DRPT2008 6-9 April 2008 Nanjing China

Pitch Angle Control for Variable Speed Wind

Turbines
Jianzhong Zhang, Ming Cheng, Zhe Chen, Xiaofan Fu

Abstract--Pitch angle control is the most common means for

adjusting the aerodynamic torque of the wind turbine when wind
speed is above rated speed and various controlling variables may
be chosen, such as wind speed, generator speed and generator
power. As conventional pitch control usually use PI controller, the
mathematical model of the system should be known well. A fuzzy
logic pitch angle controller is developed in this paper, in which it
does not need well known about the system and the mean wind
speed is used to compensate the non-linear sensitivity. The fuzzy
logic control strategy may have the potential when the system
contains strong non-linearity, such as wind turbulence is strong, Fig. 1 Structure of a model of variable-speed wind turbine
or the control objectives include fatigue loads. The design of the
fuzzy logic controller and the comparisons with conversional can have a dramatic effect on the power output. The purpose of
pitch angle control strategies with various controlling variables the pitch angle control might be expressed as follows [1-3]:
are carried out. The simulation shows that the fuzzy logic
controller can achieve better control performances than 1) Optimizing the power output of the wind turbine.
conventional pitch angle control strategies, namely lower fatigue Below rated wind speed, the pitch setting should be at
loads, lower power peak and lower torque peak.
its optimum value to give maximum power.
Index Terms--Blade aerodynamic, Fatigue load, Fuzzy logic 2) Preventing input mechanical power to exceed the
control, Pitch angle, Wind turbine design limits. Above rated wind speed, pitch angle
control provides a very effective means of regulating
the aerodynamic power and loads produced by the
I. INTRODUCTION rotor.
3) Minimizing fatigue loads of the turbine mechanical
R ECENT years, a fast growing in wind energy is
experienced in whole world. At the end of 2006, the
installed capacity of wind farms in China reaches 2600 MW,
component. It is clear that the action of the control
system can have a major impact on the loads
which is increased by 105% compared with 2005. Wind power experienced by the turbine. The design of the controller
may become Chinas third major power supply by 2020, with must take into account the effect on loads, and the
an expected installed capacity of 40 GW. controller should ensure that excessive loads will not
Pitch-adjusting variable-speed wind turbines have become result from the control action. It is possible to go further
the dominating type of yearly installed wind turbines in recent than this, and explicitly design the controller with the
years. There are usually two controllers for the variable-speed reduction of certain fatigue loads as an additional
wind turbines which are cross-coupled each other, shown as in objective.
Fig. 1. In low wind speed below rated value, the speed The rain-flow counting method is widely used in the analysis
controller can continuously adjust the speed of the rotor to of fatigue data which can assess the fatigue life of a structure
maintain the tip speed ratio constant at the level which gives subject to complex loading. From the view of the cycle
the maximum power coefficient, and then the efficiency of the counting method during a given time history, simple
turbine will be significantly increased. Pitch angle regulation is
comparisons of the fatigue life for the turbine components can
required in conditions above the rated wind speed when the be carried out.
rotational speed is kept constant. Small changes in pitch angle In this paper, conventional pitch angle control strategy, in
which various controlling variables may be used, is discussed.
As the conventional pitch control strategy usually uses
This work was supported in part by NSFC under Project 50777008, the proportional and integral (PI) controller, the knowledge of the
Jiangsu High-Tech. Project under Grant BG2005035 and Graduate Student system dynamic is need [4-7]. However, fuzzy logic controller
Innovation Plan of Jiangsu Province, China.
Jianzhong Zhang, Ming Cheng and Xiaofan Fu are with the school of does not need a well known system and is potential for the
electrical engineering, Southeast University, Nanjing China. Email: system contains strong non-linearity. The fuzzy logic control
seuzjz@seu.edu.cn, mcheng@seu.edu.cn. for wind turbines has been proposed in some papers [8-10].
Zhe Chen is with the institute of energy technology, Aalborg University,
Aalborg Denmark. Email: zch@iet.aau.dk.
Refs. [8-9] propose a fuzzy logic controller for maximum

978-7-900714-13-8/08/ 2008DRPT 2691

 DRPT2008 6-9 April 2008 Nanjing China

power point tracking (MPPT) control and Ref. [10] proposes a speed.
fuzzy logic pitch angle controller for power system
stabilization. In this paper, on the basis of previous three
purposes, a fuzzy logic pitch angle controller is designed and
analyzed. The simulations and comparisons of the different
pitch angle control strategies are carried out and the
conclusions are drawn at last.

