A NELI

“Knowledge to make your life easier”

Knowledge to make your life easier

Electric Machines
Basic Theory ©

Jorge Cardenas
First Edition

Follow us in

https://www.linkedin.com/company/adneli-consultant/?viewAsMember=true

www.adneli.com ADB-001
 Electric Machines – Basic Theory

Editorial: Adneli Consultant, S.L.

Copyright© May 2021 by Adneli Consultant, S.L.

www.adneli.com
[email protected]

You can find complementary information here:

Visit our website: www.adneli.com/store

Copyright © 2021
“All rights to this book are retained by Adneli Consultant, S.L. This book is allowed to reproduce
individual figures (except the ones belonging to others and mentioned in the specific references) in
other publications, and their use for any third party provided proper acknowledgment is given to the
author.”
License: CC-BY-SA 4.0 International - Creative Commons, Attribution Share-alike

2
 Electric Machines – Basic Theory

About the Author

I received my Engineering Degree from the Universidad de Ingeniería (Peru) in 1977 and my MBA
from the Universidad Politécnica de Madrid (Spain) in 1998.

I began my career with the Utility Electroperu (Peru) in 1978, as a Protection & Control engineer. In
1987 I moved to ABB (Spain) as HV equipment Sales Engineer and was then promoted to a Control
Design Engineer. In 1989 I joined General Electric (GE), where I have held several positions until 2020.
Currently, I work as Advisor for Adneli Consultant in Spain. I have authored and co-authored more
than 70 articles (English, Spanish and Russian) on protective relaying and topics related to Grid
Modernization (papers published in some of the most important congresses around the world: Europe,
Asia, USA): See some of them in https://www.researchgate.net/profile/Jorge_Cardenas11. I am a
member of the CIGRÉ WG B5.31 and WG B5.43 and a contributor of the magazines GE P&C Journal,
Pacworld (USA), IET (UK), and Energía (Spain). I am a regular speaker in congress and conferences
in Europe. I have made several contributions in the design of new products related to Generator, Bus,
Line, Transformer, Motor, feeders, Network Protection. I have two co-patents: the first one on a new
Power System inter-area oscillation detector and the second one on an Early warning system for
Cybersecurity alert and automatic protective actions. I have also written a book on Philosophy See:
www.adneli.com/store, a book on “Relay Protection, Control, and Information Management in the
Modern Power Systems,”, a book “Line Protection in High Voltage AC Transmission Systems,” and I
am glad to present this book entitled “Electric Machines – Basic Theory.”

I have worked with some of the major Utilities in the world as PG&E (USA), Scottish Power (UK),
Enel Terna (Italy), Iberdrola, Gas Natural, Red Electrica & Endesa (Spain), CFE (Mexico), МЭС
Сибири (Rusia), CTEEP (Brazil), IEC (Israel), Fingrid (Finland), TenneT (Netherlands), RTE (France),
EDP & REN (Portugal), ESKOM (South Africa), WAPDA (Pakistan), TEIAS (Turkey), Sonelgas
(Algeria), ONE (Morocco). I have worked also with some of the major Industries in Oil &Gas as Rasgas
(Qatar) and SWCC (Saudi Arabia).

I have worked as well in projects with some of the main laboratories in the world as CESI (Italy), Kema
(Holland), CEPRI & NCEPRI (China) and with Universities as “La Sapienza” in Rome, Italy, and UPV
in the Basque Country, Spain.

Specialties: Negotiation, Project Management, Consulting in Electrical Engineering focused on Power

System Protection and Assets Management. In recent years I have led R&D in the synchrophasor
application for Utilities and now I work on projects related to Grid Modernization.

3
 Electric Machines – Basic Theory

Foreword
This book is a summary of the basic theory of Electrical Machines and aims to support other books such
as the ones entitled. “Relay Protection, Control, and Information Management in the Modern Power
Systems,” and “Line Protection in High Voltage AC Transmission Systems,” where despite the
existence of an extensive theoretical description, it does not include the complete basic theory (some
has been already developed and explained), because it is not the objective of an application book to be
used by engineers with a certain level of training and experience.

The technical aspects that the author considers to be the most important for aid in the better
understanding of other application books on electrical protections and other disciplines have been
included. Some of the topics developed in application books sometimes present difficulties in
understanding due to not having available some basic theoretical aspects that are sometimes forgotten.
The author hopes to achieve the desired goals and expects to receive any suggestions that allow to
improve the content. Be free to send them to [email protected]. Thanks.

Jorge Cardenas

4
 Electric Machines – Basic Theory

Contents
Chapter I-ELECTROMAGNETISM...................................................................................................... 7
1.1. Basic definitions ..................................................................................................................... 7
1.2. The second fundamental interaction – Electromagnetism ...................................................... 7
1.3. MFF Waveforms .................................................................................................................... 9
1.4. Main Applications in Electricity .......................................................................................... 10
1.4.1. Transformers .................................................................................................................... 10
1.4.2. Overhead Transmission lines ........................................................................................... 14
1.4.3. Rotating machines ............................................................................................................ 15
Chapter II-TRANSFORMATIONS ..................................................................................................... 18
2.1. Definition.............................................................................................................................. 18
2.2. Sequence components in a Three-Phase Power System ....................................................... 19
2.2.1. Symmetrical Components ................................................................................................ 19
2.2.2. Clarke and Park Components ........................................................................................... 23
2.2.3. Transformation for modal analysis ................................................................................... 30
2.2.4. Laplace Transformation.................................................................................................... 31
Chapter III-SYNCHRONOUS MACHINE PARAMETERS’ DERIVATION. PART 1 .................... 34
3.1. Synchronous Machine Derivation ........................................................................................ 34
3.2. Generator Model................................................................................................................... 35
3.3. Theory of Operation ............................................................................................................. 35
3.3.1. Representation of the d-q non-reduced order model .................................................... 37
Chapter IV-Synchronous Machine. Parameters Derivation. Part 2 ...................................................... 46
4.2. Effect of saliency in stability ................................................................................................ 51
Chapter V-SYNCHRONOUS MACHINE. PARAMETERS DERIVATION. PART 3 ..................... 53
5.1. The correspondence of Stator EFM to Rotor quantities ....................................................... 53
5.2. The Swing Equation ............................................................................................................. 55
5.3. Torque and Power Referred to the Rotor ............................................................................. 56
Chapter VI-SYNCHRONOUS MACHINE. PER UNIT VALUES ..................................................... 59
6.1. Per-unit Values ..................................................................................................................... 59
6.1.1. Application in transformers .............................................................................................. 59
6.2. Pu values of the d-q voltage equations in Synchronous machines: ...................................... 62
6.2.1. Per unit values for Stator in a synchronous generator ...................................................... 63
6.2.2. Per unit values for the rotor in synchronous machines ..................................................... 64
6.3. Base change .......................................................................................................................... 65
Chapter VII-ASYNCHRONOUS MACHINE DERIVATION. PART 1 ............................................ 67
7.1. Asynchronous Machine Derivation ...................................................................................... 67
Chapter VIII-ASYNCHRONOUS MACHINE DERIVATION. PART 2 ........................................... 72
8.1. Steady-State Analysis ........................................................................................................... 72
8.1.2. Application Example - Type II wind turbine variable slip generator ............................... 77

5
 Electric Machines – Basic Theory

Chapter II-CLASSICAL AUTOMATIC CONTROL .......................................................................... 78

9.1. Classical Control Theory ...................................................................................................... 78
9.1.1. Mathematical Models of Control Systems ....................................................................... 79
9.2. Software tools for control analysis ....................................................................................... 93
Chapter X-MODERN CONTROL THEORY ...................................................................................... 95
10.1. Modern Control Theory.................................................................................................... 95
10.1.1. The state equations ....................................................................................................... 96
10.1.2. Output Equations .......................................................................................................... 97
10.1.3. State Space Representation........................................................................................... 98
10.1.4. Stability and natural response characteristics of a continuous time LTI system ........ 100
Chapter XI-WIND ENERGY ............................................................................................................. 102
11.1. Wind Energy................................................................................................................... 102
11.2. Wind Generators. Overview ........................................................................................... 103
11.2.1. Type I (SCIG) Wind turbine Generator...................................................................... 103
11.2.2. Type II (WRIG) Adjustable Speed Generators .......................................................... 106
11.2.3. Doubly Fed Induction Generator (DFIG)-based Wind Turbines2_8____....................... 107
11.2.4. Full Converter Generator (IG)-based Wind Turbines50 .............................................. 114
Chapter XII-PHOTOVOLTAIC SYSTEMS AND BATTERIES...................................................... 119
12.1. Photovoltaic Systems- .................................................................................................... 119
12.1.1. Maximum power point tracking ................................................................................. 121
12.2. Battery ............................................................................................................................ 122
12.2.1. Modeling of a Battery................................................................................................. 122
Disclaimer .......................................................................................................................................... 124

6
 Electric Machines – Basic Theory

1
1.1.
Chapter I-ELECTROMAGNETISM

Basic definitions
There are four fundamental interactions known to exist (fundamental forces, are the interactions that
do not appear to be reducible to more basic interactions): 1) the gravitational, 2) the electromagnetic
3) the strong and 4) the weak interactions. The first two produce significant long-range forces and the
last two produce forces at minuscule, subatomic distances and govern nuclear interactions.

1.2. The second fundamental interaction – Electromagnetism

Electromagnetism is a branch of physics involving the study of electromagnetic force, a type of
physical interaction that occurs between electrically charged particles. The electromagnetic force is
carried by electromagnetic fields composed of electric fields and magnetic fields.

Figure 1.1 Electromagnetic field.

Let´s assume a current circulating through a conductor. This current will create a magnetic field
surrounding the conductor. We can express the mathematical relationship of the magnetic field in a
point with the following equation:

=∮ ⃗∙ ⃗ 1.1

⃗: magnetic field vector (other definitions are Magnetic flux density and Magnetic induction as well)
Where

I: current circulating through a conductor

: vacuum permeability. Air permeability is assumed the same.

As ⃗// ⃗ and B is uniform in all point of the area under analysis:

=∮ .
= .2
1.2
1.3

7
 Electric Machines – Basic Theory

= 1.4

Electromagnetic phenomena can be defined in terms of the electromagnetic force, by the Lorentz
force law in its vector form1:

⃗= ⃗ + (!⃗ X ⃗) 1.5

⃗: force on the particle.

Where

: particle electric charge.

⃗: electric field vector.
!⃗: particle´s velocity.
⃗: magnetic field vector (other definitions are Magnetic flux density and Magnetic induction as well)
⃗: Electric force.
$!⃗ ∙ ⃗%: Magnetic force.
X: operator to multiply phasors.

1.2 Force generated by the interaction of the magnetic field with a particle.

Using the definition of the cross product, the magnetic force can also be written as a scalar equation:

&'()*+,-. = / 0123 1.6

Where 3 is the angle between the velocity of the particle and the magnetic field.

1
Wikipedia

8
 Electric Machines – Basic Theory

1.3 Rotating spire with i current circulating through.

If instead a single conductor, we have a spire rotating counterclockwise with a continuous current
flow (because of the particle movement) through a conductor of length l exposed to a uniform
magnetic field will give us an equivalent expression for the magnetic force.
⃗ = 4⃗ X ⃗ 1.7
When the spire is rotating because of the torque produced by the magnetic force, a voltage is induced
between the terminals of the spire. The value of this voltage at each side of the spire is given by the
equation:
5⃗ = 6⃗ X ⃗ 1.8

1.3. MFF Waveforms

In physics, the magnetomotive force (MMF) is a quantity appearing in the equation for the magnetic
flux in a magnetic circuit. It is the property of certain substances or phenomena that give rise to
magnetic fields:

The magnetic flux can be expressed through the following equation:

7= X8= ∙ 9 ∙ :;03 1.9

magnetic flux, and 3 is the angle between the magnetic field lines and the normal (perpendicular) to
Where B is the magnitude of the magnetic flux density, S is the area of the surface crossed by the

S.

The magnetic flux is related to the current

7 = < ∙ ℜ = Magnetic flux

-
1.10
>

And the magnetomotive force (MMF) will be:

?&& = < ∙ ∙ ℜ@ = < ∙ 1
-
ℜ>
1.11

1: current circulating through the spire

Where

<: number of turns in the coil.

9
 Electric Machines – Basic Theory

1.4. Main Applications in Electricity

1.4.1. Transformers
Electromagnetic induction, the principle of the operation of the transformer, was discovered
independently by Michael Faraday in 1831 and Joseph Henry in 1832. Only Faraday furthered his
experiments to the point of working out the equation describing the relationship between
Electromotive Force (emf) and Magnetic Flux now known as Faraday's law of induction2:

/=
ABC
A,
1.12

where

/ : Magnitude of the emf in volts.

7D : Magnetic Flux through the circuit in webers.

A transformer is a passive electrical device that transfers electrical energy from one electrical circuit
to another, or multiple circuits by magnetic coupling with no moving parts. An alternating current in
any winding of the transformer produces a time-varying magnetic flux in the transformer's core,
which induces a time-varying electromotive force (voltage) across any other windings wound around
the same core. Transformers are used to convert electrical energy between high and low voltages, to
change impedance, and to provide electrical isolation between circuits3.

A simple transformer consists of two electrical conductors called the primary winding and the
secondary winding. These two windings can be considered as a pair of mutually coupled coils. Energy
is coupled between the windings by the time-varying magnetic flux that passes through (links) both
primary and secondary windings.

Figure 1.4 Basic Transformer1-3.

1.1.4.1. Analysis of Magnetically coupled circuits.

From Figure 1.4

7E = 7EF + 7E' + 7 ' 1.13

2
Wikipedia.
3
Paul C. Krause, Oleg Wasynczuk, Scott D. Sudhoff. “Analysis of Electric Machinery and Drive Systems.” IEEE Press Power Engineering
Series.

10
 Electric Machines – Basic Theory

7 =7 F+7 ' + 7E' 1.14

Important Definitions assuming linear magnetic circuits:

7EF = <E ∙ ℜG : Leakage flux coil 1

-
1.15
GH

7 F = < ∙ ℜI : Leakage flux coil 2

-
1.16
IH

7E' = <E ∙ G : Magnetizing flux coil 1

-
ℜ>
1.17
7 ' = < ∙ ℜI : Magnetizing flux coil 2
-
1.18
>

ℜ = : Reluctance.
F
∙J
1.19

: mean of the equivalent length of the magnetic path.

K: Cross-sectional area of the magnetic path.
: permeability
ℜEF : Reluctance of leakage path 1
ℜ F :Reluctance of leakage path 2
ℜ@ : Reluctance of the path of the magnetizing fluxes
< ∙ 1: magnetomotive force (MFF). It represents the potential that a hypothetical magnetic charge

7E : Flux linking coil 1

would gain by completing the loop. Ampere-turns.

7 : Flux linking coil 2

We can write:

7E = <E ∙ G + <E ∙ G + < ∙ I

- - -
ℜ ℜ ℜ
1.20
GH > >

7 = < ∙ ℜ + < ∙ ℜ + <E ∙ ℜ

-I -I -G
1.21
IH > >

The voltage equations can be expressed as:

/E = E 1E +
ALG
A,
1.22
/ = 1 + A,I
AL
1.23

Where

ΨE = 7E <E = <E ∙ G + <E ∙ G + <E ∙ < ∙ I : Flux linkage in coil 1

- - -
ℜ ℜ ℜ
1.24
GH > >

Ψ = 7 < = < ∙ ℜ + < ∙ ℜ + <E ∙ < ∙ ℜ : Flux linkage in coil 2

-I -I -G
1.25
IH > >

NEE = <E ∙ ℜ + <E ∙ ℜ = NEF + NE' : self-inductance of coil 1

E E
1.26
GH O

N =< ∙ +< ∙ =N F+N ':

E E
ℜIH ℜO
self-inductance of coil 2 1.27
NE = <E ∙ < ∙ ℜ
E
: Mutual inductance in coil 1 1.28
>

N = < ∙ <E ∙ ℜ
E
E : Mutual inductance in coil 1 1.29
>

NE = PI ∙ NE' = PG ∙ N
P P
' 1.30
G I

NEF : Leakage inductance in coil 1

N F : Leakage inductance in coil 2

11
 Electric Machines – Basic Theory

The inductance in matrixial form can be written as:

NEF + NE' ∙ NE'

PI
N NE
Q = R EE S = T PG U
PG
N E N ∙N N F+N
1.31
PI ' '

And the flux linkages:

ΨE = NEF ∙ 1E + NE' (1E + PI ∙ 1 )

P
1.32
G

Ψ =N F∙1 +N ' (1 + P ∙ 1E )
PG
1.33
I

And in matrixial form:

V=Q∙W 1.34

From Equations 1.22, 1.23, 1.31 and 1.34

/E = E 1E + NEE ∙ + NE ∙
-G -I
X X
1.35
/ = 1 +N ∙ +N E∙
-I -G
X X
1.36

The obtained equivalent circuit is shown in Figure 1.5

Figure 1.5 Equivalent circuit for a single-phase transformer

We can select the coil 1 or coil 2 as a reference, then choosing coil 1 as a reference and moving equations
from coil 2 we have:

1Y = I 1
P
P
1.37
G

/ Y = PG / ,
P
and 1.38
I

Ψ Y = PG Ψ
P
1.39
ΨE = NEF ∙ 1E + NE' (1E + 1 )
I
Y

Ψ Y = NYF ∙ 1 Y + NE' (1E + 1 Y )

1.40
1.41
= ZPG [
Y P
1.42
I

NYF = Z G [ NF
P
PI
1.43

/E = E 1E +
ALG
A,
1.44

12
 Electric Machines – Basic Theory

/Y = 1 +
Y Y A\I]
A,
1.45

The new equivalent circuit referred to as coil 1 is shown in Figure 1.6

Figure 1.6 Transformer equivalent circuit referred to coil 1.

1.1.4.2. Effect of saturation

In the development of the above equations, we have assumed parameters being linear, but when the
magnetic flux circulates through a ferromagnetic material such as iron, the magnetic flux density
exceeds the maximum relative permeability ^ , where is the vacuum permeability. In such
circumstances an increase in applied external magnetic field H cannot increase the magnetization of
the material further, so the total magnetic flux density B more or less levels off. The magnetization
flux density remains nearly constant despite of increment of H (proportional to the current circulating
through the circuit where the magnetic field is produced) and is said to have saturated.

From Figure 1.6 and assuming no saturation, we can obtain the following equations:

_E = NEF ∙ 1E + _'
_ Y = NYF ∙ 1 Y + _'
1.46

_' = NE' ∙ (1E + 1 Y )

1.47
1.48

In saturation _' is not linear but depends on the saturation characteristic of the material being used as
the core of the magnetic circuit. We can define a new term `(_' ) associated with the saturation curve
and _' will become:

_' = NE' ∙ (1E + 1 Y ) − `(_' ) 1.49

13
 Electric Machines – Basic Theory

Figure 1.7 Typical Saturation curve3.

1.4.2. Overhead Transmission lines

Balanced single-circuit three-phase lines can be studied using the same equations used for
transformers, but in this case, we have flux interactions produced by three currents shifted 120º being
the air the magnetic material where we will assume a permeability equal to 0 .

Figure 1.8 Model representing self and mutual impedances in a transposed transmission line.

Nc : self inductance
N' : mutual inductance

It is customary to describe the voltage drop along a transmission line in the form of partial differential
equations, e.g., for a single-phase line as:

− Ae = f. 1 + N A,
Ad A-
1.50

To facilitate the analysis, the voltage drops can be expressed in the form of phasor equations for AC
steady state conditions at a specific frequency where:

gc = fee + h2 ` ∙ Nc
g' = fei + h2 ` ∙ N'
1.51
1.52

fee = self-resistance of the respective phase (aa, bb, cc).

Where

fei = mutual resistance between phases (ab, bc, ca).

14
 Electric Machines – Basic Theory

In the case of the three-phase balanced and transposed line in Figure 1.8, the impedance matrix
becomes:

fmm + hnNmm fmo + hnNmo fm: + hnNm: gc g' g'

jkl = Tfmo + hnNmo foo + hnNoo fo: + hnNo: U=pg' gc g' q
fm: + hnNm: fo: + hnNo: f:: + hnN:: g' g' gc
1.53

If we assume symmetrical space between conductors or transposed lines, then:

fmm = foo = f:: m2 fmo = fo: = fm: 1.54

The voltages in the three phases can be calculated using the equation 1.55

gc g' g'
jrstu l = pg' gc g' q jvstu l
g' g' gc
1.55

1.4.3. Rotating machines

The spire of Figure 1.3 produces an MFF that is a square wave, as we can observe in Figure 1.9

Figure 1.9 MFF produced by the spire of Figure 1.3

By adding more spires uniformly distributed, the MMF wave distribution can evolve from a square
wave toward a sinusoid as we can see in Figure 1.10.

Figure 1.10 MMF waveform due to several spires4.

4
Kundur (1994).

15
 Electric Machines – Basic Theory

Finally, with multiple spires uniformly distributed we can have each phase producing a sinusoidally
distributed MMF wave in the three phases. This is not entirely true and for that, there are a certain
amount of voltage and current harmonics produced during the electromagnetic conversion.

The MFF due the three phases shifted 120º can be described with the following equations:

??&( = w ∙ 1( ∙ cos (3 )
??&( = w ∙ 1( ∙ cos (3 − 2 /3)
1.56

??&( = w ∙ 1( ∙ cos (3 + 2 /3)

1.57
1.58

With balanced sinusoidal phase currents and time origin when 1( is maximum or minimum
(reference):

1( = ' ∙ cos (|c X)

1} = ' ∙ cos (|c X − 2 /3)
1.59

1} = ' ∙ cos (|c X − 2 /3)

1.60
1.61

|c = 2 `: angular frequency of stator currents in electrical radian/s, being `: the frequency.

Where

Figure 1.11 Evolution of the rotational magnetic field with multiple spires in each winding.

Having sinusoidal waveforms, the total MFF due to the three-phase is given by: (Kundur, 1994).

??&,~, = ??&( + ??&} + ??&.

2 2 2 2
1.62
= w ∙ ' ∙ jcos(|c X) ∙ cos(3 ) + cos •3 − € ∙ cos •|c X − € + cos •3 + € ∙ cos •|c X + €l
3 3 3 3
??&,~, = w ∙ ' ∙ cos(3 − |c X)
•
1.63

The MFF has a sinusoidal spatial distribution with a constant amplitude and a space-angle |c X what is
function of time. The MFF is continuously rotating at constant speed |c electrical radians/s called
synchronous speed.

1.4.3.1. Complementary definitions

The synchronous speed is a multiple of the effective mechanical speed of the machine, and it depends
on the number of poles p in the machine. Both are related as follows:

|c' = |c : mechanical speed in radians/s.

‚

Vendors usually give this value in rpm where:

16
 Electric Machines – Basic Theory

2c = |c' = `: rpm
ƒ E
‚

We need to point out that for balanced operation the MMF wave to stator currents is stationary with
respect to the rotor.

17
 Electric Machines – Basic Theory

2
2.1.
Chapter II-TRANSFORMATIONS

Definition
Transformations are mathematical tools that facilitate the solution of complex problems in
Engineering. In the case of the analysis of electromagnetic phenomena, one of the variables that
introduce a great difficulty in the solution of the equations is the time variable. Some transformations
allow creating an alternative space where time or its influence can be eliminated, facilitating the
solution of complex equations. Other transformations allow us to analyze asymmetric phenomena as
unbalanced currents, decomposing circuits into a set of new ones associated with balance currents that
together represent the original one.

The simplest transformation is the one that converts a periodic sinusoid into a phasor

Figure 2.1 Phasor equivalent of a periodic sinusoid.

Table 2.1 Phasor representation.

Polar Form Rectangular Complex form Exponential Phasor form

|…|∠3 ‡ + hˆ |…|(:;0 3 + h012 3) |…|‰ Š‹ … or …

form form

Being |…| = Œ‡ + ˆ where h = √−1

Other transformations can be classified as follows:

• Sequence components in a three-phase power system.

• Clarke and Park transformations.
• Modal Analysis transformation.
• Laplace transformation.

18
 Electric Machines – Basic Theory

These transformations have applications in the different components of the Power System (lines,
transformers, generators and motors mainly).

2.2. Sequence components in a Three-Phase Power System

To facilitate the analysis in unbalanced faults, the three-phase power system can be broken down into
several symmetrical sequence components (Dr. Charles Fortescue. 1918), and so, applying the
superposition principle, several important relations can be found:

2.2.1. Symmetrical Components5

phasors, •( , •} , •. , and the components which are to replace them, can be represented in the same
Since the angles between phasors revolving at the same rate are fixed, a set of three voltage or current

vector diagram with any current or voltage phasor revolving at the same rate, as reference phasor. A
balanced and symmetrical set of three phasors have equal magnitude with phase shift of 120º.

Figure 2.2 Balanced and symmetrical three-phase phasors.

In any unbalanced or non-symmetrical part of the system, the symmetrical components methodology
applies to any polyphase system (this methodology is based on “create three balanced networks”
representing together the unbalance network). Because of the widespread use of three-phase systems
and the greater familiarity which electrical engineers have with them, they are used in practically all
the applications related to short circuit calculations (the methodology used is now called the “method
of symmetrical components”). Symmetrical component equations were developed by Dr. Charles
Fortescue and reported in a paper in 1918.

2.2.1.1. Sequence components-fundamentals

Let us assume three phasors of an unsymmetrical but balanced system, “rotating counterclockwise”.
The sum of the three phasors is zero. When three or more phasors are in a balanced system, they can
be represented in a polygon. See Figure 2.3.

