See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/277733834

Variable Speed wind Turbine Control System: Design & Model

Article · January 2010

CITATION READS
1 4,514

1 author:

Wajeeh Ul Hasan
Shaheed Zulfikar Ali Bhutto Institute of Science and Technology, Karachi Campus, Karachi, Pakistan.
4 PUBLICATIONS 6 CITATIONS

SEE PROFILE

All content following this page was uploaded by Wajeeh Ul Hasan on 10 June 2015.

The user has requested enhancement of the downloaded file.

 Variable Speed wind Turbine Control System: Design & Model
Wajeeh Ul Hasan1 & Dr. Faraz Junejo2
SZABIST
Karachi.

The commercial production of wind turbines started in

Abstract:At present, the wind turbine is the most common
early 1980s. Since then frequent enhancements are utilized
alternative energy source that is being utilized to produce
to increase the ability of wind turbines to absorb maximum
electricity around the globe. The rotor accompanying the
possible wind energy. Generally these enhancements
blades is thecomponent of wind turbine that transforms the
include use of more impressive rotors, larger blades, better
kinetic energy of wind to electrical energy. The control of
use of composite materials and higher towers. Since 1980,
rotor frequency is the area of problem to be solved. The
the efficiency of a wind turbine has increased by a factor of
frequency control can be achieved by controlling the
more than 200 [2]. Keeping Betz theory in mind which
nacelle and rotor blades for the optimum performance of
state that the most optimal wind turbine can only capture
wind turbine, because they have direct impact on wind
59.3% of kinetic energy from wind, the reliability of wind
turbine’s performance. The trend of wind speed shows that,
turbines are very high up to 98%, with operating
it is not constant and it varies throughout the operation of
availabilities(the proportion of time in which they are
a wind turbine. Therefore, it is required to build a
solutionthat will be used to control the rotor’s pitch and available to operate) [2]. Wind turbines are the only source
nacelle’s direction to maintain fixed frequency of rotor. that can produce electricity at such higher availability.
The speed of rotor can be controlled by changing pitch Generally, todays modern wind turbine’s rotor
angle and nacelle direction to maintain the wind turbine contains three blades. The speed and power of rotor is
rotor frequency for generator. A sophisticated model is managed by either yaw or pitch ruling, where stall ruling
needed to be built to control both parameters. For the control the mechanical rotation of blades while pitch ruling
testing and validation of the model simulation has been manage the blade angle. Blades are manufactured from
performed. composite materials like fiberglass and polyester or
fiberglass and epoxy.
Keywords:Wind Turbine, Controller, Feedback Control
System

1. INTRODUCTION

Use of wind energy is not a new thing. It has been

used since at least 3000 years. Initially wind energy was
used to crushing grain or driving water. The most common
and oldest example of using wind energy as an essential
source of power for long-term is sailing ships. The
technology that empowers winds energy as a source of Figure 1: Illustrating the Principal of Betz Law
power is known as windmills or wind turbines. In 13th The fuel for wind power station is wind energy.
century a revolutionary era originated in the field of wind The commercial value of wind farms are badly affected by
turbines and horizontal-axis wind turbines became an just slight changes of wind energy.
integral part of rural economy [1]. The wind turbine is the It is observed that when the average wind speed
most common source of alternative source that is being doubles, the power of wind turbines increases by factor of
utilized to produce electricity around the globe. Such wind eight. Therefore, the small changes in wind speed could
turbines were used to drive cheap fossil-fuelled engines and affect wind turbines performance very badly. For example,
then feast rural electrification. In late 19th century, a 12 the change of average wind speed from 6 m/s to 10 m/s will
KW DC wind turbine known as “The Brush Machine“ (to affect the energy production performance of a wind turbine
generate electricity) was built by Charles F. Brush in the over 130% [3]. Therefore, it is necessary to have thorough
USA and its exploration carried out by LaCour in Denmark and consistent information about the power and direction of
[1]. wind with its blow regularity at suggested site, to build an
efficient wind turbine [2].

