Induction Generator And AC/DC/AC Converter Model

For Wind Energy Conversion

Amir Husin Tanady

(09884986)

A thesis submitted for the partial fulfillment of the degree

of Bachelor of Engineering (Electrical)

Induction Generator and AC/DC/AC Converter Model for Wind
Energy Conversion

Tanady

Amir Husin

26th October 2001 Dr. W. W. L. Keerthipala

Bachelor Of Engineering Electrical

This report documents the analysis and simulation of a self-

excited induction generator and the conversion scheme
(AC/DC/AC converter) using PSCAD/EMTDC simulation
software. The report basically separated into several parts. First
two sections give a brief introduction and the literature review
the project. Thirdly, the modelling of the self-excited induction
generator and the AC/DC/AC converter was evaluated and
analysed. Next, the recommendations on the future work were
given. And finally, the conclusion is made.

Induction Generator; Self-Excitation; Power Electronic

Converters; AC/DC/AC Converter; Wind Driven Electrical
Generators; Wind Energy Conversion; Renewable Energy;
PSCAD; EMTDC
Mr. Amir Husin Tanady
U31/ 132 Mounts Bay Road
West Perth – 6005
Western Australia

26th October 2001

Professor J. L. Hullett
School of Electrical and Computer Engineering
Curtin University of Technology
P. O. Box U1987
Perth, W. A. 6000

Dear Sir,

Re: Final Dissertation For Final Year Project

Please find the enclosed report on the project titled “Induction Generator and
AC/DC/AC Converter Model for Wind Energy Conversion”, in partial fulfilment of
the requirements of the Electrical Project unit for the award of Bachelor of
Engineering (Electrical) – Honours.

Yours Sincerely

Amir Husin Tanady

Student No.: 09884986
EXECUTIVE SUMMARY

This report elaborates the analysis and simulation of a self-excited induction

generator and the conversion scheme (AC/DC/AC converter) using

PSCAD/EMTDC simulation software.

The report basically separated into several parts. Firstly, a general introduction of

wind power, wind turbine, and the PSCAD/EMTDC software was given. Secondly,

the background information including the wind/diesel hybrid system, the Australian

wind farms, the potential of offshore wind farms, the wind turbine generators in

particular the analysis of the induction machine behaviour, and past investigation

that had been conducted by previous students as well as other research papers was

summarised. Thirdly, the modelling of the self-excited induction generator and the

AC/DC/AC converter was evaluated and analysed. Next, the recommendations on

the future work were given. And finally, the conclusion is made.

-i-
ACKNOWLEDGEMENTS

The implementation of this project was possible with the help and advices from

many people. The project could not have reached at this stage without the assistance

of these people.

Firstly, I would like to thank the project supervisor Dr. W. W. L. Keerthipala for his

support, guidance, patience, motivation, and advice throughout the project. Many

thanks to Mr. Z. Cielma and M. Fowler for their kindness and support on the

technical aspect of the software simulation, induction machine operation and the

experiment on the induction machine in the laboratory.

Finally, many thanks to all my friends and everyone for giving me their valuable

opinions and support throughout the project.

- ii -
 TABLE OF CONTENTS

Executive Summary .................................................................................................... i

Acknowledgements.................................................................................................... ii

Table of Contents...................................................................................................... iii

Nomenclature............................................................................................................ vi

List of Figures .......................................................................................................... vii

List of Tables ............................................................................................................ xi

1. Introduction............................................................................................................ 1

1.1 Power From the Wind.......................................................................................1

1.2 Wind Turbine ....................................................................................................3

1.3 PSCAD/EMTDC Software ...............................................................................6

2. Background Information........................................................................................ 7

2.1 Wind – Diesel Systems .....................................................................................7

2.2 Wind Farms in Australia...................................................................................8

2.3 Offshore Windfarms .........................................................................................9

2.4 Wind Turbine Generators .................................................................................9

2.4.1 Why Induction Generator ........................................................................10

2.4.2 Induction Machine Analysis ....................................................................12

2.5 Power Electronics ...........................................................................................15

2.5.1 AC to DC Conversion..............................................................................15

2.5.2 DC to AC Conversion..............................................................................22

2.5.3 DC – DC Converter .................................................................................23

2.6 Past Investigations ..........................................................................................26

2.6.1 Simulation of Wind-Diesel Hybrid Systems............................................26

2.6.2 Self-excited Induction Generator.............................................................28

- iii -
 2.6.3 Problem Formulation ...............................................................................30

3. Modelling of the System...................................................................................... 32

3.1 Self-Excited Induction Generator Model........................................................32

3.2 AC/DC/AC Converter Model .........................................................................41

3.2.1 AC to DC Conversion..............................................................................41

3.2.2 DC to DC Conversion Using Buck-Boost Converter ..............................49

3.2.3 DC to AC Conversion Using PWM VSI .................................................52

3.2.4 AC/DC/AC Converter..............................................................................59

4. Results Of The Complete System ........................................................................ 71

4.1 System 1..........................................................................................................71

4.2 System 2..........................................................................................................75

4.3 System 3..........................................................................................................78

5. Future Work ......................................................................................................... 82

6. Conclusion ........................................................................................................... 85

References

Appendix A - PSCAD

Appendix B – Test Machine

Appendix C – Diode Rectifier

Appendix D – Half-Controlled Rectifier

Appendix E – Buck-Boost Converter

Appendix F – PWM VSI

Appendix G – Diode Rectifier with PWM VSI

Appendix H – Half-Controlled Rectifier with PWM VSI

Appendix I – Diode Rectifier, Buck-Boost Converter, and PWM VSI

Appendix J – Complete System 1 Design

- iv -
Appendix K – Complete System 2 Design

Appendix L – Complete System 3 Design

-v-
NOMENCLATURE

FFT – Fast Fourier Transform

GTO Thyristors – Gate-Turn-Off Thyristors

IGBT – Insulated Gate Bipolar Transistors

PSCAD – Power System Computer Aided Design

PWM – Pulse Width Modulated

RMS – Root Mean Square

SEIG – Self-Excited Induction Generator

- vi -
 LIST OF FIGURES

Fig. 1.1: (a) Main Components of Horizontal-axis Wind Turbine ............................ 5

Fig. 1.1: (b) Cross-section of a Typical Grid-connected Wind Turbine.................... 5

Fig. 1.1: (c) Cross-section of a Nacelle in A Grid-connected Wind Turbine [14] .... 5

Fig. 2.1: Torque vs Speed Characteristics of Squirrel-cage Induction

Generator [3]............................................................................................. 12

Fig. 2.2: Per-Phase Equivalent Circuit of An Induction Machine ........................... 13

Fig. 2.3: Alternative Form for Per-Phase Equivalent Circuit .................................. 14

Fig. 2.4: Power Flow Diagram................................................................................. 14

Fig. 2.5: 3-φ Diode Bridge Rectifier ........................................................................ 16

Fig. 2.6: 3-φ Rectifier Circuit With a Constant DC Current (Ls = 0) ...................... 17

Fig. 2.7: Rectified Voltage Waveform After Diode Bridge..................................... 17

Fig. 2.8: Input AC Line Current of 3-φ Diode Rectifier .......................................... 18

Fig. 2.9: Controlled 6-Pulse Full-Bridge Rectifier .................................................. 20

Fig. 2.10: Effect of α on Vd ..................................................................................... 20

Fig. 2.11: 3-φ Inverter Circuit Diagram ................................................................... 22

Fig. 2.12: Control Signal for DC-DC Converter...................................................... 24

Fig. 2.13: Buck-Boost Converter Circuit Diagram.................................................. 25

Fig. 2.14: Cúk Converter Circuit Diagram .............................................................. 25

Fig. 2.15: Proposed System ..................................................................................... 31

Fig. 3.1: Induction Generator Parameters ................................................................ 33

Fig. 3.2: SEIG Model in PSCAD............................................................................. 34

Fig. 3.7: Simulation Results of Self-Excited Induction Generator Parameters ....... 36

Fig. 3.3: RMS Stator Line Voltage .......................................................................... 37

Fig. 3.4: Stator Frequency........................................................................................ 38

- vii -
Fig. 3.5: RMS Stator Voltage At Higher Compensation Levels.............................. 38

Fig. 3.6: Stator Frequency At Higher Compensation Levels................................... 39

Fig. 3.7: PSCAD Model of Diode Bridge Rectifier................................................. 42

Fig. 3.8: Voltage and Current At Rectifier Output .................................................. 42

Fig. 3.9: Half-Controlled 6-Pulse Full-Bridge Rectifier Circuit.............................. 43

Fig. 3.10: Thyristor Firing Angle (α) Circuit .......................................................... 44

Fig. 3.11: Alpha Train Pulse Signal Generator Circuit Diagram............................. 45

Fig. 3.12 Comparator Outputs and AND Gate Output For Alpha ........................... 46

Fig. 3.12: Thyristor Train Pulse Waveform............................................................. 47

Fig. 3.13: T3 and T5 Firing Angle Circuit............................................................... 47

Fig. 3.14: VPn, VNn, and Vd Waveforms................................................................... 48

Fig. 3.15: Buck-Boost Converter Circuit Diagram.................................................. 49

Fig. 3.16: Firing Angle For GTO............................................................................. 50

Fig. 3.17: Gating Signal For GTO ........................................................................... 50

Fig. 3.17: Input Voltage Vd and Output Voltage Vo of Buck Boost Converter...... 51

Fig. 3.18: PWM VSI Supplying Resistive Load...................................................... 52

Fig. 3.19: PWM Circuit ........................................................................................... 53

Fig. 3.20: PWM Pattern For IGBT Switching ......................................................... 53

Fig. 3.21: PWM Patterns in One Leg of VSI........................................................... 54

Fig. 3.22: Inverter Voltage Output With Respect To Zero Potential Between

Capacitors on DC Side ............................................................................ 56

Fig. 3.23: Inverter Phase Voltage Output With Respect To DC Neutral................. 56

Fig. 3.24: Inverter Phase Voltage Output With Respect To Neutral Ground .......... 57

Fig. 3.25: FFT of PWM VSI Output Phase Voltage................................................ 58

Fig. 3.27: Rectifier Output Voltage and Current of 1st Design................................ 60

- viii -
Fig. 3.28: Inverter Output Vinva and Vlinvab of 1st Design ......................................... 61

Fig. 3.29: PSIMDemo Model of Diode Rectifier With PWM VSI ......................... 62

Fig. 3.30: FFT of Output Phase Voltage Of Diode Rectifier PWM VSI

Converter Design..................................................................................... 62

Fig. 3.31: Half-Controlled Rectifier With PWM VSI Converter Design ................ 63

Fig. 3. 32: Half-Control Rectifier Output Voltage and Current Of 2nd Design ....... 64

Fig. 3.33: Inverter Output Phase Voltage and Line-To-Line Voltage of 2nd

Design...................................................................................................... 64

Fig. 3.34: FFT of Inverter Output Phase Voltage Of 2nd Design............................. 65

Fig. 3.35: Diode Rectifier With Buck-Boost Converter and PWM VSI

Converter Design (3rd Design)................................................................. 65

Fig. 3.36: (a) Output Voltage Of Diode Rectifier; (b) Output Voltage Of Buck-

Boost Converter....................................................................................... 66

Fig. 3.37: Inverter Output Phase Voltage And Line-Line Voltage Waveform

Of 3rd Converter Design .......................................................................... 67

Fig. 3.38: Buck-Boost Converter Conventional Current Flow Diagram ................. 67

Fig. 3.39: Current Flows In Buck-Boost Converter................................................. 68

Fig. 3.40: Modified Buck-Boost Converter ............................................................. 68

Fig. 3.41: Current Flows in Modified Buck-Boost Converter ................................. 69

Fig. 3.42: Inverter Output Phase Voltage And Line-Line Voltage Of 3rd

Converter Design..................................................................................... 70

Fig. 4.1: Complete System 1 Circuit Diagram......................................................... 72

Fig. 4.2: RMS Voltage Of SEIG.............................................................................. 72

Fig. 4.3: Rectifier Output Voltage Of System 1 ...................................................... 73

Fig. 4.4: Inverter Output Phase Voltage And Line-to-Line Voltage Of System 1 .. 73

- ix -
Fig. 4.5: FFT of Inverter Output Phase Voltage Of System 1 ................................. 74

Fig. 4.6: Complete System 2 Circuit Diagram......................................................... 75

Fig. 4.7: Modified Alpha Circuit ............................................................................. 76

Fig. 4.8: Half-Control Rectifier Output Voltage Ripple Of System 2 ..................... 76

Fig. 4.9: Inverter Output Line-To-Line Voltage Of System 2................................. 77

Fig. 4.10: FFT Of Inverter Output Phase Voltage OF System 2 ............................. 77

Fig. 4.11: Complete System 3 Circuit Diagram....................................................... 78

Fig. 4.12: Rectifier Output And Buck-Boost Converter Output Voltages Of

System 3 .................................................................................................. 80

Fig. 4.13: Inverter Output Line-To-Line Voltage Of System 3 ............................... 80

Fig. 4.14: FFT Of Inverter Output Phase Voltage Of System 3 .............................. 81

Fig. 5.1: Blanking Time In PWM VSI..................................................................... 83

-x-
 LIST OF TABLES

Table 3.1: Induction Machine Parameters ................................................................32

Table 3.2: Induction Machine Parameters (Per-Unit Values)...................................33

Table 3.3: Maximum Overvoltages Magnitude ........................................................39

- xi -
1. INTRODUCTION

Wind is abundant almost in any part of the world. Its existence in nature caused by

uneven heating on the surface of the earth as well as the earth’s rotation means that

the wind resources will always be available.

