Frequency Control For DFIG

DFIG Based Wind Turbine Contribution
to System Frequency Control

by

Mansour Jalali

A thesis
presented to the University of Waterloo
in fulfillment of the
thesis requirement for the degree of
Master of Applied Science
in
Electrical and Computer Engineering

Waterloo, Ontario, Canada, 2011

Mansour Jalali 2011

ii
Authors Declaration

I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including
any required final revisions, as accepted by my examiners.
I understand that my thesis may be made electronically available to the public.

iii
Abstract

Energy is one of the most important factors that continue to influence the shape of civilization in
the 21
st
Century. The cost and availability of energy significantly impacts our quality of life, the
health of national economies and the stability of our environment. In recent years there has been a
significant global commitment to develop clean and alternative forms of energy resources and it
is envisioned that by 2020 10% of world energy will be supplied from renewable resources, and
there is an expectation that this value will grow to 50% by 2050.
Among renewable energy resources, wind generation technology has matured
considerably, and wind is fairly distributed around the globe and therefore available to world
communities. In the last decade, wind generation has been the fastest growing energy source
globally. However more penetration of wind energy into existing power networks raises concern
for power system operators and regulators. Traditionally wind energy convertors do not
participate in frequency regulation or Automatic Generation Control (AGC) services, and
therefore large penetration of wind power into the power systems can result in a reduction of total
system inertia and robustness of the frequency response to the disturbances.
The research presented in this thesis covers some of the operational and design aspects of
frequency control and AGC services in power systems with mixed generation resources. The
thesis examines the operation of the Doubly Fed Induction Generator (DFIG) with a modified
inertial loop control considering single-area and two-area frequency control, both primary control
and AGC. The thesis presents new, small-perturbation, linear, dynamic, mathematical models for
the simulation of primary regulation services and AGC services for single-area and two-area
power systems with a mix of conventional and non-conventional DFIG-based wind generators. In
order to improve the performance of the frequency regulation and AGC services of the above
systems, a parameter optimization technique based on the minimization of the Integral of Squared
Errors (ISE) is applied to determine the optimal settings for the proportional-integral (PI)
controller gains of the DFIG machines.
The thesis presents analytical studies with various perturbations to demonstrate the
effectiveness and participation of DFIG-based wind generators in frequency support services and
draws some important conclusions. Variation in DFIG penetration levels, and wind speed levels
(strong wind and weak wind) on system frequency control performance, has also been examined
in the thesis.

iv
Acknowledgements

First and foremost, I express my deepest and greatest appreciation to my supervisor, Professor
Kankar Bhattacharya. Dr. Bhattacharya has consistently provided invaluable guidance and
suggestions in my research project as well as over my entire course program.
I am thankful to Professor Claudio A. Caizares and Professor El. Shatshat for reading this thesis
and providing their insightful comments and suggestions, which helped to improve this work.
I am thankful to the ABB Inc Canada, Department of Protection and Substation Automation
which have supported me with partial fund for my research and studies.

v
Dedication

This thesis is dedicated to my parents and my children Zhina and Alan.

vi
Table of Contents
Authors Declaration ...................................................................................................................ii
Abstract .................................................................................................................................... iii
Acknowledgements .................................................................................................................... iv
Dedication................................................................................................................................... v
Table of Contents ....................................................................................................................... vi
List of Figure ............................................................................................................................ vii
List of Tables ............................................................................................................................. ix
Nomenclature ............................................................................................................................. xi
Abbreviations ........................................................................................................................... xv
Chapter 1 Introduction................................................................................................................. 1
1.1 Motivation ............................................................................................................................. 1
1.2 Current State of Wind Energy Globally ................................................................................. 2
1.2.1 Wind Power in Canada and Ontario .................................................................................... 2
1.2.2.Wind Power in the United States: ....................................................................................... 4
1.2.3.Wind Power in the United States: ....................................................................................... 5
1.3 Objective of this Thesis ......................................................................................................... 6
1.4 Organization of the Thesis: .................................................................................................... 7
Chapter 2 Wind Power overview ................................................................................................. 8
2.1 Status of wind Turbine Technology ....................................................................................... 8
2.2 Power of the Wind ............................................................................................................... 11
2.3 Pitch control ........................................................................................................................ 13
2.4 1 Assumption and constraint in participation of DFIG in frequency regulation ..................... 19
2.4 Literature Review ................................................................................................................ 17
2.4 Wind Energy Converter System Functional structure: .......................................................... 14
2.4.1 WECS with Fixed Speed Wind Turbines Type A .............................................................. 14
2.4.2 WECS with Variable speed Wind Turbines type B............................................................ 15
2.4.3 WECS with Doubly-Fed Induction Generator (DFIG) Based Wind Turbine ...................... 15
2.4.4 WECS with full converter ................................................................................................. 16
2.5 Simulation: .......................................................................................................................... 19
2.5.1 Active Pitch Control: ........................................................................................................ 19
2.6 Conclusion: ......................................................................................................................... 22
Chapter 3 Primary Frequency Regulation and AGC in a Single-Area System with
DFIG-Based Wind Turbine ................................................................................................... 23

vii

3.1 Introduction ......................................................................................................................... 23
3.2 DFIG-Based Wind Turbine Control Model .......................................................................... 26
3.3 Dynamic Model of Primary Frequency Regulation with DFIG-Based Wind Turbines .......... 28
3.4 Dynamic Model of AGC with DFIG-Based Wind Turbines ................................................. 30
3.5 Analytical Studies ............................................................................................................... 34
3.5 Optimal Tuning of DFIG-Based Wind Turbine Controller Parameters ................................. 31
3.6.1 Primary Regulation ........................................................................................................... 34
3.6.2 Automatic Generation Control .......................................................................................... 38
3.7 Conclusion .......................................................................................................................... 44
Chapter 4 Primary and Secondary Frequency Control in Two Area Systems with
DFIG- Based Wind Turbine Support: .................................................................................... 45
4.1 Introductions: ...................................................................................................................... 45
4.2 Dynamic model - Primary Frequency Regulation with DFIG based Wind Turbine
two area controls: ..................................................................................................................... 47
4.3 Dynamic model of Secondary Frequency Regulation with DFIG based Wind Turbine - two
area controls: ........................................................................................................................ 49
4.4 Optimal Tuning of DFIG Based Wind Turbine Controller Parameters : ................................ 52
4.4.2 Secondary control analytical studies: ................................................................................ 62
4.5 Analytical studies: ............................................................................................................... 55
4.5.1 Primary control analytical studies: .................................................................................... 55
Chapter 5 Conclusion and Future Work ..................................................................................... 71
Bibliography ............................................................................................................................. 73
Appendix A............................................................................................................................... 76
Appendix B ............................................................................................................................... 77

viii
List of Figures

Figure 2.1 Drag device based on Persian design, used in the sail boat [26] ................................... 8
Figure 2.2 Vertical Axis wind turbine [27] .................................................................................. 9
Figure 2.3 Horizontal Axis wind turbine [28] .............................................................................. 9
Figure 2.4 Illustration of forces around the moving blade [30] ................................................... 12
Figure 2.5 WECS with fixed speed wind turbine ....................................................................... 15
Figure 2.6 WECS with fixed speed wind turbine and variable resistor........................................ 15
Figure 2.7 WECS with DFIG based wind turbine ...................................................................... 16
Figure 2.8 WECS with synchronous generator and convertor ..................................................... 17
Figure 2.9 Plots of simulation of wind speed step perturbation for induction type wind turbine .. 20
Figure 2.10 Plots of simulation of wind speed step perturbation for induction type wind turbine 21
Figure 3.1 Principle of DFIG Inertial Emulation Control [38] .................................................... 25
Figure 3.2 Power system dynamic model overview with mixed generation ................................ 26
Figure 3.3 DFIG-based wind turbine control based on frequency change [35], [39] .................... 27
Figure 3.4 Primary frequency regulation block diagram with DFIG based WT ........................... 28
Figure 3.5 Dynamic model for AGC studies with DFIG-based wind turbines ............................. 30
Figure 3.6 Optimal tuning of DFIG controller parameters for 20% wind penetration .................. 33
Figure 3.7 Primary frequency regulations with and without DFIG...34
Figure 3.8 Primary frequency regulations with and without DFIG35
Figure 3.9 Generator responses in primary frequency regulation with and without..36
Figure 3.10 Generator response in primary frequency regulation with and without DFIG ........... 36
Figure 3.11 Generator response in primary frequency regulation with and without DFIG ........... 37
Figure 3.12 DFIG generator response in primary frequency regulation....................................... 37
Figure 3.13 DFIG generator response in primary frequency regulation....................................... 38
Figure 3.14 DFIG-Based WT with AGC Simulation model in Simulink .................................. 39
Figure 3.15 AGC with 2% load increment with and without DFIG ............................................ 40
Figure 3.20 DFIG generation with AGC and 2% load increment ................................................ 43
Figure 4.1 linear model of frequency control two Area Systems................................................. 45
Figure 4.2 Linear dynamic model of primary frequency control of two area systems with DIFG 47

ix
Figure 4.3 Linear model of secondary frequency control of two area systems with DIFG-based
wind turbines ............................................................................................................................ 51
Figure 4.4.Schematic diagram for optimal tuning of DFIG controller parameters ....................... 53
Figure 4.5.Tuning DFIG controller parameters for 50% wind penetration .................................. 54
Figure 4.6.Simulink model for frequency response studies for two-area controller with ............. 56
Figure 4.7.Primary frequency regulation tie-line incremental power .......................................... 57
Figure 4.8. Primary frequency regulation for 2% load change area 1 .......................................... 58
Figure 4.9.Primary frequency regulation for 2% load change area 2 ........................................... 58
Figure 4.10. Primary frequency regulation for 2% load change area 1 ........................................ 59
Figure 4.12. Primary frequency regulation for 2% load change tie line ....................................... 60
Figure 4.14 Primary frequency regulation for 2% load change area 2 ......................................... 61
Figure 4.15 Primary frequency regulation for 2% load change tie line........................................ 61
Figure 4.16 Secondary frequency regulation for 2% sudden load tie line power ......................... 62
Figure 4.17 Secondary frequency regulation for 2% sudden load tie line power ......................... 63
Figure 4.22 Secondary frequency regulation for 2% sudden load change area-2 ......................... 65
Figure 4.27 Area error for 2% sudden load change in area-1 ..................................................... 68

x
List of Tables

Table 1.1 Wind farm projects and installations in Canada. [7][8] ................................................. 3
Table 1.2 Installed wind capacity in Ontario [9]........................................................................... 4
Table 1.3 Ongoing wind development projects in Ontario [9] ...................................................... 4
Table 1.4 Total wind generation capacity in EU by scenario and year (GW)[3] ............................ 6
Table 2.1 Available wind turbines in the market [29] ................................................................. 11
Table 3.1 Optimal parameters of the controller for different wind penetration ............................ 33
Table 4.1 Optimal DFIG controller settings parameters area-1 ................................................... 54
Table 4.2 Optimal DFIG controller settings parameters area-2 ................................................... 55

xi
Nomenclature
A State space matrix
B
1
Frequency bias factor area 1
B
2
Frequency bias factor area 2
C
p
Efficiency Coefficient
D load damping factor
D
1
load damping factor area 1
D
2
load damping factor area 2
E
K
Kinetic Energy
E
K0
Initial

Kinetic Energy
J Objective function
H

Power system inertia constant
H
e
Wind generation inertia constant
H
1

Power system inertia constant area 1
H
2

Power system inertia constant area 2
K
agc
AGC integral

control gain
K
df
Derivatives controller gain
K
I1
AGC integral

control gain area 1

K
I2
AGC integral

control gain area 2
K
p
Power system gain
K
pf
Proportional controller gain
K
p1
Power system gain

K
p2
Power system gain

K
wi
DFIG integral speed controller gain
K
wi1
DFIG integral speed controller gain area 1

K
wi2
DFIG integral speed controller gain area 2

K
wp
DFIG proportional speed controller gain
K
wp1
DFIG proportional speed controller gain area 1
K
wp2
DFIG proportional speed controller gain area 2

P Perturbation vector
P
mech
Wind turbine Mechanical Output
pu per unit
P
wind
Wind power

R Regulation droops

xii
R
r
Rotor Radius
R
1
Regulation droop area 1
R
2
Regulation droop area 1
s Laplace variable

