress in the shaft. Hint: The rotational stiffness of a shaft, k , is

given by kθ =
JG
l
SOLUTION

(a) The applied rotor torque is Q = = = 4.384 MNm / rad

Ωη
5,000,000

⎟(0.9)
P
⎛ 12.1 ⎞
2π ⎜
⎝ 60 ⎠
π (D )4 π (0.5)4
(b) The polar moment of inertia is J = =
= 0.00614 m4

JG (0.00614)(80 x 109 )
32 32
The rotational stiffness is kθ = = = 49.12 MNm/rad
(
The angular deflection is θ = Q / kθ = 4,384,000 / 49.12 x 10 = 0.08925 rad = )
l 10
6

5.11°

(c) The strain energy in the shaft is:

U= Qθ = (1 / 2 )(4.384 MNm / rad )(0.08925 rad ) = 195.6 kJ

1
2

(d) The maximum stress is σ max = Qc / J = (4,384,000 )(0.5 / 2 ) / 0.00614 = 178.5 MPa

6.3 For a helical coil spring, the spring constant, k, can be expressed in terms of the
spring's dimensions and its material properties:

k=
Gd 4
64 R 3 N c

where: G = modulus of elasticity in shear, N/m2 (lb/in2); d = diameter of wire, m (in); R =

radius of coil, m (in), and N c = number of coils. The average value of G for spring steel is
Problems 97

6
11.5x10 lb/in2 (7.93x1010 N/m2). The coil radius is based on the distance from the
longitudinal axis of the spring to the center of the wire making up the coil.
A spring is needed to return the aerodynamic brakes on a wind turbine blade to their
closed position after deployment. The spring should exert a force of 147 lb. (653.9 N)
when extended 6 inches (0.1524 m) (brake deployed) and a force of 25 lb. (111.2 N) when
the brake is closed. Space limitations dictate that the diameter of the spring be equal 1.5
inches (0.0381 m). A spring of wire gage 7 (0.177 inches or 4.5 mm diameter) is available.
How many coils would be required for this spring to have the desired force? How much
would the spring coil weigh? Assume a density for steel of 498 lb/ft3 (76 815 N/m3).

SOLUTION

The spring constant should be:

147 − 25
k= = 20.3 lb / in

653.9 − 111.2
6
= = 3561 N / m
0.1524

The radius of the coil is:

R = Dcoil / 2 − d wire / 2 = 1.5 / 2 − 0.177 / 2 = 0.6615 in

= 0.0381 / 2 − 0.0045 / 2 = 0.0168 m

The number of coils should be

Nc = = = 29.9 coils ≈ 30 coils
G d4 (11.5 x10 6 lb / in 2 )(0.177 in) 4
64 R k (64)(0.6625 in) 3 (20.3 lb / in)
3

= ≈ 30 coils
(7.93 x1010 N / m 2 )(0.0045) 4
(64)(0.0168 m) 3 (3561 N / m)

Using a weight density for steel, ρ s , of 489 lb/ft3 (76,815 N/m3) the weight of the coil is:

Wc = π ( d wire / 2) 2 ( 2)(π )( Rcoil )( N c )( ρ ) = π Rcoil N c ρ

1 2
2
= π (0.177 in) 2 (0.6615 in)(30)(489 lb / ft 3 ) /(12 in / ft )3 = 0.87 lb
1 2
2
= π 2 (0.0045 m) 2 (0.0168 m)(30)(76,815 N / m 3 ) = 3.87 N
1
2

6.4 The basic rating load for ball bearings, LR , is that load for which a bearing should
perform adequately for at least one million revolutions. A general equation relates the basic
98 Wind Energy Explained

rating load for ball bearings [with balls up to 1 inch (0.083 m diameter)] to a few bearing

L R = f c [n R cos(α )]0.7 Z 2 / 3 D 1.8

parameters:

m), depending on D cos(α ) / d m ; D = ball diameter, inches (m); α = angle of contact of

where f c = constant, between 3,500 and 4,500 (260,200 to 334,600 for ball diameter in

the balls (often 0 deg); d m = pitch diameter of ball races [= (bore + outside diameter)/2];
nR = number of rows of balls; Z = number of balls per row.
If two bearings are to operate for a different number of hours, the rating loads must be
adjusted according to the relation:

L2 / L1 = (N1 / N 2 )1 / 3

A bearing used on the output shaft of a wind turbine's gearbox has the following
characteristics: OD = 4.9213 inches (0.125 m), ID (bore) =2.7559 inches (0.07 m). The
bearing has 13 balls, each 11/16 inches (0.01746 m) in diameter. Find the basic rating load
of the bearing. Suppose that the gearbox is intended to run for 20 years, at 4000 hrs/yr.
How does the bearing load rating change? Assume that f c = 4500. The angle of contact of
the balls is equal to zero. The shaft is turning at 1800 rpm.

SOLUTION

The equation for basic load rating yields:

C = f c (n R cos(α ))0.7 Z 2 / 3 D1.8 = (4500 )(1)0.7 (13)2 / 3 (11 / 16 )1.8 = 12700 lb

= (334,600 )(1)0.7 (13)2 / 3 (0.01746 )1.8 = 56,400 N

If the gearbox is intended to operate for 20 years, 4000 hrs/year, the bearing will
experience (20 yr)(4000 hrs/yr)(1800 rpm)(60 rev/hr)= 8.64x109 revolutions. The relative

( )
loading is

L2 / L1 = (N1 / N 2 )1 / 3 = 106 / 8.64 x109 = 0.0847

1/ 3

Thus the load rating for the bearing should be

L2 = (12,700)(0.0487) = 625 lb
= (56,400)(0.0487 ) = 2750 N

6.5 A wind turbine is to be mounted on a three-legged truss tower 30 m high. The turbine
has a mass of 5000 kg. The rotor diameter is 25 m. The tower is uniformly tapered and has
a mass of 4000 kg. The tower legs are bolted in an equilateral pattern, 4 m apart, to a
Problems 99

concrete foundation. Each leg is held in place by six ¾ inch bolts with coarse threads. The
wind is blowing in line with one of the tower legs and perpendicular to a line through the
other two legs (which are downwind of the single leg). The wind speed is 15 m/s. Find the
torque in the bolts such that the upwind tower leg is just beginning to lift off the
foundation. Assume that the rotor is operating at the maximum theoretical power
coefficient. Hint: the torque in a bolt of this type can be approximately related to the load
in the bolt by the following equation:

Q B = 0.195 d W

where QB = bolt torque, Nm; d = nominal bolt diameter, m; W = load in bolt, N.

SOLUTION

The thrust load at the conditions given is:

T=
81
ρπR 2V 2 =
81
(1.22)(π )(12.5)2 (15)2 = 59,900 N
92 92

The tower and turbine weight on each leg is:

FT = (mtower + mturbine )(g ) / 3 = (5,000 + 4,000 )(9.81) / 3 = 29,400 N

Taking moments about the two down wind legs:

(FT + N B FB )Llegs − (T )(H ) = 0

(( ))
This implies that:

(T )( H ) / Llegs − FT (59,900 N )(30 m) / 4 3 / 2 − 29,400

FB = = = 81,600 N / bolt
NB 6

The torque is then:

QB=(0.195)(.75in)(.0254 m/in)(81,600 N)=303 Nm = 223 ft lb

6.6 An inventor has proposed a multi-blade wind turbine. The downwind rotor is to have a
diameter of 40 ft (12.2 m). Each blade is 17.6 ft (5.36 m) long, with a thickness of 3 inches
(7.62 cm) and width of 8 inches (20.3 cm). Assume that the blades have a rectangular
shape. The blades are to be made of wood, which can be assumed to have a modulus of
elasticity of 2.0 x106 lb/in2 (1.38x1010 Pa) and a weight density of 40 lb/ft3 (6280 N/m3).
The rotor turns at 120 rpm. Ignoring other potential problems, are any vibration problems
to be expected? (Ignore rotational stiffening.) Explain.
100 Wind Energy Explained
Problems 101

SOLUTION

Each blade will be excited as it passes behind the tower. The speed of ration is 120 rpm,
which corresponds to 12.57 rad/s. The blades are clearly flexible. Use the Euler method
(Chapter 4) or the Engineering Mini-Codes that may be found on the accompanying web
site for a cantilevered beam to see if the blade natural frequency is close to that of the
excitation. Finding the appropriate terms and converting to consistent units:

I= = = .000868 ft 4
b h 3 (8 / 12)(3 / 12) 3

(.203)(.0762)3
12 12

= = 7.48 x10 − 6 m 4
12

ρ ′ = ρ Ac = (40 lb / ft 3 )(8 / 12 ft )(3 / 12 ft ) /(32.2 lb / slug ) = 0.2071 slug / ft

( ) (
= 6280 N / m 3 (0.203 m )(0.0762 m ) / 9.81 m / s 2 = 9.9 kg / m )

(0.000868 ft )(2.0 x10 lb / in )(144 in ) = 12.5rad / s

The first mode natural frequency is

(β L )12
ω1 = =
4 6 2 2
/ ft 2
ρ ′ 17.6 2
E I 3.525

(7.48x10 )( )
2
L 0.02071 slug / ft
−6
= = 12.5 rad / s
3.525 m 4 1.38 x1010 N / m 2
5.36 2 9.9 kg / m

The blade natural frequency is very close to the excitation which the blades would
experience. This might be expected to be a problem.

6.7 A gearbox is to be selected for providing speedup to a generator on a new turbine. The
turbine is to have a rotor diameter of 40 ft (12.2 m) and to be rated at 75 kW. The generator
is to be connected to the electrical grid, and has a synchronous speed of 1800 rpm. A
parallel shaft gearbox is being considered. The gearbox has three shafts (an input,
intermediate, and output shaft) and four gears. The gears are to have a circular pitch of
0.7854 inches (1.995 cm). The characteristics of the gears are summarized in Table B.7.
Find the diameter, rotational speed and peripheral velocity of each gear. Find the tangential
forces between gears 1 and 2 and between gears 3 and 4.
102 Wind Energy Explained

Table B.7 Gear teeth

Gear Shaft Teeth

1 1 96
2 2 24
3 2 84
4 3 28

SOLUTION

The diameter of each gear is found from the circular pitch:

d = N p /π

where :
p = Circular pitch
N = Number of teeth
The relative rotational speeds, n, of each gear is given by the inverse of the ratio of the
diameters:

=
n1 d 2
n2 d1

The peripheral speed is found from.

V = (RPM )(π )(1 / 60 s / min )(d )

The results are summarized in the table below.

Gear Shaft Teeth d (in) d (m) n (RPM) V (ft/s) V (m/s)

1 1 96 24 0.61 150 15.7 4.79
2 2 24 6 0.152 600 15.7 4.79
3 2 84 21 0.533 600 55.0 16.8
4 3 28 7 0.178 1800 55.0 16.8

The tangential force is found from the power divided by the peripheral speed of the
gear:

Ft =
P
V

Converting to consistent units, the tangential force between Gears 3 and 4 is:
Problems 103

P (75 kW )(737.3 ( ft lb / s ) / kW )
Ft ,3 − 4 = = = 1006 lb

(75 kW )(1000 W / kW ) = 4464 N

v4 55 ft / s

=
16.8 m / s

The force between Gears 1 and 2 is

P (75 kW )(737.3 ( ft lb / s ) / kW )
Ft ,1− 2 = = = 3520 lb

(75 kW )(1000 W / kW ) = 15,700 N

v1 15.7 ft / s

=
4.79 m / s

6.8 Plate clutches are often used as brakes on wind turbines. According to "uniform wear
theory", the load carrying capability of each surface of a plate clutch can be described by

coefficient of friction, μ , times the average radius, rav , of the friction surface.
the following equation for torque. The torque, Q, is simply the normal force times the

ro + ri
Q=μ Fn = μ rav Fn
2

where: ro = outer diameter; ri = inner diameter; μ = coefficient of friction; Fn = normal

force.
A clutch type brake is intended to stop a wind turbine at a wind speed 50 mph (22.4
m/s). The brake is to be installed on the generator shaft of the turbine. The turbine has the
following characteristics: rotor speed = 60 rpm, generator power at 50 mph (22.4 m/s) =
350 kW, generator synchronous speed = 1800 rpm. The brake will consists of four
surfaces, each with an outer diameter of 20 inches (0.508 m) and an inner diameter of 18
inches (0.457 m). The coefficient of friction is 0.3. The brake is spring applied. It is
released when air at 10 lb/in2 (68.95 kPa) is supplied to a piston which counteracts the
springs. The effective diameter of the piston is 18 inches (0.457 m). What is the maximum
torque that the brake can apply? Should that be enough to stop the turbine if the generator
is disconnected from the grid? Ignore inefficiencies.

SOLUTION

The force that the springs apply must be equal to the force in the piston which can release
the brake. This is given by:

Fn = pairπR 2piston = (10 lb / in 2 )(π )(9 in) 2 = 2545 lb

= (68950 Pa )(π )(0.457 m / 2 )2 = 11,310 N

The maximum torque is:

104 Wind Energy Explained

ro + ri 20 / 2 in + 18 / 2 in
QB = N s μ Fn = (4)(0.3) (2545 lb) /(12 in / ft ) = 2418 ft lb

(4)(0.3)⎛⎜ 0.508 / 2 m + 0.457 / 2 m ⎞⎟(11,310 N ) = 3274 Nm

2 2

⎝ 2 ⎠

where Ns = number of surfaces.

The rotor torque at the speed of the generator is given by:

PRotor (350 kW )(44236 ( ft lb / min ) / kW )

QRotor = = = 1369 ft lb
Ω rotor (1800 rpm)(2 π rad / rev)
(350 kW )(1000 W / kW )
= = 1857 Nm
(1800 rpm)(2 π rad / rev)(1 / 60 s / min )

Based on the above calculations, the brake should be able to stop the turbine.

6.9 A cantilevered steel pipe tower, 80 ft (24.38 m) tall is being considered for a new wind
turbine. The weight of the turbine is 12 000 lb (53.4 kN). The pipe under consideration has
a constant outer diameter of 3.5 ft (1.067 m) and a wall thickness of 3/4" (1.905 cm). The
turbine is to have three blades. The nominal rotor is 45 rpm. Would the tower be
considered stiff, soft or soft–soft? Suppose that a tapered tower were being considered
instead. How would the natural frequency be analyzed, using the methods discussed in this
book? Assume that the density of steel is 489 lb/ft3 (76.8 kN/m3) and its elasticity is 30 x
106 lb/in2 (2.069 x 1011 Pa).

SOLUTION

Use the approximate equation suggested in this chapter to find the first mode natural
frequency:

f0 =
2π (0.23mTower + mTurbine )L3
1 3E I

( ) = π (3.5 )
The moment of inertia, I, is found from:

π D04 − Di4 − 3.375 4

I= = 0.997 ft 4
4

( ) = 0.00856 m
64 64
π 1.067 4 − 1.0289 4
= 4
64

The cross sectional area of the tower is:

Problems 105

Ac =
(
π D02 − Di2 ) = π (3.5 2
− 3.3752 )
= 0.6750 ft 2

( )
4 4
π 1.06682 − 1.0289 2
= = 0.0624 m 2
4

( )( )
The density of steel is 489 lb/ft3 (76.8 kN/m3). The weight of the tower is then:

Wtower = H tower Ac ρ steel = (80 ft ) 0.6750 ft 2 489 lb / ft 3 = 26,400 lb

( )(
= (24.38 m ) 0.0624 m 76.8 kN / m = 117.4 kN
2 3
)
The natural frequency is:

f0 =
2π (0.23 mtower + mturbine )H 3
1 3E I

=
(3)(30 x106 lb / in 2 )(144 in 2 / ft 2 )(0.997 ft 4 ) = 1.07 Hz
2π ((0.23)(26,400 lb ) + (12,000 lb ))(80 ft )3 (1 / 32.2 lb / slug )
1

(3)(2.069 x1011 N / m 2 )(0.00856 m 4 )

((0.23)(117.4 kN ) + (53.4 kN ))(24.38 m )3 (1000 N / kN )(1 / 9.81 m / s 2 )
= 1.07 Hz
2π
1

The rotor frequency is:

fR = = = 0.75 Hz
N rotor 45
60 60

The blade passing frequency is 3 times the rotor frequency, or 2.25 Hz. Because the
tower frequency is between the two it may be considered a soft tower.
For a tapered tower, the Myklestad method, developed in Chapter 4, could be used.

