History and Development of Power Electronics Systems

Prof. (Dr.) B. Pradhan

Director, GIET, Baniatangi, Bhubaneswar -752060

(Former Professor, Mechanical Engg., IIT, Kharagpur)
Email : bpradhan@gietbbsr.com, bpradhan@mech.iitkgp.ernet.in

ABSTRACT: In the days of exponential growth of science and technology, the development of power electronics
systems assumes a leading role in use, distribution and control of various energy resources including the renewable
energies. Several electronic devices are being innovated for domestic as well as industrial requirements. The
effective use of all such devices depend significantly on how they are being designed and made use of. To this affect,
power electronics acts as the smartest grid to harness them well. This paper deals with the history and development
of power electronics systems in brief. The further growth and development of the subject are highlighted.

1. POWER ELECTRONICS : Definition and its
difference from classical electronics
Power electronics is the application of solid-state
electronics for the control and conversion of electric
power. It also refers to a subject of research in electrical
engineering which deals with design, control,
computation and integration of nonlinear, time varying
energy processing electronic systems with fast
dynamics.
Power electronic converters can be found wherever
there is a need to change voltage, current or frequency
of electric power. The power range of these converters
is from some milliWatts (as in a mobile phone) to
hundreds of megawatts in an HVDC transmission
system.
With "classical" electronics, electrical currents and
voltage are used to carry information, whereas with
power electronics, they carry power. Thus, the main
metric of power electronics becomes the efficiency.
2. HISTORY AND DEVELOPMENT
The first very high power electronic devices were
mercury-arc valves. In modern systems the conversion
is performed with semiconductor switching devices
such as diodes, thyristors and transistors, as pioneered
by R. D. Middlebrook and others beginning in the
1950s. In contrast to electronic systems concerned with
transmission and processing of signals and data, in
power electronics substantial amounts of electrical
energy are processed. An AC/DC converter (rectifier) is
the most typical power electronics device found in many
consumer electronic devices, e.g. television sets,
personal computers, battery chargers, etc. The power
range is typically from tens of watts to several hundred
watts. In industry a common application is the variable
speed drive (VSD) that is used to control an induction
motor. The power range of VSDs start from a few
hundred watts and end at tens of megawatts.
3. EARLY LAND MARKS
Power electronics started with the development of
mercury arc rectifier. Invented by Peter Cooper Hewitt
in 1902, the mercury arc rectifier was used to convert
alternating current (AC) into direct current (DC). In
1933 selenium rectifiers were invented.
In 1947, the bipolar point-contact transistor was
invented by Walter H. Brattain and John Bardeen under
the direction of William Shockley at the Bell Telephone
Laboratory. Then in 1948, the invention of the bipolar
junction transistor by Shockley all at once reduced the
cost and size while increasing the efficiency of
transistors beginning a revolution in semiconductor
electronics. Shortly after, in the 1950s, semiconductor
power diodes became available and started replacing
vacuum tubes. Then in 1956, the Silicon Controlled
Rectifier (SCR) was introduced by General Electric
marking the point where semiconductor power
electronics really began.
In 1960s the switching speed of BJTs allowed for
DC/DC converters to be possible in high frequency,
with the MOSFET introduced in 1960. In 1976 power
MOSFET becomes commercially available. Then in
1982 the Insulated Gate Bipolar Transistor (IGBT) was
introduced.
4. NOTABLE PEOPLE AND PATENTS IN
POWER ELECTRONICS
Julius Edgar Lilienfeld : 1925, patent in Canada
describing a device similar to a MESFET and accepted
as a patent in the US entitled “Method and apparatus for
controlling electric current.”

Prof. B. Pradhan Page No. 01

 1928, US patent entitled “Device for controlling electric
current.”
Walter Houser Brattain and John Bardeen and
William Shockley: Invented the transistor. (1956 Nobel
Prize in Physics), 1947: demonstrated a point contact
transistor for the first time.
John Bardeen (American) : 1972 fundamental theory
of conventional superconductivity known as the BCS
theory (won the Nobel prize a second time)
William Bradford Shockley Jr. (American) : switches can approximate this ideal property and so
Commercializing the new transistor design.
Gordon Kidd Teal : Production of extremely
puregermanium single crystals. Together with Morgan
Sparks invented a process to produce a Bipolar Junction
Transistor (BJT) by modifying the fabrication process
of the transistor. This made it more possible to use the
transistors in everyday life.
Bimal K. Bose : Has authored more than 200 papers
and holds 21 U.S. patents in the field of power
electronics.
Frede Blaabjerg : Author or coauthor of more than 300
publications in various research fields including a book
Control in Power Electronics. These include power
electronics, static power converters, ac drives, switched
reluctance drives, modeling, characterization of power
semiconductor devices and simulation, wind turbines,
and green power inverters.
Ned Mohan : Five textbooks, 13 patents and has
written over 200 technical articles. Contribution in the
areas of renewable energy and the next generation of
wind generators and storage.
John G. Kassakian : Expertise in the semiconductor.
R. David Middlebrook : Differential amplifiers,
Design-orientated circuit analysis and measurement
techniques.
Fred C. Lee : High-frequency power conversion,
distributed power systems, renewable energy, high-
density electronics packaging and integration, and
modeling and control.
Bernard de Fornel : Automatic control for electrical
machines.
5. POWER ELECTRONICS DEVICES :
The capabilities and economy of power electronics
system are determined by the active devices that are
available. Their characteristics and limitations are a key
element in the design of power electronics systems.
Formerly, the mercury arc valve, the high-vacuum and
gas-filled diode thermionic rectifiers, and triggered
Prof. B. Pradhan
devices such as the thyratron and ignitron were widely
used in power electronics. As the ratings of solid-state
devices improved in both voltage and current-handling
capacity, vacuum devices have been nearly entirely
replaced by solid-state devices.
Power electronic devices may be used as switches, or as
amplifiers. An ideal switch is either open or closed and
so dissipates no power; it withstands an applied voltage
and passes no current, or passes any amount of current
with no voltage drop. Semiconductor devices used as
most power electronic applications rely on switching
devices on and off, which makes systems very efficient
as very little power is wasted in the switching devices.
By contrast, in the case of the amplifier, the current
through the device varies continuously according to a
controlled input. The voltage and current at the device
terminals follow a load line, and the power dissipation
inside the device is large compared with the power
delivered to the load.
Several attributes dictate how devices are used. Devices
such as diodes conduct when a forward voltage is
applied and have no external control of the start of
conduction. Power devices such as silicon controlled
rectifiers and thyristors (as well as the former mercury
valve and thyratron) allow control of the start of
conduction, but rely on periodic reversal of current flow
to turn them off. Devices such as gate turn-off
thyristors, bipolar junction transistors (BJT), and
MOSFET transistors provide full switching control and
can be turned on or off without regard to the current
flow through them. Transistor devices also allow
proportional amplification, but this is rarely used for
systems rated more than a few hundred watts. The
control input characteristics of a device also greatly
affect design; sometimes the control input is at a very
high voltage with respect to ground and must be driven
by an isolated source.
As efficiency is at a premium in a power electronic
converter, the losses that a power electronic device
generates should be as low as possible.
Devices vary in switching speed. Some diodes and
thyristors are suited for relatively slow speed and are
useful for power frequency switching and control;
certain thyristors are useful at a few kilohertz. Devices
such as MOSFETS and BJTs can switch at tens of
kilohertz up to a few megahertz in power applications,
but with decreasing power levels. Very high power
(hundreds of kilowatts) at very high frequency
(hundreds or thousands of megahertz) is still the area
where vacuum tube devices predominate. The use of
faster switching devices minimizes energy lost in the
transitions from on to off and back, but may create
Page No. 02
 problems with radiated electromagnetic interference.
Gate drive (or equivalent) circuits must be designed to
supply sufficient drive current to achieve the full
switching speed possible with a device. A device that
doesn't get sufficient drive to switch rapidly, may be
destroyed by excess heating.
Practical devices have non-zero voltage drop and
dissipate power when on, and take some time to pass
through an active region until they reach the "on" or
"off" state. These losses are a significant part of the total
lost power in a converter.
Power handling and dissipation of devices is also a
critical factor in design. Power electronic devices may
have to dissipate tens or hundreds of watts of waste
heat, even switching as efficiently as possible between
conducting and non-conducting states. In the switching
mode, the power controlled is much larger than the
power dissipated in the switch. The forward voltage
drop in the conducting state translates into heat that
must be dissipated. High power semiconductors require
specialized heat sinks or active cooling systems to keep
their junction temperature from rising too high; exotic
semiconductors such as silicon carbide have an
advantage over straight silicon in this respect, and
germanium, once the main-stay of solid-state electronics
is now little used due to its unfavorable properties at
high temperature.
Semiconductor devices exist with ratings up to a few
kilovolts in a single device. Where very high voltage
must be controlled, multiple devices must be used in
series, with networks to equalize voltage across all
devices. Again, switching speed is a critical factor since
the slowest-switching device will have to withstand a
disproportionate share of the overall voltage. The
former mercury valves were available with ratings to
100 kV in a single unit, simplifying their application in
HVDC systems.
The current rating of a semiconductor device is limited
by the heat generated within the dies and the heat
developed in the resistance of the interconnecting leads.
Semiconductor devices must be designed so that current
is evenly distributed within the device across its internal
junctions (or channels); once a "hot spot" develops,
breakdown effects can rapidly destroy the device.
Certain SCRs are available with current ratings to 3000
amperes in a single unit.
Solid-state devices:
Diode, Silicon-controlled rectifier (SCR), Thyristor,
Gate turn-off thyristor (GTO), Triac , Bipolar junction
transistor (BJT), Power MOSFET, Insulated gate
bipolar transistor (IGBT), MOS Controlled Thyristor
(MCT), Integrated gate-commutated thyristor (IGCT)
Prof. B. Pradhan
DC/AC converters :
DC to AC converters produce an AC output waveform
from a DC source. Applications include adjustable
speed drives (ASD), uninterruptable power supplies
(UPS), static VAR compensators, active filters, flexible
AC transmission systems (FACT), voltage
compensators, and photovoltaic generators. Topologies
for these converters can be separated into two distinct
categories: voltage source inverters and current source
inverters. Voltage source inverters (VSIs) are named so
because the independently controlled output is a voltage
waveform. Similarly, current source inverters (CSIs) are
distinct in that the controlled AC output is a current
waveform.
Being static power converters, the DC to AC power
conversion is the result of power switching devices,
which are commonly fully controllable semiconductor
power switches. The output waveforms are therefore
made up of discrete values, producing fast transitions
rather than smooth ones. The ability to produce near
sinusoidal waveforms around the fundamental
frequency is dictated by the modulation technique
controlling when, and for how long, the power valves
are on and off. Common modulation techniques include
the carrier-based technique, or pulse width modulation,
space-vector technique, and the selective-harmonic
technique.
Voltage source inverters have practical uses in both
single-phase and three-phase applications. Single-phase
VSIs utilize half-bridge and full-bridge configurations,
and are widely used for power supplies, single-phase
UPSs, and elaborate high-power topologies when used
in multicell configurations. Three-phase VSIs are used
in applications that require sinusoidal voltage
waveforms, such as ASDs, UPSs, FACTs, and VAR
compensators. They are also used in applications where
arbitrary voltages are required as in the case of active
filters and voltage compensators.
Current source inverters are used to produce an AC
output current from a DC current supply. This type of
inverter is practical for three-phase applications in
which high-quality voltage waveforms are required.
A relatively new class of inverters, called multilevel
inverters, has gained widespread interest. Normal
operation of CSIs and VSIs can be classified as two-
level inverters, due to the fact that power switches
connect to either the positive or to the negative DC bus.
If more than two voltage levels were available to the
inverter output terminals, the AC output could better
approximate a sine wave. It is for this reason that
multilevel inverters, although more complex and costly,
offer higher performance.
Page No. 03
 Each inverter type differs in the DC links used, and in
whether or not they require freewheeling diodes. Either
can be made to operate in square-wave or pulse-width
modulation (PWM) mode, depending on its intended
usage. Square-wave mode offers simplicity, while PWM
can be implemented several different ways and
produces higher quality waveforms[.
Voltage Source Inverters (VSI) feed the output inverter
section from an approximately constant-voltage source..
The desired quality of the current output waveform
determines which modulation technique needs to be
selected for a given application. The output of a VSI is
composed of discrete values. In order to obtain a smooth
current waveform, the loads need to be inductive at the
select harmonic frequencies. Without some sort of
inductive filtering between the source and load, a
capacitive load will cause the load to receive a choppy
current waveform, with large and frequent current
spikes.
AC/AC converters:
Converting AC power to AC power allows control of
the voltage, frequency, and phase of the waveform
applied to a load from a supplied AC system . The two
main categories that can be used to separate the types of
converters are whether the frequency of the waveform is
changed. AC/AC converter that don’t allow the user to
modify the frequencies are known as AC Voltage
Controllers, or AC Regulators. AC converters that allow
the user to change the frequency are simply referred to
as frequency converters for AC to AC conversion.
Under frequency converters there are three different
types of converters that are typically used:
cycloconverter, matrix converter, DC link converter
(aka AC/DC/AC converter).
AC voltage controller:
The purpose of an AC Voltage Controller, or AC
Regulator, is to vary the RMS voltage across the load
while at a constant frequency. Three control methods
that are generally accepted are ON/OFF Control, Phase-
Angle Control, and Pulse Width Modulation AC
Chopper Control (PWM AC Chopper Control). All
three of these methods can be implemented not only in
single-phase circuits, but three-phase circuits as well.
• ON/OFF Control: Typically used for heating loads
or speed control of motors, this control method
involves turning the switch on for n integral cycles
and turning the switch off for m integral cycles.
Because turning the switches on and off cause’s
undesirable harmonics to be created, the switches
are turned on and off during zero-voltage and zero-
current conditions (zero-crossing), effectively
reducing the distortion.
Prof. B. Pradhan
• Phase-Angle Control: Various circuits exist to
implement a phase-angle control on different
waveforms, such as half-wave or full-wave voltage
control. The power electronic components that are
typically used are diodes, SCR’s, and Triac’s. With
the use of these components, the user can delay the
firing angle in a wave which will only cause part of
the wave to be outputted.
• PWM AC Chopper Control: The other two control
methods often have poor harmonics, output current
quality, and input power factor. In order to improve
these values PWM can be used instead of the other
methods. What PWM AC Chopper does is have
switches that turn on and off several times within
alternate half-cycles of input voltage.
Matrix converters and cycloconverters:
Cycloconverters are widely used in industry for ac to ac
conversion, because they are able to be used in high-
power applications. They are commutated direct
frequency converters that are synchronized by a supply
line. The cycloconverters output voltage waveforms
have complex harmonics with the higher order
harmonics being filtered by the machine inductance.
Causing the machine current to have fewer harmonics,
while the remaining harmonics causes losses and torque
pulsations. Note that in a cycloconverter, unlike other
converters, there are no inductors or capacitors, i.e. no
storage devices. For this reason, the instantaneous input
power and the output power are equal.
• Single-Phase to Single-Phase Cycloconverters:
Single-Phase to Single-Phase Cycloconverters
started drawing more interest recently because of
the decrease in both size and price of the power
electronics switches. The single-phase high
frequency ac voltage can be either sinusoidal or
trapezoidal. These might be zero voltage intervals
for control purpose or zero voltage commutation.
• Three-Phase to Single-Phase Cycloconverters:
There are two kinds of three-phase to single-phase
cycloconverters: 3φ to 1φ half wave
cycloconverters and 3φ to 1φ bridge
cycloconverters. Both positive and negative
converters can generate voltage at either polarity,
resulting in the positive converter only supplying
positive current, and the negative converter only
supplying negative current.
With recent device advances, newer forms of
cycloconverters are being developed, such as matrix
converters. The first change that is first noticed is that
matrix converters utilize bi-directional, bipolar
switches. A single phase to a single phase matrix
Page No. 04
 converter consists of a matrix of 9 switches connecting
the three input phases to the tree output phase. Any
input phase and output phase can be connected together
at any time without connecting any two switches from
the same phase at the same time; otherwise this will
cause a short circuit of the input phases. Matrix
converters are lighter, more compact and versatile than
other converter solutions. As a result, they are able to
achieve higher levels of integration, higher temperature
operation, broad output frequency and natural bi-
directional power flow suitable to regenerate energy
back to the utility.
The matrix converters are subdivided into two types:
direct and indirect converters. A direct matrix converter
with three-phase input and three-phase output, the
switches in a matrix converter must be bi-directional,
that is, they must be able to block voltages of either
polarity and to conduct current in either direction. This
switching strategy permits the highest possible output
voltage and reduces the reactive line-side current.
Therefore the power flow through the converter is
reversible. Because of its commutation problem and
complex control keep it from being broadly utilized in
industry.
Unlike the direct matrix converters, the indirect matrix
converters has the same functionality, but uses separate
input and output sections that are connected through a
dc link without storage elements. The design includes a
four-quadrant current source rectifier and a voltage
source inverter. The input section consists of bi-
directional bipolar switches. The commutation strategy
can be applied by changing the switching state of the
input section while the output section is in a
freewheeling mode. This commutation algorithm is
significantly less complexity and higher reliability as
compared to a conventional direct matrix converter.
DC link converters:
DC Link Converters, also referred to as AC/DC/AC
converters, convert an AC input to an AC output with
the use of a DC link in the middle. Meaning that the
power in the converter is converted to DC from AC with
the use of a rectifier, and then it is converted back to AC
from DC with the use of an inverter. The end result is an
output with a lower voltage and variable (higher or
lower) frequency. Due to their wide area of application,
the AC/DC/AC converters are the most common
contemporary solution. Other advantages to AC/DC/AC
converters is that they are stable in overload and no-load
conditions, as well as they can be disengaged from a
load without damage.
Prof. B. Pradhan
Hybrid matrix converter:
Hybrid matrix converters are relatively new for AC/AC
converters. These converters combine the AC/DC/AC
design with the matrix converter design. Multiple types
of hybrid converters have been developed in this new
category, an example being a converter that uses uni-
directional switches and two converter stages without
the dc-link; without the capacitors or inductors needed
for a dc-link, the weight and size of the converter is
reduced. Two sub-categories exist from the hybrid
converters, named hybrid direct matrix converter
(HDMC) and hybrid indirect matrix converter (HIMC).
HDMC convert the voltage and current in one stage,
while the HIMC utilizes separate stages, like the
AC/DC/AC converter, but without the use of an
intermediate storage element.
6. APPLICATIONS:
List of common applications for converters:
• AC Voltage Controller: Lighting Control; Domestic
and Industrial Heating; Speed Control of Fan,Pump
or Hoist Drives, Soft Starting of Induction Motors,
Static AC Switches (Temperature Control,
Transformer Tap Changing, etc.)
• Cycloconverter: High-Power Low-Speed
Reversible AC Motor Drives; Constant Frequency
Power Supply with Variable Input Frequency;
Controllable VAR Generators for Power Factor
Correction; AC System Interties Linking Two
Independent Power Systems.
• Matrix Converter: Currently the application of
matrix converters are limited due to non-
availability of bilateral monolithic switches capable
of operating at high frequency, complex control law
implementation, commutation and other reasons.
With these developments, matrix converters could
replace cycloconverters in many areas.
• DC Link: Can be used for individual or multiple
load applications of machine building and
construction.
Power electronic systems for electronic devices :
• DC/DC converters are used in most mobile devices
(mobile phones, PDA etc.) to maintain the voltage
at a fixed value whatever the voltage level of the
battery is. These converters are also used for
electronic isolation and power factor correction.
• AC/DC converters (rectifiers) are used every time
an electronic device is connected to the mains
(computer, television etc.). These may simply
Page No. 05
 change AC to DC or can also change the voltage Some examples, but not all, of power electronics
level as part of their operation. applications are shown bellow:
• AC/AC converters are used to change either the
voltage level or the frequency (international powe
power
adapters, light dimmer). In power distribution
networks AC/AC converters may be used to
exchange power between utility frequency 50 Hz
and 60 Hz power grids.
• DC/AC converters (inverters)) are used primarily in
UPS or renewable energy systems oor emergency
lighting systems.. When mains power is available, it
will charge the DC battery. If the mains fails, an
inverter will be used to produce AC electricity at
mains voltage
age from the DC battery.
Application of power electronics for the control and
conversion of electric power.: An inverter thyristor tower 16.8 m tall in a hall at Baltic Cable
AB in Sweden
Applications of power electronics range in size from a
switched mode power supply in an AC adapter
adapter, battery
chargers, fluorescent lamp ballasts, thro
through variable
frequency drives and DC motor drives used to operate
pumps, fans, and manufacturing machinery, up to
gigawatt-scale high voltage direct current power
transmission systems used to interconnect electrical
grids.
Motor drives are found in pumps, blowers, and mill
drives for textile, paper, cement an and other such
facilities. Drives may be used for power conversion and
for motion control.. For AC motors, applications include
variable-frequency drives, motor soft starters and A battery charger is an example of a piece of power
electronics
excitation systems..
In hybrid electric vehicles (HEVs), power electronics
are used in two formats: series hybrid and parallel
hybrid. The difference between a series hybrid and a
parallel hybrid is the location of the electric motor in
comparison to the internal combustion engine (ICE).
The power electronic devices used in electric vehicles
consist mostly of dc/dc
dc converters for battery charging,
and dc/ac converters to power the propulsion motor.
Electric trains are a big movement in public
transportation, especially in rural areas. Th
These trains use
power electronic devices to obtain power, as well as for
vector control using pulse width modulation (PWM)
rectifiers. These devices are used, sincee the trains obtain
their power from the lines they transport along. Another
new usage for power electronics is in elevator systems. A PCs power supply is an example of a piece
pie of power
These systems may use thyristors, inverters, permanent electronics, whether inside or outside of the cabinet
magnet motors, or various hybrid systems that
7. “SMART GRID” : A novel power electronics
incorporate PWM systemss and standard motors motors. The
system for integrating renewable energy with
uses of power electronics discussed above are just some
existing power grid
examples of the use of power electronics in residential
and public applications.

Prof. B. Pradhan Page No. 06

 The Smart Grid is a concept in which renewable energy
systems are integrated into the existing power grid.
Power electronics are used to convert power generated
by distributed energy systems into usable power for the
grid.
Electric power can be generated by wind turbines and
hydroelectric turbines by using induction generators.
Unfortunately, the wind and water flows may not be
fully controlled and can cause variances in the
frequency at which the power is generated. Power
electronic devices are utilized in these systems to
1.
convert the generated ac voltages into high-voltage
direct current (HVDC). The HVDC power can be more
easily converted into three phase power that is coherent2.
with the power associated to the existing power grid.
Through these devices, the power delivered by these
systems is cleaner and has a higher power factor
associated to them. In many wind power systems, the
optimum torque on the generator is obtained through a
gearbox. On the other hand, some systems use direct
drive technologies which reduce the size of the power3.
electronic device and remove the gearbox from the
system.
4.
Electric power can be generated through photovoltaic
cells by using power electronic devices. The produced
power is usually then transformed by inverters. Inverters
are divided into three different converter types: central
converters, module-integrated converters, and string
converters. Central converters can be connected in5.
either parallel and/or in series on the DC side of the
system. For photovoltaic farms, only one central
converter is used for the entire system. Module-
Prof. B. Pradhan
integrated converters are connected in series on the DC
side of the system, and in series on the AC side. There
are normally several modules within the photovoltaic
system, since the system requires these converters on
both the DC and the AC terminals. A string converter is
used in a system that utilizes photovoltaic cells that are
facing different directions. It is used to convert the
power generated to each string, or line, in which the
photovoltaic cells are interacting.
8. REFERENCES
Issa Batarseh, "Power Electronic Circuits" by John
Wiley, 2003.
S.K. Mazumder, "High-Frequency Inverters: From
Photovoltaic, Wind, and Fuel-Cell based Renewable-
and Alternative-Energy DER/DG Systems to Battery
based Energy-Storage Applications", Book Chapter in
Power Electronics handbook, Editor M.H. Rashid,
Academic Press, Burlington, Massachusetts, 2010.
V. Gureich "Electronic Devices on Discrete
Components for Industrial and Power Engineering",
CRC Press, New York, 2008, 418 p.
Editor: Semikron, Authors: Dr. Ulrich Nicolai, Dr.
Tobias Reimann, Prof. Jürgen Petzoldt, Josef Lutz:
Application Manual IGBT- and MOSFET-power
modules, 1. edition, ISLE Verlag, 1998, ISBN 3-
932633-24-5 online version
R. W. Erickson, D. Maksimovic, Fundamentals of
Power Electronics, 2nd Ed., Springer, 2001, ISBN 0-
7923-7270-0
Page No. 07
 Key Note-01

Some Practical Applications of Image Processing

Prof. T. K. Basu

Ex-Professor, Electrical Engineering

IIT, kharagpur-721302

Modern digital technology has made it possible to manipulate multi-dimensional signals

with systems that range from simple digital circuits to advanced parallel computers. As we all
know, computer vision is the science and technology of machines that can see. As a scientific
discipline, computer vision is concerned with the theory for building artificial systems that obtain
information from images. The image data can take many forms, such as a video sequence, views
from multiple cameras, or multi-dimensional data from a medical scanner. As a technology
discipline, image processing is important to this and yes, it is beneficial. Image processing
methods are extremely numerous and varied in terms of activities as in terms of size. To
illustrate our statement, here is a non-exhaustive list of application domains: medical and
biological purposes, security and surveillance, multimedia, photo and video, robotics, biometrics
and remote sensing, etc. Moreover, in a situation of simultaneous existence of small and
medium-sized businesses, including start-ups, many big companies can be easily identified by
adoption of image processing technologies (like Philips, Canon, EADS, Thalès …). It is often
seen that some great firms (for example, automobile or optical industries such as Renault,
Essilor, etc.) also work with images for some of their products or services.

This lecturer deals with some practical applications of image processing in several sectors
of industries. The utilities of image processing technologies in Rural and Agro-based industrial
applications are being highlighted.

Prof. T.K. Basu Page No. 08

 Key Note-02

Distributed Generations
Prof. (Dr.) P. K. Hota
Principal,
College of Engineering and Technology
Ghatikia, Bhubaneswar , Odisha

Why
Distributed Generations? Types
• Electric utility restructuring • Single shaft
• Public environmental policy • Split shaft
• Expanding power demand
• Small distributed generators are in great need to Single-shaft
satisfy on-site customer energy needs. • High-speed single shaft unit with the compressor
• Major improvements have been achieved through and turbine mounted on the same shaft as the
decades of intensive research electrical alternator
• Speed: 50,000 to 120,000 rpm
Obvious advantages
• Small-modular units Split-shaft
• Economic • Uses a power turbine rotating at 3600 rpm and a
• Operational conventional generator connected via a gear box
• Environmental
• Performance Fuel Cell
• Produces power electrochemically by passing a
DERs (Distributed Energy Resources) hydrogen rich gas over an anode and air over a
• Micro-turbines cathode & introducing an electrolyte in between
• Fuel Cells to enable exchange of ions
• The effectiveness is strongly dependent upon the
Both of them are promising as they can operate on electrolyte to create the chemical reactivity
multiple fuels with needed for ion transport
• low emissions
• High efficiency Features of Fuel Cells
• Efficiency: 35-60%
• High reliability
• Low to zero emissions
Micro-turbines • Quiet operation
• Small • High reliability due to limited number of moving
• Simple-cycle gas turbines parts
• Output (25 to 300KW)
Types of Fuel Cells
Techniques incorporated into the larger machines to • Polymer Electrolyte Fuel Cell (PEFC)
improve performance can be typically found in • Alkaline Fuel Cell (AFC)
micro-turbines as well. • Phosphoric Acid Fuel Cell (PAFC)
• Molten Carbonate Fuel Cell (MCFC)
Techniques • Solid Oxide Fuel Cell (SOFC)
• Recuperation
• Low NOX technologies Developments of Fuel Cells
• Potential use of advanced materials such as • PAFC:- Successfully commercialized, 200 KW
ceramics for hot section parts PAFC first commercialized
• SOFC and MCFC are the second generation fuel
cells (2002)
• PEFC:- still under development and testing phase

Prof. (Dr.) P. K. Hota Page No. 09


Developments of Fuel Cells
• PAFC:- Successfully commercialized, 200 KW
PAFC first commercialized
• SOFC and MCFC are the second generation fuel
cells (2002)
• PEFC:- still under development and testing phase
Micro-turbine & FC modeling issues
• Their dynamic models are necessary to deal with
issues in system planning, operation and
management
• Both of them have several unique properties from
a modeling point of view
Micro-turbine
• Extremely low inertia and damping
Fuel Cell
• Electrical response time of the power section is
generally fast, being mainly associated with the
speed at which the chemical reaction is capable of
storing the charge that has been drained by the
load
• Conversely the chemical response time of the
reformer is usually slow, being associated with
the time for the FC stack to modify the chemical
reaction parameters after a change in the flow of
reactants
Micro-turbine modeling
• Turbine: High speed single shaft or split shaft
• Both of them are small gas turbines
• Alternator or conventional machine:
• Single-shaft rotor: either two or four pole
permanent magnet design
• Stator: conventional copper wound design
• Split-shaft: a conventional induction or
synchronous machine is mounted on the power
turbine via a gear box
Power Electronics
• Single-shaft: alternator generates a very high
frequency, three-phase signal ranging from 1500-
4000 Hz.
• The high frequency voltage is first rectified and
then inverted to a normal 50 or 60 Hz voltage
• Split-shaft: Power inverters are not needed
Recuperator
• It is a heat exchanger
• Transfers heat from the exhaust gas to the
discharge air before it enters the combustor
Prof. (Dr.) P. K. Hota
Key Note-02
• This reduces the amount of fuel required to raise
the discharge air temperature to that required by
the turbine
Control & Communication Systems
• Full control of turbine
• Power inverter & start-up electronics
• Instrumentation
• Signal conditioning
• Data logging
• Diagnosis
• User’s control communication
Assumptions for deriving simplified micro-turbine
model
• System operation is under normal operating
conditions
• i.e., start-up, shut-down, fast dynamics (faults,
loss of power, etc) are not included
• Micro-turbine’s electro-mechanical behavior is
our interest
• Recuperator is not included in the model as it is
only a heat exchanger to raise the engine
efficiency.
• Also due to recuperator’s very slow response
time, it has little influence on the time scale of our
dynamic simulation.
• The gas turbine’s temperature control and
acceleration control are of no significance under
normal operating conditions (Thus, they are not
included in the turbine model)
• Most micro-turbines do not have governors and so
not included in the model
Power Works micro-turbine diagram
Control System
• Real power Proportional-Integral (PI) control
function
• The controlled real power Pin applied to the
turbine.
Page No. 10
 Key Note-02

• The standard transformation between synchronous

rotating reference frame and abc variables are
used to transform mechanical equations to
electrical equations.
• Transformation from frequency domain to time
domain.
• Transformation from abc to qd0 variables
(electrical equations)
• Transformation from qd0 to abc
• Transformation from time to frequency domain

Modelling of single-shaft design

• Electrical alternator Control system model
• Air Compressor
• Turbine
• Shaft turns between 50K and 120K rpm
• Frequency generated in the range 1500 to 4000
Hz.