T

Tie line synchronizing Coefficient

T
a
DFIG turbine time constant
T
a1
DFIG turbine time constant area 1

T
a2
DFIG turbine time constant area 2

T
h
conventional

generation governor time constant Transducer time constant
T
h1
conventional

generation governor time constant Transducer time constant area 1
T
h2
conventional

generation governor time constant Transducer time constant area 2

T
p
Power system time constant
T
p1
Power system time constant area 1

T
p2
Power system time constant area 2
T
r
Transducer time constant
T
r1
Transducer time constant area 1
T
r2
Transducer time constant area 2
T
t
conventional

generation turbine time constant Transducer time constant
T
t1
conventional

generation turbine time constant Transducer time constant arae1
T
t2
conventional

generation turbine time constant Transducer time constant area 2
T
w
washout filter time constant
T
w1
washout filter time constant for DFIG area 1
T
w2
washout filter time constant for DFIG area 2
V
rel
Relative wind Speed
V
tip
Tip Speed
V
wind
Wind Speed
X

State vector matrix
Angle of attack
W
Penetration index
. Blade Angle
air
Air density
Angle of Incidence angle of incidence between the plane of the rotor and V
rel
e
r
Rotor Frequency
e

Wind Turbine speed

xiii
e
max
Wind turbine cut-out speed maximum
e
mech
Angular velocity of rotor rotation
e
min
Wind turbine cut-out speed minimum
e
m,meas
wind turbine measured mechanical speed
e
s
System Frequency
turb
Angular velocity of the Rotor
e
1
Wind Turbine speed area 1
e
2
Wind Turbine speed area 2
e
1max
Wind turbine cut-out speed maximum area1
e
2max
Wind turbine cut-out speed maximum area2
e
1min
e
2min
Tip Speed Ratio
State space matrix
1
Tie line Voltage angle area1
2
Tie line Voltage angle area2
f Incremental power frequency
f
1
Incremental power frequency area1 change
f
2
Incremental power frequency area2 change
APD Incremental active power demand
APD
1
Incremental active power demand area1
APD
2
Incremental active power demand area 2
AP
f
Incremental value of total active power mixed generation
AP*
f
Incremental Wind Turbine power set point (reference) based on frequency and
speed change
AP*
f
Incremental Wind

Turbine power set point (reference) based on frequency
AP
g
Incremental value of conventional generation
AP
NC
Incremental Wind Turbine active Power output (non-conventional generation)
AP
NC1
Incremental DFIG active Power output area 1
AP
NC2
Incremental DFIG active Power output area 2
AP
NC,ref
Incremental Wind Turbine active Power reference (non-conventional generation)
AP
NC1,ref
Incremental Wind Turbine active Power reference area1

xiv
AP
NC2,ref
Incremental Wind Turbine active Power reference area2
AP
*
e
Incremental Wind Turbine power set point (reference) based on speed change
AP
12
Incremental power transferred from neighboring area (area 1 to area 2)
e Incremental wind turbine speed
e
1
Incremental wind turbine speed area 1
e
2
Incremental wind turbine speed area 2
X
1
Measured incremental frequency change (after Transducer)
X
1-1
Measured incremental frequency change for DFIG (after Transducer) area 1
X
1-2
Measured incremental frequency change for DFIG (after Transducer) area 2
X
2
Measured incremental frequency change (after wash out filter)
X
2-1
Measured incremental frequency change (after wash out filter) for DFIG area 1
X
2-2
Measured incremental frequency change (after wash out filter) for DFIG area 2
X
3
Incremental DFIG active power based on mechanical speed change
X
3-1
Incremental DFIG active power based on mechanical speed change area 1
X
3-2
Incremental DFIG active power based on mechanical speed change area 2

xv
Abbreviations

AGC Automatic Governor Control
ACE
1
Control error area 1

ACE
2
Control error area 2

CanWEA Canadian Wind Energy Association
DFIG Doubly Fed Induction Machine
GHG Green House Gas
EEA European Environment Agency
EU European Union
GW Giga-watt
MW Megawatt
MTP Main Tracking Power
rpm Rotation per minute
TSR Tip Speed Ratio
WECS Wind Energy Converter System

1
Chapter 1
Introduction
1.1 Motivation
The reduction of economic dependency on fossil fuel-based energy has been among the top
priority goals of regulators and their governments around the world. During recent years fossil
fuel resources are limited and have a significant adverse impact on the environment by raising the
level of CO
2
in the atmosphere and contributing to global warming. Among renewable sources of
energy, wind is one of the most promising technologies. It has already been in use for a
significant period of time and, compared to other forms of alternative energy resources, has the
greatest potential to reduce the conventional generation. The proportion of wind-based generation
in total energy production mix has been growing continuously in many parts of the world.
It has been reported [1] that renewable energy will provide as much as 10% of the
worlds energy supply by 2020, and will increase to as much as 50% by 2050. Canada has
outlined a future strategy for wind energy that would reach a capacity of 55,000 MW by 2025,
fulfilling 20% of the countrys energy needs [2]. The European Union plans to produce 22% of its
electricity from renewable resources by the year 2010 [3].
Although wind-based generation is fairly well known and has been used for several years,
but because of the low penetration levels of wind turbines, focus of the industry has been on
turbine protection aspects. However, with the increasing amounts of wind energy into the
network, new challenges with regards to the functioning of the current power grid are surfacing,
especially in area of grid stability, balance, security, planning, cross-border transmission, and
market design. A wind source is unpredictable. Therefore, efficient integration of large amounts
of variable sources of wind turbines into the existing electrical networks can significantly impact
the design, operation, and control of the network.
The work carried out and presented in this thesis is an attempt to study and examine the
role of variable speed based wind turbines, in particular the Doubly Fed Induction Generators
(DFIG), in frequency regulation and control with different levels of wind penetration into the
system. In particular the issue of optimal tuning of the DFIG control system in single-area and
two-area frequency control problems has been presented as a means to help improve the
performance of the system.

2
1.2 Current State of Wind Energy Globally
There is an increasing commitment from global leaders and policy makers to reduce Greenhouse
Gas (GHG) emissions. Efforts are being made to increase the contribution of renewable sources
of energy in the energy supply mix. Several countries have already formulated policy frameworks
to ensure that renewable resources play a major role in future energy scenarios.
1.2.1 Wind Power in Canada and Ontario
Early development of wind energy in Canada was primarily concentrated in Ontario, Quebec and
Alberta. Throughout the late 1990s and since the beginning of the 21
st
Century all Canadian
provinces have pursued wind power development to supplement their provincial energy grids.
Alberta built the first commercial wind farm in Canada in 1993. British Columbia was the last
province to add wind power to its grid with the completion of the Bear Mountain Wind Park in
November 2009 [4]. With steady population growth in a country, comes growth in energy
demand. Canada has seen wind power as a way to diversify its energy supply, and to help moving
away from its traditional reliance on fossil fuel based thermal plants and hydroelectricity. In
provinces like Nova Scotia, where only 12% of the electricity comes from renewable resources
[5], the development of wind energy projects can provide a measure of electricity security that
some jurisdictions are lacking. In the case of British Columbia, it is envisaged that wind energy
will help close the electricity deficit gap that the province is facing in the next decade, and reduce
its reliance on importing power from other jurisdictions that may not use renewable energy
sources. An additional 2004 MW of wind power is scheduled to come on-line in Quebec between
2011 and 2015 [6].
In 2008, the Canadian Wind Energy Association (CanWEA), a non-profit trade
association, outlined a future strategy for wind energy that envisages a capacity of 55,000 MW by
2025, fulfilling 20% of the countrys energy needs. The plan, Wind Vision 2025 [1], could create
over 50,000 jobs and represent around CDN $165 million annual revenue. CanWEAs target
would make the country a major player in the wind power sector and would create around
CDN$79 billion of investment opportunities. It would also eliminate an estimated 17 megatons of
GHG missions annually [1]. Table 1.1 shows the list of wind farm projects and installations in
Canada.

3
Table 1.1 Wind farm projects and installations in Canada. [7][8]
Province/
Territory
Current
Installed Power
(MW)
Planned under
construction
(MW)
Expected 2015
Capacity
(MW)
Alberta 590 409 999
British Columbia 102 107.7 272.7
Manitoba 104 138 242
New Brunswick 195 114 309
Newfoundland 54.4 0 54.4
Nova Scotia 59.3 244 303.3
Ontario 1161.5 647.2 1808.7
Prince Edward 151.6 0 151.6
Quebec 659 2671.5 3330.5
Saskatchewan 171.2 24.75 195.95
Yukon 0.81 0 0.81
Total 3,248.81 4,419.15 7,667.96

Ontario is at the forefront of wind generation in Canada with over 1,100 MW of installed
capacity on the transmission system. In Ontario, wind generation capacity has increased from 15
MW in 2003 to 1,100 MW in 2009; enough electricity to supply the needs of some 300,000
homes. Currently there are seven large-scale wind farms in operation in Ontario (Table 1.2).
These facilities are dispersed across the province which helps to mitigate the impact of local
weather conditions on wind energy production. There is also another 68 MW of wind generation
located within distribution service areas through Ontario Power Authority contracts [9].Table 1.3
shows the planned MW power generation projects in Ontario.

4

Table 1.2 Installed wind capacity in Ontario [9]
Wind Farm Capacity (MW) Operational
Amaranth ITownship of Melancthon 67.5 Mar. 2006
Kingsbridge Norfolk and Elgin Countries 39.5 Mar 2006
Port Burwell (Erie Shores) Norfolk and Elgin Countries 99 May 2006
Prince I Sault Ste Marie District 99 Sep 2006
Prince II Sault Ste Marie District 90 Nov 2006
Ripely Township of Melancthon 76 Dec 2007
Port Alma (Kruger) Port Alma 101.2 Oct 2008
Amaranth II Township of Melancthon 132 Nov 2008
Underwood (Enbridge) Bruce County 181.5 Feb 2009
Wollfe Island Township of Frontenac Island 197.8 Jun 2009

Table 1.3 Ongoing wind development projects in Ontario [9]
Wind Farm Capacity (MW) Expected Operation
Byran Wind Project 64.5 2010
Gosfield Wind Project 50 2012
Greenwich Wind Farm 99 2011
Kruger Energy Chatham 101.2 2011
Raleigh Wind Center 78 2010
Talbot Wind Farm 99 2011

1.2.2 Wind Power in the United State
At the end of 2009, the installed capacity of wind power in the United States was just over 35
GW [10] [11], making it the world leader ahead of Germany. Wind power accounts for about 2%
of the electricity generated in the United States [12].
Over 9,900 MW of new wind power capacity was brought online in 2009, up from 8,800
in 2008. In 2009, the added new capacity was enough to power the equivalent of 2.4 million
homes or generate as much electricity as three large nuclear power plants [13]. These new
installations place the US, on a trajectory to generate 20% of the nations electricity from wind
energy by 2030 [11]. Growth of MW capacity in 2008 channeled some $17 billion into the

5
economy, positioning wind power as one of the leading sources of new power generation in the
country, along with natural gas. New wind projects completed in 2008 accounted for about 42%
of the entire new power-producing capacity added in the US during the year [14] At the end of
2008, about 85,000 people were employed in the US wind industry[15], and GE Energy was the
largest domestic wind turbine manufacturer [15]. Wind projects boosted local tax bases, and
revitalized the economy of rural communities by providing a steady income stream to farmers
with wind turbines on their land [16] Wind power in the US provides enough electricity to power
the equivalent of nearly 9 million homes, avoiding the emissions of 57 million tons of carbon
each year and reducing expected carbon emissions from the electricity sector by 2.5% [17].
Texas, with 9,410 MW of capacity, has the most installed wind power capacity of any US state,
followed by Iowa with 3,053 MW [11]. The Roscoe Wind Farm (780 MW) in Texas is the
world's largest wind farm [18].

1.2.3 Wind Power in the European Union
In 2008, according to European Wind Energy Association (EWEA), there were 5,000 wind
turbines, with a total capacity of 64.93 GW in the European Union, generating 142 TWh of
electricity which took an investment of 11 billion euro [19]. In 2008, the European Union
produced 4.2% of its electricity from the wind [19][20]. As of 2009, the leading countries are
Germany (25 GW) and Spain (16.7 GW). The European Environment Agency (EEA) report [21]
states that potential of wind energy is sufficient to power Europe many times over. The report
highlights Europes wind power potential in 2020 to be three times greater than its expected
electricity demand, rising to a factor of seven by 2030 [22].
In 2008, 8.48 GW (8.1 onshore and 0.37 GW offshore) of wind energy capacity was installed in
the European Union compared to 27 GW in the world [19][22]. The market for European wind
power capacity grew significantly in 2006, according to statistics from the EWEA. A total of
7,588 MW of wind power capacity, worth some 9 billion Euros, was installed in the European
Union (EU) in 2006, an increase of 23% compared to 2005. Currently more than 25,000 wind
farms are operating throughout Europe, and capacity is expected to double by 2015. According to
EWEA, the industry will be worth $109 billion Euro by 2020. Research from a wide variety of
sources in various countries indicates that support for wind power is consistently between 70 and
80 per cent amongst the general public.
Wind power capacity estimates were reported for all EU twenty seven countries as well
as Norway, Switzerland, Croatia and some of the Balkan states [3] as given in table 1.4.