II. WIND TURBINE MODELING

A. Power Capture of Wind Turbines

The power in the wind is proportional to the cube of the wind
speed and may be expressed as
P = 0.5 A w3 (1)
Fig. 3 Mechanical power versus rotor speed characteristics
where is air density, A is the area swept by blades and w is
wind speed. A wind turbine can only extract part of the power The typical power control regions of wind turbine are shown
from the wind, which is limited by the Betz limit (maximum in Fig. 4. The turbine starts operating when the wind speed
59%). This fraction is described by the power coefficient of the exceeds cut-in wind speed. The power captured by the turbine
turbine, Cp, which is a function of the blade pitch angle and the increases with the wind speed increasing. At the set point of
wind speed, the generating power reaches the rated power of
tip speed ratio. Therefore the mechanical power of the wind
the turbine. If the wind speed continues to rise, the generator
turbine extracted from the wind is
output power remains constant at the design limit. Due to safety
Pw = 0.5C p ( , ) A w 3 (2) consideration, the turbine is shut down at speeds exceeding
where Cp is the power coefficient of the wind turbine, is the cut-out wind speed.
blade pitch angle and is the tip speed ratio. The tip speed
ratio is defined as the ratio between the blade tip speed and the
1
wind speed w
R
=
0.8
(3)
w
Power [p.u]

0.6
where is the turbine rotor speed and R is the radius of the
wind turbine blade. 0.4

0.2
V cut in V rated V cut-out

0
0 5 10 15 20 25 30
Wind speed [m/s]

Fig. 4 Power control regions of wind turbines

B. Blade Aerodynamics
The wind velocity conditions at a blade cross-section are
illustrated in Fig. 5. The velocity components, include the wind
speed, the flow factors and rotational speed of the rotor, will
determine the angle of attack. Using angle of attack, the forces
on a blade element is calculated by means of two-dimensional
aerofoil characteristics. Then the blade element momentum
(BEM) method can be used to predict the blade forces
Fig. 2 Power coefficient characteristics produced due to the interaction of blades with the wind.
The inflow wind velocity perpendicular to the rotor plane is
Thus any change in the rotor speed or the wind speed w. When the wind passes through the rotor plane, this wind
induces change in the tip speed ratio leading to power speed becomes reduced by an amount aw due to axial
coefficient variation. In this way, the generated power is interference. The blade rotates at angular velocity . Thus, a
affected. Fig. 2 shows a group of typical Cp- curves where blade element at a distance r from the rotor axis will be moving
optimum values of tip speed ratio, opt, correspond to the at a speed r in the rotor plane. When the wind passes through
maximum power coefficient, Cp.max. Fig. 3 shows that the the rotor plane and interacts with the moving rotor, a tangential
mechanical power converted from the turbine blade is a velocity of the wake a' r is introduced. Then the net
function of the rotational speed, and the converted power is tangential flow velocity experienced by the blade element is
maximized at the particular rotational speed for various wind (1+a') r. The resultant relative velocity at the blade is

2692
 DRPT2008 6-9 April 2008 Nanjing China

W = w2 (1 a 2 ) + 2 r 2 (1 + a) 2 (4) 25

where W is the resultant relative velocity at the blade, a and a'

is the flow factors and r is the distance of blade element from 20

the rotor axis. The resultant relative velocity gives rise to

Blade Angle [degree]

aerodynamic forces on the blade, therefore a lift force is 15

1
FL = cW 2C L (5) 10
2
and a drag force is
5
1
FD = cW 2CD (6)
2 0
where CL is lift coefficients, CD is drag coefficients and c is the 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind Speed [m/s]