Requires a power electronic interface which

consists of:
• AC to DC rectifier
• DC bus capacitor
• DC to AC inverter

Assumptions Dynamic response of micro-turbine system.

• Adequate energy storage on the DC bus so that
micro-turbine can be seen as an ideal, no-load PM
generator with constant DC potential.
• The AC to DC converters are full wave rectifiers
with constant current output.
• All losses are neglected
Evaluation of stand-alone performance
• (Split-shaft microturbine model)
• Constant rated voltage of 1.0 p.u.
• Power demand of 0.7 p.u.
• At t = 0.5s, there is a step increase of power
demand from 0.7 to 1.0 p.u.
• Figure shows the dynamic response of the
mechanical power Pm input and total 3-phase
electrical power output Pe

Observations
• The initial response time for the step change is
around 10s. The delay is mainly due to the turbine
response time.
• For electrical power response, the oscillation is
significant with a time period around 20s. This is
mainly due to the small inertia and damping in the
Main blocks in micro-turbine model turbine.

Prof. (Dr.) P. K. Hota Page No. 11

 Key Note-02

Fuel Cell Control Scheme (Cont.)

• A DC to DC boost converter before the DC/AC
PWM inverter is suggested.
• FC inverter has power modulation capability that
enhance the voltage level of the system.
• The modulation index controller increases the
value of modulation index in order to enhance
voltage regulation.
• The values of modulation index remain between 0
Solid oxide fuel cell connected to infinite bus and 1.

Solid oxide fuel cell connected to infinite bus.

• The O/P power as a function of the firing angle is
shown.
• The O/P power can reach 1.4 p.u. by changing the
firing angle relative to the infinite bus.

Fuel-cell control scheme.

Hybrid Fuel-Cell Power System

• Fuel Cell Applications: 1) DG Power System
2) Electric Vehicle
SOFC active power versus firing angle
Shortcomings:
• It cannot store energy
• The response is slower
• Output voltage fluctuates with load variations
• Difficult to cold-start
Cold-start
• At the beginning of a FC cold-start, the stack is
dry and its temperature is low.
• If the FC is forced to power a heavy load during
this period, it could be destroyed.
• Therefore, a secondary energy source needs to be
introduced into the FC power system to power the
load during the cold-start stage.

Fuel cell and auxiliary source Components of hybrid FC Power System

• Fuel Cell
Fuel Cell Control Scheme • Isolated Unidirectional Converter (UDC)
• Power Control: Done by adjusting the firing angle • BDC
of the inverter for fast transient variations. • Inverter
• Voltage Control: Done by adjusting the • Battery
modulation index of the converter, which affects • The FC & Battery are connected to the same bus
the magnitude of the converter output voltage. through the appropriate UDC & BDC,
• Power control also can be achieved through respectively.
control of the input fuel flow, but this has a large • The BDC is employed to limit battery
time constant and will not affect the transient charge/discharge current
behaviour.

Prof. (Dr.) P. K. Hota Page No. 12


• When the battery is charged, power flows from
the DC bus to the Battery & BDC operates in
buck mode.
• When the battery is discharged, power flows from
battery to DC bus, & the BDC operates in boost
mode.
Advantages
• It can optimize power management & improve
system efficiency.
• During cold-start stage, the battery powers the
load & the FC provides no energy; thus the FC is
easy to cold-start.
• FC can not respond immediately when the load
varies. A battery provides or absorbs dynamic
energy such that the dynamic characteristic of the
system can be improved.
• Due to the introduction of the battery, peak power
can be produced allowing the power rating of the
FC to be decreased, thus reducing the total system
cost.
Hybrid fuel-cell power system.
Grid Connected PV System with MPPT Control
(Max. Power Point Tracking)
• The PV array consists of no. of individual photo-
voltaic cells that are connected in a series and
parallel array to obtain a unit with a suitable
power rating.
• The DC O/P current of PV array IPV is converted
into AC and injected into the grid through
inverter.
• The relation between VPV & IPV is highly non-
linear & dependent on the solar irradiance
incident on the PV array.
• The combination of VPV and IPV that maximizes
the O/P of the inverter depends on the irradiance
& is also affected by the temperature of the cells.
• A MPPT system is therefore always implemented
to optimize the efficiency of the photo-voltaic
energy conversion.
Prof. (Dr.) P. K. Hota
Key Note-02
• This tracking system adjusts the inverter voltage
reference signal Vref and hence, the DC voltage
VPV at the O/P of the solar array.
MPPT Algorithm
(Perturbation & Observation) technique
• It requires only the measurements of VPV & IPV .
• Capable of tracking the max. power point (MPP)
quite accurately through variations in irradiance &
temperature.
• As the name indicates the P&O method works by
perturbing VPV and observing the impact of this
change on the O/P power of the PV array.
• At each cycle, VPV & IPV are measured to
calculate P(k).
• This value of P(k) is compared to that of P(k-1)
calculated at the previous cycle.
• If the O/P power has increased, VPV is adjusted
further in the same direction as in the previous
cycle.
• If the O/P power has decreased, VPV is perturbed
in the opposite direction as in the previous cycle.
• VPV is thus perturbed at every cycle.
• When the max. power point is reached, VPV
oscillates around the optimal value VOPT .
Grid-connected PV system.
I-V and P-V characteristics of a PV array for three
irradiance levels
Page No. 13
 Key Note-02

Wind Energy Conversion System • Between rated wind speed & 25 m/s, the power
• Turbine rotor delivered is limited in order to avoid overloading
• A gear box on the WT system.
• A generator • Over the cutout wind speed, the turbine has to be
• A power electronic system stopped in order to avoid damages.
• A transformer for grid connection • During the optimal efficiency wind speed range,
• WTs capture the power from wind by means of the wind generator may be adjusted to follow the
the turbine blades & convert it to mechanical maximum power point.
power. • Methods to perform MPPT control for WTs:
• It is important to control & limit the converted • TSR control
mechanical power during higher wind speeds. • PSF (Power signal feedback) control
• The power limitation may be done either by stall • HCS (Hill climbing searching) is similar to MPPT
control, active stall or pitch control. • HCS is ineffective for large WT, since difficult to
• The power may be smoothly limited by rotating adjust speed fast.
the blades either by pitch or active stall control,
while the power from a stall-controlled turbine Wind Farm
shows a small overshoot & a lower power O/P for • Equipped with PE converters
higher wind speed. • Can perform both active & reactive power control
• The low-speed, high-torque mechanical power is • Operates the WTs in variable speed to maximize
converted to electrical power by using a gear box the capture energy
& a generator with standard speed. • Reduces the mechanical stress & noise.
• The generator converts the mechanical power into • IG is used.
electrical power, which being fed into a grid • STATCOM is used to provide reactive power
possibly through power electronic converters and control
a transformer with circuit breaker & electricity • Example: Denmark:-(160 MW wind power
meters. station)
• The two most common types of electrical
machines used in WTs: 1) IG, ii) Sync. Generator.
• IG with cage rotor can be used in the fixed-speed
WT due to damping effect.
• IG in such cases, suffer from the voltage
instability mostly, found in large-scale wind farm.
• A wound rotor IM provides partial variable-speed
WT with the help of small PE converter for
increased energy yield at wind speeds below the
rated speed.
• The use of sync. Generator leads to the
requirement for a full rated PE conversion system
to decouple the generator from the network.
• Sync. M/C powered by WTs may not be directly
connected to AC grid because of the requirement
for significant damping in the drive train.
• Multiple pole Sync. Generator either with PM
excitation or with an electromagnet can be driven
by a WT rotor without a gearbox or with a low
ratio gearbox. Main components of a wind turbine system
(Relation between wind speed & power generated
by WT)
• The blades start to move around 4m/s & optimal
aerodynamics efficiency is achieved up to the
rated wind speed about 15 m/s.

Prof. (Dr.) P. K. Hota Page No. 14

 Key Note-02

Power characteristics of fixed-speed wind turbines. (a)

Stall control. (b) Active stall control. (c) Pitch
control.

Output power of a wind turbine as a function of the

wind speed.

Cage-induction-generator-based fixed-speed wind

turbine with power electronic soft-starter.

Block diagram of the TSR control.

Wound rotor induction generator with a rotor Block diagram of the PSF control.
resistance converter

Prof. (Dr.) P. K. Hota Page No. 15

 Key Note-02

Fundamentals of Hybrid Generation System

• Stand-alone hybrid generating systems are usually
used to supply isolated area or locations
interconnected to a weak grid.
• They combine several generation modules,
typically assimilating different renewable energy
sources.
• The application of these hybrid topologies reduces
the probability of energy supply shortage.
• With the incorporation of energy storage, it allows
to eliminate the back ground diesel generator
which is commonly required in generation system
based on a single renewable energy source.
• Many electric generation hybrid system (EGHS)
frequently combine solar & wind energy sources
Block diagram of the HCS control.
taking advantage of their complementary nature
with a lead-acid battery bank to overcome periods
Bio-gas Generation System of scarce generations.
• The air-flow & purified bio-gas flow in pipelines
• The wind generation module is constituted by a
in front of a gas engine (GE).
wind mill, a multipole PM synchronous generator
• A single line equivalent circuit connected between (PMSG), a rectifier, and a DC/DC converter ti
the O/P terminals of an induction machine and the interface the generator with the DC bus.
power grid inside the large food mill is shown.
• The converter commands the voltage on the
• The normal operating procedure for starting up PMSG terminals, indirectly controlling the
the GE-IG set is first to connect the stator operation point of the wind turbine and
windings of IG to the distribution system through consequently its power generation.
a manual-controlled electromagnetic switch.
• The solar module comprises several PV panels
• When the switch is closed, the IG operates as both connected to the DC bus via a DC/DC converter.
an induction motor & a GE starter, absorbing a
• Similar to the wind subsystem, the converter
large starting current to produce sufficient starting
controls the operation point of the PV panels.
torque to drive the crank of GE.
• The DC bus collects the energy generated by both
• When the bio-fuel mixed with proper amount of
modules & delivers it to the load &, if necessary,
air in the combustion chamber of GE is ignited at
to the battery bank.
correct instants, the GE can normally start-up &
• Its voltage is imposed by the battery bank which
maintain a sustained rotation to drive the shaft of
comprises lead-acid batteries connected in a
the IG.
serial/ Parallel array.
• When the speed of the IG is higher than its sync.
• The load could be an AC or DC load.
speed (1800 rpm), the slip of the IG becomes
• In case of AC load, a voltage inverter is required.
negative & the IG can deliver electrical power to
the distribution system.

Schematic diagram of the studied BGS.

Hybrid generation system.

Prof. (Dr.) P. K. Hota Page No. 16

 Modeling, Design Analysis and Simulation of Small
Signal Control Strategies on a STATCOM for
Reactive Power Compensation
S.K.Das* and Dr.J.K.Moharana**
*Faculty member and Research Scholar, Department of Electrical Engineering,
Hitech Institute of Technology, Bhubaneswar, Odisha, India and
**Professor and Head, Department of Electrical Engineering,
Dhaneswar Rath Institute of Engineering and Management Studies, Tangi, Cuttack, Odisha, India.
Email:*sarat_bls@yahoo.co.in and **jkrushna@gmail.com

ABSTRACT
The statcom (static synchronous compensator) is being increasingly popular in power system applications. in
general, reactive power compensation for power factor and stability of the utility system can be improved. a simple
dq transformation and steady state and transient analysis are achieved to characterize the open loop system. two
control schemes of the statcom are proposed without using independent dc voltage source. the small signal scheme
controls the phase angle as well as modulation index of the switching pattern and with small perturbation of
reference current (reactive current of load), the dc voltage nearly remains constant. the draou scheme controls only
the phase angle. the two schemes provide fast response of the statcom compensation of the reactive power and
becoming the stable of the system. the small signal control scheme is better than drauo scheme with respect to
duration from transient to steady state operation of the statcm and voltage transient overshoot. all responses are
obtained trough matlab simulink tool box.
Index Terms—STATCOM, small signal model, Jacobian matrix
I. INTRODUCTION the synchronous motor controls the amount of VAR
absorbed/injected.In a similar way, the firing instant
In recent years power systems have become very
of the 3-phase inverter controls the VAR flow into or
complex with interconnected long distance
out of the STATCOM. Large numbers of capacitor
transmission lines. The interconnected grids tend to
banks or any other passive elements are no more
become unstable as the heavy loads vary dynamically
required. Only a fixed set of capacitor provides the
in their magnitude and phase angle and hence power
required VAR control, with a rapid control of bus
factor. Commissioning new transmission systems are
voltage and improvement of utility power factor. It
extremely expensive and take considerable amount of
offers several advantages over conventional
time to build up. Therefore, in order to meet
thyristorised converters [2] in terms of speed of
increasing power demands, utilities must rely on
response. The penalty paid for this improvement is in
power export/import arrangements through the
terms of introduction of some harmonics, which
existing transmission systems. Power electronic
requires separate handling using active filtration
devices are gaining popularity for applications in the
techniques. Moran et al [3] have shown in details how
field of power transmission and distribution systems.
the utilization of SPWM techniques reduces harmonic
The reactive power (VAR) compensation and control
distortion. It has also been shown that an increase of
have been recognized [1] as an efficient & economic
modulation index reduces the size of the link reactor
means of increasing power system transmission
and stress on switches which are significant issues in
capability and stability. The FACTS (Flexible AC
practical implementation. The modeling and analysis
Transmission Systems) devices, such as STATCOM
of STATCOM steady state and dynamic performance
has been introduced more recently which employs a
with conventional control method have been studied
VSI with a fixed dc link capacitor as a static
Schauder and Mehta [4]. In [6] the dynamic responses
replacement of the synchronous condenser. In a
and steady state behavior of STATCOM with
traditional synchronous condenser, the field current of

S.K.Das and Dr.J.K.Moharana Page No. 17

SVPWM has been studied and the advantages of can be explained to be inductive if the load is
introducing SVPWM inverter with higher values of capacitive.
MI are highlighted. Starting with an established
steady state open loop model of the STATCOM, the B. Modeling
present paper goes on to develop closed loop model
The modeling of the STATCOM, though well known,
for investigating transient performance of the
is reviewed in the lines below, for the sake of
STATCOM by using small signal scheme. First, in
convenience. The modeling is carried out with the
Section III, a small signal model is developed to
following assumptions:
establish the effect of the variation of the modulation
1) All switches are ideal
index 'm' and the phase angle 'α ' (between the 2) The source voltages are balanced
STATCOM terminal voltage and the voltage at point
of common coupling (PCC)) on the STATCOM 3) Rs represents the converter losses and the losses
terminal current and the dc link voltage. Thereafter, a of the coupling inductor
closed loop control scheme has been proposed, so that
DC link voltage may be kept unchanged. This scheme
is both an extension and a significant improvement of
the scheme suggested by Cho et al [7]. The results
obtained by this method have been compared and
appropriate conclusions have been drawn.

II . MODELING OF THE STATCOM AND

ANALYSIS
A. Operating principle
As is well known, the STATCOM is, in principle, a
static (power electronic) replacement of the age-old Fig.1: Schematic diagram of STATCOM
synchronous condenser. Fig.1 shows the schematic
diagram of the STATCOM at PCC through coupling
inductors. The fundamental phasor diagram of the
STATCOM terminal voltage with the voltage at PCC
for an inductive load in operation, neglecting the
harmonic content in the STATCOM terminal voltage, is
shown in Fig.2. Ideally, increasing the amplitude of the
r
STATCOM terminal voltage Voa above the amplitude
r
of the utility voltage Vsa causes leading (capacitive)
r
current IC to be injected into the system at PCC.
Fig.2: Phasor diagram for inductive load operation
r r
IC _ a , the real component of I C , accounts for the
losses in the resistance of the inductor coil and the 4) The harmonic contents caused by switching
power electronic converter. Ideally, if the system losses action are negligible
r The 3-phase stationary abc coordinate vectors with
can be minimized to zero, IC _ a , would become zero,
1200 apart from each other are converted into αβ
r
and IC would be leading at perfect quadrature. Then, 2-phase stationary coordinates (which are in
r r quadrature). The α axis is aligned with a axis and
Voa, which is lagging and greater than Vsa , would also leading β axis and both converted into dq two-
r phase rotating coordinates. The Park’s abc to dq
be in phase with Vsa . The STATCOM in such a case
transformation matrix is
operates in capacitive mode (when the load is
inductive). Similarly, the operation of the STATCOM

S.K.Das and Dr.J.K.Moharana Page No. 18

 Cos(ϖt ) Cos(ϖt − 2π / 3) Cos(ϖt + 2π / 3) idc = S T ic,abc (10)
2
K=  Sin(ϖt ) Sin(ϖt − 2π / 3) Sin(ϖt + 2π / 3)  (1)
3  = m [0 0 ] i 
T (11)
 1/ 2 1/ 2 1/ 2  i
dc
1
 cq
i
cd
i
co 
The actual proposed circuit is too complex to analyze The voltage and current related in the dc side is given by
as a whole, so that it is partitioned into several basic dv dc
i dc = C (12)
subcircuits, as shown in Fig.1. The 3-phase system dt
voltage Vs,abc lagging with the phase angle α to the Now replacing (11) in (12) yields
dv dc m
= i cd (13)
STATCOM output voltage Vo,abc and differential form dt C
of the STATCOM currents are defined in (2) and (3). The complete mathematical model of the STATCOM in
  dq frame is obtained as given in (14)
 V sa 
 Sin ( ω t − α )  (2)
2  2π  (14)
=  V sb  =  R 
V V s  Sin ( ω t − α − ) − L
s
− w 0 
s , abc   3  3   i cq   s
  i cq   − Sin α 
 V sc  d   =  w Rs m   + Vs  Cos α 
 Sin ( ω t − α + 2 π  i cd − − i cd
 dt    Ls    
3 
Ls Ls
 v dc     v dc   0 
m
 0 0 

Ls
d
dt
(ic ,abc ) = − R s ic ,abc + V s ,abc − Vo ,.abc (3)

C. Steady State Analysis
C 

where , Vs ,ϖ, Rs and Ls have their usual The steady state equations of Icq, I cd , Vdc, Pc , Qc are
connotations. The above voltages and currents are given in (15-16) and their responses with the
transformed into dq frame to give, parameters given in Table 1, are shown in Fig.3 and
Fig.4.
Vs ,qdo = Vs − Sinα [ Cosα 0 ]T (4)
I cq = −
V s Sin α
, I cd = 0
Rs (15)
d  
Ls (icq ) = −Rs icq − wLs icd + Vsq − Voq (5a) V dc =
Vs
M


Cos α −
wL s
Rs
Sin α 

dt
d 2
V s
Ls (icd ) = wLs icq − Rs icd + Vsd − Vod (5b) Pc = [1 − Cos (2 α )]
dt 2 R s (16)
The switching function S of the STATCOM can be V s
2
defined as follows Q c = Sin (2 α )
2 R s
  Table.1
S a   Sin ( wt ) 
 π  (6)
S = Sb  =
2 2 Sl. Parameters Sym Values
m  Sin ( wt − )
  3  3  No. bol
 S c   Sin ( wt + 2 π )  1 Frequency f 50 Hz
 3  2 Angular Frequency ϖ 314 rad/sec
The modulation index, being constant for a 3 RMS line-to-line Voltage
programmed PWM, is given by,
Vs
4 Coupling Resistance
V o , peak 2
(7)
Rs 1.0 Ω
MI = = m
v dc 3 5 Coupling Inductance Ls 5.0mH

The STATCOM output voltages and dq 6 DClink capacitor C 500 µ F

transformation are given by, 7 Modulation Index M 0.979
8 Phase angle α
Vo.abc = Svdc (8) m 50
9 Load Resistance 52 Ω
T RL
Vo ,qdo = m 0 1 0[ ] v dc (9) 10 Load Inductance 126mH
LL
The dc side current in the capacitor and its dq 11 Load Power factor 0 . 79
φ
transformation may be written as,

S.K.Das and Dr.J.K.Moharana Page No. 19

 Fig.3: Steady State response of Icq Icd and Vdc Fig.6: Transient response of icd

Fig.7: Transient response of vdc

Fig.4: Steady State response of Pc and Qc
It can be seen that the transient responses take about
one and half power cycle to reach at their steady state
The theoretical results of (15-16) are verified by the values.
simulation results.
III. SMALL SIGNAL SCHEME
D. Transient Analysis A. Modeling
To know the dynamic characteristics of STATCOM
The transient responses of the system in (14) are system, the small signal analysis is to be done. For a
simulated in MATLAB for an initial condition of given operating point, small signal equivalent circuit is
vdc = 100V and the parameters given in Table.1 and derived based on the following assumption:
i) The disturbance is small,
are shown in Fig.5 to Fig.7. ii) Hence, the second order terms (products of
variations) are negligible,
iii) The phase nominal value of angle α is small.
With the above assumptions eqns. (9) and (11)
can be rewritten as,
Vod + vˆod (t ) = MVdc + Mvˆdc + m
ˆ Vdc + m
ˆ vˆdc (17)

I dc + iˆdc = MIcd + Miˆcd (18)

Sinαˆ = αˆ , Cos αˆ = 1 and α0 =0 (19)

By using (5), (17), (18) and (19), and applying Laplace
Transformation, we have,
Fig.5: Transient response of icq

S.K.Das and Dr.J.K.Moharana Page No. 20

 sLs + Rs 0   iˆcq   − Vsαˆ (s) 
ϖLs
 − ϖL  
 s sLs + Rs M   iˆcd  = − Vdc mˆ (s) (20)
 0 −M sC vˆdc   0 
The important transfer functions of the states of the
STATCOM in small signal model can be derived as
iˆcq (s) V ϖLs C
dc ˆ 2 2
= (21a) icq ( s ) = − V s ( L s Cs + R s C + M ) (21b)
mˆ ( s) A(s) αˆ ( s ) A( s )
Fig.10: Step response of v̂dc for small change of m̂ and
iˆcd (s) − VdcC (sLs + Rs )s ˆ − VdcϖL s Cs
= (22a) icd ( s ) = (22b) αˆ
mˆ (s) A(s) αˆ ( s ) A( s ) C.Closed loop response
vˆ dc ( s ) − MV dc ( sL s + R s )
(23a) vˆdc ( s ) − MVsϖLs
(23b) To achieve fast dynamic response without using
= =
mˆ ( s ) A( s ) αˆ ( s ) A( s ) independent dc voltage source, it is required that the
2 3 2 2 2 2 2
A ( s ) = CL s s + 2 L s CR s s + [ C { R s + (ϖ L s ) } + M L s ] s + Mcapacitor
R voltage vdc be kept constant by controlling
s
the phase angle. Simultaneously the load reactive power
B. Open loop responses should also be compensated by controlling the
modulation index of the converter. If the capacitor
The iˆcq , iˆcd and v̂dc in (21-23) are simulated in
vˆdc(s) should be zero giving,
voltage is constant,
MATLAB with the same parameters given in Table.1
and also αˆ = −5u(t ) and mˆ = 0.1u(t) .The responses are VsϖLs
mˆ (s) = − αˆ (s) (24)
shown in Fig.8 to Fig. 10. Vdc (sLs + Rs )
This principle though suggested by Cho et al [7], its
implementation is not correctly presented by the
authors there. The PI controller for controlling the
reactive current of the STATCOM with independent
control of dc voltage is proposed as shown in Fig.11.
The parameters of PI controller have been determined
by root locus method and their values are K pi = −0.02

and Kii = −6.66. Using the parameter values given in

Fig.8: Step response of iˆcq for small change of m̂ and αˆ Table.1 and taking reference reactive current as the
reactive component of the load current, the whole
system is simulated in SIMULINK.

Fig.9: Step response of iˆcd for small change of m̂ and αˆ

Fig.11: Proposed iˆcq control block

It is seen that the power factor improves from 0.79 as per the
R-L load taken to unity after one and half power cycle and all
relevant responses are given in Fig.12 to Fig. 14 (for phase

S.K.Das and Dr.J.K.Moharana Page No. 21

A). Fig.15 shows that the maximum overshoot of dc link
voltage is to 525 volts only. Fig 16 and 17 show the
corresponding dynamics of 'α ' and ' m' respectively.

Fig.15: DC link voltage Vdc

v i
Fig.12: System voltage sa and current sa before compensation

Fig.16:Phase angle α variation

v
Fig.13: System voltage sa and compensating current ca i

Fig.17: modulation index ‘m’ variation

Fig.14: System voltage vsa and current isa after compensation

With the small perturbation of reference current
(reactive current of load),very small change in DC link
voltage is obtained as shown in Fig.19 and when the
reference current goes to its original value then DC link
voltage immediate jumps to its steady state value.
Fig.18: Load current of ilq with STATCOM current icq

S.K.Das and Dr.J.K.Moharana Page No. 22


Fig.19: Vdc with change of reference
V. CONCLUSION
In this paper, the STATCOM model has been
linearized by small signal scheme [7] with nominal
value of ' α ' and 'm' . The system with controller
operates transient to steady state value on 0.03 sec. in
case of small signal scheme. The transient overshot
nearly 525V in case of small signal scheme. With the
small perturbation of reference current (reactive current
of load),very small change in DC link voltage is
obtained as shown in Fig.19 and when the reference
current goes to its original value, the DC link voltage
immediate jumps to its steady state value. On overall,
the small signal scheme is better for the operation of the
STATCOM in compensating mode. It must be pointed
out that all the results present here are on the basis of
the assumption that the DC link voltage has stabilized
to some significant value before the control can be
applied. However, in practice the issue of the charging
the dc link voltage to the required value is quite
significant. In most cases, there is a separate charging
circuit for the dc link voltage. The authors are working
on plausible method of eliminating such an extra
starting arrangement, so that the controller may become
operational while the dc link voltage at minimal value.
V. REFERENCE
[1] A.T. Johns, A.Ter-Gazarian and
D.F.Wame,”Flexible ac transmission systems
(FACTS)”, IEE Power and Energy Series,
London, U.K.
[2] R.M.Mathur and R.K. Varma, “Thyristors-based
FACTS Controllers for Electrical Transmission
Systems, IEEE Press”, Wiley-Interscience
Publication.
S.K.Das and Dr.J.K.Moharana
[3] L.T. Moran, P.D.Ziogas and G.Joos, “Analysis
and Design of a Three-Phase Synchronous
Solid- State Var Compansator”, IEEE Trans.
Industry Application, Vol. 25, No. 4, 1989, pp.
598-608.
[4] C.Shauder and H.Mehta, “Vector analysis and
control of advanced static VAR compensators”,
IEE Proc, 140, No. 4, July 1993.
[5] A. Draou, M. Benghanem and A. Tahiri,
“Multilevel Converter and VAR Compensation”,
Power Electronics Handbook”, pp.599-611,
Academic Press, 2001.
[6] M.Sengupta,J.K Moharana and A.Sengupta,“
Study on an Advanced Static VAR Compensator
switched from a Space Vector PWM inverter –
Analysis, simulation and comparison with the
conventional sinusoidal PWM, NPEC 2003, IIT
Bombay,16-17 Oct 03 pp 72-78.
[7] G.C.Cho,N.S.choi,C.T.Rim,and G.H.Cho,
“Modeling,Analysis and Control of STATIC
VAR Compensator using three level
inverter”,IEEE,Ind.Socity,Annual
Meet,page837-843,1992
[8] G.C.Cho,N.S.choi,C.T.Rim,and G.H.Cho,
“Analysis and Controller Design of STATIC
VAR Compensator Using Three-Level GTO
Inverter”,IEEE,Transations on Power
Electronics,Vol.11,No.1,page57-65,January
1996.
[9] D.M.Brod and D.W.Novotny, “Current control
of VSI- PWMinverter”IEEETrans.Industrial
Appl, Vol.IA-21,pp.562- 570, July/Aug. 1985.
[10] S.Buso, L.Malesani and P.Mattavelli,
“Comparison of Current Control Techniques
for Active Filter Application” IEEE Trans.
Industrial Electronics, Vol.45, No.5, pp.722-
729, October 1998.
[11] P.S.sensarma, K.R.Padiyar and
V.Ramnarayanan, “Analysis and Performance
Evaluation of a Distribution STATCOM for
Compensating Voltage Fluctuations”, IEEE
Transaction on Power Delivery, Vol.16, No.2,
pp.259- 264, April 2001
[12] A.M. Kulkarni and K.R. Padiyara, “Design of
Reactive Current and Voltage Controllers of
Static Condenser”, Power and Energy
System,Vol.19,No.6,pp.397-410,1997
Page No. 23
 Design of a Gearless, Axial Flux Wind Power Generator
L.Sreedhar1, V. Santosh Kumar2
Associate Professor, Dept. of E.E.E, Pydah College of Engineering and Technology, lollasreedhar@yahoo.co.in1.
Associate Professor, Dept. of E.I.E, GITAM UNIVERSITY. vsantoshpatnaik97@gmail.com2

ABSTRACT
With rapid development of wind power technologies and significant growth of wind power capacity installed
worldwide, various wind turbine concepts have been developed. The wind energy conversion system is demanded to
be more cost-competitive, so that comparisons of different wind generator systems are necessary. The design aspects
of an axial flux wind generator are described. A simple wind turbine is proposed to be developed with cost effective
and technology effective methods, using materials like PVC tubes for blade design, Nylon shaft for Rotor assembly,
an Axial flux excitation system that uses neodymium magnets in place of the regular rotor poles. Emf producing
coils are proposed to be setup in the stator, for the same advantage as expected in a commercial alternator. The issue
of gear box has been dispensed with in small power turbine design to further simplify the design.
Keywords Rotor assembly, Neodymium magnets, Charge controller, Wind direction sensor and Wind generator
design.