6

Table 1.4 Total wind generation capacity in EU by scenario and year (GW)[3]
Year 2005 2008 2010 2015 2020 2030
Low Growth Scenario 42.2 57.2 72.3 103.3 143.9 203.3
Medium Growth Scenario 42.2 66.5 90.0 143.7 205.8 279.6
High Growth Scenario 42.2 78.1 108.2 185.0 263.4 351.1

The EU targets to meet 22 percent of its electricity from renewable energy sources by the
end of 2010. Wind commercially is viable and as economically competitive renewable source, is
a major player in meeting this target.
1.3 Objectives of This Thesis
In light of the above discussions, it is apparent that wind energy is one of most favorable clean
energy alternatives to replace fossil fuel based energy globally. Fast growth and high penetration
of wind energy in existing networks requires full integration of wind turbines from different
perspectives such as operation, scheduling, control, and market aspects. Frequency control and
regulation service provision from wind generation is very important and challenging compared to
that from conventional sources of energy since availability of the wind energy source is not
controllable. It is evident that without a through strategy, full integration of wind energy with
high penetration would not be possible, and therefore studies are required. With that said,
following is a summary of the objectives of this work:
- Provide mathematical models of the network with mixed generation and various
penetration scenarios to examine the control strategy for primary and secondary
frequency control.
- Develop a mathematical model for network with mixed generation in single area
frequency control, and provide the method to improve the dynamic response of the
system.
- Develop a mathematical model for network with mixed generation in two area frequency
control, and provide the method to improve the dynamic response of the system.
- Compare the classical frequency control with mixed generation network, and examine
and compare the responses for various penetration scenarios.

7
1.4 Organization of This Thesis
Chapter 2 provides the Wind power overview and WECS modeling. Chapter 3 develops
mathematical model for single area control system with mixed generation, and tuning the wind
machine speed controller, in order to improve frequency response of the network for small
perturbation. Chapter 4 develops the mathematical model for two area control system with mixed
generation, and tuning of the wind machine speed in each area, in order to improve frequency
response of the network for small perturbation, and to minimize the area error and tie line
disturbance for various penetration scenarios. Chapter 5 draws the important contributions of the
thesis, and outlines the scope for future work.

8
Chapter 2
Wind Power overview
2.1 Status of wind Turbine Technology
Wind energy conversion systems can be classified into turbines that depend on aerodynamic drag
and turbines that depend on aerodynamic lift. The early Persian vertical axis wind turbine shown
in Figure 2.1 utilized the drag principal. The history of wind power shows a general evolution
from use of simple light devices driven by aerodynamic drag forces, to heavy, material-intensive
drag devices and finally to the use of light material, and efficient aerodynamic lift devices in the
modern era. The earliest known use of wind power, (sail boat form ancient Persia) used this
technology which had a significant impact on the pursuing wind power development.

Figure 2.1 Drag device based on Persian design, used in the sail boat [26]

Modern wind turbines are predominately based on aerodynamic lift. Lift devices use
blades that interact with incoming wind. The force resulting from the blade (airfoil) interaction
with the air flow consists of the drag force component in the direction of the flow and a force
component that is perpendicular to the drag, namely the lift. Magnitude of the lift force could be
several times larger than drag force depending on the design of the airfoil blade, which provides
the driving force for the rotor. Wind turbine based on aerodynamic lift can be further classified
according to the orientation of spin axis, as vertical axis turbines Figure 2.2 and horizontal axis
turbine Figure 2.3. In the vertical axis wind turbine, the main rotor shaft is set vertically. The
advantages of this arrangement are that generators and gearboxes can be placed close to the
ground and as shown in Figure 2.2, these devices do not need to be pointed into the wind
direction [27].

9

Figure 2.2 Vertical Axis wind turbine [27]

Figure 2.3 Horizontal Axis wind turbine [28]

Figure 2.3 shows a horizontal axis or propeller-type wind turbine in which the axis of the
rotor's rotation is parallel to the wind stream and the ground. The wind passes over both surfaces
of the airfoil shaped blade, faster over the longer (upper) side of the airfoil than the shorter side

10
(lower), hence creating a lower-pressure area above the airfoil with respect to the underside. The
pressure differential between top and bottom surfaces results in what is called aerodynamic lift.
Since the blades of a wind turbine are constrained to move in a plane with the hub as its center,
the lift force causes rotation about the hub. In addition to the lift force, a drag force perpendicular
to the lift force impedes rotor rotation. This approach currently dominates wind turbine
applications in the industry. As shown in Figure 2.3 this type of turbine consists of a tower, a
nacelle that is mounted on the top of the tower and contains a generator, gearbox and the rotor.
Different mechanisms exist for pointing the nacelle towards the wind direction, or move
the nacelle out of the wind in the case of high wind speed. For small wind turbines, the rotor and
the nacelle are oriented into the wind with the help of tale vanes. On the larger wind turbines,
nacelle and rotor are electrically yawed into or out of the wind, in response to a control signal
from wind vane. Most of horizontal axis turbines built today are with two- or three-blades,
although some may have more blades. The number of blades is directly linked to the Tip Speed
Ratio (TSR) which is the ratio of blade tip speed to wind speed, as given:
wind
turb
V
R e
= (2.1)
Where
turb
is angular velocity or frequency of rotation, R is radius of the aerodynamic rotor and
V
wind
is wind speed. The TSR is of vital importance in the design of wind turbine generators. If
the rotor of the wind turbine turns too slowly, most of the wind will pass undisturbed through the
gap between the rotor blades. Alternatively, if the rotor turns too quickly, the blurring blades will
appear like a solid wall to the wind. Therefore, wind turbines are designed with optimal TSR to
extract as much power out of the wind as possible. When a rotor blade passes through the air it
leaves turbulence in its wake. If the next blade on the spinning rotor arrives at this point while the
air is still turbulent, it will not be able to efficiently extract power from the wind. However, if the
rotor spans a little more slowly, the air hitting each turbine blade would no longer be turbulent.
Therefore the TSR is chosen so that the blades do not pass through turbulent air.
The optimum TSR depends on the number of blades in the wind turbine rotor. The fewer
the number of blades, the faster the wind turbine rotor needs to turn to extract maximum power
from the wind. A two-bladed rotor has an optimum TSR of around 6, a three-bladed rotor around
5, and a four-bladed rotor around 3. A well designed typical three-bladed rotor would have a tip
speed ratio of around 6 to 7. Three-bladed turbines provide lower noise, and better torque
characteristics compared with two-bladed wind turbines.

11
In tune with growth of the industry, the wind energy technology is also changing. One
apparent change is the shift towards offshore installations enabling the industry to envision large
and ambitious installations to take place.
Another trend in the industry is to go for larger machines since the industry is growing from MW
to multi MW scale. Bigger turbines are cheaper on per KW basis. Several manufacturers are
manufacturing turbines with 5 MW size. Some of these machines have rotors with 125 m blades
weighing around 9 tones. In another attempt to make offshore wind farms more profitable,
Norway plans to build the world's largest turbine standing 533 feet tall with a rotor diameter of
475 feet. It will also be the single largest generating unit with capacity of 10 MW, to power over
2,000 homes, making it three times more powerful than current turbines [25].

Table 2 1 Available wind turbines in the market [29]
Application Hub- height Diameter Power rating
Commercial 50-90 m 47-90 m 0.660-2.00 MW
Medium 35-50 m 13-30 m
Residential 18-37 m 1-13 m Below 30KW

2.2 Power from Wind
From a physical point of view, the static characteristics of a wind turbines rotor can be described
by the relationships between the total power in the wind and the mechanical power of the wind
turbine. These relationships are readily described starting with the incoming wind in the rotor
swept area. It can be shown that the kinetic energy of a cylinder of air of radius R travelling at
wind speed V
wind
corresponds to a total wind power P
wind
within the rotor swept area of the wind
turbine. This power, wind, can be expressed by (2.1) as given below [30],

3 2
2
1
wind air wind
V R P t = (2.1)
where
air
is the air density (1.225 kg/rn
3
), R is the rotor radius and V
wind
is the wind speed.
It is not possible to extract all the kinetic energy from the wind since this would mean
that the air would stand still directly behind the wind turbine. This would not allow the air to flow
away from the wind turbine, and clearly this cannot represent a physical steady-state condition.
The wind speed is only reduced by the wind turbine, which thus extracts a fraction of the power
in the wind. This fraction is expressed as the power efficiency coefficient, C
p
, of the wind turbine.
Therefore the mechanical power output of the wind turbine P
mech
considering the definition of C
p

can be stated as given by (2.2) [30],

12

wind p mech
P C P = (2.2)

3 2
2
1
wind air p mech
V R C P t = (2.3)
It can be shown that the theoretical static upper limit of Cp is 16/27 (approximately
0.593); which is the maximum theoretically possible value (approximately 59%) that can be
extracted from the kinetic energy of the wind. This is known as Betzs limit [31]. For comparison,
a modern three bladed wind turbine has an optimal Cp value in the range of 0.52-0.55 [31] when
measured at the hub of the turbine.
From a physical point of view the power, P
mech
that is extracted from the wind will
depend on rotational speed, wind speed and blade angle, . Therefore, P
mech
and C
p
are functions
of these parameters.
) , V , ( f P
wind turb mech
| e = (2.4)
The forces of the wind on a blade section and thereby the possible energy extraction will
depend on the angle of incidence between the plane of the moving rotor blades and the relative
wind speed V
rel
(see Figure 2.4) as seen from the moving blades. Simple geometrical
considerations, which ignore the wind turbulence created by the blade tips show that the angle of
incidence is determined by the incoming wind speed V
wind
and the speed of the blade. The blade
tip is moving at speed V
wind
, equal to (
turb*
R). This is illustrated in Figure 2.4. The highest values
of C
p
are typically obtained for values in the range 8 to 9 (i.e. when the tip of the blades move 8
to 9 times faster than the incoming wind). This means that the angle between the relative air
speed as seen from the blade tip and the rotor plane is rather a sharp angle. Therefore, the angle of
incidence is most conveniently calculated as:
)
v
arctan( ) arctan(
turb
wind
e
= =
1
(2.5)
R V
turb tip
e =
wind
V
wind tip rel
V V V + =

Figure 2.4 Illustration of forces around the moving blade [30]

13
where V
tip
= tip speed;
turb
= turbine rotational speed; R = rotor radius; V
rel
= relative wind
speed; V
wind
= wind speed; = angle of attack; = angle of incidence between the plane of the
rotor and V
rel
; = blade angle

On modern wind turbines, it is possible to adjust the pitch angle of the entire blade
through a servo mechanism. If the blade is turned, the angle of attack between the blade and the
relative wind V
rel
will be changed accordingly. Again, it is clear from a physical perspective that
the forces of the relative wind on the blade, and thereby the energy extraction, will depend on the
angle of attack between the moving rotor blades and the relative wind speed V
rel
as seen from
the moving blades. Hence C
p
can be expressed as a function of and :
) , ( f C
p
| = (2.5)
C
p
is a characteristic of the wind turbine that can be obtained by blade element method, look-up
table, or analytical approximation.

2.2.1 Pitch control [31]
Whenever the average of wind speed changes over a propeller, the propellers pitch needs to be
controlled. Commercial wind turbines are designed to produce optimum power with 15 m/s of
wind speed. However, the wind speed always fluctuates up and down around this optimum. To
generate the optimum power, the turbine blades have to adjust accordingly. This adjustment
comes from turning the blades around their longitudinal axis (to pitch). When the wind speed
decreases, the blade pitch is adjusted such that it exposes more surface area to the wind.
Conversely, when wind speed picks up, the blade pitch is adjusted such that it exposes less
surface area to the wind. If a blade is not designed for stall, increased wind speeds will force the
rotor to turn faster without a pitch control mechanism. The pitch mechanism allows the wind to
flow around the blade as smoothly as possible. To do this, air particles cannot hit the blade head
on, rather they must flow almost tangent to the blade just as in an airplanes wing operating in the
air.
There are two kinds of pitch control mechanism. The first is called "Active Pitch
Control" where the rotor blades turn around their longitudinal axis (to pitch) by a computer
controlled mechanism. This type of pitch control requires expensive equipment; however it
provides good pitch control. Active pitch controls are used in one third of the large turbines
currently installed. The second pitch control mechanism is called "Passive" or "Stall Pitch

14
Control". In this type of design the blade does not rotate around its longitudinal axis, but is
designed such that it naturally creates a stall and low tip speeds. This type blade requires precise
blade design and structurally strong towers.

2.3 Wind Energy Converter System Functional Structure

One of the important components in the WECS is the gearbox placed between the main shaft and
the generator. Its task is to increase the rotational speed of the rotor blades to the generators
rotation speed of 1000 or 1500 revolutions per minute (rpm). Capacitor banks are used to provide
the reactive power required by the induction generator power electronic converters used to
facilitate the coupling of wind turbine with variable power source to the grid network with the
constant frequency. Therefore from the functional point of view WECS could be classified into
WECS with fixed speed wind WECS with variable-speed wind turbines.

