Documente Academic
Documente Profesional
Documente Cultură
INSTITUTE OF TECHNOLOGY
Advisor
Dr. Mengesha Mamo
A Thesis submitted to the School of Computing and Electrical Engineering of Bahir Dar
University in partial Fulfillment of the Requirements for the Degree of Master of Science in
Power System Engineering.
January 2014,
Institute of Technology
____________________________________ __________
____________________________________ __________
____________________________________ __________
____________________________________ _________
This is to certify that the above declaration made by the candidate is correct to the best of my
knowledge.
Finally, but most importantly, I would like to thank my beloved family members, who always
give me strong inspirations, moral supports, and helpful suggestions.
i
Table of content
Contents page
Acknowledgement .......................................................................................................................i
Abstract ...................................................................................................................................... x
1.Introduction ............................................................................................................................. 1
ii
3.Electric Traction .................................................................................................................... 15
iii
4.3. Mathematical Model Equation of Eleven Level CHBMLI ............................................... 48
4.4. Analysis of Switching Angles and Output THD of 11-Level CHBMLI by Newton-
Raphson Method ................................................................................................................... 57
5.1. Simulation of Three phase induction motor driven by 11 level CHBMLI ........................ 71
6.1. Conclusion...................................................................................................................... 75
References ................................................................................................................................ 77
Appendices ............................................................................................................................... 81
Appendix A. ............................................................................................................................. 81
Appendix B. .............................................................................................................................. 81
iv
List of table
Table 3.1: Switching states in one leg of the three-level diode clamped inverter ........................ 25
Table 3.2: Switching states of the five level inverter.................................................................. 26
Table 3.3: Comparison of power component requirements per phase leg among three ............... 38
Table 4.1: Output voltage levels and their switching states for eleven-level inverter .................. 48
Table 4.2: The number of eliminated harmonic components of SHEPWM technique. ............... 53
Table 4.3: content of matric , , and from MATLAB program by using the Newton‟s
method ...................................................................................................................................... 60
Table 4.4: The switching angles of 7-level cascaded inverter..................................................... 60
Table 4.5: The switching angles of 9-level cascaded inverter..................................................... 61
Table 4.6: The switching angles of 11-level inverter. ................................................................ 61
Table 4.7: Voltage THD of 7- level cascaded inverter ............................................................... 62
Table 4.8: Voltage THD of 9- level cascaded inverter ............................................................... 62
Table 4.9: Voltage THD of 11- level cascaded inverter ............................................................. 63
v
List of figure
vi
List of Acronyms
AC Alternating Current
ASD Adjustable Speed Drive
CHBMLI Cascaded H-bridge Multilevel Inverter
DC Direct Current
ER Electric Railway
HV High Voltage
vii
MLC Multilevel Converter
MW Mega Watt
viii
List of Symbols
C Capacitor
DC Voltage
, , Fourier Coefficients
Switching Angle
Modulation Index
S Switch
ix
Abstract
Multilevel inverters have been attracting the attention of researchers both in the academia and the
industry in the recent decade for high-power and medium-voltage energy control. They can
synthesize switched waveforms with lower levels of harmonic distortion than an equivalently
rated two-level converter. In railway application, multilevel inverters can be directly supplied
from caternary voltage eliminating transformer on board the train, reducing deadweight of the
train and saving space on the train.
There are mainly three types of multilevel inverters; these are a) diode- clamped, b) flying
capacitor and c) cascade multilevel inverter (CMLI). Among these three, the cascaded multilevel
inverter is best suited for railway traction application because of its modular structure and use of
devices at low rating. This thesis presents the application of the cascaded multilevel inverter in a
transformerless railway traction drive. When these inverters are used for transformerless railway
traction drive directly, the THD contents in output voltage of inverters affects the drive
significantly as the performance of drive depends very much on the quality of voltage applied. In
this thesis, the THD contents of 7, 9 and 11 level cascade multilevel inverters have been
analyzed. The THD depends on the switching angles for different units of multilevel inverters;
therefore, the switching angles are calculated first by using Newton-Raphson method where
certain number of harmonic components has been eliminated. It has been found that the eleven
level CMLI satisfies this limit at modulation index of one while seven and nine level CMLI
violates this limit at all modulation index value.
In order to validate the analytical result, IGBT based three-phase cascaded eleven level inverter
using selected harmonic elimination (SHE) PWM technique is simulated with the use of
MATLAB/SIMULINK at modulation index equal to one. The FFT analysis is done and
performance of induction motor fed by eleven level CMLI and ideal AC source has been
compared.
Using calculated switching angles, THD analysis is carried out analytically as well as using
MATLAB simulation and found to be 5.0623% and 6.37% respectively. Both results are in close
agreement with IEEE-519 standard which is 5%.
This eliminates the need of a transformer in the railway traction drives and also results in the
reduction in the Total Harmonic Distortion of the voltage to be supplied to the traction motors.
x
Chapter one
1. Introduction
1.1. Overview
Railways are basically categorized to Electrified and Diesel-electric systems. This classification
dictates a majority of their characteristics. Based on input voltage source, electrical railway
system can be categorized in to AC traction system and DC traction system. With utilization of
voltage source inverters, AC traction motors are presently common in electric railway traction
applications. Several AC motors such as switched reluctance, permanent magnet synchronous and
asynchronous motors are being used widely in electrified traction applications. Meanwhile,
induction motors are being used as the most popular AC motors for the electrified traction
systems
Several European railway networks supply their rolling stock from single phase 15 or 25 kV with
60, 50, or Hz frequencies. Therefore, the railway vehicles are equipped with a transformer
to reduce the line voltage to a more convenient level for the motor and semiconductors. At such
variable frequencies, the medium voltage transformer suffers from poor efficiency and high
weight. Several research activities have been going on to substitute the transformer by power
electronic concepts to improve its performance.
In this thesis, analysis and modeling of cascaded multilevel inverter fed induction motor with
transformerless approach is studied for traction applications. This system consists of a cascaded
converter and inverter which are serving different jobs such as producing three-phase, 50 Hz, 380
V output voltage for the induction motors through the input voltage with any amplitude and
frequency. This structure is used for converting single phase AC voltage of the catenary to DC
and converting the DC voltage to three phase AC voltage with a SHE PWM controlled inverter.
In such a structure, harmonic currents injected into the power system during the SHEPWM
switching process are low. Using such a structure can reduce the necessity of using a transformer
for feeding the induction motor and also reduce the amounts of harmonics injected to the power
system greatly.
1
1.2. Research Background
Multilevel converters inherently tend to have a smaller dv/dt due to the fact that switching
involves several smaller voltages. Furthermore, switching at the fundamental frequency (50Hz)
will also result in decreasing the number of times these voltage changes occur per fundamental
cycle. There are three topologies of MLCs. They are,
a) Diode clamped multilevel converter, b) Flying capacitor multilevel converter and
c) Cascaded multilevel converter
2
The diode clamped and the flying capacitor type MLCs are fed from a single DC source. The DC
voltage is equally divided employing a string of capacitors. In the diode clamped MLC, diodes
are employed to clamp the voltages to produce a staircase voltage, hence the name diode clamped
MLC. The voltage clamping to produce a staircase is achieved via capacitors in a flying capacitor
type MLC.
Due to the presence of a large number of capacitors in the flying capacitor MLC, the device is
bulky and the operation and control of the device is complex. Hence, the diode clamped and
cascaded MLC are the two topologies that are extensively employed for various applications.
The key issue for multilevel converter modulation is the harmonics elimination. The multilevel
fundamental switching method inherently provides the opportunity to eliminate certain higher
order harmonics by varying the times at which certain switches are turned “on” and turned “off”,
which is also called varying the switching angles.
Harmonic elimination is performed for several reasons. The first reason is harmonics are a
source of EMI. Without harmonic elimination, designed circuits would need more protection in
the form of snubbers or EMI filters [1]. As a result, designed circuits would cost more. The
second reason is that EMI can interfere with control signals used to control power electronics
devices and radio signals. The third reason is that harmonics can create losses in power
equipment. For example, harmonic currents in an electrical induction motor will dissipate power
in the motor stator and rotor windings. There will also be additional core losses due to harmonic
frequency eddy currents. The fourth reason is that harmonics can lower the power factor of a
load. Increased harmonic content will decrease the magnitude of the fundamental relative to the
magnitude of the entire current. As a result, the power factor would decrease [1].
The total harmonic distortion(THD) of the output voltage from a converter is limited by standards
specifications. IEEE-519 standard for the harmonic limit in the electric network is 5% for all
voltage levels below 69 kV and 3% for all voltage above 161kV [9]. Therefore, Harmonics must
always be limited below threshold levels prescribed by standards [9], both in their THD and
individual magnitudes. The THD of the output voltage can be reduced by eliminating or filtering
out the harmonics from the system. The harmonics in the system can be filtered by passive low
pass filters or the harmonics can be eliminated by harmonic elimination technique. The harmonic
elimination method involves solving complex trigonometric equations. The voltage magnitude of
3
the DC sources impacts the elimination of harmonics in the output voltage. Harmonic elimination
is simple when the DC source voltage magnitudes are equal and constant.
The second advantage of multilevel converters concerns switch ratings. Since multilevel
converters usually utilize a large number of DC voltages, several switches are required to block
smaller voltages. Since switch stresses are reduced, required switch ratings are lowered.
The third advantage of multilevel converters concerns system reliability. If a component fails on a
multilevel converter, most of the time the converter will still be usable at a reduced power level.
Furthermore, multilevel converters tend to have switching redundancies. In other words, there
might be more than one way to produce the desired voltage [1].
The fourth advantage of multilevel converters concerns application practicality. Some of the
applications of the multilevel converters include compressors, fans, grinding mills, rolling mills,
conveyers, blast furnace blowers, mine hoists, reactive power compensations, high voltage direct
current (HVDC) transmission, Flexible Alternating Current Transmission System(FACTS), wind
energy conversion, electric traction , railway traction , Hybrid Electric Vehicle. The multilevel
converters are not only for high power applications like HVDC etc but also for low power
requirements like in renewable energy sources. These converters can be easily interfaced with
the renewable energy sources like photo voltaic cells, fuel cells, and wind energy conversion [3].
Application of the multilevel inverters in electric railway traction is a recent development. Thus,
power electronics is contributing toward a greener and cleaner world.
One disadvantage of multilevel converters is that they require more devices than traditional
converters. The system cost may increase (part of the increased cost may be offset by the fact
switches with lower ratings are being used). Using more devices also means the probability of a
system failure will increase. Another disadvantage of multilevel converters concerns control of
the switches. The increased number of switches will result in more complicated control.