I. INTRODUCTION called ‘drag’. The combination of lift and drag causes

Consumption of energy based on fossil fuels is
considered to be the major factor for global warming
and environment degradation. The utilization of
naturally occurring renewable energy sources as an
alternative energy supply has been assuming more
importance. Research on performance improvement of
and cost reduction in such nonconventional energy
conversion systems is being accorded the highest
priority. The wind energy is a green energy and is
becoming one of the most appealing energy today.
Small scale wind turbines (less than 20 kW) are
typically designed and manufactured by relatively
small mechanical engineering companies. One
component that causes particular problem is the
electrical generator, as suitable generators are not
always readily available. For smaller turbines the
manufacturer is inevitably looking at a specially built
generator. Hence smaller wind turbine developers
would like to employ a direct drive generator (without
gearbox) whilst having a generator design that is
mechanically simple, so enabling them to build their
own if necessary.
II. WORKING OF WIND TURBINE
When the wind blows, air pocket of low pressure air
forms on the downwind side of the blade. Low
pressure air pockets then pull the blade toward it,
causing the rotor to turn. This is called ‘lift’. The force
of the lift is actually much stronger than the wind’s
force against the front side of the blade, which is
the rotor to spin like propeller. The rotor of the turbine
is attached to a hub that is mounted on a turning shaft.
The transmission is attached to a shaft which turns a
generator that makes electricity. The electricity thus
obtained can depend on the end user’s need, and can
be either alternating or direct. Direct current can be
generated by driving brushless D.C generator attached
to the windmill’s rotor. If the system is designed to
produce a.c power, a constant angular velocity and
force is desirable. Unfortunately, the wind velocity is
neither constant in magnitude or direction nor is it
constant from the top to the bottom of a large rotor.
This imposes severe cyclic loads on the turbine blades,
creating fatigue problems. There are also times when
the wind velocity is too low to produce significant
power. This means that some sort of energy storage
system or back up energy system must be utilized. The
wind systems must also be shut down when the wind
velocity is too high so as to prevent damage to the
machine.
A 150w to 200w vertical turbine will be able to cover
10 to 15% of a household’s electricity use. On the
basis of power available, wind power plants are
grouped into three categories:
Small size: - Power output: 10-50 KW and rotor
diameter of 1-16m.
Medium size: - Power output: 50-500 KW and rotor
diameter of 16-50m.

L. Sreedhar & V. Santosh Kumar Page No. 24


         4

 
 
Large size: - Power output: 500-5000 KW and rotor
diameter of 50-130m.
Wind turbines are classified into two general types:
Horizontal axis and Vertical axis. HAWT is the most
common wind turbine design. In addition to being
parallel to the ground, the axis of blade rotation is
parallel to the wind flow. As long as the rotor blades
are faced in the direction of the wind, the turbine can
efficiently convert the wind’s kinetic energy to
electrical energy.
Some machines are designed to operate in an upwind Fig 3.1Wind flow through the opening
mode, with the blades upwind of the tower. In this From the above equation it is clear that long blades of
case, a tail vane is usually used to keep the blades turbine rotor will produce more power but there are
facing into the wind. Other designs operate in a technical limitations for doing so. Therefore several
downwind mode so that the wind passes the tower turbines with small rotor are used to produce required
before striking the blades. Without a tail vane, the power. Also the wind power should be above 10m/s
machine rotor naturally tracks the wind in a downwind for power extraction. The mass flow rate of air that
mode. passes through the swept area of wind turbine varies
III. ANALYSIS OF WIND TURBINE with the wind speed and air density.

3.1 WIND POWER CALCULATIONS
The moving wind possesses kinetic energy in it, and
can be harnessed for other applications by suitable
The power given in the equation (4) is an
instantaneous maximum power because the wind
speed varies with time.
Therefore instantaneous power of wind (
 )

1
conversion system. Like solar energy, the wind energy

     5

is also associated with problems because of its

availability in dilute form. Let m be the mass flow rate
of air moving with velocity v through opening
provided at any time. The kinetic energy associated
with this moving air is;
2
The wind speed varies continuously and the average
power available from wind for a specific period is

    1


taken in consideration;

     6

  

Where   the average wind velocity
This is the maximum power associated within the

  The instantaneous velocity of wind

wind. Fig shows the motion of wind through the
opening of area (A) flowing with velocity (V). The
discharge rate (Q) of air through the opening at 1 and The above power contained in the flowing wind is not
2 is given as;
  
practically extracted by aero generator and depends
upon the wind speed, turbine type, and spillage losses
etc. the spillage loss occurs between the blades as the
Mass flow rate (kg/s) of air through the opening,
       2

air spills through the gap between blades.


The Power developed in aero generator = power

   
 ,"#$
coefficient power available in wind
From equations 1 and 2

        3
Where  = power coefficient


IV. WIND GENERATOR DESIGN &
kg/ and changes by 10% with altitude. Therefore
As the density of air at sea level is 1.25
FABRICATION
power developed by wind depends upon the swept 4.1 BLADE DESIGN
area (A) and wind velocity (V). If, D the diameter of
The primary part of a wind turbine is its blade.
opening the above equation (3) reduces to;
While designing the blade, we should consider the

L. Sreedhar & V. Santosh Kumar Page No. 25

parameters such as the “drag”, “tip speed ratio” and
the “aerodynamic profile” of the whole blade. So, for
high scale wind turbine applications the blades are
designed very specifically to follow the Bernoulli’s
Principle. The basic blade design of a wind turbine can
be observed from fig 4.1.
For the small scale wind generator purpose, the blade
design has been modified using a 4 inch PVC pipe
measuring a length of 2 feet.
Fig4.1 General Blade Design. (Cross-sectional view)
4.2 HORIZONTAL SHAFT WITH BEARING
The support frame of the wind turbine starts with the
horizontal shaft on which the blade hub and the rotor
are arranged. The shaft should be tough enough to
handle the stress of the blades and the rotor disc. The
centre of the shaft is fitted into the inner rays of the
bearing which ensures free rotation of the rotor. In
commercial wind turbines, the horizontal shaft is
geared.
As the project is for a small scale purpose, we’ve
considered the usage of a bearing to ensure complete
free rotation of the rotor and blades.
Fig 4.2 Horizontal shaft with bearing.
L. Sreedhar & V. Santosh Kumar
4.3 ROTOR: P . M EXCITATION
This project consists of an indigenously built generator
consisting of a permanent magnet rotor and a stator
consisting of the coils. The rotor comprises of 8
equally spaced rare earth Neodymium alloy magnets
(NdFeB) with the grade of N35. The magnetic field
strength of each magnet is 0.35 Tesla. The rotor disc is
a 6 inch iron disc. The reason for considering a Ferro-
magnetic material for the rotor is to provide the
minimum possible reluctance path between the
magnets.
Fig4.3 Rotor with magnets arranged on it
4.4 STATOR- COIL ASSEMBLY
The next part of the generator is the stator, on which
we have the coil assembly. The coils taken are 30
SWG of 700 turns each. The coils have been arranged
on a disc shaped flexi-glass. To cut out the reluctance
to as much extent as possible, the coils have been
arranged on a metal ring. Each of the coil consists of a
stack of soft iron core laminations which provides
better continuity for the flux lines from the magnets on
the rotor to the coils on the stator. The air gap between
stator and the rotor is a mere 1-2 mm.
The figure below (fig.17) shows the complete stator
arranged with all the coils and the metal ring.
Fig4.4 Stator Coil Arrangement
Page No. 26
4.5 WIND VANE voltage from the wind generator is less than or within
a tolerable window of 12V, the voltage flows through
The second to last part in the complete assembly
a rechargeable battery connected and charges it. If the
of the wind turbine is the wind vane. The wind vane’s
voltage is more than the tolerable voltage, the relay
main purpose is to sense the direction of wind and
trips and the extra energy is dumped through an
direct the whole system to move accordingly. The
external dump resistor or a dummy load. LED
design of the wind vane also follows tthe Bernoulli’s
indicators
ndicators are provided to know whether the circuit is
principle.
in dumping mode or in charging mode. Fig.20 shows
the charge controller circuit.

Fig4.5 The wind vane

4.6 VERTICAL MAST
The final and the main part of the support frame for
the wind turbine is the vertical mast which provides
the base support for the whole system. In commercial
wind generators, the verticall masts are long and large Fig4.7 The charge controller circuit
hollow tubes, which provide a path for all the wiring
and even contain spiral steps through which the
maintenance engineers access the generator system
which is present at the top of the vertical mast.
For our small scale project purpose, we have used a
Stainless Steel hollow tube providing the support for
our system.

(a) Charge ON (b) Dump ON

Fig4.6 The vertical mast with the reed sensor arrangement
4.7 WIND GENERATOR CHARGE
CONTROLLER
The chargee controller circuit is 555 timer based. The
charge controller’s purpose is to control the flow of
charge throughout the system. It consists of a 555
timer and a 12V relay. The working is that, if the
(b)
4.8 WIND GENERATOR DIRECTION SENSOR
SENS
To know the direction of wind, we have arranged a
circuit consisting of reed relays and LED indicators
through which we can know the direction of wind. The
support frame consists of a magnet which is held
directly above the reed sensors. Once the magnetmagne
passes the reed sensor, the reed closes the switch and
the supply from one end to the other of the reed switch
leads to the respective LED to glow. So, to know all
the 8 cardinal directions, we have taken 8 reed sensors
and 8 LED indicators connected to it respectively. The
figures 4.10(a) and 4.10 (b) show direction sensor
arrangements.

L. Sreedhar & V. Santosh Kumar Page No. 27

 V. SYSTEM SIMULATIONS USING
PROTEUS
5.1 Simulation of Charge Controller
Charging Mode (Charge-LED is ON)

Fig4.8 Direction sensor arrangement

4.9 WIND GENERATOR VOLTAGE

MEASUREMENT

The voltage measurement circuit comprises

of a Programmed Microcontroller AT89C51, an
ADC0804, a crystal oscillator and a LCD 16x1
Display. The circuit displays the voltage generated Fig 5.1(a): charge controller circuit in charging mode
across the coils. Discharging Mode (Dump LED is ON)

Fig4.9 The voltage measuring circuit Fig 5.1(b): Charge Controller in discharging mode

4.10 PROPOSED & FINAL MODEL

5.2 Simulation of Direction Sensor
(Screenshot : ‘EAST’ direction LED is ON)

Fig4.10 The proposed model, & the final assembled wind turbine. Fig5.2: Simulation of direction sensor circuit.

L. Sreedhar & V. Santosh Kumar Page No. 28

5.3 Simulation of Digital Voltmeter
Fig5.3: Digital Voltmeter Simulation using Proteus.
VI. CONCLUSION
After initial review of the background of wind
generators, a wind turbine along with the supporting
auxiliary apparatus was designed and fabricated. It
was found to be delivering electrical power as
expected.
Fabrication of mechanical parts and their
assembly posed few problems, which were resolved
using alternative design or by change of material parts.
Auxiliary support circuits like Wind Direction Sensor,
Charge Controller and Digital Voltmeter are designed
and tested successfully.
In the excitation system, (i.e. rotor consisting
of Neodymium magnets N35, showed remarkable
improvement in coil voltages with the addition of an
iron plate to reduce reluctance of the magnetic path.
These results were comparable to the calculated
values.
The design and fabrication experience would
be very useful in determining the selective procedures,
to be adopted for the fabrication of higher capacity
Wind Turbine generators, if necessary.
L. Sreedhar & V. Santosh Kumar
VII. REFERENCES
[1] Non-conventional Energy Resources by S.Hasan
Saeed, D.K.Sarma (Katson Pub.)
[2] Non conventional Energy Resources by
B.H.Khan,Tata Mcgraw-Hill Publications
[3] “Analysis of Magnetic field in an Axial flux
permanent magnet synchronous Generator using
an analytical method” by T.F.Chan.
[4] IEEE Paper : “Analysis of Magnetic Field in an
Axial-Flux Permanent-Magnet Synchronous
Generator Using an Analytical Method” by T. F.
Chan, Member, IEEE1, Shuming Xie2, and L. L.
Lai, Senior Member, EEE31Department of EE,
The Hong Kong Polytechnic University, Electrical
Engineering Institute, Shenyang University of
Technology, China, 3Energy Systems Group, City
University, London, UK.
[5] Paper: “Design of an Axial Flux Permanent
Magnet Wind Pwer Generator” by L. Soderlund,
A. Koski, H. Vihriala, J-T. Eriksson and R.
Perala.
[6] IEEE Paper: “Axial Flux Permanent-Magnet
Machine under Optimum Control Strategy for
Wind Power Generation” by Roberto H.
Moncada, Raúl A. Rodríguez, Juan A. Tapia, and
Thomas M. Jahns*, Elec. Eng. Dept. University of
Concepcion,Chile.
[7] IEEE Paper: “A Novel Wind Energy System” by
Tze-Fun Chan, Member, IEEE, and Loi Lei Lai,
Fellow, IEEE
[8] Paper: “Design and Testing of a Permanent
Magnet Axial Flux Wind Power Generator” by
Garrison F. Price, Todd D. Batzel, Mihai
Comanescu, and BruceA. Muller Pennsylvania
State University, Altoona College.
Page No. 29
 Comparative Analysis of Different Optimization
Techniques for Selective Harmonic Elimination of
Cascaded Multilevel Inverter with equal DC Sources
Anup Kumar1 and A. Dasgupta2
Assistant Professor, Gandhi Institute For Education and Technology, Bhubaneswar1
Professor, Kilinga Institute For Industrial Technology,Bhubaneswar2
ABSTRACT-
In this paper, elimination of lower order harmonics in a cascade multilevel inverter has been taken into account by
using different optimization techniques. Each separate H-Bridge has been given equal dc sources. Switching angles
of each level of cascaded multilevel inverter has been optimized by solving non-linear transcendental equation.
Comparison of THD’s has been made for different optimization techniques. Theoretical results are verified by
simulation of a 7-level 3-phase cascade multilevel inverter. Experimental analysis has been shown for 5-level single
phase cascade multilevel inverter. Results show that the proposed method effectively eliminates specific lower order
harmonics, and the output voltage is resulted in low total harmonic distortion (THD).
Keywords
Multilevel inverter (MI), Newton Rapson Method (NRM), Mathematical Theory of Resultant, particle swarm
optimization (PSO), transcendental equations, selective harmonic elimination (SHE)

I. INTRODUCTION Different techniques have been implemented to solve

Multilevel voltage source inverters are widely used for
high voltage and high power applications.. The
principle of operation of this inverter is usually based
on synthesizing the desired output voltage waveform
from several steps of voltage, which is typically
obtained from dc voltage sources [3]. Among different
topologies of multilevel inverters, the cascade
multilevel inverter has received special attention due
to its modularity and simplicity of control [1]. In
cascaded multilevel inverter, s number of separate dc
sources generates (2s +1) numbers of levels.
In case of equal dc sources, in order to control the
output voltage and to eliminate the lower order
harmonics different modulation techniques has been
adopted such as sinusoidal PWM technique. But this
technique doesn’t eliminate specific order of
harmonics and large percentage of harmonic are still
present. This results in high value of THD. But to
eliminate specific order of harmonics such as 3rd, 5th,
7th, 11th & 13th harmonics, non-linear transcendental
equations have to be solved for the determination of
switching angle. These switching angles determines
the on time and off time of each switch of multilevel
inverter. Solution of non-linear transcendental
equation is not an easy task .It requires a lot of
mathematical computation.
transcendental equations. Newton Raphson Method is
one of them. This technique requires good initial
guess. But this method gives satisfactory result up to
7-level multilevel inverter. For higher level multilevel
inverter this method is not applicable. This method is
also time consuming.Mathematical theory of resultant
(Polynomial Method) is an important technique to
solve non-linear transcendental equation. In this
method, each equation has to be converted into
equivalent polynomial. This method gives good result
for low level multilevel inverter. This method is quite
successful up-to 9- level multilevel inverter with equal
dc sources. But for higher level multilevel inverter it is
very difficult to solve trigonometric equations and
very difficult task to convert the transcendental
equations into equivalent polynomial.
II. POWER TOPOLOGY OF CASCADED
MULTILEVEL INVERTERS

Anup Kumar and A. Dasgupta Page No. 30

Fig.1.(a) Topology of a single-phase cascaded
inverter. (b) Staircase output phase voltage [3].
Fig. 1 (a) shows the structure of a single-phase Algorithm For N-R Method:-
cascaded multilevel inverter. With s number of
separate dc sources (2s+1) levels of output voltage can 1. Guess the initial values of switching angles with
j=0.
Assume @ A  ,@1A @2A @3A 9^C
be synthesized. Fig. 1 (b) shows the staircase output
voltage waveform generated by the multilevel inverter.

Calculate value of F(@ A) = F0

Ѳ1, Ѳ2, Ѳ3………. Ѳs represent the switching angles of
each level of multilevel inverter. The advantage of this 2.
topology is that optimized circuit layout and packing Fj = [f1j, f2j, f3j]
is possible because each H-bridge has same structure
cos@1E
0 cos@2E
0 cos @3E

and there is no requirement of clamping diodes and Fj =

D cos : @1E
0 cos : @2E
0 cos : @3E
F
voltage balancing capacitors. But one major
cos′ : @1E
0 cos′ : @2E
0 cos ′ : @3E

disadvantage of this process is that separate dc sources

are required for each H-bride.
3. Linearise equation
HI A
The number of output phase-voltage levels in a
F0+G K L@ A  C G K =
HI
HJ HJ
cascade multilevel inverter is 2s+1, where as ‘s’ is the
MNOPQRS
0 MNOPQTS
0 MNOP QUS

D MPNOPV : QR MPNOPV : QUS

F
number of separate dc sources. To obtain the 3-phase
0 MPNOPV : QTS
0
MPW NOPVW : QRS
0 MPXNOPVW : QTS
0 MPXNOP VW : QUS

configuration, the outputs of 3 single phase cascaded

inverters are connected in Y or ∆.
And d@ A = ,L@1A , L@2A , L@3A 9
HI A
III. HARMONIC ELIMINATION WITH
4. Solve the d@ A  YZ G K C M [ A

EQUAL DC SOURCES
HJ
By applying Fourier series analysis, the staircase 5. Update the initial values
output voltage of multilevel inverter with equal dc
sources can be described as follows:- α j+1 = αj + dαj
6. Repeat the steps from 2 to 5 and the solutions must

∑1
23,,4,5,… 4 Vdc/sπ ,CossѲ1
0 CossѲ2
0 0 \ α1 ] α2 ] α3 ] ^/2
V (ωt) = satisfied the conditions

CossѲ3
0 7 … . . 0 CossѲs
9 : Sinωt
(1)
Where:- n and n’ represents the harmonics to be
Switching angles Ѳ1 to Ѳs must satisfy the following eliminated.
condition : α 1 , α 2 and α 3 are the switching angles to be
determined
Ѳ1 ≤Ѳ2 ≤ Ѳ3 ≤ Ѳ4 ≤……………. Ѳs ≤ Ѳ/2 (2)
Note:- Above given algorithm is applicable to the 7-
The number of harmonics that can be eliminated from level Cascaded Multilevel Inverter.
the output voltage of the inverter is (S-1 ). Where S
represents the number of separate dc sources. It should
be noted that elimination of triple harmonics with S. Modu Alpha1 Alpha2 Alpha3 %
three phase power system is not applicable because No lation THD
Index (α1) (α2) (α3)
these harmonics will be eliminated from line-to-line
voltage. 01 0.50 39.4252 56.2529 80.09949 11.2332
IV. NEWTON RAPSON METHOD (NRM) 02 0.60 33.4951 54.75757 67.10481 10.3726
03 0.70 18.3104 54.1012 64.3480 10.0943
In numerical analysis, Newton’s Method is also
known as the Newton’s –Raphson Method (N.R.M.), 04 0.80 11.50499 28.71664 57.1067 7.5778
named after Issac Newton & Joseph Newton, is a
Table 1:- Switching Angles & % of THD 3 phase 7
method for finding successively better approximations
levels M.I.
to the roots (or zeros) of a real valued functions. This
method can also be extended to complex functions and
to system of equations.

Anup Kumar and A. Dasgupta Page No. 31

 Fig 2:- Switching Angles & % THD vs.
Modulation Index of 3 phase 7 level M.I. Fig 4:- Switching Angles & % THD vs Modulation
Index of 3 phase 7 level M.I.

Fig 3:- FFT Analysis of 3 Phase 7- Level Cascaded

Multilevel Inverter
Fig 5:- FFT Analysis of 3 Phase 7-Level Cascaded
V. MATHEMATICAL THEORY OF M.I.
RESULTANT (POLYNOMIAL METHOD)
VI. PARTICLE SWARM OPTIMIZATION
The Mathematical Theory of Resultant is used to (PSO)
compute the switching angles in a multilevel converter
so that it produces the required fundamental voltage PSO was first introduced by Kennedy and Eberhart in
while at the same time cancels out unwanted order of 1995 [1]. This optimization technique is based on the
harmonics. It is shown that for a range of the social behaviour of the particles such as their
modulation index the switching angles can be chosen movement in search of food. PSO is a computational
to produce the desired fundamental while at the same method that optimizes a problem by iteratively trying
time eliminating the lower order harmonics. to improve a candidate solution. It optimizes a
problem by having a population of candidate solutions
The Mathematical Theory of Resultant of Polynomials and moving these particles around in search space
is a systematic procedure for finding the roots of according to simple mathematical formulae over these
systems of polynomial equations. particle position and velocity. Each particle is
S. Modulation Alpha1 Alpha2 Alpha3 % determined by two vectors in D-dimensional search
No Index (α1) (α2) (α3) THD space: the position vector X1 = [xi1, xi2,Xi3,…………xiN]
and the velocity vector Vi = [vi1, vi2.,……..viN]. Each
01 0.5 20.446 56.124 89.679 11.5430 particle in the swarm refines its search space through
its present velocity, previous experience, and the
02 0.6 11.819 41.712 85.716 12.3682
experience of the neighbouring particles. The best
03 0.7 18.305 44.119 64.361 11.1536 position of the particle i found so far is called the
04 0.8 4.439 30.003 89.868 10.7771 personal best and is denoted by Pi = [Pi1, Pi2, Pi3,…….,P-
iN]. The best position of the particle i in the entire
05 0.9 12.366 12.761 38.131 9.5871 swarm is called the global best position of the particle
and is denoted by Pg = [Pg1, Pg2,Pg3,…….,PgN]. At first,
Table 2:- Switching Angles & THD of 3 phase 7 levels M.I. the velocity of the ith particle in the nth dimension
space is updated using equation (3) and then position
of the particle is updated using equation (4).
Vidk+1 = w *Vidk + c1 * rand1 * (Pbestid - xidk) + c2 *
rand2 *(Gbestgd- xidk) (3)

Anup Kumar and A. Dasgupta Page No. 32

Xidk = xidk + Vidk+1 (4)
Where ‘w’ is the inertia weight parameter which
controls the global and local exploration capability of
the particle, rand1 and rand2 are the random generation
of angles between 0 - Ѳ/2 and c1 and c2 are the
cognitive and social parameters. For the proposed
method c1 = 2 and c2 = 2.
A large inertia weight factor is used during initial
exploration and its value is gradually reduced as the
search proceeds. The concept of time –varying inertial
weight (TVIW) is given by
W = (w max – w min)* _`abcde f _`ab
ghijklm
+ wmin
W max = 0.9; W min = 0.4
The minimization of THD in multilevel inverter is
achieved by using this PSO algorithm because of its
simplicity in nature and easy to implement,
computational efficient. The procedure for getting
optimized results the objective function is taken as the
THD equation.
P.S.O. Algorithm
Step 1: Create the random initial population size of
switching angles by considering their limitation 0 to
π/2.
Step 2: Initialize the velocity, Pbest, Gbest, iteration
count for computing switching angles.
Step 3: Update the iteration count.
Step 4: Update the velocity and position according to
the equations for movement of particles.
Step 5: Evaluate the fitness or objective function by
using equation.
Step 6: Update the value of Pbest and Gbest.
Step7: Is criterion achieved then go for next step,
otherwise repeat the step 3 to step 6 for best solution.
Step 8: Select the best solution of fitness value.
VII. FORMULATING THE PROBLEM
Selective Harmonic Elimination (SHE) Equations for
3 phase 7 levels CSMLI for equal dc sources are as
follows:-
Cos (Ѳ1) + Cos (Ѳ2) + Cos (Ѳ3) = 3*M
Cos (5Ѳ1) + Cos (5Ѳ2) + Cos (5Ѳ3) = 0
Cos (7Ѳ1) + Cos (7Ѳ2) + Cos (7Ѳ3) = 0
Anup Kumar and A. Dasgupta
Where M is called as modulation index and it is given
by M = V1/ V1max and V1max = 3*(4Vdc/π).
These equations are non-linear transcendental
equations that have been solved by using PSO.
The procedure to solve the SHE equations to find the
optimum switching angles of (2s+1) level CSMLI is as
follows.
1. Set the population size and initialize a population
of the particles with random position between 0
and Ѳ/2. For (2s + 1) level, s numbers of separate
dc sources are required. Thus there will be s
switching angles Ѳ1, Ѳ2, Ѳ3……. Ѳs.
2. The objective of this study is to minimize the
specific harmonics that could be achieved by
finding the switching angles which satisfy the
non-linear transcendental equations. In this case,
THD has been taken as the objective function
because the main objective is to reduce the %
THD and to maintain the fundamental harmonics.
nop
Fundamental voltage is given by
V1=4* [cos
Ѳ
(Ѳ1)+cos(Ѳ2)+cos(Ѳ3)+……………+ cos
(Ѳs)].The objective function considered in this
paper is given below:-
:
qnr s n4r sn5s ntr s7………………….s ntr
n
100.THD has
f(Ѳ) =
been calculated up-to 39th
harmonics.
3. Compare ith particle fitness evaluation with its
personal best.
4. Identify the particle in the entire swarm with
best success so far .
5. Change the velocity and position of the particle
according to equation (3) & (4).
6. In this case 30 iteration has been taken. At each
iteration, positions are updated and new solution
has been created and new population is generated.
Fig 6:- Switching Angles & %THD vs Modulation
Index of 3 phases 7 level M.I.
Page No. 33
S. Modula Alpha Alpha Alpha % THD
No tion 1 2 3 (Up to 39th
Index (α1) (α2) (α3) Harmonics)

01 0.5 17 48 67 11.4739
02 0.7 18.71 44.61 64.41 10.90
03 0.9 5 15 35 5.4189
Table 3:- Switching Angles & THD
Simulation Result
Switching angles are determined using PSO
Technique for modulationn index 0.9. These switching Fig5.7:- FFT Analysis of Line to Line Voltage with
angles are used for giving the gate pulses to the IGBT start time= 0.1 sec & stop time=0.2 sec
and the output voltage & current waveforms are
analyzed using MATLAB.
Switching Angles:-
Alpha1 (α1) = 5 degree
Alpha2 (α2) = 15 degree
Alpha3 (α3) = 35 degree
Voltages applied
lied to separate D.C. sources:
sources:-
V1 = 50v Fig 8:- FFT Analysis of 3 Phase 7-level
7 M.I.