The conventional ways of generating electricity using non renewable resources such

as coal, natural gas, oil and so on, have great impacts on the environment as it

contributes vast quantities of carbon dioxide to the earth’s atmosphere which in turn

will cause the temperature of the earth’s surface to increase, known as the green

house effect [2]. Hence, with the advances in science and technology, ways of

generating electricity using renewable energy resources such as the wind are

developed. Nowadays, the cost of wind power that is connected to the grid is as

cheap as the cost of generating electricity using coal and oil. Thus, the increasing

popularity of green electricity means the demand of electricity produced by using

non renewable energy is also increased accordingly.

1.1 Power From the Wind

Kinetic energy from the wind is used to turn the generator inside the wind turbine to

produced electricity. There are several factors that contribute to the efficiency of the

wind turbine in extracting the power from the wind. Firstly, the wind speed is one of

the important factors in determining how much power can be extracted from the

wind. This is because the power produced from the wind turbine is a function of the

-1-
cubed of the wind speed. Thus, the wind speed if doubled, the power produced will

be increased by eight times the original power. Then, location of the wind farm plays

an important role in order for the wind turbine to extract the most available power

form the wind.

The next important factor of the wind turbine is the rotor blade. The rotor blades

length of the wind turbine is one of the important aspects of the wind turbine since

the power produced from the wind is also proportional to the swept area of the rotor

blades ie. the square of the diameter of the swept area. Hence, by doubling the

diameter of the swept area, the power produced will be four fold increased. It is

required for the rotor blades to be strong and light and durable [2]. As the blade

length increases, these qualities of the rotor blades become more elusive. But with

the recent advances in fiberglass and carbon-fibre technology, the production of

lightweight and strong rotor blades between 20 to 30 meters long is possible. Wind

turbines with the size of these rotor blades are capable to produce up to 1 megawatt

of power.

The relationship between the power produced by the wind source and the velocity of

the wind and the rotor blades swept diameter is shown below.

π 2 3
Pwind = dD v wind
8

The derivation to this formula can be looked up in [2]. It should be noted that some

books derived the formula in terms of the swept area of the rotor blades (A) and the

air density is denoted as ρ.

-2-
Thus, in selecting wind turbine available in the market, the best and efficient wind

turbine is the one that can make the best use of the available kinetic energy of the

wind.

Wind power has the following advantages over the traditional power plants.

• Improving price competitiveness,

• Modular installation,

• Rapid construction,

• Complementary generation,

• Improved system reliability, and

• Non-polluting.

1.2 Wind Turbine

There are two types of wind turbine in relation to their rotor settings. They are:

• Horizontal-axis rotors, and

• Vertical-axis rotors.

In this report, only the horizontal-axis wind turbine will be discussed since the

modeling of the wind driven electric generator is assumed to have the horizontal-axis

rotor.

The horizontal-axis wind turbine is designed so that the blades rotate in front of the

tower with respect to the wind direction i.e. the axis of rotation is parallel to the wind

-3-
direction. These are generally referred to as upwind rotors. Another type of

horizontal axis wind turbine is called downwind rotors which has blades rotating in

back of the tower. Nowadays, only the upwind rotors are used in large-scale power

generation and in this report, the term “horizontal-axis wind turbine” refers to the

upwind rotor arrangement.

The main components of a wind turbine for electricity generation are the rotor, the

transmission system, the generator, and the yaw and control system. The following

figures show the general layout of a typical horizontal-axis wind turbine, different

parts of the typical grid-connected wind turbine, and cross-section view of a nacelle

of a wind turbine.

(a) (b)

-4-
 (c)
Fig. 1.1: (a) Main Components of Horizontal-axis Wind Turbine

(b) Cross-section of a Typical Grid-connected Wind Turbine

(c) Cross-section of a Nacelle in A Grid-connected Wind Turbine [14]

As can be seen in figure 1 (c), the nacelle consists of several components. They are

the generator, yaw motor, gearbox, tower, yaw ring, main bearings, main shaft, hub,

blade, clutch, brake, blade and spinner. Other equipment that is not shown in the

figure might include the anemometer, the controller inside the nacelle, the sensors

and so on.

The generator is responsible for the conversion of mechanical to electrical energy.

Yaw motor is used power the yaw drive to turn the nacelle to the direction of the

wind. The gearbox is used to connect the low-speed shaft (main shaft in the figure) to

the high-speed shaft which drives the generator rotor. The brake is used to stop the

main shaft from over speeding. The blades are used to extract the kinetic power from

the wind to mechanical power i.e. lifting and rotating the blades. The tower is made

-5-
from tubular steel or steel lattice and it is usually very high in order to expose the

rotor blades to higher wind speed.

1.3 PSCAD/EMTDC Software

In the initial stage, familiarization of the software and understanding the operations

involved in the software in order to conduct the project is necessary. The software

used in the project is the PSCAD/EMTDC software.

PSCAD is a group of programs which provides a flexible interface to

electromagnetic transient simulation based on EMTDC. A more detailed explanation

of this software can be looked up in appendix A.

-6-
2. BACKGROUND INFORMATION

2.1 Wind – Diesel Systems

In the early days, diesel generators were used to supply local load in most remote

areas. Single diesel generator can be used to supply small load, while for larger load

multiple diesel generators are employed. The most reliable and efficient diesel

generators are commonly used to supply the base load so that other generators can be

shut down at any time in case of regular maintenance, engine failure or other

emergencies.

The main advantage of the diesel systems is that they have been proven to be highly

dependable in many remote areas if they are maintained correctly. On the other hand,

diesel generators should be operated above minimum load in order to maintain their

efficiency and to minimize engine wear and fuel consumption. In many instances,

diesel generators are incorrectly sized and inefficiently controlled due to the

changing consumer loads which regulate the diesel’s operational constraints. The

main disadvantage of diesel generators is that they are costly to operate. Their

reliance on the diesel fuel proved to be costly since the fuel needs to be transported

to the diesel plant in order for the diesel produce electricity and the cost of diesel fuel

itself is likely to increase in the future. Moreover, diesel generators like many other

generators need to be well maintained particularly when they are not operated under

their minimum load. Hence, the cost of the diesel operation and maintenance which

-7-
requires replacing parts of the diesel generator itself is expensive considering the

remote location of the diesel plant.

Wind – diesel system provides a solution to this particular problem. The wind-diesel

hybrid system is commonly applied in many remote areas where there is a constant

wind source. When the demand is low, the wind power is used to supply the load.

Diesel generators are only used when the customer demand is high. In this case, the

wind power is also used to supply the demand in order to safe the diesel fuel

consumption providing there is enough wind to produce electricity. In other case,

diesel generators are used when there is a shortfall due to the availability of the wind

source.

2.2 Wind Farms in Australia

Wind power is currently used at a number of locations within Western Australia. In

1994, the largest installation within the State is located at Ten Mile Lagoon near

Esperance and was the first commercial wind farm within Australia. The wind farm

utilizes nine 225 kW Vestas variable pitch wind turbines which is connected to

Esperance’s conventional diesel power station via a 33kV power line. The total

capacity of the wind farm is 2MW. The wind farm is owned and operated by

Western Power.

Another recent wind farm that is now being planned in Western Australia is located

at Albany. It is a project that has been in the planning stages for over ten years and in

-8-
29th June 2000 Western Power Managing Director David Eiszele said the $45 million

Albany wind farm was given final approval by the State’s Energy Minister, the Hon.

Colin Barnett. The completion of the wind farm is predicted to finish in July 2001.

The project represented a positive commitment by Western Power and made a

significant contribution to the reduction of Australia’s greenhouse gases.

2.3 Offshore Windfarms

Offshore windfarms provides large electric generation since the speed of the

available wind is higher compared to the inland wind. The installation of the offshore

windfarms is relatively close to the shoreline since the cost of transmission line to

transfer the electricity to the inshore is expensive. One of the largest offshore

windfarms is the one based in Netherlands. The capacity of electricity generated is

17 MW (each unit of 600 kW).

Major challenges in constructing offshore windfarms are the monitoring of the

offshore installation since it is subject to the local weather. The weather plays an

important part since the construction cannot be continued if there are any

disturbances such as thunderstorm, cyclone, and so on.

2.4 Wind Turbine Generators

Wind turbine generator converts mechanical energy to electrical energy. The

electricity generated by the wind generator is usually fed to a transformer outside the

-9-
wind turbine or inside the nacelle (refer to figure 1.1 (c)) to raised the voltage in

order to be transmitted to the local grid or power distribution stations. The supply

frequency of the wind generator is usually 50 Hz for most part of the world and 60

Hz for the electrical standard in America.

Like many other generators, the wind generator tends to get hot. Thus, a cooling

system is introduced in the wind turbine. Some manufacturers use water-cooling

system to cool down the generators. The advantage is that it provides some electrical

efficiency advantages [25].

In wind turbine industry, two types of generators are used. They are:

• Synchronous generator, and

• Asynchronous (Induction) generator.

2.4.1 Why Induction Generator

Induction generator is commonly used in the wind turbine electric generation due to

its reduced unit cost, brushless rotor construction, ruggedness, and ease of

maintenance. Moreover, induction generators have several characteristics over the

synchronous generator. The speed of the asynchronous generator will vary according

to the turning force (moment, or torque) applied to it. In real life, the difference

between the rotational speed at peak power and at idle is very small approximately 1

percent. This is commonly referred as the generator’s slip which is the difference

- 10 -
between the synchronous speed of the induction generator and the actual speed of the

rotor.

slip (s) = ns - n

This speed difference is a very important variable for the induction machine. The

term slip is used because it describes what an observer riding with the stator field

sees looking at the rotor which appears to be slipping backward [35]. A more useful

form of the slip quantity results when it is expressed on a per unit basis using

synchronous speed as the reference. The expression of the slip in per uni is shown

below.

ns − nr
s=
ns

A four-pole, 50 Hz generator will run idle at 1500 rpm according to the following

formula.

120 f
ns =
p

If the generator is producing its maximum power, it will be running at 1515 rpm. A

useful mechanical property of the generator is that it will increase or decrease its

speed slightly if the torque varies and hence will be less tear and wear on the gearbox

as well as in the system. This is one of the important reasons to use asynchronous

(induction) generator compared to a synchronous generator on a wind turbine.

- 11 -
2.4.2 Induction Machine Analysis

The following figure shows the torque vs speed characteristic of typical squirrel cage

induction machine.

Fig. 2.1: Torque vs Speed Characteristics of Squirrel-cage Induction Generator [3]

In the figure, it can be seen that when the induction machine is running at

synchronous speed at the point where the slip is zero i.e. the rotor is spinning at the

same speed as the rotating magnetic field of the stator, the torque of the machine is

zero. If the induction machine is to be operated as a motor, the machine is to

operated just below its synchronous speed. On the other hand, if the induction

machine is to be operated as a generator, its stator terminals should be connected to a

constant-frequency voltage source and its rotor is driven above synchronous speed

(s<0) by a prime mover such as the wind turbine shaft. The source fixes the

synchronous speed and supplies the reactive power input required to excite the air-

gap magnetic field and hence the slip is negative.

- 12 -
The following figure shows the per-phase equivalent circuit of the induction

machine.