There are four popular modulation methods for multilevel converter control, space vector control
method, selective harmonic elimination method, space vector PWM method and sinusoidal PWM
method. Among these methods, space vector control method, space vector PWM method, and
sinusoidal PWM method cannot completely eliminate the specified harmonics. Only selective
harmonic elimination method can completely eliminate the specified harmonics. Due to this
reasons selective harmonic elimination (SHE PWM) method is selected for controlling the
4
switching angle of cascaded H-bridge multilevel inverter for transformerless railway traction
drive application.
The space vector PWM method and sinusoidal PWM method will cause high switching loss and
low efficiency due to the high switching frequency. A disadvantage for the space vector PWM
method and the space vector control method is that they cannot be applied to multilevel
converters with unequal DC voltages.
For current-source drives, two topologies have found industrial applications in high power ranges:
the load-commutated inverter (LCI) and the PWM-CSI. The LCI has been utilized for many years
presenting simple converter topology, low manufacturing cost, and reliable operation. Its main
problems include low input power factor and distorted input current waveforms, which these
problems are overcame by the newer technology of PWM-CSI [13].
5
Medium voltage drives
cycloconverter
ac
Current source Voltage source
PWM current
source inverter
Multilevel inverter High power
Load commutated
2-Level
inverter
Flying
capacitor Symmetric (equal
DC source)
Asymmetric
(unequal DC)
)source)
Fig 1.1: Classification of converters topologies for high-power drives
On the other hand, high-power voltage-source inverters (VSI), have attracted more markets and
developed significantly over the last decades, compared to current-source topologies. The
voltage-source drives are divided in two groups of high-power two-level and multilevel inverters.
The simplest topologies of VSIs are single-phase half-bridge inverter which generates a 2 level
square-wave voltage and full-bridge inverter which generates 3-level waveform. These classical
inverters were limited to low or medium power applications due to the limitations on the power
6
semiconductor voltages and currents. The series connection of semiconductor switches enabled
high two level VSI for high-power applications.
However, the addition of some other power components, like diodes or capacitors, and utilization
of more complex switching methods permitted a more interesting use of these additional
components to enhance the quality of input and output currents and voltages, originating the
family of multilevel VSI technology.
Although multilevel inverters were basically developed to reach higher voltage operation, before
being restricted by semiconductor limitations, the extra switches and dc sources (supplied by dc-
link capacitors) could be used to generate different voltage levels, enabling the generation of
stepped waveform with less harmonic distortion, reducing dv/dt and common-mode voltages.
These characteristics have made them popular for high-power medium-voltage applications but
the large number of semiconductor switches in these inverters, result in a reduction both of the
reliability and efficiency of the drive [14]. Therefore, many power electronic researchers have
made great effort in developing multilevel inverters with the same benefits and less number of
semiconductor devices.
Nowadays, the induction motors are used in the inverter-driven electric train. It can withstand the
mechanical shock, high temperature and the vibration due to the harmonics present in the supply
voltage of the inverter. Due to the absence of the brushes in IMs, the maintenance is low and the
weight is reduced. It has better torque characteristics [15]. These favorable characteristics of IMs
served as a motivation to use IMs in the traction drive. First, the silicon-controlled rectifier (SCR)
or thyristor was developed, which made the voltage-regulation requirement for DC traction
7
motors much easier. Then the gate-turnoff thyristor was developed, which made the three-phase
induction-motor inverter drive become a standard. Now the insulated-gate bipolar transistor
(IGBT) is taking over from the GTO and is expected to consolidate the supremacy of the inverter
drive and further improve its performance.
stock from single phase 15 or 25 kV with 60, 50, or Hz frequencies. The Ethiopian under
construction railway network also supply the rolling stock from single phase 25kV with 50Hz
frequency. Therefore, the railway vehicles are equipped with a transformer to reduce the line
voltage to a more convenient level for the motor and semiconductors. The transformer used in the
electric locomotive is the most heavy and expensive equipment. It requires a large floor space and
harmonic currents in the transformer lead to increase in temperature because of the losses. The
alternating magnetic field caused by the harmonic current lead to unwanted noise and vibration,
thus making it uncomfortable for the passengers. So this thesis proposed to eliminate
magnetically coupled bulky transformer using cascaded multilevel converter method. This thesis
typically focused on the analysis and modeling of cascaded multilevel inverter for transformerless
electric railway traction drive part to get the desired voltage and frequency for induction motor
drive application. This transformerless drive increases efficiency and reduces the cost of the
traction drive.
8
1.4. Objective
It reduce the cost, weight, large floor space , unwanted noise and vibration effect by using
power electronics transformer rather than using the conventional type transformer. It also
increases the efficiency of the traction drive.
Since multilevel voltage source inverter is recently applied in many applications such as
ac power supplies, static VAR compensators, flexible AC transmission systems,
renewable energy sources, active power filters and drive system, so having knowledge in
such area can contribute to transfer technology of the above listed applications in
Ethiopia.
The study mainly focuses on analyzing and modeling of cascaded H-bridge multilevel
inverter drive for electric railway traction application.
9
1.7. Methodology
Literature review: A number of journals, article and papers on different types of multilevel
inverter and modulation techniques for electrical train drive application have been reviewed.
Data collection: Standard data that related to Ethiopian electrical railway train system have been
collected from Ethiopian Railways Corporation (ERC).
Analysis and modeling of eleven level cascaded H-bridge multilevel inverter: CHBMLI with
SHEPWM modulation technique have been modeled for electrical train drive application. The
switching angles have been analyzed using Newton-Raphson method.
Simulation of eleven level cascaded H-bridge multilevel inverter: using the calculated
switching angles by Matlab/Simulink software, simulation of 11 level CHBMLI have been
performed.
Result and analysis: The simulated results are compared with the results obtained using a
mathematical model to check the validity of the result.
10
Chapter Two
2. Literature review
This section reviews the technical literature related to the topics multilevel inverters and
multilevel modulation techniques dealt in this research work.
Leopoldo G et al(2008)[4] presented that the railway electric traction requires high voltage
operation. This can be fulfilled by the series and/or parallel combination of various semi-
conductor devices. But because of the differences in their inherent characteristics, it will damage
the devices. This limitation can be overcome with the help of multilevel converter. The output
voltage of multilevel inverter has low harmonic content (THD) in comparison to that of a two
level inverter.
S.kouro et al (2010)[3] Proposed with the help of the multilevel inverter, the transformation of
the voltage level can be done without the help of the bulky transformer. This results in
transformerless traction drives. Thus, the multilevel inverter prevents the motor damage and
thereby increases the efficiency of the drive.
Shahnia and Sharifian(2005) [5] presented that the multilevel inverters can be directly connected
to the high voltage supply and can step down the voltage. Thus, it eliminates the need of the
transformer. This transformerless drive increases efficiency and reduces the cost of the traction
drive.
11
Manjrekar and Steimer (2000)[14] introduced a NPC PWM inverter and examined in detail about
the theory, design and performance of NPC PWM. Diode clamping inverter suffers from DC link
unbalance and indirect clamping of the inner devices. This inverter is unable to control real power
flow in multilevel inverter.
Verma and Thakur (2011)[24] proposed a FCM inverter with experimental results which had
been carried out to prove the validity and effectiveness of the scheme. In this inverter an
excessive number of storage capacitors are required and controlling circuit is complex.
Song et al (2006)[25] presented a generalized multilevel inverter topology with self voltage
balancing. The existing multilevel inverters such as diode-clamped and capacitor-clamped
multilevel inverters can be derived from the generalized inverter topology. Moreover, the
generalized multilevel inverter topology provides a true multilevel structure that can balance each
DC voltage level automatically without any assistance from other circuits. Thus, in principle
providing a complete and true multilevel topology that embraces the existing multilevel inverters.
Meynerd and Foch(1992)[29] presented a general structure for cascaded power converters in
which any numbers of H-bridge cells having any number of voltage levels are in series connected
to form an inverter phase leg. In this paper, it was experimentally verified by cascading a five-
level H-bridge cell with a three-level H-bridge cell.
Rodriguez et al (2002)[31] discussed the most important topologies such as diode clamped
inverter, capacitor clamped and cascaded multi cell with separate DC sources. The most relevant
control and modulation methods for this family of converters are also presented.
Chiasson and Knzie (2003)[41] designed and implemented control of a multilevel converter using
resultant theory which compute the switching angle to produce the required fundamental voltage
while at the same time it cancel out specified higher order harmonics. The experimental results
are validated and compared with simulation results.
S.kouro et al (2010)[3] investigated a survey on cascaded multilevel inverters. The survey of
different topologies, control strategies and modulation techniques used by the inverters are
12
investigated which reveals several remarkable features of cascaded multilevel inverter such as
high degree of modularity, high power quality and the control of power flow in the regenerative
version.
Lee et al (1998)[37] presented the general analytical solution for carrier based PWM to
mathematically identify the harmonic cancellation that occurs in various PWM implementations
and converter topologies. In this paper, the opportunities for harmonic elimination in multilevel
inverter systems by carrier phase shifting are investigated and the optimum phase shift is
identified.
Patel and Hoft (1973)[39] developed cascaded and hybrid inverters using discontinuous PWM
with phase-shifted carriers within each H-bridge of each multilevel phase leg. The new
modulation strategy gives a similar improved harmonic performance for cascaded and hybrid
inverters.
Chiasson et al (2004)[40] presented analytical solutions for determining the spectral
characteristics of multicarrier based multilevel PWM. The proposed control strategy is to reduce
the undesirable harmonics in order to improve the performance and efficiency of multilevel
inverter. The analytical solutions have been cross verified extensively with simulation results
match with each other.
N.V.Suryanarayan (2005)[22] proposed a novel multilevel carrier based PWM control method for
GTO inverter in low index modulation region. Two PWM control based on adding a bias
13
reference voltage of the inverter and switching patterns are presented. The fluctuations of the
neutral point voltage are also reduced using this method.
Enjeti and Lindsay(2003)[7] proposed a novel algorithm by which switching angles can be
obtained in real-time in the context of step modulation for multilevel inverters. With the
algorithm, the voltage Total Harmonic Distortion (THD) is minimized, which is proven by the
mathematical derivation and verified by the THD comparison.