2.3.1 WECS with Fixed Speed Wind Turbines Type A
Normally for the fixed speed wind turbine, induction generator types A [30] squirrel-cage
induction generator is used as shown in Figure 2.5. The rotor of a fixed-speed wind turbine
rotates at a fixed speed determined by the frequency of the grid, gear ratio, and the pole pairs
of generator. A fixed-speed wind turbine is connected to the grid through a soft-starter. Since
the induction generator absorbs the reactive power from the grid, a capacitor bank is
necessary to provide for the reactive power compensation. A gear box is used to transform
power from the turbine with lower-rotational speed to the generator rotor with high-rotational
speed. The generator terminal voltage is increased with a step-up unit transformer to a
medium voltage level.
Although this type of wind turbine has the advantage of being simple and relatively cost-
efficient compared to other types of wind turbines, but the reactive power cannot be
controlled, therefore any sever short-circuit near the wind farm may cause significant voltage
dip in the weak network.
Since the rotational speed of this type is limited to a very narrow range determined by the
slip of the induction generator, any fluctuations in mechanical torque due to wind speed
variation will impact the network where the wind farm is connected.

15

Gear
Box
Point of coupling
Flow of wind
Turbine Blade
Gen
Capacitor
Bank
Grid

Figure 2.5 WECS with fixed speed wind turbine

Risk of loss of synchronism in case of voltage dips is high because both field and stator
windings are supplied from the same bus.
2.3.2 WECS with Variable Speed Wind Turbines Type B
In the case of an induction generator with a variable rotor resistance (Type B wind turbine)
Figure 2.6, it is possible to vary the speed over a somewhat wider range. However, the speed
range is still rather limited and the speed cannot be controlled directly. Hence, looking at it
from a control system perspective, this type of wind turbine may essentially be considered as
a fixed-speed wind turbine.

Gear
Box
Point of coupling
Flow of wind
Turbine Blade
Gen
Capacitor
Bank
Variable Resistor
Grid

Figure 2.6 WECS with fixed speed wind turbine and variable resistor

2.3.3 WECS with DFIG Based Wind Turbine
Doubly-Fed Induction Generator (DFIG) drives Figure 2.7 and the full load converter connected
generator drive are the two most frequently applied variable speed generator drive concepts.

16

Gear
Box
Grid
Point of coupling
Flow of wind
Turbine Blade
Gen
DC Link
AC
AC
DC
DC
Voltage Source converter

Figure 2.7 WECS with DFIG based wind turbine
Other variable-speed generator drive types are also available in the market, but currently they are
not used for wind turbines. It would probably be possible to use written pole synchronous
generators for instance, as a way of obtaining variable-speed capability in a generator drive. In
short, theoretically all types of variable speed high-power drives electrical, mechanical or
hydraulic with an electrical generator somewhere in the drive system can be used in wind turbine
systems.

2.3.4 WECS with Full Converter
Variable-speed generator drives enable the wind turbine control system to adapt the rotational
speed of the rotor to the instantaneous wind speed over a relatively wide range. The electrical
system has a fixed frequency though. A generator drive connecting a variable-speed mechanical
system to a fixed frequency electrical system must therefore contain some kind of a slip or
decoupling mechanism between the two systems
In variable-speed wind turbine with full converter, Figure 2.8, the generator can be a
squirrel-cage induction generator or a synchronous generator which is connected to the grid
via a power electronic converter as shown in Figure 2.8. The entire power output from
generator goes through the converter; therefore the converter is rated at full power. The
voltage level and the reactive power can be regulated by using power electronic converters.

17

Gear
Box
Grid
Point of coupling
Flow of wind
Turbine Blade
DC Link
AC
AC
DC
DC
converter
Gen

Figure 2.8 WECS with synchronous generator and convertor

2.4 Literature Review
Several researchers have carried out work on DFIG-based wind turbines to demonstrate their
possibility to provide separate active and reactive power control. Therefore, DFIG-based wind
turbines are the preferred choice for network operators in comparison to fixed-speed machines in
which active and reactive power control is not independent.
In this section a modest attempt is made to review some of the pertinent research papers
related to this study with emphasis on DFIG-based wind turbines and their role and contributions
to system frequency control.
In [32], the concept of releasing the kinetic energy of a DFIG-based wind turbine when
the frequency of the power system is reduced in order to prevent the reduction of system inertia is
presented. The paper suggests the use of additional power set-point in the control system of the
wind turbine. This additional power reference is calculated based on the rate of frequency change
and inertia of the network. The additional power released by the DFIG-based wind turbine
supports the primary frequency regulation and thus prevents reduction of system inertia.
In [33], the experimental results of engineering and design of a DFIG-based wind turbine,
using back-to-back PWM voltage-source converters in the rotor circuit, is presented. A vector-
control scheme is used for the supply-side of the PWM converter to maintain the quality of the
supply source by drawing sinusoidal current from the network. The same scheme in the rotor side
provides optimal speed tracking for a wide speed range of operation.
In [34], a model is developed to represent the two most important variable-speed wind
turbine concepts in power system dynamic simulations. The first concept is the use of a DFIG
with a back-to-back voltage source converter feeding the rotor winding. And the second concept
is a direct-drive synchronous generator which is grid coupled through a diode rectifier and
voltage source converter or through a back-to-back voltage source converter.

18
In [35], impact assessment of high penetration of wind energy to the Irish network is
presented. Because of the small size of the network there could be significant impacts of small
wind generation penetration. The paper studies two scenarios wherein all or none of the wind
turbines provide an inertial response following a frequency transient.
In practice, the truth lies somewhere in between, and wind farms typically comprise a
mix of fixed-speed turbines which can provide only limited inertial support and variable-speed
machines which do not provide inertial support if they are not equipped with auxiliary inertial
feedback loop [2]. It has been shown in [6] that if wind turbines provide no inertia in times of low
system demand, the robustness of the system is jeopardized. On the contrary, if wind turbines can
participate in system inertia provisions, the frequency effect could be significantly mitigated.
In [36], a control scheme is proposed that allows DFIG-based wind turbines to participate
effectively in system frequency regulation. In this approach, wind turbines operate according to a
de-loaded optimum power extraction curve such that the active power provided by each wind
turbine increases or decreases during system frequency changes. The control strategy defined for
the wind turbine to supply primary frequency regulation capability exploits a combination of
control of the static converters and pitch control, adjusting the rotor speed and the active power
according to the de-loaded optimum power extraction curve. Results obtained considering a small
isolated power system are presented to demonstrate the effectiveness of the approach, the
drawback of this method is the loss of revenue accrued because the wind generator is not able to
generate its maximum power.
In [37], the inertial response of wind turbines employing induction machine based
generators is presented. A model of a Field-Oriented Controlled DFIG, based on a fifth-order
induction generator model is described. The model is implemented in a reference frame that
allows the factors affecting the inertial response of the DFIG to be easily examined. A
comparison between the inertial response of a squirrel-cage and DFIG-based wind-turbine
generator is performed using the developed models. It is found that the inertial response of a
DFIG employing field-oriented controlled (FOC) strongly being influenced and effected by band
width of rotor current control.
In [38], the concept of extracting the kinetic energy of DFIG machines by reducing the
speed of the rotor is presented, and details of an addition power set-point in the DFIG, which is
used in order to emulate the inertia of the wind turbine, is formulated. It is also observed that, as a
result of supporting the frequency, the rotational speed of the DFIG decreases and therefore the
power output drops considerable when the frequency control support is ended. This drop in power
is undesirable, especially when all DFIG-based wind farms demonstrate this behavior. The

19
authors conclude that variable-speed wind turbines are able to support primary frequency control
and emulate inertia by applying additional control loops. For that purpose, the kinetic energy
stored in the hidden inertia of the turbine blades is used.

2.4.1 Assumptions and constraint in participation of DFIG in frequency regulation

In this study it has been assumed that the mechanical power of DFIG based wind turbine is
constant (wind speed) and therefore uncertainty of wind power generation has not been
considered. It is also assumed that in the power system with mixed generation, conventional
generation is able to supply the additional demand during the disturbances and increase the
production as needed. Maximum Power Tracking (MPT) system which tracks the optimal
mechanical speed and permits temporary deviation from optimal speed is described in Chapter 3
and 4.
2.5 Simulations
2.5.1 Active Pitch Control

Figure 2.9 and 2.10 presents the behavior of induction type generator type A, being driven by a
wind turbine as described in section 2.3.1. The turbine is controlled by a wind governor, during
the wind speed step perturbation. Simulation is developed with PSCAD professional program
wind turbine and governor type MOD2 (generic model), which is available in the library of this
program is used.
Before t=50 sec wind speed is equal to optimal wind speed 15 m/s and power output of
the wind farm and mechanical torques as shown in Figure 2.10 are in the steady state condition.
In t=50 sec after about 2 m/s wind speed perturbation is applied and therefore the wind speed
raises to 17 m/s. thereafter pitch control system attempts to maintain the same power output by
controlling the pitching angle, the output with about 20 second of perturbation returned to steady
state by raising the from about 15.6 deg to 18 deg. The active power of induction machine
towards to network is shown with the negative sign.
Figure 2.10 similar to 2.9 presents the wind speed reduction steps and attempts of control
system to maintain the wind farm output constant.

20

Figure 2.9 Plots of simulation of wind speed step perturbation for induction type wind turbine

21

Figure 2.10 Plots of simulation of wind speed step perturbation for induction type wind turbine

22
2.6 Conclusion

This chapter presents a back ground review of the evolution of wind turbine technology and
current status in its development. Thereafter, the mathematical relations for power from wind
turbines are discussed and the importance of pitch control is highlighted. Subsequently, various
classes and type of Wind Energy Converter Systems (WECS) are discussed in brief. A moderate
review of literature pertaining to frequency control contribution of DFIG based wind generations
is performed. Finally PSCAD simulation is performed that lay done the foundation for the
research which has done in the next two chapters.

23
Chapter 3
Primary Frequency Regulation and AGC in a Single-Area System with
DFIG-Based Wind Turbine
3.1 Introduction
In spite of the increasing penetration of wind turbines into the power grid, the frequency
regulation and AGC tasks are mainly undertaken by conventional generation units. The goal of
frequency regulation and AGC is to keep the frequency within specified limits through primary
and secondary control of governor.
Traditionally, Wind Energy Conversion Systems (WECS) do not participate in system
frequency support. Hence, wind turbines do not increase or decrease their production when the
frequency falls or rises respectively, meaning that they do not contribute to the system inertia.
With high rate of growth in wind generation capacity, its contribution to total generation mix is
increasing but the total system inertia participating in frequency control is decreasing to
inadequate levels for appropriate system frequency recovery following a power imbalance or grid
disturbance. Therefore, further penetration of wind energy in the network calls for contributions
from wind generators to frequency regulation and AGC in order to prevent reduction of total
system inertia of the network.
Not much research has been reported on analyzing the effects of wind generator
participation to frequency regulation and AGC in power systems. One of the methods discussed
in several papers [32], [36], [38] with earlier types of turbines (such as fixed-speed turbines)
propose preventing the wind turbines from supplying their maximum available power in normal
situations so as to maintain a reserve margin that can be utilized for frequency control. More
recently, extracting the kinetic energy stored in the mechanical system of wind turbines with
variable-speed generators has been the subject of research [34], [35], [36]. The control system
based on this concept, utilizes the ability of DFIG-based wind turbines to produce power with
variable mechanical speed and extracting the kinetic energy to support the primary frequency
regulation, has been based on the Wind Turbine Inertia Control model [32]. The block diagram of
such a control system is shown in Figure 3.1.
Similar to conventional generators, wind turbines have a significant amount of kinetic
energy stored in the rotating mass of their rotor and blades. An important feature of DFIG
machines is the possibility for their active and reactive power outputs to be controlled as required
by system operators. Although the steady-state active power delivered to the grid by a WECS
depends on the wind speed, the power can be dynamically controlled to a certain extent by

24
utilizing the stored mechanical energy. This is due to the capability of these machines to work at
asynchronous speeds. This way, the wind energy will transfer efficiently for a given wind speed,
while the mechanical stress is relieved to some extent.
DFIG machines with slip-ring rotors are generally designed with three ac windings in the
stator and rotor. The stator is normally connected directly to the grid and the rotor is supplied via
a frequency converter. Thus the angular velocity of stator rotating field can be written as,
r
r
mech
s
s
p p
e
e =
e
(3.1)
In (3.1), p
s
and p
r
denote the number of stator and rotor poles respectively. As per (3.1),
the system frequency e
s
is equal to sum of the angular velocity of mechanical rotation (e
mech
) and
rotor current frequency (e
r
). Depending on the direction of the supply frequency, this machine
can operate in over- or under- synchronous speed. The amount of kinetic energy released from the
shaft of the wind turbine when the speed is reduced, given by AE
k
, can be formulated from the
relations below:
2
0 0
2
1
mech k
J E e = (3.2)
2
1
2
0 1 0
2
1
2
1
mech mech k k
J J E E e e = (3.3)
|
|
.
|
\
|
= A
2
0
2
1
0
1
mech
mech
k k
E E
e
e
(3.4)
E
k0
is dependent on wind speed which varies between zero to 1.0 per unit and e
mech1

cannot be smaller than the minimum mechanical rotational speed of the DFIG-based wind
turbine. Furthermore, the instantaneous power extracted from the wind turbine cannot exceed the
maximum allowed value, as per manufacture data of the machine. These constraints, as stated in
(3.5)-(3.8), have to be regarded in the control system.