6.10 A wind turbine has a rotor with a polar mass moment of inertia, J, of 55,000 kg m2.
The equivalent stiffness of the drive train, k , is 45,000 kNm/rad. The turbine is operating
in high winds, such that the angular deflection, , in the drive train is 0.05 radians.
Suddenly a brake is applied at the far end of the drive train, bringing it to a stop. What is
the natural frequency of vibration (in Hz) of the rotor/drive train system as it “rings”
against the brake? Write the equation for the angular deflection as a function of time.
Ignore any damping.
106 Wind Energy Explained

SOLUTION

The natural frequency is found from:

ω = kθ / J = (45,000,000)(55,000) / 45 = 28.6 rad/s

Converting to Hz, the natural frequency is f = ω / 2π = 28.6 / 2π = 4.55 Hz

The equation for angular deflection is:

θ = θ 0 cos(ωt ) = (0.05)cos(28.6 t )

6.11 An operating turbine is having a problem. The emergency brake is applied, bringing
the rotor to a stop in one second. The turbine has a rotor diameter of 80 m. The tower
height is also 80 m. The mass of the rotor is 35,000 kg. The turbine is running at 20 rpm
before being stopped.
Estimate the overturning moment at the base of the tower due to the application of the
brake. Use the impulse-momentum theorem, and assume that all the mass is concentrated
at a distance equal to one half of the radius from the center of the rotor.

SOLUTION

Start from Newtons’ 2nd Law:

∑ F = ma = m dt
dV

Where F = force, m = mass, a = acceleration, and V = velocity.

Newton’s 2nd Law may be written in impulse momentum form as:

∫ ∑ Fdt = m(V 2 − V1 )

In the case of rotation, the above equation is:

∫ ∑ Fdt = mr (ω 2 − ω1 )

Where ω = angular speed of rotor. Assuming an average force of F over the one second
stopping period, and that the final speed (ω2) is zero, the relation is:

F = mrω1 / Δ t

Assuming that the mass is concentrated at ½ the radius, the average force would be:
Problems 107

F = mrω1 / Δ t = (35000 )(20 )(20)(2π / 60 ) = 1,467 kN

Assuming the force to act at hub height, the corresponding overturning moment would be:

M OT = F h = (1,467 kNm )(80 m ) = 117,300 kNm

B.7 Chapter 7 Problems

7.1 A turbine is being designed to operate in winds up to 15 m/s (at a hub height of 40 m).
The turbine is to have a rotor diameter of D = 45 m. Consider an extreme operating gust
which it might be expected to experience at least once every 50 years. What is wind speed
at the peak of the gust? For how long would the wind speed exceed that of the nominal
maximum operating wind speed? Graphically illustrate the hub height wind speed during
the gust. Assume a higher turbulence site.
The IEC 50-year gust for turbines with hub heights greater than 30 m is defined by:

⎧U − 0.37 U gust 50 sin (3π t / T )[1 − cos(2π t / T )], for 0 ≤ t ≤ T

U (t ) = ⎨
⎩U , for t < 0 and t > T

where T = 14 s and

⎛ σx ⎞
U gust 50 = 6.4⎜⎜ ⎟⎟ .
⎝ 1 + 0.1( D 21) ⎠

SOLUTION

From Equation 7.1, the standard deviation of the wind is found from:

σ x = I ref (0.75U hub + 5.6)

At a higher turbulence site, I 15 = 0.16 and σ x = 2.7

The peak value of the gust is then

⎛ ⎞
U gust 50 = 6.4⎜ ⎟ = 14.2 m / s
⎜ 1 + 0.1( 45 ) ⎟
2.7
⎝ 21 ⎠

The time series can be plotted as shown below and is determined by using the design gust
equation:
108 Wind Energy Explained

⎧U − 0.37U gust 50 sin (3π t / T )(1 − cos(2π t / T )) for 0 ≤ t ≤ T

U (t ) = ⎨
⎩U for t < 0 and t > T

30

25
Wind Speed, m/s

20

15

10

0
0 2 4 6 8 10 12 14
Time, seconds

From the analysis, the peak wind speed is 25.52 m/s.

The wind speed drops from nominal before it begins to rise to its peak. The point at
which the wind speed begins to rise is found from:

15 = 15 − 0.37 U gust 50 sin(3πt / T )(1 − cos(2πt / T )

In other words, the wind speed begins to rise when t is such that

sin(3πt / T )(1 − cos(2πt / T ) = 0

This occurs when t = T/3 = 4.67 sec. Since the gust is symmetric over the 14 second
interval, the length of time that the wind speed exceeds the nominal is 14-(2)(4.67) = 4.67
sec.

7.2 The maximum hourly average wind speeds (m/s) recorded at Logan Airport over a
sixteen year period are shown in the Table B.8. Using this data, find the 50 yr and 100 yr
expected maximum hourly wind speeds.
Problems 109

Table B.8 Annual maximum wind speeds

Year Maximum
1989 17.0
1990 16.5
1991 20.6
1992 22.6
1993 24.2
1994 21.1
1995 19.0
1996 20.6
1997 18.5
1998 17.0
1999 17.5
2000 17.5
2001 17.0
2002 16.0
2003 17.0
2004 16.5

SOLUTION

Use the Gumbel distribution (Equation 2.71):

⎛ ⎛ − (x − μ ) ⎞ ⎞
F (x )=exp⎜⎜ − exp⎜⎜ ⎟⎟ ⎟
⎟
⎝ ⎝ β ⎠⎠

First find the average and the standard deviation of the data.

U max = 18.6 m / s σ U max = 2.5 m / s

σ 6
The Gumbel parameters are: β = μ = U max − 0.577 β
π

Therefore: β = 1.916 μ = 17.541

We need to find the wind speed, Umax, which is exceeded once in 50 years or once in 100
yrs. That is, the wind such that 1-F(Umax) = 1/50 = 0.02 and 1-F(Umax) = 0.01. In other
words, F(Umax) = 0.98 or F(Umax) = 0.99. One can make a table in Excel (see Gumbel Dist
[Link]):
110 Wind Energy Explained

Umax F(Umax)
24 0.966
24.5 0.974
25 0.980
25.5 0.984
26 0.988
26.5 0.991
27 0.993
27.5 0.994
28 0.996
28.5 0.997
29 0.997
29.5 0.998
30 0.998

Thus the 50 year maximum wind is 25 m/s, and the 100 year wind is 26.5 m/s.
The figure below illustrates the data and the model.

1.2

0.8

Gumbel Fit
CDF

0.6
Data

0.4

0.2

0
0 5 10 15 20 25 30 35

Hourly Maximum Wind Speed, m/s

7.3 Show that maximum stresses due to flapping in a wind turbine blade are independent
of size for turbines of similar design. For simplicity, assume a rectangular blade and an
ideal rotor.
Problems 111

SOLUTION

Maximum stresses are found from the bending moments, the moment of inertia and the
maximum thickness of the blade. Without loss of generality, the blades may be considered
to be rectangular, with thickness h and width b. In the flapwise direction, the bending
moment at a given wind speed U is:

1 ⎛2R⎞ 1
M A, F = ⎜ ⎟CT ρπ R U
B⎝ 3 ⎠ 2
2 2

where
B= Number of blades
CT = Rotor thrust coefficient
The ratio of the stresses, σ AF = M c / I , due to the aerodynamic flapping moment in
blades of different lengths is:

⎛1 ⎛ 2R ⎞ 1 ⎞ ⎛1 ⎛ 2R ⎞ 1 ⎞
⎜ CT ⎜ 1 ⎟ ρπ R12U 2 ⎟ ⎜ CT ⎜ 2 ⎟ ρπ R22U 2 ⎟
σ AF ,1 ⎜ B ⎝ 3 ⎠2 ⎟ /⎜ B ⎝ 3 ⎠ 2 ⎟
=
σ AF , 2 ⎜ ⎟ ⎜ ⎟
⎜⎜ ⎟⎟ ⎜⎜ ⎟⎟
b1 h12 / 6 b2 h22 / 6
⎝ ⎠ ⎝ ⎠
⎛ R13 ⎞ ⎛ R23 ⎞
=⎜ ⎟ /⎜ ⎟ =1
⎜ b h2 ⎟ ⎜ b h2 ⎟
⎝ 1 1 ⎠ ⎝ 2 2⎠

Note that here c/I = (h/2)/(bh3/12)=1/(bh2/6)

7.4 Derive the relationship for blade bending stiffness (EI) as a function of rotor radius
(assuming that the tip speed ratio remains constant, the number of blades, airfoil, and blade
material remain the same at different rotor radii and that geometric similarity is maintained
to the extent possible).

SOLUTION

Stiffness is moment per unit deflection (EI). If the same materials are used E, a material
property, stays the same. For simplicity, consider the blade root to be approximated by a

moment of inertia about the flapping axis is I = c t 3 / 12 . If the radius is doubled, then the
rectangular cross-section of width c (corresponding to the chord) and thickness t. The

moment of inertia goes up by a factor 16. In fact for any constant cross section, if all
dimensions are scaled using R, the area moment of inertia increases by a factor of R4. Thus
the blade flap bending stiffness, EI, scales proportional to R4.
112 Wind Energy Explained

7.5 A wind turbine is being designed to supply power to a load which is not connected to a
conventional electrical grid. This could be a water pumping turbine, for example. This
problem concerns the estimate of the power curve for this wind turbine–load combination.
The power coefficient vs. tip speed ratio of many wind turbine rotors can be described
by the following simple third-order polynomial:

⎛ 3 C p ,max ⎞ 2 ⎛ 2 C p ,max ⎞ 3
Cp = ⎜ 2 ⎟λ −⎜ ⎟λ
⎜ λ ⎟ ⎜ λ3 ⎟
⎝ max ⎠ ⎝ max ⎠

where: C p , max = maximum power coefficient and λmax = the tip speed ratio corresponding
to the maximum power coefficient.

[i.e. C p (λ = 0) = 0) ], b) the slope of the power coefficient curve is zero at λ = 0 and c) the
Using the assumptions that a) the power coefficient = zero at a tip speed ratio of zero

slope of the power coefficient curve is also zero at tip speed ratio λmax , derive the above
relation.

speed, N L , which is related to the rotor speed by the gear ratio, g, so that N L = g N R .
The load the wind turbine is supplying is assumed to vary as the square of its rotational

Using the rotor speed as reference, the load power is:

PL = g 2 k N R2

where k is a constant.
A closed-form expression can be derived for the power from the turbine to the load, as
a function of rotor size, air density, wind speed, etc. Find that expression. Ignore the effect
of inefficiencies in the turbine or the load.

30 mph (13.4 m/s): Cp, max = 0.4, λmax = 7 , R = 5 ft (1.524 m), gear ratio = 2:1 speed up,
For the following turbine and load, find the power curve between 5 mph (2.24 m/s) and

rated load power = 3 kW, rated load speed = 1800 rpm.

SOLUTION

The general form for the polynomial is

C p = c1 + c2λ + c3λ2 + c4λ3

Since C p = 0 when λ = 0 then c1 = 0

The derivative of the polynomial is:

= c2 + 2 c3λ + 3 c4λ2
dC p
dλ

By the same logic as before, since dC p / dλ = 0 when λ = 0 then c2 = 0 and:

Problems 113

Then C p = c 3 λ 2 + c 4 λ3

= 2 c3 λ + 3 c 4 λ 2
dC p
dλ
and

Because C p (λ ) = C then C p , max = c3λ2max + c4 λ3max and because the slope at

λmax is zero: 0 = 2 c3λmax + 3 c4 λ max .

max p , max
2

With two equations in two unknowns, the remaining constants may be found directly:

c4 = − and c3 = −
3 C p , max
λ2max
2 C p , max
λ3max

Thus,

⎛ 3 C P ,max ⎞ 2 ⎛ 2 C P ,max ⎞ 3
Cp = ⎜ 2 ⎟λ − ⎜ ⎟λ
⎜ λ ⎟ ⎜ λ3 ⎟
⎝ max ⎠ ⎝ max ⎠

Rotor power is given by

PR = C p ρ π R 2V 3
1
2

Using the expression for Cp, and expressing the tip speed ratio in terms of the rotor
speed and wind velocity the rotor power is:

⎡⎛ 3 Cp, max ⎞⎛ N ( 2π / 60) R ⎞ 2 ⎛ 2 Cp, max ⎞⎛ N ( 2π / 60) R ⎞3 ⎤ 1

PR = ⎢⎜ ⎟⎜ R ⎟ − ⎜⎜ ⎟⎜ R ⎟ ⎥ ρπ R V
⎢⎣⎜⎝ λmax ⎟⎠⎝ ⎠ ⎝ λmax ⎟⎠⎝ ⎠ ⎥⎦ 2
2 3
2 3
V V

Ignoring efficiciencies, the rotor power must equal the load power:

⎡⎛ 3 Cp, max ⎞⎛ N (2π / 60) R ⎞ 2 ⎛ 2 Cp, max ⎞⎛ N ( 2π / 60) R ⎞ 3 ⎤ 1

g 2 k N R2 = ⎢⎜ ⎟⎜ R ⎟ − ⎜⎜ ⎟⎜ R ⎟ ⎥ ρπ R V
⎢⎣⎜⎝ λmax ⎟⎠⎝ ⎠ ⎝ λmax ⎟⎠⎝ ⎠ ⎥⎦ 2
2 3
2 3
V V

To simplify, it is convenient to define some constants:

⎛ Cp , max ⎞ 1
a1 = ⎜ 2 ⎟ ρ π 3 R4
⎜ λ ⎟
⎝ max ⎠ 600
114 Wind Energy Explained

⎛ Cp , max ⎞
a 2 = −⎜ 3 ⎟ ρ π ⎛⎜ 2 π ⎞⎟ R 5
3

⎜ λ ⎟ ⎝ 60 ⎠
⎝ max ⎠

[ ]
The equation equating rotor and load power is then:

g 2 k N R2 = a1V N R2 + a2 N R3

Solving this equation yields:

k g 2 − a1V
NR =
a2

Substituting the above equation for speed into the load power equation yields the
expression for the power as a function of wind speed:

⎡ g 2 k − a1V ⎤
P = g2 k⎢
2

⎥
⎣⎢ a2 ⎦⎥

For the specific example turbine and load, the equations above yield:

⎛ Cp , max ⎞ 1
a1 = ⎜ 2 ⎟ ρ π 3 R4
⎜ λ ⎟ 600
⎝ ⎠
)(0.002378 slug / ft 3 )(π 3 )(5 ft )4 = 0.00062698 lb s 2
max

= 2 (
0 .4 1

)(1.225 kg / m 3 )(π 3 )(1.524 m )4 = 0.002788 N s 2

7 600
= 2 (
0 .4 1
7 600

⎛ Cp , max ⎞
a 2 = −⎜ 3 ⎟ ρ π ⎛⎜ 2 π ⎞⎟ R 5
3

⎜ λ ⎟ ⎝ 60 ⎠
⎝ ⎠

( )
max

⎛ 2π ⎞
⎟ (5 ft ) = −0.00003127 lb s 2 ft
⎛ 0.4 ⎞
= −⎜ 3 ⎟ 0.002378 slug / ft 3 π ⎜
3

⎝7 ⎠ ⎝ ⎠
5

( )
60
⎛ 2π ⎞
⎟ (1.524 m ) = −0.00004237 N s 2 m
⎛ 0.4 ⎞
= −⎜ 3 ⎟ 1.225 kg / m 3 π ⎜
3

⎝7 ⎠ ⎝ 60 ⎠
5
Problems 115

( )
k = 3000 / 1800 2 = 0.000926 W / rpm 2
= (3000W / 1.3558W /( ft lb / s) ) 1800 2 ft lb / s rpm 2

The resulting power curve is shown below:

3500.00

3000.00

2500.00

2000.00
P, W

1500.00

1000.00

500.00

0.00
0.00 5.00 10.00 15.00 20.00
U, m/s

7.6 Assume that current 50 m blade designs have masses of 12,000 kg. Assume also that at
the design aerodynamic load, the blade tip deflects in the flap-wise direction 15 m from the
blade root axis and the resulting root bending moment is 16 MNm. A new blade testing
laboratory is to be designed that will handle blades up to 85 m long. In order to design the
new laboratory, the expected mass, tip deflection and root bending moment of 85 m blades
that will be designed in the future need to be estimated. Use the scaling relationships
derived in the text or derive additional scaling relationships to estimate these quantities.
Assume that the tip speed ratio remains constant, the number of blades, airfoil, and blade
material remain the same at different rotor radii and that geometric similarity is maintained
to the extent possible. Assume that the blade has a uniform cross section and a uniform
distributed aerodynamic load of q. Note that in that case, the tip deflection, v, with an be
determined from

v=
qR 4
EI
SOLUTION
116 Wind Energy Explained

Mass and Aerodynamic Bending Moment: From previously derived relationships, the
weight (or mass) and the aerodynamic root bending moment both scale proportional to R3.
So the expected blade mass is:

(12,000 kg)(85/50)3 = 59,000 kg.

The expected aerodynamic bending load at the root is:

(16 MNm)(85/50)3 = 78.6 MNm

Deflections: Since:

v=
qR 4
EI

The scaling of q, the distributed load along the blade (in N/m) and EI need to be
determined.