Figure 2.3 A balanced system can be represented as a polygon.

phasofrom is different to zero. We can make a “trick” by creating an artificial vector •• to build a
Let´s now assume three phasors of an unsymmetrical and unbalanced system. The sum of the three

polygon.
5
Edith Clarke: “Circuit Analysis of AC Power Systems”, General Electric Series, 1961

19
 Electric Machines – Basic Theory

Figure 2.4 An unbalanced system cannot be represented as a polygon.

VR0

Figure 2.5 Artificial 3• phasor to balance the system “creating” a polygon.

Add them up, as in Figure 2.5, i.e.,

•( + •} + •. = •• 2.1

So, we see that,

•( + •} + •. + •• = 0 2.2

defining

• = • ••
E
2.3

Then

•( + •} + •. + 3• = 0
(•( + • ) + (•} + • ) + (•. + • ) = 0
2.4
2.5

Defining

•′( = (•( + • )
•′} = (•} + • )
•′. = (•. + • )
2.6

Then

•′( + •′} + •′. = 0 2.7

Conclusion: We obtain an unsymmetrical set of balanced voltages that sum zero by adding • from
each original phasor, where • is 1/3 of the resultant phasor, illustrated in Figure 2.5.

20
 Electric Machines – Basic Theory

V´c
V0 V´a
Vc
V´b V0
V0
Vb Va
Figure 2.6 Adding V0 from unsymmetrical phasors.

Now, the system has become an unsymmetrical balanced system.

Figure 2.7 A balanced system where the sum of the three phasors is zero.

This balanced system can break down into two a-b-c symmetrical sets or in an a-b-c and in an a-c-b
system, both symmetrical.

Figure 2.8 Adding 2 symmetrical sets.

The three a-b-c phasors will be called positive sequence and the a-c-b ones, negative sequence.
The sum of these unsymmetrical phasors is zero, since we just added two phasor sets that sum zero,
i.e.,

•′(E + •′}E + •′.E = 0

•′( + •′} + •′. = 0
V(Y + V}Y + V.Y = 0
2.8

21
 Electric Machines – Basic Theory

Conclusion: In this example, we can represent any unsymmetrical set of 3 phasors that sum zero as
the sum of 2 constituent symmetrical sets:

• A positive (a-b-c) sequence set.

• A negative (a-c-b) sequence set.

Extension: We can represent any unsymmetrical set of 3 phasors as the sum of 3 constituent sets, each
having 3 phasors:

• A positive (a-b-c) sequence set.

• A negative (a-c-b) sequence set.
• An equal set.

Three sets we call respectively:

Positive (•(E , •}E , •.E ).

Negative (•( , •} , •. ).
•

Zero (•( , •} , •. ) sequence components.

•
•

The implication of this is that any unsymmetrical set of 3 phasors •( , •} , •. , can be written in terms of
the above sequence components in the following way:

•( = •(E + •( + •(
•} = •}E + •} + •(
•. = •.E + •. + •(
2.8

Equations 2.16 can be standardized, but first, we must describe a mathematical operator that is essential.

2.2.1.2. The “a” operator

The operator “j” is familiar to us, since is used in complex numbers.

“j” is a vector with a magnitude and an angle equal to:

h = 1∠90° 2.9

In the same way, we are going to define the “a” operator as:

m = 1∠120° 2.10

And develop the following relationships:

m = 1∠ − 120°
m• = 1∠0°
2.11

m– = 1∠120° = m
2.12
2.13

We also have that:

1 + m = −m = 1∠60° 2.14

22
 Electric Machines – Basic Theory

a
1+a

1
Figure 2.9 Illustration of 1+a.

Note that:

−m = −1∠240° = 1∠60° 2.15

1 + m = −m = 1∠ − 60°
Similarly, we may show that:

1 − m = √3∠ − 30°
2.16

1 − m = √3∠30°
2.17

m − 1 = √3∠150°
2.18

m + m = 1∠ − 180°
2.19
2.20

V0

V0

V0

V0

V0
V0

2.10 Summary of the sequence components.

2.2.2. Clarke and Park Components6

On 1943, Edith Clarke in his book “Circuit Analysis of AC Power Systems,” Volume I, Chapter X
developed a system called α, β and 0 components. Initial ideas come from a paper written on 1929 by
R.H. Park, “Two-Reaction Theory of Synchronous Machines,” where three-phase system AC
6
Edith Clarke; “Circuit Analysis of A-C Power Systems, Volume I, Symmetrical and Related Components.” June 1943.

23
 Electric Machines – Basic Theory

quantities were reduced into two DC quantities shifted both 90º. These components were called 0.
The 0 transform is essentially an extension of the Clarke transform, applying an angle
transformation to convert from a stationary reference frame to a synchronously rotating frame. The
synchronous reference frame can be aligned to rotate with the voltage (e.g., used in voltage source
converters) or with the current (e.g., used in current source converters).

With phase a reference in a three-phase system, the α, β and 0 components of current and voltages are
defined as follows:

• α components in phases b and c are equal; they are opposite in sign and of half the magnitude
of α component of phase a.
• β component in phases b and c are equal in magnitude and opposite in sign; in phase a, they
are zero.
• 0 components are identical in the three phases (similar to zero sequence components).

Components of current flow into a three-phase circuit in phase a, returning one-half in phase b and
one-half in phase c, while β components are circulating currents in phases b and c. 0 components are
zero sequence components taken over from symmetrical components without change, except in
notation. E. Clarke decided to call 0 components for brevity and to indicate that they are to be used
with α and β components.

2.2.2.1. Relations between Phase Currents and Voltages and their α, β and 0 Components

The constant coefficients required to express a set of three vectors •( , •} , •. of a three-phase system in
terms of their α, β and 0 components are as follows:

⎡1 − − ⎤
E E
•› •(
⎢ √• ⎥
T•œ U = • ⎢0 √•
− ⎥ p•} q
• ⎢E E ⎥ •.
2.21
E
⎣ ⎦

Defining:

⎡1 − − ⎤
E E

⎢ √• ⎥
£ = • ⎢0 √ − ⎥
•

⎢E E ⎥
2.22
E
⎣ ⎦

We see that Equation 2.21 can be written as follows:

•› •(
•
T œ U = £ p•} q
• •.
2.23

We may also obtain the α, β and 0 quantities from the abc (phase) quantities:

•( •›
•
p•} q = £ T œ U
¤¥

•. •
2.24

Where

24
 Electric Machines – Basic Theory

1 0 1
⎡ E √• ⎤
£¤¥ = ⎢− 1⎥
⎢ E ⎥
2.25
⎣− − 1⎦
å

The corresponding equations for currents are as follows:

› ( ( ›
T œ U = £ p }q p } q = £¤¥ T œ U 2.26
. .

The above Clarke transformation preserves the amplitude of the electrical variables to which it is
applied.

Having three-phase symmetrical currents

1( = √2 •¦§ ∙ cos (3)

1} = √2 •¦§ ∙ cos (3 − )
2.26
•
2.27
1. = √2 •¦§ ∙ cos (3 + )
•
2.28

And applying T, we obtain

1› = √2 •¦§ ∙ cos (3)

1œ = √2 •¦§ ∙ sin (3)
2.29

1 =0
2.30
2.31

The Park transformation (named after Robert H. Park), where the rotor is the new reference converts
vectors from the abc reference frame (stator) to the dq0 reference frame (rotor), to take advantage of
DC parameters when vectors from stator reference are converted to rotor reference, that simplifies the
calculation during dynamic studies. The primary value of the Park transform is to rotate the reference
frame of a vector at an arbitrary frequency. The Park transform shifts the frequency spectrum of the
signal such that the arbitrary frequency now appeit’s as "DC" and the old DC appears as the negative
of the arbitrary frequency. The Park transformation matrix is as follows:

cos (3) sin (3) 0

ª« = p−012(3) cos(3) 0q
0 0 1
2.32

The “normalized” Park transformation ¬ is the Clarke and Park transforms combined:

⎡cos (3) cos (3 − •

) cos (3 +
•
)⎤
⎢ ⎥
Original Park transformation: ¬ = ª« ∙ £ = • ⎢ 012(3) sin (3 − •
) sin (3 + • ) ⎥
⎢ E E E ⎥
⎣ ⎦

cos (3) 012(3) 1

And its inverse: ¬¤¥ = • ·cos (3 − •
) sin (3 − •
) 1¸
cos (3 + ) sin (3 + ) 1
• •

25
 Electric Machines – Basic Theory

£ and ¬ are not unitary because £¹ ≠ £¤¥ and ¬¹ ≠ ¬¤¥ ; because that, the transformation has not the
power invariant characteristic as we can see below:

/( , 1( /A 1A
»(}. = /( 1( + /} 1} + /. 1. = p/} q p1} q = T¬¤¥ p /¼ qU ·¬¤¥ p1¼ q¸
/. 1. /½ 1½
2.33

1A
»(}. = j/A /¼
/½ l¾¬¤¥ ¿¹ ¬¤¥ p1¼ q
1½
2.34

0 0⎤
•
⎡
⎢ ⎥
¾¬¤¥ ¿ ¬¤¥ = ⎢0 0⎥, then:
¹ •

⎢ E⎥
⎣0 0 • ⎦
0 0⎤
•
⎡ 1A
/ / / ⎢ ⎥
»(}. = j A ½ l ⎢0 0⎥ p1¼ q = (/A 1A + /¼ 1¼ ) + /½ 1½
• • E
¼
•
⎢ E ⎥ 1½
2.35

⎣0 0 • ⎦

»(}. = /( 1( + /} 1} + /. 1. = (/A 1A + /¼ 1¼ ) + /½ 1½ 7, using the original Park transformation.

• E
•

The 3/2 factor comes from the constant used in the transformation. Note that the q and d voltages,
currents, flux linkages, and electric charges are dependent upon the angular velocity of the frame of
referethe nce, being the total power independent of the frame of reference.

For Power invariant transformation, T unitary becomes8:

⎡1 − − ⎤
E E

⎢ √• ⎥
£ = Á ⎢0 √•
− ⎥
•
⎢E E ⎥
2.36
E
⎣√ √ √ ⎦

Using per unit values, the “normalized” unitary power invariant Park transformation ¬ becomes:

⎡1 − − ⎤
E E
cos (3) sin (3) 0
⎢ √• ⎥
¬ = p−012(3) cos(3) 0q ∙ Á ⎢ 0 √•
− ⎥
•
0 0 1 ⎢E E ⎥
2.37
E
⎣√ √ √ ⎦

⎡ cos (3) cos (3 − • ) cos (3 + • ) ⎤

⎢−012(3) −sin (3 − ) −sin (3 + )⎥ 9
¬ = Á• ⎢ • • ⎥
⎢ ⎥
2.38
E E E
⎣ √ √ √ ⎦
7
/½ = 3/ and 1½ = 31 , where / and 1 are zero sequence components.
8
W.A. Lewis, “A Basic Analysis of Synchronous Machine – Part I,” AIEE Trans, Vol. 77, pp. 436-456, 1958.
9 The original Park transformation is not a power invariant transformation and does not result in a d-q-0 reciprocal (symmetric) inductance

sequence component would appear to be obtained applying a scale factor of Œ1/3) to its power invariant value. Despite of this advantages,
matrix. Equation A.28 facilitates solution of the normalized equations allowing symmetry of the inductance d-q-0 matrix. The modified zero

there are several discussions on the fact that original Park transformation reflects more closely the physical features of the machine (Kundur
1994).

26
 Electric Machines – Basic Theory

Or when d-axis leads the q-axis:

⎡cos (3) cos (3 − •

) cos (3 + •
)⎤
⎢ sin (3 + ) ⎥⎥
¬ = Á ⎢ 012(3) sin (3 − •
)
•
•
⎢ E ⎥
2.39
E E
⎣ √ √ √ ⎦

Figure 2.11 Axis d-q-0 referenced to phase a.

frame (rotor) rotating with the synchronous speed nc which will be along the axis of phase a at t=0. If
Where θ is the instantaneous angle of an arbitrary ω frequency. We will assume a rotating reference

θ is the angle by which the rotor direct axis is ahead of the magnetic axis of phase a, then

3 = nc X + Ã + rad 2.40

Being à the synchronous torque angle (rotor angle); value dictated by the power generated.

The main field-winding flux is along the direction of the d axis of the rotor. It produces an efm that
lags this flux by 90º. Therefore, the machine efm E is primarily along the rotor q axis.

Being the inverse:

⎡ cos (3) sin (3)

E
√ ⎤
⎢ ⎥
¬¤¥ = ¬¹ = Á• ⎢cos (3 − • ) sin (3 − • )
E
√ ⎥
⎢ E⎥
2.41

⎣cos (3 + • ) sin (3 + • ) √ ⎦

finally, we may write:

W ÄÅ = ¬ ∙ Wstu and Wstu = ¬¤¥ ∙ W ÄÅ 2.42

27
 Electric Machines – Basic Theory

We note that ¬¤¥ = ¬¹ , which means that the transformation in per-unit values is orthogonal, and also,
power invariant, and we should expect to use the same power expression in either the a-b-c or the d-q-
0 frame of reference

Æ = /( 1( + /} 1} + /. 1. = !¹stu Wstu = (¬¤¥ ! ÄÅ ) (¬ W ÄÅ )

¹ ¤¥
= !¹ ÄÅ (¬¤¥ )¹ ¬¤¥ W ÄÅ 2.43

Æ = !¹ ÄÅ ¬¬¤¥ W ÄÅ = !¹ ÄÅ W ÄÅ = /A 1A + /¼ 1¼ + / 1 2.44

We can write similar expressions for voltages and flux linkages (Ψ = <7 :;0 n X)

! ÄÅ = ¬!stu , V ÄÅ = ¬Vstu 2.45

In the case of generators, we need to consider three particularities:

1. The transformation from the three-phase windings (N) to a system with two-phase windings

Kundur, 1994; Krause et al., 2002) use Œ2⁄3. As this is the most used widely, we will consider
(N’). Various authors select the ratio N/N’ equal to 2/3, but majority (as Fitzgerald et al., 1992;

it as the valid ratio.

2. The active and reactive powers computed in Clarke´s domain with the transformation shown
above are not the same as those computed in the standard reference frame. This is because T is
not unitary.
3. When a synchronously reference frame is used (to facilitate the analysis in synchronous
machines), both stator and rotor fluxes as seen as stationary. An additional transformation
(Park) is needed where the rotor is the new reference to transfer three-phase stator and rotor
quantities into a single rotating reference frame to eliminate the effect of time-varying
inductances and transform the system into a linear time-invariant system10.

Figure 2.12 Transformation from a three-phase system to a bi-phase system.

The most common representation that facilitates the visualization of parameters analysis is using the
d-axis ahead of the q-axis. This changes the sign of the second row in Equation 2.38:
10
Wikipedia

28
 Electric Machines – Basic Theory

Figure 2.13 d-axis ahead of q-axis

With the currents in the stator

1( = √2 •¦§ ∙ cos (|c X)

1} = √2 •¦§ ∙ cos (|c X − )
2.46
•
2.47
1. = √2 •¦§ ∙ cos (|c X + )
•
2.48

2 2
Applying the d-q transformation, we have:
⎡cos (3) cos (3 − ) cos (3 + )⎤
⎢ 3 3 ⎥
A 2 2 2 (
T ¼ U = ⎢ 012(3) sin (3 − ) sin (3 + ) ⎥ p } q
3⎢ 3 3 ⎥
⎢ 1 1 1 ⎥
.

⎣ √2 √2 √2 ⎦
1A = √2 •¦§ ∙ cos (|c X)
1¼ = −√2 •¦§ ∙ sin (|c X)
2.49

1 =0
2.50
2.51

The inverse transformation

cos (3) 012(3) 1

⎡ 2 2 ⎤
( 2 ⎢cos (3 − ) sin (3 − ) 1⎥ A
p }q = ⎢ 3 3 ⎥ T ¼U
3 2 2
⎢ ⎥
⎣cos (3 + 3 ) sin (3 + 3 ) 1⎦
.

1( = A ∙ cos(|c X) + ¼ 012(|c X) + 2.52

1} = A ∙ cos Z|c X − [+ ¼ 012 Z|c X − [+
• •
2.53
1. = A ∙ cos Z|c X + •
[+ ¼ 012 Z|c X + •
[+ 2.54

In balanced systems =0

to bi-phase system: 3 È/2 (peak value):

The current relationships taking into consideration the current equivalent from the three-phase system

N′1A = <√2 ∙ cos (|c X)

•
•¦§ 2.55
< Y 1¼ = − <√2 ∙ sin (|c X)
•
•¦§
1 =0
2.56
2.57

29
 Electric Machines – Basic Theory

Using P] = Œ2⁄3:
P

1A = √3 •¦§ ∙ cos (|c X)

1¼ = −√3 •¦§ ∙ sin (|c X)
2.58

1 =0
2.59
2.60

With currents

1( = √2 •¦§∙ sin (|c X)

1} = √2 •¦§ sin (|c X −
∙ )
2.61
•
2.62
1. = √2 •¦§ ∙ sin (|c X + )
•
2.63

We will obtain:

1A = √3 •¦§ ∙ sin (|c X)

1¼ = √3 •¦§ ∙ cos (|c X)
2.64

1 =0
2.65
2.66

2.2.3. Transformation for modal analysis

Propagation of High-frequency signals as traveling waves (TW) occurs in the other conductors, even
though only one conductor is energized. Modal analysis is a mathematical tool, similar to the
symmetrical components technique used for analysing unbalanced faults on three-phase (50 or 60 Hz)
power systems. Like Clarke components (simile is closer than with symmetrical components), modal
analysis is a mathematical tool whose components or “modes” can be generated and measured
separately. The modal theory is based on the premise that there are as many independent modes of
propagation on multiconductor circuits, as there are conductors involved in the propagation of
energy11.

There are five characteristics of natural modes in a three-phase line:

a) Any set of phase-conductor currents or voltages existing at any point on a lossy, reflection-free
three-phase line can be resolved into three sets of natural-mode components.
b) At any point on a line, the mode components must add up to the actual phase quantities. Also, the
total power derived from phase currents and voltages must be equal to the sum of mode powers
c) The ratio of mode voltage to mode current (the mode “characteristic impedance”) is constant on
each phase conductor.
d) Each mode propagates with a specific attenuation per unit length and a specific velocity of
propagation.
e) A set of phase components corresponding to one mode only cannot be resolved into other modes.
The modes are independent, and there is no inter-mode coupling on a uniform line.

There are three modes of propagation:

1. Mode 1 is a high attenuation mode that is propagated on all three phases with the ground
return.
2. Mode 2 is a medium-attenuation mode that is propagated on one outside phase and returns on
the other outside phase. There is no mode current in the center phase.
3. Mode 3 is the least attenuated of the three modes. The energy is propagated on the two outer
phases and returns to the center phase.
11
Applied Protective Relaying. ®Westinghouse Electric Corporation, 1976.

30
 Electric Machines – Basic Theory

To use matrix algebra in determining modal quantities, the following assumptions are made:

a) Line is transposed, so the impedance matrix is symmetrical.

b) No effect of frequency.
c) Instantaneous currents (phase or modal quantities) in the three-phase are either in phase or 180º
out of phase.

Figure 2.14 Simplified presentation of the propagation modes.

From Figure 2.14, we can define a transformation matrix, as in the case of Clarke components, but for
propagation modes:

1 1 1 1 1 1
8 = p1 0 −2q and 8¤¥ = p3/2 0 −3/2q
E

1/2 −1 1/2
•
1 −1 1
2.67

matrices kstu are of the form:

We can apply this transformation to a special case with balanced impedances, i.e., the impedance

gc g' g'
k = kstu = pg' gc g' q
g' g' gc
2.68

where

gc = Self-impedance of each conductor (assumed equal in the three conductors)

g' = Mutual impedance (three-phase line transposed) is equal between conductors.

The matrix for modal impedances is the same as the matrix impedances obtained with other
transformations as the Clarke or symmetrical components ones.

gc + 2g' 0 0
kÉ = p 0 gc − g' 0 q
0 0 gc − g'
2.69

2.2.4. Laplace Transformation

The Laplace transformation, named after its inventor Pierre-Simon Laplace, is an integral transform
that converts a function of a real variable t (often time) to a function of a complex variable s (complex
frequency). The Laplace transform reduces a linear differential equation to an algebraic equation,
which can then be solved by the formal rules of the algebra. After the algebraic equation has been
combined (added, multiplied) with other algebraic equations, the result to a new differential equation
can then be solved by applying the inverse Laplace transformation.

ℒË`(X)Ì = Í `(X)‰ ¤c, X = &(0)

Î
2.70

31
 Electric Machines – Basic Theory

Figure 2.15 Time domain and frequency domain transformation6

This transformation has several properties, some of them are as follows12:

a) Linearity

ℒËm`(X) + oÏ(X)Ì = mℒË`(X)Ì + +oËÏ(X)Ì = m&(0) + oÐ(0) 2.71

ℒ ¤E Ëm&(0) + oÐ(0)Ì = mℒ ¤E Ë&(0)Ì + oℒ ¤E ËÐ(0)Ì = m`(X) + oÏ(X) 2.72

b) Derivatives of `(X)

ℒË`′(X)Ì = 0ℒË`(X)Ì − `(0) = 0&(0) − `(0)

⋮
2.73

ℒÒ` (*) (X)Ó = 0 (*) &(0) − 0 (*¤E) `(0) − 0 (*¤ ) ` Y (0) − ⋯ − ` (*¤E) (0) 2.74

c) Derivatives of F(s)

ℒ ¤E Ë&′(0)Ì = −X`(X)
⋮
2.75

ℒ ¤E Ò& (*) (0)Ó = (−1)* X * `(X) 2.76

d) Integrals of `(X)

ℒ ÕÍ `(Ö) Ö× = &(0)
, E
c
⋮
2.77

ℒ ÕÍ … Í `(Ö)( Ö)* × = cÙ &(0)

, , E
2.78

e) Integrals of F(s)

ℒ ¤E ÒÍ &(Ö) ÖÓ = `(X)
Î E
,
⋮
2.79

ℒ ¤E ÒÍ … Í &(Ö)( Ö)* Ó = , Ù `(X)

Î Î E
2.80

12
https://www.efunda.com/math/laplace_transform/rules.cfm

32
 Electric Machines – Basic Theory

f) Convolution

ℒ ¤E Ë&(0)Ð(0)Ì = Í `(Ö)Ï(X − Ö) Ö = `(X) ∙ Ï(X)

,
2.81

Some examples:

Constant function `(X) = 1; the Laplace transformation is .

E
c
i.
Linear function `(X) = X; the Laplace transformation is cI
E
ii.
The power function `(X) = X * ; the Laplace transformation is
*!
cÙÛG
|…|‰ Š‹
iii.
iv. We can express phasor of Figure 2.1 in exponential form with C=1, then:

‰ Š‹ = :;03 + h0123 2.82

Equation 2.82 is also known the Euler´s formula13

If we have a function `(X) = sin(mX), using Equation 6.36 we obtain:

sin(mX) =
+ ÜÝÞ ¤+ ßÜÝÞ
Š
2.83

The Laplace transformation of sin(mX) then becomes:

ℒËsin (mX)Ì = Í $‰ − ‰ ¤Š(, %‰ ¤c, X=

E Î Š(, (
Š ( I àcI
2.84
ℒËcos (mX)Ì = (ℝ‰(0) > 0)
c
( I àcI
2.85

In a similar way and using Equation 2.70 and the properties, we can obtain the Laplace transformation
of other functions.

13
Kim Thibault https://mathvault.ca/laplace-transform/

33
 Electric Machines – Basic Theory

3
3.1.
Chapter III-SYNCHRONOUS
MACHINE PARAMETERS’
DERIVATION. PART 1
Synchronous Machine Derivation

Synchronous Generators or Alternators are used to convert mechanical power derived from steam,
gas, hydraulic-turbine, or wind to AC electric power. Synchronous generators are the primary source
of electrical energy we consume today, and large AC power networks rely almost exclusively on
synchronous generators.

A conductor winding, through which a DC current circulates continuously, mechanically propelled by

an external element, moves across producing a rotating electro-magnetic field. This electromagnetic
field interacts with the stator winding (named sometimes alternator), and in this way, sinusoidal
voltage (AC) is produced across the terminals of the induced circuit (Stator). See Figure 3.1.

The rotor magnetic field may be produced by induction (in a "brushless" generator), by permanent
magnets (usually in very small machines), or by a rotor winding energized with direct current through
slip rings and brushes. Automotive alternators invariably use brushes and slip rings, which allow
control of the alternator-generated voltage by varying the current in the rotor field winding.
Permanent magnet machines avoid the loss due to magnetizing current in the rotor but are restricted in
size owing to the cost of the magnet material. Since the permanent magnet field is constant, the
terminal voltage varies directly with the speed of the generator. Brushless AC generators are usually
larger machines than those used in automotive applications.

Generators, due to their construction and to the needs of the industry itself, generally produce three-
phase voltage. We will therefore limit ourselves to an analysis of three-phase circuits, fundamentally
in the operating frequency of 50 and 60 Hz, standardized worldwide values.

A rotating generator has two basic components:

• Armature or Stator.
• Field circuit or Rotor.

a) Stator winding b) Rotor Winding

Figure 3.1 Basic Parts of a Synchronous Motor (Source: Wikipedia).