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 44
 Generally wind turbines start working at the wind 2. LITERATURE SURVEY
speed of 3m/s to 4m/s, but the full power operation
normally requires 10 m/s to 15 m/s of wind speed [3]. This 2.1 Control Theory
range of wind speed is known as cut-in speed. At the speed Control systems basically deal with the
25 m/s wind turbine goes in shutdown state. The wind influencing actions of energetic systems that are needed to
speed at which shut down state occurs is known as cut-out run on specific conditions. To meet these required
speed. conditions, it is necessary to come up with a small and
The three main components of wind turbine are sophisticated program that could handle the input
tower, rotor and nacelle. Rotor is mounted on tip of tower. parameters for desired output features and could built in a
Normally rotor contains three blades that helps rotor to microchip. These desired outputs are known as reference.
transform energy absorbed from wind into mechanical The program that control input parameters is called
energy and then with the help of generator that mechanical controller. It could be in the form of software or hardware.
energy transform into electrical energy. Generally, rotor’s Such programs are when found in the form of hardware or
blades helps rotor to rotate at the constant speed of about chip, known as microcontroller.
15 to 30 revolutions per minute (rpm) [3], it could be vary
as per requirement. The ratio of wind speed to the rotation
of blades is known as Tip Speed Ratio. As higher the ratio
is, the faster the rotation of blades are at the given wind
speed.

Figure 3: Feedback Control System

2.2 CONTROL STRATEGIES

Control systems have different strategies for the
stability of systems. The most common strategies of control
systems are:

Adaptive control
Figure 2: Sample Structure of a Nacelle of a Horizontal
Hierarchical control
Axis Wind Turbine

Intelligent control
The rotation of blades is caused by wind energy

Optimal control
and controlled by the nacelle which can face the blades to

Robust control
the wind (yaw control) and by changing blades angle (pitch

Stochastic control, etc.
control) to optimal absorption of available wind. The
nacelle is the generative part of the wind turbine that 2.2.1 Adaptive Control System:
contains generator, controller, gearbox, wind vane and an Adaptive control applied by perceiving
anemometer to monitor the wind speed and wind direction. operational identification of process factors, in that way
It has already seen earlier that slight change of wind speed attaining robust properties [4]. The adaptive control
cause large changes in frequency of wind turbine will badly systems are used in complex systems by automatically
affects its performance.For this purpose, it is necessary to finding its way to control the mechanism.
build such a sophisticated control model that can control
wind turbines rotor and blades. Therefore in this study the 2.2.2Hierarchical Control System:
focus will on this issue and try to come up with an optimal Large system’s controls are always planned in a
feedback control system in MATLAB environment. distributed hierarchy. Hierarchical controls is the branch of
control systems that replicating a divided and conquer
strategy. In hierarchical control systems, set of actuators
and principal software are arrange in a hierarchical tree [4].

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 45
2.2.3 Intelligent Control System:
To attain sophisticated goal by a system, an
Intelligent Control System is prefer to use because it could
work independently, while due to unexpected changes in
system, its components are not well-defined. Intelligent
control systems use one of different AI techniques as per
requirement, like; neural network, fuzzy logics, etc. to
control the system [4].

2.2.4Optimal Control System:

Optimal control system is the technique in which
control signal are optimized at given point [4]. Optimal Figure 4: Block Diagram of a Wind Turbine Feedback
control system considers a more comprehensive condition Control System
in which decision variable and the level of state machine In this study, a hierarchical control strategy is use
affect the change in the state variable. The future growth of to attain the wind turbine requirements. The hierarchical
state variable depends on current decision and current control model could be observed by figure 5. Figure 5
decision is based on state variables. show that main controller transfers the control initially to
pitch controller on the basis of mathematical calculations.
2.2.5Robust Control System: Pitch controller is assigned to control the pitch angle of
Robust control systems are those control rotor’s blades according to wind speed conditions.
mechanisms that controls the system by taking all If pitch controller is unable to meet the
parameters in account. It could handle all the minor requirements of wind turbine, the control will transfer to
differences that occur between the real systems and the yaw controller. Then yaw controller will try to adjust the
model that design to use [4]. rotor to face the blades to the wind. If still the required
conditions doesn’t meet then the control will transfer to
2.2.6 Stochastic Control System:
state controller that decide to switch wind turbine’s state to
Stochastic control deals in modeling and control
start-up or shut-down states. The moment when wind speed
of dynamic systems inclined by stochastic turbulences and
crosses either upper or lower speed limit, the state
indecisions. Stochastic control systems are the systems that
controller switches the wind turbine to shut-down state.
could account all noises and disturbances, which could
occur in a system [4]. In stochastic control, the abnormal
conditions and system’s efficiency is monitored and
accommodated.