If the same materials are used E, a material property, stays the same. For simplicity,
consider the blade root to be approximated by a rectangular cross-section of width c

is I = c t 3 / 12 . If the radius is doubled, then the moment of inertia goes up by a factor 16.
(corresponding to the chord) and thickness t. The moment of inertia about the flapping axis

In fact for any constant cross section, if all dimensions are scaled using R, the area moment
of inertia increases by a factor of R4. Thus the blade flap bending stiffness, EI, scales
proportional to R4.

The uniform bending load is due to thrust, which, from the text, scales as R2.

Thus, the deflection scales as:

= R2
R2R4
R4

The expected deflection of an 85 m blade would be:

(15m)(85/50)2 = 43.35 m

7.7 The data file [Link] contains 10 minutes of 50 Hz teeter angle

data from a two-bladed 250 kW wind turbine with a down-wind rotor.
a. Determine the average, maximum, minimum and standard deviation of the data.
b. Graph the frequency distribution of the data (a histogram) in bins that are 0.5
degrees wide. If the hub includes dampers to damp teeter motion greater than 3.5
degrees, how many times in this 10-minute data set did the teetering hub contact
the teeter dampers?
Problems 117

c. Using the MiniCodes, generate and graph the power spectral density for the data
using a segment length of 8192 data points. What is the dominant frequency of
oscillation of the data?

SOLUTION

a. The statistics of the data (in degrees) are (see

[Link]):

Average -0.00259
Maximum 4.108
Minimum -4.021
Standard
Deviation 0.987
118 Wind Energy Explained

b. The teetered hub contacts the dampers a total of 58 times in this ten minute data
set. The frequency distribution and its graph are provided below.

Damper
BinCenter Count Percentage Contacts
-4.75 0 0.0 0
-4.25 2 0.0 2
-3.75 28 0.1 28
-3.25 84 0.3
-2.75 226 0.8
-2.25 472 1.6
-1.75 1097 3.7
-1.25 2173 7.2
-0.75 3955 13.2
-0.25 7127 23.8
0.25 6985 23.3
0.75 3774 12.6
1.25 2058 6.9
1.75 1163 3.9
2.25 529 1.8
2.75 197 0.7
3.25 103 0.3
3.75 24 0.1 24
4.25 4 0.0 4
30001 100 58

25
Percent of Occurences

20

15

10

0
-4.25 -3.75 -3.25 -2.75 -2.25 -1.75 -1.25 -0.75 -0.25 0.25 0.75 1.25 1.75 2.25 2.75 3.25 3.75 4.25

Teeter Angle, deg.

c. The dominant oscillation frequency is 1 Hz. The PSD is provided below.

Problems 119

1.00E+02
1.00E+01
1.00E+00
Teeter Angle PSD, (deg/Hz)^2

1.00E-010.001 0.01 0.1 1 10 100

1.00E-02
1.00E-03
1.00E-04
1.00E-05
1.00E-06
1.00E-07
1.00E-08
1.00E-09
1.00E-10
Frequency, Hz

7.8 The file [Link] includes data from full scale testing of a 250 kW two-
bladed constant-speed wind turbine with an induction generator. From kW and RPM data
in [Link] determine the rated slip (in %) of the generator:

(RPM at rated kW- Rated RPM)/(Rated RPM).

Rated RPM = 1800, Rated kW = 250 kW

SOLUTION

From a linear fit to the data, the generator speed at rated power, 250 kW is 1826.92 rpm.

From the formula in the problem solution the slip is 1.5% (see [Link])

From the slope of a linear fit to the data it can be seen that there is a slight offset in the
RPM transducer circuit.
120 Wind Energy Explained

120

100

80
Power, kW

60

40

20

1802 1804 1806 1808 1810 1812 1814

Generator RPM

The actual no-load generator speed is 1802 rpm. Using this as the rated slip, the slip is
1.38% at 250 kW.

B.8 Chapter 8 Problems

8.1 Wind power has been used for centuries for productive purposes. One early yaw
control system for windmills, invented by Meikle about 1750, used a fan tail and gear
system to turn the rotor into the wind. The windmill turret supported the rotor that faced
into the wind. The fan tail rotor, oriented at right angles to the power producing rotor, was
used to turn the whole turret and rotor (see Figure B.8). The gear ratio between the fan tail
rotor shaft rotation and the turret rotation was about 3000:1. Explain the operation of the
yaw orientation system, including the feedback path that oriented the rotor into the wind.

Fan tail rotor

Main rotor

Power shaft

Figure B.8 Meikle yaw control

Problems 121

SOLUTION

The torque from the rotor drove the mill and resulted in a net torque on the power shaft.
That torque would drive the turret, including the fan tail rotor out of the wind. As the yaw
error increased, the aerodynamic torque on the fan tail rotor would produce a counter
torque, to stabilize the yaw position. The system would operate with a yaw error that
depended on the design and wind speed. The yaw error sensor (the fan tail) produced a
torque that increased as the yaw error increased from the desired set point (no yaw error).
The 3000:1 gear reduction provided the force amplification to turn the turret.

8.2 A supervisory control system, as part of its tasks, is to monitor gearbox operation and
the need for gearbox maintenance or repairs. What information should be collected by the
supervisory controller and what information should be reported to the system operators?

SOLUTION

Numerous variations on input information and reporting might be acceptable, but important
input sensor information should include operating temperatures (bearing or lubricant
temperatures), gearbox vibration, lubricant level, and measures of lubricant circulation
system operation (flow, pressure, or power, etc.). Operators should be appraised of total
operating hours on the gearbox and lubrication systems, peak vibration and temperature
levels, low lubricant levels, and hours since the last regular maintenance work. Additional
information might include files of sensor data collected at frequent intervals, long term
temperature trends, and correlations between other operating measures and gearbox
operation (gearbox temperatures as a function of power, etc.).

8.3 A variable speed wind turbine control system uses blade pitch and speed variations to
provide constant power to the grid. What are the tradeoffs between fluctuations in rotor
speed and the time response of the pitch control system?

SOLUTION

In a variable speed wind turbine, the converter is set to provide constant power or torque as
wind speed fluctuations cause rotor speed changes. The blade pitch is used to smooth out
the rotor speed fluctuations. A slower time response of the pitch control system would
allow the rotor speed to increase or decrease more than with a faster pitch control system.
This has the advantage of possibly reducing pitch system loading and wear, but may result
in undesirable excursions in rotor speed. Faster pitch system response increases pitch
motor power requirements and loads in the pitch system and blades.

8.4 An accelerometer on an experimental wind turbine is used to measure tower vibration.

Measurements indicate that the sensor is very sensitive to temperature variations. To
complete a series of tests in cold weather the test engineers rig up a quick electrical heating
element and controller to keep the sensor's temperature constant. The system includes 1) a
small electronic chip that provides a millivolt output that varies with temperature, 2) an
electronic circuit, 3) a transistor, 4) a resistance heater, and 5) a housing to encase the chip,
122 Wind Energy Explained

the heater, and the accelerometer. The electronic circuit provides a voltage output related to
the difference between the voltages at the two inputs. A small current (milliamps) flows
into the transistor that is a function of the circuit output voltage. The transistor uses that
current to provide up to two amps of current to the heater. The heater maintains the
temperature in the enclosure at 70 F.

a) What is the process that is being influenced by the control system?

b) What elements play the roles of the sensor(s), the controller, the power amplifier, and
the actuator?
c) What if any are the disturbances in the system?

SOLUTION

a) The process that is being influenced by the control system is the energy balance between
the accelerometer and components inside the enclosure and the environment on the outside
of the enclosure. The introduction of the heating element and the controller adds a heat
source inside the enclosure, increasing the temperature inside the enclosure and increasing
the heat transfer from the enclosure.
b) The chip is the sensor that responds to temperature changes and the electronic circuit
is the controller that responds to differences between the enclosure temperature and the
desired enclosure temperature. The transistor is the power amplifier. It takes the small
current from the circuit and controls a much larger current that is capable of providing heat.
The actuator is the resistance heating element.
c) The heat transfer from the outside of the enclosure is a disturbance to the system.
This is mainly a function of ambient temperature, but also of wind speed and possibly the
presence of snow, ice, rain, or dirt.

8.5 An accelerometer on an experimental wind turbine is used to measure tower vibration.

Measurements indicate that the sensor is very sensitive to temperature variations. To
complete a series of tests in cold weather the test engineers rig up a quick electrical heating
element and controller to keep the sensor's temperature constant. The closed loop transfer
function of the system is:

= 2
T 0.1
Tref s + 0.5s + 0.1

Plot the step response of the system to a step increase of the reference temperature of
one degree F.
Hint: The step response is formed by multiplying the transfer function by Tref = 1/s.
The solution is determined by taking the inverse Laplace transform of the resulting
equation:

⎛1⎞
T= ⎜ ⎟
0.1
s 2 + 0.5s + 0.1 ⎝ s ⎠
Problems 123

First it must be converted into a sum of terms, using the method of partial fraction
expansion:

⎛1⎞ ⎛ A⎞ Bs + C
T= ⎜ ⎟ = ⎜ ⎟+
0.1
s 2 + 0.5s + 0.1 ⎝ s ⎠ ⎝ s ⎠ s 2 + 0.5s + 0.1
Then the inverse Laplace transform of each of the terms can be determined. The fraction
with the second order denominator may need to be broken into two terms to find the
inverse Laplace transform.

SOLUTION

The step response of the system is the response for:

Tref =
1
s
or

T (s ) = ⎜ 2
⎛ ⎞⎛ 1 ⎞
⎟⎜ ⎟
0.1
⎝ s + 0.5s + 0.1 ⎠⎝ s ⎠
The inverse Laplace transform of this transfer function can be found using a partial fraction
expansion:

⎛ ⎞⎛ 1 ⎞ A Bs + C
T ( s) = ⎜ 2 ⎟⎜ ⎟ = + 2
0.1
⎝ s + 0.5s + 0.1 ⎠⎝ s ⎠ s s + 0.5s + 0.1
Multiplying through and collecting terms, one finds that A = 1, B = -1, and C = -0.5. A few
manipulations result in:
s + 0.5 s + 0.25
T ( s) = − = − + 1.29
s s 2 + 0.5s + 0.1 s (s + 0.25)2 + 0.194 2 (s + 0.25)2 + 0.1942
1 1 0.194

The inverse Laplace transfer of this function is:

T (t ) = 1 − e ( −t / 4) cos(0.194t ) + 1.29e ( −t / 4) sin(0.194t )

The graph looks like:

124 Wind Energy Explained

1.2

1.0

Temperature, degrees F
0.8
Temperature controller step response
0.6

0.4

0.2

0.0
0 5 10 15 20 25 30
Time, seconds

8.6 The transfer function for a pitch control system is:

Θ m (s) ⎛ ⎞
=⎜ ⎟
Θ m, ref ( s) ⎝ s 3 + s 2 + s + K ⎠
K

The closed-loop system response to a step command to pitch the blades 1 degree is:

⎛ ⎞⎛1⎞

( )
Θ m ( s) = ⎜ ⎟⎜ ⎟ =
K K
⎝ s +s +s+ K ⎠⎝ s⎠
( s + a) s 2 + bs + c
3 2
s

a) Calculate and plot the closed-loop time domain system step response for K = 0.5, the
initial choice of the designer. Comment on the closed-loop system response, including
damping, overshoot, and response time.
b) For no good reason, the control system designer decides to increase the gain of the
system, K. Calculate the closed-loop time domain system step response for K = 3, the new
choice of the designer.
c) What differences are evident between the time responses, each with a different gain?

Hint: perform a partial fraction expansion of the general form of the closed-loop
response:

( )
d
( s + a ) s + bs + c s
2

and find the inverse Laplace transform of each term, using the a, b, c, d variables. This
symbolic form will be handy, as it will be used twice in the solution.

s 3 + s 2 + s + d , using the appropriate K. The values of s at the zero crossing is –a. The
For parts a) and b), find the real root, a, of the denominator by graphing

( )
second order root is found using long division:

= s 2 + (1 − a )s + 1 − a + a 2 s
s3 + s2 + s + d
( s + a)
Problems 125

( )
Thus, the roots of the denominator are s, (s + a) and

s 2 + bs + c = s 2 + (1 − a )s + 1 − a + a 2

Insert the solutions for a, b, c, d into the general solution and plot the result.

SOLUTION

a) For K = 0.5

Θ m (s) =
( )
0.5
s + s + s + 0.5 s
3 2

From a graph of s + s + s + 0.5 , one root of the denominator is (s +.6478). Long division
3 2

provides the other imaginary root, so the whole transfer function is:

(s + 0.6478)(s )
Θ ( s) =
0.5
2
+ 0.3522s + 0.7718 s

which is of the form:

(s + a )(s 2 + bs + c )s
Θ (s ) =
d

Using a partial fraction expansion:

( ) ( )
Cs + D
= + + 2
(s + a ) s + bs + c s s (s + a ) s + bs + c
d A B
2

( ) ( )
or

d = A(s + a ) s 2 + bs + c + B s 2 + bs + c s + (Cs + D )(s + a )s

( ) ( ) ( ) ( )
so

d = A s3 + (a + b )s 2 + (c + ab )s + (ac ) + B s 3 + bs 2 + cs + C s3 + as 2 + D s 2 + as

and, therefore, equating coefficients on the left and right hand sides:

0 = (A + B + C )
0 = ( A(a + b ) + Bb + Ca + D )
s3 :

0 = ( A(c + ab ) + Bc + Da )
2
s :
1
s :
s : 0
d = Aac
So

A=
d
ac
126 Wind Energy Explained

d (b − a )
C=
(
c c − ab + a 2 )
⎛d ⎞
B = −⎜ + C ⎟
⎝ ac ⎠

⎛ A(c + ab ) + Bc ⎞
D = −⎜ ⎟
⎝ a ⎠
Also, the transfer function:

( )
Cs + D
Θ m ( s) = + + 2
s (s + a ) s + bs + c
A B

can be broken down into functions for which the inverse Laplace transform can be
determined:

(D − C b 2 )
( ) ( )
s+b 2 c − b2 4
Θ m ( s) = + +C +
s (s + a ) (s + b 2)2 + c − b 2 4 c − b 2 4 (s + b 2 ) + c − b 4
A B
2 2

(D − C b 2) e −bt 2 sin⎛ t
Using those inverse Laplace Transforms:

θ (t ) = A + Be − at + Ce −bt 2 cos⎛⎜ t c − b 2 4 ⎞⎟ + ⎜ c − b 2 4 ⎞⎟
⎝ ⎠ c −b2 4 ⎝ ⎠

For K = 0.5
a = 0.6478
b = 0.3522
c = 0.7718
d = 0.5
And
A =1
C = −.199
B = −.801
D = −.589
A graph of the resulting response is:
Problems 127

1.2

Non-dimensionalized Pitch Angle

1.0

0.8
System Response K=0.5
0.6

0.4

0.2

0.0
0 4 8 12 16 20 24 28
Time, seconds

The step response has some overshoot and is not highly damped, but is reasonably close
to the commanded position after 4 seconds. The response settles at the desired position
after about 18 seconds.
b) For K = 3

Θ m ( s) =
(s )
3
3
+ s2 + s + 3 s
From a graph of s + s + s + 3 , one root of the denominator is (s +1.57474). Long
3 2

division provides the other imaginary root, so the whole transfer function is:

Θ m ( s) =
(
(s + 1.57474) s − 0.57474s + 1.90506 s )
3
2

Using the same partial fraction expansion and transformation developed in part a), the

(D − C b 2) e −bt 2 sin⎛ t
system time response is:

θ (t ) = A + Be − at + Ce −bt 2 cos⎛⎜ t c − b 2 4 ⎞⎟ + ⎜ c − b 2 4 ⎞⎟
⎝ ⎠ c −b 4 2 ⎝ ⎠

For K=3
a = 1.57474
b = −0.57474
c = 1.90506
d =3
And
A =1
C = −.640
B = −.360
D = −.200
A graph of the resulting response shows unstable behavior:
128 Wind Energy Explained

Non-dimensionalized Pitch Angle

80

40

-40 System Response K=3

-80

0 4 8 12 16 20 24 28
Time, seconds
c) The response with K = 3 is an unstable response whereas the response with K = 0.5 is
a stable one.

8.7 A wind turbine manufacturer wants to design a yaw drive control system. To minimize
wear on the drive gears the yaw is to be locked with a yaw brake until the 10 minute time-
averaged yaw error is more than some specified amount (the "yaw error limit"). At that
point the yaw drive would move the turbine to face the previously determined 10 minute
time-averaged wind direction.
a) What consequences does the choice of yaw error limit and averaging time have on
machine operation?
b) What approach would you take to determining the quantitative tradeoffs between
yaw error limit and other factors?