34
The rotor of the generator is driven by a prime mover. A DC current is flowing in the rotor winding
which produces a rotating magnetic field within the machine and the rotating magnetic field induces a
three-phase voltage in the stator winding of the generator.

An elementary two-pole three-phase generator is illustrated in Figure 3.3. The stator contains three
coils aa´, bb´, cc´, displaced from each other by 120 electrical degrees.

be approached to sinusoidal functions of 7c , having a relatively good approximation of sinusoidal

The waveforms of the MMF´s produced in the stator are built using a discrete structure, but they can

airgap MMF with respect to 7c . This is possible assuming that windings are distributed sinusoidally.
This is not entirely true and for that, there are a certain amount of voltage and current harmonics
produced during the electromagnetic conversion.

3.2. Generator Model

∆Ptie
∆PC

∆PV
∆PG , ∆QG

Figure 3.2 Generator Model

3.3. Theory of Operation

When the rotor is excited to produce an air gap magnetic flux 7 per pole and is rotating at constant
angular speed w, the magnetic field cuts the stator conductors and three voltages, which are displaced
by 120º are induced in the three coils aa´, bb´ and cc´. When these three coils (or windings) are
connected to a three-phase balanced load, the resulting currents are also displaced by 120º (See Figure
3.3). These currents will, in turn, each produce a magnetic field and the resultant magnetic field in the

7c ) and the magnetic field produced by the field winding (the rotor magnetic field, 7 ).
airgap is the combination of the magnetic fields produced by the stator currents (stator magnetic field,

The flux linkage of the coil varies with the position of the rotor magnetomotive force axis |t (mmf -
force produced by current through a coil of wire), where |t is measured in electrical radians from the
coil aa´ magnetic axis. As shown in Figure 1.3, when angle is at zero degrees (|t=0) the current in

www.adneli.com
 Electric Machines – Basic Theory

phase a is at its positive maximum ( È), producing a component of the stator magnetic field, 7( ,

and c are at their negative half maxima (− È/2) producing two components of the stator magnetic
where the magnitude is proportional to the number of ampere-turns NIm and the currents in phases b

field, 7} and 7. , whose magnitudes are proportional to the number of ampere-turns NIm/2 along the
axes of phases b and c, respectively. The flux linkage in phase a will be zero at nX = /2. At angle of
-120º (nX = − • ), in coil bb´ the current in phase b is at its positive maximum, while currents in
phases a and c are at their negative half maxima (− È/2). At -240º (nX = − ), in coil cc´ the
•

half maxima (− È/2). The total flux 7c produced by the three phases in the stator is the summation
current in phase c is at its positive maximum, while currents in phases a and b are at their negative

of the three fluxes, that is:

7c = 7( + 7} + 7. 3.1
Phase A
Current Phase Relationship in the stator windings
Phase B

Phase C
250
200
150
100
50
Amplitude

0
0 30 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570
-50
-100
-150
-200
-250
Degrees

Figure 3.3 Currents produced in the stator windings.

Figure 3.4 Three-coil representation of the stator.

currents in the armature. Assuming current in phase a lagging the generated emf ‰( by an angle 3 =
In normal conditions, the generator operates at synchronous speed, this results in three-phase balanced

|X, the instantaneous armature currents are:

1( = √2 •¦§ ∙ cos (3)

1} = √2 •¦§ ∙ cos (3 − )
3.2
•
3.3
1. = √2 •¦§ ∙ cos (3 + •
) 3.4

36
 Electric Machines – Basic Theory

3.3.1. Representation of the d-q non-reduced order model

phase a, carrying a current equivalent to 3 È/2 (peak value) and rotating as synchronous speed nc . In
In steady state for a balanced system, we can simplify the analysis using a single coil aligned with

this case, in phase a flux linkage Ψ( will vary as the cosine of angle |t.

Ψ( = <7 :;0 | X 3.5

where:

<: Number of turns

7: Air gap flux per pole
Ψ( : Flux linkage

Windings in the electrical machines are magnetic coupled electrical circuits with coefficients that
determine the current magnitudes circulating through the windings. This characteristic allows us the
analysis from the perspective of electrical circuits. Equation 3.6 relates to this relationship, being L
(inductance) the parameter that relates the current with the flux.

Ψ( = 7 ∗ < = N ∗ 1 = Í(‰ ∗ X) 3.6

Figure 3.5 Transformation (d-q) in stator and rotor circuits of a synchronous generator.

3.3.1.1. Flux linkage equations

In Figures 3.5 and 3.6 we note that we have a network consisting of six mutually coupled coils. These
are the three phase windings aa´,bb´,cc´, the field winding FF´, and the two damper windings DD´and
QQ´.

We write the flux linkage equations for these six circuits (abc for the Stator and FDQ for the Rotor) as

Ψ( N(( N(} N(. N(ä N(å N(æ 1

⎡ Ψ ⎤ ⎡ N}( N}} N}. N}ä N}å
⎤ (
N}æ ⎥ ⎡ 1} ⎤
⎢Ψ ⎥ ⎢N N.æ ⎥ ⎢ 1. ⎥
}
N.} N.. N.ä N.å
⎢ . ⎥ = ⎢⎢ .( ⎢ ⎥ Wb turns
⎢ Ψä ⎥ ⎢ Nä( Nä} Nä. Nää Näå Näæ ⎥ ⎢ 1ä ⎥
⎥
3.7
⎢Ψå ⎥ ⎢Nå( Nå} Nå. Nåä Nåå Nåæ ⎥ ⎢1å ⎥
⎣Ψæ ⎦ ⎣Næ( Næ} Næ. Næä Næå Nææ ⎦ ⎣1æ ⎦

37
 Electric Machines – Basic Theory

Figure 3.6. Flux linkage and voltage equations.

the fact that the inductances are functions of time (dependency of angle 3 = |X + Ã + ). We could
The solution of the two systems of equations (Stator and Rotor) referred to the Stator is complicated by

take some advantages in solving them in phase quantities (referenced to Stator), when space harmonics
in the magnetic field distribution had not to be taken into account, with no harmonic’s presence,
transform them from phase quantities to d-q-0 quantities, that is to refer all equations to the Rotor offers
a simpler approach, as the projection of the rotating fluxes onto the field axis appears as stationary
during steady state operation. This transformation was first proposed by Blondel, and further developed
by Doherty, Nickle and Park (it is popularized as Park´s transformation). Is has been utilized in the
majority of books and this transformation is identical for fluxes, voltages and currents, converting phase
quantities into d-q-0 quantities, with quantities on the field structure remaining unchanged3.

Figure 3.7 Axis d-q

Assumptions14:

• The resistance of each winding is constant.

• The permeance of each portion of the magnetic circuit is constant (no saturation is considered
initially).
• The armature windings are symmetrical with respect to each other.
• The electrical and magnetic circuit of the field structure is symmetrical about the direct and
quadrature axis.
• The self-inductance of each winding in the field structure is constant.
14
EMTP Theory book.

38
 Electric Machines – Basic Theory

harmonic sinusoidal function of the rotor position 3 (second harmonic is zero if the salient is
• The self and mutual inductances of the armature windings are a constant plus a second

ignored), with the amplitude of the second harmonic component being the same for all self
and mutual inductances.

is a fundamental sinusoidal function of the rotor position 3.

• The mutual inductance between any winding on the field structure and any armature winding

• Effect of hysteresis are negligible.

• Effects of eddy currents are negligible.

Figure 3.8 Variation of the permeance with the rotor position.

Variation of the permeance can be expressed according to the following equation:
℘ = ℘ + ℘ :;0è 3.8
Where è is the angular distance from d-axis along the periphery of the rotor.
The self-inductance N(( is equal to the ratio of flux linking phase a winding to the current 1( , with
currents in other circuits equal to zero. The inductance is directly proportional to the permeance ℘ and
has a second harmonic component.

peak aligned with the axis of phase a, being the peak amplitude equal to <( 1( where <( is the effective
Assuming no harmonics presence, the MMF of phase a has a sinusoidal distribution in space with its

turns per phase.

From Figure 3.8 and doing è = 3 we have the MMF in d and q axis
?&&‚(A = <( 1( :;03 3.9
?&&‚(¼ = <( 1( cos (3 − ⁄2) = <( 1( sin3 3.10
Air-gap fluxes per pole along the two axes will be:
Ψ(A = (<( 1( :;03)℘A 3.11
Ψ(¼ = (<( 1( 0123)℘¼ 3.12

Where ℘A and ℘¼ are the permeance coefficients of the axis d and q respectively.

39
 Electric Machines – Basic Theory

As the permeance include factors that relate flux per pole with a peak value of the MFF wave, the total
air-gap flux linking in phase can be expressed according to the following equation:
Ψ(( = Ψ(A :;03 + Ψ(¼ 0123 3.13

Ψ(( = <( 1( (℘A :;0 3 + ℘¼ 012 3) 3.14

012 3 = (1 − :;023)
E
3.15

:;0 3 = (1 + :;023)
E
3.16

Ψ(( = <( 1( Z + :;023[

℘é à℘ê ℘é ¤℘ê
3.17

N(( = NF + Ψ
PÝ
-Ý ((
3.18

Where NF is the leakage inductance.

Defining

NJ = <( ; ND = <( = N'

℘é à℘ê ℘é ¤℘ê
3.19

Based on the functional relation of Ψ(( and the rotor angle è , we can conclude that the self-inductance
in the phase a of the stator (neglecting loses) can be expressed according to the equation:
N(( = NF + NJ + N' :;023 3.20
Or
N(( = = Nc + N' :;023
\ÝÝ
-Ý
3.21

Where

N(( = Total self-inductance in phase a.

N' = Magnetizing inductance. It is assumed the same for all phases.

From Figure 3.7 and having Ψ( = 7 ∙ < = N ∙ 1 for phase a, and N' as the magnetizing inductance, Nc
inductance is essentially constant, independent of 3; however, the magnetizing inductance is a
periodic function of 3. That is, N' = `(3) = N' (3).

N' (3) = ℜ
PI
O (‹)
3.22

Reluctance ℜ' is maximum (permeance ℘' (è) =

E
ℜO (‹)
is minimum. From Figure 3.7, the air gap is
minimum) when 3 = 0º and N' is maximum. When 3 = , N' is minimum and the same happens
with 3 = , and so on, while it is maximum with 3 = , 2 and so on. Hence, the magnetizing
• ë

inductance varies between maximum and minimum positive values twice per revolution of the
rotating member (rotor).

The mutual inductance ab can be found by evaluating the air gap flux _}( linking phase b when only
phase a is excited. As phase b is shifted by (3 − 2 ⁄3) from phase a, we simply replace this new
angle instead 3 in the Equation 3.23.

Ψ}( = Ψ(A :;0(3 − 2 ⁄3) + Ψ(¼ 012(3 − 2 ⁄3) 3.23

Finally, we can obtain a similar equation

40
 Electric Machines – Basic Theory

N(} = N}( = −N(}c + N' :;02(3 + 2 ⁄3) 3.24

Where we have N(}c =

ìí
3.25

We can simplify the nomenclature using the one from the reference15

N(} = N}( = −Nc' + N' :;02(3 + 2 ⁄3) 3.26

As we can observe above, these variations may be adequately approximated by sinusoidal functions.

Similarly, we can define the corresponding Stator self-inductances equations for phases b and c.

N(( = Nc + N' :;0 2 3 Henrys

N}} = Nc + N' :;0 2 (3 − 2 ⁄3) Henrys
3.27

N.. = Nc + N' :;0 2 (3 + 2 ⁄3) Henrys

3.28
3.29

The Rotor self-inductances (field, direct and quadrature components) are constant; not a function of 3.

Nää = Nä Nåå = Nå Nææ = Næ Henrys 3.30

Doing a similar approach as for Stator self-inductances, Stator mutual inductances can be defined in a

component Nc' , is assumed the same for all phases.

similar way, considering the summation of fluxes from the individual phases. Non rotor position

N(} = N}( = −Nc' + N' :;02(3 + 2 ⁄3) Henrys

N}. = N.} = −Nc' + N' :;0(2 3) Henrys
3.31

N.( = N(. = −Nc' + N' :;02(3 − 2 ⁄3) Henrys

3.32
3.33

The Rotor Mutual inductances are constant; not a function of 3.

Näå = Nåä = N•' Näæ = Næä = 0 Nåæ = Næå = 0 Henrys 3.34

Stator-to-rotor mutual inductances are functions of the rotor angle 3.

From the phase windings to the field winding, Nä' is assumed the same for all phases.

N(ä = Nä( = Nä' :;0 3 Henrys

N}ä = Nä} = Nä' :;0( 3 − 2 ⁄3) Henrys
3.35

N.ä = Nä. = Nä' :;0( 3 + 2 ⁄3) Henrys

3.36
3.37

From the phase windings to the damper winding D, Nå' is assumed the same for all phases.

N(å = Nå( = Nå' :;0 3 Henrys

N}å = Nå} = Nå' :;0( 3 − 2 ⁄3) Henrys
3.38

N.å = Nå. = Nå' :;0( 3 + 2 ⁄3) Henrys

3.39
3.40

From the phase windings to damper windings Q, Næ' is assumed the same for all phases.

N(æ = Næ( = Næ' 012 3 Henrys

N}æ = Næ} = Næ' 012( 3 − 2 ⁄3) Henrys
3.41
3.42

15
Paul M. Anderson, A.A. Fouad. “Power System Control and Stability.” IEEE Power Systems Engineering Series, 1993.

41
 Electric Machines – Basic Theory

N.æ = Næ. = Næ' 012( 3 + 2 ⁄3) Henrys 3.43

The signs on mutual terms depend upon assumed current directions and coils orientation.

The matrix of equations becomes:

⎡ Nc + N' :;0 2 3 −Nc' + N' :;02(3 + •

) −Nc' + N' :;02(3 − •
) Nä' :;0 3 Nå' :;0 3 Næ' 012 3 ⎤
Ψ( ⎢ ) ⎥ ⎡ 1( ⎤
⎡ Ψ ⎤ ⎢−Nc' + N' :;02(3 + •
) Nc + N' :;0 2 (3 −
•
) −Nc' + N' :;0(2 3) Nä' :;0( 3 −
•
) Nå' :;0( 3 −
•
) Næ' 012( 3 −
⎥ 1}
•
⎢ } ⎥ ⎢−Nc' + N' :;02(3 − ) −Nc' + N' :;0(2 3) Nc + N' :;0 2 (3 + ) Nä' :;0( 3 + ) Nå' :;0( 3 + ) Næ' 012( 3 + )⎥ ⎢ 1. ⎥
⎢ Ψ. ⎥ = ⎢ • • • • • ⎥⎢ ⎥
⎢ Ψä ⎥ ⎢ ⎥ ⎢ 1ä ⎥
Nä' :;0 3 Nä' :;0( 3 − ) Nä' :;0( 3 + ) Nä N•' 0
3.44
⎢Ψå ⎥ ⎢ • •
⎥ ⎢ 1å ⎥
⎣Ψæ ⎦ ⎢ Nå' :;0 3 Nå' :;0( 3 − ) Nå' :;0( 3 + ) N•' Nå 0 ⎥ ⎣ 1æ ⎦
⎢ ⎥
• •

⎣ Næ' 012 3 Næ' 012( 3 − ) Næ' 012( 3 + ) 0 0 Næ ⎦

• •

3.3.1.1. Transformation of inductances

Nearly all terms in the matrix are time varying, since 3 is a function of time. Four of the off-diagonal
terms vanish (equal to zero). The time-varying inductances can be simplified by referring all quantities
to a rotor frame of reference through a Park´s transformation applied to the a-b-c partition.

⎡cos (3) cos (3 − • ) cos (3 + • )⎤

Having

1 1 1
⎢ ⎥
¬ = Á• ⎢ 012(3) sin (3 − • ) sin (3 + • ) ⎥, îï = p1 1 1q
⎢ E ⎥ 1 1 1
3.45
E E
⎣ √ √ √ ⎦
Vstu 16 V ÄÅ
Impedance matrix RV S can be expressed in d-q components R S using the following
ðñ V ðñ
expression:

V ÄÅ ¬ 0 Vstu ¬ 0 Qss Qsò ¬¤¥ 0 ¬ 0 Wstu

R S=R S RV S=R SR SR SR SR S
V ðñ 0 îï ðñ 0 îï Qòs Qòò 0 îï 0 ó• W ðñ
3.46

Where

N(( = stator-stator inductances.

N(• , N(• = stator-rotor inductances.
N•• = rotor-rotor inductances.

¬ Å Vstu V ÄÅ
R S RV S=R S
Å îï ðñ V ðñ
3.47
¬ 0 Wstu W ÄÅ
R S RW S=R S
0 îï ðñ W ðñ
3.48
¬ 0 Qss Qsò ¬¤¥ 0 Q ÄÅ Qõ
R SR SR S=ô ¹ ö
0 îï Qòs Qòò 0 îï Qõ Q ðñ
3.49

Q¹õ is the transposed of the stator mutual inductance Qõ

Doing all calculations, finally we obtain d-q components in terms of the 6x6 matrix:

16
Letters in bolt indicate matrix or vector arrays.

42
 Electric Machines – Basic Theory

ΨA NA 0 0 ÷Nä' ÷Nå' 0 1A
⎡Ψ ⎤ ⎡ 0 N¼ 0 0 0 ÷Næ' ⎤ ⎡ 1¼ ⎤
⎢ ¼⎥ ⎢ ⎥⎢ ⎥
⎢Ψ ⎥ = ⎢ 0 0 N 0 0 0 ⎥ ⎢1 ⎥
⎢ 0 ⎥ ⎢1ä ⎥
⎢ Ψä ⎥ ⎢ ÷Nä' 0 0 Nä N•'
⎥
Web turns 3.50
⎢Ψå ⎥ ⎢÷Nå' 0 0 N•' Nå 0 ⎥ ⎢1å ⎥
⎣Ψæ ⎦ ⎣ 0 ÷Næ' 0 0 0 Næ ⎦ ⎣1æ ⎦

where the new constants are as follows:

NA = Nc + Nc' + (3⁄2)N' Henrys

N¼ = Nc + Nc' − (3⁄2)N' Henrys
3.51

N = Nc − 2Nc'
3.52

÷ = Œ3⁄2
Henrys 3.53
3.54

3.3.1.2. Voltage Equations

In Figure 3.9 we have omitted mutual inductances, but we will include them in the analysis. If we
assume the machine as a generator, the general equation for the voltage on the terminals is as follows:

! = −øW − + !ù
V
¹
0 0 0 0 0 1( Ψ( ⁄ X
3.55
/( ⎡ ⎤
⎡0 0 0 0 0 ⎤ ⎡ 1} ⎤ ⎢
(
⎡ /} ⎤ Ψ} ⁄ X⎥
⎢ / ⎥ ⎢ }
⎥ ⎢ ⎥
0 0 . 0 0 0 1. Ψ. ⁄ X⎥ /
⎢ . ⎥ = −⎢ ⎥⎢ ⎥ − ⎢ + ú * û Volts
⎢ X⎥
−/
⎢ ä⎥ ⎢ 0 0 0 ä 0 0 ⎥ ⎢ 1ä ⎥ ⎢ Ψä ⁄
⎥
0
3.56
⎢ 0 ⎥ ⎢0 0 0 0 0 ⎥ ⎢1å ⎥ Ψå ⁄ X⎥
⎣ 0 ⎦
å ⎢
⎣0 0 0 0 0 æ ⎦ ⎣1æ ⎦ ⎣ Ψæ ⁄ X⎦
We can do the following assumptions:

( = } = . = , then: òstu = øîï 3.57

Equation 3.56 becomes in matrixial simplified form:

!stu òstu 0 Wstu Vstu⁄ ¹ /

R! S = −R 0 S R
ò ðñ W ðñ S − R S + ú *û
ðñ V ðñ ⁄ ¹ 0
3.58

−/ä 1ä Ψä
where

! ðñü ô 0 ö W ðñü p1å q V ðñ = pΨå q

0 1æ Ψæ
3.59

Equation 3.58 has time variant coefficients in the Ψ⁄ X term, but these coefficients can be eliminated
from the stator by using the Park`s transformation.

¬ 0 !stu !
We have R SR S = R! S, then equation 3.58 becomes:
ÄÅ
0 îï ! ðñ ðñ

¬ Å òstu Å Wstu ¬ Å Vstu⁄ ¹ ¬ Å !ù

= −R SR Å S R
ò ðñ W ðñ S − R S R S+R Sú û
Å îï Å îï V ðñ ⁄ ¹ Å îï Å
3.60

¬ Å òstu Å Wstu ¬ Å òstu Å ¬¤¥ Å ¬ Å Wstu

R
Å îï
SR Å ò ðñ S RW ðñ S = R Å îï S R Å ò ðñ
SR
Å
SR SR
îï Å îï W ðñ
S3.61

43
 Electric Machines – Basic Theory

¬ Å òstu Å ¬¤¥ Å òstu Å

R SR Å ò SR S=R Å ò S
Å îï ðñ Å îï ðñ
3.62

¬ Å Vstu⁄ ¹ ¬. Vstu⁄ ¹
R SR S = ô ö
Å îï V ðñ ⁄ ¹ V ðñ ⁄ ¹
3.63

By definition V ÄÅ = ¬Vstu, then V ÄÅ ⁄ ¹ = ¬ Vstu ⁄ ¹ + ¬⁄ ¹(Vstu ) 3.64

¬ Vstu⁄ ¹ = V ÄÅ ⁄ ¹ − ¬⁄ ¹(¬¤¥ )V ÄÅ ) 3.65

⎡:;03 − :;03 + 0‰23 − :;03 − 0‰23⎤

E å E å

⎢ ⎥
¬ = Á• ⎢ 0123 − 0123 −
E å
:;03 − 0123 + :;03 ⎥
E å

⎢ E ⎥
3.66
E E
⎣ √ √ √ ⎦
⎡−0123 0‰23 + :;03 0‰23 − :;03 ⎤
E å E å

⎢ ⎥
¬⁄ ¹ = |Á ⎢ :;03 − E :;03 + √• 0‰23 − E :;03 − √• 0‰23⎥
•
⎢ E ⎥
3.67
E E
⎣ √ √ √ ⎦
:;03 0123
E
⎡ √ ⎤
⎢ E⎥
¬¤¥ = Á• ⎢− :;03 + 0‰23 − 0123 − :;03 √ ⎥
E √ • E √ •

⎢ E E⎥
3.68

⎣− :;03 − 0‰23 − 0123 + :;03 √ ⎦

å E å

0 −1 0 ΨA |ΨA
¬⁄ ¹(¬¤¥ )V ÄÅ = | p1 0 0q TΨ¼ U = p−|Ψ¼ q
0 0 0 Ψ 0
3.69

Where

|ΨA = |NA 1A + |÷Nä' 1ä + |÷Nå' 1å

|Ψ¼ = |N¼ 1¼ + |÷Næ' 1æ + |÷N½'
3.70
3.71

Finally, Equation 3.60 can be written as:

/A 0 0 |N¼ |÷Næ' |÷N½' 1A

⎡−/ä ⎤ ⎡ ⎤ ⎡1 ⎤
0 0 0 0 0
⎢ 0 ⎥ ⎢ ä ⎥ ⎢ ä⎥
⎢ ⎥ = −⎢ 0 0 å 0 0 0 ⎥ ⎢1å ⎥
⎢ / ⎥ −|N −|÷N −|÷N 0 0 ⎥ ⎢ 1¼ ⎥
¼ ⎢ A ä' å'
⎢ 0 ⎥ ⎢ 0 0 0 0 æ 0 ⎥ ⎢1æ ⎥
⎣ / ⎦ ⎣ 0 0 0 0 0 + 3 ½ ⎦ ⎣1 ⎦
NA ÷Nä' ÷Nå' 0 0 0 1A ⁄ X
⎡ ÷N Nä N•' 0 0 0 ⎤⎡ ⎤
1ä ⁄ X ⎥
⎢ ä' ⎥⎢
⎢÷Nå' N•' Nå 0 0 0 ⎥⎢ 1å ⁄ X⎥
−⎢ 0 0 0 N¼ ÷Næ' ÷N½' ⎥ ⎢ 1¼ ⁄ X ⎥
⎢ ⎥⎢ ⎥
3.72

⎢ 0 0 0 ÷Næ' Næ ?ý ⎥ ⎢ 1æ ⁄ X⎥
⎣ 0 0 0 ÷N½' ?ý N + 3N½ ⎦ ⎣ 1 ⁄ X⎦

Where:

44
 Electric Machines – Basic Theory

1 = •þ : Zero sequence Current.

-

N : Zero sequence inductance in the stator.

Normalizing in pu values, we have the following equation:

/A 0 0 |N¼ |÷Næ' |÷N½' 1A

⎡−/ä ⎤ ⎡ ⎤ ⎡1 ⎤
0 0 0 0 0
⎢ 0 ⎥ ⎢ ä ⎥ ⎢1ä ⎥
⎢ ⎥ ÆÖ = − ⎢ 0 0 å 0 0 0 ⎥ ÆÖ ∙ ⎢ å ⎥ ÆÖ
/ ⎢ 1¼ ⎥
⎢ ¼ ⎥ ⎢−|NA −|÷Nä' −|÷Nå' 0 0 ⎥
⎢ 0 ⎥ ⎢ 0 0 0 0 æ 0 ⎥ ⎢1æ ⎥
⎣ / ⎦ ⎣ 0 0 0 0 0 + 3 ½⎦ ⎣1 ⎦
NA ÷Nä' ÷Nå' 0 0 0 1 ⁄ X
⎡ ÷N ⎤ ⎡ A ⎤
⁄
⎢ ä'
N ä N •' 0 0 0
⎥ ⎢ 1ä X ⎥
⎢÷Nå' N•' Nå 0 0 0 ⎥ ⎢ 1å ⁄ X⎥
−⎢ 0 0 0 N¼ ÷Næ' ÷N½' ⎥ ÆÖ ∙ ⎢ 1¼ ⁄ X ⎥ ÆÖ
⎢ ⎥ ⎢ ⎥
3.73

⎢ 0 0 0 ÷Næ' Næ ?ý ⎥ ⎢ 1æ ⁄ X⎥
⎣ 0 0 0 ÷N½' ?ý N + 3N½ ⎦ ⎣ 1 ⁄ X⎦

Both; inductance and speed values are given in pu values and |‚ = where |• = rotor rated speed.
The complete description of the per-unit values is given in Article number AD-006. www.adneli.com.