) (
0
Speed Wind f E
k
= (3.5)
) ( 0 . 1 0
0
Speed Wind f pu E
k
= s s (3.6)
1
min
mech mech
e e s (3.7)
max k k k
E E E s + A
0
(3.8)
Figure 3.1 depicts the controllers of DFIG-based wind turbines that try to keep the
turbine at its optimal speed in order to produce the maximum power. The controller provides a
power set-point (P
*
e
) that is based on measured speed (e
m,meas
) and measured electrical power

25
(P
el,meas
). The power set-point is an input for the converter control that realizes the torque and
power by controlling the generator rotor currents. An additional control signal is added (P
*
f
)
which adapts the power set-point as a function of the deviation and the rate of change of the grid
frequency. The emulated inertia of the additional control signal is proportional to its controller
parameters (K
df
and K
pf
). The support to primary frequency control depends on when this
additional loop is activated, as the grid frequency exceeds certain limits, by adding this signal to
the torque equation to set the torque demand. As the system frequency drops, the set point torque
is increased and the rotor slows down and kinetic energy is released.
P
dt
d
K
df
K
pf
PI
-
+
-
dt
df
f A
m,ref
f
grid
P
el,meas
f
ref
+
-
-
+
+
m,meas
m
*
dt
df
p A
*
f
p
A
A
*
f
p A
*
p
e
A
*
f
p
e
A
Converter
Control

Figure 3.1 Principle of DFIG Inertial Emulation Control [38]

The power reference point (AP
*
f
) has two components, AP
*
f
the additional reference
point based on frequency changes as described earlier and AP
*
e
which is based on optimum
turbine speed as a function of wind speed, as given below [39]:
f K
dt
df
K P
pf df
*
f
A = A (3.9)
dt ) ( K ) ( K P
*
wi
*
wp
*
}
e e + e e = A
e
(3.10)
* *
f
*
f
P P P
e e
A A A + = (3.11)

K
df
and K
pf
are the controller gains for the derivative and proportional controllers
respectively. Considering the two components of the DFIG power set-point (3.11), AP
*
e
changes

26
in relative terms, slowly, compared to the frequency derivation power set-point AP
*
f
. Therefore,
at t=0 when disturbance occurs, we can assume AP
*
e
to be zero. Also, considering the instant
conversion of set point to power by converter, we could assume AP
NC
= AP
*
f
and therefore,
0 + A = A
e
*
f
*
f
P P (3.12)
f K
dt
df
K P P P
pf df
*
f
*
f NC
A = A = A = A
e
(3.13)
3.2 DFIG-Based Wind Turbine Control Model
Figure 3.2 shows the overall transfer-function block diagram of a power system comprising a
conventional generator providing frequency regulation as well as a non-conventional DFIG-based
wind turbine generator contributing to frequency regulation. The incremental active power
demand APD subtracted from the incremental values of conventional generation AP
g
and wind
generation AP
NC
equals the power transferred from the neighboring area, AP
12
as given in (3.9).
f , NC g
P PD P P P A = A A A + A
2 1
(3.14)

R
1
t
sT + 1
1
h
sT + 1
1
P
P
sT
K
+ 1

g
P A
h
P A f A
+

+
12
P
ref
P A
PD A
NC
P
Conventional
Generation
Electrical
System
Non-conventional
Generation
+
f
P A

Figure 3.2 Power system dynamic model overview with mixed generation

From Figure 3.2 and (3.8), (3.9) and (3.13),
fD
H
T
P
2
= (3.15)
D
K
P
1
= (3.16)

27
f D PD P P P f D P
dt
f d
f
H
NC g f
A A A A + A = A A =
A
12
2
(3.17)
A + A A A =
A
|
|
.
|
\
|
+

*
12
*
2
) (
2
D
pf g
H
df
D K PD P P
dt
f d
K
f
H
(3.18)

Figure 3.3 shows the dynamic model adopted for the study of frequency regulation [3]
with DFIG-based wind turbine generator. This model is in essence taken from the DFIG inertia
emulation model presented in Figure 3.2.

r
sT + 1
1
R
1
w
w
sT
T
+ 1
a
sT + 1
1
+
wp
K
s
K
wi
e
sH 2
1
*
e A
max
e A
min
e A
f A
ref NC
P

A
+
Frequency Measurement Washout Filter Droop
Mechanical inertia
Speed Controller
WT model
*
f
P A
*
P
e
A
1
X A
2
X A
e A
max NC
P

0
) Out ( NC
P A
max
e
min
e
3
X A

Figure 3.3 DFIG-based wind turbine control based on frequency change [35], [39]

The contribution of the DFIG towards system inertia can be seen in (3.17). The
differences between the models described in Figure 3.1 and that in Figure 3.3 [3] is the additional
reference power setting which is build based on the change in frequency using a washout filter
with time constant T
w
, which relies on a conventional primary regulation performance in a
transient.
2
1
X
R
P
*
f
A = A (3.19)
R is the droop constant as used conventionally and
2
X A is the frequency change measured
where the wind turbine is connected to the network.

28
3.3 Dynamic Model of Primary Frequency Regulation with DFIG-Based Wind
Turbines

The dynamic performance of a power system comprising a conventional prime-mover model, a
non-reheat type turbine and a DFIG-based wind turbine is represented by the small-perturbation
model shown in Figure 3.4 [3]. This model simulates the primary frequency regulation after a
disturbance and includes conventional system parameters such as the load damping factor (D), the
regulation droop (R), the inertia H, governor time constants T
h
and T
t
of the system equivalent
unit (governor and turbine).
The system behavior depends on the choice of network parameters and DFIG-based wind
turbine speed controllers are K
wp
and K
wi
. If several generators are connected to the system, the
equivalent regulation droop can be determined from the (3.20) as follows:
n
R
.....
R R R R
1 1 1 1 1
3 2 1
+ + + + = (3.20)

t
sT + 1
1
h
sT + 1
1
P
P
sT
K
+ 1

Pg A Ph A
f A
+

+
NC
P A
ref
P A
PD A
sTr + 1
1
R
1
sTa + 1
1
+

wp
K
s
K
wi
e
sH 2
1
+
+
*
e A
+
e A
NC
P A
ref , NC
P A
w
w
sT
T
+ 1
R
1
max
e
min
e
DFIG Mechanical inertia
model
DFIG Speed regulator
Additional power set point
based on frequency
change for DFIG
AGC secodary
regulation
Conventional Primary
regulation
Electrical Network
Wind Turbine 1
X A
2
X A
3
X A
P
max
e
min
e

Figure 3.4 Primary frequency regulation block diagram with DFIG based WT

29
In Figure 3.4, AP
h
is the incremental hydraulic governor valve position change, is
incremental speed of the wind turbine, f is the incremental frequency change, X
1
is frequency
increment after being measured by transducer, X
2
frequency change after the washout filter and
X
3
is the incremental change in DFIG integral speed control. The incremental model is obtained
by linearizing the system around a nominal operating point. The dynamic model in state-space
form can be obtained from the transfer-function model given below:
p X A
dt
X d
I + =
(3.21)
X is the state-vector, p is the perturbation and A, I are state and perturbation matrices
respectively. The state-equations (3.21) can be written in expanded vector-matrix form as
follows:
| |
T
NC 2 1 g h
P X X X f P P X e A
3
=
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
=
0 0 0 0 0 0 0
2
1
2
1
0 0 0 0
1 1 1
0 0
0 0 0
1 1 1
0 0 0 0
1 1
0 0 0 0
1
0 0 0 0 0 0
0 0 0 0 0
1
0
wi
e e
a a
wp
a a
w r
r
p
p
p p
p
t t
h h
K
) H * ( ) H * (
0 0
T T
K
T ) T * R (
0 0
T T Tr
0 0
T Tr
0 0
T
K
T T
K
0
T
1 -
T
1
) RT ( T
1 -
A
| |
T
D
P p 0 0 0 0 0 0 0 A =
T
p
p
T
K
(
(

= 0 0 0 0 0 0 0 I

Dynamic simulation for primary regulation with and without the DIFG-based wind turbine
is carried out. The result is shown in Section 3.6.1.

30
3.4 Dynamic Model of AGC with DFIG-Based Wind Turbines

The small perturbation dynamic model considered for the single-area power system for AGC
studies is shown in Figure 3.5. The model is similar to that described in the context of primary
control except that a secondary control is added at the governor reference to regulate the set-
point.
t
sT + 1
1
h
sT + 1
1
p
p
sT
K
+ 1

g
P A
h
P A f A
+

+
NC
P A
ref
P A
PD A
r
sT + 1
1
R
1
a
sT + 1
1
+

wp
K
s
K
wi
e
sH 2
1
+
+
*
e A
+
e A
s
K
AGC
ref , NC
P A
w
w
sT
sT
+ 1
R
1
max
e
min
e
DFIG Mechanical inertia
model
DFIG Speed regulator
Additional power set point
based on frequency
change for DFIG
AGC secodary regulation
for the Area
Conventional Primary
regulation
Electrical Network
Wind Turbine
NC
P A
1
X A
2
X A
3
X A
P

Figure 3.5 Dynamic model for AGC studies with DFIG-based wind turbines

The dynamic model in state-space form is developed from the transfer-function
representation by defining the state-vector X, given by:
| |
T
NC 2 1 ref g h
P X X X f P P P X A e A
3
=
p X A
dt
X d
I + =
(3.20)
X is the state-vector and p is the perturbation vector. The matrices A and I are state and
perturbation matrixes respectively. The state-equation (3.20) can be written in expanded vector-
matrix form as follows:

31
| |
T
NC 2 1 ref g h
P X X X f P P P X A e A
3
=
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
=
0 0 0 0 0 0 0 0
2
1
2
1
0 0 0 0 0 0 0
1 1 1
0 0 0 0 0
0 0 0
1 1 1
0 0 0
0 0 0
1 1
0 0 0 0
0 0 0 0
1
0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
0 0 0 0 0
1
Wi
e e
a a
wp
a a
w
p
p
p p
p
AGC
t t
h h h
K
) H * ( ) H * (
T T
K
T ) T * R (
T Tr Tr
Tr Tr
T
K
T T
K
K
T
1 -
T
1
) RT ( T
1
0
T
1 -
A
| |
T
PD p 0 0 0 0 0 0 0 0 =
T
P
P
T
K
(

= I 0 0 0 0 0 0 0 0

Dynamic simulation is carried out with and without the DIFG-based wind turbine for
sudden load increase.
3.5 Optimal Tuning of DFIG-Based Wind Turbine Controller Parameters
The objective of this section is to determine the optimal settings of the speed control parameters
of the DFIG-based wind turbine so as to improve its participation in frequency control in
response to system disturbances. The Integral of Squared Error (ISE) [40] technique is used for
obtaining the optimum values of K
wp
and K
wi
. In the state-space model (3.20), a step perturbation
of 0.02 per unit is considered, which remains constant during the dynamics of the process.
In order to examine the impact of the growth of wind energy in the total energy supply
mix of the system, a Penetration Index (
W
) is defined as follows:

100 =
sources all from generation Total
generation wind Total
W
o

32
Four cases of DFIG penetration are considered:
-
W
= 5%
-
W
= 10%
-
W
= 20%
-
W
= 50%

The initial values of the state variable, which in this study, are the incremental changes
from their nominal values, will be zero since the system is stable. For t, the states will
approach a new steady-state and thus at any time t, the state-vector X(t) represents the deviations
of the state variable from its steady-state value. Therefore, the dynamic performance of the
system, for a given set of control parameters, can be measured using a quadratic performance
index denoting the squared of the deviations of the state variables, as follows:

| |dt ) t ( X ...... ) t ( X ) t ( X ) t ( X J
n
}
+ + + + =
0
2 2
3
2
2
2
1
(3.22)

In this work, the square of the frequency deviation errors is considered as the
performance index for tuning the controller parameters, as stated below.

| |dt ) t ( f J
}
=
0
2
A
(3.23)

For computational purposes, the performance index J is calculated by summing the
discrete values at very small intervals, over a considerable period until the steady-state is
achieved, as given below.

| |
=
+ =
K
k
) t k t ( f J
1
0
2
A A
(3.24)

The optimal values of the DFIG speed controller parameters K
WP
and K
WI
are determined
by searching for the minimum value of J. At first, K
WI
is fixed at a certain value and K
WP
is varied
over a wide range. It is observed that J decreases as K
wp
is increased, attains a minimum, J
Min
, and
then increases again. The values of K
wp
and K
wi
that yields the minimum of J
Min
(J
*
Min
), are the
optimal controller parameters. Figure 3.6 shows the plot of J versus K
WP
for several values of K
WI

33
considering ,20% wind turbine penetrations in the system respectively following a 0.02 per unit
step load perturbation.