Maswood et al(2001)[8] presented that Selective Harmonic Elimination Pulse Width Modulation
(SHE-PWM) has been intensively studied in order to achieve low THD. The common
characteristic of the SHE-PWM method is that the waveform analysis is performed using Fourier
theory. Sets of non-linear transcendental equations are then derived, and the solution is obtained
using an iterative procedure, mostly by Newton-Raphson method.
The above literature review reveals that there exists a need for reducing the total harmonic
distortion and enhance the output voltage in multilevel inverter. Hence, in this work, minimizing
the total harmonic distortion and enhancing the output voltage is performed using SHEPWM
modulation techniques for railway traction drive application.
14
Chapter Three
3. Electric Traction
3.1. Introduction
The development of electric and electric vehicles will offer many new opportunities and
challenges to the power electronics industry, especially in the development of the main traction
motor drive. Many current and future designs will incorporate the use of induction motors as the
primary source for traction in electric vehicles.
Large electric drives will require advanced power electronic inverters to meet the high power
demand (1-50MW) required of them [3]. The cascaded inverter with separate dc source closely
fits the needs of all-electric vehicles because it can use the onboard batteries or fuel cells to
synthesize a sinusoidal voltage waveform to drive the main vehicle traction motor. Where
generated ac voltage is available, a back-to-back cascaded H-bridge multilevel converter can be
used to output variable frequency ac voltage for driven motor.
Multilevel inverters also solve problems with present 2-level pulse width modulation (PWM)
adjustable speed drives (ASD‟s). Adjustable speed drives usually employ a front-end diode
rectifier to convert utility ac voltage to variable frequency and variable voltage for motor control.
In this chapter, the railway electrification and its traction drive has been discussed. Comparison
of the steam and diesel traction drive has been done with the electric traction. Different traction
units have been discussed. Traction motor and various other issues have been studied here.
In earlier days, the steam engine drives and the diesel engine drives were used [16]. Their
characteristics and drawbacks are discussed below.
15
o No Electromagnetic interference
But it has the following limitations:
o Even though the train may be idle, the fire has to be banked. This results in low thermal
efficiency.
o It requires adequate supply of feed water at regular interval. Also the supply of water and
coal add to the cost.
o To put the steam engine into operation, steaming time is required.
o The steam engine is greatly influenced by the firing rate of the coal.
o Power weight ratio is low.
o Its centre of gravity is high due to the presence of boiler.
o It causes environmental disaster with high carbon emissions.
o The man power requirement is more as compared to the electric drive.
o It cannot be used in underground railways because of the smoke.
16
Similarly, the electric drive has some limitations. These include:
17
solutions. Firstly some countries devoted a specific power system to their ERs and made other
customers isolated from power quality disturbances produced by railway. The other solution is to
use DC transmission systems for trains. This strategy is working and different standard and a few
not standard DC voltage levels are used to supply railway system. In this approach railway is
connected to a main power system through an AC\DC rectifier and all of the power turbulences
are cut off at a point of common coupling by this high voltage rectifier. The other solution was to
design railway train to work as AC load but not necessarily with standard frequency of 50/60 Hz.
There are different voltage levels for this type of railway power feeding.
B. Harmonic distortion
There are speed drives, power conversion equipment or frequency converters that inject harmonic
in to railways supplying power system. These harmonics can disturb other power systems or lead
to high frequency electromagnetic fields incompatible with close equipment as well as train
signaling system. There are some classic solutions for harmonic distortion reduction like third
sequence harmonic eliminator transformers. An active power filter is a modern solution for these
problems such as unbalance, harmonic distortion and low power factor problems.
C. Flicker
As the train passes between two adjacent substations voltage sag may happen and affect other
customers electrical light performance so called flicker. There are some structural solutions
regarding to all of these challenges which are used in different countries. In Germany and
Sweden low frequency systems are used which are low frequency power system in Germany and
frequency conversion in Sweden. However in England, France, USA and Africa higher voltages
has reduced the problem in different manners which are connecting railway supply system to
main network in a high voltage point of common coupling or designing railway system to work
under a higher catenary voltage.
18
3.5.3. Input Voltage Source
As mentioned in train systems classification there are some voltage sources which are used based
on trains exact application. Concentrating on input voltage source, railways can be categorized as
below:
A. DC Voltage Source
The main advantage of DC traction system is simplicity in control. DC voltage supplied ERs can
be split in two approaches. Firstly voltage level of the feeders and secondly the measure which is
used to feed the train. Standard DC voltage levels under load over the world are 600VDC, 750
VDC, 1500 VDC, 3000VDC. There are some non-standard voltage levels as well.
In another approach DC ERs can be classified in which main DC power supply is connected to
trains. Overhead line system and Conductor rail system is main sub categories which can be
manufactured in different schemes. Although the feeder structure for voltage level of 3000VDC
and more is limited to overhead lines. DC voltage supplied railway systems are recommendable
for low power applications. Higher DC voltages for higher power railway application are not
practical because of insulation and clearance restrictions. The power system interface of DC
feeder systems is limited to some harmonics injected to power system at substations that can be
attenuated by appropriate filters. The other problem with DC railway supply system is ground
return current that may be cause of corrosion.
B. AC Voltage Source
Standard AC voltage supply for ERs are 25 kV, 50 Hz which suppose to dominate Europe
railways as a regional standard to unify European railway system [20] or 25 kV, 60 Hz for
countries with 60 Hz power system ex. Japan, US and 15kV, 16.7Hz in Germany, Austria and
Scandinavia and 50kV 50/60 Hz or 2x25kV [21], 50/60 Hz using auto transformers for very high
power applications all installed as overhead single phase lines. Due to less insulation restriction
with higher frequency, much higher voltage level is feasible in AC power transmission to trains,
leading to remarkable reduction in power lose.
Distance between substations and power of each substation depending on trains load, speed and
catenary characteristics varies among 20 to 80 miles. Basically the higher feeder voltage the
longer distance between substations is allowed. Figure 3.1 is an illustration of an electric
locomotive. It can be seen that it consists of the transformer, the main rectifier, the main inverter
19
and the three phase AC motors [21]. This thesis is mainly concerned with the main rectifier, main
transformer, an inverter and three phase AC motors. So these are discussed here in brief.
Main transformer is used to step down the supply voltage before it can be used by the train. Main
rectifier is mainly used for the conversion of AC to DC. Main Inverter is mounted on the trains to
provide alternating current from direct current. It is used for three phase drives. Three phase AC
motors are used to generate tractive force to provide accelerate to the train. Earlier DC motors
were used but nowadays, three phase AC motors are used.
20
o They should be protected from dirt and dampness.
All these above requirements are not fulfilled by any single motor. For DC system, DC series and
compound motors are used. For single phase AC system, AC series motor is used and for three
phase system, induction motors are used [22].
DC series motors have high starting torque and variable speed characteristics. For this reason, it is
very much used for electric traction. But because of presence of brushes and commutation
problem, three phase induction motors are used.
Nowadays, the three phase induction motor is generally used as traction motor because of the
following reasons:
o Absence of brushes.
o Robustness and reliability with low maintenance
o Simple cooling arrangement with enclosed frames
o High uniform torque with inherent overload management
o High power to weight ratio
o High voltage operation
o Low cost to power ratio
o High maximum speeds
o Inherent regenerative braking capability
o Steep torque speed characteristics
22
Single phase
1M
Overhead
distribution system
c
50 HZ, 25kV 2M
AC Supply
Cascaded Cascaded
Multilevel Rectifier Multilevel
Inverter
6M
Fig 3.2: Schematic diagram of the drive system of induction motors used in railway traction
system.
23
The multilevel inverter structures are the main focus of discussion in this section; however, the
illustrated structures can be implemented for rectifying operation as well. Although each type of
multilevel converters share the advantages of multilevel voltage source inverters, they may be
suitable for specific application due to their structures and drawbacks. Operation and structure of
some important type of multilevel converters are discussed in the following sections.
In this thesis, three capacitor voltage synthesis-based multilevel inverters are introduced, i.e.
24
Table 3.1: Switching states in one leg of the three-level diode clamped inverter
The voltage across each capacitor for an N level diode clamped inverter at steady state is /
, N-level output phase voltage and a -level output line voltage. Although each
active switching device is required to block only a voltage level of , the clamping diodes
require different ratings for reverse voltage blocking.
Vdc/2 S1
C1
D1
Vdc/4 S2
C2
S3
D‟1 D2
D3 S4
Vdc
a
n
D‟2 S‟1
Vdc/2 S1
C1
Van
C3
D1 S2 S‟2
-Vdc/4
Van
a
Vdc D‟3
D‟1 S‟1 S‟3
C4
C2 -Vdc/2 S‟2 -Vdc/2
S‟4
0 0
(a)
(b)
Fig 3.3: Topology of the diode-clamped inverter (a) three-level inverter, (b) five -level inverter
25
In general for an N level diode clamped inverter, for each leg 2(N-1) switching devices, (N-1) *
(N-2) clamping diodes and (N-1) dc link capacitors are required. By increasing the number of
voltage levels the quality of the output voltage is improved and the voltage waveform becomes
closer to sinusoidal waveform. However, capacitor voltage balancing will be the critical issue in
high level inverters. When N is sufficiently high, the number of diodes and the number of
switching devices will increase and make the system impracticable to implement. If the inverter
runs under pulse width modulation (PWM), the diode reverse recovery of these clamping diodes
becomes the major design challenge.
A. Operation of DCMLI
Fig 3.3(a) shows a three-level diode-clamped converter in which the dc bus consists of two
capacitors, C1, C2. For dc-bus voltage , the voltage across each capacitor is /2 and each
device voltage stress will be limited to one capacitor voltage level /2 through clamping diodes.
To explain how the staircase voltage is synthesized, the neutral point n is considered as the output
phase voltage reference point. There are three switch combinations to synthesize three-level
voltages across a and n.
1) Voltage level = /2, turn on the switches S1 and S2.
2) Voltage level = 0, turn on the switches S2 and S1'
3) Voltage level =- /2 turn on the switches S1',S2'.
Fig 3.3(b) shows a five-level diode-clamped converter in which the dc bus consists of four
capacitors, C1, C2, C3, and C4. For dc-bus voltage , the voltage across each capacitor is /4
and each device voltage stress will be limited to one capacitor voltage level /4 through
clamping diodes.
Table 3.2: Switching states of the five level inverter
26
Four complementary switch pairs exist in each phase. The complementary switch pair is defined
such that turning on one of the switches will exclude the other from being turned on. In this
example, the four complementary pairs are (S1 –S1'), (S2- S2'), (S3 – S3'), and (S4 - S4').