For balanced systems, it is assumed that 1 = 0, then:

/A 0 0 |N¼ |÷Næ' 1A
⎡−/ä ⎤ ⎡ ⎤ ⎡1 ⎤
0 0 0 0
⎢ 0 ⎥ ⎢ ä ⎥ ⎢ ä⎥
⎢ ⎥ ÆÖ = − ⎢ 0 0 å 0 0 ⎥ ÆÖ ∙ ⎢1å ⎥
⎢ /¼ ⎥ ⎢−|NA −|÷Nä' −|÷Nå' 0 ⎥ ⎢ 1¼ ⎥
⎣ 0 ⎦ ⎣ 0 0 0 0 æ ⎦ ⎣1æ ⎦
NA ÷Nä' ÷Nå' 0 0 1 ⁄ X
⎡ ⎤ ⎡ A ⎤
÷N
⎢ ä'
Nä N•' 0 0
⎥ ⎢ 1ä ⁄ X⎥
÷N N N 0 0 ⁄
− ⎢ å' •' å ⎥ ÆÖ ∙ ⎢⎢ 1å X⎥ ÆÖ
⎢ 0 0 0 N¼ ÷Næ' ⎥ 1 ⁄ X⎥
3.74
⎢ ¼ ⎥
⎣ 0 0 0 ÷Næ' N æ ⎦ ⎣ 1æ ⁄ X⎦

45
 Electric Machines – Basic Theory

4 Chapter IV-Synchronous Machine.

Parameters Derivation. Part 2

For most stability studies, it is assumed that = 0, then the general voltage d-q-0 equations of the
4.1. Synchronous Machine for Steady State Analysis

synchronous machine are according to the Equation 4.1:

/A 0 0 |N¼ |÷Næ' 1A
⎡−/ä ⎤ ⎡ ⎤ ⎡1 ⎤
0 0 0 0
⎢ 0 ⎥ ⎢ ä ⎥ ⎢ ä⎥
⎢ ⎥ ÆÖ = − ⎢ 0 0 å 0 0 ⎥ ÆÖ ∙ ⎢1å ⎥
⎢ /¼ ⎥ ⎢−|N A −|÷N ä' −|÷N å' 0 ⎥ ⎢ 1¼ ⎥
⎣ 0 ⎦ ⎣ 0 0 0 0 æ ⎦ ⎣1æ ⎦
NA ÷Nä' ÷Nå' 0 0 1 ⁄ X
⎡ ⎡ A ⎤
÷Nä' Nä N•' 0 0 ⎤ ⎢ 1ä ⁄ X⎥
⎢ ⎥
− ⎢÷Nå' N•' Nå 0 0 ÆÖ ∙ ⎢ 1å ⁄ X⎥ ÆÖ
⎥ ⎢ 1 ⁄
⎢ 0 0 0 N ÷N æ' ⎥ X⎥
4.1
¼
⎢ ¼ ⎥
⎣ 0 0 0 ÷Næ' Næ ⎦ ⎣ 1æ ⁄ X⎦

We can obtain from Equation 4.1 the steady-state relations between the d-q-0 voltages and currents, by
setting:

1å = 1æ = 1½ = 0
• All derivatives to zero.
•

/A 0 |N¼ 1A
−/
p äq = − T 0 ä 0 U p1ä q
/¼ −|NA −|÷Nä' 1¼
4.2

The resulting equations are:

/A = − 1A − |N¼ 1¼ = − 1A − ‡¼ 1¼
/¼ = − 1¼ + |NA 1A + |÷Nä' 1ä = − 1¼ + ‡A 1A + √3
4.3
4.4

From Park’s relation !stu = ¬¤¥ ∙ ! ÄÅ . In a balanced system / =0. This provides that

⎡ cos (3) sin (3)

E
/( √ ⎤ /A
⎢ ⎥ /
p/} q = Á• ⎢cos (3 − • ) sin (3 − • ) p ¼q
E
√ ⎥
/. ⎢ E⎥ /
4.5

⎣cos (3 + • ) sin (3 + • ) √ ⎦

In a balanced system it is sufficient to analyze only one phase. In our case, we will analyze the phase a,
then from Equation 4.5:

46
 Electric Machines – Basic Theory

/( = Á ¾/A :;0 3 + /¼ 012 3¿

•
4.6

where 3 is the angle of the d-axis given by 3 = |• + Ã + /2 4.7

Substituting /A , /¼ from Equations 4.3 and 4.4, we obtain:

/( = Á ¾$− 1A − ‡¼ 1¼ % :;0(|• X + à + /2) + $− 1¼ + ‡A 1A + √3 % 012(|• X + à + /2)¿

•
4.8

We can write: 012(|• X + Ã + /2) = :;0(|• X + Ã) then, we have that:

/( = Á• ¾$− 1A − ‡¼ 1¼ % :;0(|• X + à + /2) + $− 1¼ + ‡A 1A + √3 % :;0(|• X + Ã)¿4.9

Equation 4.9 give us instantaneous values, that means that /( magnitude is a peak quantity, then the
RMS values:

•( = ¾$− 1A − ‡¼ 1¼ % :;0(|• X + à + /2) +$− 1¼ + ‡A 1A + √3 % :;0(|• X + Ã)¿ 4.10

E
å

Converting to phasor notation, we have:

•( = ∠(Ã + /2) + ∠(Ã) + ∠(Ã)

$¤ -é ¤eê -ê % ¤ -ê àeé -é
å å
4.11

Combining terms in , , ‡A and ‡¼ yields:

•( = − ú ∠(à + /2) + ∠(Ã)û − ‡¼ ∠(à + /2) + ‡A ∠(Ã) + ∠(Ã)

-é -ê -ê -é
å å å å
4.12

∠(Ã + /2) = h∠Ã, and defining the RMS equivalent d- and q-axis currents reflected to the stator as:

= =
-é -ê
A √• ¼ √•

We have the phasor notation of Equation 4.12 with RMS values

•( = − ¾h A ∠à + ¼ ∠ÿ − h‡¼ ¼ ∠à + ‡A A ∠à + ∠(Ã) 4.13

The quantity h A ∠Ã + ¼ ∠Ã is the phasor of the stator current decomposed into the d- and q-axis, i.e.,

( = h A ∠Ã + ¼ ∠Ã = A + ¼ 4.14

Where h affecting the A term provides the 90º rotation ahead of the q-axis for the d-axis component of
the current.

Thus Equation 4.13 becomes:

•( = − ( − h‡¼ ¼ ∠à + ‡A A ∠à + ∠à 4.15

Solving for ∠Ã, we have:

= ∠à = •( + ( + h‡¼ ¼ ∠à − ‡A A ∠à 4.16

47
 Electric Machines – Basic Theory

¼ ∠Ã = ¼, A
= h A ∠Ã ⇒ = A ∠Ã ⇒ −h = A ∠Ã
Now let’s focus on the last two terms of Equation 4.16. and then we need to determine
E
Recall that A Š A A

Therefore, we can write:

= ∠à = •( + ( + h‡¼ ¼ + h‡A A 4.17

E
q axis
d axis
θ jX q I q
Iq
δ
β Va I a Ra
φ jX d I d
Id Ia Ref

Figure 4.1 Graphic representation of Equation 4.17.

If we have, from the power flow solution, •( and ( , we can compute the first part of Equation 4.17. As
location of the q-axis is not known, A and ¼ is not known as well, but we know that ( = A + ¼ , so
we can find A or ¼ in function of ( .

Equation 4.17 can be converted in Equation 4.18

= ∠à = •( + ( + h‡¼ ¼ +h‡¼ A −h‡¼ A + h‡A A 4.18

JAA+A c }, (.,+A

Regrouping terms:

= ∠à = •( + ( + h‡¼ $ ¼ + A % + h A $‡A − ‡¼ % 4.19

E
jX d I d Ea q axis
d axis j( X d − X q )I d

θ
Iq jX q I q jX q I a
δ
β Va I a Ra
φ
Id Ia Ref

Figure 4.2 Graphic representation of Equation 4.19.

Equation 4.19 can be expressed as follows:

48
 Electric Machines – Basic Theory

= ∠à = •( + ( + h‡¼ $ ¼ + A % + h A $‡A − ‡¼ % = ( + h A $‡A − ‡¼ % 4.20

Ý

where the first part of eq. 4.19 is given by:

( = •( + ( + h‡¼ ( 4.21

with ( = A + ¼.

( is also on the q-axis because it has the same q-axis components as .

Defining as the power factor angle where ( lags •( . Let’s also define as the angle of •( , relative
to the reference. Then ∠ ( = − . Figure 4.2 illustrates this relation.

From the phasor diagram, we can observe that

∠ A = Ã − 90°
| A | = | ( | 012( − + Ã)
4.22
4.23

From the above relations, we can determine A, and finally we can find the value for E.

= ∠à = ( + h A $‡A − ‡¼ % 4.24

Finally, we have:

A = | A |∠Ã − 90
= ¼ ∠Ã
4.25
¼
1A 1¼
4.26

A = ¼ =
√3 √3

In Figure 4.2

( = ¼ + A =$ ¼ + h A %‰ Š = ¼ ∠Ã + h A ∠Ã = ¼ ∠Ã + A ∠Ã + 90 4.27

But ∠ A = Ã − 90°, or A = | A |∠Ã − 90. The implication is that A = −| A | for lagging currents as
below:

A = A ∠Ã + 90 = − A ∠Ã + 90 = A ∠ − 180∠Ã + 90 = A ∠Ã + 90 − 180 = A ∠Ã − 90

On the other hand, we have related a phasor ( getting its rotation from the sinusoidal variation of the
alternating currents with A and ¼ that are direct currents, but they are also rotating phasors whose
variation comes from the rotor motion.

For round rotor machines ‡A = ‡¼ , then Equation 4.19 becomes:

= ∠à = •( + ( + h‡¼ ( ( ) 4.28

The equivalent graphic representation for stead-state analysis is shown in figure 4.3.

49
 Electric Machines – Basic Theory

d axis jX d I d E
q axis
θ
Iq jX q I q jX q I a
δ
Va I a Ra Ref
φ
Id Ia
Figure 4.3 Graphic representation for round rotor machines.

In case of salient pole machines, assuming = 0, we have the Equation 4.17

= ∠à = •( + h‡¼ ¼ + h‡A A 4.29

jX d I d E
q axis
d axis
θ
jX q I q jX q I a
Iq
δ Va Ref

Id Ia
Figure 4.4 Equation 4.17 graphically represented.

Using E as reference with =0 in the above relations and the result is the classical power equations.

= | |∠0° •( = |•( |∠ − à = |•( |‰ ¤Š = |•( | :;0 à − h|•( | 012 à 4.30

The output power can be calculated with the Equation 4.31.

9~ = •( = •( $ +
∗
¼%
∗
, ( A 4.31

From Equations 4.29 and 4.30

= | |∠0° = |•( | :;0 Ã + h‡A =

| |¤|dÝ | .~c
A ⇒⇒ A Šeé
4.32
0 = −h|•( | 012 à + h‡¼ =
|dÝ | c-*
¼ ⇒⇒ ¼ eê
4.33

Equations 4.32 and 4.33 in Equation 4.31 yields:

∗
9~ = (|•( |∠ − Ã) • + €
| |¤|dÝ | .~c |dÝ | c-*
, Šeé eê
4.34

50
 Electric Machines – Basic Theory

∗
9~ = $|•( |(:;0 Ã − h 012 Ã)% • −h €
|dÝ | c-* | |¤|dÝ | .~c
, eê eé
4.35

9~ = $|•( |(:;0 Ã − h 012 Ã)% • +h €

|dÝ | c-* | |¤|dÝ | .~c
, eê eé
4.36

The real part »~ , will be:

»~ = |•( | R + S
|dÝ | .~c c-* | | c-* ¤|dÝ | .~c c-*
, eê eé
4.37

Rearranging the order of the term’s yields:

»~ = + |•( | :;0 Ã 012 Ã Re − e S

| ||dÝ | c-* E E
, eé
4.38
ê é

012 2Ã = 2:;0 Ã 012 Ã 4.39

»~ = + Re − e S 012 2 Ã
| ||dÝ | c-* |dÝ |I E E
, eé
4.40
ê é

Similarly, we may derive from ~ ,, as:

= − |•( | R + S
| ||dÝ | .~c .~cI c-*I
~ , eé eé eê
4.41

For round-rotor (‡A = ‡¼ ), Equations 4.40 and 4.41 become two familiar equations frequently used to
evaluate the generator stability:

»~ = 012(Ã)
| ||dÝ |
, eê
4.42

= :;0(Ã) −
| ||dÝ | |dÝ |I
~ , eê eê
4.43

4.2. Effect of saliency in stability

Figure 4.5 Equations 4.40 and 4.42 represented in a plane.

51
 Electric Machines – Basic Theory

From Figure 4.5, we can conclude:

• Second harmonic component shift »'(e and this occurs for Ã<90 degrees in salient pole

»'(e is greater for a salient-pole machine relative to a round-rotor machine of similar

machines.
•
characteristics. At equal level, a salient-pole machine will typically have more decelerating
energy available than a corresponding round-rotor machine, consequently, saliency tends to
improve stability.

52
 Electric Machines – Basic Theory

5
5.1.
Chapter V-SYNCHRONOUS
MACHINE. PARAMETERS
DERIVATION. PART 3

The correspondence of Stator EFM to Rotor quantities

Using the appropriate base values, we can express the voltage equations in per unit values, according
to the initial circuit from Figure 5.1

I a
rF
vF IF LF
ra
L aa
rb Ib
rD n' Lbb
LD
vD = 0 ID rG Lcc

LG rc Ic
rQ
vQ = 0 LQ vn
IQ
Ig
va v b v c

Figure 5.1 Equivalent circuit of a Synchronous machine.

With the following set of equations:

! = −øW − + !ù
V
¹
0 0 0 0 0 1( Ψ( ⁄ X
5.1
/( ⎡ ⎤
⎡0 0 ⎤ ⎡ 1} ⎤ ⎢
(
⎡ /} ⎤ 0 0 0 Ψ} ⁄ X⎥
⎢ ⎥
0 ⎢ 1. ⎥ ⎢
}
⎢ / ⎥ 0 0 0 0 Ψ. ⁄ X⎥ /
⎢ . ⎥ = −⎢ ⎥⎢ ⎥
0 ⎥ ⎢ 1ä ⎥ − ⎢ + ú *û
.
⎢−/ä ⎥ ⎢0 0 0 0 Ψä ⁄ X⎥ 0
0 ⎥ ⎢1å ⎥ ⎢ ⎥
ä
volts 5.2
⎢ 0 ⎥ ⎢0 0 0 0 Ψå ⁄ X⎥
⎣ 0 ⎦
å ⎢
⎣0 0 0 0 0 æ ⎦ ⎣1æ ⎦ ⎣ Ψæ ⁄ X⎦

Doing the corresponding Park transformations, and the per-unit conversions the circuit of Figure 5.1
can be expressed according to the circuit of Figure 5.2

53
 Electric Machines – Basic Theory

rF
vF IF LF

rD

vD = 0 ID LD

rQ
vQ = 0 LQ
IQ

Figure 5.2 Equivalent circuit of the synchronous machine from Figure 5.1.

With the corresponding equations

/A 0 0 |N¼ |÷Næ' 1A
⎡−/ä ⎤ ⎡ ⎤ ⎡1 ⎤
0 0 0 0
⎢ 0 ⎥ ⎢ ä ⎥ ⎢ ä⎥
⎢ ⎥ ÆÖ = − ⎢ 0 0 å 0 0 ⎥ ÆÖ ∙ ⎢1å ⎥
⎢ /¼ ⎥ ⎢−|N A −|÷N ä' −|÷N å' 0 ⎥ ⎢ 1¼ ⎥
⎣ 0 ⎦ ⎣ 0 0 0 0 æ ⎦ ⎣1æ ⎦
NA ÷Nä' ÷Nå' 0 0 1 ⁄ X
⎡ ⎤ ⎡ A ⎤
÷N
⎢ ä'
Nä N•' 0 0
⎥ ⎢ 1ä ⁄ X⎥
÷N N N 0 0 ⁄
− ⎢ å' •' å ⎥ ÆÖ ∙ ⎢⎢ 1å X⎥ ÆÖ
⎢ 0 0 0 N¼ ÷Næ' ⎥ 1 ⁄ X⎥
5.3
⎢ ¼ ⎥
⎣ 0 0 0 ÷Næ' Næ ⎦ ⎣ 1æ ⁄ X⎦

Where we have removed the /* component, as our interest is a balanced system for steady-state and
stability studies.

Equations 5.3 can be expressed in their matrixial form:

! = −(ò + | )W − Q4 pu 5.4

Where

ò = Diagonal matrix of resistances.

= Matrix of speed voltage inductance coefficients.
Q = Symmetric matrix of constant inductances.
4 =
W
¹

Assuming that inverse of Q exists, we may write in function of the currents.

4 = −Q¤¥ (ò + | )W − Q¤¥ ∙ ! pu 5.5

Equation 5.5 has the state-space form. Despite that it does not express the entire system behavior, as we
need to add the torque equations, but it can be schematically described by Figure 5.2.

54
 Electric Machines – Basic Theory

5.2. The Swing Equation

Consider a synchronous generator developing an electromagnetic torque + and running at the

synchronous speed |c' . If ' is the driving mechanical torque, then under steady-state operation
with losses neglect a steady-state:

' = + 5.6

A change from steady state due to a disturbance result in an accelerating torque ( if ' > + or
decelerating if ' < +

( = ' − + 5.7

The rotor angular velocity can be obtained as follows:

|' = = |c' +
A‹O A O
A, A,
5.8

where 3' is the angular displacement of the rotor with respect to the stationary reference axis on the

with constant angular velocity |c'

stator. The Angular reference is chosen relative to a synchronously rotating reference frame moving

And the rotor acceleration

=
A I ‹O AI O
A, I A, I
5.9

If is the combined moment of inertia of the prime mover + generator, neglecting frictional and
damping forces, from laws of rotation we have:

=
A I ‹O
( A, I
5.10

3' = |c' X + Ã' 5.11

The general equation then becomes:

− =0
AI O
A, I ( 5.12

If we multiply this equation by |'

|' = |' − |'

AI O
A, I ' + 5.13

Since power » = |

n' = »' − »+
AI O
A, I
5.14

|' = ? =Inertia constant. M is related to the kinetic energy of the rotating masses,

= |' = ?|'
E E
5.15
?= 5.16
O

55
 Electric Machines – Basic Theory

Since |' does not change by a large amount before stability is lost, M is evaluated at the synchronous
speed and it is considered to remain constant;

?= 5.17
O

The swing equation becomes:

? = »' − »+
AI O
A, I
5.18

And in terms of electrical power angle,

? A, I = »' − »+
AI
‚
5.19

where:

Æ=
à = Ã'
‚
the number of poles of a synchronous generator

The swing equation is usually expressed in per-unit values referred to the base power 9D

Defining a new term known as the H (constant per unit inertia constant)

= =
-*+,-. +*+ )i -* ¦ (, (,+A c‚++A
'(.ℎ-*+ (,-*) -* ¦dJ §C
5.20

and |c' = (2/Æ)|c in terms of electrical angular velocity, the expression becomes:

= »'(‚ − »+(‚
AI
A, I ) ) 5.21

|c = 2 ` 5.22

H units are in seconds, Ã in radians and ` is the nominal frequency (60 or 50 Hz).

5.3. Torque and Power Referred to the Rotor

We can write in pu values the following general equation

»~ , = /( 1( + /} 1} + /. 1. = !¹stu Wstu = (¬¤¥ ! ÄÅ ) (¬ W ÄÅ )

¹ ¤¥
= !¹ ÄÅ (¬¤¥ )¹ ¬¤¥ W ÄÅ 5.23

From the general equation

/A 0 0 |N¼ |÷Næ' 1A
⎡−/ä ⎤ ⎡ ⎤ ⎡1 ⎤
0 0 0 0
⎢ 0 ⎥ ⎢ ä ⎥ ⎢ ä⎥
⎢ ⎥ ÆÖ = − ⎢ 0 0 å 0 0 ⎥ ÆÖ ∙ ⎢1å ⎥
⎢ /¼ ⎥ ⎢−|NA −|÷Nä' −|÷Nå' 0 ⎥ ⎢ 1¼ ⎥
⎣ 0 ⎦ ⎣ 0 0 0 0 æ ⎦ ⎣1æ ⎦
NA ÷Nä' ÷Nå' 0 0 1 ⁄ X
⎡ ⎤ ⎡ A ⎤
÷N
⎢ ä'
Nä N•' 0 0
⎥ ⎢ 1ä ⁄ X⎥
÷N N N 0 0 ⁄
− ⎢ å' •' å ⎥ ÆÖ ∙ ⎢⎢ 1å X⎥ ÆÖ
⎢ 0 0 0 N¼ ÷Næ' ⎥ 1 ⁄ X⎥
5.24
⎢ ¼ ⎥
⎣ 0 0 0 ÷Næ' N æ ⎦ ⎣ 1æ ⁄ X⎦

56
 Electric Machines – Basic Theory

/A and /¼ : Direct and quadrature components of the terminal voltage, respectively.

/ä and 1ä : Field voltage and current, respectively.
1A and 1¼ : Direct and quadrature components of the terminal current, respectively.
| and ÷Nä' : Machine speed and mutual inductance between the windings, respectively.

To calculate the power output, we can reduce Equation 5.24, to obtain the /A and /A terms (assume
no energy losses).

»~ , = !¹ ÄÅ ¬¬¤¥ W ÄÅ = !¹ ÄÅ W ÄÅ = /A 1A + /¼ 1¼ + / 1 5.25

Assuming a balanced network: / = 0, 1 = 0, then

/A −Ψ¼ ΨA¼
ú/ û = | R S -ô ö
¼ ΨA Ψäåæ
5.26

»~ , = /A 1A + /¼ 1¼ = (1A ΨA + 1¼ Ψ¼ ) +(1¼ ΨA − 1A Ψ¼ )| 5.27

»~ , = /A 1A + /¼ 1¼ 5.28

According to Concordia17, the torque is obtained from the rate of change of stator magnetic field
energy

= = = ¾(1¼ ΨA − 1A Ψ¼ )|¿ = 1¼ ΨA − 1A Ψ¼
$!"#Hé
+
!"#Hé
‹
5.29

1A
⎡1 ⎤
ΨA NA ÷Nä' ÷Nå' 0 0 ⎢ ä⎥
RΨ S = R 0 1
0 0 N¼ ÷Næ' S ⎢ 1å ⎥
⎢ ¼⎥
¼
5.30

⎣1æ ⎦

Equation 5.30 in 5.29:

1A
⎡1 ⎤
⎢ ä⎥
= jNA 1 ÷Nä' 1 ÷Nå' 1 −N ¼ 1 −÷Næ' 1 l ⎢1å ⎥
⎢ 1¼ ⎥
‰ 5.31

⎣1æ ⎦

We can express the total acceleration torque adding the damping torque A = %n (D is a damping
constant) as:

= − − = − +−
&#'
( ' 3 ' 5.32
For generator action, the rotor field leads the armature field by the forward torque or a prime mover;
for motor action, the rotor field lags behind the armature field due to the retarding torque of shaft
load (mechanical)18.
The relation between à and | is given by:

17
Concordia, C. “Synchronous Machines,” Wiley, New York, 1951.
18
Kundur, 1994

57
 Electric Machines – Basic Theory

à = | − 1. 5.33

Combining these equations into one system, the simplified synchronous machine model is given in the
state space form by:

⎡ A⎤ ⎡ −Q¤¥ (ò + | ) Å
⎤ ⎡ A⎤ ⎡ ⎤
⎢ ä⎥ ⎢ ⎥ ⎢ ä ⎥ ⎢−Q¤¥ •⎥
⎢ ⎥=⎢ ⎥ ⎢ ¼⎥ + ⎢ ⎥
⎢ ¼ ⎥ ⎢ − ìé ê −
ì(O ê ìê é
−
å
0 ⎥⎥ ⎢ | ⎥ ⎢ ' ⎥
5.34
⎢| ⎥ ⎢ ƒ C
⎣ ⎦ ⎣ −1 ⎦
0⎦ Ã
ƒ ƒ
⎣Ã⎦ ⎣ 0 0 0 1
C C C

58
 Electric Machines – Basic Theory

6
6.1.
Chapter VI-SYNCHRONOUS
MACHINE. PER UNIT VALUES

Per-unit Values19

In the power systems analysis, a per-unit system is the expression of system quantities as fractions or
percentages of a defined base unit quantity. This process presents the advantage to simplify the
analysis in an interconnected power system having different voltage levels, as with percentages is
possible to refer all impedances to a single voltage level, understanding that power is invariant and
independent from any voltage.