Figure 3.6 Optimal tuning of DFIG controller parameters for 20% wind penetration

It can be observed from Table-3.1 that the optimal settings of the controller gains are
slightly affected by the wind penetration levels and the performance is expected to worsen
with a higher penetration of wind generation, since J
*
Min
is higher for 50% penetration than
that with 20% penetration.

Table 3.1 Optimal parameters of the controller for different wind penetration
Wind Penetration K
wi
K
wp
J
*
Min

5% 0.05 1.00 0.0151
10% 0.05 1.00 0.0161
20% 0.1 1.23 0.1209
50% 0.1 1.98 0.1235

34
3.6 Analytical Studies
3.6.1 Primary Regulation

Simulations are carried out considering the dynamic model presented in Figure 3.4 for a 0.02 per
unit load perturbation with and without DFIG wind turbine to examine its contribution in primary
frequency regulation, different plots for theses simulations are presented in Figure 3.7 to 3.11.
During the simulations it has been assumed that DFIG-based wind turbines are in their optimal
mechanical speed with the maximum power obtainable form the wind, and wind speed remain
constant during the simulation.
Figure 3.7 to 3.8 presents the frequency plots following a 0.02 per unit step load
perturbation with and without the DFIG. It is observed that the frequency response following the
disturbance is improved, by way of lower frequency peak excursion when DFIG participation is
considered. It is also observed that when the wind penetration level increases from 5% to 50%,
although the lower frequency peak excursion is improved by higher penetration but the frequency
response deteriorates by way of increased settling time and higher steady-state frequency error,. It
can be generally concluded that with lower penetrations 5 % there is no significant improvement
compare to no-DFIG case.

Figure 3.7 Primary frequency regulations with and without DFIG

35

Figure 3.8 Primary frequency regulations with and without DFIG

Figure 3.9 and 3.11 shows the responses of the conventional generator and the DFIG
providing primary regulation services. It can be seen that when the load increases at t=0, the
DFIG instantly releases its kinetic energy by reducing the mechanical speed instantly therefore
can increases its output to participate in primary frequency regulation. Thereafter DFIG output
decreases since the speed is no longer at the optimal and power extracted from the wind is
reduced. As the DFIG speed controller acts and the optimal speed is recovered, the DFIG power
output returns to its nominal.
It is also observed that the DFIG does not contribute to the frequency support after
steady-state is reached, and only provides frequency regulation during the transient phase. The
conventional generator, on the other hand, regulates its output and attains a new steady-state
operation level while providing primary regulation service.

36

Figure 3.9 Generator responses in primary frequency regulation with and without

Figure 3.7 Generator response in primary frequency regulation with and without DFIG

37

Figure 3.8 Generator response in primary frequency regulation with and without DFIG

Figure-3.12 and 3.13 shows the relation of the DFIG-based wind turbine incremental
power output and mechanical speed change, during the participation of the DFIG in primary
frequency regulation services. It can be seen that the additional power output of DFIG after load
perturbation is obtained by reduction of mechanical speed.

Figure 3.9 DFIG generator response in primary frequency regulation

38

Figure 3.10 DFIG generator response in primary frequency regulation

3.6.2 Automatic Generation Control
Figure 3.14 shows the MATLAB Simulink
model for examining the secondary frequency

control or AGC effects of the DFIG-based wind turbine generator. In this model wind turbine
mechanical speed limit during week wind or very strong wind is defined. DFIG maximum power
output constraint is implemented, and finally both strong and weak wind mechanical source has
been considered. Wind turbine speeds has to be within the threshold of cut out speed is been
implemented.
The energy extraction from DFIG machine is permitted only when the ultimately the
mechanical speeds lay within cut out speeds, and speed controller will not permit the speed to be
lower than minimum speed.

39

Figure 3.11 DFIG-Based WT with AGC Simulation model in Simulink

Simulations are carried out considering the dynamic model presented in Figure- 3.5 for a
0.02 per unit load perturbation with and without DFIG wind turbine to examine its contribution in
AGC service, The simulations are carried out for two levels of wind penetration, 20% and 50%
and the corresponding optimally tuned parameters for the DFIG controller, obtained in Section-
3.5, are used. As mentioned earlier, it has been assumed that DFIG-based wind turbines are in
their optimal mechanical speed with the output corresponding to the maximum power obtainable
form the wind, and wind speed remain constant during the simulation and the conventional
generator is able to supply the additional load after settlement of the transient.
Different plots for theses simulations are presented in Figure 3.15 to 3.18. Figure 3.15
shows the frequency responses with AGC for 5% wind power penetration with and without
DFIG. It is observed that the frequency responses for the case with and without DFIG very close,

40
and it is improved, by way of lower frequency peak excursion when DFIG participation is
considered. It is also observed that by using the DFIG tuned speed controller parameters the
settling time for cases 5%, 10%, 20% and 50% wind penetration are improved. It can be generally
expected that with the higher wind penetration the settling time will be longer.
Figure 3.15 to 3.18 shows low penetration of wind energy will not have negative impact
on frequency transient settling time between 5% to 20 %, while with further wind power
penetration 50% wind power frequency transient settling time will deteriorates, therefore it
generally can be concluded that the performance of AGC following a small load perturbation can
be improved with participation of DFIG. Peak frequency excursion will be reduced in direct
relation to the level of DFIG penetration while frequency settling time compare to no DFIG-based
wind turbine is not improved.

Figure 3.12 AGC with 2% load increment with and without DFIG

41



42


Figure 3.19 and 3.20 shows the responses of the DFIG supporting AGC services. It can
be seen that when the load increases at t=0, the DFIG instantly releases its kinetic energy by
reducing the mechanical speed instantly therefore can increases its output to participate and
support AGC. Thereafter as has been indicated in the earlier section DFIG output decreases since
the speed is no longer at the optimal and power extracted from the wind is reduced. As the DFIG
speed controller acts and the optimal speed is recovered, the DFIG power output returns to its
nominal.

43

Figure 3.19 DFIG generation with AGC and 2% load increment

Figure 3.16 DFIG generation with AGC and 2% load increment

44
3.7 Conclusion
Traditionally, wind turbine is not participating in the frequency regulation. With further growth of
wind energy, operators are demanding more and more participation of wind turbine in the
ancillary services especially frequency regulation. Doubly fed induction generators with the
ability of delivering power with different mechanical speed in one side, and the ability to
instantly reduce the speed, release the stored mechanical energy is able to support the
conventional generator in regulating frequency. This will reduce the frequency dip even more
efficiently whenever DFIG, PI controller is tuned. The extraction of energy from DFIG will be
coordinated with AGC of conventional generators which as result, damping of overshoot and
frequency excursion can be improved significantly. With further penetration of wind energy in
the electrical network with a target of growing to 20% of installed based power in North America
and most of the European countries, PI controller and AGC characteristics should be tuned, in
order to optimize the use of DFIG machine and their performance in participation of frequency
regulation.

45
Chapter 4
Primary and Secondary Frequency Control in Two Area Systems with
DFIG- Based Wind Turbine Support

4.1 Introduction
Multiple-area frequency control issues for an interconnected power system are far more
significant and complicated than those associated with isolated networks. In isolated networks
(single area) it is assumed that the frequency is uniform across the network resulting in single-
area control. Certainly this is not the case in large interconnected system where numerous control
areas can be identified. These control areas are traditionally associated with vertically integrated
utilities in charge of well defined geographical areas, and with their own secondary frequency
controls.

1
1
R
1
1
1
h
sT +
1
1
1
p
p
sT
K
+

1 g
P A 1 h
P A
1
f A
+

+
1 ref
P A
1
PD A
Electrical
System Area 1
1
1
1
t
sT +
1 f
P A
s
K
AGC1
1
B
s
T
o
t 2
2
1
1
h
sT +
2
2
1
p
p
sT
K
+

2 g
P A
2
f A
+

+
2
ref
P A
Conventional
Generation Area 2
Electrical
System Area 2
2
1
1
t
sT +
2 f
P A
s
K
AGC2
-1
-1
2
1
R
2
B
Conventional
Generation Area 1
2
PD A
1
ACE
2
ACE
12
P A
12
P A
2 h
P A
+
+
+
+
+
1
f A
2
f A
21
P A

Figure 4.1 linear model of frequency control two Area Systems

46
The dynamic performance of a two-area interconnected system can be analyzed
using a small-perturbation transfer-function model (Figure 4.1). This model includes both
the primary and secondary frequency regulation controls after a disturbance. The system
performance largely depends on the integral control gains (K
AGC1
and K
AGC2
) and the
frequency bias factors (B1 and B2), in addition to system parameters and operating
conditions.
Under normal operating conditions each area strives to carry its own load, and a
portion of the neighboring areas load through the tie-line if agreed upon and scheduled
in advance. Similar to a single-area each system frequency control area is characterized
by the same frequency, therefore incremental frequency changes are represented by
1
f A
and
2
f A respectively.
During normal operation, tie-line power flow obtained (4-1):
) ( Sin
X
V V
P
o o
o o
o
2 1
2 1
12
o o = (4-1)
Where
1
and
2
are the angles of the tie end voltages V
1
and V
2
respectively, and P
12
is
the power in the tie line. For small value of
1
and
2
the tie line power incremental can be written
as follows:

) ( T P
2 1 12 12
o A o A ~ A (4-2)
) ( Cos
X
V V
T
o o
o o
2 1
2 1
12
o o =
A
(4-3)

Analogous to electric stiffness concept of synchronizing machines, (4-3) is considered as the
synchronizing torque coefficient.
For two area-control system, the Area Control Error (ACE) for each area consists of
linear combination of frequency and tie line error, as given below:

1 1 12 1
f B P ACE A A + = (4-4)
2 2 21 2
f B P ACE A A + = (4-5)

47
4.2 Dynamic Model of Primary Frequency Regulation with DFIG- Based Wind
Turbines
The small perturbation model in the form of transfer-function block diagram for a two-area
frequency control system comprising a simple conventional prime-mover model, with a non-
reheat type turbine and DFIG-based wind turbine in each area is shown in Figure 4.2.
1
1
R
1
1
1
h
sT +
1
1
1
p
p
sT
K
+

1 g
P A
1 h
P A 1
f A
+
1 ref
P A
1
PD A
Electrical
System Area 1
1
1
1
t
sT +
1 f
P A
s
T 2
o
t
2
1
1
h
sT +
2
2
1
p
p
sT
K
+

2 g
P A 2
f A
+

+
2
ref
P A
Conventional
Generation Area 2
Electrical
System Area 2
2
1
1
t
sT +
2 f
P A
-1
2
1
R
Conventional
Generation Area 1
2
PD A
12
P A
2 h
P A
1
f A
21
P A
1
1
1
r
sT +
1
1
1
w
w
sT
T
+
1
R
1

1
wp
K
s
K
wi
1
1
1
1
a
sT +
1
2
1
e
sH
+
ref , NC
P
1
A
*
1
e A
1
e A
1 NC
P A
1 NC
P A
1 1
X

A
1 2
X

A
1 3
X

A
2
1
1
r
sT +
2
2
1
w
w
sT
T
+
2
R
1

2 wp
K
s
K
wi 2
s T 1
1
2 a
+
s H
e2
2
1
2 NC
P A
ref , NC
P
2
A
*
w
2
A
2
e A
2
f A
+
+
+

+
+
+
2 2
X

A
2 1
X

A
2 3
X

A
Tie line model
DFIG model Area 1
DFIG model Area 2
1 NC
P A
min 1
e
0
Max , NC
P
1
max , NC
P
2
0
min 2
e
max 2
e
max 1
e
2 NC
P A
2 NC
P A

Figure 4.2 Linear dynamic model of primary frequency control of two area systems with DIFG

48

The dynamic model in state-space form can be obtained from the transfer-function model
as given below:
p X A
dt
X d
I + =
(4.6)
X is the state-vector and p is the perturbation vector. The matrices A and I are state and
perturbation matrixes respectively. The state-equation (4.6) can be written in expanded vector-
matrix form as given below:
T
NC g h
NC g h
P X X X f P P
P X X X P f P P
X
(
(
=