Although each active switching device is only required to block a voltage level of / (m-1), the
clamping diodes must have different voltage ratings for reverse voltage blocking. Using D1' of
Fig 3.3(b) as an example, when lower devices S2'- S4', are turned on, D1' needs to block three
capacitor voltages, or 3 /4, and D1 needs to block /4. Similarly, D2 and D2' need to block
2 /4, and D3 needs to block 3 /4. Assuming that each blocking diode voltage rating is the
same as the active device voltage rating, the number of diodes required for each phase will be (m-
1)*(m-2). This number represents a quadratic increase in m.
B. Features of Diode clamped MLI
1) High-Voltage Rating Required for Blocking Diodes:
Although each active switching device is only required to block a voltage level of /(m - l), the
clamping diodes need to have different voltage ratings for reverse voltage blocking. Using D1' of
Fig 3.3 (b) as an example, when all lower devices, S1'-S4' are turned on, D1' needs to block three
capacitor voltages, or 3 /4. Similarly, D2 and D2' need to block 2 /4, and D3 needs to block
3 /4. Assuming that each blocking diode voltage rating is the same as the active device voltage
rating, the number of diodes required for each phase will be (m - 1) x (m - 2). This number
represents a quadratic increase in m. When m is sufficiently high, the number of diodes required
will make the system impractical to implement.
2) Unequal Device Rating,
In table 3.2 it can be seen that switch S1 conducts only during = , while switch S4
conducts over the entire cycle except during - . Such an unequal conduction duty
requires different current ratings for switching devices. When the inverter design is to use the
average duty for all devices, the outer switches may be oversized, and the inner switches may be
undersized. If the design is to suit the worst case, then each phase will have 2 x (m - 2) outer
devices oversized. In comparison with the traditional transformer coupling multipulse converters
using six-step operation for each converter, such unequal conduction duty is indeed an
advantageous feature because the six-step operation needs maximum duty in each device and
circulating currents between converters through transformers.
27
3) Capacitor Voltage Unbalance:
In most applications, a power converter needs to transfer real power from ac to dc (rectifier
operation) or dc to ac (inverter operation). When operating at unity power factor, the charging
time for rectifier operation (or discharging time for inverter operation) for each capacitor is
different. Such a capacitor charging profile repeats every half cycle, and the result is unbalanced
capacitor voltages between different levels. The voltage unbalance problem in a multilevel
converter can be solved by several approaches, such as replacing capacitors by a controlled
constant dc voltage source such as pulse-width modulation (PWM) voltage regulators or batteries.
The use of a controlled dc voltage will result in system complexity and cost penalties. With the
high power nature of utility power systems, the converter switching frequency must be kept to a
minimum to avoid switching losses and electromagnetic interference (EMI) problems. When
operating at zero power factors, however, the capacitor voltages can be balanced by equal charge
and discharge in one-half cycle. This indicates that the converter can transfer pure reactive power
without the voltage unbalance problem
C. Advantages and Disadvantages of DCMLI
Advantages:
o All of the phases share a common dc bus, which minimizes the capacitance requirements
of the converter. For this reason, a back-to-back topology is not only possible but also
practical for uses such as a high-voltage back-to-back inter-connection or an adjustable
speed drive.
o The capacitors can be pre-charged as a group.
o Efficiency is high for fundamental frequency switching.
o When the number of levels is high enough, harmonic content will be low enough to avoid
the need for filters
Disadvantages
o Real power flow is difficult for a single inverter because the intermediate dc levels will
tend to overcharge or discharge without precise monitoring and control.
o The number of clamping diodes required is quadratic ally related to the number of levels
[1], which can be cumbersome for units with a high number of levels.
28
3.8.2. Flying Capacitor Structure
The capacitor clamped inverter alternatively known as flying capacitor was proposed by Meynard
and Foch in 1992 [30]. The structure of this inverter is similar to that of the diode-clamped
inverter except that instead of using clamping diodes, the inverter uses capacitors in their place.
The flying capacitor involves series connection of capacitor clamped switching cells. This
topology has a ladder structure of dc side capacitors, where the voltage on each capacitor differs
from that of the next capacitor. The voltage increment between two adjacent capacitor legs gives
the size of the voltage steps in the output waveform. Figure 3.4 shows the three-level and five-
level capacitor clamped inverters respectively.
Vdc/2 S1
C4
Vdc/4 S2
C3
C4
C3 S3
C3 S4
n C2
Vdc
a
C1
S‟1
C2 S1 C2
Van
C4
S2 S‟2
-Vdc/4
Van
n C1 a
Vdc
S‟1 S‟3
C4
C2 S‟2 S‟4
0 -Vdc/2
0
(a)
(b)
Fig 3.4: FCMLI circuit topologies, (a) 3-level inverter (b) 5- level inverter
29
A. Operation of FCMLI
In the operation of flying capacitor multi-level inverter, each phase node can be
connected to any node in the capacitor bank . Connection of the a-phase to positive
node V3 occurs when S1 and S2 are turned on and to the neutral point voltage when S2 and S1‟
are turned on. The negative node V1 is connected when S1' and S2' are turned on. The clamped
capacitor C1 is charged when S1 and S1' are turned on and is discharged when S2 and S2' are
turned on. The charge of the capacitor can be balanced by proper selection of the zero states. In
comparison to the three-level diode-clamped inverter, an extra switching state is possible. In
particular, there are two transistor states, which make up the level V3. Considering the direction
of the a-phase flying capacitor current for the redundant states, a decision can be made to
charge or discharge the capacitor and therefore, the capacitor voltage can be regulated to its
desired value by switching within the phase. As with the three-level flying capacitor inverter, the
highest and lowest switching states do not change the charge of the capacitors. The two
intermediate voltage levels contain enough redundant states so that both capacitors can be
regulated to their ideal voltages.
Similar to the diode clamped inverter, the capacitor clamping requires a large number of bulk
capacitors to clamp the voltage. Provided that the voltage rating of each capacitor used is the
same as that of the main power switch, an N level converter will require a total of (N-1) * (N-
2) / 2 clamping capacitors per phase in addition to the (N-1) main dc bus capacitors.
Unlike the diode-clamped inverter, the flying-capacitor inverter does not require all of the
switches that are on (conducting) in a consecutive series. Moreover, the flying-capacitor inverter
has phase redundancies, whereas the diode-clamped inverter has only line-line redundancies.
These redundancies allow a choice of charging/discharging specific capacitors and can be
incorporated in the control system for balancing the voltages across the various levels.
The voltage synthesis in a five-level capacitor-clamped converter has more flexibility than a
diode-clamped converter. Using Fig 3.4 (b) as the example, the voltage of the five-level phase leg
a output with respect to the neutral point n (i.e. ), can be synthesized by the following switch
combinations.
1) Voltage level = /2, turn on all upper switches S1 - S4.
2) Voltage level = /4, there are three combinations.
30
a) Turn on switches S1, S2, S3 and S1‟.
3) Voltage level = 0, turn on all upper switches S3, S4 and lower switches S1‟, S2‟.
4) Voltage level =- /4, turn on all upper switches S1 and lower switches S1‟,S2‟ and
S3‟
5) Voltage level =- /2, turn on all lower switches S1‟ , S2‟,S3‟ and S4‟
B. Features of FCMLI
The major problem in this inverter is the requirement of a large number of storage capacitors.
Provided that the voltage rating of each capacitor used is the same as that of the main power
switch, an m-level converter will require a total of (m - 1) x (m - 2)/2 auxiliary capacitors per
phase leg in addition to (m - 1) main dc bus capacitors. With the assumption that all capacitors
have the same voltage rating, an m-level diode-clamp inverter only requires (m - 1) capacitors. In
order to balance the capacitor charge and discharge, one may employ two or more switch
combinations for middle voltage levels (i.e., 3 /4. /2, and /4) in one or several
fundamental cycles. Thus, by proper selection of switch combinations, the flying-capacitor
multilevel converter may be used in real power conversions. However, when it involves real
power conversions, the selection of a switch combination becomes very complicated, and the
switching frequency needs to be higher than the fundamental frequency. In summary, advantages
and disadvantages of a flying capacitor multilevel voltage source converter are as follows.
Advantages
Compared to the diode-clamped inverter, this topology has several unique and attractive features
as described below:
o Added clamping diodes are not needed.
o It has switching redundancy within the phase, which can be used to balance the flying
capacitors so that only one dc source is needed.
31
o The required number of voltage levels can be achieved without the use of the transformer.
This assists in reducing the cost of the converter and again reduces power loss.
o Unlike the diode clamped structure where the series string of capacitors share the same
voltage, in the capacitor-clamped voltage source converter the capacitors within a phase
leg are charged to different voltage levels.
o Real and reactive power flow can be controlled.
o The large number of capacitors enables the inverter to ride through short duration outages
and deep voltage sags.
Disadvantages
o Converter initialization i.e., before the converter can be modulated by any modulation
scheme the capacitors must be set up with the required voltage level as the initial
charge. This complicates the modulation process and becomes a hindrance to the
operation of the converter.
o Control is complicated to track the voltage levels for all of the capacitors.
o Precharging all of the capacitors to the same voltage level and startup are complex
o Switching utilization and efficiency are poor for real power transmission.
o Since the capacitors have large fractions of the dc bus voltage across them, rating of
the capacitors are a design challenge.
o The large numbers of capacitors are both more expensive and bulky than clamping
diodes in multilevel diode-clamped converters.
o Packaging is also more difficult in inverters with a high number of levels.
a
C1
a
C1 C2
a
C1 n C2 n C3 n
Fig 3.5: Single phase structures of Cascaded inverter (a) 3-level, (b) 5-level, (c) 7-level
33
A. Operation of CMLI
The converter topology is based on the series connection of single-phase inverters with separate
dc sources. Fig 3.5 shows the power circuit for one phase leg of a three-level , five-level and
seven-level cascaded inverter. The resulting phase voltage is synthesized by the addition of the
voltages generated by the different cells. In a 3-level cascaded inverter each single-phase full
bridge inverter generates three voltages at the output: + , 0, - (zero, positive dc voltage,
and negative dc voltage). This is made possible by connecting the capacitors sequentially to the
ac side via the power switches. The resulting output ac voltage swings from - to + with
three levels, -2 to +2 with five-level and -3 to +3 with seven-level inverter. The
staircase waveform is nearly sinusoidal, even without filtering. For a three-phase system, the
output voltage of the three cascaded converters can be connected in either wye (Y) or delta (Δ)
configurations. Fig 3.6 shows an IGBT based three phase eleven level cascaded inverter which is
used in the modeling of the traction drive.