A per-unit system provides units for power, voltage, current, impedance, admittance, and speed
mainly, but it can be extended to other parameters as magnetic flux. All quantities are specified as
multiples of selected base values.

The per-unit value of any quantity is defined as follows:

Öm2X1Xˆ 12 Ɖ − Ö21X =
(., (F ¼ (*,-,i
}(c+ )(F + ~* ¼ (*,-,i
6.1

The system power and voltage equations must be invariant, whether they are in per unit or in MKS
units and all mutual inductances must be capable of representation in tee circuits.

6.1.1. Application in transformers

A simple transformer consists of two electrical conductors called the primary winding and the
secondary winding. These two windings can be considered as a pair of mutually coupled coils. Energy
is coupled between the windings by the time-varying magnetic flux that passes through (links) both
primary and secondary windings.

Figure 6.1 Basic Transformer1.

19
Hadi Saadat: Power System Analysis. McGraw-Hill International Editions, 1999.

59
 Electric Machines – Basic Theory

6.1.1.1. Analysis of Magnetically coupled circuits.

From Figure 6.1, the voltage equations can be expressed as:

/E = E 1E +
ALG
A,
6.2
/ = 1 + I
AL
A,
6.3

The inductance in matrixial form can be written as:

NEF + NE' ∙ NE'

PI
N NE
Q = R EE S = T PG U
PG
N E N ∙N N F+N
6.4
PI ' '

And the flux linkages:

ΨE = NEF ∙ 1E + NE' (1E + I ∙ 1 )

P
P
6.5
G

Ψ =N F∙1 +N ' (1 + PG ∙ 1E )
P
6.6
I

In matrixial form:

V=Q∙W 6.7

The obtained equivalent circuit is shown in Figure 1.5

Figure 1.5 Equivalent circuit for a single-phase transformer

From Equations 1.22, 1.23 and 1.31, the voltages on the primary and secondary have the following
Equations:

/E = E 1E + NEE ∙ + NE ∙ A,I
A-G A-
A,
6.8
/ = 1 +N ∙ I+N E∙ G
A- A-
A, A,
6.9

As the voltage equations are expressed in three dimensions (v, i, t), choosing three base quantities
involving three dimensions allows obtaining the remaining ones combining the quantities selected.

•D = Usually, the rated line to neutral voltage = •• in volt RMS.

9D = Stator rated VA/phase, VA RMS.
|D = Generator rated electrical speed = |• in elec. rad/s.

60
 Electric Machines – Basic Theory

= dC in Amp. RMS
§
Current = D 6.10
C

Impedance = gD = = |D ND in Ohms.
dC
6.11
C

ND =
dC
( C∙ C)
= Henry. 6.12
XD =
E
in s. 6.13
_D = •D
C
D Wb turn. 6.14

Selecting three base values for each winding:

•E+ , E+ , XE+
• +, +, X
for winding 1 6.15
+ for winding 2 6.16

Applying base values to Equations 6.10 to 6.14 we obtain the per-unit values for parameters in
Equations 6.8 and 6.9

= ,G , = , I , NEE‚ = ì GG , N = ì II
ì ì
E‚ ‚ ‚ 6.17
GC IC GC IC

Examining the mutual coefficients, preserving reciprocity:

NE D /•ED X D =N E ED /• D XED 6.18

Having NE = N E Equation 6.18 becomes:

D /•ED X D = ED /• D XED 6.19

= , , =
dGC GC dIC IC §GC §IC
,GC ,IC ,IC
6.20
GC

As XED = X D = XD , Xℎ‰2 9ED = 9 D = 9D 6.21

For mutual inductances analysis, using Equation 6.4, Equation 6.7 can be written in the following
form:

ΨE N NE 1E N 0 1E N NE 1
R S = R EE S R S = R EF S R S + R E' S R ES
Ψ N E N 1 0 N F 1 N E N ' 1
6.22

Injecting 1E = ED and 1 = 0, we have the following flux linkages:

_E' = NE' ED and _ ' =N E D 6.23

_E'‚ = _E' / _ED = NE' /NED and _ '‚ =_ '/ _ D =N E ED /N D D 6.24

Injecting 1E = 0 and 1 = D, we have in pu values:

_ '‚ =_ '/ _ D =N ' /N D and _E'‚ = _E' / _ED = NE D /NED ED 6.25

Since 9ED = 9 D = 9D, ED NED = DN D 6.26

Then, Equation 6.25 can be written as:

N ' /N D =N '‚ = NE ED /N D D 6.27

61
 Electric Machines – Basic Theory

From Equations 6.24 and 6.27, we conclude:

N '‚ = NE'‚ = N'‚ 6.28

The voltage equations and the tee circuit equivalent are shown below:

/E‚ = ú E 1E + NEF ∙ + N' ∙ Z G + I [û ÆÖ

A-G A- A-
A, A, A,
6.29
/ =ú 1 +N ∙ + N ∙ Z + [û ÆÖ
A-G A-G A-I
‚ F A, ' A, A,
6.30

Figure 1.6 Transformer equivalent tee circuit in per-unit values.

6.2. Pu values of the d-q voltage equations in Synchronous machines:

From Equation 3.5420, we can write:

/A = − 1A − n_¼ − _A ⁄ X 6.31

/¼ = − 1¼ + n_A − _¼ ⁄ X 6.32

−/ä = − ä 1ä − _ä ⁄ X 6.33

0 = −/å = − å 1å − _å ⁄ X 6.34

0 = −/æ = − æ 1æ − _æ ⁄ X 6.35

Where;

_A = NA 1A + ÷Nä' 1ä + ÷Nå' 1å 6.36

_¼ = N¼ 1¼ + ÷Næ' 1æ 6.37

_ä = ÷Nä' 1A + Nä 1ä + N•' 1å 6.38

_å = ÷Nå' 1A + N•' 1ä + Nå 1å 6.39

_æ = ÷Næ' 1¼ + Næ 1æ 6.40

20
AD-003. “Synchronous Machine Parameters’ Derivation Part 1.” Adneli Consultant, 2021.

62
 Electric Machines – Basic Theory

To obtain the per-unit (pu) representation of the voltage equations, we need to consider the following
base values for the Stator and Rotor:

6.2.1. Per unit values for Stator in a synchronous generator

Let´s define the RMS quantities e •∠ volt., and ∠. Amp. The three-phase apparent power will be
9• = 3• ∙ , and the active power »• ‚ = 3• ∙ ∙ cos ( − .). The per unit apparent power is
given by:

9• =d = 3 ∙ •‚ ∙
•d∙
‚ ‚
C∙ C
6.41

To obtain the d and q axis quantities, we can define the phase voltages as:

/( = √2••¦§ ∙ sin (3 + )
/} = √2••¦§ ∙ sin(3 + − )
6.42
•
6.43
/. = √2••¦§ ∙ sin (3 + + )
•
6.44

Applying the Park transformation:

! ÄÅ = ¬ ∙ !stu 6.45

√3••¦§ ∙ sin ( )
! ÄÅü T√3• U
•¦§ ∙ cos ( )
0
6.46

And for currents:

√3 •¦§ ∙ sin (.)

W ÄÅü T√3 U
•¦§ ∙ cos (.)
0
6.47

Where the per-unit values for voltages and currents are:

/A‚ = é = √3•‚ ∙ sin ( )

)
d
6.48
C
/¼‚ = = √3•‚ ∙ cos ( )
)ê
d
6.49
C

1A‚ = = √3 ∙ sin (.)

-é
‚ 6.50
C

1¼‚ = = √3 ∙ cos (.)

-ê
‚ 6.51
C

/A‚ + /¼‚ = 3 ∙ V‚
1A‚ + 1¼‚ = 3 ∙ I‚
6.52
6.53

»• ‚ = /A‚ ∙ 1A‚ + /¼‚ ∙ 1¼‚

»• = 3 ∙ •‚ ∙ ‚ ∙ cos( − .) = 3 ∙ •‚ ∙ ∙ cos( )
6.54
‚ ‚ 6.55

63
 Electric Machines – Basic Theory

E
q axis
d axis
θ jX q I q
Iq
δ
β Va I a Ra
φ jX d I d
Id Ia Ref

Figure 6.1 Parameters for a generator in steady-state.

/A 0 0 |N¼ |÷Næ' 1A
⎡−/ä ⎤ ⎡ ⎤ ⎡1 ⎤
0 0 0 0
⎢ 0 ⎥ ⎢ ä ⎥ ⎢ ä⎥
⎢ ⎥ = −⎢ 0 0 å 0 0 ⎥ ∙ ⎢1å ⎥
⎢ /¼ ⎥ ⎢−|N A −|÷N ä' −|÷N å' 0 ⎥ ⎢ 1¼ ⎥
⎣ 0 ⎦ ⎣ 0 0 0 0 æ ⎦ ⎣1æ ⎦
NA ÷Nä' ÷Nå' 0 0 1 ⁄ X
⎡ ÷N ⎡ A ⎤
Nä N•' 0 0 ⎤ ⎢ 1ä ⁄ X ⎥
⎢ ä' ⎥
− ⎢÷Nå' N•' Nå 0 0 ∙ ⎢ 1å ⁄ X⎥
⎥
⎢ 0 0 0 N¼ ÷Næ' ⎥ ⎢ 1¼ ⁄ X ⎥
6.56
⎢ ⎥
⎣ 0 0 0 ÷Næ' Næ ⎦ ⎣ 1æ ⁄ X⎦

Dividing Equations 6.8 to 6.12 by the base value of impedance gD

=− 1 − − _A ⁄ X , =− − −
)é \ê E )é é \ê E A \é
,C A dC C ìC C C A, ìC C
6.57
,C ,C ,C ,C C

/A‚ = (− 1A − |_¼ − _A ⁄ X) p.u.

E
6.58
C

/¼‚ = (− 1¼ + |_A − _¼ ⁄ X) p.u.

E
6.59
C

6.2.2. Per unit values for the rotor in synchronous machines

To choose the rotor circuit base quantities we will do the following assumptions.

• Reciprocity between per unit mutual inductances between different windings.

• Per unit mutual flux linkages between stator and rotor circuits in each axis (d-q) are equal.
• All circuits have the same power (VA) base.

Rotor base quantities must meet the conditions of equal base power (9D ), equal base current ( eD ),,
equal ??&D , or (_eD ).
and

This means a power stator-rotor power balance:

•cD cD =•D D 6.60

64
 Electric Machines – Basic Theory

Subscript r is used to indicate any rotor circuit (field or amortisseur) and s any stator circuit.

Based on the above considerations we can write:

cD Nc' = D ÷Nä' , being ÷ = Œ3⁄2 6.61

The base current for rotor circuits related to the base current for the stator circuits will be:

D = (Nc' /÷Nä' ) cD 6.62

Making ÷ä = ÷Nä' /Nc' we can calculate the following relationships as base values for rotor circuits

D = (1/֊ ) cD 6.63

• D = (÷ä )•cD 6.64

f =Z [ = ֊ fcD
d^C
D 6.65
^C

where f D : base rotor resistance and fcD : base stator resistance

N = (• D XD )/ = ÷ä Z [ Z [ = ÷ä NcD : inductance base of the rotor circuit.

dC E
D D 6.66
C ^

Nc D = (•cD XD )/ D = ÷ä NcD : mutual inductance base. 6.67

Thus, we can conclude that the per-unit d axis mutual inductance of any rotor circuit is the same as the
magnetizing inductance of the stator.

÷Nä'‚ = ÷Nå'‚ = N•'‚ = Nc'‚ 6.68

Then, the per-unit equations for rotor voltages are shown below

Z−/ä = − ä 1ä − _ä ⁄ X[ pu
E
6.69
C

Z0 = −/å = − å 1å − _å ⁄ X[ pu
E
6.70
C

Z0 = −/æ = − æ 1æ − _æ ⁄ X[ pu
E
6.71
C

6.3. Base change21

The impedance of individual generators and transformers, as supplied by the manufacturer, are
generally in terms of percent or per-unit quantities based on their ratings. Per analysis, all impedances
must be expressed in per unit on a common system base. For that, the impedance of a circuit element
expressed in terms of a particular base MVA (or KVA) must be expressed in terms of a different base
MVA (or KVA).

For example, suppose a 500 MVA transformer having 10% (or 0.1 pu) reactance and a 1000 MVA
transformer having 16% reactance (both expressed on their rated MVA as a base) are used in the same
system. For calculations including both of those transformers, they must be converted to a common
MVA base.

21
General Electric. “Industrial Power Systems Data Book”, 1957.

65
 Electric Machines – Basic Theory

=
$+ ¤ *-, ~0'c ~* *+1 }(c+ ¦dJ P+1 }(c+ ¦dJ
$+ ¤ *-, ~0'c ~* ~FA }(c+ ¦dJ 2FA }(c+ ¦dJ
6.72

Likewise, a machine rated at one voltage may be used in a circuit at a different voltage. If this latter
voltage is selected as the base voltage, the per-unit impedance of the machine must be changed to the
new base voltage.

= (*+1 }(c+
$+ ¤ *-, ~0'c ~* *+1 }(c+ d (~FA }(c+ d)I
$+ ¤ *-, ~0'c ~* ~FA }(c+ d d)I
6.73

In consequence:
»‰ − Ö21X ;ℎÈ0 ;2 2‰n om0‰ ÷• = (»‰ − Ö21X ;ℎÈ0 ;2 ; om0‰ ÷•) (*+1
(~FA }(c+ d)I
}(c+ d)I
6.74

Example:
Generator Transformer Load
1000 KVA 1000 KVA 1000 KVA

13.8 kV 2.3 kV
13.2/2.4 kV = 5.5
Base kV Base MVA Base Amps Base Ohms
High Voltage 13.8 1.0 41.6 190
Low Voltage 2.5 1.0 230 6.25

Ratio (HV/LV) 5.5 1.0 1/5.5 (5.5)²

66
 Electric Machines – Basic Theory

7
7.1.
Chapter VII-ASYNCHRONOUS
MACHINE DERIVATION. PART 1

Asynchronous Machine Derivation

As in the case of Synchronous Generators are used to convert mechanical power to AC electric power, or
electrical power into mechanical power The asynchronous machine is a type of AC motor that runs at a speed
different than the synchronous speed. When the speed is lower than the synchronous speed the asynchronous
machine operates as an electric motor converts electrical power from the stator circuits into mechanical power,
and when the speed is higher than the synchronous speed operates as an electric generator converts mechanical
power into electrical one through the stator circuits.
In Figure 7.1 we have depicted the electrical representation of an asynchronous machine.

Figure 7.1 Schematic diagram of a three-phase induction machine stator circuit (left) and rotor one (right)22.
Initial Assumptions:

• For simplicity, symmetrical two-pole, three-phase windings

• Slotting effects are neglected: No harmonic influence is assumed.
• Linear and lumped inductances. No saturation effect.
• The flux distribution is radial in the air gap and density is uniform.
• No iron losses.

Voltage Equations23-24:

/(c = (c 1(c +
ALÝ
A,
7.1
/}c = }c 1}c +
AL3
A,
7.2
/.c = .c 1.c +
AL4
A,
7.3
/( = 1( +
ALÝ^
( A,
7.4
/} = 1} +
AL3^
} A,
7.5

22
Kundur (1994).
23
AD-003, “Synchronous Machine Parameters’ Derivation Part 1.” Adneli Consultant, 2021. www.adneli.com.
24
Paul C. Krause, Oleg Wasynczuk, Scott D. Sudhoff. “Analysis of Electric Machinery and Drive Systems.” IEEE Press Power
Engineering Series.

67
 Electric Machines – Basic Theory

/. = . 1. + 4^
AL
A,
7.6
The resistance of the stator and rotor are as follows:

(c = }c = .c = c
= = =
7.7
( } . 7.8

Then, Equations 7.1 to 7.8 in matrix form

!stu5 = ø5 Wstu5 + Vstu5

!stuø = ø5 Wstuø + Vstuø
7.9
7.10

Vstu5 Q5 Q5ø 6stu5

R S=R SR S
Vstuø (Q5ø )¹ Qø 6stuø
7.11

For symmetrical stator windings, the self-inductances N(c(c , N}c}c and N.c.c are equal and will be denoted as
Ncc , where Ncc = NFc + N'c , being

NFc = Leakage inductance

N'c = Magnetizing inductance

The self-inductances of the rotor winding can be expressed similarly N = NF + N' .

(phases are shifted by ∅ = 2 /3). The leakage inductance is essentially constant, independent of ∅ ; however,
The self-inductances are all constant; however, a mutual coupling exists between stator phases and rotor phases

the magnetizing inductance is a periodic function of ∅. That is, N' = `( ∅) = N' (∅).

N' (∅) =
PI
ℜO (∅)
7.12

If we assume that the mutual inductance varies in proportion to cos(∅), the mutual inductance between phases
will be:

Necic = N'c cos (∅c )∅ = − N'c

E
ü /• 7.13

Where x and y are combinations of phases a, b and c.

In case of the rotor, we have a similar value, but considering a different < value in the N' (∅) calculations.

Ne = N' cos (∅ )∅^ ü = − N'

E
i /• 7.14

The self-inductances can be expressed in matrixial form.

⎡ Ncc − N'c − N'c ⎤

E E

⎢ E ⎥
Q5 = ⎢− N'c Ncc − N'c ⎥
E

⎢ E ⎥
7.15

⎣− N'c − N'c Ncc ⎦

E

⎡ N − N' − N' ⎤
E E

⎢ E ⎥
Qø = ⎢− N' N − N' ⎥
E

⎢ E ⎥
7.16

⎣− N' − N' N
E
⎦

the angle 3, then:

The value of the mutual inductance between the stator and rotor is depending on the rotor position, defining by

68
 Electric Machines – Basic Theory

⎡ cos (3 ) cos (3 + )
•
cos (3 −
•
)⎤
⎢ ⎥
Q5ø = Nc ⎢cos (3 − • ) cos (3 ) cos (3 + )⎥
⎢ ⎥
•
7.17

⎣cos (3 + ) cos (3 − ) • •
cos (3 ) ⎦

Where

Nc (3 ) =
P P^
ℜO (‹^ )
7.18

All rotor variables may be referred to the stator winding by the following turn ratios:

WYstuø = W
P^
P stuø
7.19
!Ystuø= !stuø
P^
P
7.20
Vstuø = ^ Vstuø
Y P
P
7.21

8øY = Z [ 8ø
P
P^
7.22

8øY = Z [ 8ø
P
P^
7.23

9Y ø = Z [ 9 ø
P
P^
7.24
= 95ø = Q5ø
P Qõ5
9Y5ø
P^ ì ^
7.25

⎡N′F + N'c − N'c − N'c ⎤

And
E E

⎢ ⎥
Q′ø = ⎢ − N'c N′F + N'c − N'c ⎥
E E

⎢ ⎥
7.26

⎣ − N'c − N'c N′F + N'c ⎦

E E

The flux linkage equations become:

Vstu5 Q Q′5ø 6stu5

R S=R 5 ¹ SR S
V′stuø (Q′5ø ) Q′ø 6′stuø
7.27

Doing the transformations with the classical Park Transformation:

⎡cos (3) cos (3 −

•
) cos (3 + )⎤
•
⎢ ⎥
¬5 = ⎢ 012(3) sin (3 − ) sin (3 + ) ⎥
⎢ E ⎥
• • •
7.28
E E
⎣ ⎦

With the inverse:

cos (3) 012(3) 1

¬¤¥ cos (3 − ) sin (3 − ) 1¸
5 = · • •
cos (3 + ) sin (3 + ) 1
7.29

• •

In the rotor, the transformation to apply will be:

69
 Electric Machines – Basic Theory

Figure 7.2 Phasors rotation referenced to phase a.

⎡cos ( ) cos ( −
•
) cos ( + )⎤
•
⎢ ⎥
¬ø = ⎢ 012( ) sin ( − ) sin ( + ) ⎥
⎢ E ⎥
• • •
7.30
E E
⎣ ⎦

Where: =3−3

cos ( ) 012( ) 1
And

¬¤¥ cos ( − ) sin ( − ) 1¸

ø = · • •
cos ( + ) sin ( + ) 1
7.31

• •

Ncc 0 0
Q5 = p 0 Ncc 0 q = Ncc v
0 0 Ncc
7.32

N 0 0
Qø = p 0 N 0 q=N v
0 0 N
7.33

Ncc = NFc + N'c 7.34

N′ = N′F + N'c 7.35

V Ä5 ¬5 Q5 ¬¤¥ ¬5 Q′5ø ¬¤¥

ø 6 Ä5
ô ö=p ¤¥ q ô6′ ö
5
V′ Äø ¬ø (Q′5ø )¹ ¬¤¥ ¬ø Q′ø ¬ø
7.36
5 Äø

Defining

|c = 3 (synchronous velocity) and | = 3 (rotor angular velocity) and the original voltage equations:
A A
A, A,

!stu5 = ø5 Wstu5 + Vstu5

!stuø = ø5 Wstuø + Vstuø
7.37
7.38

The transformed d-q voltage equations yield:

/Ac = c 1Ac − |c Ψ¼c + ΨAc 7.39

70
 Electric Machines – Basic Theory

/¼c = c 1¼c + |c ΨAc + Ψ¼c

/′A = ′ 1′A − (|c − | )Ψ′¼ + Ψ′A
7.40

/′¼ = ′ 1′¼ + (|c − | )Ψ′A + Ψ′¼

7.41
7.42

Transforming the flux linkages3_25-26:

ΨAc = NFc 1Ac + N' (1Ac + 1′A )

Ψ¼c = NFc 1¼c + N' $1¼c + 1′¼ %
7.43

Ψ′A = N′F 1′A + N' (1Ac + 1′A )

7.44

Ψ′¼ = N′F 1′¼ + N' $1¼c + 1′¼ %

7.45
7.46

Being:

N' = N'c
•
7.47

The equivalent circuits in Figure 7.3 show the voltage and flux linkage equations.

Figure 7.3 d-q equivalent circuits of asynchronous machines

25

Some authors as Kundur use a different nomenclature. E.g., Ncc = NFc + N' , N' = ? , 1A instead of 1′A , and N = N′F + N' , where
A balance condition has been assumed.

Nc = Ncc − N' is the stator leakage reactance and N' the magnetizing reactance, being the equations:
26

ΨAc = Ncc 1Ac + N' 1A

Ψ¼c = Ncc 1¼c + N' 1¼
ΨA = N 1A + N' 1Ac
Ψ¼ = N 1¼ + N' 1¼c

71
 Electric Machines – Basic Theory

8
8.1.
Chapter VIII-ASYNCHRONOUS
MACHINE DERIVATION. PART 2

Steady-State Analysis

1(c = √2 ∙ cos (|c X)

Having the currents in the stator
•¦§
1}c = √2 •¦§ ∙ cos (|c X − )
8.1
•
8.2
1.c = √2 •¦§ ∙ cos (|c X + )
•
8.3

2 2
Applying the d-q transformation in the stator, we have:
⎡cos (3) cos (3 − ) cos (3 + )⎤
1Ac ⎢ 3 3 ⎥
2 2 2 1(c
1
p ¼c q = ⎢ 012(3) sin (3 − ) sin (3 + ) ⎥ p1}c q
3⎢ 3 3 ⎥ 1
1 c
⎢ 1 1 1 .c
⎥
⎣ √2 √2 √2 ⎦
1Ac = √2 •¦§ ∙ cos (|c X)
1¼c = −√2 •¦§ ∙ sin (|c X)
8.4

1 c=0
8.5
8.6

The inverse transformation in the stator:

cos (3) 012(3) 1

⎡ 2 2 ⎤ 1
1(c 2 ⎢cos (3 − ) sin (3 − ) 1⎥ Ac
p1}c q = ⎢
3
3 3 ⎥ p1¼c q
1.c ⎢ 2 2 ⎥ 1 c
⎣cos (3 + ) sin (3 + ) 1 ⎦
3 3

1(c = 1Ac cos(|c X) + 1¼c 012(|c X) 8.7

1}c = 1Ac cos Z|c X − •
[ + 1¼c 012 Z|c X − •
[ 8.8
1.c = 1Ac cos Z|c X + •
[ + 1¼c 012 Z|c X + •
[ 8.9

1c = 1(c and in phasor form:

̅ = Ac
c
̅ + h ¼c
̅
̅ = 1Ac ⁄√2 and ¼c
̅ = 1¼c ⁄√2
8.10
Where Ac

Having:

/Ac = c 1Ac − |c Ψ¼c + ΨAc

/¼c = c 1¼c + |c ΨAc + Ψ¼c
8.11
8.12

72
 Electric Machines – Basic Theory

/′A = ′ 1′A − (|c − | )Ψ′¼ + Ψ′A

/′¼ = ′ 1′¼ + (|c − | )Ψ′A + Ψ′¼
8.13
8.14

Under steady state condition, derivatives terms disappear, and Equations 8.11 to 8.14 become:

/Ac = c 1Ac − |c Ψ¼c

/¼c = c 1¼c + |c ΨAc
8.15

/′A = ′ 1′A − (|c − | )Ψ′¼

8.16

/′¼ = ′ 1′¼ + (|c − | )Ψ′A

8.17
8.18

Where:

ΨAc = NFc 1Ac + N' (1Ac + 1′A )

Ψ¼c = NFc 1¼c + N' $1¼c + 1′¼ %
8.19

Ψ′A = N′F 1′A + N' (1Ac + 1′A )

8.20

Ψ′¼ = N′F 1′¼ + N' $1¼c + 1′¼ %

8.21
8.22

/Ac = c 1Ac − |c (NFc + N' )1¼c − |c N' 1 Y ¼

/¼c = c 1¼c + |c (NFc + N' )1Ac + |c N' 1 Y A
8.23

/′A = ′ 1′A + (|c − | )(NY F + N' )1′¼ + (|c − | )N' 1¼c

8.24

/′¼ = ′ 1′¼ + (|c − | )(NY F + N' )1′A + (|c − | )N' 1Ac

8.25
8.26

•;c = fc c̅ + h|c (NFc + N' ) c̅ + h|c N' ̅

•;c = fc c̅ + h|c NFc c̅ + h|c N' ( ̅ + c̅ )
8.27

•;c = fc c̅ + h<c c̅ + h<' ( ̅ + c̅ )

8.28
8.98

•; = f ̅ + h(|c − | )(NF + N' ) ̅ + h(|c − | )N' ̅

0 = f ̅ + h(|c − | )NF ̅ + h(|c − | )N' ( ̅ + ̅ )
8.30

0 = f ̅ + h(|c − | )NF ̅ + h(|c − | )N' ( ̅ + ̅ )

8.31
8.32

0=
( ¤ ^)
8.33

0= + h|c NF ̅ + h|c N' ( ̅ + ̅ )

•^ ^̅
c
8.34
0= + h< ̅ + h<' ( ̅ + ̅ )
•^ ^̅
c
8.35

In balanced systems c =0

Similar transformations apply to flux linkages and stator voltages.