2 2 2 3 2 2 2 1 2 2 2
1 1 1 3 1 2 1 1 12 1 1 1
e A A A A A A A
e A A A A A A A A A

(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
t t

=
2 2
2
2
2 2 2 2
2
2 2 2
2 2
2
2
2 2
2
2
2
2 2
2 2 2
1 1
1
1
1 1 1 1
1
1 1
1 1
1
1
1
1
1
1
1 1
1
1 1
1 1
2
1
2
1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0
1 1 1
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0
1 1
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0
1
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0
1 1
0 0 0 0 0 0 0 0 0
0 0 0 0 0
1
0
1
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
2
1
2
1
0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
1 1 1
0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0
1
1 1 1
0 0
0 0 0 0 0 0 0 0 0 0 0 0 0
1 1
0 0
0 0 0 0 0 2 0 0 0 0 0 0 0 2 0 0
0 0 0 0 0 0 0 0 0 0 0 0
1
0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0
1
1
0
1
e e
a
wp
a a a
wi
w r r
r r
p
p
p p
p
p
p
t t
h h
a
wp
a a
wi
r r
r r
o
P
p o
P
p
P
p
P P
p
t t
h h
H * H *
T
K
T T R * T
K
T T T
T T
T
K
T T
K
T
K
T T
R * T T
He * He *
T
K
T T R * Ta
K
Tw T T
T T
T
T
K
T
T
K
T
K
T T
K
T T
R * T T
A

T
p
p
p
p
T
K
T
K
(
(
= I 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2
2
1
1

| |
T
2 1
PD PD 0 p 0 0 0 0 0 0 0 0 0 0 0 0 0 0 =

49

The system behavior depends on the network parameters, tie line, conventional system
parameters such as damping factor of the load (D
1
and D
2
) , the regulation droop (R
1
and R
2
), the
inertia (H
1
and H
2
), governor and turbine time constants T
h
& T
t
respectively. The constants K
P

and T
P
are evaluated as K
P
= 1/D and T
P
= 2H/fD. DFIG based turbine speed controllers are K
wp

and K
wi
.

4.3 Dynamic model of Secondary Frequency Regulation with DFIG based
Wind Turbine - Two Area Controls

The small-perturbation transfer-function block diagram of the two area power system for
secondary frequency regulation is Figure- 4.3.
The dynamic model in state-space form is developed from the transfer-function
representation as follows:

p X A
dt
X d
I + =
(4.6)
Where,

T
NC g h ref
NC g h
P X X X f P P P
P X X X P f P P
X
(
e A A A A A A A A
e A A A A A A A A A
=

2 2 2 3 2 2 2 1 2 2 2 2
1 1 1 3 1 2 1 1 12 1 1 1
T
p
p
p
p
T
K
T
K
(
(
= 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2
2
1
1
I
| |
T
2 1
PD PD 0 p 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 =

Similar to previous section, X is the state-vector and p is the perturbation vector. The
matrices A and I are state and perturbation matrixes respectively. The state-equation (4.11) can
be written in expanded vector-matrix form as follows:

50
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(

=

1 1
2
2
2 2 2 2
2
2 2 2
2 2
2
2
2 2
2
2
2
2 2
2 2 2 2
2 2 2
1 1
1
1
1 1 1 1
1
1 1 1
1 1
1
1
1
1
1
1
1 1
1
1 1
1 1 1 1
1 1 1
2
1
2
1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0
1 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0
1 1
0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0
1
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0
1 1
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0
1
0
1 1
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
2
1
2
1
0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
1 1 1
0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1
0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1
0 0 0
0 0 0 0 0 2 0 0 0 0 0 0 0 0 2 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0
1
0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1
0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1
0
1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
e e
a
wp
a a a
wi
w r r
r r
p
p
p p
p
p
p
t t
h h h
AGC AGC
e e
a
wp
a a a
wi
w r r
r r
p
p
p
p
p
p
p p
p
h t
h h h
AGC AGC
H H
T
K
T T R T
K
T T T
T T
T
K
T T
K
T
K
T T
R T T T
B K K
H H
T
K
T T R T
K
T T T
T T
T
T
K
T
T
K
T
K
T T
K
T T
R T T T
K B K
A
t t

51
1
1
R
1
1
1
h
sT +
1
1
1
p
p
sT
K
+

1 g
P A
1 h
P A 1
f A
+
1 ref
P A
1
PD A
Electrical
System Area 1
1
1
1
t
sT +
1 f
P A
s
T 2
o
t
2
1
1
h
sT +
2
2
1
p
p
sT
K
+

2 g
P A 2
f A
+

+
2 ref
P A
Conventional
Generation Area 2
Electrical
System Area 2
2
1
1
t
sT +
2 f
P A
-1
2
1
R
Conventional
Generation Area 1
2
PD A
12
P A
2 h
P A
1
f A
21
P A
1
1
1
r
sT +
1
1
1
w
w
sT
T
+
1
R
1

1
wp
K
s
K
wi
1
1
1
1
a
sT +
1
2
1
e
sH
+
ref , NC
P
1
A
*
1
e A
1
e A
1 NC
P A
1 NC
P A
1 1
X

A
1 2
X

A
1 3
X

A
2
1
1
r
sT +
2
2
1
w
w
sT
T
+
2
R
1

2 wp
K
s
K
wi 2
s T 1
1
2 a
+
s H 2
1
2 e
2 NC
P A
2 ref , NC
P A
*
w
2
A
2
e A
2
f A
+
+
+

+
+
+
2 2
X

A
2 1
X

A
2 3
X

A
Tie line model
DFIG model Area 1
DFIG model Area 2
1 NC
P A
min 1
e
0
Max , NC
P
1
max , NC
P
2
0
min 2
e
max 2
e
max 1
e
2 NC
P A
2 NC
P A
s
K
AGC1
1
B
s
K
AGC2
-1
2
B
1 ACE
2 ACE
12
P A
+
+
+
+

Figure 4.3 Linear model of secondary frequency control of two area systems with DIFG-based
wind turbines

52
4.4 Optimal Tuning of DFIG Based Wind Turbine Controller Parameters
The objective of this section is to determine the optimal set of parameters of the
Proportional-Integral controller of the DFIGs in the two-area power system. The ISE
technique discussed in Chapter-3 is used for obtaining the optimum settings for K
wp
and K
wi

parameters of DFIG in each area. A quadratic performance index denoting the squared of the
deviations of the state variables, is considered as follows:
| |dt ) t ( X ...... ) t ( X ) t ( X ) t ( X J
n
}
+ + + + =
0
2 2
3
2
2
2
1
(4.12)
In this chapter, the square of the ACE is considered as the performance index for tuning
the controller parameters in each area, as stated below.
dt ) ACE ( J
}
=
0
2
1 1
(4.13)
dt ) ACE ( J
}
=
0
2
2 2
(4.14)
For computational purposes, the performance index J is calculated by summing the
discrete values at very small intervals, over a considerable period until the steady-state is
achieved, as given below.
| |
=
+ =
K
k
) t k t ( ACE J
1
0
2
1 1
A A
(4.14)
| |
=
+ =
K
k
) t k t ( ACE J
1
0
2
2 2
A A
(4.15)
The optimal settings of the DFIG controller parameters in the two areas are obtained by
using a heuristic iterative scheme that seeks to minimize the integral squared error of the ACE in
each area. The schematic step-wise flow diagram of the optimal tuning process is shown in
Fig.4.4 and the optimal values of the DFIG speed controller parameters K
wp1,
K
wi1
and K
wp2,
K
wi2

are determined. As per the flow diagram, first we arbitrarily chose the DFIG controller
parameters of both areas, and then sequentially determine the optimal settings for Area-1 and
Area-2 controllers using ISE technique. A check for convergence of the optimal parameters is
carried out to ensure the program arrives at the optimal settings for the two DFIG controllers.

53
Initialize Area-1 and Area-2 DFIG controller
parameters
Initialize
Using ISE Technique, obtain optimal DFIG
controller for Area-1. Revise initial controller
settings of Area-1.
Using ISE Technique, obtain optimal DFIG
controller for Area-2. Revise initial controller
settings of Area-2.
Check for convergence: Are the revised
controller settings of each area within tolerance
limits of initial settings?
Stop
No
Yes
Convergence
attained ?

Figure 4.4.Schematic diagram for optimal tuning of DFIG controller parameters

54
Figure 4.5 shows a typical working result of the optimization procedure for a 50%
wind penetration case, how the DFIG controller parameters converge to their optimal
settings, while the objective of minimization of ACE is achieved. Tables 4-1 and 4-2
present the results of the optimal parameter tuning for 20% to 50% penetration of wind
generation in area-1 and area-2 respectively.

Figure 4.5.Tuning DFIG controller parameters for 50% wind penetration

Table 4.1 Optimal DFIG controller settings parameters area-1
Penetration K
wi1
K
wp1
J
1

20% 0.2 1.2 0.0507
50% 0.2 1.7 0.0797

55
Table 4.2 Optimal DFIG controller settings parameters area-2
Penetration K
wi2
K
wp2
J
2

20% 0.1 1.0 0.0499
50% 0.1 1.0 0.0765

4.5 Analytical Studies
4.5.1 Primary Frequency Control
Simulations were carried out based on the dynamic model shown in Figure- 4.2 for 0.02 per unit
step load perturbation in both areas, with and without DFIG wind generation to examine the
impact of wind generator participation in primary frequency regulation. The corresponding
MATLAB SIMULINK model is given in Figure 4.6.
During the simulations in both area, it has been assumed that DFIG-based wind turbines
are in their optimal mechanical speed with the maximum power obtainable form the wind, and
wind speed remain constant during the simulation. Different plots for theses simulations are
presented in Figure 4.7 to 4.9. The parameters used for this simulation has been shown in
appendix B.
Figure 4.7 presents power increment change for tie-line power for 20%, 50% wind power
penetration and also without wind power generation, It is observed that tile-line generation
increment response following the disturbance is improved with participation of DFIG, by way of
lower tie-line power increment. It is also observed that when the wind penetration level increases
from 20% to 50% the response is deteriorates by way of increased settling time. It can be
generally concluded that with 20% DFIG-based wind power penetration, primary frequency
regulation and the settling time is improved, compare to no-WT case, which the steady state
errors are very close.

56

Figure 4.6.Simulink model for frequency response studies for two-area controller with
DFIG-based wind Turbine

57

Figure 4.7.Primary frequency regulation tie-line incremental power

Figure-4.8 to 4.15 presents the frequency plots following a 0.02 per unit step load
perturbation in both areas for 5%, 10%, 20% and 50% penetration of DFIG based WT in
comparison with non-DFIG based WT which is not participating in frequency regulation. It is
observed that the frequency response following the disturbance is improved, by way of lower
frequency peak excursion when DFIG participation is considered. It is also observed that when
the wind penetration level increases from 5% to 50%, although the lower frequency peak
excursion is improved by higher penetration but the frequency response deteriorates by way of
increased settling time and higher steady-state frequency error. It can be generally concluded that
with lower penetration, the settling time is improved, compare to no-DFIG case, which the steady
state errors are close for both of the areas.

58

Figure 4.8. Primary frequency regulation for 2% load change area 1

Figure 4.9.Primary frequency regulation for 2% load change area 2

59



60

Figure 4.12. Primary frequency regulation for 2% load change tie line


61

Figure 4.14 Primary frequency regulation for 2% load change area 2

Figure 4.15 Primary frequency regulation for 2% load change tie line

62
4.4.2 Secondary Frequency Control
Simulations are carried out considering the dynamic model shown in Figure 4.3 for a 0.02 per
unit step load perturbation in both areas considering the optimally tuned DFIG controllers, to
examine their impact on secondary frequency control.
The simulations are carried out for four levels of wind penetration, 5%, 10%, 20% and
50% and the corresponding optimally tuned parameters for the DFIG controller, obtained in
Section-4.3, are used. As mentioned earlier, it has been assumed that DFIG-based wind turbines
are in their optimal mechanical speed with the output corresponding to the maximum power
obtainable form the wind, and wind speed remain constant during the simulation and the
conventional generator is able to supply the additional load after settlement of the transient.
Different plots for theses simulations are presented in Figure 4.16 to 4.26. Figure4.16 to
4.19 shows the tie-line incremental power change for four cases of 5%, 10%, 20% and 50% of
wind power penetration with and without DFIG based WT. It is observed that response of AGC
with support of DFIG compare to no DFIG based WT to minimize the tie-line power incremental
change is improved, by way of shorter transient time settlement and lower transient power
excursion peak.