SOLUTION

a) Decreased averaging time and decreased yaw error limits would increase the actions of
the yaw drive. A larger yaw error limit or a longer averaging time would reduce the
operating hours of the yaw drive but increased time at higher yaw errors. High yaw error
operation decreases energy capture and may increase machine loads and, consequently,
component fatigue. High yaw error operation might also increase electrical power
fluctuations, depending on the generator system, and tower torsional fatigue, depending on
the yaw system design. Thus, tradeoffs exist between the operating hours of the yaw drive
and energy capture and machine life.
b) Two approaches can be used to analyze the consequences of different yaw error
limits: time series simulations and statistical approaches.
The first issue to recognize is that the time between yaw orientation corrections may be
very site dependent. In turbulent conditions or as weather fronts pass, more yaw
corrections may be required because the wind direction is changing rapidly or frequently.
In flat sites with constant prevailing winds, few corrections may be required. On the other
hand, at sites with diurnal sea and land breezes the turbine may have to rotate at least once
a day. This rotation might happen in few or in many steps.
Thus, an analysis might start with a representative time series of wind direction data.
From this one could develop 1) a file of running averages, based on the running averaging
time to be investigated for use in the controller, and 2) a histogram of averaged yaw errors
that would occur between yaw motions. These histograms could be used with turbine
Problems 129

models to determine loads, electrical power, and energy capture for different yaw errors
and to develop overall estimates of component fatigue, yaw system operating hours, and
energy capture.

8.8 A pitch control system is being designed for a wind turbine. The response of the pitch

[ ]
control system to a unit step command has been determined to be:

θ p = θ p ,ref 1 − 1.67e −1.6t cos(1.2t − 0.93)

where θp is the pitch angle change and θp,ref is the magnitude of the commanded pitch angle
change. In order to shut down the turbine in high winds the pitch must change by 16
degrees. This motion is opposed by three torques, those due to friction, the pitching
moment and inertial. The total friction torque, Q friction , is assumed to be a constant 30 Nm,
and the total pitching moment, Q pitching , in high winds is assumed to be a constant 1000
Nm. Finally, the inertial torque is a function of the inertial moments: Qinertia = Jθ&&p . The
.
total moment of inertia, J, of the blades is 100 kg m2. The total power needed to pitch the

( )
blades is:

Ptotal = θ& p Jθ&&p + Q friction + Q pitching

What is the peak power required to pitch the blades 16 degrees (0.279 radians)?

SOLUTION
Note: The problem statement in the text mixes two sets of nomenclature. In this solution
and problem description, the correct nomenclature has been used.

The power to overcome inertia is:

Pinertia = θ& p Qinertia = θ& p Jθ&&p

The power to overcome friction is:

Pfriction = θ& p Q friction

The power to overcome the pitching moment is:

Ppitching = θ& p Q pitching .

( )
Thus, the total power required to pitch the blades is

Ptotal = θ& p Jθ&&p + Q friction + Q pitching

The pitching velocity, under the specified conditions, is:

130 Wind Energy Explained

(
θ& p = θ p, ref 2.672e −1.6t cos(1.2t − .93) + 2.004e −1.6t sin (1.2t − .93) )
After differentiating and collecting terms, the pitching acceleration, under the specified

(
θ&&p = θ p, ref − 1.8704e −1.6t cos(1.2t − .93) − 6.4128e −1.6t sin (1.2t − .93) )
conditions, is:

The total contributions to the required power can be determined using the expressions
above. The total power and the contributions to the power are illustrated in the following
graph.
240

Pitch System Power Requirements

200

160
Power, Watts

120

Total Power
80 Pitching Power
Fiction Power
40 Inertia Power

0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8

Time, seconds

The maximum power required is about 243.6 W. Most of the required power, in this
case, is needed to overcome the pitching moment.

8.9 The transfer function for the shaft torque, Q S (s ) , of a 250 kW grid-connected, fixed-

αW (s ) , is:
pitch wind turbine with an induction generator as a function of the aerodynamic torque,

49(s + 96.92)

(s + 97.154)(s 2 + 1.109s + 48.88)

=
Q S (s)
αW ( s )

system, a second-order system and a term of the form K (s + a ) . The transfer function for
a) This transfer function can be characterized as being the product of a first-order

second-order systems can be expressed as:

Problems 131

ωn2
G ( s) =
s + 2ζω n s + ω n 2
2

where ω n is the system natural frequency and ζ is the damping ratio. What is the natural
frequency and damping ratio of the second-order system component?
b) The transfer function for first-order systems can be expressed as:

1τ
G ( s) =
s +1 τ

where τ is the time constant of the first-order system. What is the time constant of the
first-order system component?

SOLUTION

a) The transfer function for second order systems can be expressed as:

ωn2
G ( s) =
s 2 + 2ζω n s + ω n 2

where ω n is the system natural frequency and ζ is the damping ratio.

The transfer function can be written in the form:

( )
⎛ ⎞
= (s + 96.92) 97.154 ⎜⎜ 2 ⎟
(s + 97.154) ⎝ s + 2(0.0793)6.99s + 6.992
6.99 2
αW ( s) 97.154 ⎟
QS ( s) 1.002
⎠
In this form it can be seen that the natural frequency of the system is 6.99 rad/s and the
damping ratio is 0.0793.
b) From the equation above, the time constant for the first order component is 0.0103
seconds.

8.10 A proportional–integral–derivative (PID) controller can be designed using the

[ ]
electrical circuit pictured in Figure B.9. The differential equation for the controller is:

∫ ∫ e(t )dt ⎥⎦⎥

⎡⎛ R ⎤
g (t ) = − K P e(t ) + K D e&(t ) + K I e(t )dt = − ⎢⎜⎜ 2 + 1 ⎟⎟ e(t ) + (R 2 C1 ) e&(t ) +
C ⎞ 1
⎣⎢⎝ 1
R C 2 ⎠ R1C 2

where g(t) is the controller output, and R and C are the resistances and capacitances of the
respective circuit elements, e(t) is the error signal that is input to the controller and the
controller constants are K P , K I , and K D .
132 Wind Energy Explained

R2 C2
C1

R1

Figure B.9 PID Controller

If R1 is 10000 Ohms, what values of R2, C1 and C2 would be required for K P , = 10, K I =
100 and K D = 0.01. Note resistances are in Ohms and capacitances are in Farads.

SOLUTION

Note, the value for KD was mistakenly listed as 100 in the problem statement in the text.
The correct problem statement is included here.

[ ]
From

∫ ∫ e(t )dt ⎥⎦⎥

⎡⎛ R ⎤
g (t ) = − K P e(t ) + K D e&(t ) + K I e(t )dt = − ⎢⎜⎜ 2 + 1 ⎟⎟ e(t ) + (R2C1 ) e&(t ) +
C ⎞ 1
⎣⎢⎝ 1
R C 2 ⎠ R1C 2

KI = = 100
1
R1C2
so C2 =10-6 Farads

Combining yields:

K D = R2C1

And

⎛ KD C ⎞
⎜⎜ + 1 ⎟⎟ = K P
⎝ R1C1 C 2 ⎠

Solving for C1, one obtains:

C12 − C1 (C2 K P ) + =0
K D C2
R1
Problems 133

Substituting the values for R1, C2, KD and KP, one can solve for C1 and R2:
R2 = 100,000 Ohms
C1 = 10-7 Farads

8.11 If it is required that a digital control system determine the system behavior at
frequencies of at least 10 Hz, a) what is the minimum sampling frequency that can be used
and b) what cut-off frequency should be used to filter the input data?

SOLUTION

Sampling at 20 Hz is required to correctly represent dynamics at 10 Hz, the Nyquist

frequency. Filtering frequencies above 10 Hz is required to make sure that higher
frequency components do not affect the signal.

8.12 The transfer function of a blade pitch controller meant to control the mean rotor
power above rated on a constant speed wind turbine is approximated as:

Θ
=
0.5
Θ ref s + 0.5

Using partial fraction expansions, calculate the step response of the system. Thus, assume
that:
Θ ref =
1
s

What is the step response (the time for the blade pitch to get from 10% to 90% of the
desired position)?

SOLUTION

From the problem statement:

Θ=
1 0.5
s s + 0.5
Using partial fraction expansions:
0.5 = +
A B
s s + 0.5

Multiplying by the denominator of the transfer function:

0.5 = A(s + 0.5) + Bs

Or
134 Wind Energy Explained

0.5 = 0.5 A + ( A + B )s

Thus, equating coefficients of the terms:

A+ B = 0
A =1

So:

B = −1
Then:

Θ= −
1 1
s s +1
The inverse Fourier transform of this is:

θ = 1 − 1e −t
A graph of the system response is shown in the following figure.

1.2

0.8
Phi, radians

0.6

0.4

0.2

0
0 2 4 6 8 10 12
Tim e, seconds

t = − ln (1 − θ )
This can be solved for t:

For θ = 0.1, t = 0.105 s; for θ = 0.9, t = 2.303 s. Thus, the step response is 2.2 seconds.

8.13 An engineer is studying the results of changing the blade pitch control loop gain in
the controller that maintains the turbine rotor speed and generator power. The results of her
simulations are provided in [Link], [Link] and
[Link]. Note that the time steps for the case with a gain of -100 are less
Problems 135

than in the other data sets. Determine the average and standard deviation of the blade pitch
angle (degrees), the generator power (kW) and the rotor speed (rpm). Also determine the
average and standard deviation of the blade pitch rate (deg/sec). What difference do more
negative gains make? What is happening when the gain is -100?

SOLUTION

The analysis is included in [Link]. The results are:

Generator Power, kW
Gain
-1 -10 -100
Average 1393.5 1419.2 1385.9
Std. Dev. 90.8 16.0 52.6

Blade Pitch, deg.

Gain
-1 -10 -100
Average 19.4 19.3 19.5
Std. Dev. 3.1 3.4 7.4

Rotor Speed, rpm

Gain
-1 -10 -100
Average 20.5 20.5 20.3
Std. Dev. 0.7 0.1 0.3

Blade Pitch Rate, deg/sec

Gain
-1 -10 -100
Average 0.00 0.00 0.04
Std. Dev. 0.92 3.76 80.08

More negative gains (at least up to -10) improve the control in the sense that the rotor
speed and generator power variations decrease. On the other hand, the blades much be
pitched more actively and the blade pitch variability and pitch rate variability increase. At a
gain of -100, the pitch rates are on the order of 80 degrees per second and the controller is
loosing control of the rotor speed (the rotor speed and generator power variations increase).

8.14 If the rotor speed controller on a variable speed turbine can accurately maintain the
rotor tip speed ratio at its optimum value, then the rotor CP during constant tip speed
operation will always be at its maximum. In the real world, the control system cannot
achieve this level of control. Suppose that the range of tip speeds over which the turbine
136 Wind Energy Explained

operates during constant tip speed operation is described by a normal probability density
distribution described by:
⎡ (b (λ − λ0 ))2 ⎤

P(λ ) =
−⎢ ⎥
b ⎣⎢ ⎦⎥
2π
2
e

In this equation, b is a parameter that describes the width of the distribution of tip speed
ratios. Suppose also that the CP(λ) within a reasonable distance from the optimum tip speed
ratio can be described as:

C P (λ ) = −a(λ − λ0 )2 + C P,max

In this equation, a is a parameter that determines the curvature at the top of the CP – λ

∫C
curve. The average CP that is described by this type of operation is then:

P (λ )P(λ )dλ
Use Excel to numerically integrate the equation above to determine the average CP and the
percentage of optimum CP when b is either 1 or 2 and when a is either 0.01 or 0.02. When
a is fixed at 0.02, how much is the CP improved when the distribution of the tip speeds is

curve has less curvature? Assume that C P ,max = 0.47 .

narrower? When b is fixed at 1, how much is the CP improved when the top of the CP – λ

SOLUTION

See TrackingOptimumTipSpeed [Link] for the analysis. The results are:

a 0.01 0.02 0.01 0.02

b 1 1 2 2
0.44 0.46 0.46
Cp 0.46 9 7 4
Percent of Cp,max 97.8 95.6 99.4 98.8

When a is 0.02 the CP(λ) is less flat near the optimum tip speed ratio and the CP is 3.4%
greater when the distribution of the tip speeds is narrower (the tip speed is more tightly
controlled). When b is 1 (a wider range of tip speeds), the CP improves by 2.1% when the
top of the CP – λ curve has less curvature.

b) For a 5 MW turbine, with R = 63 m, λopt = 8, and CP,max = 0.50, and ρ = 1.225 kg/m3,
8.15 a) Derive Eq. 8.14 in the text. Hint, Power=Torque x Speed.

ρπR 5C p , max
calculate the proportionality constant:

( )
k=
2 λopt 3
Problems 137

SOLUTION

a) The maximum rotor power is:

Pmax = ρAU 3C p , max
1
2

ΩR
From the definition of tip speed ratios:
U=
λ
Substituting this into the initial equation, for conditions at CP,max one gets

1 ⎛⎜ ΩR ⎞⎟
Pmax = ρA
3

2 ⎜⎝ λopt ⎟⎠
C p , max

And using the definition of area, A = πR 2 , and the definition of power, P = Ω Q, one gets:

⎛ ΩR ⎞
Qopt Ω = ρπR 2 ⎜ ⎟ C
3

⎜ λopt ⎟ p , max
1
2 ⎝ ⎠
Collecting terms and designating the optimum torque as the reference torque for the
controller one gets:

ρπR 5C p , max
( )3
Qref = Ω2
2 λopt

b) For the stated conditions:

ρπR 5C p , max 1.225π 635 (0.50 )

( )3
k= = = 1864912 Nms2
2 λopt 2(8)3

B.9 Chapter 9 Problems

9.1 A wind farm is being considered for a ridge top site. Name 10 or more issues that
might be considered in evaluating this site.

SOLUTION

Considerations might include: mean wind speed, variations of wind speed over the site,
turbulence intensity, topographic effects on directional characteristics of the wind,
available area for turbines, the presence of nearby wind farms, ease of access, competing
land use issues, slope of the terrain, endangered species, environmentally sensitive areas
(e.g. vernal pools), visibility, proximity to houses, grid strength, need for a grid extension
138 Wind Energy Explained

and cost of any grid extension, legal issues, visual pollution concerns, extreme weather, the
possibility of danger to hikers, size of turbine that can be transported to the site, cost of
road building, site preparation, and turbine transport, availability of cranes, access for
maintenance personnel, interference with microwave transmissions, erosion and run-off
from road and site work, and the availability of accurate enough wind data for energy
capture projections.

9.2 Four identical wind turbines that are lined up in a row 12 rotor diameters apart are
experiencing wind parallel to the row of wind turbines. Use Katić’s wake model to
determine the speed of the wind approaching each of the wind turbines. Assume that k =
0.10 and that the thrust coefficient is 0.7.

SOLUTION

The most upwind turbine (which will be called #1) experiences the free stream wind speed,
U0 .

( ) ( )
At 12 rotor diameters, the next turbine (#2) experiences a velocity deficit of:

1 − 1 − CT 1 − 1 − 0.7
1− = = = 0.039
( + 2(0.1)12 )2
UX
U0 ⎛ X⎞
⎜1 + 2 k ⎟
2
1
⎝ D⎠

Thus #2 experiences a wind speed of 0.961 U0.

Turbine #3 experiences consequences from both turbines #1 and #2. The wind speed at
turbine #3 due to the wake of turbine #2, which sees the wake of turbine #1 is 0.9612 =
0.924 U0. Similarly, the wind speed at turbine #4 is 0.9613 = 0.888 U0.
The equation for interacting wakes is not needed here as that assumes that spreading
wakes of turbines not directly upstream start to merge a distance downstream.

9.3 A wind farm developer has identified a site with unique winds. They blow all of the
time from one direction and at one speed, 15 m/s. The site has enough room for two rows
of turbines perpendicular to the prevailing winds. She is wrestling with which size of
turbines and how many to put in each row. All turbines being considered have the same
hub height. She has developed two options.

Option 1 – This is an array of 24 1.5 MW turbines in two rows. The turbines just fit
onto the site:

• Front row – 12 turbines, each with a rotor diameter of 60 m and a rated power output of

• Second row 500 m behind the first one (only 8.33 rotor diameters) – 12 turbines, each
1.5 MW in 15 m/s wind speeds.

with a rotor diameter of 60 m and a rated power output of 1.5 MW in 15 m/s wind
speeds.
Problems 139

Option 2 – This option includes slightly smaller turbines in the first row. This allows a
few more turbines in the first row and reduces the velocity deficit at the second row
because of the smaller rotor diameters in the first row. In addition, because of the reduced
concerns of abutters, these smaller turbines can be moved 100 m closer to the front edge of
the site, further reducing the wake effect. The second row is the same as in Option 1:

• Front row – 15 turbines, each with a rotor diameter of 50 m and a power output of 1.0

• Second row 600 m behind the first one –12 turbines, each with a rotor diameter of 60 m
MW in 15 m/s wind speeds.

and a power output of 1.5 MW in 15 m/s wind speeds.