A B C
Fig 3.6: Power circuit of three phase cascaded H-bridge eleven level inverter using IGBT
34
Fig 3.7 shows the output phase voltage waveform of an eleven level inverter. The number of
output phase voltage levels m in a cascade converter with s separate DC sources is m = 2s + 1.
Load balance control for each H-bridge and each DC source can be acquired by rotating the
switching angles to the H-bridges [2].
Vph
π/2 π 2π
ωt
Fig 3.7: Output phase voltage waveform of an eleven level cascaded multilevel inverter
B. Features of CMLI
For real power conversions, (ac to dc and dc to ac), the cascaded-inverter needs separate dc
sources. The structure of separate dc sources is well suited for various renewable energy sources
such as fuel cell, photovoltaic, and biomass, etc. .
In summary, advantages and disadvantages of the cascaded inverter based multilevel voltage
source converter can be listed below
35
C. Advantages and Disadvantages of CMLI
Advantages
o Modular in structure so packing and circuit layout is easier
o No clamping diodes present as in NPC
o No voltage balancing capacitors present in FC
o Low voltage switching devices required
o No EMI problem
o Less common mode voltage
o Less dv/dt
o Suitable for medium voltage , high power applications
o Separate DC sources eliminates the need of the voltage balancing circuits
o With the increase in the number of the level, the staircase waveform approximates to a
sinusoid
o It can work at reduced power level when one of its cell or SDCSs is damaged
o Soft switching techniques can be applied to CMI
o No transformer required as in multi-pulse inverters
o It makes Induction Motor more accessible / safer and open wiring possible for most of an
induction motor power system.
o The regulation of the DC buses is simple
o Modularity of control can be achieved. Unlike the diode clamped and capacitor
clamped inverter where the individual phase legs must be modulated by a central
controller, the full-bridge inverters of a cascaded structure can be modulated separately.
o Requires the least number of components among all multilevel converters to achieve
the same number of voltage levels.
Because of these advantages, the CMI converter has been used in this thesis for traction drive
application.
36
Disadvantages
o Communication between the full-bridges is required to achieve the synchronization of
reference and the carrier waveforms.
o Needs separate dc sources for real power conversions, and thus its applications are
Somewhat limited.
37
Table 3.3: Comparison of power component requirements per phase leg among three
multilevel Inverters
Inverter Diode-Clamp Flying-capacitors Cascaded-inverters
Configuration
Main switching 2 2 2
Devices
Main diodes 2 2 2
Clamping diodes 0 0
DC bus capacitors /2
Balancing 0 /2 0
Capacitors
38
Parameters such as switching frequency, distortion, losses, harmonic generation, and speed of
response are typical of the issues which must be considered when developing modulation
strategies for a particular family of converters.
39
Multilevel Modulation Control
(a) (b)
Fig 3.9: Multi-carrier control (a) control signal and carrier signals; (b) output voltage
40
A number of cascaded cells in one phase with their carriers shifted by an angle and using the
same control voltage produce a load voltage with the smallest distortion [33]. A very common
practice in industrial applications for the multilevel converter is the injection of a third harmonic
in each cell to increase the output voltage [32]. Another advantageous feature of multilevel
SPWM is that the effective switching frequency of the load voltage is much higher than the
switching frequency of each cell, as determined by its carrier signal.
The advantage of the method is it is very simple. But it has two disadvantages. The first is the
method cannot completely eliminate the low order harmonics. Therefore the low order harmonics
cause loss and high filter requirements. The second is the high switching frequency causes high
switching loss and low efficiency.
41
3.9.4. Space Vector Control
A conceptually different control method for multilevel converters, based on the space-vector
theory, has been introduced, which is called space vector control (SVC) [38].
The control strategy works with low switching frequencies but does not generate the mean value
of the desired load voltage in every switching interval, which is the principle of space-vector
PWM method. The key idea in SVC is to deliver to the load a voltage vector that minimizes the
space error or distance to the reference vector.
This method is simple and attractive for high number of levels. The advantages of SVC are the
computation is very easy and the switching frequency is very low, near the fundamental
switching frequency. But the disadvantage is as the number of levels decreases, the error in terms
of the generated vectors with respect to the reference will be higher [31].
V(t) = ω
π
where s is the number of DC sources, and , are the level of DC voltages. The switching
angle must be satisfy the condition 0 π/2. However, if the switching
angles do not satisfy the condition, this method no longer exists. If = = = , this is called
equal DC voltages case. To minimize harmonic distortion and to achieve adjustable amplitude of
the fundamental component, up to s-1 harmonic contents can be removed from the voltage
waveform. In general, the most significant low-frequency harmonics are chosen for elimination
by properly selecting angles among different level converters, and high-frequency harmonic
components can be readily removed by using additional filter circuits. To keep the number of
eliminated harmonics at a constant level, all switching angles must satisfy the condition 0
π/2, or the total harmonic distortion (THD) increases dramatically.
Due to this reason, this modulation strategy basically provides a narrow range of modulation
index, which is one of its disadvantages [31].
42
In order to achieve a wide range of modulation indexes with minimized THD for the synthesized
waveforms, a generalized selective harmonic modulation method [34] was proposed, which is
called virtual stage PWM [1]. An output waveform is shown in Figure 3.10.
Vout
0 ωt
π/2 3π/2 2π
1
3
4
V(t)= π
ω (3.2)
Where s is the number of DC sources, and , are the level of DC voltages. In this
expression, the positive sign implies the rising edge, and the negative sign implies the falling
edge. Similar to the fundamental switching frequency method, the switching angle must be satisfy
the condition 0 π/2. However, if the switching angles do not satisfy the
condition, this method no longer exists.
Therefore, the modulation control problem is converted into a mathematic problem to solve the
following equations for a three-phase system. Here, m is modulation index.
cos cos cos ) cos =m
cos cos cos cos =0
43
switched “on” and “off” several times per fundamental cycle. The switching pattern decides what
the output voltage waveform looks like.
When the multilevel fundamental switching method is used, all of the DC voltages are typically
involved, where all of the switches are turned “on” and “off” only once per fundamental cycle.
The multilevel fundamental switching method also refers to exactly one switching pattern.
For fundamental switching frequency method, the number of switching angles is equal to the
number of DC sources. However, for the Virtual Stage PWM method, the number of switching
angles is not equal to the number of DC voltages. For example, in Figure 3.10, only two DC
voltages are used, whereas there are four switching angles.
Bipolar Programmed PWM and Unipolar Programmed PWM could be used for modulation
indices too low for the applicability of the multilevel fundamental frequency switching method.
Virtual Stage PWM can also be used for low modulation indices. Virtual Stage PWM will
produce output waveforms with a lower THD most of the time [1]. Therefore, Virtual Stage
PWM provides another alternative to Bipolar Programmed PWM and Unipolar Programmed
PWM for low modulation index control.
Vout
Vdc
π/2 π 3π/2 2π
0 ωt
3
2
1
-Vdc
44
Chapter Four
45
S11 S13
A
V1
S14 S12
S21 S23
V2
S24 S22
S31 S33
V3
S34 S32
S41 S43
V4
S44 S42
S51 S53
N
V5
S54 S52
Fig 4.1: Power circuit of single phase cascaded eleven level inverter for traction drive
V0
V1+V2+V3+V4+V5
V1+V2+V3+V4
V1+V2+V3
V1+V2
V1
π/2 π 2π
0 ωt
Fig 4.2: Output voltage of single phase cascaded eleven-level inverter for traction drive
46
4.2. Working principle of Eleven Level Inverter
The model output voltage waveform of eleven-level cascaded multilevel inverter is shown in the
fig 4.2. The maximum output phase voltage is given as .
The steps to synthesize the eleven-level voltage waveforms are as follows.
1. For an output voltage level 0, no switch in the H-bridges is turned on.
2. For an output voltage level , turn on the switches , , , , , .
3. For an output voltage level , turn on the switches as mentioned in step 2 and
47
Table 4.1: Output voltage levels and their switching states for eleven-level inverter
Switches
1 1 1 1 1
1 1 1 1 1
0 0 0 0 0
0 0 0 0 0
0 1 1 1 1
1 1 1 1 1
0 0 0 0 0
0 0 0 0 0
0 0 1 1 1
1 1 1 1 1
0 0 0 0 0
0 0 0 0 0
0 0 0 1 1
1 1 1 1 1
0 0 0 0 0
0 0 0 0 0
0 0 0 0 1
1 1 1 1 1
0 0 0 0 0
0 0 0 0 0
48
waveform including the level zero. As per the Fourier theorem the periodic output voltage V (ωt)
can be described by a constant term plus an infinite series of sine and cosine terms of frequency
nω, where n is an integer.
Therefore V (ωt) in general, can be expressed as
(4.1)
For periodic functions with odd quarter-wave symmetry, it can be shown that the equations
determining the Fourier coefficients simplify to the following expressions [43]:
(4.2)
for all n, (4.3)
for n even, (4.4)
and
When a periodic function possesses odd quarter-wave symmetry, the average value is zero. The
reason is due to the fact the function is odd. Also, odd symmetry results in all of the cosine
harmonics being zero. The half-wave symmetry of the periodic function forces the even sine
harmonics to be zero. Furthermore, in order to determine the amplitude of the odd sine
harmonics, one only needs to integrate over one-fourth of a fundamental period.
Since the output of the multilevel inverter is odd quarter-wave, (4.2) thru (4.5) can be used in
calculating the Fourier coefficients. As one can see from these equations, only the odd sine
harmonics can be nonzero.
Using the output of the multilevel inverter given in Figure 4.2, equation (4.5) then becomes
ω ω ω ω ω ω
ω ω
ω ω
49
Assuming all sources are of equal value , equation (4.6)
becomes
ω ω ω ω ω ω
ω ω ω ω
Where and are the switching angles and is the voltage of each dc source.
Performing the required integrations, (4.7) then becomes
ω ω ω
π π π π
π
ω ω
π
π
Finally, using the fact that when n is an odd integer and simplifying, (4.9)
becomes
or
Finally, the Fourier series of the quarter-wave symmetric s H-bridge cell multilevel waveform is
written as follows
(ωt) =
50
Where
is the switching angles, which must satisfy the following condition
3. The amplitude of all even harmonics equal zero. Thus, only the odd harmonics in the
quarter-wave symmetric multilevel waveform need to be eliminated. The switching angles
of the waveform will be adjusted to get the lowest THD in the output voltage based on
IEEE-519 standards.