Defining

|c = 3 (synchronous velocity) and | = 3 (rotor angular velocity) and the original voltage equations:
A A
A, A,
In the rotor, the transformation to apply will be:

⎡cos ( ) cos ( −
•
) cos ( + )⎤
•
⎢ ⎥
A (
T ¼ U = ⎢ 012( ) sin ( − ) sin ( + ) ⎥ p } q
⎢ E ⎥
• • •
8.36
E E .
⎣ ⎦

Where: =3−3

73
 Electric Machines – Basic Theory

cos ( ) 012( ) 1
And
( A
p } q = • ·cos ( − •
) sin ( − •
) 1¸ T ¼ U
cos ( + ) sin ( + ) 1
8.37
.
• •

1( = A cos( ) + ¼ 012( ) 8.38

1} = A cos Z − [+ ¼ 012 Z − [
• •
8.39
1. = A cos Z + •
[+ ¼ 012 Z + •
[ 8.40

1( = ' cos (|c X + .)

1} = ' cos (|c X + . − 2 /3)
8.41

1. = ' cos (|c X + . + 2 /3)

8.42
8.43

1( ' cos (|c X + .)

p1} q = T ' cos (|c X + . − 2 /3)U
1. cos (| X + + 2 /3)
8.44
' c .

⎡cos (3) cos (3 − • ) cos (3 + • )⎤

Applying the transformation
1Ac 1( ' cos (|c X + .)
1 1 ⎢ ⎥
p ¼c q = ¬ p } q = • ⎢ 012(3) sin (3 − • ) sin (3 + • ) ⎥ T ' cos (|c X + . − 2 /3)U 8.45
1½ 1. ⎢ E E E ⎥ ' cos (|c X + . + 2 /3)
⎣ ⎦
The d-axis is assumed to be 90º ahead of the q-axis in the direction of rotation. If the d-axis is so

time is |c X + , then 3 = |c +
chosen that it is 90º ahead of the phase a axis at time t=0, its displacement from phase a axis at any

1Ac = ' cos Z. + [=− ' 012(.) 8.46

1¼c = − ' sin Z. + [=− ' :;0(.) 8.47

8.1.1. Equivalent Circuit27

An induction motor has two magnetically coupled circuits: the stator and the rotor. It is similar to the
transformer equivalent circuit and can be treated for analysis essentially as a transformer whose
secondary is rotating and short-circuited.

Figure 8.7 Overview of the circuit equivalent in a motor.

The motor has balanced three-phase circuits; consequently, the single-phase representation is
sufficient for the majority of analysis needed.

27
George G. Karady & Keith Holbert: “Energy Conversion and Transport, Chapter 7, Induction Motors.” EEE 360.

74
 Electric Machines – Basic Theory

The induction motor has stator leakage reactance, stator copper loss as series components, and iron
loss and magnetizing inductance as shunt ones. The rotor circuit likewise has rotor leakage reactance,
rotor copper (aluminum) loss and shaft power as series elements.

A magnetizing reactance (=õ) and a resistance connected in parallel represent the magnetic field
generated. The resistance (òu) represents the eddy current and the hysteresis losses in the iron core.
RS I R _T XR RR S
XS

IS IR
VGRID RC Xm VS VR

Figure 8.8 Single-phase equivalent circuit of a three-phase induction motor.

In this circuit, the magnetizing reactance generates a flux that links with both the stator and the rotor
and induces a voltage in both circuits. The magnetic flux rotates with constant amplitude and
synchronous speed. This flux cuts the stationary conductors of the stator with the synchronous speed
and induces a 60 Hz or 50 Hz voltage in the stator windings. The RMS value of the voltage induced in
the stator can be expressed as follows:

•§ =
P> BOÝ? >
√
8.48
Where

<§ = Number of turns in the Stator.

7'(e = Peak value of the magnetic flux.
|§ = synchronous speed.

The flux rotates with the synchronous speed and the rotor with the motor speed, consequently, the
flux cuts the rotor conductors with the speed difference between the rotating flux and the rotor ones.
The speed difference is calculated using the slip equation:

(|§ − |' ) = |§ 0
Then: 0 =
8.49
( >¤ O)
8.50
>

The induced voltage in the rotor can be expressed by the following equation:

•• = =
P BOÝ? ( > ¤ O ) P BOÝ? > c
√ √
8.51

The division of the rotor and stator induced the following voltage results:

•• = •0 = ••§ 0
P
P> §
8.52

This speed difference determines the frequency of the rotor current:

` ~, = = >¤ O
= >c
= 0`§ 8.53

The rotor circuit leakage reactance is then:

75
 Electric Machines – Basic Theory

<•¤' = N• |• = N• |§ 0 = <• 0 8.54

The relation between the rotor current and the rotor-induced voltage is calculated by the loop voltage
equation:

•• = ••§ 0 = • (f• + h<• 0) 8.55

The division of this equation with the slip yields:

••§ = Z c + h<• [
•
• 8.56

RS I R _T XR RR S
XS

IS IR
VGRID RC Xm VS VR

Figure 8.9 Modified equivalent circuit of a three-phase induction motor.

The rotor impedance can be transferred to the stator side. This eliminates the ideal transformer used
for the previous representation.
X S = wS LS RS I R _T X R _T RR _ T S

IR _T
VGRID RC Xm VS

Stator Rotor
Air gap
Figure 8.10 Simplified equivalent circuit of a three-phase induction motor.

The last modification of the equivalent circuit is the separation of the rotor resistance into two parts:

= f•_& + f•_&
• _A (E¤c)
c c
8.57

The second term of the Equation 8.57 corresponds to the resistance that depends on the slip and
represents the outgoing mechanical power, or the load, being the first term, the losses.

f•_&
(E¤c)
c
8.58

76
 Electric Machines – Basic Theory

X S = wS LS RS I R _T X R _T RR _ T

IR _T
RR _ T (1 − S ) S
VGRID RC Xm VS

Figure 8.11 Final single-phase equivalent circuit of a three-phase induction motor.

From the equivalent circuit and a basic knowledge of the operation of the induction motor, we observe
that the magnetizing current component and the iron loss of the motor are voltage-dependent, and not
load-dependent. Consequently, the full voltage starting current of a particular motor is voltage and
speed-dependent, but not load-dependent.

8.1.2. Application Example - Type II wind turbine variable slip generator

The variable-slip generator is essentially an induction generator with a variable resistor in series with
the rotor circuit. The resistor is controlled by a high-frequency semiconductor switch. When the wind
generator operates below rated wind speed and power acts as a conventional fixed-speed induction
generator. When the operation is above rated, control of the resistance allows the air-gap torque to be
controlled and the slip speed to vary, so that behavior is then similar to that of a variable-speed
system28.
CONTROLS
MID-RANGE TO
RATED POWER

OPERATION
START-UP &
PFCC

NORMAL

Figure 8.1 Variable slip generator.

28
Olimpo Anaya-Lara, Janaka Ekanayake, Nick Jenkins, Phill Cartwright, Mike Hughes,” WIND ENERGY GENERATION Modelling and
Control.” 2009 John Wiley & Sons, Ltd.

77
 Electric Machines – Basic Theory

9 Chapter II-CLASSICAL AUTOMATIC

CONTROL

Control theory deals with the control of dynamical systems in engineered processes and machines.
The objective is to develop a model or algorithm governing the application of system inputs to drive
the system to the desired state, while minimizing any delay, overshoot, or steady-state error and
ensuring a level of control stability; often to achieve a degree of optimality.

9.1. Classical Control Theory29

Fundamentally, there are two types of control loops: open-loop control and closed-loop (feedback)
control.

In open-loop control, the control action from the controller is independent of the "process output" (or
"controlled process variable" - PV). A good example of this is a central heating boiler controlled only
by a timer, so that heat is applied for a constant time, regardless of the temperature of the building.
The control action is the timed switching on/off of the boiler, the process variable is the building
temperature, but neither is linked.

In closed-loop control, the control action from the controller is dependent on feedback from the
process in the form of the value of the process variable (PV). In the case of the boiler analogy, a
closed-loop would include a thermostat to compare the building temperature (PV) with the
temperature set on the thermostat (the setpoint - SP). This generates a controller output to maintain the
building at the desired temperature by switching the boiler on and off. A closed-loop controller,
therefore, has a feedback loop that ensures the controller exerts a control action to manipulate the
process variable to be the same as the "Reference input" or "set point". For this reason, closed-loop
controllers are also called feedback controllers.

The classical control theory is based on a simple input-output description of the plant and represented
as a transfer function. The methods used do have not any knowledge of the interior structure of the
plant and limit us to single-input single-output (SISO) systems and allow only limited control of the
closed-loop behavior when feedback control is used30.

29
Wikipedia
30
Derek Rowell. “Analysis and Design of Feedback Control Systems. State-Space Representation of LTI Systems,” October 2002

78
 Electric Machines – Basic Theory

Figure 9.1 Example of a single industrial control loop with a block diagram of a negative feedback
control system using a feedback loop to control the process variable by comparing it with the desired
value and applying the difference as an error signal to generate a control output to reduce or eliminate
the error.

9.1.1. Mathematical Models of Control Systems31

Block diagrams depicting the transfer functions of an element or group of elements of an excitation
system constitute a mathematical model of the physical system. A good model for a control system
exhibits the same gain and phase characteristics.

9.1.1.1. Frequency Response of a Circuit Represented by a Block Diagram

A block diagram that is a simplified representation of a static excitation system is shown in Figure
9.1.

V i (s ) Ve (s) 300 V0 (s)

G(s) =
1 + 0.7s

Vf (s)

0.02s
H(s) =
1+ s
Figure 9.1 Block Diagram of a feedback control system

The input and output signals can be electrical (V, I) analogs of some physical quantities (force,
acceleration), or they can be mixed (with consistent units).

The output transfer function is defined as the ratio32 of the output to the input signal.

If both signals are voltages, then the function

Ð(0) = = Eà
d (c) •
d# (c) .Bc
9.1

31
“Excitation System Dynamic Characteristics”. IEEE Committee Report, 1973.
32
“Terminology for Automatic Control” ANSI C85.1-1963.

79
 Electric Machines – Basic Theory

This is the abstract mathematical expression for a voltage ratio. To know what combination of
physical elements could produce it, we can consider the circuit of Figure 9.2b which has R and jwL
associated with an ideal transformer. For that, the ratio of output to the input voltage is:

= =
d • .• •
d# •àŠ1ì EàŠ1
C 9.2

If the ratio L/R is chosen to be 0.7, then the Equation 9.2 becomes:

= 300 ‡ Eà
d E
d# .Bc
9.3

This is identical to Equation 9.1.

Ve (s) 300 V0 (s)
G(s) =
1 + 0.7s

L
= 0.7 sec .
R

Ve V0

Figure 9.2 Block Diagram and physical equivalents.

9.1.1.2. Laplace Transformation

The Laplace transform (Pierre-Simon, marquis de Laplace, French: 23 March 1749 – 5 March 1827)
is a mathematical transformation that converts time-domain functions into complex-frequency
functions in a manner analogous to the transformation of real numbers into algorithms, but in this case
the new space for analysis eliminates time simplifying the solution of the function. Analysis in the
frequency domain facilitates calculation because it permits the use of algebraic operations to replace
those of integration and differentiation. Algorithms when transformers are from real numbers permit
the use of the simpler operations of addition and subtraction to replace multiplication and division. In
brief, the purpose of transformation is to simplify the calculation process, maintaining the result
identical.

When the inputs and outputs of either block are functions of time, their ration G(s) and H(s) are

for steady-state and transient analysis = D + hn and for steady-state analysis D = 0. The real part of
expressed as functions of ” s” rather than of “t” . For steady-state circuit analysis s=jw. In general,

the frequency is not significant for the solution.

When mathematical operations are completed in the complex-frequency domain, the inverse
transforms of the resulting expressions convert them to the time domain.

80
 Electric Machines – Basic Theory

9.1.1.3. Transfer Functions of a Feedback System

The system of Figure 9.1 is a feedback control system. It has two transfer functions of major interest.

the output voltage to the Laplace transform of the input voltage(• (0)/•- (0)). The second is the
The first is the output of “closed-loop transfer function” which is the ratio of the Laplace transform of

so the output voltage is •* (0) and the input voltage is •+ (0). In this way, the open-loop transfer
“open-loop transfer function”. The loop is opened by disconnecting the lower block and the summer,

function is •* (0)⁄•+ (0). If there are other inputs to the summer, the output transfer function is the
ratio of the Laplace transforms of the output voltage and the sum of all input voltages.

= Ð(0) (0) = Eà ‡ = (Eà

d! (c) • . c ƒc
d# (c) .Bc Eàc .Bc)(Eàc)
9.4

Figure 9.3 Time domain and frequency domain transformation1

The closed-loop transfer function is as follows:

• (0) = Ð(0)¾•- (0) − •* (0)¿ 9.5

Where:

•* (0) = (0)• (0) 9.6

Equation 9.6 into 9.5

= Eà½(c)
d (c) ½(c)
d" (c) (c)
9.7

As we can see, the open-loop transfer function is part of the denominator of the closed-loop transfer
function and almost all feedback systems can have their closed-loop transfer functions manipulated
into the form of Equation 9.7. This is an important aspect that will help us in the analysis.

9.1.1.4. The Bode Diagram

The transient response of a system can be graphically described showing how the magnitude of the
output signal varies with time for any given input signal, usually a step function or an impulse
function. The frequency response is described by how the output signal varies as the frequency of a
sinusoidal input signal is varied.

81
 Electric Machines – Basic Theory

Hendrik Wade Bode, a Bell System research engineer (1905–1982), while working at Bell Labs in the
United States in the 1930s, developed the method for describing the frequency response of a circuit.
The Bode diagram provides a clear picture of how gain and phase angle between input and output
signals of a linear system varies with frequency. It is also a convenient method for using measured
frequency response data obtained from field or factory tests, to construct representative transfer
functions.

Let´s start with the following equation:

= (0) = å(c)
d (c) P(c)
d" (c)
9.8

(0)= Transfer function

Where

For steady-state sinusoidal input voltage, the complex frequency D + hn becomes the known
frequency jw.

Equation 9.8 can be written:

(hn) =
P(Š1)
å(Š1)
9.9

Since all terms of Equation 9.9 are complex values, they can be written in terms of their magnitudes
(gains) and phase angles.

| (hn)|∠∅ &(Š1) =
|P(Š1)|∠∅E(ÜF)
|å(Š1)|∠∅G(ÜF)
9.10

Taking the natural logarithms of both sides of Equation 9.10.

2| (hn)| + h∅ &(Š1) = 2|<(hn)| + h∅P(Š1) − 2|%(hn)| − h∅å(Š1) 9.11

Writing the real terms and imaginary terms separately, we have:

2| (hn)| = 2|<(hn)| − 2|%(hn)| 9.12

∅ &(Š1) = ∅P(Š1) − ∅å(Š1) 9.13

As 2 = 2.3 ;ÏE , then, we multiply both sides of Equation 9.12 by 20/2.3, then:

20 ;ÏE | (hn)| = 20 ;ÏE |<(hn)| − 20 ;ÏE |%(hn)| 9.14

As 20 ;ÏE | | = o, gain of the absolute value in Decibels, then Equation 9.14 can be rewritten:

A} (hn) = <A} (hn) − %A} (hn) 9.15

The logarithmic expression of Equation 9.14 possesses certain simple asymptotic properties which
make it possible to construct their complete frequency characteristics without several plotting points.
Also, the overall gain of a transfer function composed of several transfer functions in series, is simply
the sum of the gains of the individual transfer functions when the gains are in logarithmic form. The
use of Decibels, simplify the notation, but all relations are logarithmic ones.

82
 Electric Machines – Basic Theory

9.1.1.4.1. Bode Diagrams for Closed-Loop Transfer Functions

Consider the data in Figure 9.2. Equation 9.1 becomes:

<A} (hn) = 20 ;ÏE |Ð(hn)| = 20 ;ÏE H H = 20 ;ÏE

• •
EàŠ .B1 ŒEà( .B1)I
9.16

<A} (hn) = 20 ;ÏE 300 − 20 ;ÏE Œ1 + (0.7n) 9.17

reference. w=2πf. For values (0.7n) ≪ 1, i.e., for very low frequencies:
“j” is no longer necessary for magnitude evaluation and only we will keep in mind that it is 90º from

<A} (hn) = 20 ;ÏE 300 − 0 = 20 ∗ 2.477 = 49.5 o 9.18

This value is called DC or static gain of high-frequency values of w such that (0.7n) ≫ 1, the
Equation 9.17 becomes:

<A} (hn) = 20 ;ÏE 300 − 20 ;ÏE (0.7n) 9.19

So that the high frequency asymptote is a linear function of ;ÏE (0.7n). When w=1/0.7, Equation
9.19 yields <A} (hn) = 20 ;ÏE 300. High and low-frequency asymptotes intersect at this point.

When w=10 x 1/0.7, Equation 9.19 yields <A} (hn) = 20 ;ÏE 300 − 20, so that the slope of the high
frequency asymptote is -20 dB/decade, where a decade is any interval on the frequency scale covering
10:1 ratio.

Note that where the two asymptotes intersect, w=1/0,7, the actual curve is 3 db lower than the
intersection of the asymptotes. The actual curve is dotted.

The frequency characteristics of the phase angle can also be drawn.

<(hn) = =
• • ∠ º
ŒEà( .B1)I ŒEà( .B1)I ∠,(*ßG ( .B1)
9.20

The phase angle of <(hn) is ∅P(Š1) = −Xm2¤E (0.7n)

At the limit when w = 0, ∅P(Š1) = 0º; and when n = ∞, ∅P(Š1) = 0º . When w = 1/0,7, ∅P(Š1) =
9.21

−45º and an approximation for the asymptote is to take -45º/decade.

Now, we will consider the frequency characteristics of the denominator %(hn) = 1 + Ð(hn) (hn)

20 ;ÏE %(hn) = 20 ;ÏE j1 + Ð(hn) (hn)l 9.22

%A} = 20 ;ÏE H1 + EàŠ ‡ =

• . Š1 .B(Š1)I àƒ.BŠ1àE
.B1 EàŠ1
H (EàŠ .B1)(EàŠ1)
9.23

The roots of the numerator are found to be (jw) = -9.4192, -0.151, so that the expression becomes:

=
.B(Š1àM.–EM )(Š1à .EëE)
(EàŠ .B1)(EàŠ1)
9.24

83
 Electric Machines – Basic Theory

Figure 9.4 Bode Plot of <A} (hn).

Taking the absolute magnitude of each of the values in parenthesis, we can write Equation 9.24 as:

I I
ÁEàZ F [ ÁEàZ F [
%A}(Š1) = 20 ;ÏE T N.OGNI .GPG
U
ŒEà( .B1)I ŒEà(1)I
9.25

The logarithmic form of Equation 9.25 can be expressed as follows:

%A}(Š1) = 20 ;ÏE Á1 + Z [ + 20 ;ÏE Á1 + Z [ − 20 ;ÏE Œ1 + (0.7n) −

1 1
M.–EM .EëE
20 ;ÏE Œ1 + (n) 9.26

The bode plot that we obtain for different components is the following one:

84
 Electric Machines – Basic Theory

Figure 9.5 Bode Plot of A} (hn)

If we want to construct the transfer function, we can use the bode plot (if the frequency response of a
suitable transfer function were known). The parameters of the transfer function could be derived from
the frequency response parameters – the breakpoints and the slopes. A convenient procedure could be
to obtain, by field measurements, the response of the magnitude and phase angle of the output voltage
as the frequency of the sinusoidal voltage to the piece equipment is varied.

9.1.1.4.2. Bode Diagram of the Open-Loop Transfer Function

The open-loop transfer function of Figure 9.1 is given in Equation 9.4.

= Ð(hn) (hn) =
d! (Š1) ƒŠ1
d# (Š1) (Eà .BŠ1)(EàŠ1)
9.27

From that equation, we can obtain the gain and phase as a function of frequency, being the gain:

Ð(hn) (hn)A} = 20 ;ÏE 12n − 20 ;ÏE Œ1 + (0.7n) − 20 ;ÏE Œ1 + (n) 9.28

And the phase as function of frequency, is:

∅½(Š1) (Š1) = 90º − Xm2¤E (0.7n) − Xm2¤E (n) 9.29

The diagram plotted is shown in Figure 9.6.

85
 Electric Machines – Basic Theory

Figure 9.6 Open-loop gain and phase frequency domain response.

9.1.1.5. Complex Frequency Domain

The time response of a circuit can be obtained by taking the inverse Laplace transform of the output of
the signal expressed in the complex frequency domain, but this is often impractical for a complicated
circuit.

There is a whole field of analysis of feedback control systems, first introduced by W.R. Evans known
as the Root-Locus Method33-34.

Consider Equation 9.1.

Ð(0) = = Eà
d (c) •
d# (c) .Bc
9.30

If s = -1/0.7 = -1.4286, then the denominator will equal zero, and the value of • (0)⁄•+ (0) = ∞. Any
value of s which makes a function infinite, is known as a pole of that function. Since 0 = D + hn, the
pole of the function can be located on the complex-frequency plane. In the case of Equation 9.1, D = -
1.4286 and jw = 0. The corresponding plot of the pole can be seen in the Figure 9.5.

33
W.R. Evans, Control System Dynamics, McGraw Hill Book Co., Inc., New York, New York, 1954.
34
J.G. Truxal, Automatic Feedback Control System Synthesis, McGraw-Hill Book Co., Inc., New York, New York, 1955.

86
 Electric Machines – Basic Theory

•
Eà .Bc
Figure 9.7 S plane plot of

d (c)
If S = ∞ in Equation 9.30, then the denominator will be infinite, and the value of d (c) will equal to
#
zero . Any value of s which makes a function zero is known as a “zero” of that function. Either D or
jw can equal infinite. All such values of D + hn lie on a circle at infinite. See Figure 9.7.
. c
Eàc
Now, we consider Equation

This function has one zero and one pole. The zero is located at σ = 0, jw = 0 and is shown as a circle
at the origin. The pole is located at σ = -1, jw = 0. These points are shown in Figure 9.8.

. c
Eàc
Figure 9.8 S plane plot of

Finally, consider Equation 9.4.

= Ð(0) (0) = Eà ‡ = (Eà

d! (c) • . c ƒc
d# (c) .Bc Eàc .Bc)(Eàc)

This equation has two zeros, at s = 0, s = ∞; and two poles, at s = -1.4286 and at s = -1. The two zeros
and two poles are shown in Figure 9.9.

87
 Electric Machines – Basic Theory

Figure 9.9 S plane of Ð(0) (0) = ‡ Eàc = (Eà

• . c ƒc
Eà .Bc .Bc)(Eàc)

Consider another example: A function which has a denominator 0 + 20 + 5. This can be factored
into (0 + 1 + h2)(0 + 1 − h2). This function will have poles at s = -1+j2 andf s = -1-j2 as well as
zeros all along the circle at infinity. See Figure 9.10.
6

0
-6 -4 -2 0 2 4 6

-2

-4

-6
Q
Figure 9.10 S-Plane plot of cI à càë

Laplace transform theory tells us that the perpendicular distances of a pole from the σ-axis and the jw-
axis have important physical implications. (See Figure 9.10). If the distance of the pole from the σ-
axis is zero, i.e., if the pole is on the σ-axis, the time-response of the function is non-oscillatory; the
greater the distance of the pole from the σ-axis, the higher will be the frequency of oscillation of the
time-response function. If the distance of the pole from the jw-axis is zero, i.e., if the pole is on the
jw-axis, the time response of the function does not decay. The greater the distance of the pole from
the jw-axis in the left half of the plane, the more rapid will be the decay of the time-response of the
function; the greater the distance of the pole from the jw-axis in the right half of the plane, the more

88
 Electric Machines – Basic Theory

rapid will be the growth of the time-response of the function. This later information tells us that all
poles must lie to the left of the jw-axis or on the jw-axis if the function is to be stable.