chord length of the blade. Fig. 6 Generic blade pitch angle

In the pitch-adjusting variable-speed wind turbines, the
angle of attack, , decreases when the pitch angle, , increases, The conventional blade pitch angle control strategies are
as shown in Fig. 5. The lift force, FL, decreases as well and this shown in Fig. 7. The pitch angle reference, ref, is controlled by
causes reduction of the mechanical power of the wind turbine. the input values, which may be as follows:
According to the pitch angle control, the initial angle below 1) Wind speed, as shown in Fig. 7(a). Ideally, the pitch
rated wind speed is the optimized value, opt. The optimized angle reference can be obtained from the curve of the
pitch angle can be defined using the BEM method from the pitch angle versus wind speed, as shown in Fig. 6. This
incoming wind and it is in very few degrees around zero. As control strategy is simple as the wind speed is directly
shown in Fig. 4, the power of the wind turbines will increase at measured. However, this is not an appropriate
the increasing wind. The wind turbine must be protected procedure, since it is not possible to measure the wind
against mechanical overloads and possible risk of damages at speed precisely.
strong wind. This is achieved by pitching the blades into the 2) Generator rotor speed, as shown in Fig. 7(b). The
position where a part of incoming wind will pass by the wind controlling rotor speed is compared with its reference.
turbine. The power will be kept at its rated value at this region. The error signal is then sent to the PI controller and
By pitching, the thrust is reduced. Fig. 6 shows the curve of the produces the reference value of the pitch angle, ref.
pitch angle versus incoming wind speed which is computed 3) Generator power, as shown in Fig. 7(c). The error
with use of the BEM method for the given 1.5 MW wind signal of the generator power is sent to a PI controller.
turbine. The PI controller produces the reference pitch angle ref.
The non-linear variation of the pitch angle versus wind
speed for large wind speeds, illustrated in Fig. 6,
implies the necessity of a non-linear control. When the
wind speed is close to the rated, the sensitivity of
aerodynamic torque to pitch angle is very small. Thus a
much larger controller gain is required here than at
higher wind speeds, where a small change in pitch can
have a large effect on torque. Frequently the torque
sensitivity changes almost linearly with pitch angle,
and so can be compensated by varying the overall gain
of the controller linearly in inverse proportion to the
pitch angle. Such a modification gain with operating
Fig. 5 Blade element velocities and force at a blade dross-section. point is termed as gain scheduling [5].
The conventional pitch angle control strategy with gain
III. CONVENTIONAL PITCH ANGLE CONTROL scheduling performs well when the system dynamic is not
Adjusting the pitch angle of the blades, as shown in Fig. 2, strong. However, the sensitivity of aerodynamic torque to pitch
angle varies in a different way which is treated as linear
provides an effective means of regulations or limiting turbine
variation in gain scheduling, and because of its effect on
performance in strong wind speeds. To put the blades into the
necessary position, pitch servos are employed which may be turbine dynamics which couples strongly with the pitch
controller, it may be necessary to modify the gain scheduling
hydraulic or electrical systems. During normal operation, blade
further to ensure good performance in all winds.
pitch adjustments with rotational speeds of approximately
5-10/s are expected [11]. Here the chosen pitch rate is 8/s
which avoids excessive loads during normal regulation
procedures.

2693
 DRPT2008 6-9 April 2008 Nanjing China

(a) shown in Fig. 8. The input and output membership functions

are shown in Fig. 9. Triangular symmetrical membership
functions are suitable for the input and output, which give more
sensitivity especially as variables approach to zero value. The
width of variation can be adjusted according to the system
parameters.
(b)

Fig. 8 Fuzzy logic control strategy

(c)
As shown in Fig. 8, the proposed fuzzy logic controller is
Fig. 7 Pitch angle control strategy. (a) Wind speed; (b) generator rotor speed; (c) based on the power deviation from its reference value P, its
generator power. variation during a sampled time (P) as follows:
IV. FUZZY LOGIC CONTROL P = Pref Pg
(7)
(P ) = Pn Pn 1
Rule based fuzzy logic controllers are useful when the
system dynamics are not well known or when they contain where Pref is the rated power of the system and Pg is the
significant non-linearities, such as the un-stationary wind measured generator power.
contains large turbulence. The advanced control strategies are The mean wind speed wm, which used as the third input
more favor as reducing fatigue loads is one of the control variable, is useful to compensate the non-linear sensitivity of
motivations. pitch angle to the wind speed. Table I lists the control rules for
Fuzzy logic controllers apply reasoning, similar to how the input and output variable. In the proposed fuzzy system,
human beings make decisions, and thus the controller rules nine fuzzy sets have been considered for variables: negative
contain expert knowledge of the system. The big advantages of large (NL), negative medium large (NML), negative medium
fuzzy logic control when applied to a wind turbine are that the (NM), negative small (NS), zero (ZE), positive small (PS),
turbine system neither needs to be accurately described nor positive medium (PM), positive medium large (PML), positive
does it need to be linear. The design process for a fuzzy logic large (PL). At lower mean wind speed, the low sensitive of the

(a) (b)

(c) (d)
Fig. 9 Membership functions of fuzzy logic controller.(a) Input signal error of the power; (b) Input signal variation of power error; (c) Input signal wind speed;
(d) Output signal variation of pitch angle.