V2 = 50 v
V3 = 50 v
Type of Load = Star connected Resistive Load
Resistance = 1 ohm.
Gate Source = Pulse Generator.
MATLAB Stop Time = 0.2 second.
Fig 9:- Comparison of THD’s of Various Optimization
Technique of 3 phases 7 levels M.I.
VIII. EXPERIMENTAL RESULTS RESULT
Hardware implementation of 1 phase 5 level multi level
inverter requires two H-bridges.
bridges. Each H-Bridge
H consists
of 4 MOSFETs. Therefore, number of MOSFETs
requires are eight. Each H-Bridge
Bridge is given separated dc
sources. Therefore, two separate dc sources are required.
To trigger the MOSFET gate pulses are required, these
gate pulses are given from the output of the optocoupler.
And the inputs of these optocouplers are fed from the
Fig 7:- Output Voltage & Current Waveform of 3 ATMEGA 16 micro controller. Pulse generation
Phase 7-Level M.I. programming is accomplished in ATMEGA 16. The
programming has been done for port D. Output voltage
waveform has been observed in the digital storage
oscilloscope. FFT analysis of the output voltage
waveform is carried out in a computer by loading the

Anup Kumar and A. Dasgupta Page No. 34

output voltage waveform obtained in the CRO through
software SW206 signal analyzer.
Sl Name of the Component Quan
No tity
1. 6N60 N-CHANNEL MOSFET 08
2. Resistive load, 470Ω, 1W 01
3. MCT 2E, optocoupler 08
4. Resistor, 10Ω 08
5. Resistor, 100Ω 08
6. ATMEGA16, Microcontroller chip, 40pin 01
7. 7805 Voltage Regulator 01 Fig 12 :- Output Voltage Waveform of 1 Phase 5
8. Connecting wires and vero board Levels CSMLI Taken From MSO
Table 5:- Components Used in the hardware
implementation of 1 phase 5 levels M.I.
Fig 10 shows the experimental set up of the 1 phase 5
level cascaded multi level inverter. In this set up two H-
bridges are used to construct it. Each H-bridge is given
separate equal dc sources of 2 volts. For triggering the
MOSFETs, gate pulses are fed to the each gate terminals
of the MOSFET from the output of the optocoupler. The
ATMEGA 16 is used to generate the pulses for each
switching device. The ATMEGA 16 programming was
loaded into ATMEGA 16 chip by using software
WINAVR. Fig 9.4 shows the gate pulses given to the
each gate terminal of the MOSFET.

Fig 13 :- Output Voltage Waveform of 1 Phase 5

Levels CSMLI Taken From M.S.O.

Fig 10:- Experimental setup of 1 phase 5 levels

Cascaded Multilevel inverter

Fig 14 :-FFT Analysis of Output Voltage Waveform of

1 Phase 5 Levels CSMLI.
IX. CONCLUSION

The PSO has been proposed to solve the SHE problem

with equal dc sources in cascaded multi level inverter.
The proposed method has optimized the switching
angles as well as voltages of each H-bridge. This
Fig 11:- Triggering Pulses from the ATMEGA 16

Anup Kumar and A. Dasgupta Page No. 35

optimization technique is giving better result than [9] The text book Power Electronics, Circuit,
NRM & Polynomial method. Devices and Applications, 3rd Edition by
Muhammad H. Rashid.
X. REFERENCES
[10] Handbook Of Power Electronics edited by
[1] H.Taghizadeh and M.Tarafdar Hagh Muhammad H. Rashid.
:”Harmonic Elimination of Multi level [11] P. Palanivel S.S. Dash: “Selective harmonic
Inverter Using Particl Swarm Optimization elimination in multilevel inverter”, Published
Technique”, 2008 IEEE. in IET Power Electronics Received on 31st
[2] M.Tarafdar Hagh, H.Taghizadeh and Kaveh March 2010 & Revised on 31st March 2011.
Razi: “Harmonic Minimization in Multi level [12] Manyuan Ye, Pinggang Song, Chaowei
inverter Using Modified species Based Zhang: “Analysis of multilevel inverter”,
Particle Swarm Optimization”, IEEE Dept. of Electrical Engineering, East China
transaction on power electronics, vol 24, no jiaotong University 2008 IEEE.
10, oct 2009.
[13] J. N. Chiasson, L. M. Tolbert, K. J.
[3] H. Taghizadeh and M.Tarafdar Hagh: McKenzie, and Z. Du, “The use of power sums
“Harmonic Elimination Of Cascaded Multi to solve the harmonic elimination equations
level inverter With Non Equal DC Sources for multilevel converters,” EPE J., vol. 15, no.
using Particle Swarm Optimization”, IEEE 1, pp. 19–27, Feb. 2005.
Transactions on Industrial Electronics, vol.
57, no. 11, November 2010. [14] L. M. Tolbert, J. N. Chiasson, K. McKenzie,
and Z. Du, “Elimination of Harmonics in a
[4] A. Mathuramalingam, M.Balaji and multilevel converter with non equal DC
S.Hamavathi: “Selective Harmonic sources,” inProc. IEEE Appl. Power Electron.
Elimination Method For Multi level Inverter”, Conf., Miami, FL, Feb. 9–13, 2003,pp. 589–
Proceedings Of India International 595.
Conference On Power Electronics 2006.
[5] Alan J. Watson, Patric W.Wheeler: “A
Complete Harmonic Elimination Approach to
DC link Voltage Balancing For a Cascaded
Multi level Rectifier”, IEEE Transaction on
Industrial Electronic, Vol 54, No. 6, Dec. 2007
[6] Prashanth Neelugonda, A. Dasgupta, B.Kumar
and J. yadagiri: “Harmonic Minimization in
multi level Inverter by Using Particle Swarm
Optimization”, int joural on Recent Trends in
Engineering & Technology, Vol 5, No 02, Mar
2011
[7] Jagdish Kumar, Biswarup Das and Pramod
Agarwal: “selective Harmonic Elimination
For Multilevel inverters , fifteen national
Power Conference”, IIT Bombay, December
2008
[8] Jagdish Kumar, Biswarup Das and Pramod
Agarwal: “Harmonic Reduction Technique
For a Cascaded Multi level Inverter”,
International journal On Recent Trends in
Engineering, 2009

Anup Kumar and A. Dasgupta Page No. 36

 Modeling, Design and Simulation of Current and
Voltage Linear Controller of a STATCOM for
Reactive Power Compensation
I.
S. K. Sethy* and Dr. J. K. Moharana**

*Faculty member and Research Scholar, **Professor and Head

Department of Electrical Engineering, Dhaneswar Rath Institute of Engineering and
Management Studies, Tangi, Cuttack, Odisha, India.
Email:*sushanta_sethy1975@yahoo.co.in and **jkrushna@gmail.com

ABSTRACT--

The STATCOM (STATic synchronous COMpensator) is a shunt connected voltage source converter using
self-commutating device and can be effectively used for reactive power control. Its principle of operation is
similar to that of a synchronous condenser. This paper describes the modeling of STATCOM along with
design of current controller and voltage controller. The design of controllers for the converters can be
realized in two ways. The first method is a non-linear realization, which results in simple control rules with
faster dynamics. The second method is a linear method, which requires system modeling. The second
approach is adopted and simulated waveforms are presented.
Index Terms-: Controller design, PI Controller, STATCOM,

I. INTRODUCTION the synchronous condenser. In a traditional

synchronous condenser, the field current of the
In recent years power systems have become very
synchronous motor controls the amount of VAR
complex with interconnected long distance
absorbed/injected and hence in a similar way, the
transmission lines. The interconnected grids tend to
firing instant of the 3-phase inverter controls the
become unstable as the heavy loads vary
VAR flow into or out of the STATCOM. Large
dynamically in their magnitude and phase angle and
numbers of capacitor banks or inductor banks are no
hence power factor. Commissioning new
more required. Only a fixed set of capacitor provides
transmission systems are extremely expensive and
the required VAR control, with a rapid control of bus
take considerable amount of time to build up.
voltage and improvement of utility power factor. It
Therefore, in order to meet increasing power
offers several advantages over conventional
demands, utilities must rely on power export/import
thyristorised converters [2] in terms of speed of
arrangements through the existing transmission
response. The penalty paid for this improvement is
systems. Power electronic devices are gaining
in terms of introduction of some harmonics, which
popularity for applications in the field of power
requires separate handling using active filtration
transmission and distribution systems. The reactive
techniques. Moran et al [3] have shown in details
power (VAR) compensation and control have been
how the utilization of Sinusoidal Pulse Width
recognized [1] as an efficient & economic means of
Modulation (SPWM) techniques reduces harmonic
increasing power system transmission capability and
distortion. It has also been shown that an increase of
stability. The FACTS (Flexible AC Transmission
modulation index reduces the size of the link reactor
Systems) devices, such as STATCOM has been
and stress on switches which are significant issues
introduced more recently which employs a VSI with
in practical implementation. The modeling and
a fixed DC link capacitor as a static replacement of
analysis of STATCOM steady state and dynamic

S. K. Sethy and Dr. J. K. Moharana Page No. 37

performance with conventional control method have r
been studied Schauder and Mehta [4] using (capacitive) current Ica to be injected into the system
nonlinear controller. In [6] the dynamic responses at PCC as shown in Fig.2.
and steady state behavior of STATCOM with Space
2.2. Modeling
Vector Pulse Width Modulation (SVPWM) has been
studied and the advantages of introducing SVPWM The modeling of the STATCOM, though well known,
inverter with higher values of modulation index are is reviewed in the lines below, for the sake of
highlighted. convenience. The modeling is carried out with the
following assumptions:
The controllable reactive power allows for a rapid
control of bus voltage and power factor at the system 5) All switches are ideal
or at the load end. To compensate for the distorted 6) The source voltages are balanced
current drawn by the rectifiers from the utility grid, the
STATCOM and its current controller must have the 7) Rs represents the converter losses and the
capability to track source PWM (Pulse Width losses of the coupling inductor
Modulation) converters. The linear control is more
suitable for STATCOM application reported in [7-8].
The present paper suggests the design of a linear
current controller and voltage controller on the basis of
gain and time constant adjustment along with the
parameter of the coupling inductor and storage
capacitor.
The present paper goes on to develop closed loop Fig.1: Schematic diagram of STATCOM
model for investigating transient performance of the
STATCOM by using controller parameter. First, in
Section 2 focuses on state space model of the
STATCOM with the system. Secondly, in Section 3, a
current and voltage controllers are designed. The
simulated responses with the designed controller
parameters are presented. This scheme is both an
extension and a significant improvement of the
scheme suggested by Shauder et al [4] and Sensarma
et al [9].The results obtained have been compared and Fig.2: Phasor diagram for inductive load
appropriate conclusions have been drawn. operation

II. MODELING OF THE STATCOM AND 8) The harmonic contents caused by switching
ANALYSIS action are negligible

2.1. Operating principle The 3-phase stationary abc coordinate vectors

with 1200 apart from each other are converted into
As is well known, the STATCOM is, in principle, a
static (power electronic) replacement of the age-old αβ 2-phase stationary coordinates (which are in
synchronous condenser. Fig.1 shows the schematic quadrature). The α axis is aligned with a axis
diagram of the STATCOM at PCC through coupling and leading β axis and both converted into dq
inductors. The fundamental phasor diagram of the two-phase rotating coordinates. The Park’s abc
STATCOM terminal voltage with the voltage at PCC
to dq transformation matrix is
for an inductive load in operation, neglecting the
harmonic content in the STATCOM terminal voltage,  Cos (ϖ t ) Cos (ϖ t − 2 π / 3 ) Cos (ϖ t + 2π / 3 ) 
is shown in Fig.2. Ideally, increasing the amplitude of 2
 Sin (ϖ t ) Sin (ϖ t − 2π / 3 ) Sin (ϖ t + 2π / 3 )  (1)
r K =
3 
the STATCOM terminal voltage V oa above the  1/ 2 1/ 2 1/ 2 
r The actual proposed circuit is too complex to analyze as a
amplitude of the utility voltage V sa causes leading whole, so that it is partitioned into several basic sub-
circuits, as shown in Fig.1. The 3-phase system voltage

S. K. Sethy and Dr. J. K. Moharana Page No. 38

 vs,abc lagging with the phase angle α dv dc m
to the STATCOM = i cd (9)
dt C
output voltage vo,abc and differential form of the
The complete mathematical model of the STATCOM
STATCOM currents are defined in (2) and (3). in dq frame is obtained as given in (10)
 
 Sin ( ω t − α )  (2)
 v sa   2π 
 Rs 
− −w 0 
2
v =  v sb  = V s  Sin ( ω t − α − )
   icq   Ls   icq  − Sinα 
s , abc 3 3 
 v sc  
 Sin ( ω t − α + 2 π  d  
icd = w
R
− s −
m    Vs
icd +  Cosα  (10)
 3 
dt    Ls Ls    Ls  
vdc   m  vdc   0 
Ls
d
dt
(ic ,abc ) = − R s ic,abc + v s ,abc − v o,.abc (3)  0
 C
0 


C. Steady State and transient Analysis

where , Vs ,ϖ, Rs and Ls have their usual
The detailed steady state and transient responses with
connotations. The above voltages and currents are
the Table.1 are given in Fig.3-6 and responses suggest
transformed into dq frame the static and dynamic conditions of the STATCOM.
It can be seen that the transient responses take about
d
Ls (icq ) = − R s icq − wL s icd + v sq − v oq (4a) one and half power cycle to reach at their steady state
dt values.
d Table.1
Ls (icd ) = wL s icq − R s icd + v sd − v od (4b)
dt Sl Parameters Symb Values
The switching function S of the STATCOM can be ol
defined as follows
1 Frequency f 50 Hz
  2 Angular Frequency ω 314 rad/sec
 Sin ( wt )  3 RMS line-to-line Voltage 230V
Sa  Vs
2  
S = Sb  = m  Sin ( wt −
2π
) (5)
S  4 Coupling Resistance Rs 1.0 Ω
 c
3
 3
2π

 Sin ( wt + ) 5 Coupling Inductance Ls 5.0mH
 3 
6 DClink capacitor C 500 µ F
The modulation index, being constant for a
7 Modulation Index M 0.979
programmed PWM, is given by,
8 Phase angle α m 50
v o , peak 9 Load Resistance
MI = =
2
m (6) RL 52 Ω
V dc 3 10 Load Inductance LL 126mH
The STATCOM output voltages in dq 11 Load Power factor φ 0.79
transformation are

v o , qdo = m 0 1 0 [ ]T v dc (7)

The dc side current in the capacitor in dq

transformation
T
i = m [0 1 0 ] i i i  (8)
dc  cq cd co 

The voltage and current related in the dc side is given

by Fig.3: Steady state responses of Icq , I cd and V dc

S. K. Sethy and Dr. J. K. Moharana Page No. 39

 [u ] = 
v oq   i cq 
 , [y ] = i  (12)
 v od   cd 
The block diagram of the STATCOM in d-q
transformation as per (11) is shown in Fig.7.The
instantaneous voltage of the system and the
STATCOM are independent, but the active and the
reactive currents are coupled with each other through
the reactance of the coupled inductor. So it is very
Fig.4: Steady state responses of Pc and Qc essential to decouple the active and reactive current
from each other and design the controller for tracking
the required value.

Fig.5: Transient responses of icq in capacitive and Fig.7: Equivalent Diagram on a.c.side of STATCOM
inductive 3.1. DESIGN OF CURRENT CONTROLLER.
The current controller design for the above system can
be done using the strategy [8-9] attempts to decouple
the d and q axes equations, so that the MIMO system
reduces to two independent Single Input Single Output
(SISO) system. Hence, the control inputs vod and

v oq are configured as

*
Fig.6: Transient responses of vdc in capacitive and v oq = − v oq − wL s i cq + v sq
(13)
inductive *
v od = − v od + wL s icd + v sd
III. DESIGN OF CONTROLLERS FOR
STATCOM The equation (14) can be obtained by replacing (11)
by (13). Hence each row of (14) is independent of
With the assumption of the system voltage and each other and thus defines an independent SISO
STATCOM output voltage are in phase and hence the system. Conventional frequency-domain design
equation (10) can be modified as given in equation methods can now be directly applied for current
(11) controller. Taking the Laplace transformation of both
sides of (14) and rearranging terms are given by (15)
 Rs 
− L −ω  and their decoupled SISO system is shown in Fig.8.
d icq 
& i  1   v   v   (11)
=  s   cq  +  sq  −  oq 
dt i&cd   Rs  icd  L
s v sd  vod 
 Rs 
ω − − 0   * 
 
Ls  i&cq   Ls icq  1  voq  (14)
i&  =    +
 cd   0 − Rs  icd  Ls v * 
So the equation (11) is a Multiple Input and Multiple  L s   od 
Output (MIMO) system and its input and output are
given in equation (12)

S. K. Sethy and Dr. J. K. Moharana Page No. 40

 I q (s) 1 I d (s) 1 Thus the system behaves like a first order with an
Gq (s) = = , Gd (s) = = (15) apparent time constant as
* *
Voq(s) Rs + sLs Vod(s) Rs + sLs
Ls
τi = (21)
K
The gain of K can be adjusted such a way that if it is
increased too high then the system behaves as second
order, otherwise responses very slow. Hence the
numerical values for K p and Ki are decided from the

circuit parameters Ls and Rs from the required value

Fig.8: Current control of inverter of equivalent of K. So the parameters of PI controller are defined as
decoupled SISO systems
For similar dynamic behaviour of the d and q- axis KR s
K p = K, Ki = (22)
currents, both the d and q- axis controllers are Ls
identical and its transfer function is given in (16)
Ls
Icq(s) Icd (s) 1 where, τi = which is taken as 0.3mseconds and
G (s) = = = (16) Kp
i * *
Voq(s) Vod(s) Rs + sLs with the parameters given in Table-1, value of

The transfer function of a PI controller is Kpi =16.9 and Kii = 3.3×103 are calculated.

  K These parameters are used in d and q- axis current

G pi ( s ) = K  1 + 
1 i (17)
= Kp + controller. The structure of the effective closed loop
 sτ  s
 i  system is shown in Fig.9 and is replicated in both the
K
d and q- axis current controllers. Bode plot of the
With K p = K , K = . The transfer function in system with controller is shown in Fig.10, which is a
i
τ
i first order system.
open loop of PI controller associated with the transfer
function on the a.c. system is
 1 
[G pi ( s ).G i ( s ) ] = K 1 + s τ1 


Rs
Ls


(18)
 i  1 + s 
 Rs 

Ls
While taking τ = and on simplification reduces to Fig.9: Effective closed loop current control system
i Rs
The relation between dc voltage v and dc current
[G pi (s).Gi (s)]= sLK (19)
idc is
dc
s
The closed loop transfer function is 1
v dc = ∫ i dc dt (23)
C

1 The transfer function can be written as

T = (20)
L
1+ s s V
G v ( s ) = dc =
1
(24)
K I dc sC

S. K. Sethy and Dr. J. K. Moharana Page No. 41

 C
3.2. DESIGN OF VOLTAGE CONTROLLER where , τ v = and taking τv =1msecond
K * K dc
and with the parameters of Table. I, the value of
Kdc =1.08

Fig.10: Bode plot of the system with PI controller

Neglecting the power loss in the source resistance and Fig.12: DC link voltage control loop
power losses in the switches, balancing the power on
Then Proportional Integral controller is considering for
both sides,
the voltage control. Hence, the transfer function of PI
controller in (18) is associated with the transfer
vsdicd = vdcidc (25) function on dc side is
From the above equation, we have   1 
idc vsd Vs 230 [Gv ( s ).G pi ( s )]ol = K 1 + sτ1  
 sC 
(29)
= = = = 0.46 (26)  v 
icd vdc Vdc 500
The DC bus voltage is maintained at 400 volts. After taking τ v = C and on simplification
V dc  1 + sτ 
With as the reference, the voltage control loop is [G v ( s ).G pi ( s ) ] = K  2 2 v  (30)
 s τv
ol
shown in Fig.12 and it consists of inner d- axis 
current control loop. The active power is supplied by The transfer function in closed loop
the d-axis current which is nothing but the ripple
current of the capacitor. To make the steady state error  
 
of the voltage loop zero Proportional control is 1 + sτ v
adopted here and it produces the reference d-axis
[G (s).G
v pi ]
( s ) cl =
 s 2τ v2


(31)

current for the control of the d-axis current. The  1 + sτ v + 

 K 
design of voltage controller is as follows:
So the system behaves like a second order system. As
The open loop transfer function of DC bus voltage
controller is τ v2
τ v >> and magnitude plot in Fig.10 shows the
G op =
K * K dc
(27) K
sC initial slop at break point is approximately –
20db/decade and hence it reduces to first order
The closed loop transfer function with unity feed back system. The value of K can be determined form root
gain is locus with approximate settling time as
1 K
Gcl = (28) K pv = K = 0 .15 , K vi = = 200 (32)
sC C
1+
K * K dc

S. K. Sethy and Dr. J. K. Moharana Page No. 42

 As per the value of icq obtained in steady state and
transient, the authors are interested to control the
reactive current as per the load. The simulations of the
reference reactive current of 15A and reference
DC_link voltage of 500V are shown in Fig.13 and 14.
These Figs show that the output current and voltage
are properly follow the reference values. Fig.15: Implementing scheme

Fig.13: Current control with reference Fig.16: System voltage and system current before
compensation

Fig.14: Voltage Control with reference

IV. SIMULATIONS USING ABOVE Fig.17: System voltage and system current after
CONTROLLER compensation

The control scheme for controlling DC link voltage as

well as d and q axes current of STATCOM
simultaneously as shown in Fig.15 is implemented with
MATLAB SIMULINK with the parameters given in
Table. I. The PI controller is applied at DC link voltage
and reference current for controlling q axis current of
STATCOM is generated from the q axis of the linear
load currents. The generated reference output voltages of Fig.18: System voltage and STATCOM current using
d and q -axes are transformed to α and β axis and
then abc axes. All the relevant outputs are shown in
Fig.16 to Fig.19. The significance of the Fig.17 is the
improvement of power factor to unity from 0.79 as per
Fig.16 after one power cycle with a under shoot of 30A
of the system current before coming to unity power Fig.19: DC link voltage
factor.Fig.18 shows the dynamics of STATCOM current
with the same under shoot of the system current. Fig.19
shows that the maximum overshoot of DC link voltage is
to 640 volts and remains constant at 400 volts after one V CONCLUSION
power cycle. The complete analysis and models of reactive current
and voltage controllers of the STATCOM application
are presented. The controllers are designed on the
basis of gain adjustment and time constant. The
simulated figures are given which have been
controlled the desired values. The settling time of the

S. K. Sethy and Dr. J. K. Moharana Page No. 43

system by using the PI controller is faster than other
controllers. In this paper, the proposed scheme is
easier to implement compared to [4,9].However, in
practice the issue of the charging the DC link voltage
to the required value is quite significant. In most cases,
there is a separate charging circuit for the DC link
voltage. The authors are working on a plausible
method of eliminating such an extra starting
arrangement, so that the controller may become
operational while the DC link voltage is at a low
value.
VI. REFERENCE
[1] A.T. Johns, A.Ter-Gazarian and
D.F.Wame,”Flexible ac transmission systems
(FACTS)”, IEE Power and Energy Series,
London, U.K.
[2] R.M.Mathur and R.K. Varma, “Thyristors-
based FACTS Controllers for Electrical
Transmission Systems, IEEE Press”, Wiley-
Interscience Publication.
[3] L.T. Moran, P.D.Ziogas and G.Joos, “Analysis
and Design of a Three-Phase Synchronous
Solid- State Var Compansator”, IEEE Trans.
Industry Application, Vol. 25, No. 4, 1989, pp.
598-608.
[4] C.Shauder and H.Mehta, “Vector analysis and
control of advanced static VAR compensators”,
IEE Proc, 140, No. 4, July 1993.
[5] A. Draou, M. Benghanem and A. Tahiri,
“Multilevel Converter and VAR
Compensation”, Power Electronics Handbook”,
pp.599-611, Academic Press, 2001.
S. K. Sethy and Dr. J. K. Moharana
[6] M.Sengupta,J.K Moharana and A.Sengupta,“
Study on an Advanced Static VAR
Compensator switched from a Space Vector
PWM inverter –Analysis, simulation and
comparison with the conventional sinusoidal
PWM, NPEC 2003, IIT Bombay,16-17 Oct 03
pp 72-78.
[7] D.M.Brod and D.W.Novotny, “Current control
of VSI- PWMinverter”IEEETrans.Industrial
Appl, Vol.IA-21,pp.562-570, July/ Aug. 1985.
[8] S.Buso, L.Malesani and P.Mattavelli,
“Comparison of Current Control Techniques for
Active Filter Application”IEEETrans.Industrial
Electronics, Vol.45, No.5, pp.722-729, October
1998.
[9] P.S.sensarma, K.R.Padiyar and
V.Ramnarayanan, “Analysis and Performance
Evaluation of a Distribution STATCOM for
Compensating Voltage Fluctuations”, IEEE
Transaction on Power Delivery, Vol.16, No.2,
pp.259- 264, April 2001
[10] A.M. Kulkarni and K.R. Padiyara, “Design of
Reactive Current and Voltage Controllers of
Static Condenser”, Power and Energy
System,Vol.19,No.6,pp.397-410,1997
Page No. 44
 A Comparative study of Back-Propagation
Propagation and RBF
ANN Based Bearing Fault Detection of Induction Motor
P. S.Puhan1 , M.M.Patra2 Dr G.Panda3
1
Electrical Engineering Department, B.I.E&T, Bhadrak ee-mail:
mail: puhan_samal@rediffmail.com
2
Electrical Engineeringg Department, B.I.E&T, Bhadrak ee-mail:
mail: mana_manjari@yahoo.co.in
3
Electrical Engineering Department, I.G.I.T. , Saranga ee-mail:
mail: p_gayadhar@yahoo.com

ABSTRACT
The paper investigates the effectiveness of different Artificial Neural Network structure like Feed
Fee Forward Back
Propagation (FFBP) and Radial Basis Function (RBF) as a classifier for bearing faults detection in Induction Motor.
The steady state motor current with Park’s Transformation has been used for discrimination of inner race and outer
race bearing
ng defects. The RBF neural networks shows very encouraging results for multi-class
multi classification
problems is hoped to set up a base for incipient fault detection of induction motor which will be simple, fast and
overcome the limitations of traditional tec
techniques.
Keywords— Induction Motor, Bearing Fault, Park’s Transform, Back Propagation (FFBP), Radial Basis Function
(RBF).