A set of nonlinear equation of single phase eleven-level inverter corresponding to equation (4.13)
can be given as follows:
The nonlinear equation of the fundamental component:
= (4.14a)
= (4.14b)
= (4.14c)
= (4.14d)
= (4.14e)
51
For an output voltage waveform of five full-bridge cell circuit, five switching angles, namely
and , need to be known. It is implied that five equations are required to solve such
switching angles. Importantly, the expected solutions must be less than π/2.
From nonlinear equations system (4.14), equation (4.14a) is the nonlinear equation of the
fundamental component of an output voltage waveform, whereas (4.14b), (4.14c), (4.14d), and
(4,14e) are the nonlinear equation of the , the , the and the component of the
output voltage, respectively. On the right hand side of these equations, they are the amplitudes of
the components, which can be controlled. Basically, the lowest odd harmonic components should
be eliminated from a single-phase system. In a three phase system, because of 120 electrical
degree phase shift, the lowest non-triplen harmonic components need to be removed from phase
voltage. All even harmonics are not existed because of the symmetric characteristic of the
waveform. To control the amplitude of the fundamental component, (4.14a) will be considered.
The modulation index for the multilevel waveform is given as
M= (4.15)
where
h1 is the amplitude of the fundamental component.
s is the number of dc sources or H-bridge cells.
and is the voltages of the dc sources.
From (4.14a) and (4.15), the following equation can be obtained.
π
= (4.16)
From equation (4.16), varying the modulation index value can control the amplitude of the
fundamental component. The other s-1 nonlinear equations, which are the undesirable harmonic
components, can be eliminated. Usually, they are set to be zero. Therefore, in a single phase
multilevel inverter, the lowest s-1 odd harmonics can be removed, and in a three phase multilevel
inverter, the lowest s-1 non-triplen harmonics can be eliminated. The phenomenon is
mathematically proven as follows. Consider the third harmonic of the produced output.
= (4.17)
The voltages in the other two phases are shifted by 120 degrees each, and the equation is given as
follows.
52
=
= (4.18)
and
= ) (4.19)
Hence
= =
In a three-phase system,
= =0
Similarliy,
= =0
Similarly, the other triplen harmonics cancel out each other. Hence, in a three phase system, one
has to deal only with the other odd number harmonics.
Table 4.2 shows the number of harmonic contents which can be eliminated as function of the
number of dc sources, s, in both single-phase and three-phase system.
Table 4.2: The number of eliminated harmonic components of SHEPWM technique.
Single-phase Three-phase
The number of dc sources S s(per phase)
The number of output 2s + 1 phase voltage 4s + 1 line to line voltage
Voltage levels levels
The harmonics in the output The lowest s -1 odd harmonics The lowest s -1 non-triplen
Phase voltage, which need Odd harmonics
To be eliminated
53
4.3.3. The Newton-Raphson Method
The set of equation in (4.14) are nonlinear in nature Newton Raphson method is used to solve
these equation [28], initially guesses are taken and then following steps are perform to calculate
the switching angles and iterations are perform till the harmonics content is reduce up to four
decimal places, which is suitable for implementation in a computer program.
Generally, the system of nonlinear equation in s variables can be represented as
=
=
= (4.20)
Then, (4.13) can be written in vector notation as
F =k (4.21)
Where
F=
k=
and F, , K are s x1 matrices
In a multivariable nonlinear system, a set of independent variables is formed in matrix format and
the statement of the algorithm of Newton‟s method can be shown as follows:
1) Guess a set of initial values for with j =0
Suppose = (4.22)
+ d =k (4.24)
54
Where
and
d =
4) Solve d from (4.24) by
d = INV (4.25)
55
4.3.4. The Total Harmonic Distortion Calculation
The total harmonics distortion (THD) is mathematically given by
THD = (4.27)
Where
is the amplitudes of the fundamental component, whose frequency is and
is the amplitudes of the harmonics at frequency
From (4.11), the amplitude of the fundamental and harmonic components of the quarter-wave
symmetric multilevel waveform can be express as:
= (4.28)
THD = (4.29)
THD (4.30)
By using (4.30), therefore, output voltage THD of multilevel inverter waveform can be
calculated. Theoretically, to get exact THD, infinite harmonics need to be calculated.
However, it is not possible in practice. Therefore, certain number of harmonics will be given. It
relies on how precise THD is needed. Usually, n = 63 is reasonably accepted.
56
4.4. Analysis of Switching Angles and Output THD of 11-Level CHBMLI by
Newton-Raphson Method
In this thesis, 7-level, 9-level and 11-level CMLIs are employed to generate ac output voltage
producing different magnitudes of THD at different values of modulation indices for comparison
purpose. The THD produced by 7-level and 9-level inverter is more than IEEE-519 standard [9]
for all values of m. The THD produced by 11-level inverter is satisfying this standard for
modulation index of one (not for all values), 11-level cascaded H-bridge multilevel inverter is
selected for induction motor drive of traction application at modulation index of one.
By using the Newton-Raphson method and MATLAB program, the switching angles of the
output voltage waveform and output THD are calculated. The output waveform of eleven-level
cascaded H-bridge multilevel inverter is shown in Fig 4.3. Since three phase induction motors are
used in traction system, three-phase system is assumed, and modulation index of output voltage,
m, is 1.
From the waveform shown in Fig 4.3, five unknowns, , , , and need to be known .
Because of a three-phase system, the lowest four odd harmonics, i.e the , , and
, should be eliminated.
V0
π/2 π 2π
ωt
4
1
2
3
57
To control the fundamental amplitude and eliminate the harmonics, five nonlinear
equations can be set up as follows:
= (4.32a)
2) The nonlinear system matrix,
= (4.32b)
and
(4.32c)
T= (4.32d)
Then, equations (4.31a) to (4.31e) can be rewritten in the following matrix format:
F =T (4.33)
58
By using matrices (4.32a) to (4.32d) and the Newton‟s method, the statement of algorithm, which
is suitable for implement in a computer program, is shown as follows:
1) Guess a set of initial values for with j = 0
Assume = (4.34)
2) Calculate the value of
F = (4.35)
3) Linearize equation (4.33) about
+ =T (4.36)
and
d =
4) Solve d from equation (4.36), i.e.
d = INV (4.37)
To implement this algorithm in a computer, MATLAB programming is used (see Appendix A).
After running the program, each iteration result is tabulated in table 4.3. In this case, the process
is repeated until all elements in matrix are less than 0.0001. The final solutions, in degree, are:
7.8649, 19.3815, 29.6656, 47.7057, 63.2464 (4.40)
Because these switching angles satisfy the condition (4.39), they are the correct solutions. By
using these switching angles in the waveform, the , , and harmonics will be
eliminated with the modulation index equal to 1. As the results, the first harmonics content,
which shows in the waveform, is the harmonic.
59
Table 4.3: content of matric , , and from MATLAB program by using the Newton‟s
method
Iteration index
(degree)
j =0 5 4.2165 1.4662
13 -0.0972 7.8433
25 0.8855 4.7000
40 -0.3869 6.3987
55 0.3166 10.5566
Using the same procedure, the switching angles of 7-level, 9-level, and 11-level with different
modulation index, are shown in Table 4.4 to 4.6, respectively.
Table 4.4: The switching angles of 7-level cascaded inverter.
Modulation
Index, M
1 11.6817 31.1783 58.5774
0.95 13.8210 37.2095 61.9509
0.90 17.5104 43.0523 64.1395
0.85 22.7654 49.3798 64.5562
0.80 29.2355 54.4383 64.4844
0.75 34.8935 54.4622 68.5500
0.70 38.3413 53.9297 73.9648
0.65 39.3876 55.5215 78.8979
0.60 39.4298 58.5839 83.1042
60
Table 4.5: The switching angles of 9-level cascaded inverter.
Modulation
Index, M
1 10.0154 22.1424 40.7521 61.7681
0.95 11.5499 27.3939 46.725 64.442
0.90 12.4367 34.5875 48.8091 68.8855
0.85 19.0991 39.7221 55.5860 66.9784
0.80 24.6998 45.5307 57.0398 68.8886
0.75 30.0144 49.2484 57.1585 72.8307
0.70 36.1183 47.8768 61.0723 76.2975
0.65 34.4790 51.4007 62.9388 83.8302
0.60 37.0314 51.023 67.1599 86.0159
Modulation
Index, M
1 7.8598 19.3725 29.6522 47.68 63.2122
0.95 13.3907 20.9886 36.299 58.915 59.3507
0.90 7.6592 27.5706 40.7891 52.5598 73.0391
0.85 17.7312 32.7053 50.0119 57.8089 68.37
0.80 22.3419 39.2785 52.6866 59.3192 70.9645
0.75 28.4564 45.3922 50.9961 63.5194 73.2291
0.70 34.3709 44.6208 54.1495 65.3723 77.917
0.65 35.7274 44.883 56.8276 68.0465 83.6168
0.60 35.3424 46.9528 58.5799 72.6121 87.8373
The phase and line-to-line voltage THD are calculated by using the following formula
THD = (4.41)
Where
s is the number of H-bridge inverters per phase
n is the harmonic order
, ,… are the calculated switching angles
In case of line voltage THD calculation using equation (4.41), all triplen harmonic components
are set to zero. To get a precise result, the harmonic components up to the are used to
calculate the voltage THD. From the calculated switching angles in table 4.4 to 4.6, the line and
61
phase voltage THD of 7-level, 9-level, and 11-level inverter as function of the modulation index,
are shown in table 4.7 to 4.9.
7 level(s=3)
Modulation index M THDphase(%) THDline(%)
1 12.2146 7.8950
0.95 15.4137 8.4329
0.90 20.4685 12.2248
0.85 27.9162 9.2206
0.80 36.7156 10.8829
0.75 42.0658 11.1212
0.70 45.2950 12.4183
0.64 46.8596 12.1812
0.60 47.3293 12.9483
9 level(s=4)
Modulation index M THDphase(%) THDline(%)
1 9.3633 6.4238
0.95 12.7828 7.8288
0.90 17.0568 8.6673
0.85 25.3925 8.2808
0.80 32.7299 7.3595
0.75 38.6554 9.6889
0.70 43.5572 8.6434
0.64 42.6945 9.889
0.60 44.6478 9.4710
62
Table 4.9: Voltage THD of 11- level cascaded inverter
11 level(s=5)
Modulation index M THDphase(%) THDline(%)
1 7.5873 5.0623
0.95 15.4734 5.9535
0.90 13.9569 6.5977
0.85 24.7241 4.9028
0.80 31.4369 7.1752
0.75 38.2359 6.6307
0.70 42.6908 5.8025
0.64 43.4610 7.2716
0.60 43.2388 7.2666
The line and phase voltage THD of 7-level, 9-level, and 11-level inverter as function of the
modulation index, are shown in Fig 4.4 to 4.6, respectively.