9.1.1.6. Determining Time response from Frequency response

Transient response is meant the description in real-time of how a circuit responds to a disturbance.
Figure 9.11 shows a typical time (transient) response of the output of a system to a step increase in
input to the system.

Figure 9.11 Typical time response of the system to a step increase in input

The time response of the output of a feedback control system may be obtained by taking the inverse
Laplace transform of the function found by multiplying the Laplace transform.

VR (s) Ve (s) Vf (s) VT (s)

400 1 1
1 + 0 .02 s 1 + 0 .8 s 1 + 5s

0 . 03 s
1 + s

Figure 9.12 Typical excitation control system at no load.

The approximation technique is based on estimating the values of the poles of the closed-loop transfer
function from the system open-loop transfer function and estimating only a few of the coefficients of
the approximate partial fraction expansion of the closed-loop transfer function. The approximate poles
along with the approximate coefficients, are used to estimate the principal characteristics (rise time,
overshoot, and settling time) of the time response

89
 Electric Machines – Basic Theory

The gain information offered by the asymptotes of the open-loop Bode plot enables one to determine
quickly which closed-loop poles are near the beginning, which are near to the end, and which are near
the middle of migration along the root loci.

G ( jω )

wC P2 P3
P0 P1 Z 1 LOG ω

+ jω
−σ

V 0 (t )
Vi (t )

Figure 9.13 Comparison of the frequency domain, S-Plane, and time domain

Using the Final Value and Initial Value theorems, Chen35 suggested that the closed-loop poles be
divided into three groups corresponding to three ranges of the frequency response (Figure 9.13(a)) and
time response characteristics (Figure 9.13(c)), respectively The Bode plot in Figure 9.13(a) represents
the open-loop frequency response of the feedback control system shown in Figure 9.12. This Bode
plot represents the open-loop frequency response of the feedback control system shown in Figure
9.12. It should be noted that the ±15 DB and selected to divide the frequency range into the three

35
K. Chen, “A quick method for Estimating Closed-loop poles of Control Systems”. AIEE transactions, Applications and Industry, vol. 76,
pt. II, pp. 80-87, May, 1957.

90
 Electric Machines – Basic Theory

regions corresponding to the pole migration in the s-plane, is arbitrary. It was suggested by Chen
based on engineering experience in the application of the Initial and Final values corresponding to
many typical control systems.

In Figure 9.13, the range I corresponds to the locus of closed-loop poles near the beginning of
migration, i.e., near the open-loos. Range I also corresponds to a sub-transient in the time domain and

response Ð(hn) (hn) ≪ 1: therefore:

to the frequency response below the -15db band on the Bode plot. In the range of the frequency

(0) = ≈ Ð(0)
dA ½(c)
d Eà½(c) (c)
9.31

And the closed-loop pole ≈-p3, the open-loop pole corresponding to the break-point on the Bode plot
slightly above the 15db line. In terms of the root-loci, this simply means that the closed-loop pole
which starts from –p3 has not moved far from s=-p3 for the gain in range I of the Bode plot. This
close-loop near –p3 corresponds to a transient term that decays rapidly and the overall system
response is not affected significantly if the closed-loop pole is assumed not to have moved from –p3
at all. Poles other than –p3 may be neglected in a closed-loop.

At the other extreme, Range III of Figure 9.13 corresponds to closed loop poles near the end of

frequency response, above the +15db line on the Bode plot Ð(hn) (hn) ≫ 1 and the following
migration, i.e., near the open-loop zeros (approaching steady-state). In the corresponding range of the

approximation may be written:

(0) =
d# E E
d Eà½(c) (c)
≈ ½(c) (c)
9.32

(0), which are the same as the

dA
d
From this approximation, it is evident that the closed-loop poles of
(0), are approximately the open-loop zeros of G(s)H(s).
d#
d
poles of
As shown in Figure 9.13(b), one approximately closed loop near is near the open loop zero, s=-z1 in
range III. The open-loop zero corresponds to the breaking point on the Bode plot above the 15db line.
In terms of pole migration along the root loci, this means that one close-loop pole has moved from –
p1 very close to –z1 but has not quite reached –z1. Since s=-z1 is also a closed-loop zero, there is a
“dipole” (formed by a zero and a pole close to each other) near s=-z1. The dipole has a significant
effect on the settling time of the system since, as shown in Figure 9.13(c), it corresponds to a small
but long transient term producing a “tail” in the system response.

Chen has shown that in the case of a dipole close to the origin of the s plane the closed-loop pole is
approximately located at:

9 = −(1 + w Y )gE 9.33

corresponding to gE .
Where K’ is the gain obtained from the open-loop frequency asymptote plot at the break frequency

Finally, Range II of Figure 9.13(b) corresponds to the locus of closed-loop poles in the middle of
migration. Range II also corresponds to the main transient including equal rise time and overshoot.
Concerning the Bode plot, it corresponds to the frequency response in the region of the cross-over
frequency and within the ±15bd band.

The overall open-loop transfer function in terms of its poles and zeros may be written as:

Ð(0) =
QYY(§à,G )
(§à$ )(§à$G )(§à$I )(§à$S )
9.34

91
 Electric Machines – Basic Theory

Where w YY = is the gain which satisfies Ð(0) (0) = −1 along the root loci.
Q$ $G $I $S

In the mid-range of the frequency response, it may be considered that » < »E < gE ≪ |0| ≪ »• and
,G

Ð(0) ≈
QYY⁄$S
hence the following approximation may be written:
)
§(cà$I
9.35

The approximate closed-loop transfer function is then:

½(c) QYY⁄$S
≈ §(cà$
Eà½(c) I )àQYY⁄$S
9.36

From this approximation, the approximate complex closed loop poles are obtained as the roots of the
equation:

9 +» 9+ =0
QYY
$S
9.37

Approximation of the closed-loop transfer function within the ±15db does not always result in a
quadratic for the characteristic equation; however, for most stable systems, the approximation will
usually not be higher than the third order. Several common variations of open-loop frequency
response characteristics near the cross-over frequency along with their corresponding transfer
functions are tabulated in reference35.

The techniques for approximating the transient response from the frequency response can best be
illustrated by applying them to an excitation control system. Figure 9.10 represents a typical
excitation control system off-line, (generator at no load) assuming that the synchronous machine may
be represented by a single time constant.

9.1.1.7. Frequency Response performance Criteria

Rise time, overshoot, and settling time are the principal performance criteria concerning the transient
or time response of a feedback control system.

The main performance criteria concerning the frequency response of a feedback control system
include the bandwidth, phase margin and gain margin.

Bandwidth is the interval separating two frequencies between which both the gain and the phase
difference (of sinusoidal ouas tput referred to sinusoidal input) remain within specified limits. For
excitation control systems and many of their components the lower frequency is zero; the higher

1/√2 of its low-frequency value.

frequency is usually defined as the frequency at which the closed-loop frequency response falls to

Bandwidth is related to the cross-over frequency of the open-loop frequency response of the system
and to the system transient response (rise time). Bandwidth is also indicative of the highest input
frequency that can be handled by the system, i.e., it approximately describes the filtering
characteristics of the system.

Phase and gain margin are both measures of the relative stability of a control system. A feedback
control system will be unstable if the phase shift is greater than 180º at the frequency at which the
gain is unity, i.e., at the crossover frequency on the Bode plot of the system open-loop characteristic.

Phase margin is equal to 180º minus the absolute value of the loop phase angle at a frequency where
the loop gain is the unity. A phase margin of 40º or more is generally considered good design practice.
Similarly gain margin is defined as the reciprocal of the gain at the frequency at which the phase

92
 Electric Machines – Basic Theory

angle reaches minus 180º. A gain margin of 6 dB or more is generally considered good design
practice.

9.2. Software tools for control analysis

There are diverse software tools for classical and modern control analysis. Vendors as Matlab®,
Etap®, among others provide software modules that facilitate us the task. In the next example we are
utilizing the open-source software Octave. https://www.gnu.org/software/octave/index.

Ï=
E
∙c I à•∙cà–
Function:

Figure 9.14 Gain and phase margins36.

36
Root Locus Design and SISO Design Tools. Octave®. https://octave.sourceforge.io/.

93
 Electric Machines – Basic Theory

Figure 9.15 Pole Zero Map12

Figure 9.16 Step response12.

94
 Electric Machines – Basic Theory

10
10.1. Modern Control Theory37
Chapter X-MODERN
CONTROL THEORY

Modern control theory solves many of the limitations of classical control, by using a more complete
description of the plant dynamics. The so-called state-space (SS) description provides the dynamics as
a set of coupled first-order differential equations in a set of internal variables known as state variables,
together with a set of algebraic equations that combine the state variables into physical output
variables, rather than by one or more nth-order differential or difference equations.

The principle of SS is based on transforming a motion equation of order n into n equations of 1st
order. The concept of the state of a dynamic system refers to a minimum set of variables, that fully
describe the system and its response to any given set of inputs. The state variables are an internal

which any output variables TW (¹) may be computed.

description of the system which completely characterize the system state at any time t, and from

characterized by the response of the set of n variables UW (¹), where the number n is defined
This definition asserts that the dynamic behavior of a state-determined system is completely

to be the order of the system.

The system shown in Figure 10.1 has r inputs ÖE (X) to Ö (X), and m output variables ˆE (X), … ˆ' (X).
If the system is state-determined, knowledge of its state variables (‡E (X ), ‡ (X ), … ‡* (X )) at some
initial time X , and the inputs ÖE (X) to Ö (X) for t ≥ X is sufficient to determine all future behavior of
the system.

The state variables represent storage elements in the system which leads to derivative equations
between its input and output; it could be physical or mathematical variables.

Figure 10.1 System Inputs and Outputs.

37
Derek Rowell. “Analysis and Design of Feedback Control Systems, State-Space Representation of LTI Systems”, 2002.

95
 Electric Machines – Basic Theory

+ 17 A, + 11ˆ = 13 A, + 23Ö
AS i AI i Ai A
A, A, I
S +5 10.1

Equation 10.1 represents a system of order 3, then it has 3 states and 3 storage elements.

10.1.1. The state equations

The standard form of the state equations is expressed through a mathematical description of the
system as a set of n coupled first order ordinary differential equations, known as state equations, in

(‡E (X), ‡ (X), … ‡* (X)) and the system inputs ÖE (X) to Ö (X). The state variables defined must be
which the time derivative of each state variable is expressed in terms of the state variables

linearly independent, i.e., no state variable can be written as a linear combination of the other state
variables or the system will not be able to be solved.

The behavior of a dynamic system, such as a power system, may be described by a set of n first order
nonlinear ordinary differential equations of the following form:

UW = VW (U¥ , UW , … , Uù ; Y¥ , YW , … , Yø ; ¹) W = ¥, W, … , ù 10.2

Where n is the order of the system and r is the number of inputs.

‡E
= ‡E = `E (‡, Ö, X)
X
= ‡ = ` (‡, Ö, X)
AeI
A,
10.3
∶=:
‡*
= ‡* = `* (‡, Ö, X)
X

variables is written as a state vector ‡(X) = j‡E (X), ‡ (X), … ‡* (X)l& and the set of r inputs is written
It is common to write the set of the above functions in a vector form, in which the set of n state

as an input vector Ö(X) = jÖE (X), Ö (X), … Ö (X)l& . Each state variable is a time varying component of
the column vector x(t).

Uù = Vù (U, Y, ¹) 10.4

‡E ÖE `E
Where

‡ Ö `
U=· ⋮ ¸ Y=·⋮ ¸ V=· ¸
⋮
‡* Ö `*

The column vector x is referred to as the state vector, and its entries ‡- as state variables. The column
vector u is the vector of inputs to the system. These are the external signals that influence the
performance of the system. Time is denoted by t, and the derivative of a state variable x with respect
to time is denoted by U. If the derivatives of the state variables are not explicit functions of time, the
system is said to be autonomous. In this case, Equation 10.4 simplifies to:

U = V(U, Y) 10.5

We are often interested in output variables that can be observed in the system. These may be
expressed in terms of state variables and the input variables in the following form:

T = [(U, Y) 10.6

96
 Electric Machines – Basic Theory

ˆE ÏE
Where

ˆ Ï
\=·⋮¸ ]=· ⋮ ¸
ˆ* Ï*

The column vector y is the vector of outputs, and g is a vector of nonlinear functions relating state
and input variables output variables

We will restrict our attention to systems that are linear and time-invariant (LTI), that is systems
described by linear differential equations with constant coefficients.

Equation 10.4 can be written as:

‡E = mEE ‡E + mE ‡ + ⋯ + mE* ‡* + oEE ÖE + ⋯ + oE Ö

‡ = m E ‡E + m ‡ + ⋯ + m * ‡* + o E ÖE + ⋯ + o Ö
.

‡* = m*E ‡E + m* ‡ + ⋯ + m** ‡* + o*E ÖE + ⋯ + o* Ö

. 10.7

Where the coefficients m-Š and o-Š are constants that describe the system. This set of n equations
defines the derivatives of the state variables to be a weighted sum of the state variables and the system
inputs.

‡E mEE mE ⋯ mE* ‡E oEE oE ⋯ oE ÖE

A ‡ m E m ⋯ m * ‡ oE o ⋯ o Ö
T U =· ⋮ ¸T ⋮ U + · ¸T ⋮ U
A, ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮
‡* ⋯ m** ‡* o* Ö
10.8
m*E m* o*E o* ⋯

Or

U = ^U + Y 10.9

A is an n × n square matrix of the constant coefficients aij, and B is an n × r matrix of the coefficients
bij that weight the inputs38.

10.1.2. Output Equations

A system output is defined to be any system variable of interest. A description of a physical system in
terms of a set of state variables does not necessarily include all of the variables of direct engineering
interest. An important property of the linear state equation description is that all system variables may
be represented by a linear combination of the state variables xi and the system inputs ui. An arbitrary
output variable in a system of order n with r inputs may be written:

ˆE = :EE ‡E + :E ‡ + ⋯ + :E* ‡* + EE ÖE + ⋯+ E Ö
ˆ = : E ‡E + : ‡ + ⋯ + : * ‡* + E E + ⋯+
Ö Ö
.

ˆ' = :'E ‡E + :' ‡ + ⋯ + :'* ‡* + 'E ÖE + ⋯ + Ö

. 10.10
'

Or in matrix form
38
We use bold-faced type to denote vector quantities. Upper case letters are used to denote general matrices while lower case letters denote
column vectors.

97
 Electric Machines – Basic Theory

ˆE :EE :E ⋯ :E* ‡E EE E ⋯ E ÖE
ˆ :E : ⋯ :* ‡ ⋯ Ö
T ⋮ U=· ⋮ ⋮ ¸T ⋮ U + · E ¸T ⋮ U
⋮ ⋮ ⋮ ⋮ ⋮ ⋮
ˆ' ⋯ :'* ‡* Ö
10.11
:'E :' 'E ' ⋯ '

Or

T = _U + ðY 10.12

Where C is an m×n matrix of the constant coefficients cij that weight the state variables, and D is an
m × r matrix of the constant coefficients dij that weight the system inputs.

For many physical systems the matrix D is null, and Equation 10.12 can be expressed as:

T = _U 10.13

10.1.3. State Space Representation

The complete system model for a linear time-invariant system consists of:

• A set of n state equations, defined in terms of the matrices A and B.

• A set of output equations that relate any output variables of interest to the state variables and
inputs and expressed in terms of the C and D matrices.

The system can be written in the form:

U = ^U + Y
T = _U + ðY

The matrices A and B are properties of the system and are determined by the system structure and
elements. The output equation matrices C and D are determined by the particular choice of output
variables.

The block diagram representation corresponding to the above equations is shown in Figure 4.18.

Figure 10.2 Block diagram representation of the linear state-space equations39

39
https://en.wikipedia.org/wiki/User:Cburnett

98
 Electric Machines – Basic Theory

Let´s draw a block diagram for a single general second order, single input – single output system:

‡ mEE mE ‡E oE
R E S = úm m û ú‡ û + Ro S Ö(X)
‡ E
10.14

‡
ˆ(X) = j:E : l ú E û + Ö(X)
‡ 10.15

We will follow this process:

1. Draw n integrator (9 ¤E ; Í ) blocks and assign a state variable to the output of each block.
2. At the input to each block (which represents the derivative of its state variable) draw a
summing element.
3. Use the state equations to connect the state variables and inputs to the summing elements
through scaling operator blocks.
4. Expand the output equations and sum the state variables and inputs through a set of scaling
operators to form the components of the output.

The result is shown in Figure 10.3

Figure 10.3 Block diagram for a state-equation based second-order system (Equations 10.14 and
10.15).

As second example, we have an electromechanical system where we have written the equivalent
blocking drawing (Figure 10.4 (a))40.

Ve (s)

(a) (b)
Figure 10.4 Electromechanical system

40
Dr. Mostafa Abdel-geliel. “Control System I EE 411 – State Space Analysis”. Lecture 11.

99
 Electric Machines – Basic Theory

The equations that relate the different blocks will be as follows:

‡E = ‡
‡ + `‡ = w, ‡•
10.16

•+ − w} ‡ = f( ‡• + N( ‡•
10.17
10.18

And in the matrixial form:

‡E 0 1 0 ‡E 0
p‡ q = p0 −`/ w& ⁄ q p‡ q + p 0 q /+
‡• 0 −w} ⁄N( −f( ⁄N( ‡• 1⁄ N(
10.19

‡E
ˆ = 3 = ‡E = j1 0 0l p‡ q
‡•
10.20

Equation 10.19 has the form of U = ^U + Y and Equation 10.20 the form of T = _U + ðY.

10.1.4. Stability and natural response characteristics of a continuous time LTI system

Stability and natural response characteristics of a continuous-time LTI system (i.e., linear with
matrices that are constant with respect to time can be studied from the eigenvalues of matrix A). The
stability of a time-invariant state-space model can be determined by looking at the system's transfer
function in factored form. It will then look something like this:

Ð(0) = ÷
(c¤`G )(c¤`I )(c¤`S )
(c¤‚G )(c¤‚I )(c¤‚S )(c¤‚O )
10.21

The denominator of the transfer function is equal to the characteristic polynomial found by taking the
determinant of s I – A.

a(0)41 = |5v − ^| 10.22

The roots of this polynomial (the eigenvalues) are the system transfer function's poles (i.e., the
singularities where the transfer function's magnitude is unbounded). These poles can be used to
analyze whether the system is asymptotically stable or marginally stable. An alternative approach to

Lyapunov stability (if the solutions that start out near an equilibrium point ‡+ stay near ‡+ forever,
determining stability, which does not involve calculating eigenvalues, is to analyze the system's

then ‡+ is Lyapunov stable. More strongly, if ‡+ is Lyapunov stable and all solutions that start out
near ‡+ converge to ‡+ is asymptotically stable.

The zeros found in the numerator of Ð(0) can similarly be used to determine whether the system is the
minimum phase (In control theory and signal processing, a linear, time-invariant system is said to be
minimum-phase if the system and its inverse are causal and stable).

The system may still be input–output stable (see BIBO stable) even though it is not internally stable.
This may be the case if unstable poles are canceled out by zeros (i.e., if those singularities in the
transfer function are removable).

Example:

41
ln circuit analysis, the symbol a is commonly used to denote flux linkage. Here, we will use it to denote eigenvalues. Letters in bolt
indicate matrix or vector arrays. Kundur (1994).

100
 Electric Machines – Basic Theory

0 1 0
^=p 0 0 1q
−6 −11 −6
10.23

a −1 0
|bv − ^| = p0 a −1 q
6 11 a + 6
10.24

= a• + 6a + 11a + 6 10.25

= (a + 1)(a + 2)(a + 3) = 0 10.26

The eigenvalues of A are the roots of the characteristic equation, or -1, -2 and -3.

Software tools as Octave42, also provide the possibility of eigenvalues calculation.

42
Octave®. https://octave.sourceforge.io/.

101
 Electric Machines – Basic Theory

11
11.1. Wind Energy43
Chapter XI-WIND ENERGY

Wind energy is the kinetic energy of air in motion, also called wind. Without considering any
decrease in the wind velocity when it reaches the rotor, the total wind energy of a mass m through an
imaginary surface with half-opening the time t can be expressed with the following Equation:

= È/ = (K/Xc)/ = (KXc)/ •
E E E
11.1

Where

È = K/Xc: mass of air.

/: wind speed.
c: density of air.
K/X: the volume of air passing through area A (considered perpendicular to the direction of the wind).

As Power is energy per unit time, the wind power incident on the rotor area of a wind turbine is:

»= = (Kc)/ •
E
,
11.2

But because it is not possible to transform all the kinetic energy of the wind into mechanical energy,

a power ratio …‚ to the wind power equation (11.2). Theoretically its maximum value is 16/27, Burgos
the amount of energy that can be transferred from the wind to the blades comes it is done by applying

et al. (2001). So, to determine the mechanical energy or mechanical power captured under stationary
conditions, the following equation, Petersson (2005) is applied:

»' = c ∙ K ∙ …‚ (a, d) ∙ / •
E
11.2a
a=| 11.2b
)

Where:

…‚ (a, d) = power coefficient.

a = characteristic speed.
d = incident angle referred to the blades.
= radius of the circumference drawn by the tips of the blades.
| = angular speed of the low-speed shaft.

The mechanical power of a wind turbine that can be extracted highly depends on wind speed and with
the speed of the rotor in the generator, as it is shown in Figure 11.1. For each wind speed, there is
therefore an optimum turbine speed for wind power extraction.
43
Wikipedia

102
 Electric Machines – Basic Theory

Figure 11.1 Example of wind turbine power curves44

11.2. Wind Generators. Overview

11.2.1. Type I (SCIG) Wind turbine Generator

The first commercial wind turbines were constructed according to the so-called “Danish concept,”
which is depicted in Figure 11.2 (Type I SCIG). Here wind energy is transformed into electrical
energy using a simple (squirrel cage) induction machine directly connected to a three-phase collector
system (power grid).

Figure 11.2 Wind Turbine Generator Type I.

The operating principle of these turbines is the one developed in article AD-007 – Asynchronous
Machine Parameters’ Derivation. www.adneli.com. Here is a summary.

44
Christian Dufour, Jean Bélanger. “Real-Time Simulation of Doubly Fed Induction Generator for Wind Turbine Applications,” Opal-RT
Technologies inc. 2004.

103
 Electric Machines – Basic Theory

Figure 11.3 Schematic diagram of a three-phase induction machine stator circuit (left) and rotor one
(right)45.

Voltage Equations46-47:

/(c = (c 1(c +
ALÝ
A,
11.3
/}c = }c 1}c + A,
AL3
11.4
/.c = .c 1.c + A,
AL4
11.5
/( = ( 1( + A,
ALÝ^
11.6
/} = } 1} + A,
AL3^
11.7
/. = . 1. + A,
AL4^
11.8

The resistance of the stator and rotor are as follows:

(c = }c = .c = c
= = =
11.9
( } . 11.10

d-q voltage equations referred to the stator for type SCIG wind turbine are the same as the ones
developed for asynchronous machines and are as follows:

/Ac = c 1Ac − |c _¼c +

A\é
A,
11.11
/¼c = c 1¼c + |c ΨAc +
A\ê
A,
11.12

/′A = ′ 1′A − (|c − | )Ψ′¼ +

LYé^
A,
11.13
/′¼ = ′ 1′¼ + (|c − | )Ψ′A +
LYê^
A,
11.14

45
Kundur (1994).
46
AD-003, “Synchronous Machine Parameters’ Derivation Part 1.” Adneli Consultant, 2021. www.adneli.com.
47
Paul C. Krause, Oleg Wasynczuk, Scott D. Sudhoff. “Analysis of Electric Machinery and Drive Systems.” IEEE Press Power
Engineering Series.

104
 Electric Machines – Basic Theory

With the flux linkages transformed5_48-49:

ΨAc = NFc 1Ac + N' (1Ac + 1′A )

Ψ¼c = NFc 1¼c + N' $1¼c + 1′¼ %
11.15
11.16

Ψ′A = N′F 1′A + N' (1Ac + 1′A )

Ψ′¼ = N′F 1′¼ + N' $1¼c + 1′¼ %
11.17
11.18

N' = N'c
•
11.19

Figure 11.4 Equivalent circuit of a Type I (SCIG) Wind turbine Generator in pu values and referenced
to the stator.

From the circuit of Figure 11.4, we can determine the power transferred through the air gap:

»(- = ′
•Y^
)(‚ c
11.20

The electrical torque developed (three-phase) is given by

= 3 Zc ^ [ ′
‚ •Y
+ 11.21

Assuming no saturation, the magnetizing circuit can be considered open, then the current becomes:

′ =
d
]
Z• à ^ [àŠ(e àeY^ )
11.22

And Equation 11.21 can be expressed by the following relation:

= Zc ^ [ f
•‚ •Y dI
+ ]^ I
g
Z• à [ à(e àeY^ )I
11.23

Figure 11.5 shows the relation between torque + and the slip s.

48
A balance condition has been assumed.
49
Some authors as Kundur use a different nomenclature. E.g., Ncc = NFc + N' , N' = ? , 1A instead of 1′A , and N = N′F + N' , where
Nc = Ncc − N' is the stator leakage reactance and N' the magnetizing reactance, being the equations:
ΨAc = Ncc 1Ac + N' 1A
Ψ¼c = Ncc 1¼c + N' 1¼
ΨA = N 1A + N' 1Ac
Ψ¼ = N 1¼ + N' 1¼c

105
 Electric Machines – Basic Theory

Figure 11.5 Typical torque-slip characteristic of an induction machine50

We can observe that type I turbines have several handicaps and limitations:

The main characteristic of type I wind turbines is their operation at a fixed speed having a high
dependency of all the mechanical subcircuits (e.g., pitch). As the speed of wind is not constant, the
efficiency is low compared with other types of turbines with the capability to operate at variable speeds.
Some phenomena as a gust of wind when hits the turbine usually provoke a fast and strong variation of
the power output together with high mechanical stresses requirements. This causes need of a robust
mechanical design leading to an expensive turbine compared with other designs.