INTRODUCTION
Induction motors are being used extensively for current analysis [10] [13]. Application of artificial
different industrial applications ranging from intensive neural networks has also been widely used in fault
care unit, defense
ense applications to the applications in detection of induction motors [7] [8] [12] [14].
power stations. Since, various operational stresses lead However in most of the applications multi layer
to inherent possibilities of incipient faults, which perceptron model
odel has been used.
cause gradual deterioration of motor leading to failure. In this article an attempt is made to study the
Therefore the issue of preventive and condition based applicability of a general purpose feed forward and
maintenance, on-lineline monitoring, fault detection, Radial Basis function (RBF) neural network for
diagnosis and prognosis are of increasing importance bearing faults detection of induction motors by
[3][4]. As the induction motor is highly symmetrical, assessing the patterns of motor line currents.. Among
the presence of any kind of fault leads to a various motor faults, proposed investigation has been
corresponding change in the interaction of flux restricted to bearing fault only, since motor reliability
between stator and rotor, resulting in changes to the studies shows that bearing faults account for over 44%
stator currents and machine vibration. Large size of the faults occurring in an induction motor. The
induction motors are equipped with mechanical present work has been n restricted on inner race and
sensors, which are primarily vibration sensors such as outer race bearing faults only.
proximity probes which are delicate and expensive.
Since it is not economically or physically feasible to
provide the same for smaller induction motors stator
current harmonics measurement is appearing as an
alternative to the vibration measurement methods.
This technique is known as Motor Current Signatur
Signature
Analysis (MCSA) and it can perform the same
detection of any incipient fault as the vibration
monitoring technique without assessing the rotating
parts. Therefore current based fault detection is
becoming an attractive area of research. Many Fig 1: Occurrence of induction motor faults
research papers
ers have been published based on stator

P. S. Puhan, M.M. Patra Dr G. Panda Page No. 45


Fig 2: (a) Outer race (b) Inner race faulted bearings
A FEED-FORWARD BACK-PROPAGATION (FFBP)
A feed-forward back-propagation neural network has
three components: an input layer, one or more hidden
layers, and an output layer. Each layer consists of one
or more neurons called nodes. In the calculation
process of problem solving, all input nodes are
collected at each hidden node after being multiplied by
weights. Later, a bias is attached to this sum,
transformed through a nonlinearity function, and
transferred to the next layer. There are several
functions such as hyperbolic tangent, sigmoid and
linear functions that can be used as transfer function.
The same procedure can be followed in the next layer
to provide the network output results consequently. As
the forward processing arrives at the output layer, the
overall error between the network output and the
actual observation is calculated. The error at the output
layer propagates backward to the input layer through
the hidden layer in the network to obtain the final
desired outputs. During the forward pass all the
weights of the networks are fixed. During the
backward pass, on the other hand, all the weights are
adjusted in accordance with an error-correction rule
[12] [14].
RADIAL BASIS FUNCTION NETWORKS (RBF)
Radial basis function RBF networks were introduced
into the neural networks literature by Broom head and
Lowe [5]. The RBF consists of three layers, namely,
input, hidden and output layer. It is a feed-forward
network and has only one hidden layer. The input
layer is made up of source nodes or sensory units that
connect the network to its environment. The second
layer applies a nonlinear transformation from the input
space to the hidden space. In most applications the
hidden space is of high dimensionality. The output
layer is linear, supplying the response of the network
to the activation pattern signal applied to the input
layer [24]. The input to each RBF neuron is treated as
a measure of the difference between data and a
P. S. Puhan, M.M. Patra Dr G. Panda
“centre”, which is a parameter of its transfer function.
The transfer function of the neuron indicates the
influence of data points at the centre. Generally this
function is Gaussian and its centers can be chosen
either randomly from the training data or they are
iteratively trained or derived using techniques like K-
means, Max-Min algorithms, Kohonen self organizing
maps [20].
The advantages of using RBF neural networks are: (i)
the RBF does not involve iterative training and hence
much of the training time is saved, (ii) the best
possible network architecture is determined according
to the input data by a well-proposed training
algorithm, (iii) it does not require the many trial tests
to determine the appropriate network architecture, (iv)
this feature is user friendly for general industrial
applications, and lastly (v) the outputs of the neural
network are able to not only perform fault detection,
but also indicate the extent of the fault (diagnosis)
[17][18].
PROPOSED METHOD
The schematic representation of the work is shown in
fig. 3. The scheme consists of four major parts, namely
(i) simulation of different induction motor faults, (ii)
data acquisition, (iii) feature extraction and (iv) Post
Processing and Diagnosis with FFBP/RBF ANN. For
identifying the faults three phase motor line currents
(Iline) at steady state is monitored and diagnosed.
Photograph of the experimental setup is presented in
fig. 4.
Machinery Fault Simulator,
Spectra Quest, USA
2-Pole, 60Hz, 0.5 HP, Load
3450RPM, 3 Ph
Induction Motor.
3- phase
Line Currents A/D Converter and
Data Acquisition
System (DAQ)
Fault diagnosis Feature Extraction
Using using Park’s
FFBP/RBF ANN Transform
Fig. 3: Schematic diagram of the work
Page No. 46

Fig 4: Photograph of the Experimental Set
Set-up
Simulation of faults
Machinery Fault Simulator (MFS), a tool for
simulating various types of induction motor faults
initially fitted with a healthy motor an
and a motor with
inner race faulted bearings of same specification have
been used for the fault simulation.
Data acquisition
The generated data corresponding to a healthy and
faulty motor are collected using three clip clip-on Hall
Effect current transducers, recorded
corded and stored using a
computer with four-channel
channel Data Acquisition system
(DAQ). The collection is done for both the healthy
motor and the motors with faulted bearings under the
same running conditions. Photograph a clip clip-on Hall
Effect current transducerr is presented in fig. 5.
Fig 5: Photograph of a clip-on
on Hall Effect current
transducer
P. S. Puhan, M.M. Patra Dr G. Panda
Signal processing
Park’s transform has been used as the signal
processing technique. It is a mathematical tool that
allows simplifying induction motor model equations.
equati
It is based in an arithmetical transformation that is able
to convert a symmetrical system with three variables
such as a, b, and c to another two orthogonal variables
such as d and q. If Asymmetric exits and are
considered a third variable (0) appearsappear which is
basically required to the common mode [13][17[19].
As a function of mains phase-variables
variables (Ia, Ib, Ic) the
motor current Park's Vector components (Id, Iq) are
represented by the equations in 1 &2.
2 1 1
ID = IA − IB − IC
3 6 6 …………...…….. (1)
1 1
IQ = IB − IC
2 2 .…………………
.…………… (2)
It has been shown in fig 7 that Id - Iq pattern differs
from each other in healthy and under faulted
conditions.
Processing and Diagnosis of faults
Post Processing and Diagnosis of faults has been done
by recognizing the pattern of Id - Iq . Artificial Neural
Networks (ANNs) has been used as a pattern
classifier.
RESULT AND ANALYSIS
The motor line currents of the healthy and the faulty
motors at steady state were collected through DAQ at
a sampling frequency of 7680. The supply frequency
was set
et to 50 Hz. The experiments were carried out
under two different loading conditions: No load and
with load. Park’s transform was applied on the
recorded data and Id - Iq thus obtained was treated as
attributes to train the ANN.
The FFBP/ RBF ANN is then used as a classifier for
the fault detection process. The Network Structure was
implemented in the MATLAB 6.5 environment. The
network architecture has been shown in figs. 8 and 9.
Both two class as well as multi-class
class classification has
been attempted. Accuracy
ccuracy levels of both the
architectures are furnished in the Table: I – VI.
Page No. 47
 Fig 9: ANN architecture of RBF used in classification.

TABLE I. CORRECT LEVEL OF CLASSIFICATION

USING FFBPN
Training Success (%) Testing Success (%)
No Load With No Load With
Load Load
Healthy 100% 96.8% 84% 100%
Inner
100% 99.6% 100% 94%
Fault
Fig 6: Stator Current Signal for Induction Motor Total 100% 98.2% 92% 97%
Number of Epochs=51, Elapsed time =6.11 sec.

TABLE II. CORRECT LEVEL OF TWO CLASS

CLASSIFICATION

USING RBF
Training Success (%) Testing Success (%)
No Load With No Load With
Load Load
Healthy 100% 100% 100% 100%
Inner
100% 100% 100% 100%
Fault
Fig 7: Id - Iq pattern of healthy
lthy and faulty motor Total 100% 100% 100% 100%
Elapsed time =0.281 sec.

TABLE III. CORRECT LEVEL OF CLASSIFICATION

USING FFBPN
Training Success (%) Testing Success (%)
No Load With No Load With
Load Load
Healthy 100% 100% 100% 100%
Outer
100% 100% 100% 100%
Fault
Total 100% 100% 100% 100%
Fig 8: ANN architecture of FFBP used in Number of Epochs=8, Elapsed time =1.43 sec.
classification.

P. S. Puhan, M.M. Patra Dr G. Panda Page No. 48

 TABLE IV. CORRECT LEVEL OF CLASSIFICATION which will be simple, fast and overcome the
USING RBF limitations of traditional techniques.
Training Success (%) Testing Success (%) ACKNOWLEDGMENT
No Load With No Load With
Load Load The authors are thankful to AICTE and TEQIP
(BESU, Shibpur unit), Govt. of India for their support
Healthy 100% 100% 100% 100%
toward the project and Dr P.Chhotopadhya (Asst Prof
Outer EE), BESUS,Shibpur. West Bengal,India.
100 % 100% 100% 100%
Fault
Total 100% 100% 100% 100% REFERENCES
Elapsed time =0.234 sec.
[1] J.Riddle, Ball Bearing Maintenance. Norman,ok:
University .of Oklahoma Press,1955
TABLE V. CORRECT LEVEL OF CLASSIFICATION
USING FFBPN [2] P.J. Tanner, B.G. Gay don, and D.M. Word.
“Monitoring generators and large motors”, IEE
Training Success (%) Testing Success (%) Proc., 133, Part B (3): 181–189, 1986.
No Load With No Load With
Load Load [3] J. Penman, M.N. Dye, A.J. Taut, and W.E. Bryan.
Healthy 100% 96.4% 89.2% 100% “Condition monitoring of electrical devices” IEE
Proc., 133, Part B(3): 164–180, 1986
Inner
100% 100% 100% 97.2%
Fault [4] P. J. Tavner and J. Penman, “Condition
Outer Monitoring of Electrical Machines”, Research
100% 100% 100% 100%
Fault Studies Press, Ltd., (1987).
Total 100% 98.8% 96.4% 99.06%
[5] D.Broomhead and D.lowe, “Multile Variable
Number of Epochs=17, Elapsed time =5.812 sec. Functional application and adaptive network”s,
complex system, vol.2, no.3, pp 321-355, 1988.
TABLE VI. CORRECT LEVEL OF CLASSIFICATION
USING RBF [6] A. M Knight et al, “Mechanical fault detection in
a medium sized induction motor using stator
Training Success (%) Testing Success (%) current monitoring”, IEEE Trans Energy
No Load With No Load With conversaion,vol-20, no.3, pp.321-355, 1988
Load Load
[7] Chow, M-Y. And S. O. Yee, “Using neural
Healthy 100% 100% 100% 100% networks to detect incipient faults in induction
Inner motors”, Journals of neural network computing 2,
100% 100% 100% 100%
Fault 1991, pp.27-32.
Outer
100 100% 100% 100% [8] I.E. Alguindigue, A.L. Buczak and R.E. Uhrig,
Fault
Total 100% 100% 100% 100% “Monitoring of rolling element bearings using
artificial neural networks”, IEEE Transactions on
Elapsed time =0.718 sec.
Industrial Electronics, Vol.40, No.2, April 1993,
pp. 209-217.
CONCLUSION
This paper has introduced a specific application of the [9] P. Yas, Parameter Estimation, Condition
Park’s transform for inner race and outer bearing fault Monitoring and Diagnosis of Electrical
detection of induction motors along with FFBPN, RBF Machines, Clarenon Press Oxford,1993.
Neural Network as a classifier. It has been observed [10] R. R. Schoen, T. G. Habetler, and F. Kamran,
that RBF network provides 100% accuracy in multi- “Motor bearing damage detection using stator
class classification with a very less training time. The current monitoring”, IEEE trans. Ind. Applicant.,
results are very much satisfactory and promising for an vol. 31, pp. 1280-1286, Nov./Dec. 1995
industrial application and is hoped is hoped to set up a
base for incipient fault detection of induction motor

P. S. Puhan, M.M. Patra Dr G. Panda Page No. 49

 Modelling and Simulation of Induction Motor with
Different Reference Frames
G.T.Chandra Sekhar1, K.Alfoni Jose2, L.V.Suresh Kumar3, B. Manmadha Kumar4
1
Assistant Professor & Sri Sivani College of Engineering, Email: gtchsekhar@gmail.com
2
Assistant Professor & Sri Sivani College of Engineering, Email: alfoni.ee08@gmail.com
3
4
Assistant Professor & GMR Institute of Technoogy, Email: jagath.mtech2008@gmail.com
Sr. Assistant Professor & Aditya Institute of Technology and Management, Email: boddmann@yahoo.co.in

ABSTRACT
In recent years the control of high-performance induction motor drives for general industry applications and
production automation has received widespread research interests. Induction motor has varied modes of operation
both under steady and dynamic states. Hence modelling of Induction motor is necessary. Here we have done 3 types
of modelling. They are Synchronous frame of reference, Stator frame of reference, Rotor frame of reference. In all
the three reference frames 3-phase parameters are converted to 2-phase parameters. In this project, MATLAB is
used for simulation. It is been observed results like 2-phase stator currents, rotor currents, 3-phase stator currents,
electric torque, mechanical speed, stator fluxes in all three reference frames.
Key Words- Induction motor, Parks’ transformation, Rotor Frame of Reference, Stationary Frame of Reference,
Stator Frame of Reference
I. INTRODUCTION asynchronous motor whose operation depends on three
electromagnetic phenomena:
In recent years the control of high-performance
induction motor drives for general industry
1. Motor Action - When an iron rod (or other magnetic
applications and production automation has received
material) is suspended in a magnetic field so that it is
widespread research interests. Induction machine
free to rotate, it will align itself with the field. If the
modelling has continuously attracted the attention of
magnetic field is moving or rotating, the iron rod will
researchers not only because such machines are made
move with the moving field so as to maintain
and used in largest numbers but also due to their
alignment.
varied modes of operation both under steady and
dynamic states. Induction motor is simply an electric 2. Rotating Field - A rotating magnetic field can be
transformer whose magnetic circuit is separated by an created from fixed stator poles by driving each pole-
air gap into two relatively movable portions, one pair from a different phase of the alternating current
carrying the primary and the other the secondary supply.
winding. Alternating current supplied to the primary
winding from an electric power system induces an 3. Transformer Action - The current in the rotor
opposing current in the secondary winding, when the windings is induced from the current in the stator
latter is short-circuited or closed through external windings, avoiding the need for a direct connection
impedance. Relative motion between the primary and from the power source to the rotating windings.
secondary structure is produced by the electromagnetic
A. Rotating Fields
forces corresponding to the power thus transferred
Rotating magnetic fields are created by poly phase
across the air gap by induction.
excitation of the stator windings. In the example below
of a 3 phase motor, as the current applied to the
II. VIEW OF INDUCTION MACHINE
winding of pole pair A (phase 1) passes its peak and
begins to fall, the flux associated with the winding
One third of the world's electricity consumption is
also begins to weaken, but at the same time the current
used for running induction motors driving pumps,
in the winding of the next pole pair B (phase 2) and its
fans, compressors, elevators and machinery of various
associated flux is rising. Simultaneously the current
types. The AC induction motor is a common form of
through the winding of the previous pole pair C (phase

G. T. Chandra Sekhar, K. Alfoni Jose, L.V.Suresh Kumar, B. Manmadha Kumar Page No. 50
3) and its associated flux will be negative and rising rc are the rotor phase a, b, c axes respectively.
(towards positive). The net effect is that a magnetic
flux wave is set up as the flux created by the stator
poles rotates from one pole to the next, about the axis
of the machine, at the frequency of the applied
voltage. The magnitude of the rotating flux wave is
proportional applied MMF.

B. Transformer Action
The stator carries the motor primary windings and
is connected to the power source. There are normally
no external connections to the rotor which carries the
secondary windings. Instead the rotor windings are Figure 1 :Three phase model of a induction Machine.

vw Is the angle by which the axis of phase a of rotor

shorted. When a current flows in the stator windings a
current is induced in the shorted secondary windings
by transformer action. The magnitude of the rotor winding leads the axis of phase A of stator winding

vw - Electrical radian
current will be proportional to the flux density B in the in the direction of rotation.

vx  y/2 vx
air gap (and the relative motion, called the slip, of the

Where vw and P are mechanical radians and the no.

rotor with respect to the rotating field). Torque is
produced by the reaction between the induced rotor

vw = zw t
currents and the air-gap flux created by the stator of poles respectively
currents.
Where ws is the synchronous speed of the revolving
C. Torque Generation (Motor Action) stator field (rad/sec).
The stator voltage equations are given by (assuming

L~2{
Y -connection for the stator)
2{  |2 Y2{ 0 } €
When the motor is first switched on and the rotor

L
is at rest, a current is induced in the rotor windings
L~2
2  |2 Y2 0 } €
(conductors) by transformer action. Another way of

L
seeing this is that the relative motion of the rotating

2‚  |2 Y2‚ 0 ƒ ˆ
$„…†
flux passing over the slower moving (initially static)
$‡
rotor windings causes a current to flow in the windings (1)
by generator action. Similarly, the rotor voltage equations are given

L~w
Once current is flowing in the rotor windings, the by(assuming Y –connected rotor)
w  |w Yw 0 } €
motor action due to the Lorentz force on the
L
L~w‰
conductors comes into effect. The reaction between

w‰  |w Yw‰ 0 } €
the current flowing in the rotor conductors and the
L
magnetic flux in the air gap causes the rotor to rotate
wŠ  |w YwŠ 0 ƒ ˆ
in the same direction as the rotating flux as if it was $„j‹
$‡
being dragged along by the flux wave. (2)
Similar to the DC machine, the torque in an
induction motor is given as below In the matrix notation, the above equations
(both for stator and rotor) are represented as

L~2{‚
2{‚  |2 Œ2{‚ 0
T = k BI

L
w‰Š  |w Œw‰Š 0
$„jl‹
Where T, k, B, I denotes torque, constant depending on
$‡
the number of stator turns, flux density, induced rotor (3)
current.
III. TRANSFORMATION OF 3-PHASE
D. Three-Phase model of Induction Motor VARIABLES INTO 2-PHASE
A.Parks’ transformation
A 3 phase induction machine is modelled as A dynamic model for the three-phase induction
shown in Figure 1.3. sA, sB, sC are the stator machine can be derived from the two-phase machine if
phase A,B ,C axes respectively. Similarly ra, rb,

G. T. Chandra Sekhar, K. Alfoni Jose, L.V.Suresh Kumar, B. Manmadha Kumar Page No. 51
 3œA
« ®
3  1
the equivalence between three and two phases is
ª œ ­
,¯9,¯9›  ª 2 $ ­ ° 1 ±
established. The equivalence is based on the equality of

ª 3 ­ 1
the mmf produced in the two-phase and three-phase

© œ
2 –¬
windings and equal current magnitudes. Assuming that
1/√2 1/√2 1/√2
each of the three-phase windings has Ns turns per phase
,¯9  q2/3 Dcos v
cos v M 2^/3
cos v 0 2^/3
F
and equal current magnitudes, the two-phase winding
sin v
sinv M 2^/3
sin v 0 2^/3

(4)
will have 1.5Ns turns per phase for mmf equality. The
three phase equations can be converted into two phase Where θ=ωt
equation using double axis frame theory. Basically We use the following power- invariant
there are three reference frames.
YA 1/√2 1/√2 1/√2 Y{
transformation

DY$ F  q2/3 Dcos v

cos v M 2^/3
cos v 0 2^/3
F °Y ±
Y– sin v
sinv M 2^/3
sin v 0 2^/3
Y‚
1. Stator reference frame

ŒA$–  ¯ Œ{‚
2. Rotor reference frame
(5)

C³  CA zw

3. Synchronously rotating reference frame. B. Typical Characteristics

CA is the rated load torque at rated speed

zw  jk
´
´jkµ
zw is the rotor base speed in mechanical rad/sec

Note:

|
1. The same notations are used to represent referred
values. Similarly, represents rotor
resistance/phase referred to stator. These
parameters are obtained from open circuit and
Fig. 2 Phasor representation of reference frame blocked rotor test.

f 0 jfo  ‘f’ 0 jfo’ “ 0 ‘f” 0 j fo” “ ¶A obtained in the open- circuit test is usually
2. The iron losses and friction and windage losses,

0 ‘fp 0 j fop “
•– 0 —•$  • ˜ 0 •‰ ˜ EAs™š
0 •Š ˜ ›EA›™š

neglected one way to account this is by using

E™š
C·¸22 .
appropriate value of loss component of torque,

2^ We usually assume ¹·2 = ¹·w (referred values)

From the above equations
•–  œ– • sin v 0 •‰ sin }v M €
3.
3
2^
0 •Š sin }v 0 €ž
3
C. Arbitrary Frame of Reference:
T§ T§
Arbitrary frames of reference are very much like
Ÿ   ¡  Ÿ¢ £¤N ¥ 0 Ÿ¦ £¤N }¥ M € 0 Ÿ¨ £¤N }¥ 0 €ž
U U
observer platforms. For the purpose of control, it is
desirable to have the system variables as dc quantities,
In addition to ‘d’ and ‘q’ quantities a zero-sequence
although the actual variables are sinusoidal. Suppose
•A  œA ,• 0 •‰ 0 •Š 9
quantity is considered as
reference frame revolving at the same angular speed as
that of the sinusoidal variable, then the differential
œA œA œA
Thus in matrix form,
« 2^ ®
speed between them is reduced to zero, resulting in the
•A 2^ •
ªœ$ cos v œ$ cos }v M € œ$ cos }v 0 €­
D•$ F  D•‰ F
sinusoidal being perceived as a dc signal from the
ª 3 3 ­
•– ª 2^ 2^ ­ •Š
reference frames. Then, by moving to that plane, it

© œ– sin v œ– sin }v M
3
€ œ– sin }v 0 € ¬
3
becomes easier to develop a small-signal equation out
of a nonlinear equation, as the quiescent or operating
•A$–  ,¯9 •‰Š
point is described only by dc values; this then leads to
the liberalized system around an operating point.

,¯9›  ,¯9
To have orthogonal transformation matrix, Rotating reference frames, the system inductance
matrix becomes independent of rotor position, thus
We know that, leading to the simplification and compactness of the
system equations. Instead of deriving the

G. T. Chandra Sekhar, K. Alfoni Jose, L.V.Suresh Kumar, B. Manmadha Kumar Page No. 52
   
« √ √ √ ®
 ª ­
transformations for each and every particular reference
¯º  » ªcoszA 
cos ƒzA  M  ˆ cos ƒzA  0 ˆ­ 7)
frame, it is advantageous to derive the general  

ª  ­


© sinzA 
sin ƒzA  M  ˆ sin ƒzA  0 ˆ ¬

transformation for an arbitrary rotating frame of
reference. The speed of the arbitrary frame of 

 ω
$Ө
reference is given by, This is also known as Kron’s reference frame

$‡
(6) Voltage Equations:
The voltage equations of 3-phase is the matrix
IV. MODELLING OF INDUCTION MOTOR IN multiplication of inverse of B and 2-phase voltages as

¼2{‚ ¼2A$–
G¼ K  ½ › G¼ K
3 DIFFERENT REFERENCE FRAMES shown below,

w‰Š wA$–
¯ 0
(8)

Where B denotes  ž
The Commonly Used Induction Motor Models are
1. Synchronous frame of reference , where zA  z2 0 ¯³

where zA  0
2. Stationary frame or stator frame of reference,

Rotor frame of reference, where ω = zw

Equations for Formation of Stator Fluxes by Using
3.
L~2$
Stator Voltage Equations are given as

 ¼2$ M |2 Œ2$ M zA ~2–

L
The Complete Common Structure of Induction Motor
Model is as Follows

 ¼2– M |2 Œ2– 0 zA ~2$

$„…¾
$‡
(9)

Equations for Formation of Rotor Flux Equations by

L~w$
Using Rotor Voltage Equations are given as

 ¼w$ M |w Œw$ M zA Mzw

~w–
L
 ¼w– M |w Œw– 0 zA Mzw
~w$ (10)
$„j¾
$‡

Equations of 2-Phase Stator and Rotor Currents in

~2$ Œ2$
Terms Of Stator and Rotor Fluxes are given as

«~ ® Œ
ª 2– ­  ,¿9 À 2– Á
ª~w$ ­ Œw$
Fig 3. Complete Structure of Induction Motor
(11)
© ~w– ¬ Œw–
Modelling
A. Synchronous Frame of Reference:

WhenzA  z2 , This is called the synchronous

Where,
¹22 0 ¹ 0
0 ¹22 0 ¹
,¿9  À Á
rotating reference frame because the d, q axes rotate at
¹ 0 ¹ww 0
synchronous speed. However, with the reference frame
rotating at the same speed as the stator and rotor space 0 ¹ 0 ¹ww
field mmf waves, the stator and rotor d,q variables are
constant quantities, where as the actual variables are at Equations for Rotor and Stator 3-Phase Currents in
50 Hz and slip frequencies respectively. The stator d- Terms of 2-Phase Rotor and Stator Currents are given
axis current gradually reduces from 50 Hz frequency as
cosv
sinv


Y{ «√ ® Y
at the instant of switching to the equivalent of a steady
 ª ­ A
°Y ±  » ª√ cos ƒv M ˆ sin ƒv M ˆ­ D Y$ F
 
Dc when at rated speed.

 ­ Y–
 
Y‚

ª
(12)
cos ƒv 0 ˆ sin ƒv 0 ˆ¬

©√
When the reference frame is rotating at
synchronous speed, both the stator and rotor are  
rotating at different speed relative to it. Equations for Speed, Torque, and Power are Given
Below
Then the park’s transform is converted as below

G. T. Chandra Sekhar, K. Alfoni Jose, L.V.Suresh Kumar, B. Manmadha Kumar Page No. 53
 ~2$ Œ2$
«~ ® Œ
Speed equation:
$´jk
 C M C³ ª 2– ­  ,¿9 À 2– Á
$‡ ~
ª w$ ­ Œw$
(20)
© ~w– ¬ Œw–
(13)

¹22 0 ¹ 0
Torque equation:

C  ‘~w$ Œw– M ~w– Œw$ “

Ã
0 ¹22 0 ¹
,¿9  À Á
¹ 0 ¹ww 0

(14)

Power equation: 0 ¹ 0 ¹ww

¯Â Equations for Rotor and Stator 3-Phase Currents in
 ‘~w$ Œw– M ~w– Œw$ “zw
Ã

«√ 1 0 ®

Terms of 2-Phase Rotor and Stator Currents are given

YA
(15)
Y{
ª  √­
° Y ±  » ª√ M  M  ­ D Y$ F
Complete Structure of Induction Motor Model in 

Y‚ Y

ª ­
Synchronous Frame of Reference is as Follows (21)
–
©√ M 
 √
 ¬
B. Stationary Frame of Reference

In case of stationary reference frame, z  0,

Speed equation can be obtained from swing equation
as follows

J jk  CÂ M C³
$´
In this reference frame the d axis is fitted to and thus

$‡
coincident with the axis of the stator phase A winding. (22)
This means that the mmf wave of the stator moves

C  ‘~w$ Œw– M ~w– Œw$ “

Ã
Expression for the torque is
over this frame at the same speed as it does over the

¯Â  ‘~w$ Œw– M ~w– Œw$ “zw
(23)
Ã
stator phase ‘A’ winding. This reference frame’s stator

d-axis variables therefore behave in exactly the same (24)
way as do the physical stator phase ‘A’ variables of
the motor itself. The results show the identical nature
Complete Structure of Induction Motor Model in
of the stator phase ‘A’ current and the stator d-axis
Synchronous Frame of Reference is as Follows
current.
Voltage Equations:
C. Rotor Reference Frame:
In case of rotor reference frame z  zw . Since
The voltage equations of 3-phase is the matrix
multiplication of inverse of B and 2-phase voltages as

¼2{‚ ¼2A$–
in this reference frame the d- axis of the reference
G¼ K  ½ › G¼ K
shown below,
frame is moving at the same relative speed as the rotor
w‰Š wA$–
¯ 0
(17)

Where B denotes  ž
phase ‘A’ winding and coincident with its axis. The

0 ¯³
behavior of the d-axis current and the phase ‘A’
current would be identical.
R/√T R/√T R/√T
Equations for Formation of Stator Flux Equations by
Æ¡  qT/U D£¤N ÇÈ É
£¤N ÇÈ É M T§/U
£¤N ÇÈ É 0 T§/U
F (25)
 ¼2$ M |2 Œ2$
$„…Ä
Using Stator Voltage Equations are given as
NOP ÇÈ É
NOPÇÈ É M T§/U
NOP ÇÈ É 0 T§/U

$‡
 ¼2– M |2 Œ2–
$„…¾
$‡
¼2{‚ ¼2A$–
G¼ K  ,¯º 9› G¼ K
(18) Voltage Equations

w‰Š wA$–
Equations for Formation of Rotor Flux Equations by (26)

L~w$
Using Rotor Voltage Equations are given as
 ¼w$ M |w Œw$ 0zw ~w–

Equations for Formation of Stator Flux Equations by

 ¼w– M |w Œw– Mzw ~w$

$„j¾
L~2$
Using Stator Voltage Equation are given as

 ¼2$ M |2 Œ2$ M zw ~2–

$‡
(19)
Where rotor voltages are ‘zero’ L

 ¼2– M |2 Œ2– 0 zw ~2$

Equations of 2-Phase Stator and Rotor Currents in
$„…¾
$‡
Terms Of Stator and Rotor Fluxes are given (27)
Equations for Formation of Rotor Flux Equations by
Using Rotor Voltage Equations are given as

G. T. Chandra Sekhar, K. Alfoni Jose, L.V.Suresh Kumar, B. Manmadha Kumar Page No. 54
 L~w$
 ¼w$ M |w Œw$
L
 ¼w– M |w Œw–
$„j¾
$‡
(28)
Where rotor voltages are ‘zero’
Equations of 2-Phase Stator and Rotor Currents in

~2$ Œ2$
Terms of Stator and Rotor Fluxes are given as
«~ ® Œ
ª 2–
­  ,¿9 À 2– Á
Fig.5 Plot between time t in seconds and speed wr

~
ª w$ ­ Œw$
(29)
© ~w– ¬ Œw–
B. Stator Frame of Reference

¹22 0 ¹ 0
0 ¹22 0 ¹
,¿9  À Á
¹ 0 ¹ww 0
0 ¹ 0 ¹ww
Equations for Rotor and Stator 3-Phase Currents in
Terms of 2-Phase Rotor and Stator Currents are given
as
Y{ 1/√2 cos zw 
sin zw 
YA
° Y ±  q2/3 À1/√2 cos zw  M 2^/3
sinzw  M 2^/3
Á D Y$ F (30)
Y‚ 1/√2 cos zw  0 2^/3
sin zw  0 2^/3
Y– Fig.6 Plot between speed Wr and Electric Torque

¯Â  ‘~w$ Œw– M ~w– Œw$ “zw

Ã
From power equation,

(31)

C  ‘~w$ Œw– M ~w– Œw$ “

Ã
Expression for the torque is

(32)
Swing equation:

J jk  CÂ M C³
$´
The swing equation is given by
$‡
Where zw is mechanical rotor speed in mechanical
(33)
Fig.7 Plot between time t in seconds and speed wr
rad/sec

C³ is the load torque

C. Rotor Frame Of Reference
J is the moment of inertia of motor and load

CÂ is electromechanical torque
Complete Structure of Induction Motor Model in
Synchronous Frame of Reference is as Follows

V. RESULTS
The obtained Speed- torque and speed- time
curves for both Synchronous frame, stator frame
and rotor frame are given in this section.
A. Synchronous Frame of Reference Fig.8 Plot between speed and Torque Te