30
25
20 Line Voltage
15 Phase voltage
10
5
0
1 0.95 0.9 0.85 0.8 0.75 0.7 0.64 0.6
Modulation Index
Fig 4.4: Voltage THD as function of the modulation index of 7-level cascaded H-bridge
Inverter
63
Voltage THD Vs. Modulation Index of 9-level inverter
50
45
40
35
THD(%)
30
25
20 Line Voltage
15 Phase voltage
10
5
0
1 0.95 0.9 0.85 0.8 0.75 0.7 0.64 0.6
Modulation Index
Fig 4.5: Voltage THD as function of the modulation index of 9-level cascaded H-bridge
Inverter
30
25
20 Line Voltage
15 Phase voltage
10
5
0
1 0.95 0.9 0.85 0.8 0.75 0.7 0.64 0.6
Modulation Index
Fig 4.6: Voltage THD as function of the modulation index of 11-level cascaded H-bridge
Inverter
64
From the figure 4.4 to 4.6, the phase voltage THD increases dramatically, when the modulation
index decreases, whereas, the line voltage THD increases slightly, when the modulation index
decreases. At the same modulation index, the line voltage THD is much less than phase voltage
THD. Similarly, the line voltage THD decreases, when the number of H-bridge inverter per phase
s, increases. Therefore, the line voltage THD can be reduced by increasing the number of
full-bridge inverter in the multilevel inverter circuit.
From figure 4.6, without any filter circuit, the line voltage of 11-level SHEPWM THD is 5.0623
percent with modulation index of 1. In practice, filter circuits may be applied; therefore, the line
voltage THD can meet IEEE-519 standard [9] without any difficulties.
65
Chapter Five
66
Discrete,
Ts = 5e-005 s.
C
powergui PgA11 PgA13
g
C
C
SA11 SA13
PgB11
SB11
PgB13 A A
g
C
C
PgC11 PgC13
C
E
SB13 SC11
E
SC13 B B
E
C C
PgA14 Three-Phase
DCA1 PgB14 PgC14 Series RLC Branch
PgA12 DCB1 PgB12
DCC1
C
PgC12
g
C
C
SA12
C
SA14 SB12
g
C
SB14
C
E
SC12
E
E
SC14
E
E
+
v
-
Voltage Measurement2
Scope
g
PgA21 PgA23
g
g
C
C
SA21 PgB21 PgB23
g
C
SA23 SB21
C
SB23 PgC21 PgC23
C
E
E SC21
E
SC23
E
PgA24
PgB24
DCA2 DCB2
PgC24
PgA22
PgB22 DCC2 PgC22
g
C
g
g
C
C
SA22
C
SA24 SB22
g
C
SB24
C
E
SC22
E
E
SC24
E
E
g
PgA31 PgA33
g
g
C
C
SA31 PgB31 PgB33
g
C
SA33 SB31
C
SB33 PgC33
C
E
PgC31 SC31
E
SC33
E
PgB34
PgA32 PgB32
PgA34
g
C
g
g
C
g
C
g
C
E
DCC3SC34 SC32
E
E
E
E
E
PgA41 PgA43
g
g
C
SA43
C
SB41 SB43 PgC41 PgC43
C
E
SC41
E
SC43
E
E
PgA44 PgB44
DCA4 PgA42 DCB4
PgB42 PgC44
PgC42
g
DCC4
g
g
C
SA42
g
SA44 SB42
g
C
SB44
g
C
E
SC42
E
SC44
E
E
E
PgA51
g
PgA53
g
SA51 PgB51
g
SA53 PgB53
g
SB51 PgC51
g
C
E
SB53 PgC53
g
C
E
SC51
E
SC53
E
PgA54
DCA5 PgA52
PgB54
DCB5 PgC54
g
PgB52
g
SA52 DCC5
g
SA54 PgC52
g
SB52
g
C
E
SB54
g
C
E
SC52
E
SC54
E
E
E
Figure 5.1: Simulation circuit for eleven level inverter using SHE PWM technique
67
Signal(Magn) Gating pulse for Switch S11,S12,S22,S32,S42,S52
1
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Positive Voltage Gating Pulse Pattern
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Signal(Magn.) Gating Pulse for Switch S31
1
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Signal(Magn.) Gating Pulse for Switch S41
1
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Signal(Magn.) Gating Pulse for Switch S51
1
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Time(sec)
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Nagative Voltage Gating Pulse Pattern
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Signal(Magn.) Gating Pulse for Switch S33
1
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Signal(Magn.) Gating Pulse for Switch S43
1
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Signal(Magn.) Gating Pulse for Switch S53
1
0.5
0
0 0.01 0.02 0.03 0.04 0.05 0.06
Time(sec)
68
Line Voltage(Vac)
400
300
200
Phase Voltage Van(V)
100
-100
-200
-300
-400
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time(sec)
(a)
200
-200
1.5
0.5
0
0 1 2 3 4 5 6 7 8
Harmonic order
(b)
Fig 5.3: (a) Simulated phase voltage and (b) THD analysis for 11-level CMLI at m=1.00
69
800
600
400
Line Voltage Vab(V)
200
-200
-400
-600
-800
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time(sec)
(a)
500
-500
0.3
Mag (% of Fundamental)
0.25
0.2
0.15
0.1
0.05
0
0 1 2 3 4 5 6 7 8
Harmonic order
(b)
Figure 5.4: (a) Simulated line voltage and (b) THD analysis for 11-level CMLI at m = 1.00
70
5.1. Simulation of Three phase induction motor driven by 11 level CHBMLI
At this study a three phase squirrel cage asynchronous machine modeled in a selectable dq
reference frame is selected from the library of MATLAB/Sims Power which is used as a load for
cascade 11-level multilevel inverter. Rotor current, stator Current and Rotor Speed are used for
comparing the effect of harmonic with ideal AC source fed induction motor.
Figure 5.5 to 5.7 shows the outputs of an ideal case with ideal AC source. The specification of
induction motor used in this simulation is shown in appendix B
30
20
Rotor Current (A)
10
-10
-20
-30
0 0.05 0.1 0.15 0.2 0.25
Time(sec)
30
20
10
Stator Current (A)
-10
-20
-30
0 0.05 0.1 0.15 0.2 0.25
Time(sec)
71
160
140
Rotor speed (rpm)
120
100
80
60
40
20
-20
0 0.05 0.1 0.15 0.2 0.25
Time(sec)
72
20
15
10
5
Rotor Current(A)
-5
-10
-15
-20
-25
0 0.05 0.1 0.15 0.2 0.25
Time(sec)
Fig 5.8: Rotor current in case of feeding with cascaded 11-level inverter
25
20
15
Stator Current (A)
10
-5
-10
-15
-20
-25
0 0.05 0.1 0.15 0.2 0.25
Time(sec)
Fig 5.9: Stator current in case of feeding with cascaded 11-level inverter
73
180
160
140
Rotor speed(rpm)
120
100
80
60
40
20
0
0 0.05 0.1 0.15 0.2 0.25
Time(sec)
74
Chapter Six
6.1. Conclusion
The multilevel inverter topology can overcome some of the limitations of the standard two-level
inverter. Output voltage and power increase with number of levels. Harmonics decrease as the
number of levels increase discussed in Chapter 3 and 4. In addition, increasing output voltage
does not require an increase in voltage rating of individual force commutated devices. In the
thesis, several multilevel voltage source inverters and their modulation topologies are introduced.
The current trend of modulation control for multilevel converters is to output high quality power
with high efficiency. For this reason, popular traditional PWM methods and space vector PWM
methods are not the best methods for multilevel converter control due to their high switching
frequency. The selective harmonic elimination method has emerged as a promising modulation
control method for multilevel converters.
For a cascaded H-bridges multilevel inverter, the switching angles can be selected such that the
output voltage THD is less than 5% as per IEEE-519 standard. It has been found that the eleven
level CMLI satisfies this limit at modulation index of one while seven and nine level CMLI
violates this limit at all modulation index value. Where adjustable speed drives are required,
modulation index has to change. Since the train is operated in maximum voltage rate for long
period of time at modulation index of one, it satisfy IEE-519 standard limit. But at starting and
stopping time the train is operated in minimum voltage rate for short period of time at modulation
index of less than one. At this modulation index output voltage THD is violate the standard limit.
Line voltage THD can be reduced by increasing the number of levels, but by considering circuit
complexity and cost, eleven level CMLI inverter at modulation index of one is selected for
electrical railway train application.
The switching angles for cascade multilevel inverters of 7, 9 and 11-level have been computed
for analysis of total harmonic distortions produced in output voltage and complexity in
computation of these angles. It has been found that complexity in computation of switching
angles increases with increase in number of levels. On other part, the THD in output voltage
75
decrease and output voltage increase with increase in number of levels. By using three-phase 11-
level CMLI, 5th,7th , 11th , and 13th order harmonics can be eliminated.
Analytical results are validated with simulation results for 11-level CMLI at modulation index of
one. The multilevel inverter improves output voltage; reduce THD and voltage stresses on
semiconductor switches which can be verified from simulation results. The Cascaded eleven level
inverter fed induction motor drive is simulated using the blocks of Simulink. The simulation
results of rotor current, stator current and rotor speed are presented. The results of 11-level CMLI
AC source drive system are compared with the results of ideal AC source drive system. At Steady
State, Performance of 11-level CMLI at modulation index of one driven induction motor is
similar to the ideal AC source driven induction motor. 11-level CMLI can be used in traction
drive system where adjustable speed drives are required to produce output voltage with reduced
harmonic content.
The cascaded eleven level converter system modeled in this thesis can be used in traction drive
consisting of six induction motors for stepping down the catenary voltage to the rated voltage of
the induction motors. Thus, this model eliminates the necessity of the transformer in railway
traction and hence lowers the cost and floor space and increases the efficiency of the traction
drive.
76
References
Drives,” IEEE Transactions on Industry Applications, vol. 35, no. 1, Jan./Feb. 1999, pp.
36-44.
[3] Samir Kouro, Mariusz Malinowski et. al., “Recent Advances and Industrial Applications of
[5] Farhad Shahnia and B.B.Sharifian, “Harmonic analysis and modelling of transformerless
electric railway traction drives,” 13th International conference on Electrical Drives and
using Resultant Theory, IEEE Trans. Contr. Syst. Technol. 11 No. 3 (May 2003), 345–
354.