2.3WIND TURBINE FEEDBACK CONTROL

SYSTEM
In implementation of a wind turbine controller,
various types of controllers are required that are assigned to
control different functions of wind turbines. Controller
objectives are to be assigned very carefully, keeping all
parameters in consideration. Because controllers might
have major effects on structural loads, therefore it is Figure 5: The Hierarchical Control Model for Wind
necessary to consider these during designing optimal Turbine
control algorithm. The objective of controller may include
regulate aerodynamic torque, gearbox operation, pitch 2.4 RESEARCH SURVEY
activity, controlling tower vibration, yaw controlling, etc. Many researchers have done extensive work on
controlling wind turbine in different regions of the world.
Engelen et al [5] defines the design tools for wind turbine
control algorithms by focusing on frequency domain
analysis and synthesis of (linearized) turbine models.

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 46
 D. J. LEITH et al encountered three important generic 3.1 SIMULATION OBJECTIVES
implementation issues of controllers of pitch regulated The primary objective of designing a wind turbine
constant speed wind turbines [6]. The encountered issues control model is to attain the wind turbine requirements.
are named as: The controlling of wind turbine is a very serious problem

accommodation of the strongly nonlinear rotor because of the complex structure of the wind turbine along
aerodynamics [6], with variable wind speed. The control of variable wind

automatic controller start-up, shut-down and speed is essential to meet the wind turbine requirements.
accommodation of velocity [6] and Wind speed can be controlled by adjusting the wind

acceleration constraints within the actuator [6]. turbine’s blades. These blades are controlling by adjusting
Alan D. Wright defines the control system which their angle as per requirement. If wind speed is faster than
applies modern state-space control design methods to a the defined speed limit, the blades angle increases, so that
two-bladed upwind machine [7]. He defines Disturbance the area of absorbing wind energy decreases. Similarly, if
Accommodating Control (DAC) method and provides wind speed slower than the defined speed limit, the blades
accountability for wind-speed fluctuations [7]. angle decreases, so that the area of absorbing wind energy
Boubekeur Boukhezzar et al use nonlinear controllers increase. Because of the hierarchical control strategy,
to improve the efficiency and to evaluate the related initially controller try to meet the requirements of wind
efficiency; he used nonlinear static and dynamic state turbine by adjusting pitch angle than the control will
feedback controllers [8]. transfer to yaw control. Therefore, in the simulation
Kathryn E. Johnson et al work on the variable speed process, the proposed controller model especially targeted
wind turbines [9]. He explores the steadiness of an adaptive to manage the blades pitch angle.
torque control rule and the gain adaptation rule which are The wind turbine controller is also responsible to
in use at Controls Advanced Research Turbine (CART) [9]. control the dynamics of nacelle. The structural load of the
Pradeep Bhatta et al focuses on the control law, wind turbine is mainly because of its nacelle. The nacelle
derived from an energy-like Lyapunov function of the wind turbine contains generator, controller, gearbox,
(Lyapunov functions are used to prove the stability of a drive train, etc. All these components of nacelle make its
static point in a dynamical system). He guarantees the weight very heavy.
system stability for all operating points of the wind energy
conversion system (WECS) [10]. 3.2 SIMMULATION MODEL
The proposed simulation model contains blade
3. SIMULATION AND MODELING height (bh), initial pitch angle (pa) and required Tip Speed
Ratio (TSR) as input variables. The output variables of this
The model of wind turbine controller is simulated
model contain pitch controlling bit (pitch), wind speed
in MATLAB environment. The effectiveness of the
(wind) and rotor rpm (rpm). The list of variables use in
proposed model is evaluated by carrying out the simulation
model can be observed by table 3.1:
experiments, in which rotor speed is controlled by
adjusting pitch angle with respect to variable wind speed.
Table 3.57: Variable names and their acronyms
To evaluate this model, it is necessary to have variable
wind speed. For this purpose, in this model wind speed is VRIABLE
VARIABLE NAME
simulated by using “Weibull Distribution”. The ACRONYM
effectiveness of controller in terms of controlling wind Rrpm Rotor RPM
turbine is depending on the effective wind energy at the Bh Blade Height
given region. Grpm Generator RPM
It was already mentioned in Chapter 1 and Ws Wind Speed
Chapter 2 that wind turbine starts operation on 3 m/s to Pa Initial Pitch Angle
4m/s and full operation available on 10 m/s to 15 m/s, as TSR Tip Speed Ratio
per requirement. Therefore, in the simulation process, it is Pbit Pitch Controller Bit
especially kept in consideration that wind speed must vary Ybit Yaw Controller Bit
in such a way that wind turbine will not switch to shut
down state. The proposed control model is based on In this proposed model some values are taken as
hierarchical control strategy. constant values. These values are depending on
requirements of desired energy level, wind turbine’s

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 47
coefficient requirements and mainly depend on the site
region. The lists of these constant values are observed by
table 3.2:

Table 3.58: List of Constant Values

NAME CONSTANT VALUE
Tip Speed Ratio 4
Rotor RPM 28 – 32
Final Pitch Angle 15

The proposed model diagram can be viewed by

figure 6:

Figure 6: Proposed Model of Wind Turbine Controller

The formula to calculate rotor rpm is derived from

the formula of Tip Speed Ratio (TSR), where TSR is kept
to be constant and wind speed variation is simulated by
using Weibull Distribution.