controller consists of (i) determining the inputs, (ii) setting up pitch angle control means that a large reaction of pitch angle to
the rules and (iii) designing a method to convert the fuzzy result the control variants will be needed than at higher mean wind
of the rules into output signal, known as defuzzification. speed. This is seen in Table I.
This paper proposed a new fuzzy logic controller for the
pitch control where one of the motivations of the controller is
to reduce the fatigue loads. The fuzzy logic control system is

2694
 DRPT2008 6-9 April 2008 Nanjing China
TABLE I.
RULES OF FUZZY LOGIC C ONTROLLER
m PS PM PL
P NL NS ZE PS PL NL NS ZE PS PL NL NS ZE PS PL
P
NL NL NML NM NM PS NL NM NM NS PS NML NM NS NS PS
NS NL NM NS PS PM NML NM NS PS PM NML NM NS ZE PS
ZE NML NS ZE PS PML NM NS ZE PS PM NM NS ZE PS PM
PS NM NS PS PM PL NM NS PS PM PML NS ZE PS PM PML
PL NS PM PM PML PL NS PS PM PM PL NS PS PS PM PML

An immediate and easy understanding of the controller logic life is resulted according to the rain-flow counting method.
can be obtained as following considerations (not consider wind Comparatively, one-second mean wind speed is used as the
speed): controlling variable and a deviation to the ideal performance is
1) If P and (P) are negative large, the output power shown.
5

angle is too large and its amplitude is growing, consequently 14

x 10

current pitch angle must be fast decreased. 12

Torque [Nm]
2) If P is negative large and (P) is positive large, the 10

output power is higher than its reference, but since its 8

amplitude is decreasing, pitch angle variation must be small. 6

3) If P is small pitch angle variation must be smoothed

4
0 5 10 15 20 25 30 35 40 45 50
6

because too large variations excite oscillatory modes. 2

x 10

1.5

V. SIMULATION RESULTS AND DISCUSSION

Power [W]
1
Ideal wind speed
Pitch angle control systems of the wind turbine were 0.5 1s mean wind speed

simulated using MATLAB/SIMULINK tool to test the control 0

strategy and evaluate the performance of the system. Table II 15
0 5 10 15 20 25 30 35 40 45 50

shows the characteristics of the wind turbine system.

Pitch Angle [degree]

10
TABLE II.
WIND TURBINE AND GENERATOR CHARACTERISTICS 5
Blade radius 35 m
Hub height 82 m 0
Gearbox ratio 6 0 5 10 15 20 25 30 35 40 45 50

Generator power 1500 kW 2.5

2.4
Speed [rad/s]

The wind model is essential to obtain realistic simulations of 2.3

the power control of the wind turbines. The model for the wind 2.2
field includes turbulence as well as tower shadow effects. Fig.
2.1
10 illustrates the hub-height wind speed variation. It has a mean 0 5 10 15 20 25
Time [s]
30 35 40 45 50

value of 11 m/s at the hub-height and a turbulence intensity of Fig. 11 Pitch angle control strategy: wind speed.
30%.
19
Fig. 12 shows the results of the conventional pitch angle
18
control strategy where generator power is used as the
17
controlling variable. The comparison of with and without gain
16
scheduling is shown that the former has rapid pitch angle
respond to the wind speed variation which is favor for
Wind Speed [m/s]

15
minimum power and torque ripples.
14
The comparisons of different strategies are shown in Fig. 13.
It is shown that proposed fuzzy logic control strategy have
13

lowest torque ripples at high wind speed, which means a

12

11
reduced fatigue loads to the turbine components. A narrowly
10
0 5 10 15 20 25 30 35 40 45 50 lower power peak and lower torque peak is also obtained for
t [s]
fuzzy logic controller. However, the over-speed is not reduced
Fig. 10 Wind speeds with a mean of 11 m/s.
by using fuzzy logic controller. The strategy using rotor speed
Fig. 11 shows the results of the conventional pitch angle as controlling variable has less over-speed while the torque and
control strategy where wind speed is used as the controlling power ripple is most serious. From the view of smoothing the
variable. It is shown that if the ideal wind speed is available, the pitch angle action, the fuzzy logic control is better than other
torque ripples at high wind speed, as shown in Fig. 11 from two control strategies, except the rotor speed control strategy.
12-40 seconds, is extremely small. This means very small
fatigue loads to the wind turbine components and a long fatigue