Rs jXs jXr
+
Iϕ
Is Ic Im Ir

Vt Rc jXm Vg Rr/S

Fig. 2.2: Per-Phase Equivalent Circuit of An Induction Machine

In this project, star-connected induction machine is evaluated. All the calculations

are in per-phase values. Hence, for a star-connected stator:

Vline
Vph = ; I ph = I line
3

In order to analyse the behaviour of an induction generator, the operation of an

induction motor must be fully understood. Once, the equivalent circuit parameters

have been obtained, the performance of an induction motor is easy to determine.

As shown in Fig. 2.3, the total power Pg transferred across the air gap from the stator

is
Rr
Pag = I 2r
s
And it is evident from figure 3 that the total rotor loss Prloss is

Prloss = I 2r R r

Therefore, the internal mechanical power developed by the motor is

Rr 1  1− s 
Pd = Pag − Prloss = I 2r − I 2r R r = I 2r R r  − 1 = I 2r R r  
s s   s 

- 13 -
From the power point of view, the equivalent circuit of figure 3 can be rearranged to

the following figure, where the mechanical power per stator phase is equal to the

power absorbed by the resistance R2(1-s)/s.

Rs jXs jXr Rr
+
Iϕ
Is Ir

Vt_ Vg 1− s 
Rc jXm Rr 
 s 

Fig. 2.3: Alternative Form for Per-Phase Equivalent Circuit

The analysis of an induction motor is also facilitated by using the power flow

diagram as shown in the following figure in conjunction with the equivalent circuit.

Pin Pag Pd Pshaft

Input per
Output
phase

PScu Pcore Prcu Prot

Stator copper Core loss Rotor copper Rotational
loss loss loss

Fig. 2.4: Power Flow Diagram

where,

Pag = Pin − PScu − Pcore

Pd = Pag − Pcore

Pout = Pshaft = Pd − Prot

- 14 -
The parameters of an induction generator can be determined by using the no-load test

and block rotor test (The steps in calculating the parameters and the test results

obtained from a 440V, 4.6A, 2.2kW induction motor are shown in appendix B).

2.5 Power Electronics

The conversion scheme used in this report is the AC/DC/AC converter. The input is

an ac sinusoidal source provided by the induction generator. The analysis of the

converter was initially carried out using pure ac sinusoidal source to observe the

waveform patterns and to ease the analysis of the converter. In latter case, the

induction generator is to be connected as the source.

2.5.1 AC to DC Conversion

Ac to dc conversion can be achieved by using the conventional methodology namely

the uncontrolled rectifier and the controlled rectifier. An uncontrolled rectifier

basically uses 6 diodes as the switches. The operation of such rectifier is discussed in

detail in section 2.5.1.1. On the other hand, the controlled rectifier, as the name

suggests, uses 6 controllable power switches such as Thyristors, GTOs, and so on.

However, only the use of Thyristors is implemented in this project.

- 15 -
2.5.1.1 Uncontrolled 6-Pulse Rectifier

As mentioned before, the uncontrolled 6-pulse rectifier uses 6 diodes as the switches

in the rectifier circuit. The typical circuit diagram for a diode rectifier is in Figure

2.5. The ac sinusoidal source is fed into the full bridge diode rectifier as shown in the

figure. The top half of the full bridge, only the diodes with its anode at the highest

potential will conduct. On the other hand, the bottom half of the full bridge, only the

diodes with its cathode at the lowest potential will conduct.

Fig. 2.5: 3-φ Diode Bridge Rectifier

As shown in figure 6, the diode bridge rectifier consists of 6 diodes, a capacitor on

the dc side and a load. The input is represented by purely 3-phase ac sinusoidal

source each displaced by 120°, and line inductance (Ls). The dc side can be replaced

by a constant dc current source Id. It is proven that replacing the dc current Id by a

load resistance Rload makes little difference in the circuit operation (Mohan, PE304).

The sequence of conduction of the diodes is from D1&D2, D3&D4, D5&D6.

- 16 -
First consider an idealised circuit (where Ls = 0) as shown in the following figure.

Fig. 2.6: 3-φ Rectifier Circuit With a Constant DC Current (Ls = 0)

The rectified output waveform after the diode bridge (vd) is determined by the

difference between vPn and vNn. The waveform of the rectified voltage vd is shown as

follows.
vd

2Vline
Vd0

0 π 0 π ωt
−
6 6
Fig. 2.7: Rectified Voltage Waveform After Diode Bridge

Thus, the average value of vd (Vd0) can be calculated by integrating the area under

the curve from -π/6 to π/6 and divide the area with the period of the waveform (π/3). In

this case, the frequency of the waveform is determined by the frequency of the

fundamental of the ac input. The frequency of vd is exactly 6 times the frequency of

the ac input since in one cycle of the ac input waveform corresponds to 6 cycles of

vd .

- 17 -
Note that the subscript ‘0’ in Vd0 symbolises zero line inductance (idealised circuit).

Hence,

Vd0 = 1.35 VLL

In the case of the ac input current waveform, consider only 1 phase (note that the ac

source is star connected). The ac input current is shown in the following figure.

vs

is1

is 120°
0
120°

60°

Fig. 2.8: Input AC Line Current of 3-φ Diode Rectifier

The duty cycle of the input line current is shown to be

D = 2/3

Then, the rms value of the input ac line current is

Is = √D Id = 0.816 Id

and the fundamental frequency component of the input ac line current (IS1) in this

idealised case can be found as

1
I s1 = 6 I d = 0.78 I d
π

The power factor can be found by using the following expression. It can be seen that

the power factor depends on the Current Harmonic Factor (CHF) and the

Displacement Power Factor (DPF).

- 18 -
 PF = DPF ⋅ CHF

DPF is defined as the phase difference between the fundamental-frequency

component of the ac sinusoidal phase voltage and the fundamental-frequency

component of the ac phase current. CHF is defined as the ratio between the rms value

of the fundamental-frequency component of the ac phase current and the total rms

value of the phase current itself.

Other effects on the converter such as the current commutation due to line

inductances, harmonic components of the input waveform, and so on, are elaborated

in great detail in [19].

2.5.1.2 Controlled 6-Pulse Full-Bridge Rectifier

Similar to the operation of the uncontrolled 6-pulse full-bridge rectifiers, the

controlled 6-pulse full-bridge rectifiers uses controllable power switches (in this

project, only Thyristors are used for uncontrolled rectification).

The aim of using controlled rectification is that it enables the designer to control the

output dc voltage to the desired values. This can be achieved by varying the time of

“forced” conduction i.e. the “turn on” of the thyristors in such a way that the average

value of vPn (VPn) and average value of vNn (VNn) adds up to yield the desired Vd at

the output. Consider the following circuit diagram of a typical controlled 6-pulse

rectifier using Thyristors.

- 19 -
 VPn

Vd

VNn

Fig. 2.9: Controlled 6-Pulse Full-Bridge Rectifier

By adjusting the firing angle of the thyristors (α), the potential difference between

the common point of the cathode of the top group of the thyristors to the ground (vPn)

and the potential difference between the common point of the anode of the bottom

group of the thyristors to the ground (vNn) can be varied and hence vd can be varied.

The effect of α on the average dc output of vd (i.e. Vd) is best described in the

following figure.

vd

vPn
Vd
vref
VPn
0
α VNn

vNn

Fig. 2.10: Effect of α on Vd

where, Vd = VPn - VNn.

- 20 -
From the above equation, knowing that VPn and VNn can be varied, then the range

of output voltages that can be obtained is

-Vd ≤ Output voltage ≤ Vd

In this project, it is assumed that the input voltage into the controlled rectifier is

higher (due to the over-voltages of the self-excited induction generator). Hence, the

proposed method of operation of the rectifier is the half-controlled rectifier which

utilises 3 thyristors and 3 diodes for the top group and the bottom group respectively.

Operation of the half-controlled 6-pulse rectifier is similar as described in Fig. 2.10.

However, the only difference is that only VPn waveform can be varied. Therefore, the

range of output voltages that can be obtained is between

0 ≤ Output voltage ≤ Vd

By using the half-controlled 6-pulse rectifier, not only the output can be varied

between the range of zero to Vd, but also from the economical and design point of

view it will certainly be cheaper to design and easier to implement.

- 21 -
2.5.2 DC to AC Conversion

The operation of a 3-φ inverter is quite similar to that of the 3-φ rectifier. The gating

signal pattern (PWM) is produced by the difference between the carrier frequency

signal (triangular) and the reference signal (sinusoidal) which is often known as

modulation ratio (ma). Moreover, the gating signal for two IGBTs in one leg must be

directly opposite of each other. Consider the 3-φ inverter circuit diagram shown

below.

Fig. 2.11: 3-φ Inverter Circuit Diagram

In one leg of phase a, when the upper IGBT conducts, the lower IGBT should not

conducts and vice versa. This goes to other phases of the inverter as well. The

modulation ratio used in this project is ≤ 1 (which is on the linear region of ma). In

this case, the peak value of the fundamental frequency component in one of the

inverter leg is

Vd
VAN1( pk ) = ma
2

- 22 -
By using simple algebra, the rms value of the fundamental frequency component of

line-to-line voltage is

3
Vline1( rms) = ma Vd
2 2

The principles and the other aspects on the inverter, such as harmonic contents, line

notching, effect of line inductances, and so on, are discussed in great detail in [19].

The waveforms obtained from the simulation in the project is expected to be similar

to those described in [19].

2.5.3 DC – DC Converter

The use of the dc to dc converter is to obtain a desired value of dc which differs from

the input. There are some types of dc-to-dc converters. They are:

• Step Down (Buck) Converter

• Step Up (Boost) Converter

• Buck-Boost Converter

• Cúk dc-dc Converter

• Full-Bridge dc-dc Converter

In these converters, one or more switches are utilised to control the dc output voltage

to equal a desired level. This can be achieved by controlling the switch on and off

durations i.e. ton and toff which is generated by comparing a signal-level control

voltage vcontrol with a sawtooth waveform as shown in the following figure.

- 23 -
 vsawtooth
vcontrol

On
Control Signal

ton Off
toff
ts

Fig. 2.12: Control Signal for DC-DC Converter

The control signal is then fed into the switches. Hence, the duty ratio of the switch

can be expressed as follows where |Vst| is the peak of the sawtooth waveform.

t on v control
D= =
Ts Vst

In this project, it is desirable to have higher or lower level of dc simultaneously and

an unidirectional power flow. Hence, only the analysis of both Buck-Boost converter

and Cúk converter is considered.

2.5.3.1 Buck-Boost Converter

As the name suggests, the Buck-Boost converter is used in applications where dc

output voltage can be either higher or lower than the input voltage. Also, the output

voltage has negative polarity with respect to the common terminal of the input

voltage. As mentioned in [19], the output-to-input voltage conversion ratio of this

converter is

V0 1
=D
Vd 1− D

- 24 -
Hence, this allows the output voltage to be higher or lower than the input voltage

depending on the switch duty ratio D. The circuit diagram of the converter is shown

in the following figure.

Id
_
+ + IL

Vd VL C VO R
L
_ _
+

IO
Fig. 2.13: Buck-Boost Converter Circuit Diagram

Basic operation of this converter is discussed in detail in [19].

2.5.3.2 Cúk Converter

The circuit diagram for the Cúk converter is shown in Fig. 2.14. Similar to the Buck-

Boost converter, the output voltage of the converter has a negative polarity with

respect to the common terminal of the input voltage. The capacitor C1 is used as the

primary means of storing and transferring energy from the input to the output.

iL1 L1 C1 L2 iL2

_ _ _ _
+ + VL1 + VC1 VL2 +

Vd C Vo R
_
+

Io
Fig. 2.14: Cúk Converter Circuit Diagram

Analysing the circuit diagram, the average voltage across the capacitor C1 at steady

state can be obtained as follows noting the average voltage across the inductors VL1

- 25 -
and VL2 are zero. The equation shows that VC1 is definitely larger than Vd and Vo

which is one of the disadvantage of using the Cúk converter.

VC1 = Vd + Vo

Similar to the Buck-Boost converter, the input-output voltage ratio is

V0 1
=D
Vd 1− D

This enables the output voltage to be higher or lower depending on the duty ratio of

the switch. The basic operation of the converter is discussed in detail in [19].

2.6 Past Investigations

2.6.1 Simulation of Wind-Diesel Hybrid Systems

A Simulation of wind-diesel hybrid system was analysed by P. K. T. Tan of Curtin

University of Technology in his final year thesis. The simulation of the hybrid

system was based on the EMTDC / PSCAD software package.