Fig.4 Plot between speed , Wr and Electric

Torque: Fig. 9 Plot between time t in seconds and speed wr

G. T. Chandra Sekhar, K. Alfoni Jose, L.V.Suresh Kumar, B. Manmadha Kumar Page No. 55
VI. CONCLUSION H 0.1499
B 0
This paper has taken the three preferred
Frames of reference within the d,q two axis theory and
obtained results has compared and concluded as ,
When single induction motor is being studied, any of VII. REFERENCES
the three reference frames can be chosen for studying
transient analysis. When the synchronous frame of [1] “Simulink Implementation of Induction Machine
reference is used, the study dc quantities both of the Model- A Modular Approach”, Burak Ozpineci,
stator and rotor d, q variables make this the preferred Leon M.Tolbert, 2003 IEEE.
frame of reference. If the stator frame of reference is
used, then the stator d-axis variable is identical to [2] “Transient Analysis of Three-Phase Induction
those of stator phase A variable. If the rotor frame of Machine Using Different Reference Frames”,
reference is used, then the rotor d-axis variable is Vivek Pahwa and K.S. Sandhu, 2009
identical to those of rotor phase A variable. When
multi-machine system is used, then synchronous frame [3] L. Tang, M. F. Rahman, ”A new direct torque
of reference is preferable than the other two reference control strategy for flux and torque ripple
frames. reduction for induction motors drive-a
MATLAB/SIMULINK model, ”IEEE International
APPENDIX electric machines and drives
conference,2001,pp.884-890.
INDUCTION MOTOR DATA IN PER UNIT
VALUE [4] “Generalised theory of electrical machines” by P.
Sb 3.97 S. Bimbra, Khanna Publishers, fifth edition.
Vb 460 [5] “Modelling, Analysis and Control of Electric
Ib 6.4412 Motor Drives” by R.Krishnan, Prentice Hall,
India, 2005.
Zb 71.4151 [6] B. Adkins and R.G. Harley, “The General Theory
Ub 1.2202 of Alternating Current Machines: Application to
Practical Problems”, Chaman and Hall, London,
Wb 376.9911 1975.
Rs 0.0248
[7] Paul C. Krause, ”Analysis of Electric
Rr 0.0188 Machinery”, McGraw-hill inc., New York, 1987
Xls 5.25
[8] “Dynamic Model of Induction Motor for Vector
Xlr 4.57 Control”, Dal Y. Ohm, Drivetech.
Xm 139
[9] “Modelling And Simulation of The Three-Phase
Lls 0.0735 Induction Motor Using Simulink”, K.L. Shi, T.F.
Llr 0.064 Chan.
Lb 0.1894
Lm 1.9464
Wrb 188.4956

G. T. Chandra Sekhar, K. Alfoni Jose, L.V.Suresh Kumar, B. Manmadha Kumar Page No. 56
 Energy Management in Power Sector
Indian Prospective
Dr. Debakar Panigrahy1, Debasmita Panigrahy2
Professor, Dept. of M.B.A, BERHAMPUR UNIVERSITY1
Assistant Professor, Dept. of M.B.A, AITAM, TEKKALI-5322012
debasmitapanigrahy@yahoo.co.in
ABSTRACT
The electricity industry in India continues to evolve due to regulatory changes and market forces. It has
moved on from the vertically integrated system to open access market. Through overt public policy and an emerging
industry structure, the wholesale competitive marketplace has evolved. Regulatory changes are likely to be
significant, particularly with regard to how the existing system and operated in the future. These ongoing changes in
the structure and regulation of industry require changes in approach to resource planning. Given the potential for
commodity markets (both natural gas and electric) to exhibit price swings, or volatility, alternative resources plans
must be evaluated in terms of their exposure to this volatility, in addition to their long-run average costs.
Furthermore, unpredictably in the future costs of new supply alternatives arising from fuel cost( primarily natural
gas price) provides analysis leading to a comprehensive portfolio and strategy for supply acquisitions, transmission
investments and demand side management along with the consideration of environmental issues. The purpose of
IRP is to provide reliable, safe and least cost electric service to its customers. This paper pays emphasis on supply
side management and its integrated resource planning.
Keywords— IRP, supply side management, demand side management, cogeneration, renewable energy,
environmental issues.
As the deregulation of the electric power industry
I. INTRODUCTION
progresses in India, the basic structure of power
The Indian power sector has been increasing its systems will change drastically. For example, it is
installed capacity, from 30,000 MW in 1981 to over expected that small-scale distributed generation
100,000 MW in 2001. Despite this growth in supply, facilities (such as photovoltaic, fuel cell generation,
its power systems are struggling to overcome chronic etc.) and dispersed type energy storage systems (such
power shortages and poor power quality. With demand as secondary battery, electric vehicle, SMES, etc.) are
exceeding supply, severe peak (around 18%) and installed in the demand side or the power distribution
energy (around 10%) shortages continue to plague the systems. The introduction of a disaggregated,
sector. Shortages are exacerbated by inefficiencies in competitive electricity supply industry will affect the
power generation, distribution and end-use systems. appropriateness of various planning models currently
The inefficiency on the supply side is due to used. One such approach that has been significant in
unavailability of better quality of natural resources, promoting energy efficiency and demand side
less use of newer technology and lack of awareness. Management in a number of countries is Integrated
The inefficiencies in the end-use systems is due to Resource Planning (IRP).
irrational tariffs, technological obsolescence of
2.1 What is integrated resource planning?
industrial process and equipment, lack of awareness,
and inadequate policy drivers (such as energy Integrated resource planning is a planning
efficiency standards and labelling system, financial process that seeks the least cost option for meeting
incentives) in India. Another great concern is customer’s electricity demands. In determining the
biological and adverse effect of generating plants and least cost option, IRP evaluates all supply and demand
electrical equipments on human beings and side options over a forecast period from a societal
environment. perspective. IRP implies significant regulatory
oversight which can be applied by a number of means.
In its broadest context IRP includes all energy forms

Dr. Debakar Panigrahy, Debasmita Panigrahy Page No. 57

and the whole national energy market. IRP can also be
implemented for the electricity industry, or at an
individual utility level. A key feature is the regulatory
oversight of the process. IRP can take a variety of
forms. At one extreme, government may exert
significant control over the process. This may involve
establishment of an independent energy planning body
with responsibilities including:
1. Development of IRP procedures, including
costing of environmental externalities;
2. Oversight of public consultation and review of
plans;
3. Amendment of plans;
4. Setting and reviewing utility performance
criteria identified in the plans;
5. Reviewing and approving major capital
investment decisions.
At the other extreme, utility may simply be required
by government to periodically publish report on its
strategic planning process, and seek public comments.
IRP may be initiated in two main ways:
1. By governments applying IRP across one or
more major industry sectors- this could be
termed as “mandatory IRP”
2. By individual electricity businesses, applying
IRP only within that particular business- this
could be termed as “business-related IRP”.
Whatever the form of the IRP, the following elements
are common:
1. It is a continuing and iterative process, ideally
planning, implementation and evaluation;
2. It is an attempt to balance long and short term
goals, and to meet multiple goals;
3. All resource options are compared in the one
forum and a portfolio approach is promoted;
4. Public discussion and debate are involved.
Mandatory IRP works well in an electricity industry
composed of vertically integrated monopoly electricity
businesses. In an industry structure where the
functions of the electricity industry have been
unbundled into separate businesses, both mandatory
and business related IRP can still be undertaken in
Dr. Debakar Panigrahy, Debasmita Panigrahy
the natural monopoly elements of the industry, i.e. the
transmission and distribution network sectors.
2.2 Integrated resource planning in electricity
industry
Many countries are looking for "deregulating" or
"restructuring" their electric power industries. The
basic purpose of this restructuring is to open electric
power industries to competition among suppliers.
These efforts are considered to provide better quality
of electric supply to consumers at a lower price. There
is a point to consider that how the
IRP process will include the social benefits with the
deregulation of the electric power industry. The
electricity planning problem in many developing
countries is more complicated as compared to
developed countries. Generation of power is difficult
because electric power industries require large capital
investment. The developing countries have a problem
to substitute investment in cost-effective reductions in
electricity use for investments in power plants. The
combined planning for increasing electricity capacity
to meet growing demand with cost-effective
reductions in use of electricity is called integrated
resource planning. The growing demand for electricity
throughout the world is met by electric utilities by
building more power plants. Increased burning of
fossil fuels in these power plants could pollute the
atmosphere. Burning of fossil fuels in power plants
would produce more CO2 which contribute to global
warming. In developing countries due to high growth
in power generation the CO2 emissions grew fastest in
last few years. Over last two decades the CO2
emission increased by about 80% in these countries.
The combination of resources different from those
used in the past and the integration of these resources
in a cost effective way could ease the health,
economic, and environmental effects of increased
demand for electric power.
Environmental problems such as emissions of
CO2 gas from the power plants are more complex
while building the power plants to meet the increasing
demand for power. According to the environment
regulatory bodies the countries should encourage the
programs to limit the greenhouse gas emissions. IRP
process may alter the pattern and the levels of
electricity demand and constructing non- conventional
power plants. IRP takes the future assessment of
energy supply and demand. It combines them while
considering environmental and social factors, energy
efficiency and load management programs, and
Page No. 58
different limitations of the resources. Technically it
should treat the demand side resources and the supply
side resources equally.
2.3 How the integrated resource planning process
works
1. The integrated resource planning process
analyses the present situation of existing
resources, future demand for electricity and
identifies the reliability of service,
environmental impacts and economical price
for the supply.
2. IRP identifies resource options that would
compensate for discrepancies between expected
loads and existing capacities.
3. The options are evaluated considering the
economic, social and environmental factors.
4. The resources are combined to meet the
demand for electricity in such a way that satisfy
all regulatory issues and avoid risks.
5. Finally the plan, thus obtained, is implemented.
3. HOW TO DO INTEGRATED RESOURCE
PLANNING IN INDIA
3.1 Supply side management
The elementary problem faced by the power sector is
the poor financial conditions of State Electricity
Boards (SEBs) or successor entities in most states.
Over the years SEBs have been causing an
increasingly larger drain on the State Government
budgets, contributing to 10-15% of the state fiscal
deficits adversely impacting much needed investments
in the social sectors of health and education. The
power sector is operating with very low or no returns
on the equity and no contribution to future investments
from internal resources. This results in inadequate
investment in additional generation capacity which is
likely to further exacerbate the existing gap between
power supply and demand. In 1991, IPP proposals
exceeded 150,000 MW, while as of Jan 2001, just
3,500 MW of IPP power was actually operational.
Even if captive market capacity addition of 1,500-
2,000 MW per year is included, a total capacity
addition of not more than 6,000 MW a year over the
next 4-5 years is expected. This translates into US$ 6
billion of investments and several million tons of
additional pollutants but would still not be close
Dr. Debakar Panigrahy, Debasmita Panigrahy
enough to meet the targeted capacity increases of
111,500 MW by 2007.
3.1.1 Supply-Side Options
1. Existing traditional power plants
2. Cogeneration of power
3. Competitive market for selling power
4. Renewable energy sources such as bio-fuels,
solar etc. that is economic and environmentally
acceptable
3.2 Demand side management
DSM is a concept in which a power utility, such as a
vertically integrated SEB or an unbundled distribution
company, manages the demand for power among some
or all its customers to meet its current or future needs.
In India, DSM can be achieved through energy
efficiency, which is the reduction of kWh of energy
consumption or demand load management, which is
the reduction of kW of power demand or the
displacement of demand to off-peak times. In the
former category there are programs such as awareness
generation programs, customer or vendor rebates for
efficient equipment, etc. while the later includes time-
of-use tariffs, interruptible tariffs, direct load control,
etc. Specific type of programs depends on the utility
objective: peak clipping, load shifting, strategic
conservation or strategic load growth. Reductions in
energy demand and construction at the end user’s
premises can free up electricity generation,
transmission and distribution capacity at a fraction of
costs required to provide new capacity. The cost of
saved energy has been estimated to be as below as
10% of the cost added capacity for some DSM
measures.
3.2.1 Demand-Side Options
1. Use of high efficient motors, energy-efficient
appliances, lighting, refrigeration, heating,
thermal storage
2. Load management programs
3. Reduction of T&D losses
4. Tariff based on time-of-use
5. Revenue reconciliation
6. Use of available solar energy for heating and
other applications
Page No. 59
4 SUPPLY SIDE MANAGEMENT AND
INTEGRATED RESOURCE PLANNING
4.1 Existing traditional power plants
The efficiency of existing power plants should be
increased. Use of better quality of coals, new
technologies, efficient motors, modern techniques for
ash handling, etc. may increase efficiency of power
plants and reduce environmental problems.
4.2 Competitive market for selling power
Many countries have a concept of deregulating their
electric power sectors. Deregulation allows
competition among generators and creates market to
provide economic and higher quality of supply. The
needs are achieved through transmission open access
(TOA). In open access system, the use of transmission
system should be open and non discriminatory. The
transmission open access is the most recent concept to
provide better quality of service at a lower price and to
regulate the monopoly in the transmission system
thereby creating market conditions. The private
players should also be encouraged to take interest in
the business of electric power. Transmission open
access is meant by enabling third part to use the
transmission network which belongs to fully or
partially to another party for the transactions under the
regulatory concept. Transmission open access may be
an effective mean to provide reliable and economic
electric supply. For power sector reform transmission
open access is considered as a useful tool. To create
competition in the electricity business it is required
that there is an easy access to the electricity
consumers. In open access mode, the generators may
have direct contract with the consumers. In open
access system it is required that the access to
transmission should be regulated and efficient pricing
methodologies should be applied. There would be
more complexity in the future expansion of the
transmission system. Regulatory aspects would be
critical. To create competitive electric market in Indian
scenario is a desirable feature. The present status of
power supply in India is ill-conditioned. There is a
need of deregulating the Indian power sector to
achieve a higher quality of supply at a lower cost. The
power sector reform programs have paid attention
towards the unbundling of three electric companies’
viz. Genco, Transco and Disco. There is very little
redundancy in the generation and the transmission. To
create open access to the transmission system, there
should be an encouragement to generators to increase
the generation. As the generation is less than the
demand in India, a need arises to open the market for
Dr. Debakar Panigrahy, Debasmita Panigrahy
new generators. The competition in the generation
would help to reduce the cost of supply.
4.3 Cogeneration/captive power
The cogeneration or captive generation provides the
additional demand for electric power in industries. In
developing countries like India the cogeneration plants
serve a significant amount of power in industries. For
better quality of supply at lower price and to reduce
risks the most industries now have their own
cogeneration/ captive power plants.
4.3.1 Cogeneration
“A Cogeneration Facility is defined as one which
simultaneously produces two or more forms of useful
energy (e.g. electric power and steam electric pore and
shaft (mechanical) power etc.).” The Ministry of
Power resolution No. A-40/95-IPC-I date. 06.11.96.
Cogeneration is an attractive way to meet the energy
requirements of industries. Cogeneration provides
greater savings to money and the environment.
Cogeneration is not a new word in the industries.
Many of industries in earlier days had cogeneration
plants. Now again due to increasing competition in the
market to offer lower price and to reduce emissions of
air pollutants, the industries are going for efficient use
of energy. Cogeneration offers an efficient use of
energy in industries. Cogeneration facilitates them to
generate their own power and thermal energy from
waste. The opportunities for the industries having
cogeneration are
1. Energy at lower price
2. Choice of fuels
3. Improved power quality
4. Utilization of waste
5. Protection of the equipments from power
interruptions
Industries like sugar, fertilizers, paper, food processing
and textiles etc. have their cogeneration plants to
produce power and heat for the various processes. The
heat produced by the various processes can be utilized
to generate power. The waste of the industries have
used as a primary fuels.
4.3.2 Captive power production
Many industries such as steel, cement, fertilizers,
aluminum, paper and sugar have captive capacity to
meet significant part of their electricity need. The
Page No. 60
captive power production fulfils the about 30% need
of total electricity requirement of industries in India.
4.3.3 Grid connectivity
In earlier days these cogeneration plants were not
connected to the grid. The grid connectivity of power
produced by a cogeneration plant
1. Plant could run in synchronous or supply power
radial
2. Amount of power to be supplied should be larger
as compared to the line losses
3. The consumer should not far away from the
supplier
4. The cost of the laying lines and interfacing
system should be less so that it ensure profit to
the co generator
The co-generator sends power at 33 kV (up to 10 MW)
or 132 kV (above 10 MW) voltage through
transmission lines of corresponding voltage to the grid
substation. The cost of low voltage lines from the
cogeneration plant to the grid could be borne by the
co-generator. The cost of high voltage lines is very
much so it could be divided between the co-generator
and purchaser. In sugar mills in Uttar Pradesh sell
power to UPPCL and the total cost of equipments,
lines, operation and maintenance is shared between
them. The existing network may be utilized for
transmitting power from the cogeneration plant to the
grid on an agreement. Regulatory bodies have made
some rules to arrange these agreements
1. The plant shall maintain the equipments at the
generating end and the maintenance of lines and
the equipments at the substation shall be done
by the grid substation. The co-generator could
do any work with specific approval of the buyer
or regulatory commission. Installation,
operation and maintenance of interconnection
should be done in accordance with Electricity
Grid Code.
2. The system should be designed such that if any
false or abnormal operation occurs, the
cogeneration plant is isolated by circuit breaker.
3. The plant generating set should be synchronized
with the grid. The buyer would not be
responsible for any fault occur to the plant due
to failure of synchronizing or protective system.
Dr. Debakar Panigrahy, Debasmita Panigrahy
4. The power exported and imported would be
measured with two set of frequency sensitive
static meters. These meters shall satisfy all the
technical standards of Indian Electricity Rules
and the specifications of the bureau of Indian
standards.
5. The transactions of power would be done as per
instruction from the Load Control Centre.
4.3.4 Environmental issues with cogeneration
plants
Using waste as a primary fuel cogeneration offers
some environmental benefits. Cogeneration uses heat
and consumes less fuel. When the fuels are burned, air
pollution is the severe effect to be considered. The air
pollutants are mainly nitrous oxides (NOx) and
sulphur dioxide (SO2). Cogeneration reduces some of
the emissions of atmospheric pollutants associated
with the combustion of fossil fuels such as carbon
dioxide, a contributor to global warming, and sulphur
dioxide which results in acid rain. National Ambient
Air Quality Standards (NAAQS) has set standards to
meet the environmental issues for thee cogeneration
plants. Diesel generators produces much waste heat
and are not able to utilize this heat. These generators
are not environmentally feasible and may cause of
additional air pollution. There are some special
equipments required to ensure environmental
regulations. With these pollution control equipments
the cogeneration seems reliable solution for the
industries.
4.3.5 Cogeneration applications
Cogeneration systems are applied for small as well as
large industrial applications. Small cogeneration
systems are compact and economical. The systems
below 600kW capacity produce electricity and hot
water from engine waste heat which can be utilized in
some other applications. The most common uses for
the heat from cogeneration power plants are steam for
industrial processes, and space and water heating.
They can operate during peak demand to meet the
additional requirement of power. Industries consume
the largest amount of power generated. They are now
going for their own power production independent of
grid. The reasons are:
1. The industry requires continuous power supply
2. The supply should have quality and reliability
3. The supply should be economically feasible
Page No. 61
Due to the poor performance of electricity boards,
frequent power cuts and high tariff the industry the
applications of cogeneration and captive generation
are increasing day by day.
4.4 Renewable energy sources
To meet the increasing power demand the number of
power plants is also increasing. The gas emissions
from these power plants on tribute to air pollution a
great extant. The burning of fossil fuels produces
CO2 and SO2. These gases are harmful to human
being as well as environment. Use of bio-fuels as a
primary fuel in power plants reduces the level of
pollution and also utilizes the waste from industries.
These renewable sources are available in abundant in
our country and reduces the dependence on another
countries for petroleum. In IRP process these resource
options are evaluated well.
4.4.1 Power through bio-fuels: waste management
The term bio-fuels can refer to fuels for electricity and
fuels for transportation. Bio-fuels such as ethanol and
bio-diesel are becoming good supplement of
petroleum fuels. Bio fuels include alcohol, ether, ester
and other bio-products such as soybean, rapeseed and
vegetable. These fuels may be more economical and
require less processing. The bio products are now
increasingly being seen as the bio-fuels of the 21st
century. The biomass resources like agricultural waste,
municipal solid waste, and industrial waste have been
used as renewable fuels. The biomass contains high
energy that has been stored through photosynthesis.
That energy content remains when plants are
processed into other materials such as paper and
animal wastes etc. Bio-fuels are produced
domestically and this helps to create opportunities for
jobs and reduce our country’s trade deficit. This is
very beneficial for our economy. Since the bio fuels
are made from agricultural waste it offers new options
for agriculture sector. The domestic production of bio-
fuels makes us less dependent on other countries to
import oil. This improves the energy security of our
country and the energy sector would not be much
affected from the economy of the country having the
resources of crude oil. The environmental issue is not
severe with the use of bio-fuels as the bio-fuels
produce less harmful emissions. When the production
and the combustion of bio-fuels takes place they
pollute the atmosphere to little and reduce the build-up
of greenhouse gases.
Dr. Debakar Panigrahy, Debasmita Panigrahy
5. CONCLUSION
The electricity planning is not easy in India. The
challenge is not only to substitute investments in cost-
effective reductions in electricity use for investment in
power plants but also to obtain a better efficiency of
existing power plants. The issues like Demand Side
Management have been discussed since last many
years but Supply Side issue in power system has not
been given adequate attention. The types of generating
units used by electric utilities, cogeneration, captive
plants, environmental issues are motivating force to
engage in Integrated resource Planning. It is estimated
that Integrated Resource Planning of Supply Side in
power system can increase the overall efficiency of
power system by 20-30% in India. Therefore there is a
great need for Integrated Resource Planning in Supply
Side Management.
References:
1. Beecher, J.A. 1995 “Integrated Resource
Planning Fundamentals.” Journal American
Water Works Association . 87 (6): 34-48.
2. .Berman, J.S. and D.M. Logan. 1990. "A
Comprehensive Cost-Effectiveness Methodology
for Integrated Least-Cost Planning," presented
at a conference of the Electric Power Research
Institute held in Milwaukee, Wisconsin, May 2-4,
1990.
3. Power to the People: Integrated Resource
Planning in Developing Countries, L. Hill, Oak
Ridge National Laboratory.
4. A. S. Malik, and B. J. Cory, “Integrated resource
planning with consideration of dynamic costs of
thermal units”, Electric Power Systems Research
(1999) pp. 123-130.
5. Monenco Consultants & Associates. “Demand
Side Management Working Paper No. 3.” Under
contract to the National Energy Policy Office.
Bangkok. July 1991
6. Nadel, Steve, 'Utility Conservation Programs' in
State of the Art of Energy Efficiency: Future
Directions. Edited by Edward Vine and Drury
Crawley. American Council for an Energy-
Efficient Economy. Washington, DC. 1991
7. Faruqui, A and E Haites. 'Impact of Efficient
Electricity Use and DSM Programs on United
States Electricity Demand and the Environment.'
In conference proceedings from Demand Side
Management and the Global Environment.
Arlington, VA. April 1991
Page No. 62
 Power Quality Mitigation Using Dynamic
Voltage Restoration (DVR)
Balaji Prasad Padhi1, Dr.Deba Prasad Das2
Professor, Dept.of E.E.E, Gandhi Institute for Education & Technology, Baniatangi, Bhubaneswar1.
Professor, Dept.of Electrical Engg, Orissa Engineering College, Bhubaneswar2.
Email:Prasad.giet.2011@gmail.com,debadas15@gmail.com

ABSTRACT
In present days Control of most of the industrial loads is mainly based on semiconductor devices, which causes
such loads to be more sensitive against power system disturbances. Thus, the power quality problems have gained
more interest recently. In this voltage sags and swells are vital power quality problems. Therefore, different
solutions are examined to compensate these sags and swells to avoid production losses at sensitive loads. In this
project Dynamic Voltage Restorer (DVR) is discussed for these problems.
A sliding mode controller is designed and developed for single phase DVR by E. Rammohan Rao c. tal. [3]
and an attempt is made to extended it to three phase DVR. Using sliding mode control to the DVR, additional
sag/swell detection method is eliminated. This improves the dynamic response of the DVR and also DVR is able
to compensate for any variation in source voltage. Usage of sliding mode control to DVR makes it
multifunctional, such as compensation for voltage sag, swell, voltage flicker and voltage harmonics. The
validation of sliding mode control is verified threw MATLAB/SIMULINK simulation results.
Keywords: Dynamic Voltage Restorer (DVR), voltage sag, swell, voltage flicker and voltage harmonics.

I. INTRODUCTION Voltage sags are characterized by a reduction in

Utilities aim to provide their customers with an ideal
sinusoidal voltage waveform. By definition ideal
sinusoidal voltage waveform has the characteristics:
constant magnitude at the required level, constant
frequency and balanced in case of three Phase
operation. Naturally, this is not always possible
because of normal system variations and due to the
unavoidable incidents that temporarily can affect the
operation, such as short-circuits faults. On the other
hand, utilities require that customers should draw
sinusoidal current. Fig.1 illustrates the relation
between the utilities and their customers in a
systematic way. Customers place demands on the
voltage ug(t), while the utilities specify the current
ig(t).
Fig.1. Relation between utilities and customers
Significant deviation from the normal voltage is a
problem for sensitive consumers in the grid system.
Interruptions are generally considered to be the worst
case with the load disconnected from the supply.
voltage, but the load is still connected to the supply.
Sags are in most cases considered less critical
compared to interruptions, but they typically occur
more frequently. Voltage sags have in several cases
been reported to as a threat to sensitive equipment and
have resulted in shutdown, loss of production and a
hence a major burden . The concept of custom power
has been developed using advanced power electronic
equipment to ensure a high quality of supply. The
Dynamic Voltage Restorer (DVR) is the one of the
custom power devices, which has excellent dynamic
Capabilities, and it is well suited to protect critical or
sensitive loads from short duration voltage dips and
swells. With the DVR installed on a critical load
feeder, the line voltage is restored to its nominal value
within the time of a few milliseconds thus avoiding
any power disruption to the load.
POWER QUALITY PROBLEMS
The term "power quality" is still an item of
controversy between power engineers. Nevertheless it
is widely used to describe the utility-customer
interactions. The definition of power quality given in
the IEEE dictionary originates in the IEEE Std 1100
is: “Power quality is the concept of powering and
grounding sensitive equipment in a matter that is
suitable to the operation of that equipment.”

Balaji Prasad Padhi, Dr.Deba Prasad Das Page No. 63

 It is worth to mention here that the Current decreases to less than 0. 1 pu for a period time
International Standards setting organization in not exceeding 1 minute.
Electrical Engineering (the IEC) does not yet use the
2.1.2. VOLTAGE DIPS:
term power quality in any of its standard documents.
For the moment it uses the term “Electro Magnetic A voltage dip is a decrease in the RMS voltage from
Compatibility”. Instead the following definition is 0.1 to 0.9 pu at the power frequency for duration from
given in IEC 610001-1: 0.5 cycles to 1 minute. A synonym for voltage dip is
voltage sag, which is widely used in the United States.
Electro Magnetic Compatibility is ability of
Voltage dips have the potential to disrupt the sensitive
equipment or a system to function satisfactorily in its
load operation and cause a loss of production. An
electromagnetic environment without introducing
example of a voltage dip is shown in Fig.2.1.
intolerable electromagnetic disturbances to anything in
that environment. From the definitions of both power
quality and electromagnetic compatibility terms, it can
be stated that the two terms are not the same but they
have a strong overlap. The prospective of the
customers of electric power is to have an ideal voltage
waveform. Thus, the customer interest in power
quality is mainly the voltage quality. On the other
hand, utilities concern is focused on both voltage and
current quality. The power quality problems are
classified as
• Voltage quality Fig.2.1. Example of Voltage dip
• Current quality 2.1.3. VOLTAGE SWELLS:

2.1. VOLTAGE QUALITY:

Voltage quality is the quality of the product delivered
by the utility to the customers. It is concerned with
deviations of the voltage waveform from the ideal
sinusoidal waveform. Power quality problems
encompass a wide range of disturbances that can
disrupt the operation of sensitive industrial loads and
cause a loss of production. In this Section, the
following power quality problems are defined and
briefly discussed
A voltage swell is an increase in the RMS voltage
above 1.1 P.U at the power frequency for duration
from 0.5cycle to 1minute.Voltage swells are caused by
switching off large loads, switching on capacitor banks
and by earth faults in high impedance grounded
systems. Voltage swells are characterized by their
magnitude and duration. An example for a voltage
swell is depicted in Fig .2.2 the adverse effect of
voltage swells is:

• Short interruptions • Increasing the thermal stress on the load.