To evaluate her options, the developer uses Katić’s wake model to determine the wind
speed behind the first row of turbines. She assumes that k = 0.10 and that the thrust
coefficient is 8/9. She also assumes that in winds lower than the rated wind speed of 15
m/s, the power output of the turbines can be approximated by:

⎛U ⎞
P = Prated ⎜⎜ X ⎟⎟
3

⎝ 15 ⎠
a) For both options, what are the power production for each row and the total power
production?
b) If the turbines cost $800/installed kW, how much will each option cost to install (assume
that no loans are needed)?
c) If the developer’s annual operation and maintenance (O&M) costs are 7% of the
installed cost plus $10/MWh for maintenance, what are the total annual O&M costs?
d) If the developer’s income from the sale of energy is $30/MWh, what is her net income
for a year (Sales income - annual O&M costs)?
e) If net income is the deciding factor, which option should she pursue?

SOLUTION

a)-d) The results of the calculations for a) through d) are presented in the table below.
e) Based on this analysis, option 1 provides 1.9% more income per year. If net income
is the deciding factor, then option 1 should be pursued.

Option 1

# turbines U/Ufree stream wind m/s MW/turbine Total MW

Row 1 12 1 15 1.5 18
Row 2 12 0.906 13.59 1.116 13.392
Total Power Production 31.392

Installed Installed Net

MW Cost O+M Income Income
36 $28,800,000 $4,765,939 $8,249,818 $3,483,878
140 Wind Energy Explained

Option 2

# turbines U/Ufree stream wind m/s MW/turbine Total MW

Row 1 15 1 15 1 15
Row 2 12 0.942 14.13 1.255 15.06
Total Power Production 30.06

Installed Installed Net

MW Cost O+M Income Income
33 $26,400,000 $4,481,256 $7,899,768 $3,418,512

9.4 A wind farm consists of two turbines, located 100 m from each other. They are
operating in winds with a mean speed of 10 m/s and an integral time scale of 10 s.
a) What does the graph of the Von Karman spectrum look like (using log–log
coordinates)?
b) What is the function for the wind farm filter for these two turbines?
c) Suppose that the spectrum of the local wind can be expressed by:

S1 ( f ) = σ U2 40e −40 f

Assume also that each wind turbine sees the same mean wind speed, U, and that the
total average wind turbine power, PN, can be expressed as PN = NkU where k is equal
to 10 and N is the number of wind turbines. What is the standard deviation of the wind
power from the two wind turbines? How does this compare to the standard deviation
of the power from two wind turbines that experience exactly the same wind?

SOLUTION

a) The graph of the Von Karman spectrum is presented below:

100

10
Non-dimensional Variance

0.1 Von Karman Spectrum

0.01

0.001

0.0001

0.0001 0.001 0.01 0.1 1 10 100

Frequency, Hz
b) For two wind turbines, the wind farm filter is expressed as
Problems 141

∑∑ γ f)= (γ 11 ( f ) + γ 12 ( f ) + γ 21 ( f ) + γ 22 ( f ) )
N N
1 1
ij (
N2 i =1 j =1 22

where the coherence function is defined as:

γ ij (f ) = e
− 25
x ij
f
V

For xij = 100

γ 12 (f ) = γ 21 (f ) = e −250f

and

γ 11 (f ) = γ 22 (f ) = 1

Then the wind farm filter is:

∑∑ γ )= (
1 + e − 250 f )
N N
1 1
ij ( f
i =1 j =1
2
N 2

c) The variance of the fluctuating power from one wind turbine is given by

∫
∞
σ P2 ,1 = k 2 S1 ( f ) df
0

or, for k = 10:

∫
∞
⎛ − 1 − 40 f ⎞
∞
σ P2 ,1 = 4000σ V2 e − 40 f df = 4000σ V2 ⎜ ⎟ = 100σ V2
⎝ ⎠0
e
40
0

so σ P,1 = 10σ V

The variance of the total fluctuating power is given by

∫ ∫ ∑ ∑γ
∞ ∞⎧ ⎫⎪
⎪
σ P2 , N =N k S N ( f ) df = k ⎨S1 ( f ) f )⎬df
N N

0⎪ ⎪⎭
2 2 2
ij (
0 ⎩ i j

( )
From the previous section:

∑∑ γ )= 1 + e − 250 f
N N
1 1
ij ( f
i =1 j =1
2
N 2
142 Wind Energy Explained

( )
or

∑∑ γ ij ( f ) = 1 + e − 250 f
N N
N2
i =1 j =1
2

and S1 ( f ) = σ V2 40e −40 f

∫{ ( )} ∫{ ( )}
∞ ∞
so σ P2 , N = 20σ V2 N 2 k 2 e − 40 f 1 + e − 250 f df = 8000σ V2 e − 40 f 1 + e − 250 f df
0 0

∫ (e )
Then
∞
σ P2 , N = 8000σ V2 − 40 f
+ e − 290 f df

( )
0

σ P2 , N = 8000σ V2
∞
−1 −40 f
e + −1
e −290 f

( ) = 227.59σ
40 290 0

σ P2 , N = 8000σ V2 1
+ 1 2

( )
40 290 V

so σ P, N = 15.09σ V = .755 2σ P ,1

The standard deviation from two turbines experiencing the exact same wind would be
20σV. Thus, the standard deviation of the total power from the two wind turbines is 75.5%
of what would be expected with two correlated wind turbines.

9.5 Consider a wind turbine generator connected to a grid system (see Figure B.10) with a
line-neutral system voltage, VS. The voltage at the wind turbine, VG, is not necessarily the
same as VS. The distribution system resistance is R, and the distribution system reactance is
X.

Transmission VS VG
system

R X

Distribution
system

Figure B.10
Problems 143

generator is Q. Suppose R = 8.8 Ω, X = 4.66 Ω, Vs = 11 kV, P = 313 kW, Q = 162 kVAr

The real generated power is P and the reactive power consumed by the wind turbine

and the power factor at the generator is pf = 0.89. Determine the voltage difference
between the grid and the generator, the voltage drop as a percent of the grid voltage, the
fault level at the generator and the installed generator capacity as a percent of the fault
level.
144 Wind Energy Explained

SOLUTION

[ ]
The voltage at the generator can be determined from:

VG 4 + VG 2 2(QX − PR ) − VS 2 + (QX − PR )2 + (PX − QR )2 = 0

Inserting the known values, one finds:

VG = 11,178 V or 178 V

The only realistic answer is 11,178 V at the wind turbine, so:

ΔV = 178 V
In lightly loaded distribution circuits, the voltage change can be approximated as:
PR − QX
ΔV = VG − VS = = 181.8 V
VS

This is 1.65% if the nominal line voltage level. The electrical losses in the distribution

(P )
system are:

+ Q2 R
W = = 9.03 kW
2

VS 2

The fault current, I F , is:

(R )
IF = = 1104 A
VS
+ X2
2 12

and the fault level, M, is:

M = I FVS = 12.1 MVA

The turbine rating is 2.6% of the fault level.

9.6 A wind turbine is connected to an 11 kV distribution line. The magnitude of the

harmonic voltages, cn, at the point of common connection of the wind turbine and other
grid users are listed in Table B.9.
Table B.9

n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
cn, Volts 11000 120 0 85 15 0 99 151 0 12 216 0 236 80 0

What is the harmonic distortion of each harmonic and the total harmonic distortion? Are
these within the allowable IEEE 519 limits?
Problems 145

SOLUTION

The harmonic distortion caused by the specific harmonics is:

∫
T
1 v 2 dt
T n
HDn =

∫
0
T
1 v 2 dt
T F
0

where

∫ ∫ ∫
⎛ ⎛ π t ⎞⎞ ⎛ ⎛ 2π t ⎞ ⎞
vF 2 dt = ⎜⎜ vF sin ⎜ ⎟ ⎟⎟ dt = F ⎜⎜1 − cos⎜ ⎟ ⎟⎟dt
T T 2 T
1 1 v 2
0⎝
⎝ L ⎠⎠ ⎝ ⎝ L ⎠⎠
1
2
T T T
0 0

⎛ ⎛ 2πt ⎞ ⎞
= ⎜⎜ t − sin ⎜ ⎟ ⎟⎟ = F
T
vF 2 v 2
⎝ 2π
l
2T ⎝ T ⎠ ⎠0 2

Similarly

∫ ∫
⎛ ⎛ nπ t ⎞⎞
vn 2 dt = ⎜⎜ vn sin ⎜ + φ n ⎟ ⎟⎟ dt = n
T T 2
1 1 v 2
T
0
T
0⎝
⎝ L ⎠⎠ 2

Thus

∫v
T
1 2
dt
n

HDn = =
T vn

∫v
0
T vF
1 2
dt
F
T
0

Results for the measured voltages are presented below.

n vn HDn%
1 11000
2 120 1.09
3 0 0.00
4 85 0.77
5 15 0.14
6 0 0.00
7 99 0.90
8 151 1.37
146 Wind Energy Explained

9 0 0.00
10 12 0.11
11 216 1.96
12 0 0.00
13 236 2.15
14 80 0.73
15 0 0.00

The THD is:

∑ (HD )2
∞
THD =
n=2
n

The THD is 3.67% and the highest harmonic distortion of any individual harmonic is
2.15%. Thus, these harmonics are within the allowable range of IEEE 519.

9.7 The data file ‘Site1 v Site2 For [Link]’ includes five years of wind speed and
direction data from Site 1 (the reference site) and Site 2 (the candidate or project site). Data
that was determined to be incorrect (due to sensor icing or other problems, for example) is
indicated by -999. Use this data to determine the appropriate Variance MCP relationship
between the two sites using three different years of concurrent data: 1999, 2001 and 2002.
What are the slopes and offsets determined from these three different concurrent data sets?
What is the data recovery of the data sets used for the analysis expressed as a percentage of
a year worth of data? What might be the cause of the difference between the results for
different years?

SOLUTION

The calculations have been done in Site1 v Site2 For MCP_Years 1999, 2001, 2002
[Link]. The results are shown here:

Year of Concurrent Data Slope Offset Data Recovery

1999 0.998 0.214 0.958
2001 0.912 0.347 0.914
2002 0.915 0.492 0.948

There is a significant variability between the slopes, which range from 1.00 to 0.91. There
is some variability between the offsets that were determined. All of the data recovery
percentages were over 91%. The causes of the variability is unknown but could be due to
sensor changes, changes in the ground cover upwind of either site over the four year period
or variability in the relationship between the winds at the two locations.

9.8 A study determines that the maximum chord size in meters of blades can be expressed
as:
Problems 147

c = 0.0005L2 + 0.024 L + 1.4

where L is the blade length and the mass of blades in kilograms can be approximated as

m = 9.043L2 − 340 L + 6300

The clearance under the bridges on the only road to a proposed project are 5.5 m and the
trailer carrying the blades to the project must maintain a clearance between the blade and
the road of 1 m. Additionally, the overpasses on the road are only rated for loads (tractor
trailer plus cargo) of 36,000 kg. Note that the empty tractor trailer has a mass is 16,000 kg.

a) If the underpass clearances are determined by the maximum blade chord what is the
largest blade that can be transported to the project that will fit under the road underpasses
and over the bridges?

b) Final access to the project is along a narrow dirt road. The minimum radius of curvature
of the road, r, that will accommodate a blade of length L when the road width is w is:

⎛L⎞
⎜ ⎟ −w
2

⎝ 2⎠
2

r=
2w
What is the minimum radius of curvature that will accommodate the blade chosen in part a
if w = 8 m?

SOLUTION

a. The available clearance is 4.5 m between the trailer and the top of the underpasses. The
available blade weight is only 20,000 kg due to the weight of the tractor and trailer. The 4.5
m maximum chord length limits the blades to 58.5 m. The mass limit of 20,000 kg limits
the blades to about 62 m long. Thus, the underpass limit is the more restrictive and the
project must use wind turbines with blades no longer than 58.5 m long.

b. Plugging the values into the formula, the minimum radius of curvature is 68.3 m.

9.9 A developer decides to install a 45 turbine offshore wind farm. The wind map of the
proposed project is shown in the Figure B-11. The rows are 2 km apart in the north-south
direction and 1 km apart in the east-west direction. The closet turbine to land is 4 km to the
east of Eastern Point.
148 Wind Energy Explained

Figure B.11 Map of proposed project.

The wind map shows the expected range of mean hub height wind speeds at the site in each
shaded area. The Weibull shape factors found at this site are 2.4. As the permitting for the
project moves forward, the developer realizes that he will need to eliminate 4 turbines in
the higher wind speed region due to concerns at archeologically valuable shipwrecks
nearby. He also decides to eliminate all turbines within 5.7 km of Eastern Point to avoid
conflicts with lobstermen who are concerned about environmental damage to lobster beds.
What is the expected range of annual energy generation if the turbines that will be used
have a rated power of 4.0 MW above 13 m/s and has a cubic power curve from 4 m/s to
rated and a cut-out wind speed of 24 m/s?
Problems 149

SOLUTION

The final number of wind turbines in the lower wind speed area is 14 and in the high wind
area it is 17. The average capacity factors are shown in the table below. The analysis can be
found in [Link].

Minimum expected CF 0.277

Maximum expected CF 0.315

9.10 A proposed wind farm will have 18 wind turbines in a row. Each will be 8 rotor
diameters apart. The wind speeds in each 30 degree direction bin are characterized in the
Table B.10. The turbine that will be used in this project generates rated power of 2.0 MW
above 11 m/s and has a cubic power curve from 4 m/s to rated and a cut-out wind speed of
24 m/s. The array losses are assumed to be 15% when the wind is from the north (0
degrees) and 0% when the wind is from other directions. Additionally, it is assumed that
other losses (downtime, electrical losses, weather conditions, etc.) will be 20%. Under
these assumptions, what will the total generation of the wind farm be? After consulting the
manufacturer, the developer learned that the wind turbine manufacturer would not honor
the turbine warranty if the wind turbines were operated when the wind was directly along
the row of the turbines (in direction sectors bins of 0 and 180 degrees) due to excessive
wake turbulence. Given this new information, how much can the developer expect to
generate at this wind farm and preserve the warranty on his turbines?

Table B.10 Wind speed and direction data

Direction Percent of
Mean Wind Speed, m/s Weibull Shape Factor
Bin Center Time
0 7.9 2.0 8.0
30 7.0 2.1 6.0
60 6.7 2.0 4.0
90 5.6 2.1 10.0
120 6.9 2.3 8.0
150 7.2 2.1 7.0
180 8.8 2.2 7.0
210 9.2 2.1 10.0
240 9.0 2.0 8.0
270 8.3 2.1 14.0
300 8.4 2.1 12.0
330 7.6 2.2 6.0
150 Wind Energy Explained

SOLUTION

The answers are presented in the table below and the details of the analysis are in
[Link].

Project CF including array losses 0.430

Net Project CF including all losses 0.344

Project CF w/ Dir. Sector Management including array losses 0.368

Net Project CF w/ Dir. Sector Management, incl. all losses 0.294

9.11 [Link] contains average hourly electrical load (GW)

in a utility control area and the wind power that is expected to be generated by future
planned wind farms.
a. Determine the total GWh, mean, maximum, minimum and standard deviation of
the gird load and the generated wind power. Graph the load duration curve of the
utility with the existing load (before the installation of the wind plants).
b. Determine the total GWh, mean, maximum, minimum and standard deviation of
the net load once the wind power plants are installed. Graph the load duration
curve of the net load of the utility after the installation of the wind plants.
c. How much does the mean electrical load decrease with the installation of the wind
power plants? How much does the standard deviation of the load seen by the grid
operators (the net load) increase with the inclusion of wind? How much does the
peak net load change?

SOLUTION

Answers to a) and b): The statistics of the data are:

Original Wind Net

Load Power Load
Average GW 26.26 3.67 22.58
Total GWh 229,995 32,183 197,812
Standard Deviation,
GW 4.96 4.16 6.43
Max GW 43.06 12.00 43.06
Min GW 11.03 0.00 0.24

The cumulative distributions are (see [Link]):

Problems 151

50

45
40

35
30
Net Load
Load

25
Original Load
20
15

10
5

0
0 0.2 0.4 0.6 0.8 1
Cummulative Probability

c. The mean electrical net load decreased from 26.26 GW to 22.58 GW but the standard
deviation increased from 4.96 GW to 6.43 GW. Meanwhile the peak net electrical load
remains the same.