The report basically separated into two parts. In both parts, theories are first stated to

support the principles behind the working of the system. Each part then shows the

detailed construction of the relevant components followed by their simulation results.

In his project, the induction generator was modelled and analysed by its d and q axis.

In his proposed design of the hybrid system, the electricity produced by the wind

- 26 -
turbine is fed into a power electronic converter which rectifies the ac sinusoidal

voltage to dc and then inverted again to the ac bus system directly to the consumer

load.

The size of the capacitor connected on the dc bus of the converter is unrealistic

(100,000 µF) since the size of the capacitor is significant for the relative voltage

rating of the system. Also, the ripple voltage after the diode bridge does not

necessarily be ‘ripple free’. So, a specified ripple voltage after the diode bridge is

necessary to determine value of the capacitor. TOSHIBA AC/DC/AC 6 pulse

converter that is available in the laboratory was used to start the induction motor

from a 440V, 50Hz supply (the same machine was used in the experiment of the no-

load test and the blocked rotor test). The schematic diagram of the converter shows

that the capacitance value is 2200 µF for the particular power rating.

Moreover, in his thesis, the gating signal for the inverter side of the AC/DC/AC

converter is incorrect. Consider only one leg of the inverter (where the gating signal

is G1 and G4). From his result of the simulation, it shows that the gating signal G4 is

180 degrees displaced from G1. Considering that the signal G1 and G4 must be

directly opposite of each other, this result is unacceptable. Furthermore, for a 12

pulse inverter, the gating signal for one leg of one phase must be 30 degrees

displaced from the gating signal for the other leg of the same phase. This is

confirmed by using PSIM demo (power electronic software tool). Hence, his design

on the AC/DC/AC converter is inaccurate.

- 27 -
2.6.2 Self-excited Induction Generator

Stand-alone self-excited induction generators provide a significant reduction in

system initial costs and they have several advantageous due to the ruggedness and

low maintenance requirements of induction machines. Moreover, the self-excitation

of the induction generator causes voltage to collapse rapidly when overloaded. This

means that the self-excited induction generator is self-protected. On the other hand,

there are some drawbacks from using self-excited induction generator. The inability

to control the voltage and frequency when the load varies on stand-alone application

Several technical papers had been published in analysing the behaviour of self-

excited induction generator for stand-alone wind power application. Tandon et. al.

[4] studied the onset of self excitation and minimum capacitance requirements based

on the characteristic polynomial obtained from transient representation of the

machine. This method requires sets of results obtained from the numerical solution to

the characteristic polynomial satisfying certain criteria before inferring on the

minimum capacitance at which self-excitation occurs for a particular speed and load.

Malik and Mazi [20] suggested an indirect procedure, based on the steady state

equivalent circuit model to test the self-excitation of an induction generator. The

indirect methods involve solutions requiring some initial guess in a trial and error

procedure. Al-Jabri and Alolah [5] dealt with different limiting aspects based on per

phase stead state T-form equivalent circuit.

- 28 -
Chakraborty et. al. [6] analyses the excitation requirements for stand alone three-

phase induction generator at different combination of speed, load and excitation

capacitance using the inverse-model for the steady state equivalent circuit. Unlike

any other methods, this method involve no trial and error procedure and the model

used in paper is the inverse-Γ circuit model since the T-form circuit models are

actually more complex. Moreover, since the actual rotor variables are not required,

the inverse-Γ model can be applied for analysis of self –excited induction generators

with no loss of information and accuracy.

These authors [4, 5, 6, 20] only discusses minimum excitation requirement for the

induction generator to self-excite. The problems arise from the regulation of the

frequency and voltage of the induction generator. Most applications operate at

specified voltage and frequency. Tang and Zadavil [17], and Ouhrouche et. al [18]

discussed the transient behaviour of the self-excited induction generator in windfarm

application when disconnected from the grid. Both papers showed that when the

capacitor connected at the terminal of the induction generator is sufficient enough for

the self-excitation, the terminal voltage increased from its rated value to a new value

which is much higher than its rated voltage (over-voltage). In these cases, both

papers showed that the voltage could be as high as ten times its rated value.

Moreover, Ouhrouche et. al. [18] showed that the frequency at new operation

condition after disconnection from the grid was also increased. In their paper, it is

stated that it is not recommended for the induction generator to be connected back to

the grid after its disconnection from the grid itself.

- 29 -
2.6.3 Problem Formulation

In stand-alone mode, self-excitation on the induction generator depends on the

capacitor size connected on the stator terminals of the machine, speed of operation,

and the load [6]. Once self-excitation occurs, new operating conditions of the

generator i.e. both the terminal voltage and frequency at new values will be achieved.

In order to maintain the voltage and frequency of the generator at a specified value,

some means of control has to be implemented.

Feedback control is somehow cannot be implemented into the system since varying

the torque or speed of the generator, the self-excitation might not occur since the

excitation capacitance is valid only for a particular combination of speed and the

load.

Realising the problems in this particular area, this paper investigates the feasibility of

using power electronic converters in regulating the frequency and voltage of the self-

excited induction generator. In this case, the AC/DC/AC converter is modelled and

modified by the addition of dc – dc converter in the dc bus between the rectifier and

the inverter. Buck-Boost converter is used in the study since in the Cúk converter as

discussed in section 2.5.3.2, average voltage across capacitor C1 is higher as

compared to the input and output voltage. Thus, the rating of the capacitor for the

design purposes has to be higher and hence the size of the capacitor comes into

consideration as the voltage rating goes higher.

- 30 -
The proposed system is shown in the following figure. Temporary voltage source is

connected into the system at the initial stage to achieve a steady state operation of the

generator before it is isolated or used in the stand-alone mode of operation.

Temporary
Voltage Source

Power 3-φ
Prime
Mover IG Electronic Resistive
Converter Load

Self-Excitation
Capacitors
Fig. 2.15: Proposed System

- 31 -
3. MODELLING OF THE SYSTEM

3.1 Self-Excited Induction Generator Model

Self-excited induction generator model is modelled in PSCAD software package.

The parameters of the induction machine are based on the recorded results from [6].

Therefore, the machine parameters is

Parameters Value
Voltage Rating 400V
Current Rating 10.9A
Rated Power 7.5HP (5.593 kW)
Stator Resistance 1.23Ω
Rotor Resistance 1.105Ω
Stator Leakage Reactance 2.756Ω
Rotor Leakage Reactance 2.756Ω
Mutual Reactance 91.2Ω
Table 3.1: Induction Machine Parameters

Hence, with the information given, the values can be converter into it’s per unit

value. Thus, taking the voltage phase rating and the current rating as the base values,

the base impedance can be calculated as

Vbase 230.94
Z base = = = 21.1872Ω
I base 10.9

Therefore, knowing the base impedance of the machine, the machine parameters can

be converted into its per unit value. The following table summarises the machine

parameters.

- 32 -
 Parameters Per Unit Value
Stator Resistance 0.058
Rotor Resistance 0.052
Stator Leakage Reactance 0.13
Rotor Leakage Reactance 0.13
Mutual Reactance 4.3
Table 3.2: Induction Machine Parameters (Per-Unit Values)

The simulation parameters of the induction machine are shown in the following

figure. Some parameters are not provided in [6] and therefore are assumed. For

example, the model for the induction machine is assumed to have a single cage and

hence the second cage resistance for the PSCAD model is set to be infinite (in this

case higher in its per unit value).

Fig. 3.1: Induction Generator Parameters

The machine is operating at 50 Hz and hence the base angular frequency is

ω = 2πf = 2π × 50 = 314.16 rad / s

The SEIG model in PSCAD is shown in Fig. 3.2. It can be seen from the figure that

an external voltage source is connected at the stator terminal of the induction

generator. The external voltage source is connected until it is in its steady state

- 33 -
operation and then the voltage source is disconnected by sending a tripping signal to

the circuit breaker (BRK1).

Fig. 3.2: SEIG Model in PSCAD

The induction machine is driven above its synchronous speed (w = 1.1pu) so that the

machine operates as a generator. Moreover, the induction machine is operating on

speed control mode (S = 1). A 3-phase resistive load represents the load and the

excitation capacitors (86.5µF) are connected at the stator terminals of the induction

generator.

- 34 -
In the simulation of the SEIG, the minimum capacitance value to ensure the self-

excitation on the induction generator was found to be 86.5µF in which the machine

uses the parameters that are exactly similar to [6]. However, the recorded result from

[6] shows that the SEIG machine was able to self-excite at 30µF. The main reason

behind this discrepancy is due to the fact that the load is vaguely defined in the

paper. In this simulation, the connected load is assumed to be a 3-φ resistive load.

Hence, the findings from [6] in this case could not be linked with the result of the

simulation.

The induction generator is expected to deliver power both to the load and the voltage

source before the circuit breaker operates. At initial stage, the source and the shunt

capacitance supply the reactive power to the induction generator. But when the

source is disconnected from the system, only the excitation capacitors that supply the

reactive power to the generator so that the generator can deliver power to the load.

The main use of the capacitors is to maintain the air-gap voltage between the stator

and the rotor of the induction generator.

The following figure shows the simulation results obtained from the self-excited

induction generator at 3410µF capacitive compensation level. The results show

clearly that initially the real power delivered to the load is increased due to the stator

overvoltage. In this case, negative value of the output power represents the generated

power by the induction generator. On the other hand, the induction generator absorbs

the reactive power provided by both the voltage source and the excitation capacitors

before the circuit breaker operates.

- 35 -
 Source Connected Islanding

Fig. 3.7: Simulation Results of Self-Excited Induction Generator Parameters

The generator speed is also 1.1 pu as mentioned earlier to run the generator above its

synchronous speed as to generate power instead of absorbing power and the

mechanical torque is negative since it is driven externally by the wind turbine shaft.

The following figure shows the stator rms voltage at 86.5µF capacitive compensation

level. In this simulation, the external voltage source is disconnected from the

induction generator at 0.5 seconds. The result shows clearly that, after islanding, the

- 36 -
“theoretical” steady state operation is reached at very high generated voltage levels:

more than five times the rated voltage of the induction generator. Moreover, the

steady state operation is achieved at approximately 6.5 sec (with transient of 6

seconds). This result agrees with that of [18] in a way that the overvoltage will occur

after the disconnection of SEIG from the grid.

Islanding

Fig. 3.3: RMS Stator Line Voltage

Fig. 3.4 also shows the induction generator stator frequency at 86.5 µF capacitive

compensation level.

- 37 -
 Islanding

Fig. 3.4: Stator Frequency

However, at higher capacitive compensation levels, the transient of the stator rms

voltage and the frequency is smaller and the frequency is decreased as shown in Fig.

3.5 and Fig. 3.6 respectively.

1000µF

2000µF
Islanding
3000µF

Fig. 3.5: RMS Stator Voltage At Higher Compensation Levels

- 38 -
 Islanding
1000µF

2000µF

3000µF

Fig. 3.6: Stator Frequency At Higher Compensation Levels

The simulation results (maximum voltage and its corresponding time), for various

capacitive compensation levels, are summarised in Table 3.3. Note that the circuit

breaker operates at 0.5 sec

C (µF) Vmax (kV) t (sec) F (Hz)

86.5 2.08 6.7 51.98
500 1.672 0.67 49.6
1000 1.213 0.615 45.58
2000 0.94 0.608 39.615
3410 0.783 0.767 35.33
Table 3.3: Maximum Overvoltages Magnitude

The table shows that as the capacitive compensation level increased, the maximum

overvoltage magnitude at the stator terminal is decreased. Moreover, steady state

operation of the induction generator can be achieved in a shorter period of time.

Similarly, as the capacitive compensation level is increased, the operating frequency

of the induction generator is decreased.

- 39 -
Hence, in this report, the main objective of this project is in the regulation of the

output voltage magnitude and frequency to a specific value by using power

electronic converters. The design of such converters is discussed in the next section.

- 40 -
3.2 AC/DC/AC Converter Model

The analysis of the AC/DC/AC converter is divided into three sections. The first

section analyses the conversion from ac to dc with uncontrolled 6-pulse full-bridge

rectifier or with half-controlled 6-pulse full-bridge rectifier. The second section

evaluates the conversion from dc to dc using the Buck-boost converter. The third

section examines the conversion from dc to ac using a Pulse Width Modulated

(PWM) Voltage Source Inverter (VSI). Finally, the analysis of the whole system

combining different converters to achieve regulated output voltage and frequency of

SEIG is presented.