11.2.2. Type II (WRIG) Adjustable Speed Generators

As we can observe in Figure 11.1, a variable speed generator is more efficient than the one with fixed
speed, as turbine speed could be adjusted as a function of wind speed to maximize the output power.

There are other associated advantages as a minor cost with simpler pitch control, gusts of wind can be
absorbed by storing mechanical inertia of the turbine, reducing torque pulsations and so, incrementing
the Power Quality.

The variable-slip generator is essentially a wound-rotor induction generator with a variable resistor in
series with the rotor circuit as shown in Figure 11.6. Below rated wind speed and power, this acts just
like a conventional fixed-speed induction generator. Above rated, however, control of the resistance
effectively allows the air-gap torque to be controlled and the slip speed to vary, so that behavior is
then similar to that of a variable-speed system6.

Figure 11.7 shows the effect on the Torque when the rotor resistor f value is changed. Note that the
variable speed allows maximizing the power output under a variable wind speed.

50
Olimpo Anaya-Lara, Nick Jenkins, Janaka Ekanayake, Phill Cartwright, Mike Hughes. “WIND ENERGY GENERATION Modelling and
Control”.

106
 Electric Machines – Basic Theory

Figure 11.6 Variable speed generator

4 Motoring Region

3 8Rr
4Rr
2
2Rr
Torque (pu)

0
-1 -0,5 0 0,5 1
-1

-2
Generating Region
-3

-4
Slip (pu)

Figure 11.7 Torque–slip curves showing the effect of increased rotor circuit resistance

Equations that define this type of turbine are the same as the one used for the Type I (SCIG) Wind
turbine Generator.

11.2.3. Doubly Fed Induction Generator (DFIG)-based Wind Turbines2_8_51_52_53_54

The Doubly Fed Induction Generator (DFIG) is illustrated in Figure 11.8. It uses a wound-rotor
induction generator with slip-rings to transmit current between the converter and the rotor windings
and variable-speed operation is obtained by injecting a controllable voltage into the rotor at the
desired slip frequency (Holdsworth et al., 2003). The rotor winding is fed through a variable-

51
K. Kerrouche, A. Mezouar, Kh. Belgacema. “Decoupled Control of Doubly Fed Induction Generator by Vector Control for Wind Energy
Conversion System,” Department of Electrical Engineering, Faculty of Sciences Engineering, Moulay Tahar University, Saida 20000,
Algeria.
52
Dr John Fletcher and Jin Yang. “Introduction to Doubly-Fed Induction Generator for Wind Power Applications,” University of
Strathclyde, Glasgow United Kingdom.
53
Mohit Singh, Surya Santoso. “Dynamic Models for Wind Turbines and Wind Power Plants,” The University of Texas at Austin Austin,
Texas, 2008.
54
R. Pena, J.C. Clare, G. M. Asher. “A double fed induction generator using back-to-back PWM converters supplying an isolated load from
a variable speed wind turbine,” Researchgate, EE Proceedings - Electric Power Applications · October 1996.

107
 Electric Machines – Basic Theory

frequency power converter, typically based on two AC/DC IGBT-based voltage source converters
(VSCs), linked through a DC bus. The variable-frequency rotor supply from the converter enables the
rotor mechanical speed to be decoupled from the synchronous frequency of the electrical network,
thereby allowing variable-speed operation of the wind turbine. The generator and converters are
protected by voltage limits and an over-current crowbar.

Figure 11.8 Doubly Fed Induction Generator (DFIG)

Figure 11.9 Equivalent circuit of a Type III (DFIG) Wind turbine Generator in pu values and
referenced to the stator.

Figure 11.10 DFIG power relationships (Fox et al., 2007)

108
 Electric Machines – Basic Theory

From Figure 11.9, we can calculate the current ′ (current referenced to the stator) assuming a linear
circuit, that magnetizing impedance g' >>(fc + f′ ) + h(<c + <′ ).

d ¤Z ^ [
′ = (•
h]

à•Y^ /c)àŠ(e àeY^ )

11.24

power » can be calculated with the following equation:

Assuming the power relationships from Figure 11.10, and circuit from Figure 11.9 the rotor active

» = f‰ Z ′∗ [
dY^
c
11.25

′∗ = complex conjugate of the rotor current referenced to the stator

Neglecting stator and rotor losses, we have:

»(- ¤)(‚ = »c = »' − » 11.26

Equation 11.26 can be expressed in terms of the generated torque:

|c = | − »
» = −(|c − | )
11.27
11.28

Stator and rotor powers can be related through the slip:

» =− |c = −0»c
( ¤ ^)
11.29

And the mechanical power:

»' = »c + » = (1 − 0)»c 11.30

The power generated to the grid in steady state neglecting losses can be expressed as:

») = »' = »c + » = (1 − 0)»c 11.31

11.2.3.1. Rotor side converter control

The rotor-side converter (RSC) provides the excitation for the induction machine rotor using a
PWM converter that receives energy from the grid. Converter allows to control the torque hence the
speed of the DFIG and also the power factor at the stator terminals. The rotor-side converter provides

synchronous speed |c and the variable rotor speed | because of the wind speed conditions. In this
a varying excitation with frequency changes, and that change depending on the difference between the

way, the induction machine is controlled in a synchronously rotating dq-axis frame, with the d-axis
oriented along the stator-flux vector position (see Figure 11.12) in one common implementation. This
is called stator-flux orientation (SFO) vector control. Doing so, a decoupled control between the
electrical torque and the rotor excitation current is obtained. Consequently, the active power and
reactive power can be controlled independently from each other.

109
 Electric Machines – Basic Theory

Phase A
Current Phase Relationship in the stator windings
Phase B

Phase C
250
200
150
100
50
Amplitude

0
0 30 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570
-50
-100
-150
-200
-250
Degrees

Figure 11.11 Currents produced in the stator windings.

Figure 11.12 d-axis of a synchronously rotating qd0 frame, aligned with the total stator magnetic field
at the angle (time) indicated in Figure 11.11.

Because of the alignment, the following results are obtained:

_¼c = 0
_Ac = _c¤,~,(F = Ncc 1Ac + N' 1A = N' 1'c
11.32
11.33

Substituting Equations 11.32 and 11.33 into Equations 11.11 and 11.12, and assuming the winding
resistive element r is negligible, the following relationships are obtained:

110
 Electric Machines – Basic Theory

•¼c = |_Ac = |_c¤,~,(F

•Ac = 0
11.34
11.35

The active power and reactive power are controlled independently using of 1¼ and 1A respectively.
This is the vector control we will use for the rotor. See Figure 11.15.

Some authors use q-axis aligned with the total stator magnetic field at the angle (time), being the
equations:

_¼c = _c¤,~,(F = Ncc 1¼c + N' 1¼

_Ac = 0
11.36
11.37

In this case, the active power and reactive power are controlled independently by means of 1A and 1¼
respectively

Equations 11.34 and 11.35 suggest that the speed voltage •¼c is time-invariant and constant, and the
voltage across the stator d-axis is negligible; then, the stator q and d-axis currents can be controlled

11.15 and 11.16, and with _¼c = 0, the stator q- and d-axis currents are given by:
using the rotor q- and d-axis currents. Using the stator and rotor flux equations given in Equations

1¼c = Z− ì [ 1′¼
ìO
àìO
11.38
H

1Ac = Z éì àìO é^ [
L ¤ì -Y
11.39
H O

Defining the following equations:

D =1−
ìIO
ì ì^^
leakage factor 11.40
And
N = ìO equivalent inductance
ìI
11.41

The rotor voltages and flux equations are scaled to be numerically equal to the AC per phase values

/A = f 1A + DN − (|c − | )DN 1¼
A-é^
A,
11.42
/¼ = f 1¼ + DN − (|c − | )(N 1'c + DN 1A )
A-ê^
A,
11.43

_A = N 1'c + DN 1A
_¼ = DN 1¼
11.44
11.45

The stator flux angle is calculated related to two equations:

_c› = Í(/c› − fc 1c› ) X

_cœ = Í$/cœ − fc 1cœ % X
11.46
11.47
\ i
3c = Xm2¤E Z [
\j
11.48

As the inductance and flux quantities in Equations 11.38 and 11.39 are time-invariant, the stator q-
and d-axis currents can be controlled by adjusting the rotor q- and d-axis currents appropriately.

The real and reactive power in the stator winding is given by the following equations:

111
 Electric Machines – Basic Theory

»c = $/Ac 1Ac + /¼c 1¼c %

•
11.49
= $/Ac 1¼c − /¼c 1Ac %
•
c 11.50

Since the stator d-axis voltage is zero as derived in Equation 11.35, Equations 11.49 and 11.50 can be
simplified as follows:

»c = $/¼c 1¼c %
•
11.51
= − $/¼c 1Ac %
•
c 11.52

Combining with Equations 11.38 and 11.38:

[ 1′
\é NÈ
»c = − Z
•
N 0 +NÈ
11.53

[ |Ψ
Ψ 0 −NÈ 1′
=− Z
•
c N 0 +NÈ 0 11.54

Using Equations 11.53 and 11.54, we can generate the d- and q-axis rotor reference currents by using
proportional integral controllers based on the difference between the measured and desired power
quantities. Figure 11.13 illustrates how the reference rotor q-and d-axis currents are generated.

Figure 11.13 Rotor reference currents generated using a proportional integral controller (based on the
difference between measured and desired quantities).

Figure 11.15 (below) shows a schematic block diagram for the rotor-side converter control.

11.2.3.2. Grid-side converter control

The grid-side converter controls the flow of real and reactive power to the grid, through the grid
interfacing inductance of the transformer. The objective of the grid-side converter is to keep the DC-
link voltage constant regardless of the magnitude and direction of the rotor power.

The vector control method is used as well, with a reference frame oriented along with the stator
voltage vector position, enabling independent control of the active and reactive power flowing
between the grid and the converter. Figure 11.15 (upper side) shows the schematic control process
where we can observe that the d-axis current is used to regulate the DC-link voltage and the q-axis
current to regulate the reactive component in the PWM.

112
 Electric Machines – Basic Theory

Figure 11.14 AC/DC Converter.

The voltage equations to analyze this control are as follows:

/(}. = f1(}. + N + /(}.F

A-Ý34
A,
11.55

Transforming into a d-q reference frame rotating at the grid frequency |c

0 −|c
/A¼ = f1A¼ + N +NR S 1A¼ + /A¼F
A-éê
A, |c 0
11.56

/A = f1A + N − |c N1¼ + /AF

A-é
A,
11.57
/¼ = f1¼ + N + |c N1A + /¼F
A-ê
A,
11.58

Aligning the d-axis reference frame on /A makes /¼ = 0. Also, since the supply voltage has constant
amplitude, /A also has constant amplitude.

The active and reactive power flows

» = 3$/A 1A + /¼ 1¼ % = 3/A 1A
= 3$/A 1¼ + /¼ 1A % = 3/A 1¼
11.59
11.60

1A and 1¼ components of current respectively.

Which demonstrates that the real and active powers from the grid-side converter are controlled by the

The angular position of the supply voltage is computed as

3c = Í |c X = Xm2¤E Z [
)i
)
11.61
j

Where /› and /œ are the α and Clarke components of the stator voltage. Neglecting harmonics, the
active power balance at the DC-link can be expressed by the following equation:

•åk 1~c = 3/A 1A 11.62

113
 Electric Machines – Basic Theory

Figure 11.15 Vector control of a Type III (DFIG) Wind turbine Generator.

11.2.4. Full Converter Generator (IG)-based Wind Turbines50

This type of wind turbine may or may not have a gearbox and a wide range of electrical generator
types such as asynchronous, conventional synchronous and permanent magnet can be employed.

As all the power from the wind turbine is transferred through the power converter, the specific
characteristics and dynamics of the electrical generator are effectively isolated from the power
network (Fox et al., 2007). Hence the electrical frequency of the generator may vary as the wind
speed changes, while the network frequency remains unchanged, permitting variable-speed operation.
The rating of the power converter in this wind turbine corresponds to the rated power of the generator.

114
 Electric Machines – Basic Theory

Figure 11.16 Full Converter Wind Generator.

11.2.4.1. Control and dynamic performance assessment

11.2.4.1.1. Generator-side Converter Control

Load Angle Control

The load angle control strategy employs steady-state power flow equations (Kundur, 1994; Fox et al.,
2007) to determine the transfer of active and reactive power between the generator and the DC link.

Figure 11.17 Load Angle Control relationship.

In Figure 11.17 the active and reactive power flows in steady state can be defined as:

»= 012Ã)
l dÞ
el
11.63

= :;0Ã)
l ¤ l dÞ
I

el
11.64

115
 Electric Machines – Basic Theory

The operation of the generator and the power transferred from the generator to the DC link are

side converter. The magnitude, Vt, and angle, Ã) , required at the terminal of the generator-side
controlled by adjusting the magnitude and angle of the voltage at the AC terminals of the generator-

converter are calculated using Equations 11.65 and 11.66 assuming Ã) small, so 012Ã) ≈ Ã) and
:;0Ã) ≈ 1

Ã) =
$l_^#! el
l dÞ
11.65

•, = −
æl_^#! el
) 11.66
l

Where »)_ +* is the reference value of the active power that needs to be transferred from the generator
to the DC link and )_ +* is the reference value for the reactive power.

The reference value »)_ +* is obtained from the maximum power extraction curve (Figure 11.1) for a
given generator speed, | , combined with an external dispatch center that determines the maximum
power to be delivered in function of the demand needs.

the reactive power reference value can be set to zero, )_ +* = 0 (i.e., •, and ) are equal in
magnitude) in case of a generator with permanent magnets excitation. In case of electrically excited
rotors, )_ +* is given by the demand needs and generator limits.

Figure 11.18 Load Angle Control.

Vector Control

Vector control techniques are implemented based on the dynamic model of the synchronous generator
expressed in the d-q frame. As in the case of the DIFG generators, the d-q frame is defined as the d
axis aligned with the magnetic axis of the rotor (field).

116
 Electric Machines – Basic Theory

For the vector control usually W 5_ø5V ( )_ +* = 0) is se to zero and WÄ5_ø5V is derived from the
equation

£5 = m 5 WÄ5 − mÄ5 W 5
£5 = m 5 WÄ5 (W 5 = Å)
11.67
11.68

Being

m 5 = −Q 5 W 5 + Qõ WV (WV = /‰:X; ;``1‰ :Ö ‰2X) 11.69

With W 5 = Å, m 5 = Qõ WV 11.70

Where:

Q 5 = Q 5 + Qõ
Q 5 = leakage inductance.
Qõ = mutual inductance in the d-axis.

Once the reference currents W 5_ø5V and WÄ5_ø5V, have been determined from the controller, the
corresponding voltage magnitudes can be calculated using the equations:

! 5 = −ò5 W 5 + =Ä5 WÄ5

!Ä5 = −ò5 WÄ5 + = 5 W 5 +
11.71
V 11.72

Defining mV = Qõ WV Equation 11.68 becomes £5 = mV WÄ5

For a given torque reference £5« , WÄ5_ø5V = £5« /mV we can develop the control system depicted in
Figure 11.19.

Figure 11.19 Vector Control.

117
 Electric Machines – Basic Theory

11.2.4.1.2. DC-Link

The DC-link voltage, •åk , can be expressed in terms of the generator output power, ») , and the
power transmitted to the grid, »*+, as

•åk = Á Í$») − »*+, % X 11.73

k

The reference value of the active power, »*+,_ +* , to be transmitted to the network is calculated by
comparing the actual DC link voltage, •åk , with the desired DC link voltage reference, •åk_ +* . A
similar scheme as the one shown in Figure 11.15 (upper side) can be used to control the active and
reactive power delivered to the network.

118
 Electric Machines – Basic Theory

12
12.1.
Chapter XII-PHOTOVOLTAIC
SYSTEMS AND BATTERIES

Photovoltaic Systems55-56

A photovoltaic system makes use of one or more solar panel electricity. It consists of various
components which include the photovoltaic modules, mechanical and electrical connections and
mountings and means of regulating and/or modifying the electrical output.

The photovoltaic system can generate direct current electricity without environmental impact when is
exposed to sunlight. The basic building block of PV arrays is the solar cell, which is basically a p-n
junction that directly converts light energy into electricity. The output characteristic of the PV module
depends on the cell temperature, solar irradiation, and output voltage of the module. The figure shows
the equivalent circuit of a PV array with a load.

General Symbols

'‚‚
ó'‚‚ - voltage at maximum power, V
- current at maximum power, A

A : diode current.
c : reverse bias saturation current (or scale current).
•& : thermal voltage V.
k : Boltzmann's constant = (1.3806488×10−23), J.K-1
A : linearity factor (1 for ideal diode). Sometimes known as emission coefficient.
q : elementary charge = (1.602176565×10−19), C.

•‚) : Terminal voltage of PV module.

T : p-n junction absolute temperature, K.

‚) : Output current of PV module.

‚0 :
c. : Cell's short-circuit current.
Light generated current or photocurrent.

ó~. : Open circuit voltage

PV Systems

'( - mean annual solar radiation, kWh/m2

‚ - estimated peak energy delivered, kWh
k ‚ - nominal peak energy, kW
n - number of modules
n, - temperature derating factor
»'(e - maximum power, W
. - temperature of the PV cell, K

55
Wikipedia
56
Lipsa Priyadarshanee, “MODELING AND CONTROL OF HYBRID AC/DC MICROGRID,” Thesis of Department of Electrical
Engineering National Institute of Technology Rourkela-769008, India

119
 Electric Machines – Basic Theory

.- temperature of the PV cell, K

c,. - temperature of STC, 25 °C, K
‚ - peak energy of a single module, W
η - efficiency of the system
ϒ - power temperature coefficient, °C-1

Figure 12.1 Equivalent circuit of a solar cell.

From Figure 12.1:

‚) = ‚0 − A − c0 load current in amperes 12.1

óc0 = •‚) − ‚) fc voltage across the shunt branches 12.2

c0 = óc0 /f‚ 12.3

The current through the diode is given by Shockley´s equation

= − 1r
p q

c o‰
íhA
A 12.4
•& = w / 12.5

The voltage-current characteristic equation of a solar cell can be expressed by the following equation:

ê$hst Ûust %
= − ô‰ − 1ö −
• € $dst à st • %
í. .A
‚) ‚0 c •s
12.6

= − ô‰ − 1ö −
• €
p q
ÙhA
v q
‚) ‚0 c •s
12.7

conditions ‚) = 0, and the equation reduces to:

At the limit, we can determine the open circuit voltage and short circuit current. During open

120
 Electric Machines – Basic Theory

• w4 €
0= ‚0 − c ô‰ − 1ö −
p
í.hA vw4
•s
12.8

Typically, f‚ is very high compared with ó~. and the last term can be neglected, so Equation 12.8
reduces to:

ó~. = K. •& 2 ú sq + 1û 12.9

Similarly for the short circuit current, we can set •‚) = 0, giving:

= ‚0 − c ô‰ − 1ö −
• 4 € 4•
u
í.hA
c. •s
12.10

As f‚ is much higher than fc and c is small compared with ‚) , we can conclude:

c. ≈ ‚) 12.11

The series resistance (fc ), shunt resistance (f‚ ) and reverse saturation current ( c ) are dependent on
the area of the PV cell. Generally, the bigger the cell the larger c (bigger diode junction area) and the
lower fc and f‚ will be.

The characteristic equation can be used to evaluate the effect of various parameters on the
performance of the PV cell or module:

• temperature (T) - affects the cell by being part of the exponential term and the value of the
reverse saturation voltage. As the temperature increases, while the exponential will decrease the
reverse saturation voltage will increase exponentially. The next effect is to reduce the open circuit
voltage of the cell. Typically, the voltage will decrease by 0.35 to 0.5% for each degree increase in

• series resistance (fc ) - increasing has a similar effect to temperature in that the open circuit
temperature.

voltage will start to drop. Very high values of fc will in addition reduce the available short circuit

• shunt resistance (f‚ ) - decreasing will provide a greater path for the shunt current, again
current.

lowering the cell voltage.

12.1.1. Maximum power point tracking

A PV module's I-V curve can be generated from the equivalent circuit Integral to the generation of tie
I-V curve is the current ‚0 , generated by each PV cell. The cell current is dependent on the amount
of light energy (irradiance) falling on the PV cell and the cell's temperature.

As the irradiance decreases not only is the amount of power reduce, but the peak power point moves
to the left. Similarly, as the temperature of the cell increases, the power output lowers, and the
maximum power point again shifts to the left. With the maximum power point being a variable
quantity, dependent on the solar irradiance and cell temperature, modern inverters have mechanisms
to track this and always deliver the maximum possible power from a PV cell. This is called maximum
power point tracking (MPPT).

As an electronic system maximum power point tracker (MPPT) functions the photovoltaic (PV)
modules in a way that allows the PV modules to produce all the power they are capable of. Since
MPPT is a fully electronic system, it varies the module’s operating point so that the modules will be

121
 Electric Machines – Basic Theory

able to deliver maximum available power. As the outputs of the PV system are dependent on the
temperature, irradiation, and load characteristics MPPT cannot deliver the output voltage perfectly.
For this reason MPPT is required to be implemented in the PV system to maximize the PV array
output voltage57.

Figure 12.2 MPP characteristics.

12.2. Battery

In our modern society the role of batteries is important as energy carriers, because of its presence in
devices for everyday use. At the end of the 20th century the demand for batteries rapidly increased
due to the large interest in wireless devices. Today, the battery industry comes under the category of
the large-scale industry which produces several million batteries per month. Improving the energy
capacity is one major development issue, however, for consumer products, safety is probably
considered equally important today. With the introduction of hybrid electric vehicles into the market
there is technological development in the battery field which leads to reduction of fuel consumption
and gas emissions. Battery development is a major task for both industry and academic research.57

12.2.1. Modeling of a Battery

The battery is modeled as a nonlinear voltage source whose output voltage depends not only on the
current but also on the battery state of charge (SOC), which is a nonlinear function of the current and
time57. Figure 12.3 represents a basic model of the battery. Two parameters to represent the state of a
battery i.e., terminal voltage and state of charge can be written as:

•} = • − w + K‰ (¤D.Í -,.A,)
æ
æ¤Í -,.A,
12.12

9x… = 100 Z1 + [
Í -,.A,
æ
12.13

57
O. Tremblay, L. A. Dessaint, and A. I. Dekkiche, “A generic battery model for the dynamic simulation of hybrid electric vehicles,” in
Proc. IEEE Veh. Power propulsion Conf., pp. 284-289, 2007.

122
 Electric Machines – Basic Theory

Figure 12.3 Model of a Battery

•}(,, = • − f. 1 12.14

•} = no-load voltage (V)

• = battery constant voltage (V)
K = polarisation voltage (V)

Í 1X = actual battery charge (Ah)

Q = battery capacity (Ah)

A = exponential zone amplitude (V)

•}(,, = battery voltage (V)

B = exponential zone time constant inverse (Ah)−1

1 = battery current (A).

R = internal resistance (Ω)

The original Shepherd model has a non-linear term equal to w

æ
æ¤Í -,.A,
. This term represents a non-
linear voltage that changes with the amplitude of the current and the actual charge of the battery. So,
when there is a complete discharge of the battery and no flow of current, the voltage of the battery
will be nearly zero. As soon as a current circulates again, the voltage falls abruptly. This model yields
accurate results and also represents the behavior of the battery57.

123
 Electric Machines – Basic Theory

Disclaimer

This book has been conceived to show the fundamental concepts and results of experiences of
electrical engineering to bring engineers and designers the most complete and up-to-date Knowledge
reference for his use as an aid in design and application of Protection, Control, and Monitoring
equipment together with its adequate Management, to obtain the maximum benefit in its use. It
pretends to be as well a guiding the engineers in those phases that cannot be solved by formulas but
rather depend upon their judgment. Obviously, as in any other discipline, the application and use of
the equipment is the sole responsibility of the user, requiring a high degree of engineering skill and
economic sense.

While there are suggestions made in this book regarding theoretical concepts, it is not intended to
replace competent engineering criteria for each application. These suggestions, however, are not
based on theory alone; they are also based on actual technologies, design rules, installation,
operating, and maintenance experience. This book does not cover or try to cover, all variations,
which may be encountered in the installation, operation, and maintenance of the related equipment.

Should the user encounter variations not addressed, it be appreciated that the matter is reported to
the author.

The author and representatives shall have no liability for any claim of any kind, including without
limitation negligence, for any loss or damage, including without limitation special incidental, indirect
or consequential damages from the use of the recommendations herein included or from a failure of
them to perform as intended.

The author gives no warranty, either expressed or implied, or assurance about the contents or the
performance of material and software used as a reference in the different Chapters of the book of this
publication, nor does it accept any responsibility, as to the accuracy or exhaustiveness of the
information. All implied warranties and conditions are excluded to the maximum extent permitted by
law.

Adneli Consultant S.L. reserves the right to modify, change and update the technical data and
information contained in this book without notice.

We will appreciate any comment on this book: content and presentation; and also, suggestions for
further improvements. Be free to send an email to: [email protected]

124