B.10 Chapter 10 Problems

10.1 An isolated power system serving the community of Cantgettherefromhere uses a 100
kW diesel generator. The community plans to add 60 kW of wind power. The hourly
average load and power from the wind turbine over a 24 hour period is detailed in Table
B.11. The diesel generator has a no-load fuel usage of 3 liter per hour and an additional
incremental fuel use of 1/4 liter per kilowatt-hour.
For the day in question, using the hybrid system design rules, determine:
a) The maximum renewable energy that can possibly be used in an ideal system.
b) The maximum renewables contribution without storage, and the maximum with
storage.
c) The maximum fuel savings that can be achieved.
d) The minimum diesel fuel use that can be achieved with intelligent use of storage and
controls.
152 Wind Energy Explained

Table B.11 Diesel system load data

Hour Load (kW) Wind (kW) Hour Load (kW) Wind (kW)
0 25 30 12 85 45
1 20 30 13 95 45
2 15 40 14 95 50
3 14 30 15 90 55
4 16 20 16 80 60
5 20 10 17 72 60
6 30 5 18 60 48
7 40 5 19 74 50
8 50 15 20 76 55
9 70 20 21 60 60
10 80 25 22 46 60
11 90 40 23 35 55

SOLUTION

The hybrid system design rules are:

Rule 1: The maximum renewable energy that can be used is limited by the load.
Rule 2: The use of renewable energy will be further limited by temporal mismatch
between the load and the renewables.
Rule 3: The maximum possible benefit with improved controls or operating strategies is a
system approaching the fuel use of the ideal diesel generator - fuel use
proportional to the diesel-served load.
Rule 4: The maximum fuel savings arising from the use of renewables in an optimised
system is never greater than the fuel savings of an ideal generator supplying the
proportional reduction in load resulting from use of renewables.

a) Based on rule number 1, the maximum renewables contribution with any system
cannot be larger than the load. Thus the maximum renewables contribution over the day
would be 1338 kWh, the total daily energy need.
b) Based on the rules 1 and 2, the maximum renewables contribution in any hour
without storage is the same as the load, if it is smaller than the available wind power, or the
available wind power. An examination of the data shows that the maximum renewables
contribution without storage would be 819 kWh. The maximum with storage would be the
total daily load if it were less than the total available wind power, or the total available
wind power. The total daily load is 1338 kWh; the total available wind power over the day
is 913 kWh. Thus the maximum renewables contribution with storage is 913 kWh.
c) The total fuel use of the original system is 1338(1/4)+ 24(3) = 406.5 liters/day. The
maximum fuel savings occurs with optimum use of storage, controls and renewables.
Under these conditions, the diesel load with would be 1338 – 913 = 425 kWh. The fuel use
of an optimised system would be ¼ l/kWh or 106.25 l/day, a savings of 300.25 l/day.
Problems 153

d) Based on rule 3, the minimum diesel fuel use that can be achieved with intelligent
use of storage and controls (and without renewables) is that part of the fuel use that is not
proportional to the load: 24 hrs of 3 l/hr = 72 l/day.

10.2 Based on the analysis in Problem 10.1 and other input, the community of
Cantgettherefromhere has upgraded its power system. The hybrid power system includes a
100 kW diesel generator, 100 kW of installed load, 60 kW of wind power, a 100 kW dump
load, and energy storage with 100 kWh capacity. The hourly average load and power from
the wind over a 24 hour period have remained the same and are listed in Table B.11.
Assume that the mean hourly data accurately describe the load and power from the
wind and that fluctuations about the mean load are handled by the energy storage.
Determine the hourly energy flows in the system for the following system control and
operating approaches.
a) Diesel-only system – In this baseline system, the diesel provides all of the power to
the load. The diesel provides power down to 0 kW with no provision to limit the ensure a
minimum diesel load to avoid diesel engine wear.
b) Minimum diesel – In this system minimizing diesel power takes priority, except that
the diesel may not be shut off and must run at a minimum of 30 kW to ensure long diesel
engine life. Thus, the system operates by the rules:

1. The minimum diesel power is 30 kW

2. Only the minimum diesel power that is needed above 30 kW is used (no power above 30
kW is used to fill up the storage)
3. Storage can only be used between 20 and 95 kWh of capacity, to maximize storage
efficiency and battery life
4. Storage capacity starts at 50 kWh
5. The sum of the energy into the power sources in the system (diesel, wind, battery) must
equal the sum of the energy into the power sinks in the system (the load, dump load, and
battery).
6. If there is excess energy in the system it is first stored, if possible, and dumped only if
necessary.

c) Diesel shut off – In this system a rotary inverter between the batteries and the grid
provides reactive power and the diesel can be shut off. It shuts off whenever possible, but
when running, needs to be at a minimum load of 30 kW. When running it is also used to fill
up the storage to use the fuel most efficiently. The battery level is maintained between 20
and 90 kWh. This insures that there is adequate capacity to handle fluctuating loads when
there is no diesel in the system. Thus, the system operates by the rules:

1. The diesel can be shut off

2. The minimum diesel power is 30 kW when the diesel is running
3. When the diesel is running, the storage is also filled up if possible to improve diesel fuel
efficiency
4. Storage can only be used between 20 and 90 kWh of capacity, to maximize storage
efficiency and battery life
5. Storage capacity starts at 50 kWh
154 Wind Energy Explained

6. The sum of the energy into the power sources in the system (diesel, wind, battery) must
equal the sum of the energy into the power sinks in the system (the load, dump load, and
battery)
7. If there is excess energy in the system it is first stored, if possible, and dumped only if
necessary

Specifically, for each operating approach, determine how much energy is supplied by
the diesel over the 24 hour period and the reduction in diesel power compared to the diesel
only case.

SOLUTION

Diesel only system

The baseline system with just the diesel would result in the diesel providing 1338 kWh
of power over the 24 hour period.

Results for the other two systems are detailed in the charts below.
Problems 155

Minimum Diesel System

Minimum Diesel

Run diesel as little as poss., use wind, storage

Diesel - 30-100 kW, no diesel shut-off
Storage - 100kWh capacity, 20-95 kWh range
Dump - 100 kW

Hour Load Wind Diesel Dump Storage Storage Energy

Level Balance
kW kW kW kW kW kWh kW
L W D X B S L-W-D+X+B
Initial Storage Level: 50
0 25 30 30 0 35 85 0
1 20 30 30 30 10 95 0
2 15 40 30 55 0 95 0
3 14 30 30 46 0 95 0
4 16 20 30 34 0 95 0
5 20 10 30 20 0 95 0
6 30 5 30 5 0 95 0
7 40 5 30 0 -5 90 0
8 50 15 30 0 -5 85 0
9 70 20 30 0 -20 65 0
10 80 25 30 0 -25 40 0
11 90 40 30 0 -20 20 0
12 85 45 40 0 0 20 0
13 95 45 50 0 0 20 0
14 95 50 45 0 0 20 0
15 90 55 35 0 0 20 0
16 80 60 30 0 10 30 0
17 72 60 30 0 18 48 0
18 60 48 30 0 18 66 0
19 74 50 30 0 6 72 0
20 76 55 30 0 9 81 0
21 60 60 30 21 9 90 0
22 46 60 30 44 0 90 0
23 35 55 30 50 0 90 0

Totals 1338 913 770 305 kWh

Diesel kWh saved 568
156 Wind Energy Explained

Diesel Shutoff System

Diesel Shut Off

When diesel runs >=30
Run diesel as little as poss., use wind, storage
If not enough wind, storage, run diesel at max eff.
Dump as little as poss.

Diesel - 0-100 kW, diesel shut-off allowed

Storage - 100kWh capacity, 20-90 kWh range

Hour Load Wind Diesel Dump Storage Storage Energy

Level Balance
kW kW kW kW kW kWh kW
L W D X B S L-W-D+X+B
Initial Storage Level: 50
0 25 30 0 0 5 5 0
1 20 30 0 0 10 15 0
2 15 40 0 0 25 40 0
3 14 30 0 16 0 40 0
4 16 20 0 4 0 40 0
5 20 10 0 0 -10 30 0
6 30 5 0 0 -25 5 0
7 40 5 0 0 -35 -30 0
8 50 15 100 0 65 35 0
9 70 20 60 0 10 45 0
10 80 25 0 0 -55 -10 0
11 90 40 100 0 50 40 0
12 85 45 0 0 -40 0 0
13 95 45 90 0 40 40 0
14 95 50 0 0 -45 -5 0
15 90 55 80 0 45 40 0
16 80 60 0 0 -20 20 0
17 72 60 0 0 -12 8 0
18 60 48 0 0 -12 -4 0
19 74 50 0 0 -24 -28 0
20 76 55 89 0 68 40 0
21 60 60 0 0 0 40 0
22 46 60 0 14 0 40 0
23 35 55 0 20 0 40 0

Totals 1338 913 519 54 kWh

Diesel kWh saved 819
Problems 157

10.3 This problem concerns a hypothetical island wind/diesel system. Wind and load data
for the island for one year are in the files Wind_TI.txt and Load_TI.txt. Assume that the
wind turbines available are (1) and AOC 15-50, (2) one with twice the output or (3) one
with 4 times the output. The power curves for these turbines are, respectively,
AOC_pc.csv, AOC_pc_x2.csv and AOC_pc_x4.csv.
The diesel to be considered is one rated at 100 kW, with no load fuel consumption of 2
fuel units per hour and full load fuel consumption of 10 units per hour.
Using the MiniCodes, your job is to find: (1) the average wind power, (2) the useful
wind power, (3) the average dumped power, and (4) the average fuel use for the three
different turbine options, and for storage of 0 kWh, 100 kWh, and 1000 kWh.
What do the results suggest about the usefulness of storage?

SOLUTION

From the MiniCodes: (Powers in kW, Fuel usage in “units per hour”)

Storage 0 0 0 100 100 100 100 100 100

Turbines 1 2 4 1 2 4 1 2 4
P average 14.5 29 58.1 14.5 29 58.1 14.5 29 58.1
Useful 12.8 17.8 21.3 13.5 19.4 23.1 14.5 24.5 30.5
Dumped 1.7 11.2 36.8 1 9.6 35 0 4.5 27.6
Fuel use 6.7 5.2 4 6.2 4.3 3.1 5.8 3 1.3

Sample screen, for 1 turbine, 1000 kWh storage:

158 Wind Energy Explained

The results indicate that storage can have a significant effect on fuel usage when the wind
generation is high compared to the load, but not much when generation is low.

10.4 The 10 minute average wind speed at an offshore buoy is measured to be 8.5 m/s at
an elevation of 10 m. The buoy is 10 km from land in the direction of the oncoming wind.
a) If the surface roughness length is assumed to be 0.0002 m, what is the mean wind
speed at 80 m?
b) If the Charnock constant, AC, is assumed to be to be 0.018 and CD,10 is 0.0015, what
is the mean wind speed at 80 m, assuming that the friction velocity is given by
U * = U10 C D,10 ?
c) Assume that the water depth is 20 m, the wave height is 3 m, and the peak period of
the waves is 9 sec. What is the mean wind speed at 80 m in this case?

SOLUTION

U (z ) ln (z z 0 )
a) From Equation 2.34

=
U (z r ) ln(z r z 0 )

U (80 ) ln (80 0.0002)

= = 1.192
8.5 ln (10 0.0002)

So the wind speed at 80 meters is estimated to be 10.1 m/s.

b) From the given equation and Equation 10.2:

U * = U10 C D,10 = 8.5 0.0015 = 0.329

z 0 = AC = 0.018 = 0.00020
U *2 0.329 2
g 9.81

So the wind speed at 80 meters is estimated to be the same, 10.1 m/s.

( )
c) Use Equation 10.3:

z 0 = H s (1200 ) H s / LP 4.5

Where it may be assumed that the given wave height of 3 m is equal to the significant wave
height, Hs. The water depth, d = 20 m, is moderately shallow, so Equation 10.6 may be
used to estimate the wave length, LP, from the peak period, TP = 9 sec.

⎛ gT p2 ⎞ ⎛ 2 ⎞
LP = ⎜ ⎟ tanh⎜ 4π d ⎟
⎜ 2π ⎟ ⎜ T p2 g ⎟
⎝ ⎠ ⎝ ⎠
Problems 159

( )⎞⎟
The wavelength is then:

⎛ (9.81) 9 2 ⎛ 4π 2 (20 ) ⎞
LP = ⎜⎜
( )
⎜ ⎟
⎟ tanh ⎜ 9 2 (9.81) ⎟ = 110.2 m
⎝ 2π ⎠ ⎝ ⎠

( )
The roughness length is:

z 0 = H s (1200 ) H s / LP 4.5 = (3)(1200 )(3 / 110.2 )4.5 = 0.003266 m

ln (80 0.003266 )
The wind speed is:

U (80) = (8.5) = 10.2 m / s

ln (10 0.003266 )

Note: the wave length in part c) could also be determined from Equation 10.5. In that case,
LP = 105.2, z0 = 0.0004 m and U = 10.25 m/s.

10.5 Derive the equation for total inertial force on a monopile (see Equation 10.15). Note:

U& W (z ) =
π ( ( + )) cos(θ ) Assume that the depth is much greater than the wave
the equation for the acceleration as a function of depth is:
g H cosh k z d
L cosh kd
height.

SOLUTION

From the second term in Equation 10.14

πD &
FˆI = Cm ρAU& w = Cm ρ
2
Uw
4

Therefore:

πD 2 gπH cosh (k (z + d ))
FˆI = Cm ρ
4 L cosh (kd )

The total inertial force is found by integrating from the mudline to surface:

πD 2 gπH cosh(k (z + d ))
FI = ∫ Cm ρ
cosh(kd )
0

−d
dz
4 L

Let u =z + d, then dz =du, so

160 Wind Energy Explained

πD 2 gπH cosh(ku )
∫C ∫
πD 2 gπH
FI = mρ du = Cm ρ cosh (ku )du
L cosh(kd ) L cosh(kd ) 0
d 1 d

0 4 4

πD 2 gπH sinh (ku )

FI = C m ρ
L cosh (kd )
d
1
4 k
0

πD 2 gπH sinh (kd ) πD 2 gπH

FI = Cm ρ = Cm ρ tanh(kd )
4 kL cosh (kd ) 4 (2π / L )L

CmπD 2 ˆ
FI = ρ w g ζ tanh (kd ) q.e.d.
4

(Recall that the wave height is 2 x the amplitude.)

10.6 A wind turbine has rotor diameter of 90 m. It is installed on monopile with a diameter
of 4 m in diameter in water 15 m deep. The wind speed is 12 m/s , the wave height is 2 m
and the wave length is 100 m. Estimate the force due to the wind and the maximum force
due to waves. Use the Airy model for the waves and assume that the rotor thrust coefficient
is the one corresponding to the Betz power coefficient. Assume that inertia coefficient is
2.0 and the drag coefficient 1.5. (Assume that the maximum forces due to drag and inertia
occur at the same time.)

SOLUTION

The rotor force, FR, is:

U = 0.5 ⎜ ⎟(1.225)π
⎛8⎞
FR = 0.5 CT ρπ 12 = 2545 kN
D2 2 90 2 2
4 ⎝9⎠ 4

( )( )
The maximum inertial wave force with wavelength L = 100 m is:

C mπD 2 ˆ
FI = ρ w g ζ tanh (kd ) =
(1000)(9.81)(2) π 4 2 12 tanh⎛ 2π (10)⎞ = 181.6 kN
⎜ ⎟
4 4 ⎝ 100 ⎠
The maximum wave force due to drag is:

Cd D ˆ 2 ⎡ 1 ⎤
FD = ρ w g ζ ⎢ +
⎣ 2 sinh (2kd ) ⎦
⎥
kd

(1000)(9.81)(1.5)(4) (12 )⎡ 1 + (2π / 100)15

2
⎤
= = 17.0 kN
4 ⎢2
⎣ sinh ((2)((2π / 100 )15)) ⎥⎦
Problems 161

10.7 A small community has 2000 residents and each of them consumes an average of 100
liters of water per day. They have access to an aquifer 100 m below ground level and are
considering acquiring a wind electric water pump to bring the water to the surface. What
would be the rated power of wind turbine (kW) that would pump, on the average, an
amount of water equal to what the community uses? Assume that the capacity factor of the
wind turbine is 0.25 and that the efficiency of the water pump is 80%.

SOLUTION

Assuming 2000 people, consuming 100 liters/day person (0.1 m3/day person), the total
community water requirement is 200 m3/day.

( )( )
Assuming a water density of 1000 kg/m3, the average mass flow rate is:

m& = 200 m 3 / day 1000 kg / m 3 /[(24 hr / day )(3600 sec/ hr )] = 2.315 kg / s

( )
The power requirement is

Pwater = m& gh / η = (2.315 kg / s ) 9.81m / s 2 (100 m ) / 0.8 = 2839W

The rated power of the turbine would be:

PR = Pwater / cap. fac. = 2.839 kW / 0.25 = 11.35 kW

10.8 An island community has 2000 residents and each of them consumes an average of
100 liters of water per day. The community is considering acquiring a wind powered
desalination plant to produce pure water from sea water (density = 1020 kg/m3). The sea
water is 20° C and has salinity of 38 g/kg. What size wind turbine would produce, on the
average, an amount of fresh water equal to what the community uses? Assume that the
capacity factor of the wind turbine is 0.25, and that the desalination plant uses 5 times the
minimum theoretical amount of power.