Where, is Tip Speed Ratio,

is Wind Speed, Figure 7: Part of code that control pitch angle
is Rotor rpm,
is blade height. The first line of the above given chunk of code
shows that the controller works only when, rotor rpm
The main “controller” model contains the basic decreased or increased from the defined range. The second
controlling code that governs the wind turbine on the basis line of code has made a check to verify the current rotor
of input values. Basically the code works on hierarchical speed for below the desired speed. If the rotor speed is
basis. First it try to adjust rotor speed by adjusting pitch below the desired condition and pitch angle is in between
angle, but if rotor speed is not meet the required speed then its initial and final condition then the controller pass the
it transfer control to yaw controller. The proposed signal to decrease the pitch angle. After decreasing pitch
controller code that governs the wind turbine controller is angle, controller makes a check that either rotor speed
depicted by Figure 7. meets the desired requirement or not.

The second check has made to verify the current

rotor speed for above the desired speed. If the rotor speed is
above the desired condition and pitch angle is in between
its initial and final condition then the controller pass the

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 48
signal to increase the pitch angle. After increasing pitch The second check in this part of code verifies the
angle, controller makes a check that either rotor speed current rotor speed for above the desired speed and as well
meets the desired requirement or not. as the condition of pitch angle. If the rotor speed is above
the desired condition and if there is no chance to increase
pitch angle further then the controller transfer the control to
the yaw controller and pass signal to yaw controller to
change the nacelle direction. After changing nacelle
direction, controller makes a check that either rotor speed
meets the desired requirement or not.
In next section results obtained by running the
proposed controller simulation will discuss and analyze.

4. RESULT ANALYSIS
It has been discussed in Chapter 3 that in this model
wind speed is simulated by using “Weibull Distribution”
and it is especially kept in consideration that variation of
wind speed must occurs in such a way that wind turbine
will not switch to shut down state. Therefore, in the
simulation process, the proposed controller model will
project only the effectiveness of pitch angle controlling
part,and it is sure that yaw controlling part will show its
effectiveness in real-time environment. The surety of this
statement is based on the fact that in real-time environment,
many different sensors and equipment could be arranged
that can sense the output values and feedback to controller.
For example, a counter can be set that sense the current
pitch angle value and increase or decrease as per
controlling signal response. This counter can be used to
feedback the status of current pitch angle value. Likewise a
sensor can be used, which feedback wind speed or wind
direction status from anemometer which is very useful to
adjust pitch angle or nacelle direction. On the basis of
counter which feedback the pitch angle value, it is easily
decidable that whether control transfer to yaw controller is
needed or not. While in the simulation process of wind
Figure 8: Part of code that control nacelle direction turbine controller, such accommodations are not available,
therefore complete simulation of proposed controller is not
The second part of code is depicted by Figure 8, possible.
which deals with the control of nacelle direction or simply The effectiveness of the proposed controller model is
says that deals with yaw controller. The first check in this evaluated by carrying out the simulation experiments. For
part of code verifies the current rotor speed for below the this purpose, values of required parameters that are needed
desired speed and as well as the condition of pitch angle. If to run simulation are already defined in table 3.2. After
the rotor speed is below the desired condition and if there is running simulation process, three different graphs will
no chance to decrease pitch angle further then the obtained, where first graph depicted by Figure 9 shows
controller transfer the control to the yaw controller and wind speed variation, second graph depicted by Figure 9
pass signal to yaw controller to change the nacelle shows the rotor speed in revolution per minutes (rpm) and
direction. After changing nacelle direction, controller the last graph depicted by Figure 9 shows the action of
makes a check that either rotor speed meets the desired pitch controller. Figure 9 shows the variation of wind speed
requirement or not. which is simulated using “Weibull Distribution”. By
analyzing wind simulation graph it can be observed that
with the reference of discussion in Chapter 3, the variation

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 49
of wind speed is occurs in such a way that wind turbine By viewing figure 9, 10 and 11 concurrently, it could be
will not switch to shut down state. observe that when rotor speed cross the upper limit, pitch
The simulation process is running with following controller pass the signal to pitch actuator to increase the
assumptions: pitch angle. Increasing pitch angle allows rotor to slower

After every single pitch increment or decrement, down its speed. While, when rotor speed cross lower limit,
there is a delay of two seconds to observe the current pitch controller pass the signal to pitch actuator to decrease
rotor speed. Basically it depends on wind speed the pitch angle. Decreasing pitch angle allows rotor to
variation. increase its speed.