3.2.1 AC to DC Conversion

The design of the ac to dc converter is discussed in this section. Ac to dc conversion

can be achieved by:

• Uncontrolled 6-pulse full-bridge rectifier

• Half-controlled 6-pulse full-bridge rectifier

3.2.1.1 Uncontrolled 6-Pulse Full-Bridge Rectifier

The uncontrolled 6-pulse full-bridge rectifier utilises 6 diodes as its switching

devices. In the design of the diode rectifier as well as other converters, a pure

sinusoidal voltage with constant magnitude and frequency is used at the input as to

ease the design of the converters. Later in the project, the output voltage of the SEIG

- 41 -
is to be used as the input to the converters. The diode bridge rectifier circuit is shown

in the following figure.

Fig. 3.7: PSCAD Model of Diode Bridge Rectifier

A 415V, 50Hz input is fed into the diode bridge rectifier. The output voltage and

current at the output of the diode bridge rectifier is shown in the following figure.

Fig. 3.8: Voltage and Current At Rectifier Output

- 42 -
The ripple voltage at the output of the rectifier is very much dependent on the

capacitor connected at the output of the rectifier. The higher the capacitor value, the

smaller the ripple voltage at the rectifier output. In this case, a 500µF capacitor is

used at the output of the rectifier which yields a voltage ripple of approximately

25.9V.

3.2.1.2 Half-Controlled 6-Pulse Full-Bridge Rectifier

The half-controlled 6-pulse rectifier utilises 3 thyristors at the top group and 3 diodes

at the bottom group as shown in the following figure. One of the bottom group of the

diode conducts when the cathode of the diode is at the lowest potential compared to

the other two diodes. On the other hand, the thyristors can only conduct when the

firing angle (α) of the thyristors is applied.

Fig. 3.9: Half-Controlled 6-Pulse Full-Bridge Rectifier Circuit

- 43 -
The control signal for the thyristors are governed by the following simple equation. It

can be seen that the firing angle (α) is a function of the input line-to-line RMS

voltage (VLL) and the reference average dc voltage (Vd1) that is specified according

to the desired value

 2π Vd1 
α = cos −1  − 1
 3 2 VLL 

PSCAD software enables the manipulation of such equation into the circuit diagram.

This is shown in the following figure where b = α.

Fig. 3.10: Thyristor Firing Angle (α) Circuit

However, the information given by this equation is in terms of degrees and it cannot

be fed into the thyristors since the thyristors only takes a train pulse signal with a

magnitude of 1 for a short period of time to turn on. Thus, the output “b” is fed into a

more complex circuit diagram to produce a pulse train with a delay angle “b” and a

magnitude of 1. The circuit diagram is shown in the following figure.

- 44 -
 Fig. 3.11: Alpha Train Pulse Signal Generator Circuit Diagram

The input “b” is used as the phase angle for the sine function block with the

magnitude and frequency is set at 1kV and 50Hz respectively. This sinusoidal

function is then fed into a comparator block such that if A ≥ B the output is 1, and if

A < B the output is zero.

On the other hand, a reference zero phase angle sinusoidal reference must be chosen.

In this case, the phase voltage of the sinusoidal input is chosen as the reference.

Similar to the previous explanation, this sinusoidal reference is fed into a function

block such that if A ≥ B the output is 1, and if A < B the output is zero. Then, the

output of both comparators are fed into an AND gate which will only yield 1 if both

input are both high (i.e. at 1). The following diagram shows the output of both

comparator as well as the output of the AND gate as to aid the explanation.

- 45 -
 Bottom Comparator Output

Top Comparator Output

AND Gate Output

Fig. 3.12 Comparator Outputs and AND Gate Output For Alpha

Thus, the output of the AND gate is then fed into an edge detector block which only

gives a pulse output if there is a transition (zero to one) at the input of the block. This

output is then fed into an integrator which is resetable at 300Hz. The output of this

integrator yields the firing angle for T1.

With this circuit implemented by taking “b” as the input yields the pulse train as

shown below. It can be seen from the figure that the train signal has a period of 0.02

sec (i.e. 50Hz) and has a magnitude of 1.

- 46 -
 α

Fig. 3.12: Thyristor Train Pulse Waveform

The above waveform is applied into thyristor 1 (T1) at the top group of the rectifier.

For the other two thyristors (T3 and T5), the waveform of T1 is phase shifted by

120° and 240° for T3 and T5 respectively. Therefore, T1 is fed into two circuits as

shown in the following figure to achieve the firing angle for T3 and T5.

Fig. 3.13: T3 and T5 Firing Angle Circuit

After successfully design the circuit for the firing angle of the thyristors, the analysis

of the half-controlled rectifier can be conducted. As mentioned earlier, the output of

a rectifier (Vd) is a result of the difference between VPn, which is the potential

difference between the common point of the cathode of the thyristors, and VNn,

which is the potential difference between the common point of the anode of the

- 47 -
diodes. To evaluate that the circuit operation is valid, 1kV line-to-line voltage source

is connected at the input of the rectifier and the desired output of the rectifier is set to

be at 750V. The following waveform shows the output voltage of the half-controlled

rectifier, VPn, and VNn.

Fig. 3.14: VPn, VNn, and Vd Waveforms

It can be seen that the average output of the half-controlled rectifier is approximately

equal to 747V which is close to the desired output of 750V.

- 48 -
3.2.2 DC to DC Conversion Using Buck-Boost Converter

Dc to dc converter design is evaluated in this section. The Buck-Boost converter

design is mainly discussed in this section because it is assumed that the dc voltage

level can be adjusted to any desired value.

The following diagram shows the circuit diagram of a typical Buck-Boost converter.

It consists of one switching device (in this case is GTO) which enables to turn on and

off depending on the applied gating signal.

Fig. 3.15: Buck-Boost Converter Circuit Diagram

The gating signal for the GTO can be obtained by comparing the sawtooth waveform

at high switching frequency (in this case is at 20kHz) with a control voltage at

constant magnitude. The control voltage is varied such that if the input to the

converter is varied, the output of the converter is maintained at a constant desired

value. The magnitude of the control voltage can be obtained by rearranging the

following equation.

Vd 2 D v
= , where D = control
Vd1 1 − D Vst

- 49 -
where, |Vst| = magnitude of sawtooth waveform, Vd1 = input dc voltage, Vd2 =

desired output dc voltage, and D = duty cycle.

Rearranging the above equation yields

Vst
vcontrol =
Vd1
+1
Vd 2

The above equation is implemented in PSCAD and is shown in the following figure.

Desired DC Voltage

Fig. 3.16: Firing Angle For GTO

To evaluate the validity of the converter design and operation, a 1kV dc voltage is

connected at the input of the Buck-Boost converter and the desired output dc voltage

is set to be at 650V. The control voltage and the gating signal for the GTO is shown

in the following figure where Vcon is the control voltage, T is the sawtooth waveform,

and G is the gating signal for the GTO.

Fig. 3.17: Gating Signal For GTO

- 50 -
It can be seen from the figure that the control voltage is at 0.33 and hence the

corresponding gating signal for the GTO is obtained. With this gating signal, the

output voltage of the converter is shown in the following figure.

Fig. 3.17: Input Voltage Vd and Output Voltage Vo of Buck Boost Converter

The output of the Buck-Boost converter is approximately 636V which is close to the

desired output voltage of 650V. The above figure also shows that the output dc

voltage of the Buck-Boost is not entirely ripple “free”. In other words, there is a

small ripple at the output voltage which was found to be approximately 20V. This

ripple voltage magnitude is very much dependent on the size of the capacitor and

inductor values in the converter circuit.

- 51 -
3.2.3 DC to AC Conversion Using PWM VSI

Voltage source inverter using pulse width modulation topology is used in the

conversion from dc to ac. The model of this inverter in PSCAD is shown in the

following figure. It can be seen that the inverter consists of 6 IGBTs. Each leg of the

inverter represents each phase of the inverter output. Moreover, the reference signal

in the gating circuit for the IGBTs in each leg of the inverter is displaced by 120°.

This can be achieved by shifting the reference signal (sinusoidal) of the PWM.

Fig. 3.18: PWM VSI Supplying Resistive Load

The gating signal for the IGBTs can be created by comparing a sinusoidal reference

voltage with a triangular waveform. This is implemented in PSCAD as shown in the

following circuit. The triangular waveform has a frequency of 5 kHz and the

reference sinusoidal voltage operates at 50 Hz. Basic operation of the circuit is that if

the sinusoidal voltage is higher than the triangular waveform, the output of the

comparator is one. Similarly if the sinusoidal voltage is lower than the triangular

waveform, the output of the comparator is zero.

- 52 -
 Fig. 3.19: PWM Circuit

Graphical explanation on the method is shown in the following figure.

Fig. 3.20: PWM Pattern For IGBT Switching

Referring to Fig.3.18, the gating signal for G1a must be directly opposite of the G4a

as shown in the following figure. This is to avoid short circuit of the dc source. Keep

in mind that the circuits are assumed to be ideal and blanking time is not

implemented in the circuit design.

- 53 -
 Fig. 3.21: PWM Patterns in One Leg of VSI

The modulation index (ma) of the converter is varied such that if the input voltage to

the inverter is varied the magnitude of the fundamental frequency component of the

inverter output voltage is fixed. Hence, input output relationship which takes into

account the modulation index, i.e. the ratio of the sinusoidal reference magnitude

with the sawtooth magnitude, must be formulated. This can be achieved by

rearranging the following equation where (VLL1)RMS is the rms value of the

fundamental frequency component of the line-to-line voltage, |Vsin| is the magnitude

of the sinusoidal reference voltage, and |Vtri| is the magnitude of the carrier voltage

i.e. the sawtooth waveform.

(V )
LL1 RMS =
3
m a Vd , where m a =
Vsin
Vtri
2 2

Rearranging the question yields,

(
2 2 VLL1 RMS )
Vsin = ⋅ Vtri
3 Vd

- 54 -
In evaluating the design of this inverter, a 800V dc voltage is used as the input to the

inverter and the desired output phase voltage at fundamental frequency is 240V

(RMS) or 339V (peak). The inverter output phase voltages with reference to the zero

potential between two capacitors are shown in the following figure. It can be seen

that the magnitude of the waveforms is half of that of the input voltage i.e. 400 volts

dc. This is because the input dc voltage is imposed into the two capacitors on the dc

bus. Half of the input voltage is imposed in each capacitor and when the top IGBT is

turned on the voltage imposed on the output of one leg is half of the input voltage

with respect to the neutral node between the two capacitors and negative half of the

input voltage is imposed when the bottom IGBT is turned on. Furthermore, it can be

seen from the waveform that each phase is displaced by 120 degrees from each other.

120° 60° 120° 60°

- 55 -
 Fig. 3.22: Inverter Voltage Output With Respect To Zero Potential Between Capacitors on DC Side

Similarly, the inverter output voltage with respect to dc negative side are shown in

the following figure. In this case, the magnitude of the waveforms is equal to the

input voltage. Again, the phase voltages are displaced by 120 degrees. The line

voltage of the inverter output can be obtained by taking the difference between the

two phase voltages as shown in the following figure. The difference between VaN and

VbN creates the line voltage of the inverter output.

Fig. 3.23: Inverter Phase Voltage Output With Respect To DC Neutral

Finally, the phase voltage of the inverter output with respect to neutral ground is

shown in the following figure. In this figure, the phase difference between two phase

voltages are more obvious compared to that of the phase voltages describes

previously.

- 56 -
 Fig. 3.24: Inverter Phase Voltage Output With Respect To Neutral Ground

In order to determine the magnitude of the fundamental component of the inverter

output phase voltage with respect to the neutral ground, Fast Fourier Transform

(FFT) was conducted on the above waveform. The result of FFT on the waveform is

shown in the following figure. It can be seen that the magnitude of the fundamental

component of the inverter output phase voltage is approximately 340.8V with a Total

Harmonic Distortion (THD) of 29%

- 57 -
 340.8V

Fig. 3.25: FFT of PWM VSI Output Phase Voltage

- 58 -
3.2.4 AC/DC/AC Converter

In this section, the combinations of the converters discussed in the previous sections

will be evaluated. Three designs are presented in this report and they are

1. AC/DC/AC converter utilizing diode rectifier – PWM VSI

2. AC/DC/AC converter utilizing half-controlled rectifier – PWM VSI

3. AC/DC/AC converter utilizing diode rectifier – Buck-boost converter – PWM

VSI

In the analysis of this converters, a 1kV line-to-line rms three phase voltage source

acts as the input for the converters. In the next chapter, the complete system consists

of the self-excited induction generator connected to these three converters is

presented.