• Voltage dips • Causing component overheating or
• Voltage swells destruction.
• Voltage and current transients • Tripping industrial devices such as motor
• Voltage flicker derives and control relays.
• Voltage unbalance
• Phase angle imbalance and/or jump.
However, there are other power quality problems such
as over voltage, under voltage, noise and power
frequency variations, voltage and current harmonic
distortion.
2.1.1. SHORT INTERRUPTIONS:
A short interruption is the complete loss of the supply Fig.2.2. Example of voltage swell
voltage with a time period of 0.5 cycles up to 3
seconds. [IEEE- 1 159- 19951]. Another definition is
an interruption occurs when supply voltage or the load

Balaji Prasad Padhi, Dr.Deba Prasad Das Page No. 64

2.1.4. VOLTAGE AND CURRENT
TRANSIENTS:
A transient is that part of the change in a variable that
appears during transition from one steady state
operating condition to another. Transients can be
classified into two categories:
• Impulsive transients
• Oscillatory transients.
Some equipment such as power electronic devices is
very sensitive to voltage and current transients. Such
devices can be destroyed because of transients.
• Impulsive transient:
An impulsive transient is a very short duration
(ms) voltage or current increase/decrease that is
unidirectional in polarity. Impulsive transients are
characterized by their peak, rise and decay times.
Lightning, short-circuit faults and switching of heavy
loads are the main reasons for an impulsive transient.
An example of a current impulsive transient caused by
a lightning stroke is depicted in Fig.2.3
Fig.2.3 Example of impulse transient
• Oscillator transient:
An oscillatory transient is a sudden, non power
frequency Change in the steady state conditions, which
includes both polarities (positive and negative). If the
system voltage or the load current has an oscillatory
transient, this implies that its peak changes polarity
very rapidly. Oscillatory transients are described by
their predominant frequency, duration and magnitude.
Fig.2.4. Example of oscillator transient
Balaji Prasad Padhi, Dr.Deba Prasad Das
2.1.5. HORMONIC DISTORTION:
Harmonic distortion is the periodic deviation of the
voltage/current from the ideal sinusoidal waveform.
Harmonics are sinusoidal voltages/currents, which
have frequencies of multiple integer of the
fundamental frequency. Non-linear loads have the
main contribution to the harmonic distortion of the
utility voltage or the load current. A fundamental sine
wave together with 20%, 5th harmonic is shown in
Fig.2.5. Harmonics cause excessive heat in the system
components and can cause mis operation of protection
devices.
Fig.2.5. Example of Harmonics distortion
2.1.6. VOLTAGE FLICKER:
Voltage fluctuations cause the lamps to flicker. Flicker
is defined in the in European standard EN 50160 as:
Flicker is the impression of unsteadiness of visual
sensation included by a light stimulus whose
luminance or spectral Distribution fluctuations with
time. A typical voltage fluctuation waveform is
displayed in Fig.2.6. The arc furnace is one of the
most common causes of flicker. The flicker signal is
defined by its RMS magnitude as a percent of the
fundamental. The unit of flicker measuring is as the
perception of short time.
Fig. 2.6. Example of Voltage flicker
2.1.7. VOLTAGE UNBALANCE:
Voltage unbalance is the difference in phase (or
phase-to-phase) voltage magnitudes of the three-phase
system. The voltage unbalance in percent is defined by
the National Electrical Manufacturing Association
(NEMA) in as (not generally accepted) the maximum
deviation from average to the average of the three
phase-to-phase voltage. Symmetrical components are
Page No. 65
used to characterize the voltage unbalance, which
gives an index for describing the system unbalance.
This index is called the Voltage Utilization Factor
(VUF):
%VUF = (Un / Up) x 100
Where Up, is the positive sequence component and Un
is the negative sequence component of the phase-to-
phase voltage. An example of unbalanced voltages is
given in Fig.2.7. Unbalanced voltages have adverse
effects on the induction machines that are the machine
efficiency is decreased and they may lead to motor
failure. Power electronics converters suffer also from
voltage unbalance. Operation of a diode rectifier is
influenced by the unbalance in the sense that the input
current harmonics are not restricted to the
characteristic harmonics of the rectifier.
Uncharacteristic triplen harmonics such as the 3rd and
the 9th can appear. The voltage unbalance may cause
excessive currents in one or two phases, which can trip
the overload protection circuits of the Adjustable
Speed Drives (ASD). The increased current can also
cause excess heating of the diodes and decrease the
lifetime of the dc link capacitor.
Fig.2.7. Example of voltage unbalance
2.1.8.PHASE ANGLE JUMP:
Phase angle jump is a change in the phase angle of the
voltage/current. Some loads such as thyristor based
derives are sensitive to phase-angle jump, where it can
lead to wrong determination of the voltage zero
crossing. A phase angle jump of 60 degrees is shown
in Fig. 2.8.
Fig. 2.8. Example of Phase angle jump
2.2 CURRENT QUALITY:
Current quality is a complementary term to the
voltage quality. It is concerned with deviations of the
current waveform from the ideal sinusoidal waveform.
Balaji Prasad Padhi, Dr.Deba Prasad Das
In addition to the ideal current waveform (as
demanded by the utility), the current sinusoidal wave
should be in phase with supplied voltage to minimize
the transmitted apparent power and consequently the
power system ratings. Because voltage and current are
closely related, a deviation of any of them from the
ideal may (with a high probability) cause the other to
deviate from the ideal case.
3. CAUSES AND CHARACTERISATION OF
VOLATGE DIPS/SWELLS
Voltage dips are the most important power quality
disturbances in electrical system. The outage costs
associated with poor power quality are documented
and serve as an argument to study the possible use of
power conditioning equipment. The effects of voltage
dips, swells and flicker on end-users have been
depending on the severity and duration can cause
computer resets, memory loss, tripping of adjustable
speed drives, loss of motor load, and this in turn leads
to serious disruption of the production process.
The mean causes of voltage dips are short-
circuit faults in transmission and distribution systems.
Large induction motors when starting and transformers
when being energized can also cause voltage dips. In
some networks, switching large loads may result in
voltage dips. Voltage dips due to short-circuit faults
are characterized by a dip magnitude, a dip duration,
and a phase-angle jump. The voltage dip magnitude is
the retained voltage after the initiation of the dip. It
can be determined by either the RMS voltage or the
peak voltage over half a cycle or one cycle of the
fundamental frequency. The following examples for
the voltage dips due to motor start and short circuit
fault are shown in Fig. 3.1 and Fig. 3.2.
In case of voltage dips due to motor start, the
characterization needs to be slightly different, other
terms are involved such as recovery time and steady
state drop. Voltage dip duration is the duration in
which, the system voltage remains under a threshold
value, for instance 90% of the pre-dip voltage.
Fig.3.1. Voltages dip due to Motor start
Page No. 66
If the dip is due to a short-circuit fault, the dip duration
is mainly determined by the fault clearing time, which
is affected by the speed of the protection and the speed
of the circuit breakers.. The phase-angle jump is a shift
in the zero crossing of the instantaneous voltage.
Because the system voltage is a complex quantity
consisting of a magnitude and phase angle, an event
like a short-circuit fault may not only affect the
voltage magnitude but also its phase angle. Normally,
voltage dips originated because of short-circuit faults
are associated with a phase-angle jump. Monitoring
the dip with the calculation of the RMS value does not
indicate the associated phase-angle jump but
monitoring with calculation of the fundamental
voltage gives also the phase angle
Fig. 3.2. Voltage dip due to Short circuit fault
3.1. EFFECT OF VOLTAGE DIPS/SWELLS ON
EQUIPMENT BEHAVIOUR:
Voltage dips may cause problems to low-power rating
as well as high-power rating equipment that is based
on power electronics devices. Low-power equipment
involves for instance programmable logic controllers
(PLCs) and computer power supplies. Examples for
high-power equipment are adjustable speed drives;
both ac and dc drives. Not only power electronic based
equipment is sensitive to voltage dips; also
electromechanical relays and motor contractors are
reported to open their contacts when the voltage drops
below 0.5 to 0.7 pu.
3.1.1. Programmable Logic Controller
Programmable Logic Controllers (PLCs) are used in
industrial processes to monitor the status of the
devices connected as inputs of a certain process for
instance, relays, switches and sensors. They are based
on power electronics control systems and have
software in their computer memories. The software
calculates the status of the devices connected as
outputs for instance alarms, lights and fans. The power
supplies of the PLCs and computers may be sensitive
to voltage dips that can stop the entire process being
controlled.
Balaji Prasad Padhi, Dr.Deba Prasad Das
3.1.2. Power supply:
A typical configuration of the computer supply
constitutes of a single-phase diode rectifier, a
smoothing dc capacitor (Cdc) and a dc/dc converter is
shown in Fig.3.3. The dc capacitor reduces the dc
voltage ripple. The dc/dc converter regulates the dc
output voltage. When the supply voltage Us drops, the
dc voltage also drops. The drop of the dc voltage
depends on the capacitor size and the load power. If
the dc voltage drops too much, the regulated dc
voltage will be too low and errors occur in the digital
electronics. The decay of the dc voltage Udc is
calculated as,
2P  P 
t ≈ U dc  1 − t 
2
u dc ( t ) = U dc −
C dc  C dc U dc2 
Where Udc is the initial dc voltage directly before the
voltage dip, P is load power. The dc voltage ripple is
calculated (in pu with Udc as a base voltage):
ε = PT 2C U
dc dc
If the minimum voltage that the power supply can
work properly is equal to Umin then the maximum dip
duration that can be sustained is:
t max =
(1 − U 2
min )
4ε
It can be concluded that the lower the dc voltage
ripple, the longer the dip duration that the power
supply can survive.
Fig.3.3. Configuration of Computer Power Supply
3.1.3. Adjustable speed drives:
A.C. motor drives:
A three-phase diode rectifier normally feeds AC motor
drives. Typical configuration of ac drives is displayed
in Fig.3.4. The dc inductor Ldc is used to smoothen the
dc current and the capacitor (C) minimizes the ripple
of the dc voltage Udc. Controlling the ac motor is done
by controlling the magnitude and the frequency of the
Page No. 67
output voltage of the Voltage source converter.
Voltage dips in the ac supply result in reduction of the
dc voltage Udc. Consequently, the voltage source
converter may not be able to operate and the motor
control is lost. The load connected to the ac motor may
not survive the reduction in the speed or the torque due
to the voltage. The motor energy is taken from the
capacitor at the dip start until the dc voltage becomes
lower than the supply voltage during the dip. To ride
through deeper and longer dips, the capacitor size
must be increased. Increase of the capacitor size
implies increase of the cost and increase of the
required installation area
• DC motor drives:
DC motor drives are fed by a thyristor rectifier, which
is connected, to the armature windings (Ra, La) of the
motor. The field winding (Rf , Lf) is energized by a
single-phase thyristor rectifier. Voltage dips are more
severe for dc drives compared to ac drives due to the
lack of the dc capacitor. Moreover, adding a dc
capacitor will limit the control speed of the dc drive. It
is of importance to mention here that not only the
reduction of voltage magnitude causes problems but
also associated phase angle jump with voltage dips
may cause problems to the thyristor rectifier in
determining zero crossings of the ac voltage.
Fig. 3.4. Configuration of DC drives
3.2. MITIGATION OF VOLTAGE DIPS/SWELLS:
Solutions of the problems due to voltage dips/swells
can be made at three levels. Apparently, the ideal
solution is to solve these problems at the equipment
level; increasing the equipment immunity to voltage
dips/swells. It means that the equipment should be
designed to tolerate voltage dips/swells. But this is not
always possible and requires more debate between the
customers and the manufacturers. Other solutions are
related to the power system configuration and
operation. Improvements can be done in order to
mitigate voltage dips involve:
• Reducing the number of voltage dips, e.g. by
institution of tree trimming policies, adding more
Balaji Prasad Padhi, Dr.Deba Prasad Das
lightning arrestors, lowering the ground resistance
at the foot of overhead transmission and towers
and replacing overhead lines by underground
cables.
• Reducing the fault clearing time by using faster
protection schemes keeping them reliable.
• Changing the structure of the power system. It can
be noticed that most of the power system
improvements are limited because most of the
power systems in operation now have been
designed 10 or 20 years ago and changing their
structure is not an easy task. Those improvements
are more effective when considered in designing
new power systems. This fact has led to think
about a third level solution for voltage dips
associated problem, which is installation of
mitigation equipment between the distribution
system and the equipment. Development of
mitigation equipment has been related to
developments of power electronics controllers,
which in turn led to the appearance of custom
power.
3.3. POWER ELECTRONICS SOLUTIONS FOR
POWER QUALITY PROBLEMS:
Custom power is a strategy designed
primarily to meet the requirements of industrial and
commercial customers. The concept of custom power
is the employment of power electronic or static
controllers in medium voltage distribution systems
aiming to supply a reliable and higher quality power
supplies that are needed by sensitive users. Power
electronic valves are the basis of those custom power
devices such as the State Transfer Switch (STS),
active filters and converter-based devices. The
converter-based custom power devices, used to
mitigate voltage variations and interruptions, can be
divided into two main groups: shunt connected and
series-connected devices. As well as in high voltage
transmission system the FACTS devices are used for
supplying the reliable power. These are divided into
two kinds of devices:
• FACTS devices
• Custom power devices
FACTS devices:
FACTS rely on state of the art power electronic
devices and methods applied on the high voltage side
of the power network to make it electronically
controllable. From the operational point of view, it is
Page No. 68
concerned with the ability to control the path of power
flows throughout the network in an adaptive fashion.
The equipment has the ability to control the line
impedance and the nodal voltage magnitudes and
angles at both sending and receiving end of key
transmission corridors while enhancing the security of
the system. The following two FACTS devices are
discussed in this thesis.
• Static Var Compensator (SVC)
• Thyristor Controlled Series Compensator (TCSC)
Custom power devices:
Custom power focuses on low and medium voltage
distribution systems. This technology is a response to
reports of poor power quality and reliability of supply
to factories, offices and homes. Today’s automated
equipment and production lines require reliable and
cannot tolerate voltage sags, swells, harmonic
distortions, impulses or interruptions. The following
devices are custom power devices.
• Static Transfer Switch
• Uninterrupted Power Supply (UPS)
• Distribution Static Compensator (D-STATCOM)
• Dynamic Voltage Restorer (DVR)
3.3.1. FACTS DEVICES:
3.3.1.1. Static VAR Compensator (SVC) :
Static VAR compensator (SVC) is the most widely
installed FACTS devices and is shown in Fig.3.5.
They mimic the working principle of a variable shunt
susceptance and use fast thyristor controller with
setting times of only a few fundamental frequency
periods. From the operational point of view, the SVC
adjusts its value automatically in response to changes
in the operating conditions of the network. The SVC
has the ability to either draw capacitive or inductive
current from the network. By suitable control of this
equivalent reactance, it is possible to regulate the
voltage magnitude at the SVC point of connection,
thus enhancing significantly the power systems
performance. Voltage regulation at key location of the
transmission system should provide the following
benefits:
• Prevention of large variations
• Prevention of voltage instability (voltage collapse)
• Enlargement of transient (first swing) stability
limits
• Provision of power oscillations damping.
Balaji Prasad Padhi, Dr.Deba Prasad Das
The SVC may be designed in many different ways.
The following figures are the most typical
arrangements for continuously controlled SVC’s these
are
• Fixed Capacitor with TCR
• TSC with TCR
Fig.3.5. Configuration of SVC with fixed capacitor and TCR
The thyristors are the controllable elements enabling
smooth control of the TCR when operated in the range
of 90-1800. On the other hand, the TSC is a fast
switched element that achieves voltage regulation in a
stepwise fashion. When the SVC is operated in a
voltage control mode, it is the fast thyristor controlled
FACTS controller, with setting times of almost one
period in the case of the FC/TCR arrangement. The
aim of the SVC in this application is to provide
voltage regulation at the point of connection. Initially
the SVC operated in open loop mode and for this
condition, the power transfer between the SVC and
AC system should be zero. When the load is
increased/decreased and the voltage at the load point
experience a voltage sag/swell. When the load is
changes, the SVC controller operation changes to
closed loop mode in order to adjust the SVC effective
impedance Xsvc, so that it injects capacitive current into
the system to restore the voltage back to the target
value.
Fixed capacitor and TCR:
In this the SVC consists of a TCR in parallel with a
capacitor. The thyristors firing angle control enables
the SVC to have an almost instantaneous speed of
response. It is used extensively to provide fast reactive
power and voltage regulation support. It is also known
to increase system stability margin to damp power
system oscillations. It is important to note that TCR
current can be varied continuously, without steps,
between zero and a maximum value corresponding to
full conduction of thyristors. The current is always
lagging, so that reactive power can only be absorbed.
However, the TCR compensator can be biased by
shunt capacitor so that its overall power factor is
leading and reactive power is generated into the
Page No. 69
external system. The effect of adding the capacitor 3.3.1.2. THYRISTOR CONTROLLED SERIES
currents to the TCR currents is to bias the control COMPENSATOR (TCSC) :
characteristics into the second quadrant.
drant. The TCR
The TCSC is a key member of the FACTS family of
equivalent reactance will depends on firing angle of
power electronic controller that provides smooth, rapid
the thyristor.
and continuous adjustment of the transmission line
impedance. In the equivalent circuit representation of
the basic TCSC consists of TCR in parallel with a
fixed capacitor is shown in Fig.3.7. The controlling
elements are the two back to back thyristors,
connected in series with the linear reactor. The TCSC
varies the electrical length of the compensated
transmission line with little delay. Owing to this
characteristic, it may be used to provide fast active
power flow regulation. It also increases the stability
stab
margin of the system and has proved very effective in
damping SSR and power oscillations.
Fig. 3.6.Configuration of FC--TCR
The TCSC has three fundamental modes of operations
πX L as follows:
X TCR =
σ − sin σ • Thyristor – blocked mode
σ = 2 (π − α ) • Thyristor – bypassed mode
XL is the reactance of the linear inductor, and the • Thyristor operating in phase controlled mode
sigma and alpha are the thyristors condu conduction and In thyristor-blocked
blocked mode, the current through
firing angels respectively. The total effective reactance the TCR is zero and the TCSC functions as a
of the SVC including the TCR and capacitive capacitive reactance XC. In thyristor-bypassed
thyristor mode,
reactance, is determined by the parallel combination of the thyristor valves are fired with no delay and the
the both components TCSC has small inductive impedance. When the
X C * X TCR thyristor operates in phase-controlled
controlled mode, the valve
X SVC = of the firing angle determines the direction of the
X C + X TCR current through the TCR and the capacitor, enabling
Which is a function of the conduction angle ((sigma) the TCSC to work as either an inductive reactance. In
becomes this mode the thyristor firing mechanism is controlled
to vary the amount of effective reactance connected to
π * XC * X L the system.
X SVC =
X C (σ − sinσ − π * X L ) The series capacitive compensation is bypassed during
minimum loading in order to avoid transmission line
And finally as a function of the firing angle (alpha) over voltages resulting from excessive effects in the
becomes system. Conversely, series compensation
comp is fully
utilized during maximum loading.
π * XC * XL
X SVC =
X C (2(π − α ) + sin 2α − π * X L )
The reactive power supplied by the SVC given by the
QSVC =V2BUS / XSVC
Under open loop mode the reactive power supplied by
the SVC will be zero for satisfying this condition the
XSVC should be infinite.
Fig. 3.7. Configuration of TCSC with Fixed capacitor
and TCR

Balaji Prasad Padhi, Dr.Deba Prasad Das Page No. 70

3.3.2. CUSTOM POWER DEVICES:
3.3.2.1. STATIC TRANSFER SWITCH:
The static transfer switch (STS) is composed of two
thyristor blocks. Each thyristor block is composed of
three thyristor modules corresponding to the three
phases of the system. The common configuration of
the STS in industrial distribution systems is shown in
Fig.3.8. In normal operating conditions. The primary
supply feeds the load through switch 1. In case of a
fault or a voltage dip affecting the primary supply, the
load is fed from the alternative source switch 2. The
transfer time of the STS ranges from ¼ to ½ cycle of
the fundamental frequency. Thus, the duration of the
voltage dip is reduced to this transfer time, which most
of the loads can survive. The basic shortcoming of the
STS is that it continuously conducts the load current,
which leads to a considerable conducting loss,
particularly in high power applications. The
conducting losses are in the range of 1% of the load
power. Another disadvantage is that the non-faulted
feeder experiences a ¼ to ½ cycle voltage notch.
Fig. 3.8. Model of Static Transfer Switch
A hybrid static transfer switch (HSTS) has been
proposed in. The idea is that a conventional
mechanical circuit breaker (MCB), connected in
parallel with the thyristors, and conducts the current in
normal operation conditions. The current commutation
from the primary to the alternative supply takes less
than half a cycle but the total transfer time is also
dependent on the response time of the voltage dip
detection. The voltage dip detection technique should
be very fast to ensure a fast transition of the load from
the primary to the alternative supply. An example of a
voltage detection algorithm is found in. A block
diagram of this voltage dip detection algorithm is
shown in Fig.3.9. The phase-to- phase voltages Uab,
Ubc, Uca, are the inputs to the algorithm and then
transformed to the dq-synchronous reference frame.
The output of the transformation block U(dq) is
compared with the reference voltage, U(dq)ref to obtain
the error signal. The impact of voltage transients on
the error signal U(dq)error a low-pass filter. The output
Balaji Prasad Padhi, Dr.Deba Prasad Das
of the low-pass filter is compared with the tolerance
limit of the phase-to phase voltage U(dq)tol to obtain the
transfer signal of the STS. The successful transfer of
the sensitive load by an STS depends on some factors
such as the independence of the alternative supply and
the primary supply. The capacity of the two feeders,
supply availability. It is of curiosity to mention here
that the STS may not be able to mitigate voltage dips
originating in the transmission system since both
supplies (primary and alternative) can be influenced
by the fault. Another problem with the STS is the
synchronization of the primary supply with the
alternative supply. The phase angle difference between
the two supplies could lead to phase angle jumps at the
load terminals.
Fig. 3.9. Block diagram of voltage dip detection algorithm
3.3.2.2. UNINTERRUPTABLE POWER SUPPLY:
The traditional response to avoid production
interruption and outage costs has been the
uninterruptible power supply (UPS). The load power
is taken from the main power supply through a two-
stage operation: conversion (AC/DC) and inversion
(DCIAC). During a voltage dip or an interruption, the
energy released by the battery maintains the load
voltage constant. A single-line diagram for the UPS is
shown in Fig.3.10. Depending on the storage capacity
of the battery, it can supply the load for minutes or
even hours. UPS are needed where the loss of power
can cause loss greater than the UPS cost. For higher-
power loads the costs associated with losses due to the
two additional conversions and the maintenance of the
batteries become too high and this solution no longer
appears to be economically feasible.
Fig. 3.10. Single diagram of UPS
Page No. 71
3.3.2.3. DISTRIBUTION STATIC
COMPENSATOR (D-STATCOM):
Shunt-connected custom power devices such as the
Distribution Static Compensator (DSTATCOM) are
connected in shunt with the distribution system via a
coupling transformer to maintain the voltage of the
distribution feeder at the required level by supplying
or absorbing reactive power. The DSTATCOM
consists of a voltage source converter (VSC), small dc
energy storage, and an output filter as depicted in Fig.
3.11. The DSTATCOM is based on the principle that
the VSC generates a controllable AC voltage source
behind the transformer leakage reactance so that the
voltage difference across the reactance produces a
reactive power flow between the DSTATCOM and the
distribution system. The first installation of the
DSTATCOM used for reactive power Compensation
to prevent voltage flicker was installed at a timber mill
in British Columbia, Canada. A k 2MVAr
DSTATCOM has been installed for a 2.6 MVA load
with a typical power factor of approximately 0.85. By
choosing energy storage with adequate capacity, the
DSTATCOM can also exchange active power with the
distribution system, which enables the DSTATCOM
to compensate also for voltage dips. However, the
DSTATCOM cannot be used to compensate for deep
dips because in that case it will need to inject a very
large current.
Fig.3.11. Model of D-STATCOM
3.3.2.4. DYNAMIC VOLTAGE RESTORER (DVR) :
The first static series compensator (SSC),
commercially known as the dynamic voltage restorer
(DVR) has been installed on the Duke Power
distribution system to protect a sensitive textile
customer from voltage dips. The SSC is a series
connected custom power device and it is designed to
inject the missing voltage into the distribution line. Its
basic idea is to dynamically inject a voltage uc(t).
Fig.3.12 shows a simplified single-phase equivalent
circuit of a distribution feeder with a SSC. Where the
supply voltage us(t) the SSC-injection voltage uc(t) and
the load voltage uL(t) are in series. So, the SSC is
considered to be an external voltage source where the
amplitude, the frequency and the phase shift of uc(t)
Balaji Prasad Padhi, Dr.Deba Prasad Das
can be controlled. The purpose is to maintain the
amplitude of the load voltage fixed and prevent phase
jumps. From Fig. 3.12 The load voltage is:
uL(t) = ug(t) +uC(t)
Fig. 3.12. Single line diagram of DVR connected to the line
If the supply voltage uS(t) has dropped due to a voltage
dip or increased due to a voltage swell, the SSC
compensating voltage uc(t) should be chosen so that to
the load voltage remains the same as during no-
disturbance conditions. Thus the instantaneous
amplitude of uC(t) is controlled such as to eliminate
any detrimental effects of a system fault to the load
voltage as long as the disturbance does not cause the
circuit breaker to open. Voltage sags and swells are a
major problem in present distribution systems.
Dynamic Voltage Restorer (DVR) is a one of the
solution to realize this goal. Voltage sag/swell
compensation requires energy storage, the amount of
which depends greatly on the required voltage level
(VDVR). Present control strategies are either able to
minimize the needed voltage, to allow a better
utilization of the storage system or to minimize the
distortions at the load. To avoid this drawback an
optimized control strategy is used, which is able to
reduce the voltage amplitude and causes low
distortions at the load side.
The Dynamic Voltage Restorer (DVR) is the one of
the custom power devices, which has excellent
dynamic capabilities, and it is well suited to protect
critical or sensitive loads from short duration voltage
dips or swell. The basic principal of operation of DVR
is shown in Fig. 4.1.
Fig. 4.1.Basic DVR concept
Page No. 72
4.1. BLOCK DIAGRAM OF DYNAMIC
VOLTAGE RESTORER:
The block diagram of a DVR and different
components are shown in Fig.4.2. The DVR
components are the VSC, the modulation unit, the
control unit, the output filter. The injection
transformer, the energy storage and the bypass switch.
Measured voltages and currents are the inputs to the
disturbance identification. which gives signals to the
control unit to function when the measured quantities
differ from settings of the control user. The
disturbance identification module triggers the start of
voltage dip compensation when the supply voltage
comes out of a pre-defined range. Then the control
unit generates the voltage references and synthesis
them with supply voltage. The voltage references are
inputs to the modulation unit to generate the
modulating signals for the valves of the VSC. The
energy storage provides the required power to
compensate the identified voltage dip. Installing an
output filter reduces the dv/dt effect on the windings
of the Injection transformer and it is necessary to
convert the pulse modulated voltage of the VSC into a
sinusoidal voltage. The filtered voltage is injected into
the distribution system by the series-injecting
transformer.
Fig. 4.2. Block diagram of DVR
4.2. DIFFERENT CONTROL STRATEGIES OF
DVR:
A DVR must be able to react very fast on different
kinds of voltage sags and swells. The amplitude of the
load-side voltage must be restored and for most loads
large phase jumps must be avoided. Especially, the
correct compensation of single phase and three phase
voltage sags & swells is a major issue. The three basic
control strategies are
• Pre-Sag Compensation
• In-Phase-Compensation and
• The Energy-Optimized Compensation.
Balaji Prasad Padhi, Dr.Deba Prasad Das
To avoid a loss of power supply, the amplitude of the
load voltage has to be restored by the DVR. Therefore,
different strategies can be used to achieve this goal.
4.2.1.Pre-sag Compensation:
The standard solution for compensating voltage sags
is to re-establish the exact voltage before the sag.
Therefore, the amplitude and the phase of the voltage
before the sag have to be exactly restored. The
resulting vector is shown in the Fig.4.3. This
compensation leads to the lowest distortions at the
load, because the voltage at the load is not changed
due to the sag. For this strategy, a PLL will be
synchronized with the load voltage. As soon as the
failure occurs, the PLL will be locked and therefore
the phase angle can be restored. Depending on the
phase of the new grid voltage, the DVR has to deliver
higher voltage amplitude than needed in order to
restore the correct voltage magnitude. Therefore, the
system has to be designed for a higher maximum
voltage (VDVR) and less energy from the DC-Link can
be used.
Fig. 4.3. Vector diagram of Pre-sag compensation
4.2.2. In-Phase Compensation:
As already mentioned, the Pre-Sag Compensation does
not lead to minimized voltage amplitude. This can be
realized with the In-Phase Strategy, which is designed
to control the DVR with a minimum output voltage. In
Fig.4.4 the voltages for this strategy are depicted. In
contrast to the Pre-Sag version, the voltage is now
compensated in phase to the grid voltage after the sag.
Hence, the needed voltage amplitude is minimized. In
most cases, voltage sag leads to a phase jump,
therefore the distortions due to phase changes are not
minimized. As a consequence, a phase jump will occur
at the load, leading to transient currents. If a sensible
load is secured, then the In-Phase compensation
cannot be used, because it would lead to a loss of
power supply. To realize this strategy, the PLL has to
be synchronized to the grid voltage itself and therefore
will not be locked during the compensation.
Page No. 73
 sliding mode controller for DVR. As a thumb rule, the
designer has to perform the following three main tasks
for controller design.
• Sliding surface selection
• Check for existence of sliding mode and reaching
condition
• Determination of control law
Fig. 4.4 Vector diagram of In-Phase compensation
4.2.3. Energy Optimized Compensation:
Another existing control strategy is to use as much
reactive power as possible to compensate the sag.
Therefore, the DVR voltage is controlled in such a
way that the load current is in phase with the grid Fig. 4.6. Block diagram of DVR with sliding mode control
voltage after the sag. As long as the voltage sag is The state space model of DVR with reference to Fig.
quite shallow, it is possible to compensate sag with 4.6 can be written as
pure reactive power and therefore the Compensation
time is not limited. In this Fig.4.5 the voltages for the  0 1  − 1 0 
d v c   C f  v c   C f   is 
energy-optimized compensation are depicted. Beside   =  i  + 
dt  i f   − 1 − Rf 0 1   δ (t )V dc 
 L f   f 
 L f 
the enormous advantage of not requiring active Power,  Lf
this strategy has in most cases two major
disadvantages. On the one hand, the phase distortions
and on the other hand the needed voltage amplitude Where If and Is are filter inductor current and source
are quite high. Furthermore, the compensation with current respectively. δ(t) is the switching function of
pure reactive power is only possible for shallow sags. the inverter that can be 1 or –1. In order to control the
If deep sag occurs, active power is needed and in an output voltage of inverter we have to find out a
extreme case it becomes equal to the Pre-Sag suitable sliding surface which will directly affected by
Compensation. switching law. It is seen that the first time derivative
of the output (dvC/dt) = (if - is)/ Cf = θ, does not
explicitly contain the control output δ(t)Vdc, therefore
the second derivative must be calculated and is given
as
.  θ 
v.c  =  − R f 1 Rf 1 dis 1 
θ   L θ − v c − i s − + δ (t )V dc 
   f L f C f L f C f C f dt L f C f 
The phase canonical form is shown in above equation
Fig. 4.5. Vector diagram of Energy optimized compensation
and the second derivative of the output depends on the
control input δ(t)Vdc . No further time derivative is
4.3. OPERATION OF DVR WITH SLIDING needed. Considering that, evc = vCref –vC, a sliding
MODE CONTROL: surface can be chosen:
The DVR with sliding mode control, which eliminates
devc
the use of additional sag/swell detection for DVR.
This improves the dynamic response of the DVR and
( )
S evc , t = k1evc + k2
dt
=0
also DVR is able to compensate for any variation in
4.4. Control strategy for DVR control:
supply voltage. A sliding mode controller is designed
The Fig.4.7 shows the controller block
and applied to DVR. The schematic block diagram of
diagram for DVR. The reference voltage to be injected
conventional DVR is shown in Fig.4.6. It mainly
by the DVR is calculated by subtracting the reference
consists of reference voltage calculator, sliding mode
source voltage and actual source voltage. The
controller, DC voltage source, voltage source
reference source is obtained by multiplying the unit
converter and LC filter. Output of LC filter is
sine template with the peak of the reference source.
connected to the system through a transformer. This
The sine template is generated from Phase Locked
section develops a model for the development of
Loop (PLL). After generating the reference voltage for

Balaji Prasad Padhi, Dr.Deba Prasad Das Page No. 74

DVR it is subtracted. From thee actual DVR injected 6. MATLAB/SIMULINK MODEL OF SINGLE
voltage and the error is processed through sliding PHASE DVR :
mode controller. The output of sliding mode controller
is used to switch the inverter.