[7] Enjeti, P. N.Lindsay, J. F. : Solving Nonlinear Equation of Harmonic Elimination PWM in
Power Control, Electron. Lett 23 No. 12 (June 1987), 656–657.
[8] Maswood, I.Wei, S.Rahman, M. A. : A Flexible Way to Generate PWM-SHE Switching
Patterns using Genetic Algorithms, in Proc. IEEE Applied Power Electronics Conf. Expo.,
2001, pp. 1130–1134.
[9] IEEE Recommended Practices and Requirements for Harmonic Control in Electrical Power
Systems, IEEE standard 519-1992
[10] P. M. Bhagwat and V. R. Stefanovic, “Generalized Structure of a Multilevel PWM
Inverter,” IEEE Transactions on Industry Applications, Vol.IA-19, No.6 Nov./Dec., 1983,
pp.1057-1069.
77
[11] J. Rodriguez, S. Bernet, B. Wu, J. O. Pontt and S.Kouro, "Multilevel Voltage-Source-
Converter Topologies for Industrial Medium-Voltage Drives", IEEE Transactions on
Industrial Electronics, Vol. 54, No. 6, pp. 2930-2945, December 2007.
[12] T. Podlesak, D. Katsis, P. Wheeler, J. Clare, L. Empringham, and M. Bland, “A 150-kVA
vector controlled matrix converter induction motor drive,” IEEE Transactions on
Industrial Applications, Vol. 41, No. 3, pp. 841–847, May/June 2005.
[13] R. Emery and J. Eugene, “Harmonic losses in LCI-fed synchronous motors,” IEEE
Transactions on Industrial Applications, Vol. 38, No. 4, pp. 948–954, July/August 2002.
[14] M. Manjrekar, P. Steimer, and T. Lipo, “Hybrid Multilevel Power Conversion System: A
Competitive Solution for High-Power Applications”, IEEE Transactions on Industry
Applications, Vol. 36, No. 3, pp. 834–841, May 2000.
[15] R. J. Hill, “Electric railway traction Part 2 Traction drives with three-phase induction
motors”, Power Engineering Journal, pp. 143-152, Jun 1994.
[16] H. Partab, Modern Electric Traction,4th ed., Dhanpat Rai and Co. , 2003, pp. 1-10,65-68.
[17] Mauro Carpita, Mario Marchesoni, Marc Pellerin, and David Moser, “Multilevel
Converterfor Traction Application: Small Scale Prototype tests Result”, IEEE
Transaction on Industrial Electronics, vol. 55, no. 5, pp. 2203-2212, May 2008.
[18] Bhargava “Railway Electrification Systems and Configurations” Power Engineering
Society Summer Meeting, 1999. IEEE
[19] Hongjian Luo “Unification of Electrification Systems in Europe” 2004
[20] A. Mariscotti “Simplified Modelling of 2x25kV AT Railway System for the Solution of
Low Frequency and Large-Scale Problems” Power Delivery IEEE Trans. 2007
[21] WWW.railway_technical.com
[22] N. V. Suryanarayan, “Utilisation of electric power including electric drives and
traction”, New Age International (P) Limited, 2005.
[23] K. Thiyagarajah, V. T. Ranganathan, B. S. Ramakrishna Iyengan, “A high switching
frequency IGBT PWM rectifier / inverter system for AC motor drives operating from
single phase supply”, IEEE Transactions on Power Electronics, vol. 6, no. 4, pp. 576-584,
Oct. 1991.
78
[24] A.K.Verma, P.R.Thakur, K.C.Jana, and G.Buja, “Cascaded multilevel inverter for
Hybrid electric vehicles,” 2010 India International Conference on Power Electronics
(IICPE),New Delhi, 28-30 Jan 2011.
[25] P. G. Song, E. Y. Guan, L. Zhao, S. P. Liu, “Hybrid Electric Vehicles with a multilevel
cascaded converter using Genetic Algorithm” , 2006 1st IEEE Conference on Industrial
Electronics and Applications, 24-26 May 2006.
[26] Adrian David Cheok, Shoichi Kawamto, Takeo Matsumoto and Hideo Obi, “ High power
AC/DC converter and DC/AC inverter for high speed train applications”, in Proceedings
of IEEE TECON 2000, vol. 1, pp. 423-428, 24-27 Sep 2000.
[27] Y.Panda, “Analysis of cascaded multilevel inverter induction motor drives,” Master
thesis, 2002.
[28] S.Sirisukprasert, “Optimized harmonic stepped-wave form for multilevel inverter,”
Master thesis, 1999.
[29] T. A. Meynard and H. Foch, “Multilevel conversion: High voltage choppers and voltage
source inverters,” in Proc. IEEE PESC’92, 1992, pp. 397–403.
[30] R. H. Baker and L. H. Bannister, “Electric Power Converter,” U.S. Patent 3 867 643,
Feb. 1975.
[31] J. Rodríguez, J. Lai, and F. Peng, “Multilevel converters: a survey of topologies, controls
and applications,” IEEE Transaction on Industrial Electronics, vol. 49, no. 4, Aug. 2002,
pp. 724-738.
[32] P. Hammond, “A new approach to enhance power quality for medium voltage ac drives,”
IEEE Transactions on Industry Applications, vol. 33, pp. 202–208, Jan./Feb. 1997.
[33] Y. Liang and C. O. Nwankpa, “A new type of STATCOM Based on cascading voltage-
source converters with phase-shifted unipolar SPWM,” IEEE Transactions on Industry
Applications, vol. 35, pp. 1118–1123, Sept./Oct. 1999.
[34] L. Li, D. Czarkowski, Y. Liu, and P. Pillay, “Multilevel selective harmonic elimination
PWM technique in series-connected voltage converters,” IEEE Transactions on Industry
Applications,Vol. 36, no. 1, Jan.-Feb. 2000, pp. 160 – 170.
79
[35] J. Rodríguez, L. Morán, C. Silva, and P. Correa, “A high performance vector control of a
11-level converter,” in Proc. 3rd Int. Power Electronics and Motion Control Conf.,
Beijing,China, Aug. 2000, pp. 1116–1121.
[36] J. S. Lai and F. Z. Peng, “Multilevel converters – A new breed of power converters,”
IEEE Transactions on Industry Applications, vol. 32, no. 3, May/June 1996, pp. 509517.
[37] Y. H. Lee, R. Y. Kim, and D. S. Hyun, “A novel SVPWM strategy considering DC link
balancing for a multi-level voltage source converter,” IEEE APEC,1998, pp. 509–514.
[38] J. Rodríguez, P. Correa, and L. Morán, “A vector control technique for medium voltage
multilevel converters,” IEEE APEC, Anaheim, CA, Mar. 2001, pp. 173–178.
[39] H. S. Patel and R. G. Hoft, “Generalized harmonic elimination and voltage control in
thyristor converters: Part I –harmonic elimination,” IEEE Transactions on Industry
Applications, vol. 9, May/June 1973. pp. 310-317.
[40] J. N. Chiasson, L. M. Tolbert, K. J. McKenzie, Z. Du, “A Complete Solution to the
Harmonic Elimination Problem,” IEEE Transactions on Power Electronics, March 2004,
vol. 19, no. 2, pp. 491-499.
[41] J. N. Chiasson, L. M. Tolbert, K. J. McKenzie, Z. Du, “A New Approach to Solving the
Harmonic Elimination Equations for a Multilevel Converter,” IEEE Industry Applications
Society Annual Meeting, October 12-16, 2003, Salt Lake City, Utah, pp. 640-645.
[41] R.Rangarajan, “A novel cascaded multilevel converter,” Master thesis, 2005.
80
Appendices
Appendix A.
An m-file used to solve the switching angles of selective harmonic elimination pulse width
modulation (SHEPWM) is presented as follows:
%Solve switching angle of three phase 11-level SHEPWM by using the Newton-
Raphson method
clear all; close all;
E=5; %The number of dc sources
Mstart= 1; %The initial modulation index
M=Mstart*E;
Mrange=4; %Range of calculation results
p1=5*pi/180; p2=13*pi/180; p3=25*pi/180; p4=40*pi/180; p5=55*pi/180;
%Guess initial value of switching angles
p=[p1 p2 p3 p4 p5]'; %The switching angle matrix
for j=1:Mrange
t=[M*pi/4 0 0 0 0 ]'; %The corresponding harmonic amplitude matrix
df=1; i=1;
while abs(df) > 1e-4 & i < 5 %i=5 %Degree of accuracy condition
p1=p(1,:); p2=p(2,:); p3=p(3,:); p4=p(4,:); p5=p(5,:);
f=[cos(p1)+cos(p2)+cos(p3)+cos(p4)+cos(p5);
cos(5*p1)+cos(5*p2)+cos(5*p3)+cos(5*p4)+cos(5*p5);
cos(7*p1)+cos(7*p2)+cos(7*p3)+cos(7*p4)+cos(7*p5);
cos(11*p1)+cos(11*p2)+cos(11*p3)+cos(11*p4)+cos(11*p5)
cos(13*p1)+cos(13*p2)+cos(13*p3)+cos(13*p4)+cos(13*p5)]; % The nonlinear
system matrix
delf=[-sin(p1) -sin(p2) -sin(p3) -sin(p4) -sin(p5);
-5*sin(5*p1) -5*sin(5*p2) -5*sin(5*p3) -5*sin(5*p4) -5*sin(5*p5);
-7*sin(7*p1) -7*sin(7*p2) -7*sin(7*p3) -7*sin(7*p4) -7*sin(7*p5);
-11*sin(11*p1) -11*sin(11*p2) -11*sin(11*p3) -11*sin(11*p4) –
11*sin(11*p5);
-13*sin(13*p1) -13*sin(13*p2) -13*sin(13*p3) -13*sin(13*p4) –
13*sin(13*p5)];% The differential matrix
df=inv(delf)*(t-f); % Calculate solution error
p=p+df; % Update the solutions.
i=i+1;
end
end
Appendix B.
The specification and the parameters of the induction motor is: Power = 5HP, line-line
voltage=380V, frequency = 50Hz, Stator resistance R= 8Ω, leakage stator inductance Lls =0.032
H, mutual inductance Lm = 0.45H,Rotor resistance R = 6Ω , leakage rotor inductance Llr =
0.032 H, moment of inertia J=0.16 kg-m, frication factor B=0.036 kg-m/s, torque =12 Nm..
81