3.2.4.1 Diode Rectifier – PWM VSI

The combination of the diode and PWM VSI yields the AC/DC/AC converter of the

1st design as shown in the following figure. The rectification of the ac voltage using

the diode bridge rectifier is then fed into the PWM voltage source inverter.
Diode Rectifier PWM VSI

Fig. 3.26: Diode Rectifier With PWM VSI Converter Design (1st Design)

- 59 -
The output voltage and current of the rectifier (DC side voltage and current) is shown

in the following figure. The voltage ripple on the output of the rectifier was found to

be 18V (utilising 2500µF capacitor on the output of the rectifier) and the current

peak at the rectifier output is 230A.

Fig. 3.27: Rectifier Output Voltage and Current of 1st Design

The output voltage waveform of the rectifier circuit is fed into the inverter and the

resultant output phase voltage with respect to neutral ground, and the output line-to-

line voltage is shown in the following figure where Vinva is the inverter output phase

voltage, and Vlinvab is the inverter output line-to-line voltage.

- 60 -
 Fig. 3.28: Inverter Output Vinva and Vlinvab of 1st Design

The simulation result of the phase voltage waveform differs from the result of the

phase voltage of the PWM VSI shown in Fig. 3.24. In order for one to argue that this

result is acceptable, the circuit is implemented in PSimDemo Power Electronic

Software Simulation as shown in the following figure. The result of the simulation of

the phase voltage at the inverter output shows similarity to that in Fig. 3.28. From

this finding, one can guarantee that the result of the inverter output phase voltage

shown in Fig. 3.28 is correct.

- 61 -
 Fig. 3.29: PSIMDemo Model of Diode Rectifier With PWM VSI

The magnitude of the fundamental frequency component of the inverter output phase

voltage is found to be 336V with a THD of 50.8% as shown in the following figure.

336V

Fig. 3.30: FFT of Output Phase Voltage Of Diode Rectifier PWM VSI Converter Design

- 62 -
3.2.4.2 Half-Controlled Rectifier – PWM VSI

In this section, the combination of the half-controlled rectifier and PWM VSI yields

the AC/DC/AC converter of the 2nd design as shown in the following figure. The

rectification of the ac voltage using the half-controlled rectifier is then fed into the

PWM voltage source inverter.

Half-Controlled Rectifier PWM VSI

Fig. 3.31: Half-Controlled Rectifier With PWM VSI Converter Design

In this design, the input to the half-controlled rectifier is set to 2kV line-to-line rms

three phase voltage source is utilised. The desired output of the rectifier is set to 1.5

kV. Thus, the output voltage and current of the half-controlled rectifier is shown in

the following figure. The ripple voltage at the inverter output is 32.14V peak to peak

and the current peak is 98A.

- 63 -
 Fig. 3. 32: Half-Control Rectifier Output Voltage and Current Of 2nd Design

The output of the half-controlled rectifier is then used as the input to the inverter.

The output line-to-line voltage of the inverter is shown in the following figure. As

can be seen, the voltage has a period of 0.02 seconds (i.e. 20 ms) which corresponds

to a 50Hz output frequency.

Fig. 3.33: Inverter Output Phase Voltage and Line-To-Line Voltage of 2nd Design

When FFT is applied on the output phase voltage of the inverter, the magnitude of

the fundamental frequency component is 334V with a THD of 61% as shown in the

following figure.

- 64 -
 334V

Fig. 3.34: FFT of Inverter Output Phase Voltage Of 2nd Design

3.2.4.3 Diode Rectifier – Buck-Boost Converter – PWM VSI

The combination of diode rectifier, buck-boost converter, and PWM VSI yields the

AC/DC/AC converter of the 3rd design as shown in the following figure. Basically,

the output of the diode rectifier is fed into the buck-boost converter and the output of

this converter acts as the input for the inverter.

Diode Rectifier Buck-Boost Conv. PWM VSI

Fig. 3.35: Diode Rectifier With Buck-Boost Converter and PWM VSI Converter Design (3rd Design)

- 65 -
Since, the output voltage polarity of the buck-boost converter is negative with respect

to the input voltage (i.e. from the output of the diode rectifier), the PWM VSI

configuration has to be modified such that the positive node of the inverter input is

connected to the positive node of the output of the buck-boost converter.

In this design, the input to the diode rectifier is a 1kV line-to-line three phase voltage

source, the desired output of the buck-boost converter is set to 800V, and the desired

magnitude of the fundamental frequency component of the inverter output phase

voltage is set to 339V. The following figure shows the output voltage waveform of

the rectifier and the output voltage waveform of the buck-boost converter. The

simulation result shows that the buck-boost converter successfully controls the

rectified output voltage of 1.4kV to 800V.

800V

(a) (b)

Fig. 3.36: (a) Output Voltage Of Diode Rectifier; (b) Output Voltage Of Buck-Boost Converter

The output of the buck-boost converter is then fed into the PWM VSI. The inverter

output phase voltage and line-to-line voltage is shown in the following figure.

- 66 -
Fig. 3.37: Inverter Output Phase Voltage And Line-Line Voltage Waveform Of 3rd Converter Design

Similar to the previous waveform, the line-to-line voltage waveform has a

fundamental frequency of 50Hz. However, notice that the output phase voltage of the

inverter is in the negative region. This implies that the voltage across the load is also

negative. Therefore, detail analysis on this converter has to be done in order to fix the

problem. The first step is to analyse the current flows in this converter. Hence,

consider the following buck-boost converter conventional current flow diagram.

Id1 Id0

_
+ + IL Icap

Vd1 VL L C Vd2

_ In _ Id2
+

Fig. 3.38: Buck-Boost Converter Conventional Current Flow Diagram

It was found that when the GTO of the buck-boost converter is off, there In still flows

to the common point of the anode of the diode rectifier which in normal operation

there should not be any current flow when the GTO is off. Moreover, Id2 flows in the

- 67 -
negative direction for most of the time during the one switching period as shown in

the following figure where G is the gating signal for GTO.

Ton Toff

Fig. 3.39: Current Flows In Buck-Boost Converter

Therefore, knowing the source of this problem, another GTO switch and a diode is

implemented in the buck-boost converter as shown in the following figure.

Id1 Id0

_
+ + IL Icap

Vd1 VL L C Vd2

_
_ In Id2
+

Fig. 3.40: Modified Buck-Boost Converter

It can be seen that the second GTO is connected at the negative input voltage node of

the buck-boost converter to stop the current In to flow when the first GTO is off.

- 68 -
Similarly the diode is connected at the positive output voltage node of the buck-boost

to ensure the positive current flow of Id2. The simulation result of this converter is

shown in the following figure. It can be seen from the result that when Ga (gating

signal for GTO) is off, In is not flowing. Moreover, Id2 flows in the positive direction.

Fig. 3.41: Current Flows in Modified Buck-Boost Converter

Although the current flows in the buck-boost converter is corrected, there is another

discrepancy observed. There exist current spikes in the system flowing from Ido to

Icap and to Id2. In order to fix this problem, an inductor is connected at the positive

output voltage node of the buck-boost converter.

The inverter output phase voltage and line-to-line voltage waveforms after the

modification of the buck-boost converter are shown in the following figure.

- 69 -
 Fig. 3.42: Inverter Output Phase Voltage And Line-Line Voltage Of 3rd Converter Design

Applying FFT on the output inverter phase voltage yields a magnitude of

fundamental frequency component at 341V with a THD of 56.6%.

- 70 -
4. RESULTS OF THE COMPLETE SYSTEM

After the design of the three converters is completed, the next step is to combine

individual converters with the self-excited induction generator in order to obtain the

result. Thus, there are three complete system that are presented in this section and

they are

1. Self-excited induction generator connected to the first converter design (i.e.

diode rectifier with PWM VSI)

2. Self-excited induction generator connected to the second converter design

(i.e. half-controlled rectifier with PWM VSI)

3. Self-excited induction generator connected to the third converter design (i.e.

diode rectifier with buck-boost converter and PWM VSI)

4.1 System 1

The first system consisting of the SEIG and the first design of the converter is

elaborated in this section. The following figure shows the first combined system. As

can be seen, the unregulated output voltage of the self-excited induction generator is

fed into the first converter design which consists of the diode rectifier with PWM

VSI. At the initial stage, the generator is supplied by an external voltage source and

when the induction generator is at its steady state operation, the voltage source is

disconnected from the system.

- 71 -
 Fig. 4.1: Complete System 1 Circuit Diagram

The output rms voltage of the induction generator is shown in the following figure.

The result shows that when the external voltage source is disconnected at 0.4

seconds, the new steady state operating voltage of the induction generator is 1.18kV

which is achieved at 0.8 seconds. The maximum overvoltage occurred is 1.3kV at

0.57 seconds.

Islanding

Fig. 4.2: RMS Voltage Of SEIG

- 72 -
On the other hand, the output of the rectifier, as shown in the following figure,

indicates that the maximum overvoltage is 3.16kV at 0.57 seconds. The steady state

rectifier output voltage is 2.2kV which is achieved at 1.2 seconds. The dc peak-to-

peak ripple voltage is 10.68V, which is approximately 0.5% of the peak ripple

voltage.

Islanding

Fig. 4.3: Rectifier Output Voltage Of System 1

This rectified voltage is then fed into the PWM VSI and hence the phase voltage and

the line-to-line voltage are shown in the following figure. Both inverter output

voltages have a period of 20ms which corresponds to a 50Hz output frequency.

Fig. 4.4: Inverter Output Phase Voltage And Line-to-Line Voltage Of System 1

- 73 -
When FFT is applied on the phase voltage of the inverter output, the magnitude of

the fundamental frequency component is 334V with a THD of 48%

334V

Fig. 4.5: FFT of Inverter Output Phase Voltage Of System 1

- 74 -
4.2 System 2

Self-excited induction generator connected with the second design of the converter

(i.e. half-controlled rectifier with PWM VSI) yields the second system which is

shown in the following figure.

Fig. 4.6: Complete System 2 Circuit Diagram

The overvoltage at stator terminal of the SEIG is rectified to a lower value using the

half-controlled rectifier and the output of the rectifier is then acts as the input to the

inverter. In the previous design of the firing angle circuits for the thyristors (shown in

Fig. 3.11), they consider the input to the rectifier is at 50Hz operation. In this case,

the stator frequency of the induction generator varies according to the rotor speed,

excitation capacitors, and the load (as mentioned in the literature review). This

frequency variation has to be taken into consideration as well. Thus, the circuit

shown in Fig. 3.11 is modified into the following circuit which takes into account the

frequency variation of the SEIG and the disconnection time of the external voltage

source that is connected to the SEIG. The reason to consider the disconnection time

- 75 -
 of the voltage source is that the firing angle of the thyristors is at zero degrees until

the SEIG operates at its steady state. Only when the SEIG new steady state operating

condition is reached, then the firing angle obtained from the circuit shown in Fig.

3.10 is applied.

Disconnection
Time

Frequency
Monitor

Frequency
Monitor

Fig. 4.7: Modified Alpha Circuit

The following figure shows the output of the half-controlled rectifier which shows

that the dc voltage ripple peak-to-peak is approximately 43.2V which is 2.5% of the

peak voltage ripple and the dc voltage level of 1.73kV.

Fig. 4.8: Half-Control Rectifier Output Voltage Ripple Of System 2

- 76 -
With this dc level at the half-controlled rectifier output, the inverter output line-to-

line voltage is shown in the following figure which shows that the period of the

waveform is 20 msec which corresponds to 50Hz output waveform.

Fig. 4.9: Inverter Output Line-To-Line Voltage Of System 2

The magnitude of the fundamental frequency component of the inverter output phase

voltage can be obtained by applying FFT to the phase voltage waveform as shown in

the following figure. In this case, the magnitude of the fundamental frequency

component of the phase voltage is 341V with a THD of 62%

341%

Fig. 4.10: FFT Of Inverter Output Phase Voltage OF System 2

- 77 -
4.3 System 3

The final system is analysed in this section. It consists of the self-excited induction

generator connected with the third design of the converter (i.e. diode rectifier

connected to buck-boost converter and then connected to PWM VSI) as shown in the

following figure.