2695
 DRPT2008 6-9 April 2008 Nanjing China

The conventional pitch angle control strategy using different

6
x 10
2
Torque [Nm]
1.5
controlling variables can be implemented. However, fuzzy
1
logic pitch angle control strategy need not well know about the
wind turbine dynamics and when wind turbine contains strong
0.5
non-linearities, it is more favor also. The simulation results
0
0 5 10 15 20 25 30 35 40 45 50 show that the fuzzy logic controller has lowest fatigue loads
and lower torque peak and lower power peak. The pitch angle
6
x 10
2.5

2 of the fuzzy logic controller is less active than conventional

pitch angle control with power or wind speed controlling
Power [W]

1.5

1
With gain scheduling
variable, which causes less dynamic torque.
0.5 Without gain scheduling
0
0 5 10 15 20 25 30 35 40 45 50
15
VII. ACKNOWLEDGMENT
Pitch Angle [degree]

10 Jianzhong Zhang thanks China Scholarship Council for

providing a one-year joint-supervising PhD scholarship at
5
Institute of Energy Technology, Aalborg University, Denmark.
0

2.5
0 5 10 15 20 25 30 35 40 45 50
VIII. REFERENCES
2.4
[1] T. Burton, D. Sharpe, N. Jenkins, E. Bossanyi, Wind energy handbook,
Speed [rad/s]

John Wiley & Sons Ltd, Chichester, UK, 2001.

2.3 [2] DNV/Ris, Guidelines for design of wind turbines, 2nd ed., Jydsk
2.2 Centraltrykkeri, Denmark, 2002.
[3] S. Suryanarayanan, A. Dixit, Control of large wind turbines: Review and
2.1
0 5 10 15 20 25 30 35 40 45 50 suggested approach to multivariable design, Proc. of the American
Time [s] Control Conference 2005, Portland, USA, pp. 686-690.
Fig. 12 Pitch angle control strategy: generator power. [4] V. Akhmatov, Variable-speed wind turbines with doubly-fed induction
generators Part I: Modelling in dynamic simulation tools, Wind
x 10
5
Engineering, Vol. 26, No.2, 2002, pp. 85-108.
15
[5] A. D. Hansen, P. Sorensen, F. Iov, F. Blaabjerg, Control of variable speed
wind turbines with doubly-fed induction generators, Wind Engineering,
Torque [Nm]

10
Vol. 28, No. 4, 2004, pp. 411-434.
[6] T. Senjyu, R. Sakamoto, N. Urasaki, H. Higa, K. Uezato, T. Funabashi,
5
Output power control of wind turbine generador by pitch angle control
0
using minimum variance control, Electrical Engineering in Japan, Vol.
0 5 10 15 20 25 30 35 40 45 50 154, No. 2, 2006, pp. 10-18.
6

2
x 10 [7] L. Lupu, B. Boukhezzar, H. Siguerdidjane, Pitch and torque control
strategy for variable speed wind turbines, in 2006 European Wind Energy
1.5
Conference Proceedings, Athens, Greece 2006.
Power [W]

1 Mean wind speed [8] Z. Chen, S. A. Gomez, M. McCormick, A fuzzy logic controlled power
Rotor speed+PI electronic system for variable speed wind energy conversion systems,
0.5 Power+PI
Eighth International Conference on Power Electronics and Variable
Fuzzy logic
0 Speed Drives, Publication No. 475, 2000.
0 5 10 15 20 25 30 35 40 45 50
[9] A. G. Abo-Khalil, D. Lee, J. Seok, Variable speed wind power generation
15
system based on fuzzy logia control for maximum output power tracking,
35th Annual IEEE Power Electronics Specialists Conference, Achen,
Pitch Angle [degree]

10
Germany, 2004.
[10] C. Jauch, T. Cronin, P. Srensen, B. B. Jensen, A fuzzy logic pitch angle
5 controller for power system stabilization, Wind Energy, 2007, pp. 19-30.
[11] S. Heier, Grid integration of wind energy conversion systems, John Wiley
0
0 5 10 15 20 25 30 35 40 45 50
& Sons Ltd, 2nd ed., Chichester, UK, 2006.
2.5

2.4
Speed [rad/s]

2.3

2.2

2.1
0 5 10 15 20 25 30 35 40 45 50
Time [s]

Fig. 13 Comparisons of pitch angle control strategies.

VI. CONCLUSION
Pitch angle control is the most common means to control the
aerodynamic power generated by the wind turbine rotor. Pitch
angle control also has an effect on the aerodynamic loads
which may be controlled by the controller to achieve lower
torque peak as well as lower fatigue loads.

2696