SOLUTION

Since 20° C =293° K, the osmotic pressure is:

pπ = iSρ w Ru T / M
( )( )
= (1.8)(38 g / kg ) 1020 kg / m 3 8.31447 m 3 Pa / mol K (293 K ) / 58.5 g / kmol = 2,909 kPa

The minimum amount of power required is

Pmin = pπ / 3,600 = 2,909 kPa / 3,600 m 3 kPa / kWh = 0.807 kWh / m 3

(Note that 1 kPa = 1 kNm/m3 = 1 kJ/m3 =1 kWs/m3 = (1/3600) kWh/m3.)

162 Wind Energy Explained

Assuming 4 times theoretical, the power consumption is 4.04 kWh/m3.

Assuming 2000 people, consuming 0.1 m3/day person, the total community water
requirement is 200 m3/day. The average power requirement is then 807 kWh/day or 33.6
kW.

The wind turbine rated power would be PR = Pav/cap fac = 33.6/.25 = 134.5 kW. It may be
noted that the required turbine power required to desalinate water is considerably greater
than that required to pump the equivalent amount of water in Problem 10.7.

10.9 A lead acid battery has a nominal capacity of 200 Ahrs. The constant c is equal to 0.5
and the rate constant k is equal to 2.0. Find apparent capacity of the battery (charge
removed) and corresponding current when it is discharged in (i) 1 hour and (ii) 10 hours.

SOLUTION

Use Equation 10.30:

qmax ( I ) =
( )
qmax k ct
1 − e −k t + c k t − 1 + e −k t

Making the appropriate substitutions, one has:

Discharge time, hrs Capacity, Ah

1 139.6
10 190.4

10.10 A bank of 12 lead acid batteries in series is being used to supply a resistive load with
nominal power consumption of 1200 W at 120 V. The voltage constants for each battery
(in discharging) are E0 = 12 V, A = -0.05, C = -0.5, D = 1. The internal resistance is Rint =
0.1 Ω. At a certain point the batteries are 70% discharged. Find the terminal voltage of the
batteries and the actual power consumption at this point.

SOLUTION

When the batteries are 70% discharged, X = 0.7

The internal voltage is:

E = E0 + AX + CX / (D − X ) = 12 + (− 0.05)(0.7 ) + (− 0.5)(0.7 ) / (1 − 0.7 ) = 10.8V

The resistance of the electrical load is found from:

Rload = = = 12 Ω
2
VLoad ,nom 120 2
Pnom 1200
Problems 163

Find the current such that the voltage is equal for batteries and load.

The battery voltage is

Vbat = N bat (E − IRint )

The load voltage is

Vload = IRload

The two equations may be solved:

I = N bat E / (Rload + N bat Rint )

Substituting the values:

I = (12 )(10.8) / (12 + (12 )(0.1)) = 9.818 A

( )
The power is then

P = I 2 Rload = 9.818 2 (12 ) = 1157 W

Note that the actual power delivered is less than the nominal value.

10.11 A natural gas fired turbine operating on an ideal Brayton cycle is used with a wind
powered compressed air storage facility. The cycle has the following characteristics: the
temperature at the input to the compressor is 27° C and at the input to the turbine the
temperature is 727° C. The maximum pressure is 2 MPa and the minimum pressure is 100
kPa. (a) Find the heat input from the natural gas if the net output of the gas turbine is 1
MW and the compressed air from the wind turbine is not being used. (b) Now assume that
the gas turbine’s compressor is not being used, but the compressed air is coming from the
compressed air storage. The compressed air storage is being continually recharged and the
pressure remains constant at 2 MPa. All other things being the same, what would be the
net power output of gas turbine now? Assume that the constant pressure heat capacity of
air is 1.005 kJ/kg K and that the ratio of cp to cv (i.e. k) is equal to 1.4.

SOLUTION

Assume that this is an ideal Brayton cycle (as stated in the corrected problem statement).
Find the temperatures T2 and T4, using the temperature/pressure relation for isentropic
compression or expansion, such as is given by Equation 10.36:

( k −1) / k
⎛P ⎞ ⎛ 2000 ⎞
T2 = T1 ⎜⎜ 2 ⎟⎟ = 300⎜ ⎟ = 706 K
0. 4 / 1. 4

⎝ P1 ⎠ ⎝ 100 ⎠
164 Wind Energy Explained

( k −1) / k
⎛P ⎞ ⎛ 100 ⎞
T4 = T3 ⎜⎜ 4 ⎟⎟ = 1000⎜ ⎟ = 424.9 K
0.4 / 1.4

⎝ P3 ⎠ ⎝ 2000 ⎠

Using Equation 10.37, the compressor specific work is:

wcomp,in = c p (T2 − T1 ) = (1.005 kJ / kg K )(706 K − 300 K ) = 408.1 kJ / kg

Similarly, the turbine specific work is:

wturb,out = c p (T3 − T4 ) = (1.005 kJ / kg K )(1000 K − 424.9 K ) = 578 kJ / kg

The net specific work is

wnet = wturb ,out − wcomp,in = 578 kJ / kg − 408.1kJ / kg = 169.9 kJ / kg

The specific heat input is:

qin = c p (T3 − T2 ) = (1.005 kJ / kg K )(1000 K − 706 K ) = 295.4 kJ / kg

The efficiency is:

η = wnet / qin = 169.9 kJ / kg / 295.4 kJ / kg = 0.575

The total required heat is:

Qin = Wout /η = 1000 / 0.575 = 1738.8 kJ/s (kW)

(b) If the air came from the storage, there would be no compressor work, so the net specific
work would be equal to the turbine work (578 kJ/kg). The total net output would be
proportionally higher:

Wout ,CAES = (Wout ) = ⎛⎜ 578 ⎞

⎟(1000) = 3402 kW
wturb,out
wnet ⎝ 169.9 ⎠

10.12 An island community has 2000 residents, each of whom uses, on the average, 8
kWh/day. The electricity is presently produced by a diesel generator, but they are
considering acquiring a wind turbine. They are would like to incorporate pumped water
storage in their plans. An open spot on top of a hill in could be dug out to form a circular
reservoir. The reservoir could be 30 m in diameter and 5 m deep. The hill is 70 m above
the town. A reservoir of similar size could be next at the same elevation as the town. How
many hours could the reservoir supply the average electrical load of the community?
Problems 165

SOLUTION
The solution involves applying Equation 10.41. The inputs and results are summarized in
the table below.
Enegy per capita 8 kWh/person day
Population 2000 people
Av power required 666.67 kW

Reservoir diameter 30 m
Depth 5m
Volume 3,534 m^3

Head 70 m
Water density 1000 kg/m^3

Energy 2,426,998 kJ
Energy 674 kWh

Hrs 1.011

10.13 The famous dirigible Hindenburg had a volume of approximately 199,000 m3. It
was filled with hydrogen to give it buoyancy. Assume that the pressure inside was 100 kPa
and that the temperature was 15° C. How much hydrogen did the Hindenburg contain?
How much energy (kWh) would it take to produce that hydrogen by electrolysis of water,
assuming an electrolyzer efficiency of 65%? How long would it take to produce that
hydrogen using 1.5 MW turbine, operating at full power?

SOLUTION

Use the ideal gas law to find the mass of hydrogen. First convert the temperature, 15° C to

( )
288° Kelvin.

PVM (100 kPa ) 199,000 m 3 (2.016 kg / kmol )

m=
Ru T
=
( )
8.31447 m 3 kPa / kmol K (288 K )
= 16,754 kg

The energy, E, required to produce that amount of hydrogen would be:

E = (16,754 kg )(39.4 kWh / kg ) / 0.65 = 1,016,000 kWh

The time, t, that the turbine would need to run at full output would be:

t = 1500 kW / 1,016,000 kWh = 677 hr

166 Wind Energy Explained

10.14 A hydrogen powered bus has four roof top mounted storage cylinders with a total
capacity of 1.28 m3. When full the pressure is 45 MPa. How many times could the
cylinders be refilled with the hydrogen that would fill the Hindenburg, as found in Problem
10.13? Assume that the cylinders are completely emptied before being refilled and assume
that the ideal gas law is applicable.

SOLUTION

The volume, V45 Mpa, required to hold the hydrogen at 45 MPa, using the ideal gas law,

( )
would be given by the inverse ratio of the pressures times the volume of the Hindenburg:

V45 MPa = 199,000 m 3 = 422.2 m 3

100 kPa
45,000 kPa

The number of times, N, that the top mounted storage cylinders could be filled is then:

N= = 345.5
422.2 m 3
1.28 m 3

B.11 Chapter 11 Problems

11.1 The first unit of a new wind turbine costs $100,000. The system is estimated to follow
an s = 0.83 learning curve. What is the cost of the 100th unit?

SOLUTION

This problem solution is based on the application of the learning curve equation (11.1).
This is expressed as:

C(V) ⎛ V ⎞
=⎜ ⎟
b

C( V0 ) ⎜⎝ V0 ⎟⎠
b
The progress ratio, s, is given by s = 2 . For this example, s = 0.83, and thus b = -0.2688

The cost for the 100th unit, C(V) is then:

⎛V ⎞ − 0.2688
⎛ 100 ⎞
C(V) = C( V0 ) ⎜⎜ ⎟⎟ = $100,000⎜ = $29,000
b
⎟
⎝ V0 ⎠ ⎝ 1 ⎠
Problems 167

11.2 Estimate the progress ratio (s) of modern utility-scale wind turbines. Use the time
period of about 1980 to the present.

SOLUTION

The solution to this problem will require some literature research. A good reference on the
subject is given by L. Neij (‘Cost Dynamics of Wind Power’, Energy, 24, 1999, p. 375-
389). In this paper, the author estimates a value of s between 0.92 and 0.98, depending on
various turbine size and manufacturer assumptions.

11.3 A small 50 kW wind turbine with an initial cost of $50,000 is installed in Nebraska.
The fixed cost ratio is 15% and the annual operation and maintenance (CO&M) is 2% of the
initial cost. This system produces 65,000 kWh/yr (AkWh). Determine the cost of energy
(COE) for this system.

SOLUTION

The simple equation for the cost of energy (COE) is:

COE = [(Cc x FCR) + CO&M]/Ea

Thus, for this example: COE = (($50,000 x 0.15) + ((0.02) x $50,000) / (65,000) =
$0.131/kwh

11.4 A small wind machine, with a diameter of 6 m, is rated at 4 kW in an 11 m/s wind

speed. The installed cost is $10,000. Assuming Weibull parameters at the site to be c = 8
m/s and k = 2.2, it is claimed that the capacity factor is 0.38.
The interest rate is 11% and the period of the loan is 15 years. Find the cost per unit
area, per kW, and the levelized cost per kWh. You do not have to include factors such as
inflation, tax credits, or operation and maintenance costs.

SOLUTION

a) The cost per unit area is: Rated power cost/rotor area = $354/m2
b) The cost per kW is: $10,000/ 4 kW = $2500/kW
c) The levelized cost of energy, COEL, is given from Equation 11.21:

COE L =
(NPVC )(CRF)
Annual energy production

Use Equation 11.12 for the capital recovery factor:

⎧⎪r / [1 − (1+r ) − N ], if r ≠ 0
CRF = ⎨
⎪⎩1 / N , if r = 0
168 Wind Energy Explained

The capital recovery factor (CFR) for use in the COE equation is based on the discount
rate, r, and 20 year projected life, L, and is found to be 0.1175. The capital recovery factor
for the loan payments, CRFP, is found by using the loan interest rate, b, and 15 year loan
period, L, is found in a similar manner and is equal to 0.1391. The annual payment is:
Annual payment = (CRFP) (Cost) = $1391.

The net present value of costs, NPVC , is given by Equation 11.18:

⎛ 1 ⎞ ⎛ 1+i ⎞
NPVC = Pd + PaY ⎜⎜ , N ⎟⎟ +Cc f OM Y ⎜⎜ , L ⎟⎟
⎝ 1+r ⎠ ⎝ 1+r ⎠
The Y() term is evaluated according to Equation 11.19:
⎧ k − k l +1
Y (k , l ) = ∑ k = ⎨ 1 − k
l ⎪ , if k ≠ 1
⎪l, if k = 1
j
j =1
⎩

In this case, a number of terms are equal to zero. Thus:

⎛ 1 ⎞
−⎜ ⎟
16
⎛ 1 ⎞
1
1.10 ⎝ 1.10 ⎠
NPVC = PaY ⎜⎜ , N ⎟⎟ = = ($1391) (7.61) = $10,585
⎝ 1+ r ⎠ 1−
1
1.10
Assuming the capacity factor of 0.38 is correct (one could check this out given the Weibull
parameters of the problem), the annual energy production is given by:

Ea = (0.38)(4)(8760) kWh/yr = 13,315 kWh/yr

Thus, the levelized cost of energy is given by:

COEL = ($10,585)(0.1175)/(13,315) = $0.093/kWh

11.5 The estimated cost and savings from the purchase and operation of a wind machine
over 20 years are given in Table B.12. a) Compare the total costs with the total savings. b)
Find the present values of the cost and savings. Use an annual discount rate of 10%. Would
this be a good investment? c) Determine the pay back period for the wind machine. d)
What is the breakeven cost of the machine? e) Estimate the cost of energy produced by the
wind machine if it is expected to produce 9,000 kWh annually.
Table B.12 Annual wind turbine costs

Year Costs Savings Year Costs Savings

0 $9000 11 227 669
Problems 169

1 127 2324 12 241 719

2 134 1673 13 255 773
3 142 375 14 271 831
4 151 403 15 287 894
5 160 433 16 304 961
6 168 466 17 322 1033
7 180 501 18 342 1110
8 191 539 19 362 1194
9 202 579 20 251 1283
10 214 622

SOLUTION

A spreadsheet tabulation of the costs and savings for this system is given below. From this
spreadsheet, the following answers can be determined:
(a) The total costs are $13,531 and the total savings are $17,382. However, this does
not tell us if this is a good investment. That must be determined from the net present value.
(b) Using a discount rate of 10%, the following are determined:
Total PVcosts = $10,638.00
Total PVsavings = $7,967.60
NPV = $10,638.00 - $7,697.60 = - $2940.40
Therefore, this may not be a good investment since the NPV is negative.
(c) The cumulative savings first exceeds the costs in the 16th year, so this is one
measure of payback.
(d) The breakeven cost is equal to the total present value of the savings. Therefore, the
breakeven costs = $7,697.60.
(e) Annual energy production = 9000 kWh. Use Equation 11.21:

COE L =
(NPVC )(CRF)
Annual energy production

The net present value of the costs, found above, are $10,638. The CRF in this case is found
from Equation 11.12 using L= 20 and r = 0.1:

CRF = r / [1 − (1+r ) −L ] = 0.1 /[1 − (1.1) −20 ] = 0.11746

Therefore:

COE L = = $0.14 / kWh

( $10,638 )( 0.11746 )
9000 kWh
170 Wind Energy Explained

SPEADSHEET RESULTS

Present Present
Annual Cumulative Cumulative Discount Value Value
Year Costs Savings Costs Savings Factor Costs Savings
0 9000 0 9000 0 1 9000 0
1 127 2324 9127 2324 0.909 115.45 2112.73
2 134 1673 9261 3997 0.826 110.74 1382.64
3 142 375 9403 4372 0.751 106.69 281.74
4 151 403 9554 4775 0.683 103.14 275.25
5 160 433 9714 5208 0.621 99.35 268.86
6 168 466 9882 5674 0.564 94.83 263.04
7 180 501 10062 6175 0.513 92.37 257.09
8 191 539 10253 6714 0.467 89.1 251.45
9 202 579 10455 7293 0.424 85.67 245.55
10 214 622 10669 7915 0.386 82.51 239.81
11 227 669 10896 8534 0.35 79.56 234.48
12 241 719 11137 9303 0.319 76.79 229.1
13 255 773 11392 10076 0.29 73.86 223.91
14 271 831 11663 10907 0.263 71.36 218.83
15 287 894 11950 11801 0.239 68.71 214.02
16 304 961 12254 12762 0.218 66.16 209.14
17 322 1033 12576 13795 0.198 63.71 204.37
18 342 1110 12918 14905 0.18 61.51 199.64
19 362 1194 13280 16099 0.164 59.19 195.23
20 251 1283 13531 17382 0.149 37.34 190.71

11.6 Consider the application of a small 1 kW wind turbine with a capital cost of $2500.
The installation and setup cost raises its total installed cost to $4500. Assume that the
$2500 capital cost is to be paid for with a 15 year, 7% loan. Also assume that O&M costs
will be $200 per year. Estimate the (simplified) cost of energy) over the 15 year period if
the capacity factor (CF) is 0.30.