Fig. 4.11: Complete System 3 Circuit Diagram

In this system, it was observed from the simulation result that the induction generator

never self-excites at any capacitor values. Hence, further analysis was done and it

was decided to represent the load at 0.8 power factor lagging as compared to the

resistive load only. Hence, in order to achieve a 0.8 power factor lagging, inductors

have to be connected in series with the resistor in one phase for the three phases.

Thus, to calculate the size of the inductor needed, the following calculation was

made.

Zload = R load + jX L = Zload ∠θ°

- 78 -
Manipulating the equation above yields,

XL
tan θ =
R load
∴ X L = R load ⋅ tan θ
(
= 20 × tan cos −1 0.8 )
= 15Ω

Therefore,

X L = 2πfL
XL
∴L =
2πf
15
= = 0.0478H = 47.8mH
2π × 50

Thus, by connecting the inductors in series with the resistor, the self-excitation of the

induction generator is achieved. The output of the rectifier and the output of the buck

boost converter are shown in the following figure. As can be seen from the figure,

the dc level of 1.57kV from the output of the rectifier is successfully amplified to

1.8kV at the output of the buck-boost converter. The ripple peak-to-peak voltage at

the output diode rectifier is 8.8V which is 0.56% of the peak voltage. In the

simulation, the output of the buck-boost as can be seen from the following figure is

still in the transient state after 2.8 seconds. However, the steady state is reached at

1.8kV which is at approximately 4 seconds. The steady state waveform cannot be

obtained due to the huge number of sample points in the simulation and hence the

size of the simulation file and the limitation of the server computer memory size

restrict the simulation result.

- 79 -
 Fig. 4.12: Rectifier Output And Buck-Boost Converter Output Voltages Of System 3

The inverter output line-to-line voltage is shown in the following figure. It can be

seen that the period of the waveform is 20 msec which corresponds to 50Hz

frequency waveform. The FFT of the phase voltage yields a voltage magnitude of

337V at fundamental frequency of 50 Hz with a THD of 68% as shown in Fig. 4.14.

Fig. 4.13: Inverter Output Line-To-Line Voltage Of System 3

- 80 -
 343V

Fig. 4.14: FFT Of Inverter Output Phase Voltage Of System 3

- 81 -
5. FUTURE WORK

When the objectives of the projects were met, other issues were raised which allows

the improvement of this project. However, due to time constraint these issues cannot

be implemented and solved during the phase of the project. Thus, future work can be

done based on these factors.

Firstly, as can be seen from the results of the simulation across all three complete

system designs, the inverter output voltages (whether it is line-to-line voltage or

phase voltage) contain high order harmonics at higher magnitude. This can be

eliminated by proper selection of harmonic filters such as the low pass filter or a

passive filter that is tuned to the specific harmonic that needs to be filtered. Thus, a

sinusoidal waveform of 50Hz can be obtained when the filters are implemented at the

inverter output.

Secondly, harmonic cancellation can also be implemented in the converter design.

This can be achieved by using the following converter topologies.

• 12-pulse rectifier

• Multilevel inverters

In the 12-pulse rectifier, a wye–wye wye–star transformer is utilised to split the ac

voltage and fed into two-cascaded 6-pulse diode rectifier. The addition of both output

of these rectifiers makes up the total dc rectified output voltage. On the other hand,

the “diode clamped” multilevel inverters utilize combinations of diodes and IGBTs

to create a voltage output which represents the sinusoidal voltage at the fundamental

- 82 -
frequency. The PWM VSI design in this report is basically a two level inverter.

Other multilevel inverter involves three-level, five-level, and in which the output of

the inverter has lower harmonic content (i.e. lower THD) as the level goes up.

Thirdly, the effect of blanking time at the switching of the IGBTs in the inverter

operation is not considered in this report since the switches are assumed to be “ideal”

as shown in Fig. 3.2.1 which shows the switching pattern for two IGBTs in one

inverter leg. The turn-on of a switch in one inverter leg didn’t take into account

whether the other switch in the same leg is turned off. In actual situation, one IGBT

in one leg can only turned on when the other IGBT on the same leg is turned

completely off and vice versa. This is shown in the following figure.

tB

Fig. 5.1: Blanking Time In PWM VSI

- 83 -
It can be seen from the figure that when G1a switch in one inverter leg turns off, the

other switch on the same leg (G4a) turns on after a blanking time of tB seconds.

Finally, the efficiency of the AC/DC/AC converter is not analysed in this report. The

benefit of analysing the efficiency of the converter is that one can use this

information to predict how efficient is the converter in regulating the voltage and

frequency of SEIG. Also one can do comparisons on the complexity of the converter

design and the efficiency of the converter. Power dissipated or loss in the converter

circuit generates heat and this information can be further used to determine the

requirements of the types of heat sinks, size of the converter, and so on, if the actual

hardware implementation is required.

- 84 -
6. CONCLUSION

In this final year project report, the modelling of self-excited induction generator and

AC/DC/AC converter for wind energy conversion was conducted using

PSCAD/EMTDC software package.

After the brief introduction on the wind energy and the discussion on why induction

generator is more suitable in the field of wind energy conversion, the detail analysis

and the characteristic of the induction generator was elaborated in detail in which the

project objectives were determined through the problem formulation.

The report also looked into the design of power electronic converters in particular the

three designs of AC/DC/AC converters. At the beginning of the design phase of the

power electronic converters, each converter design namely the diode rectifier, half-

controlled rectifier, buck-boost converter, and PWM VSI were presented and the

proper operation of individual converter was justified before combining them into

AC/DC/AC converter. Moreover, the report examines the combinations of three

converters namely the combinations of diode rectifier with PWM VSI, the

combinations of half-controlled rectifier with PWM VSI, and the combinations of

diode rectifier, buck-boost converter, and PWM VSI.

Finally, the complete system consisting of the SEIG connected to individual

converters were presented. Satisfactory simulation results of the complete system

were obtained at the end of the project.

- 85 -
In conclusion, the self-excitation phenomenon of the induction generator plays an

important role in the wind energy conversion. The unregulated stator voltage and

frequency of the self-excited induction generator can be regulated by implementing

power electronic converters which enable to rectify the ac voltage to dc and then

invert the signal back to ac at fixed frequency of 50Hz and a magnitude at the

fundamental frequency component of 240V rms (or 339V peak). Recommendations

for future work was presented in the previous section which suggested the design of

harmonic filters such as the passive low pass filter or tuned filters, and harmonic

cancellations by utilizing other converter designs such as the 12-pulse rectifier,

multilevel inverters, and so on. Investigations on the hysteresis current controlled

voltage source inverter design can be conducted so that comparable results can be

done between the voltage controlled voltage source inverter (PWM VSI) and the

hysteresis current controlled voltage source inverter.

- 86 -
REFERENCES

[1] A. Gastli, M. Akherraz, A. Al-Badi, “Modelling and Simulation of a Variable

Speed Self-Excited Induction Generator”, International Journal of Renewable

Energy Engineering, Vol. 2, No. 3, pp 221-227, December 2000.

[2] A. Sarre, “Wind power gathers speed", Australian foundation for science,

[Online], Available: http://www.science.org.au/nova/037/037key.htm, 2000.

[3] A. E. Fitzgerald, C. Kingsley Jr., S. D. Umans, “Electric Machinery”, Mc Graw

Hill Inc. 1992.

[4] A. K. Tandon, S. S. Murthy, G. J. Berg, “Steady State Analysis of Capacitor

Self-Excited Induction Generators”, IEEE Transactions on Power Apparatus

and Systems, Vol. PAS-103, No. 3, pp 612-618, March 1984.

[5] A. Kh. Al-Jabri, A. I. Alolah, “Limits on the Performance of the Three-Phase

Self-Excited Induction Generators”, IEEE Transactions on Energy Conversion,

Vol. 5, No. 2, pp 350-356, June 1990.

[6] C. Chakraborty, S. N. Bhadra, A. K. Chattopadhyay, “Excitation Requirements

for Stand Alone Three-Phase Induction Generator”, IEEE Transactions on

Energy Conversion, Vol. 13, No. 4, pp 358-365, December 1998.

[7] C. Grantham, F. Rahman, D. Seyoum, “ A Regulated Self-excited Induction

Generator For Use In a Remote Area Power Supply”, International Journal Of

Renewable Energy Engineering, Vol 2, No 1, April 2000.

[8] C. Saunders, “Electrical Machines 303: A Study of Machine Theory”, Curtin

University of Technology 2000.

[9] C. S. Demoulias, P.S. Dokopoulos, “Transient Behaviour and Self-Excitation

of Wind-Driven Induction Generator After Its Disconnection From the Power

Grid”, IEEE Transactions on Energy Conversion, Vol. 5, No. 2, pp272-278,

June 1990.

[10] D. J. De Renzo, “Wind Power: Recent Development”, Noya Data Co. 1979.

[11] D. M. Eggleston, F. S. Stoddard, “Wind Turbine Engineering Design”, Van

Nostrand Reinhold 1987.

[12] D. M. Simmons, “Wind Power”, Noya Data Co. 1975.

[13] G. Raina, O. P. Malik, “Wind Energy Conversion Using a Self-Excited

Induction Generator”, IEEE Transactions on Power Apparatus and Systems,

Vol. PAS-102, No. 12, pp 3933-3936, December 1983.

[14] J. Arrillaga, D. B. Watson, “Static Power Conversion From Self-Excited

Induction Generators’, IEE Proceedings, Vol. 125, No. 8, pp 743-746, August

1978.

[15] J. F. Walker, N. Jenkins, “Wind Energy Technology”, John Wiley & Sons Inc.

1997.

[16] L. Ouazene, G. McPherson Jr., “Analysis of The Isolated Induction Generator”,

IEEE Transactions on Power Apparatus and Systems, Vol. PAS-102, No. 8, pp

2793-2798, August 1983.

[17] L. Tang, R. Zavadil, “ Shunt Capacitor Failures Due to Windfarm Induction

Generator Self-Excitation Phenomenon”, IEEE Transactions on Energy

Conversion, Vol. 8, No.3, pp 513-519, September 1993.

[18] M. A. Ouhrouche, et. al., “ EMTP Based Simulation of a Self-Excited

Induction Generator After its Disconnection From the Grid”, IEEE

Transactions on Energy Conversion, Vol. 13, No.1, pp 7-14, March 1998.

[19] N. Mohan, T. M. Undeland, W.P. Robbins, “Power Electronics: Converters,

Application, and Design”, 2nd ed., John Wiley and Sons Inc., 1995, Canada.

[20] N. H. Malik, A. H. Al-Bahrani, “Influence of the Terminal Capacitor on the

Performance Characteristics of a Self-Excited Induction Generator”, IEE

Proceedings, Vol. 137, Pr. C, No. 2, pp 168- 173, March 1990.

[21] N. H. Malik, A. A. Mazi, “Capacitance Requirements for Isolated Self Excited

Induction Generators”, IEEE Transactions on Energy Conversion, Vol. EC-2,

No.1, pp 62-68, March 1987.

[22] N. H. Malik, S.E. Haque, “Steady State Analysis and Performance of an

Isolated Self-Excited Induction Generator”, IEEE Transactions on Energy

Conversion, Vol. EC-1, No. 3, September 1986.

[23] N. P. Cheremisinoff, “Fundamentals of Wind Energy”, Ann Arbor Science

Publishers Inc., 1978, Michigan.

[24] O. Shtchetinine, “ Voltage Stabilization System for Induction Generator in

Stand Alone Mode”, IEEE Transactions on Energy Conversion, Vol. 14, No. 3,

pp 298-303, September 1999.

[25] P. F. Ryff, “Electric Machinery”, Prentice Hall Inc., 1994, New Jersey

[26] R. G. Slemon, “Electric Machines and Drives”, Addison Wesley Publishing

Co., 1992, USA.

[27] R. Hunter, G. Elliot, 1994, “Wind/Diesel Systems: A Guide To The

Technology & Its Implementation”, Cambridge University Press, UK

[28] S. Krohns, “Wind Turbine Generators”, Danish Wind Turbine Manufacturers

Association, [Online], Available: http://www.windpower.dk/tour/

wtrb/electric.htm, 2000.

[29] S. Krohns, “Asynchronous (Induction) Generators”, Danish Wind Turbine

Manufacturers Association, [Online], Available:

http://www.windpower.dk/tour /wtrb/async.htm, 2000.