Figure 4.166 Secondary frequency regulation for 2% sudden load tie line power

63

Figure 4.17 Secondary frequency regulation for 2% sudden load tie line power

Figure 4.18 Secondary frequency regulation for 2% sudden load change tie line power

64

Figure 4.19 Secondary frequency regulation for 2% sudden load change tie line power

Figure-4.20 to 4.26 shows frequency response of area-1 and area-2 following the same
step load perturbation. It is observed that frequency peak excursion has been improved in all
cases with DFIG based WT for 5%, 10%, 20% and 50% wind power penetration while with
higher penetration frequency transient settling time is increased compare to no DFIG WT power
case.

Figure 4.20 Secondary frequency regulation for 2% sudden load change area-1

65

Figure 4.21 Secondary frequency regulations for 2% sudden load change area-1


66



67



It can be generally concluded with higher penetration of wind power energy the performance of
AGC, initially compare to lower penetration and ultimately compare to no DFIG wind power

68
case, due to the fact that after release of kinetic energy by DFIG, its mechanical speed is reduced,
and therefore as described earlier in the chapter 3, the power output of DFIG is reduced, compare
to steady condition. Therefore for larger penetration of DFIG, this reduction of the power output
will be additional burden in AGC and conventional generation to control the load- frequency
balance.
Figures 4.27 to 4.29 presents the ACE
1
and

ACE
2
plots following a step load
perturbation. It is observed that ACE is deteriorated with high wind energy penetration.
Deviation of DFIG from optimal mechanical speed in case of high wind energy
penetration has a more negative impact on ACE as it increases the total settling time. In
the case of low wind energy penetration, a 5% contribution of DFIG in reduction of ACE
after applying 2% step load perturbation is not significant.

Figure 4.23 Area error for 2% sudden load change in area-1

69



70
4.6 Conclusion
Participation of DFIG based wind generators in frequency regulation and automatic generation
control (AGC) services in multi-area control systems is a complex problem. The DFIG contrary
to fixed speed WT could participate and improve the frequency performance and support
conventional generation in frequency regulation services by reduction in ACE in each area.
This chapter presents a small-perturbation transfer function model for a two-area power
system, each area equipped with DFIG based wind generators. As in the previous chapter,
integral squared error based optimal tuning of the DFIG controllers is carried out. However, the
optimal tuning process becomes more complex because of the presence of more parameters to be
tuned simultaneously. Therefore, a heuristic iterative optimal tuning scheme is proposed, that
helps determine the optimal parameters successfully.
Detailed simulation studies have been presented to compare the performance of the
optimally tuned DFIG controllers and their contribution to frequency regulation and AGC in the
two-area power system. Also, studies have been presented to compare the system impact of a 5%,
10%, 20% and 50% penetration of wind generation.

71
Chapter 5
Conclusion and Future Work
5.1 Conclusion
This thesis attempts to address some of the important issues associated with the integration of
wind generation with conventional generation sources in the total energy supply mix of the power
system. The thesis is particularly focused on examining the frequency regulation and Automatic
Generation Control (AGC) effects of DFIG based wind generators. The main contributions and
findings from this research are as follows:
o Small perturbation, transfer function models in state-space form are developed for the
first time, for both single-area and two-area power systems with a mix of conventional
generation and wind generation sources. The state-space models are developed to
examine both primary regulation and AGC effects in the two classes of models. DFIG
based wind generation is considered for the model development of the wind generation
sources.
o A simple parameter optimization technique based on the ISE criterion is applied to
optimally tune the DFIG controller parameters. In the case of the two-area power system,
with DFIGs in both areas, the number of parameters to be optimally tuned is larger, and
hence a sequential application of ISE criterion is presented.
o Dynamic performance analysis is carried out considering the optimally tuned DFIG
controllers to examine their contribution to system frequency support services, both
primary regulation and AGC.
o A wind generation penetration index is defined and the effect of variation in penetration
is studied for the cases of 5%, 10%, 20% and 50 % wind energy penetration in line with
global world wind energy penetration targets of 2020 and 2050, respectively

5.2 Scope for Future Work
With respect to future research directions, the following problems can be examined, based on the
research presented in this thesis:
a. The state-space model of the system developed, considering the DFIG based wind
generator can be further extended to include pitch control effects of the wind generator.
Considering the result of simulation presented in chapter 2 for wind speed changes and
overall response of pitch control system to this disturbance, in one side and secondary
frequency regulation settling time presented in chapter 3 and 4 are about the same and

72
close, therefore further study could be conducted to examine use of DFIG based wind
turbine with de-loaded operation (operation with less than maximum power) to full
participation of DGFIG in frequency regulation in transient and with increasing the
generation by pitch control.

b. This work considers the traditional proportional-integral controllers for the DFIG
machines. There is scope for research into examining other controller structures, in
particular design of adaptive DFIG controllers for frequency control contributions, which
can provide improved dynamics performances because of the intermittent nature of wind
generation.

c. There is scope for examining how wind generators can participate in frequency regulation
and AGC services based on market mechanisms and developing appropriate incentives
for DFIGs to participate in frequency regulation support.

73
Bibliography

[1] Wind Vision for Canada, Recommendations for Achieving Canadas Wind Energy
Potential, Report by the Canadian Wind Energy Association (CanWEA).
[2] Quick Facts about Wind Energy, Canadian Wind Energy Association (CanWEA),
http://www.canwea.ca.2001, http://www.canwea.ca/pdfs/CanWEA-WindVision.pdf.
[3] European of Wind Energy in the European Power Supply: Analysis, Issues and
Recommendations, 2005. Available online: http://www.ewea.org/.
[4] Canadian Wind Energy Association (CanWEA) Wind farm over view
http://www.canwea.ca/featuredWindFarm_e.php?farmId=104
[5] "Nova Scotia Power is generating cleaner, greener energy". From Nova Scotia Power
web site. http://www.nspower.ca/en/ Retrieved 2007-12-16.
[6] Qazette Montreal May 5 2008, Article Quebec picks 15 wind-power projects
[7] Project reference from CanWEA, http://www.canwea.ca/farms/index_e.php
[8] Project reference from Can WEA http://www.canwea.ca/pdf/Proposed%20projects.pdf
[9] Installed wind capacity in Ontario Reference from IESO,
http://www.ieso.ca/imoweb/marketdata/windpower.asp
[10] American Wind Energy Association (2009). Annual Wind Industry Report, Year Ending
2008 pp. 9-10.
[11] Reference from American Wind Energy Association 2009 report
http://www.awea.org/publications/reports/4Q09.pdf
[12] Peter Behr. Predicting Wind Power's Growth -- an Art That Needs More Science New
York Times, April 28, 2010.
[13] US Wind Resource Even Larger Than Previously Estimated: Government Assessment
AWEA February 18, 2010
[14] American Wind Energy Association FOR EMBARGOED RELEASE:
October 20, 2009, 3:00 PM ET
[15] American Wind Energy Association (2009). Annual Wind Industry Report, Year Ending
2008 p. 17.
[16] Office of Energy Efficiency and Renewable Energy (2010-02-04). "Installed Wind
Capacity by State". United States Department of Energy.
http://www.windpoweringamerica.gov/docs/installed_wind_capacity_by_state.xls

74
[17] Office of Energy Efficiency and Renewable Energy (2010-03-05). "US Installed Wind
Capacity and Wind Project Locations". United States Department of Energy.
http://www.windpoweringamerica.gov/wind_installed_capacity.asp
[18] Investment US b 9 of the Worlds Most Amazing Wind Farms
http://www.investmentu.com/latest-research/testimonials.php?code=X300L118 Wind
2010 report
[19] Count on Wind Energy by EWEA 2008 report
http://www.ewea.org/fileadmin/ewea_documents/documents/press/campaigns/Count_on_
Wind_Energy_Briefing.pdf
[20] Wining with European Wind by EWEA public report 2008
http://www.ewea.org/fileadmin/ewea_documents/documents/publications/
Annual_Report_2008.pdf
[21] Europe's onshore and offshore wind energy potential An assessment of environmental
and economic constraints by EEA No6/2009
http://www.eea.europa.eu/publications/europes-onshore-and-offshore-wind-energy-
potential
[22] Ocean of Opportunity by European Wind Energy Association
http://www.ewea.org/fileadmin/ewea_documents/documents/publications/reports/Offshor
e_Repo rt_2009.pdf
[23] Reference from EWEA wind is power, European Market for Wind Turbines
Grows 23% in 2006
[24] Bloombeg Business week, Scott is a reporter in BusinessWeek's London bureau
http:/www.busineesweek.com/globalbiz/content/aug2007/gb2007083_852915_page_2.ht
m
[25] Science and Technology Solar. by Jerry James Stone, San Francisco, CA
on 02.15.10 www.treehugger.com
[26] Illustrated history of wind power generation by Darrell M. Dodge, Littleton, Colorado
[27] Reference from Visual Dictionary Online
http://www.visualdictionaryonline.com/energy/wind-energy/wind-turbines
[28] Reference from encyclopedia of alternative energy www.daviddarling.info
[29] Community Wind Power fact sheet # 4 by Renewable Energy Research Laboratory,
University of Massachusetts at Amherst.
[30] Introduction to the modeling of wind turbine, Wind Power system 2005 John Wiley and
Sons Ltd ISBN 0-470-85508-8 (HB)

75
[31] Wind energy Converter System by Siegfried Heier Kassel University of Germany
Published by John Wiley and Sons Ltd
[32] Janaka Ekanayake and Nick Jenkins, Comparison of the Response of Doubly Fed and
Fixed-Speed Induction Generator Wind Turbines to Changes in Network Frequency
IEEE Transaction on Energy Conversion, Vol. 19, NO. 4, December 2004
[33] R. Pena J.C.Clare, Doubly fed induction generator using back-to-back PWM converters
and its application to variable speed wind-energy generation, Power Appl., Vol. 143, No
3, May 1996
[34] J. G. Slootweg, S. W. H. de Haan, H. Polinder, and W. L. Kling, General Model for
Representing Variable Speed Wind Turbines in Power System Dynamics Simulations,
IEEE Transaction on Power Systems, Vol. 18, NO. 1, February 2003
[35] Gillian Lalor, Julia Ritchie, Shane Rourke, Damian Flynn and Mark J. OMalley
Dynamic Frequency Control with Increasing Wind Generation
[36] Rogrio G. de Almeida and J. A. Peas Lopes, IEEE Transaction on Power Systems,
Participation of Doubly Fed Induction Wind Generators in System Frequency Regulation
Vol. 22, NO. 3, August 2007
[37] Alan Mullane, and Mark OMalley, The Inertial Response of Induction-Machine-Based
Wind Turbines, IEEE Transaction on Power Systems, VOL. 20, NO. 3, August 2005
[38] Johan Morren, Sjoerd W. H. de Haan, Wil L. Kling, and J. A. Ferreira, Wind Turbines
Emulating Inertia and Supporting Primary Frequency Control , IEEE Transaction on
Power Systems, Vol. 21, NO. 1, February 2006
[39] Juan Manuel Mauricio, Alejandro Marano, Antonio Gmez-Expsito, and Jos Luis
Martnez Ramos, Frequency Regulation Contribution Through Variable-Speed Wind
Energy Conversion Systems, IEEE Transaction on Power Systems,, VOL. 24, NO. 1,
February 2009
[40] Andreoiu and K. Bhattacharya, Robust tuning of power system stabilizers using
Layapunove method based generic algorithm, IEEE Proc.Gener Transaction Vol 149,No
5 Sep 2002

76
Appendix A
System Parameters Single Area

H
e
=3 PU. MW.sec
K
agc=
0.1

K
p
=62

HZ/PU

K
wi
=0.1

K
wp
=1.58

R=3 Hz/PU.MW

T
a
=0.2

Sec
T
h
=0.1 Sec

T
p
=10

sec

T
r
=0.1

Sec

T
t
=1 Sec

T
w
=6 Sec

77
Appendix B
System Parameters Two Area

H
e1
=3.5 PU. MW.sec
H
e2
=3.5 PU. MW.sec
K
agc1
=0.05
K
agc2
=0.05
K
p1
=62

HZ/PU

K
wi1
=0.1

K
wp1
=1.58

K
p2
=62

HZ/PU

K
wi2
=0.1

K
wp2
=1.61

R
1
=3 Hz/PU.MW
R
2
=3 Hz/PU.MW
T=0.07 PU.MW/Hz
T
a1
=0.2

Sec

T
a2
=0.2

Sec
T
h1
=0.1 Sec
T
h2
=0.1 Sec

T
p1
=10

sec
T
p2
=15

sec
T
r1
=0.1

Sec

T
r2
=0.1

Sec

T
t1
=1 Sec

T
t2
=1 Sec

T
w1
=6 Sec

T
w2
=6 Sec

Frequency Control For DFIG

Uploaded by

Copyright:

Available Formats

Frequency Control For DFIG

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Frequency Control For DFIG

Uploaded by

Copyright:

Available Formats

DFIG Based Wind Turbine Contribution

to System Frequency Control

model for examining the secondary frequency

You might also like