On each single increment or decrement, the limit of
pitch actuator to increase or decrease pitch angle is
predefined.

The wind speed variation does not allow wind
turbine to go in shut down state.

Figure 8: Pitch Angle Signaling

5. CONCLUSION
The design task of such a control system that can
efficiently control wind turbine, must address several
issues, such as variable wind speed which causes a
variation in the wind turbine’s rotor frequency i.e. rpm,
Figure 9: Wind Speed (m/s)
thereby resulting in mismatch between rotor and generator
The second graphcan be viewed by observing
rpm, which can damage the generator. In this study, a
figure 10, which shows the rotor speed. By observing
hierarchical control model namely, “Rotor Speed Feedback
figure 9 and 10 simultaneously, it can be conclude that the
Control” is proposed to resolve the problems of controlling
ideal wind speed at which rotor speed is in ideal condition
wind turbine and to maintaining its consistency i.e. rotor’s
i.e. 30 rpm is 12 m/s.
rpm; this control model is tested, and analyzed via
By observing figure 9 and 10 simultaneously, it is
simulation process.
also conclude that when wind speed goes above 13 m/s, the
The control strategy is validated using simulation only,
rotor speed also increases and cross the predefined upper
due to the unavailability of sensors and counters; it was not
limit. When wind speed goes below 11 m/s, the rotor speed
possible to validate it for real world environment. As wind
also decreases and falls down the defined lower limit.
turbines are sensitive to wind speed, the proposed control
model is used to control pitch and nacelle direction to
maintain the required rotor speed i.e. between 28 and 32
rpm. The results discussed in previous chapter shows that
proposed control model does achieve desired stability
results, hence fulfill the requirements of this study.

REFERENCES
1. Wind Energy Handbook. By T. Burton, D. Sharpe, N.
Jenkins, and E. Bossanyi, Wiley et sons., 2002
2. Wind Power Technology, link:
Figure 10: Rotor Speed (rpm) http://www.ewea.org/fileadmin/ewea_documents/docu

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 50
 ments/publications/factsheets/factsheet_technology2.p 7. Modern Control Design for Flexible Wind Turbines by
df Alan D. Wright, 1617 Cole Boulevard, Golden,
3. Wind Energy As A Power Generation Option-Chapter Colorado 80401-3393, 303-275-3000, www.nrel.gov
, link : 8. Nonlinear Control of Variable Speed Wind Turbines
http://www.eskom.co.za/content/WEF%20DSR%20- without wind speed measurement by Boubekeur
%20Ch%2003%20_Wind%20energy%20as%20a%20 Boukhezzar and Houria Siguerdidjane, 44th IEEE
power%20option_.pdf Conference on Decision and Control, and the
4. Control Theory From Wikipedia, the free European Control Conference 2005 Seville, Spain,
encyclopedia, Link: December 12-15, 2005
http://en.wikipedia.org/wiki/Control_theory 9. Stability Analysis of an Adaptive Torque Controller
5. Design Tool For Wind Turbine Control Algorithms By for Variable Speed Wind Turbines by Kathryn E.
Engelen, P. Schaak And E.J. Van Der Hooft, T.G. Van Johnson, Lucy Y. Pao, Mark J. Balas, Vishwesh
Wiggelinkhuizen, ECN Wind Energy, P.O. Box 1, Kulkarni, and Lee J. Fingersh, 1617 Cole Boulevard,
1755 ZG Petten,The Netherlands, Fax. +31 224 Golden, Colorado 80401-3393, 303-275-3000,
568214 www.nrel.gov
6. Implementation of wind turbine controllers by D. J. 10. Nonlinear estimation and control for small-scale wind
LEITH and W. E. LEITHEAD, Department of turbines by Pradeep Bhatta and Michael A. Paluszek,
Electronic and Electrical Engineering, University of Princeton Satellite Systems, Inc. 33 Witherspoon St,
Strathclyde, Glasgow G1 1QE, UK Princeton, NJ 08542, USA
{pradeep,map}@psatellite.com

Journal of Independent Studies and Research – Computing Volume 8 Number 1 January 2010 | 51

View publication stats