[30] S. Krohns, “Wind Turbines And Power Quality Issues”, Danish Wind Turbine

Manufacturers Association, [Online], Available:

http://www.windpower.dk/tour /grid/rein.htm, 2000.

[31] S. Rajakaruna, R. Bonert, “ A Technique for the Steady-State Analysis of a

Self-Excited Induction Generator with Variable Speed”, IEEE Transactions on

Energy conversion, Vol. 8, No. 4, pp 757-761, December 1993.

[32] S. Wei, A. I. Maswood, “A Novel Current Source PWM Drive Topology with

Specific Harmonic Elimination Switching Patterns”, IEEE Power Engineering

Review, pp 53-55, December 2000.

[33] S. M. Alghuwainem, “Steady-State Analysis of a Self-Excited Induction

Generator Including Transformer Saturation”, IEEE Transactions on Energy

conversion, Vol. 14, No. 3, September 1999.

[34] S. M. Alghuwainem, “Steady-State Analysis of an Isolated Self-Excited

Induction Generator Driven by Regulated and Unregulated Turbine”, IEEE

Transactions on Energy Conversion, Vol. 14, No. 3, pp 718-723, September

1999.
[35] S. S. Murthy, O.P. Malik, A. K. Tandon, “Analysis of Self-Excited Induction

Generators”, IEE Proceedings, Vol. 129, Pt. C, No. 6, pp 260-265, November

1982.

[36] T. F. Chan, “Steady-State Analysis of Self-Excited Induction Generators”,

IEEE Transactions on Energy Conversion, Vol. 9, No. 2, pp 288-294, June

1994.

[37] T. F. Chan, “Self-Excited Induction Generators Driven by Regulated and

Unregulated Turbines”, IEEE Transactions on Energy Conversion, Vol., 11,

No. 2, pp 338-343, June 1996.

[38] V. C. Toro, “Electric Machines And Power Systems”, Prentice Hall Inc., 1985,

New Jersey

[39] Z. A. Yamayee, J. L. Bala Jr., “Electromechanical Energy Devices And Power

Systems”, John Wiley & Sons Inc., 1994, Canada

APPENDIX A - PSCAD
PSCAD®/ EMTDC® Software

The behaviour of a power system under certain condition plays an important part in

the design of the system itself. For example, the behaviour of a power system

network under severe overvoltages due to a fault in the network is critical in the

design of the protection system in the network, and so on. In certain cases, these

conditions are best analysed before they occur in the power system network. That is

why the simulation using computer softwares comes into the picture.

PSCAD® stands for Power Systems Computer Aided Design. It is used as a GUI

(Graphical User Interface) for EMTDC. It is a powerful graphical user interface

which increases productivity in undertaking simulation of electromagnetic transients

studies of electrical power systems. PSCAD has many advanced features to boost

productivity, including intelligent data forms, interactive control and feedback, up-

to-date documentation of the circuit along with plots and comments, context

sensitive help, hierarchical designs, and multiple levels of zooming. The software

also serves as the front end for the RTDS (Real Time Digital Simulator).

EMTDC® stands for ElectroMagnetic Transient for DC. It was first engineered by

Dennis Woodford in 1975 since at that time the existing tools were not

comprehensive and flexible to study the Manitoba Hydro Nelson River HVDC

Power System in Manitoba, Canada. The well-known Dommel Algorithm is used in

EMTDC software. The program structure has been developed so as to allow the user
great flexibility in modelling power systems and their associated controls in great

detail.

An interpolation algorithm has been included to permit accurate modelling of

switching devices such as thyristors and GTO's. The solution of very larger power

systems has been made efficient by allowing networks to be separated into smaller

sub-networks connected by transmission lines. Solutions of these individual

subsystems have been made efficient by use of Gaussian Elimination sparse matrix

techniques.

PSCAD / EMTDC is a software simulation tool for studying the transient behaviour

of electrical networks. This software was first developed in 1976. The software

comes with a comprehensive library of electrical models consisting of all aspects of

AC and DC power systems and controls. Users are also able to create their own

model and libraries using the built-in graphical Component Workshop in the

software. Moreover, EMTDC supports models written in FORTRAN / C/C++ which

made the construction of models easy. The resources of user’s computer are the only

limitation to the size of the circuit to be simulated in PSCAD / EMTDC software.

The software supports all aspects of conducting a simulation including circuit

assembly, run-time control, analysis, and reporting. In 1988 PSCAD V1 was first

introduced on Apollo workstations. The rapid development of the software brought

the introduction of PSCAD V2 in 1994, which operates under Unix system. In 1999

PSCAD V3 was released and it operates under Windows platform.

The scope of simulation studies that can be conducted using PSCAD/EMTDC

software is as follows.

• General power system electromagnetic transient studies,

• Overvoltages in a power system due to a fault or breaker operation,

• DC transmission systems and detailed controls,

• Synchronous and induction machine torsional effects and self-excitation,

• Static Var Compensators,

• Harmonic interactions,

• Relay protection studies,

• and many more.

FILEMANAGER

The main window that appears when the user runs PSCAD is the File Manager

software module. The user’s database were presented in a project / case / file

hierarchy so that the usual database maintenance tasks such as file editing, copying,

deleting, and so on can be performed easily and directly from the File Manager.

Moreover, other PSCAD software modules can be initiated by clicking on the

appropriate button on the top right hand screen of the File Manager and each

individual modules that are currently running is shown as icons on the right hand

side of the File Manager screen under ‘process icons’ label.

DRAFT

The Draft module is where the users are able to graphically create their power system

models which are to be simulated. Simulation using either EMTDC or the RTDS can

be performed in the Draft module.

Electrical components of power systems are located on the library palette which is on

the right hand side of the Draft window. Users copy the desired components and

place them onto the canvas which is on the left hand side of the Draft window

followed by the appearance of the parameter menu of that particular component.

In the case where a large number of components are to be implemented onto the

canvas, users are able to create a sub-circuit which can be link together by exporting

the variables that are needed. Once the user has completed the layout, it can be

printed either by a PostScript based laser printer or plotter capable of accepting HP-

GL commands.

RUNTIME

The analysis of the user’s circuit can be carried out in the Runtime module. Users are

allowed to load, start, stop and analyse the specific parameters of their circuit in a

form of on-line plotting of data as it is generated in the Runtime module. Valuable

feedback from the simulation can be obtained regarding the transient or dynamic

behaviour such as set-point changes, breaker action and fault.

MULTIPLOT

The generated data by either EMTDC or RTDS can be plotted and analysed in

Multiplot module. It also enables scaling and general formatting of data.

Furthermore, Fourier analysis can be directly performed. Plots obtained from the

Runtime module can be combined and manipulated into a single page for the ease of

comparison. Moreover, some other features includes labelling, text editing, date and

so on can be added into the plots.

APPENDIX B – TEST MACHINE
 Three Phase Squirrel Cage Induction Motor Parameters From No-Load and

Locked Rotor Tests

Objective

The purpose of this experiment is to conduct no-load and locked rotor test in order to

obtain the induction machine parameters.

Equipment List

• 440 V, 4.6 A, 2.2 kW, 1420rpm, delta-connected induction motor

• 2 x Nanovip power meters

• 1 x Multimeter

• Autotransformer

Circuit Diagram

Nanovip 1

A
440 V

3-Phase
B
50Hz
C
Nanovip 2
Procedure

No-load Test

The autotransformer panel was set to 440 V. Then the motor was uncoupled from the

dc dynamometer by removing a rubber bar that connects the shafts together and the

motor was switched on to allow the motor to warm up (approximately 5mins). The

line voltage, line current, and total input power were obtained. After the motor was

switched off, the stator winding was disconnected and the resistances of each stator

winding were obtained using the multimeter. Thus, the stator resistance per phase Rs

was calculated by using the average of the three readings. Determine the rotational

losses by subtracting the input power and the stator copper losses (i.e. 3Is2Rs). Lastly,

by assuming the stator core loss and the friction and windage losses are equal in

magnitude, the values of stator core loss Rc was calculated.

Locked Rotor Test

Recouple the motor to the dc dynamometer and lock the rotor. The autotransformer

was set to 90V. Then, quickly measure the line voltage, line current, total power

input and torque as the motor was switched on. After the measurements were

obtained, the motor was switched off. Because of the locked rotor, there are no

frictional loss and at the reduced voltage, the core loss is negligible. Thus, it can be

assumed that entire power input is used to supply the stator and rotor copper loss. Rr

was then calculated from the reading as well as Xe (Xe = Xs + Xr). It can be assumed
that stator leakage reactance Xs and rotor leakage reactance Xr are equal in

magnitude.

Measurement

No-Load Test

Parameters Nanovip 1 Nanovip 2

Vl 413 V 416 V

Il 2.27 A 2.43 A

Pin 427 W 559 W

Ra 11.5 Ω -

Rb 11.7 Ω -

Rc 11.4 Ω -

Rave (Rs) 11.53 Ω -

P.F. -0.45 0.55

Locked Rotor Test

Parameters Nanovip 1 Nanovip 2

Vl 89.2 V 90.1 V

Il 4.15 A 3.94 A

Pin 9W 310 W

T 0.875 Nm -
Calculations

No-Load Test

The following figure shows the circuit diagram for no-load test.

Inl Rs jXs

Vnl jXm

Il
Vnl = 414.5 V I nl = = 1.357 A
3
Il = 2.35 A

Pnl = 559 – 427 = 132 W

Hence, the rotational loss can be calculated as follows.

Prot = Pnl − Pcu = Pnl − 3I 2nl R s

= 132 − 3(1.357) 2 (11.53) = 68.32 W

It is assumed that the stator core loss and friction and windage losses are equal in

magnitude. Hence

Psc = Prot = 68.32 W

There are some other information that could be obtained:

Pnl 132
PF = cos θ = = = 0.078
3Vnl I nl 3(414.5)(1.357)

∴ θ = 85.51° lagging

And,
Vnl 414.5
Z nl = = = 305.453∠85.51° = 23.91 + j304.516
I nl 1.357∠ − 85.51°
Hence, the no load resistance and reactance can be found as

Rnl = 23.91 Ω, and Xnl = 304.516 Ω

Moreover,

Psc 68.49
Ic = = = 0.0551 A
3Vnl 3(414.5)

Thus,
Vnl 414.5
Rc = = = 7.5227 kΩ
Ic 0.0551

Locked Rotor Test

The following figure shows the equivalent circuit diagram of the locked rotor test of

the induction machine.

Ibl Rs Rr jXs jXr

Vbl

Vbl = 89.65 V Pbl = 301 W

Il = 4.045 A T = 0.875 Nm

Il 4.045 and Pbl 301

I bl = = = 2.335 A R bl = = = 18.4 Ω
3 3 3I bl 3(2.335) 2
2

The rotor resistance referred to the stator Rr can be found once the locked rotor test

resistance is found.

Rbl = Rs + Rr
Hence,

Rr = Rbl – Rs = 18.4 – 11.53 = 6.87 Ω

The locked rotor reactance can be found as

X bl = Z 2bl − R 2bl = 38.394 2 − 18.4 2 = 33.7 Ω

Thus,

Xs = Xr = Xbl / 2 = 16.85 Ω

Moreover, the magnetising reactance can be found once the leakage reactance of the

stator is found.

Xnl = Xs + Xm

and,

Xm = Xnl – Xs = 304.516 – 16.85 = 287.67 Ω

Summary of Results

Base values:

Vbase = 440 V, Ibase = 2.656 A, Zbase = 165.67 Ω.

Parameters Values p.u.

Rs 11.53 Ω 0.0696

Xs = Xr 16.85 Ω 0.102

Rr 6.87 Ω 0.0415

Rc 7.5227 kΩ 45.41

Xm 287.67 Ω 1.7364
APPENDIX C – DIODE RECTIFIER
APPENDIX D – HALF-CONTROLLED RECTIFIER
APPENDIX E – BUCK-BOOST CONVERTER
APPENDIX F – PWM VSI
APPENDIX G – DIODE RECTIFIER WITH PWM VSI
APPENDIX H – HALF-CONTROLLED RECTIFIER WITH PWM

VSI
APPENDIX I – DIODE RECTIFIER, BUCK-BOOST

CONVERTER, AND PWM VSI

APPENDIX J – COMPLETE SYSTEM 1 DESIGN
APPENDIX K – COMPLETE SYSTEM 2 DESIGN
APPENDIX L – COMPLETE SYSTEM 3 DESIGN
APPENDIX M – TECHNICAL PAPER