SOLUTION

The answer to this problem is based on the use of Equation 11.7 The operating cost of this
wind turbine is:

Operating cost = Annual payments on loan + O&M costs

For a 15 year loan at 7% interest, the annual payment is given by:

Problems 171

Annual Payment = Loan x Capital recovery factor (CRF)

CRF is calculated via Equation 11.12 and equals 0.1098

The annual payment is $494.10 and the total operating cost is $694.10

The average energy delivered per year is: CF x Rated Power x 8760 (kWh/yr)

= 0.30 x 1 kW x 8760 hr/yr = 2628 kWh/yr

Thus, COE = $694.10/ 2628 = $0.264/ kWh

11.7 A 2.4 MW wind turbine has an installed cost of $2.4 x 106. Assume that the annual
operating and maintenance costs are 2% of the initial installed cost. The wind regime at this
site results in a capacity factor (CF) of 0.35. It is also assumed that the lifetime of this
turbine is 25 years and that the electrical output can be sold for $0.05/ kWh. For a 5%
discount rate calculate the following economic parameters for this system:

a) Simple payback period

b) The net present value of the savings (NPVS)
c) The benefit-cost ratio (B/C)
d) The internal rate of return (IRR)
e) The levelized cost of energy (COEL)

SOLUTION

a) The simple payback period is calculated via the use of Equation 11.3 The annual energy
production is equal to 0.35 x 8760 x 2400 = 7.35 x 106 kWh/yr.
The average annual return, AAR, is equal to 7.35 x 106 x 0.05 = $3.68 x 105/ yr.

Thus, the simple payback period is $2.4 x 106 / $3.68 x 105 = 6.52 yr.

b) The net present value (NPV) of the savings is equal to the net present cost of the system
subtracted from the net present value of the income from the wind turbine.

The NPVincome is found from Equations 11.14 and 11.19, using discount rate, d = 0.05, and
lifetime, L = 25, as:
⎛ ⎛ 1 ⎞ ⎛ 1 ⎞ L+1 ⎞
⎜⎜ ⎟−⎜ ⎟ ⎟
= ( AAR )Y ⎢ 5⎜ ⎝ 1+ d ⎠ ⎝ 1+ d ⎠
⎡ 1 ⎤ ⎟
, L ⎥ = $3.68 x10 ⎜ ⎟
⎣1 + d ⎦ ⎜ ⎛ 1 ⎞ ⎟
NPVincome
1− ⎜ ⎟
⎜ ⎝ 1+ d ⎠ ⎟

( )
⎝ ⎠
= 3.68 x10 5 (14.09 ) = $5.185 x10 6
172 Wind Energy Explained

The net present value of the cost of the system equals the net present value of the capital
cost ($2.4 x 106) plus the net present value of the O&M costs. The NPV of the O&M costs
are calculated via Equations 10.18 and 10.19 as:

NPVO&M = ($2.4 x 106) x 0.02 x 14.09 =$6.76 x 105

Thus, NPVcost = $2.4 x 106 + $6.76 x 105 = $3.076 x 106

And, NPVsavings = $5.185 x 106 - $3.076 x 106 = $2.11 x 106

c) The Benefit-Cost Ratio (BCR) is found from Equation 11.23

BCR = $5.185 x 106 / $3.076 x 106 = 1.69

d) The internal rate of return (IRR) is defined by Equation 11.22

By trial and error, the answer works out to be about 14%

e) The levelized cost of energy is given by Equation 11.20.

COEL = $3.076 x 106 x .07095 / 7.35 x 106 = $0.0296 /kWh

11.8 A proposed wind farm installation is based on the use of thirty 2 MW wind turbines.
The installed capital cost is estimated at $1000/kW and the annual operation and
maintenance costs are estimated to be 3% of the original capital cost. Financing for the
project is to come from two sources: 1) A 20 year loan at 7% (75%), and 2) an equity
investment that has a return of 15% (25%). The average capacity factor of the wind farm is
estimated to be 0.35. Determine the estimated cost of energy ($/kWh) from this proposed
wind farm.

SOLUTION

The solution to this problem will be based on the use of Equation 11.20 thus we need to
calculate all the annual costs plus the annual electrical production.

For the loan ($45 x 106) the annual payment is found using Equation 11.12 and is equal to
$4.25 x106/yr.

The return on the equity ($15 x 106) is equal to 15 x 106 x 0.15 = $2.25 x 106/yr

Annual O&M costs are equal to 0.03 x $60 x 106 = $1.8 x 106

Thus, the total annual costs are $(4.25 + 2.25 + 1.8) x 106 = $8.3 x 106.

The annual energy production is equal to: 30 x 2 x 1000 x 8760 x 0.35 kWh/yr =1.84 x 108
kWh/yr, so the cost of energy is calculated as $8.3 x 106/ 1.84 x 108 = $0.045/ kWh
Problems 173

11.9 This problem is intended to assess the relative economic merits of installing a wind–
diesel system rather than a diesel-only system at a hypothetical location. Assume that the
existing diesel is due for replacement and that the annual system demand will remain fixed
during the lifetime of the project.
Your problem is to calculate the levelized cost of energy for the diesel only and for the
wind–diesel system. The following system and economic parameters hold:

System
Average system load 96 kW (841,000 kWh/yr)
Diesel rated power 200 kW
Mean diesel usage- no wind turbine generator 42.4 l/hr (371,000l/yr)
Wind turbine rated power 100 kW
Average annual wind speed 7.54 m/s
Annual wind speed variability 0.47
Average wind turbine power 30.9 kW
Useful wind turbine power (based on 21.1 kW
25% lower limit on diesel loading)
Mean diesel usage with WTG 34.8 l/hr (305,000 l/yr)
Economics
Wind turbine cost $1500/kW installed
Diesel cost $3500/kW installed
System life 20 yr
Initial payment $30,000
Loan term 10 yr
Interest 10%
General inflation 5%
Fuel inflation 6%
Discount rate 9%
Diesel fuel cost $0.50/l
Wind turbine O&M cost 2% of capital cost/yr
Diesel O&M cost 5% of capital cost/yr

SOLUTION:

This problem is discussed in Chapter 8 of Wind Diesel Systems by Hunter and Elliot
(1994). In order to determine the levelized cost of energy, the net present value (or here,
net present costs) of installing the different energy supply systems must be calculated.
The net present value equation for a wind/diesel system can be found by adding fuel costs
to Equation 11.18:

⎛ 1 ⎞ ⎛ 1+i ⎞ ⎛ 1+e ⎞
NPVC = Pd + PaY ⎜⎜ , N ⎟⎟ +Cc f OM Y ⎜⎜ , L ⎟⎟ + ( FL)( FC )Y ⎜⎜ , L ⎟⎟
⎝ 1+r ⎠ ⎝ 1+r ⎠ ⎝ 1+r ⎠
where:
FL = Fuel consumption, units/yr
FC = Fuel cost, $/unit
174 Wind Energy Explained

e = energy inflation rate

The following Series Present Worth and Capital Recovery Factors may be calculated first
to facilitate solving the problem:

Series Present worth factor for interest payments, SWWFP:

⎡ 1 ⎤
Y⎢ , N ⎥ = 6.418
⎣1 + r ⎦
Series Present worth factor for fuel costs, SPWFF
⎡1 + e ⎤
Y⎢ , L ⎥ = 15.144
⎣1 + r ⎦
Series Present worth factor for O&M costs, SWWFO
⎡1+ i ⎤
Y⎢ , L ⎥ = 13.822
⎣1 + r ⎦
Capital Recovery Factor for interest payments, CRFP

⎡ 1 ⎤
1/ Y ⎢ , N ⎥ = 0.1628
⎣1 + b ⎦
Capital Recovery Factor for system income, CRFI
⎡ 1 ⎤
1/ Y ⎢ , L ⎥ = 0.1095
⎣ 1 + r ⎦
Now, the Net Present Values can be determined using the form of Equation 11.18 as given
above. The following values result:

Parameter Equation Diesel Only WTG and Diesel

Capital Cost Cc $70,000 $220,000
Initial Payment on System Pd $30,000 $30,000
Loan Cc - Pd $40,000 $190,000
Annual Payment (Ap) (Cc - Pd) CRFP $6510 $30,922
NPV of Annual Payments Ap x SPWFP $41,779 $198,448
Fuel Costs per year (year 1) FL x FC $185,500 $152,500
NPV of fuel consumed over FL x FC x SPWFF $2,803,647 $2,593,176

ΣCc x OM
system lifetime
O&M costs (year 1) OMC $3500 $6500
NPV of O&M costs over OMC x SPWFO $48,377 $89,843
system lifetime
Total NPV of system costs, TNPV $2,923,803 $2,623,176
Levelized cost of energy, (TNPV)(CRFI)/SL $0.38/kWh $0.34/kWh
Problems 175

B.12 Chapter 12 Problems

12.1 Despite its positive environmental attributes, not everyone is in favor of wind power
development. Your assignment here is to perform an internet search and summarize the
view points of three different organizations that actively oppose wind projects in one way
or another. Make sure that you list their objections. You should also comment on their
technical validity.

SOLUTION

As a start one could look at the web sites of the following organizations:

1) European Platform Against Windfarms: [Link]

2) Wind Turbine Syndrome: [Link]

3) Virginia Wind: [Link]

4) National Wind Watch: [Link]

5) Country Guardian: [Link]

12.2 Assume that you are planning the development of a small wind farm (say 10 turbines
of 500 kW) in a rural area. Describe how you would reduce the visual impact of this
project. Also prepare a list of the tools that could be used for assessing the impact of this
project on the surrounding environment.

SOLUTION

This is a literature review problem. Good sources here are references dealing with visual
impact assessment, especially ones that give some type of visual simulation. A search of
the internet should yield descriptions of available computer software programs that
accomplish this.

12.3 Researchers from Spain (see Hurtado et al. (2004) Spanish method of visual impact
evaluation in wind farms. Renewable and Sustainable Energy Reviews, 8, pp. 483-491.)
have developed a method that gives a numerical value for a factor that represents a
coefficient of the affected population.

Review this paper and give an evaluation of the use of this methodology

SOLUTION

Once one finds and reads this paper, the solution is left to the originality and thoughts of
the participating student.
176 Wind Energy Explained

12.4 Estimate the sound power level for a three-bladed, upwind wind turbine with the
following specifications: rated power = 500 kW, rotor diameter = 40 m, rotor rotational
speed = 40 rpm, wind speed = 12 m/s.

SOLUTION

This solution is based on solution of three different sound power level prediction equations
in the text:

LWA = 10 (log10 PWT ) + 50 (12.9)

LWA = 22 (log10 D ) + 72 (12.10)

LWA = 50( log10VTip ) +10 (log10 D) − 4 (12.11)

From the 1st equation:

LWA = 10 (log10 (500,000) + 50 = 56.99 + 50 = 106.99 dB

From the 2nd equation:

LWA = 22 (log10 (40) + 72= 35.25 + 72 = 107.25 dB

Using the 3rd equation:

Vtip = (2π)(20) (40/60) = 83.78 m/s

LWA = 50 log10 (83.78) +10 log10 (40) − 4 = 96.16 + 16 − 4 = 108.18 dB

12.5 Suppose the sound pressure level at a distance of 100 m from a single wind turbine is
60 dB. What are the sound pressure levels at a distance of 300, 500, and 1000 m?

SOLUTION

The solution is based on the use of Equation 12.14 from the text:

L p = Lw −10 log10 ( 2πR 2 )−αR

The MiniCodes give a Sound Power Level = 108.48 dB

Sound Pressure = 49.5 at 300 m
Sound Pressure = 44 at 500 m
Sound Pressure = 35.5 at 1000 m
Problems 177

12.6 Assuming the same turbine as in the previous problem, what are the sound pressure
levels at a distance of 300, 500, and 1000 m for 2, 5, and 10 turbines?

SOLUTION

Use Equation 12.13

Ltotal = 10 log10 ∑10

N
Li / 10

i =1

2 turbines
Sound Pressure = 52.5 at 300 m
Sound Pressure = 47 at 500 m
Sound Pressure = 38.5 at 1000 m

5 turbines
Sound Pressure = 56.4 at 300 m
Sound Pressure = 51 at 500 m
Sound Pressure = 42.5 at 1000 m

10 turbines
Sound Pressure = 59.5 at 300 m
Sound Pressure = 54 at 500 m
Sound Pressure = 45.5 at 1000 m

12.7 Two wind turbines emitting 105 dB(A) at the source are located 200 m and 240 m
away from a location of interest. Calculate the sound pressure level (in dB(A)) at the point
of interest from the combined acoustic effect of the two turbines. Assume a sound
absorption coefficient of 0.005 dB(A)/m.

SOLUTION

For the first turbine, the sound pressure level is calculated from Equation 12.14 as

(Lp)1 = 105 – 10 log10(2 x π x 2002) – 0.005x200 = 50 dB(A)

Similarly for the second turbine

(Lp)2 = 105 – 10 log10(2 x π x 2402) – 0.005x200 = 48 dB(A)

The sound pressure level from the two turbines is then calculated from Equation 12.12 as

Ltotal = 10 log10 (10 L1/10 + 10 L2/10) = 52.12 dB(A)

12.8 A wind turbine has a measured sound power level of 102 dB at 8 m/s. Plot the sound
pressure level as a function of distance using both hemispherical and spherical noise
178 Wind Energy Explained

propagation models. Do this for sound absorption coefficients of 0.0025, 0.005, and 0.010
(dB(A) m-1).

SOLUTION

The sound level prediction equation for hemispherical noise propagation is given in the
Text as Equation 12.14. A little research is needed to find the sound prediction equation
for a spherical noise prediction model (e.g., see [Link]
techguides/wtnm/).

It is similar to Equation 12.14 and is:

Lp = Lw -10 log10 (4πR2) – αR

The graphical results are shown below for the three values of sound absorption coefficient.

Hemispherical Noise Propagation

80

70

60

alpha=0.005
50 alpha=0.0025
alpha=0.010

40

30

20
0 100 200 300 400 500 600 700 800 900 1000
Distance (m)
Problems 179

Spherical Noise Propagation

80

70

60

alpha=0.005
50 alpha=0.0025
alpha=0.010

40

30

20
0 100 200 300 400 500 600 700 800 900 1000
Distance (m)

12.9 For electromagnetic interference problems once the (bistatic) radar cross-section (σb)

For a TV transmitter operating in an urban environment, σb was experimentally determined

is known, the polar coordinate r can be calculated for a required interference ratio (C/I).

to be 24 and 46.5 dBm2 in the backscatter and forward scatter regions. Determine the value
of r in both the backscatter and forward scatter regions for required values of C/I equal to
39, 33, 27, and 20 dB.

SOLUTION

This solution is based on experimental and analytical work summarized in the paper of Van
Kats and Van Rees (1989). The solution is based on the rewriting of Equation 12.19 in the
following manner:

20 log(r ) = C / I − 10 log(4π ) + 10 log(σ b ) − A2 + A1 − Ar − Δ G

For a worst case approximation, ΔG = 0 (no discrimination of the receiving antenna), and
when A1 = A2 = Ar = 0 , r can be calculated from:
20 log(r ) = C / I − 10 log(4π ) + 10 log(σ b )
= C / I − 11 + 10 log(σ b )

For the backscattering region, σ b = 24 dB , thus:

20 log(r ) = C / I + 13
180 Wind Energy Explained

This gives the following results:

C/I r (m)
39 398
33 200
27 100
20 45

For the forward scattering region, σ b = 46.5 dB , thus:

20 log(r ) = C / I + 35

This gives the following results:

C/I r (m)
39 5308
33 2660
27 1333
20 595
Problems 181

Lists of Files Used in Problems or Solutions

A summary table of the data input files is provided below.

Chapter Files for Problems Problem #

2
[Link] 2.11
[Link] 2.14
[Link] 2.15

4
[Link] 4.12

7
[Link] 7.7
[Link] 7.8

8
[Link] 8.13
[Link] 8.13
[Link] 8.13

9
Site1 v Site2 For [Link] 9.7
[Link] 9.11

10
AOC_pc.csv 10.13
AOC_pc_x2.csv 10.13
AOC_pc_x4.csv 10.13
Load_TI.txt 10.13
Wind_TI.txt 10.13
182 Wind Energy Explained

A summary table of the solution files is provided below.

Chapter Solution Files for Problems Problem #

3
VAWT_High Tipspeed Approximations_Solution.xls 3.13
VAWT_Lambda_a_Solution.xls 3.13

4
tower straight_2.csv 4.13
[Link] 4.13

7
Gumbel [Link] 7.2

8
[Link] 8.13
[Link] 8.14

9
Site1 v Site2 For MCP_Years 1999, 2001, 2002 [Link] 9.7
[Link] 9.9
[Link] 9.10
[Link] 9.11