Fig.6.1. MATLAB/SIMULINK
B/SIMULINK model of Single Phase DVR
6.1. Block diagram of Single Phase DVR
Controller:

Fig. 4.7. Control block diagram of DVR

5. SIMULATION AND RESULTS

ESULTS Fig. 6.2 Block diagram of DVR controller
DESIGN OF SINGLE PHASE DVR
DVR: 6.2. Model of Single Phase Filter:
Design specifications of sliding mode control of DVR:
Source voltage

• 120 V rms (ph-n); single phase

Injection transformer rating:
• VA rating = 500 VA; single phase
• Winding 1 and 2 parameters
• Vrms = 120/sqrt(3); single phase
R = 0.002 Fig. 6.3 .Single Phase Filter
X = 0.08
• Magnetizing parameters Rm = 500 7. SIMULATION RESULTS OF SINGLE
Xm = 500 PHASE DVR:
• Winding t/f ratio( K ) = 1:1 7.1. For Voltage Sag:
Active filter:
• Inductance (L) = 40 mH In Fig. 7.1 a Voltage Dip of 30% of the supply voltage
is applied between 80ms to 160ms. Fig.7.2 shows the
• Capacitance (c) = 40 µF
voltage injected by the DVR and Fig.7.3 shows the
Dc Link Voltage:
compensation load voltage.
• 200 V; single phase
Load:
• Resistance (R) = 50 ohm; single phase
• Inductance (L) = 50 mH
• Resistance (R) = 100 ohm; three phase
• Inductance (L) = 50 mH

PD-CONTROLLER:

• KP = 0.1
• KD = 0.00006
Fig. 7.2.Voltage Injected by the DVR

Balaji Prasad Padhi, Dr.Deba Prasad Das Page No. 75

 Fig. 7.3 Load voltage after compensation
Fig. 7.4.Source, Injected voltage of DVR and
Compensated Load voltage waveforms.
7.2. Simulation Results for Voltage Swell:
In Fig. 7.5 a voltage swell of 30% of the supply
voltage is applied between 80ms to 160ms. Fig.7.6
shows the voltage injected by the DVR. Fig.7.7 shows
the compensated load voltage.
Fig. 7.5.Source Voltage under voltage swell condition
Fig. 7.6.Voltage Injected by the DVR
Fig. 7.7.Load Voltage after compensation
Balaji Prasad Padhi, Dr.Deba Prasad Das
Fig. 7.8.Source, Injected Voltage and Compensated Load Voltages
8. CONCLUSION:
The main problems for both industrial and domestic
customers are voltage sags/swell problems. These problems
are alleviated using FACTS and Custom Power Devices. In
this project a single phase DVR using sliding mode control
[3] is verified through MATTLAB/SIMULINK simulation
results are verified and seen to be satisfactory.
REFERENCES
[1]. N.G. Hingorani “INTRODUCING CUSTOM POWER,”
IEEE spectrum, vol.32, June 1995, pp. 41-48.
[2]. Bollen, M.H.J.,” VOLTAGE SAGS IN THREE-PHASE
SYSTEMS” Power Engineering Review, IEEE, Volume:
21, Issue: 9, Sept. 2001, pp: 8 - 11, 15.
[3]. Haque, M.H.”COMPENSATION OF DISTRIBUTION
SYSTEM VOLTAGE SAG BY DVR AND D-
STATCOM”, IEEE Porto Power Tech Conference,
Volume: 1, 10th-13th Sept. 2001.
[4]. K. Chan, A. Kara, and G. Kieboom, “POWER QUALITY
IMPROVEMENT WITH SOLID STATE TRANSFER
SWITCHES,” in Proc. 8th ICHQP ’98, Athens, Greece,
Oct. 1998, pp. 210–215. [11]. Haque, M.H.” Compensation
of distribution system voltage sag by DVR and D-
STATCOM”, IEEE Porto Power Tech Conference,
Volume: 1, 10th-13th Sept. 2001.
[5]. TEQSIM International Inc., “POWER SYSTEM
BLOCKSET USER'S GUIDE”, 2001
[6]. B. Singh, K. Al-Haddad, and A. Chandra, “A REVIEW OF
ACTIVE FILTERS FOR POWER QUALITY
IMPROVEMENT,” IEEE Trans. Ind. Electron., vol.46,
Oct. 1999, pp: 960–971.
Page No. 76
Ant Colony Search Based Minimum Losses Reconfiguration
for Distribution Systems
Swamy Naidu Yerra1, Dr. G V Nagesh Kumar2

Assistant Professor, Dept. of EEE, Miracle Educational Society Group of Institutions, Vizianagaram, India, swamyyerra@gmail.com1
Assistant Professor , Dept. of EEE, GITAM University, Visakhapatnam, India, Gundavarapu_kumar@yahoo.com2

Swamy Naidu Yerra, Dr. G V Nagesh Kumar Page No. 77

Swamy Naidu Yerra, Dr. G V Nagesh Kumar Page No. 78
Swamy Naidu Yerra, Dr. G V Nagesh Kumar Page No. 79
Swamy Naidu Yerra, Dr. G V Nagesh Kumar Page No. 80
Swamy Naidu Yerra, Dr. G V Nagesh Kumar Page No. 81
Swamy Naidu Yerra, Dr. G V Nagesh Kumar Page No. 82
 Intelligent Techniques for Optimization
S. K. Satapathy, Prof, Dept. of CSE, GEC, Ahemadabad, Gujurat.
Dr. S. K. Nayak, Prof, Dept. of Electronics Science, Berhampur University, Orissa

ABSTRACT:
Optimization techniques form an important part in many of the power system studies. They are used in planning
operation and control, both in on-line and off line applications. Conventional optimization methods have been
widely used and exploited. However they are associated with several drawbacks like for instance they are either time
consuming or do not provide an optimal solution. New optimization techniques have been used which overcome the
limitations of the conventional methods. These methods are termed ‘intelligent’ as they use heuristics to guide the
search and arrive at an optimum solution. The theory of intelligent optimization techniques, namely, Evolutionary
programming, Genetic Algorithms, Particle Swarm Optimization, Ant Colony Optimization and Fuzzy systems are
introduced. Applications and case studies to Economic Dispatch and Optimal Power Flow are described and results
presented.
Keywords: Optimization, Heuristics, Search Strategies, Population, Artificial Intelligence (AI), power systems,
economic dispatch, optimal power flow.

1 INTRODUCTION economic dispatching, and optimal power flow are

In the real world, it is not an easy task to find a useful
optimal solution of a given problem because many
constraints and limitations must be taken into account
during this process. Usually, only the most important
constraints and limitations are chosen to be used
during the solution search process. Another problem is
that the solution may not be unique and it depends
directly on the weight of each constraint. Hence, many
optimization processes have been developed in the last
decades to achieve the best solution in this search
process. In addition, one of the most important points
in this process is the computational cost to find the
best solution. Sometimes, this cost may comprise the
use of a technique to solve a problem. Number of
constraints, number of variables (involved to describe
the problem) and poor convergence speed are some
examples of computational cost. In other words, the
computational problems are related to hardware
processor speed, memory capacity and numerical
techniques. However, the highly fast evolution of the
computational world (hardware and software) allows
optimization techniques, that could not be used before
to solve a specific problem, to be applied successfully
now. Specifically for power system problems,
decomposition techniques, partitioning techniques and
parallel processing are examples of recent evolution of
computational techniques.
In many power system problems, the use of
optimization techniques has been important to reduce
costs and losses of the system. Unit commitment,
some areas where these techniques have been
extensively used. For example, minimization of active
power losses is one of the biggest challenges for
power control operators.
The achievement of this goal in real-time is a critical
task. A possible solution for this problem is to use the
Dantzig and Wolfe decomposition algorithm to
partitioning the power system in many subsystems
according to a geographic basis. The optimization
process is applied to these subsystems, and the
constraints are limited to local constraints and
coupling bus constraints. Conventional search
techniques such as hill climbing are often incapable of
optimizing non-linear multi-modal functions. In such
cases, a random search method might be required.
Power system optimization is widely used in the
operation, planning and control of power systems. The
role of optimization is to minimize an objective or a
set of objectives subjected to several equality and
inequality constraints. Most of the power system
optimization problems are characterized by non-linear
objective functions and non-linear constraints. Since it
is difficult to obtain the solution of such and complex
problem, the objective functions and constrained are
linearized (smoothened) and the solution obtained
through linear programming methods. In the process
of linearizing the problem, the originality is lost and
the solution obtained is only an approximate of the
actual solution. Advanced methods, termed as
‘intelligent optimization methods’ are widely being

S. K. Satapathy, Dr. S. K. Nayak Page No. 83

used in several disciplines in engineering and science.
They are termed as they use several information, like a
population of search points, heuristics, search
direction, etc. which help achieve the best optimal
solution, without sacrificing the non-linearity of the
problem. These intelligent optimization methods are
classified as AI methods and Evolutionary
Programming methods, even though both come under
the same category.
Artificial Intelligence (AI) Methods of optimization
can be further classified as
1. Tabu Search.
2. Simulated Annealing.
3. Neural Networks.
4. Fuzzy Systems.
Evolutionary Programming methods are classified as
1. Evolutionary Computation.
2. Genetic Algorithms.
3. Particle Swarm Optimization.
4. Ant Colony Optimization.
1.2 AI BASED OPTIMIZATION METHODS.
Tabu Search:
Tabu search is a heuristic procedure that
employs dynamically generated constraints or tabus to
guide the search for optimum solutions. The tabu
search algorithm was developed independently by
Glover and Hansen for solving combinatorial
optimization problems. It is a kind of iterative search
and is characterized by the use of a flexible memory. It
is able to eliminate local minima and to search areas
beyond a local minimum. Therefore it has the ability
to find the global minimum of a multi-modal search
space. The process by which tabu search overcomes
the local optimality problem is based on an evaluation
function that chooses the highest evaluation solution at
each iteration. This means moving to the best
admissible solution in the neighbourhood of the
current solution in terms of the objective value and
tabu restrictions.
Simulated Annealing:
Simulated Annealing finds optima in a way analogous
to the reaching of minimum energy configurations in
metal annealing. Simulated annealing algorithm was
proposed and developed by Kirkpatrick in 1983.It is
based on the analogy between the solids and the
S. K. Satapathy, Dr. S. K. Nayak
problem of solving combinatorial optimization
problems. In the analogy between the combinatorial
optimization problem and the annealing process, the
states of the solid represent feasible solutions of the
optimization problem, the energies of the states
correspond to the values of the objective function
computed from these solutions, the minimum energy
state corresponds to the optimal solution to the
problem and rapid quenching can be viewed as local
optimization.
The algorithm consists of a sequence of iterations.
Each iteration consists of randomly changing the
current solution to create a new solution in the
neighborhood of the current solution. Once a new
solution is created, the corresponding change in the
cost function is computed to decide whether the newly
produced solution can be accepted as the current
solution. If the change in the cost function is negative,
the newly produced solution is directly taken as the
current solution; otherwise it is accepted according to
the Metropolis criterion based on Boltzman’s
probability.
Neural Networks:
Neural Networks which are computational models of
the brain are used for optimization by exploiting their
inherent ability to evolve in the direction of the
negative gradient of an energy function and to reach a
stable minimum of that function. Neural Networks are
modeled on the mechanism of the brain and have a
parallel distributed information processing structure.
The major abilities of the neural networks are their
ability to learn from examples and their tolerance to
noise and damage to their components. Optimization
algorithms can be solved by utilizing neural networks
that are designed inherently to minimize and energy
function. Hopfield neural network, Kohonen’s self
organizing neural network and Multi-Layer Perceptron
are widely used for optimization. In these networks,
the objective functions and constraints of the
optimization problem are coded into the energy
function of the network, which aims to reach a stable
state where its outputs yield the desired optimum
parameters.
Fuzzy Systems:
Fuzzy Systems use the uncertainty or impreciseness in
a problem to optimization the problem using
membership functions and fuzzy rules to describe the
behaviour of the system. They are widely used for
single as well as multi-objective optimization. Both
neural networks and fuzzy systems are termed as
Page No. 84
‘model-free estimators’, since they do not require the
mathematical model of the problem to be optimized.
Due to this, complex non-linear optimization problems
can be solved using these techniques
Fuzzy logic implements human experiences and
preferences via membership functions and fuzzy rules.
Fuzzy membership functions can have different shapes
depending on the designer's preference and/or
experience.
Fuzzy Optimization
Usually, optimization problems with a single-real
variable are solved using bisection methods, where the
main idea is to reduce an initial interval until a
required minimum. Different from the classical
optimization methods, the main idea in fuzzy
optimization is to optimize objective function and
constraints, simultaneously. In order to determine the
optimal point (solution point), both objective function
and constraints must be characterize by membership
functions and they must be linked by a linguistic
conjunction: “and” (for maximization) and “or” (for
minimization). Fuzzy optimization was proposed by
Bellman and Zadeh ,and the main idea is to satisfy a
fuzzy objective function and fuzzy constraints that
receive the same treatment, i.e., there is no difference
among the objective function and constraints. Another
usual procedure is the use of fuzzy numbers to define
constraints.
In classical optimization, intervals define the region to
be explored. In fuzzy optimization, this region can be
expressed using fuzzy numbers.
Fuzzy Programming
Fuzzy Linear Programming
Classical linear programming can be defined by an
optimization of a linear objective function and linear
constraints. The fuzzy linear programming has the
same structure as the classical linear programming.
The difference between these two approaches is that in
the classical approach values and operators are crisp,
while, in the fuzzy approach values and/or operators
may assume fuzzy characteristics.
Fuzzy Dynamic Programming
The idea in classical dynamic programming is to
decompose a main problem into several sub-problems
(one for each variable). Thus, the optimization of each
sub-problem is divided into a multistage decision
process. Here, all operators and values have a crisp
meaning. In the same way, a fuzzy dynamic
S. K. Satapathy, Dr. S. K. Nayak
programming can be defined as a fuzzification of all
(or part of) these elements. In a well known fuzzy
dynamic programming method, Bellman and Zadeh
have proposed to work with fuzzy constraints and
fuzzy goals to determine the sub-goals of each step of
the process, while the transformation function is
maintained crisp. Fuzzy dynamic programming is
widely used in power systems in the area f unit
commitment and generation scheduling.
Fuzzy Multi-Criteria Analysis
Description of the Problem
During any decision making process, many different
factors must be taken into account. These factors can
be heuristic or arising from numerical analysis.
Usually, the heuristic factors are due to the planner's
previous experience and have a non-numerical
structure, i.e., they can be better expressed by
linguistic values.
The problem that planners face in their daily
job is how to incorporate these linguistic values into
numerical analysis. Commonly, the computational
packages do not include the possibility to use non-
numerical values. Thus, planners have two
possibilities when using this kind of knowledge. One
is to put the linguistic numbers in numbers. The other
possibility is to forget this knowledge during the
numerical analysis and then, after getting the final
result, modify it so as to make an adaptation to take
into account the planner expertise.The problem is that
both approaches are not good. In the first one, where
planner tries to transform linguistic knowledge into
numerical values, much information is lost during this
process.
Classification of Fuzzy Multi-Criteria Analysis
The classification of fuzzy multi-criteria problems is
divided in two main types: multi-objective decision-
making and multi-attribute decision-making. In
general, the difference between these two approaches
is located in the decision space. For the former
approach, this space is continuous, and the problem is
solved by mathematical programming. For the latter
approach, the decision space is discrete, and other
approaches have been developed. The next subsection
presents an algorithm to treat this problem.
Presentation of a Multi-Attribute Decision-Making
Algorithm
Evolutionary Programming methods
Evolutionary Computation.
Evolutionary Computation
Genetic Algorithms.
Particle Swarm Optimization.
Ant Colony Optimization.
Page No. 85
1.3 EVOLUTIONARY COMPUTATION (EC)
1.3.1 INTRODUCTION
Evolutionary Algorithms mimic the process of organic
evolution, the driving process for the emergence of
complex and well-adapted organic structures. At a
simplified level, evolution can be seen as the result of
the interplay between the creation of new genetic
information and its evaluation and selection. A single
individual of a population is affected by other
individuals in the population (e.g., by competition,
mating etc.,) as well as by the environment (e.g., by
food supply and climate). The better an individual
performs under these conditions the greater its chance
to live for a longer while and generate offspring,
which in turn inherit the (disturbed) parental genetic
information. The non-deterministic nature of variation
leads to a permanent production of novel genetic
information and therefore the creation of different
offspring.
Evolutionary Computation (EC) paradigms differ from
traditional search and optimization paradigms in three
main ways:
• EC paradigms utilize a population of points in
their search.
• EC paradigms use direct “fitness” information
instead of function derivatives or other related
knowledge.
• EC methods use probabilistic, stochastic rather
than deterministic, transition rules.
1.3.2 EVOLUTIONARY COMPUTATION (EC)
PARADIGM ATTRIBUTES
Most traditional optimization methods move from one
point in the hyperspace to another, using some
deterministic rule. One of the drawbacks of this
approach is the likelihood of getting stuck at a local
optimum. EC methods, on the other hand, start with a
population of points. Thus many maxima and minima
can be explored simultaneously, lowering the
probability of getting stuck. Operators such as
crossover and mutation effectively enhance this
parallel search capability, allowing the search to
directly “tunnel through” from one promising
hyperspace region to another.
EC paradigms do not require information of
the problem, such as function derivatives. The fitness
of each member of the population is calculated from
the value of the function being optimized, and it is
common to use the function output as the measure of
S. K. Satapathy, Dr. S. K. Nayak
fitness. Fitness is the direct metric of the performance
of the individual population member on the function
being optimized.
The fact that EC methods use probabilistic
transition rules certainly doesn’t mean that a strictly
random search is being carried out. Rather stochastic
operators are applied to operations that direct the
search toward regions of the hyperspace that are likely
to have higher values of fitness.
I.3.3 IMPLEMENTATION
Regardless of the paradigm implemented,
Evolutionary Algorithms often follow a similar
procedure:
1) Initialize the population,
2) Calculate the fitness for each individual in the
population,
3) Reproduce selected individuals to form a new
population,
4) Perform evolutionary operations, such as
crossover and or mutation, on the population,
and
5) Loop to step2 until some condition is met.
The total number of individuals chosen to make up the
population is both problem and method dependent.
The fitness value is often proportional to the output
value of the function being optimized.
In many, if not most cases, a global optimum exists at
one point in the decision hyperspace. Sometimes the
global optimum changes dynamically because of
external influences; frequently there are very good
local optima as well. For these and other reasons, the
bottom line is that it is often unreasonable to expect
any optimization method to find a global optimum
within a finite time. The best that can be hoped for is
to find near optimal solutions within the specified
time. Put in another way, EC is often the second-best
way to solve a problem. So why should we be
satisfied with the second best? Well for the following
reasons:
1) Classical and customized approaches will
frequently not be feasible, and EC paradigms will
be usable in a vast number of situations.
2) EC methods are robust, i.e., the algorithm can be
used to solve many problems, with minimum
amount of special adjustments to account for
special qualities of a particular problem. Robust
methodologies are fast and easy to implement.
Page No. 86
1.3.4 GENETIC ALGORITHMS (GA)
Genetic Algorithms (GA) was introduced by Holland
in the 1970s not to solve a particular problem but to
investigate the effects of natural adaptation in
stochastic search algorithms. GA consists of a
population of possible problem solutions that get
refined over time through selection, recombination and
mutation. Traditionally these were binary encoded but
now real-value encoded GA are just as frequent. In
GA the mutation operator is the source of the random
variation and recombination happening to investigate
intermediate solutions. The selection is the primary
operator that drives the GA individuals towards
optimality. For example, if one knows that certain
genes are correlated this can be incorporated in the
recombination operator to enhance the search.
1.3.5 OPERATORS
SELECTION
Selection plays a central role in Genetic Algorithms
determining how individuals compete for gene
survival. Selection weeds out the bad solutions and
keeps the good ones. This can be done by fitness
proportional selection that assigns a selection
probability in proportion to the fitness of the given
individual. This however tends to be sub-optimal as
the effective selection strength can be changed by
adding an offset. More commonly used is tournament
selection, where a number of randomly picked
individuals are compared to each other. The individual
with the best fitness is then selected to be a part of the
next generation. The tournament size determines how
many individuals are to be compared per selected
individual. Selection in GA is usually done on the
whole original population and usually repeated for all
individuals in the population. This typical kind of
selection allows for an individual to be selected
several times and thus results in a loss of diversity.
Because of the randomness in selection most
techniques can not guarantee survival of the current
best solution. Neither can most traditional
recombination or mutation operators. Elitism provides
this guarantee by explicitly selecting the best
individual or group of individuals. Both the typical
selection and the additional elitism technique can lead
to duplicates, however of good individuals. Population
diversity decreases with duplicates, but the search can
benefit when it comes to the recombination of
individuals with good genes.
S. K. Satapathy, Dr. S. K. Nayak
RECOMBINATION
Recombination of individuals is done to investigate
the performance of new individuals that resemble
exiting ones. This is done on the genotype level of the
individuals and leads to the construction of new
intermediate solutions.
The notion of generations arises as parent individuals
recombine their genes to create offspring. Usually the
parents are removed to make room for the offspring
carrying some of their genes. Recombination is often
done by crossover, creating two offspring individuals
from two parent individuals by giving each offspring a
part of each parental gene. For binary encoded GA one
can use 1-point crossover, which creates one of two
offspring individuals by copying the gene of the first
parent up to a random point and then copying the gene
of the second parent. Vice versa is done for the second
offspring individual. This technique has been
generalized creating the n-point crossover, with 1-
point crossover being most frequently used. For real
encoded GA one often uses arithmetic crossover that
per genome entry assigns a weighted arithmetic mean
of each parent to each offspring. The crossover
probability is usually a variable GA parameter. The
recombination operator used in the Canonical Genetic
Algorithm is called Single Point Crossover.
Individuals are paired at random with a high
probability that Crossover will take place. In
affirmative case, a Crossover point is selected at
random and, say, the rightmost segments of each
individual are exchanged to produce two offspring as
illustrated in the next figure, where two 7-bit
chromosomes A and B exchange parts
(The Crossover point is p=4) resulting in the
chromosomes A’ and B’:
MUTATION
Mutation happens infrequently both in nature
and in GA to the genome of an individual altering it
slightly. Mutation can maintain population diversity
and contribute to the exploration of the search space.
Exploration by mutation is often slow compared to
recombination, but in problems where recombination
is disruptive mutation can be an important way to
explore the landscape. Even if most of the search is
being performed by recombination, mutation can be
vital to provide the diversity which recombination
needs. The probability of mutation is usually a
variable GA parameter. The figure 3.2.1.3 illustrates
the mutation operator on B, resulting in B”.
Page No. 87
1.3.6 POSITIONING WITHIN OPTIMIZATION THEORY every iteration. This is done by adding the velocity
The Flow chart for Genetic Algorithm is presented in
figure 3. GA has certain characteristics that
differentiate them from traditional optimization
methods. Here's a summary of the differences:
• GA code parameters in a bit string and not in the
values of the parameters. The meaning of the bits
is completely transparent for the GA.
• GA search from a population of points and not
from a single point.
• GA uses only the fitness function and don't need
knowledge about derivatives or problem structure.
• GA use transition probabilistic rules
(represented by the operators Selection, Crossover
and Mutation) instead of deterministic rules
1.7 PARTICLE SWARM OPTIMIZATION
(PSO)
Kennedy and Eberhart introduced Particle Swarm
Optimization (PSO) in 1995 as an alternative to
Genetic Algorithms. The PSO technique has ever since
turned out to be a competitor in the field of numerical
optimization.
1.7.1 THE PARTICLE SWARM OPTIMIZATION MODEL
A PSO consists of a number of individuals refining
their knowledge of the given search space. PSO’s are
inspired by particles moving around in the search
space. The individuals in a PSO thus have a position, a
velocity and are denoted particles. The PSO
traditionally has no crossover between individuals, has
no mutation and particles are never substituted by
other individuals during the run. Instead the PSO
refines its search by attracting the particles to positions
with good solutions. The PSO remembers the best
position found by any particle. Additionally each
particle remembers its own previously best found
position. With particle i having position x each particle
have its velocity vi updated each iteration in the
following way:

vi = (wvi +ϕ1i ( pi − xi ) +ϕ2i ( pg − xi ))

where w is inertia weight, p i is the best found
Fig 4. Concept of Modification of a search point
position for the given particle and pg is the best
position known for all particles. ϕ 1 and ϕ 2 are random vector, v i to the position vector, x i i.e.,
values different for each particle and for each
dimension. The position of each particle is updated

S. K. Satapathy, Dr. S. K. Nayak Page No. 88


Fig. 6 Flow chart of Particle Swarm Optimization
xi = xi + vi
Settings for the velocity parameters determine the
performance of the PSO to a large extent. The random
values ϕ 1 and ϕ 2 are usually set to be upper bounded
by 2.0.
Figure 6 illustrates the PSO algorithm. PSO utilizes
several searching points like Genetic Algorithm (GA)
and the searching points gradually get close to the
global optimal point using its p i and pg . The features
of the searching procedure can be summarized as
follows:
pg
(1) Initial positions of p i and are different.
However, using the different direction of p i and pg ,
all agents gradually get close to the global optimum.
(2) The modified value of the agent position is
continuous and the method can be applied to the
continuous problem. However, the method can be
applied to the discrete problem using grids for XY
position and its velocity.
(3) There is no inconsistency in searching procedures
even if continuous and discrete state variables are
S. K. Satapathy, Dr. S. K. Nayak
utilized with continuous axes and grids for XY
positions and velocities. Namely, the method can be
applied to mixed integer nonlinear optimization
problems with continuous and discrete state variables
naturally and easily.
(4) The above concept is explained using only XY axis
(2 dimensional spaces). However, the method can be
easily applied to n dimensional problem.
1.7.2 CONSTRICTION FACTOR APPROACH
The basic system equation of PSO can be considered
as a kind of difference equations. Therefore, the
system dynamics, namely, search procedure, can be
analyzed by the eigenvalue analysis. By analyzing the
eigenvalues of simplified equations the velocity of
constriction factor approach (CFA) can be expressed
as follows:
vi = χ(wvi +ϕ1i ( pi − xi ) +ϕ2i ( pg − xi ))
Where χ is known as the constriction coefficient is:
2k
χ= ,η = ϕ 1 + ϕ 2 ,η ≥ 4
2 − η + η 2 − 4η
χ = k , η< 4
For e.g., if η= 4.1, then χ = 0.73. As ηincreases
above 4, χ gets smaller. The convergence
characteristics of the system can be controlled byη.
Figure 5 shows the variation of the constriction factor
χ as a function ofη. Constriction Factor Approach
can generate higher quality solutions than PSO. The
system behavior has the following features:
1) The system does not diverge and finally can
converge.
2) The system can search different regions
efficiently by avoiding premature convergence.
Constriction Factor Approach can generate higher
quality solutions than PSO. The system behavior has
the following features:
3) The system does not diverge and finally can
converge.
4) The system can search different regions
efficiently by avoiding premature convergence.
1.8 EVOLUTIONARY PROGRAMMING (EP)
Evolutionary Programming (EP), originally conceived
by Lawrence J. Fogel in 1960, is a stochastic
Optimization strategy similar to GA. (EP) is a
computational intelligence method in which an
optimization algorithm is the main engine for the
process of three steps; namely, natural selection,
mutation and competition. It is a stochastic
Page No. 89
optimization strategy, which places emphasis on the
behavioral linkage between parents and their
offspring, rather than seeking to emulate specific
genetic operators as in GA’s. EP tends to generate
more effective and efficient searches. It operates on
populations of real values (floating points) that
represent the parameter set of the problem being
solved over some finite ranges. Each representation is
an individual in the EP population. The population is
initialized with random individuals at the start of the
EP run. The EP searches the space of possible real
values for better individuals. The search is guided by
fitness values returned by the environment. This gives
a measure of how well adapted each individual is in
terms of solving the problem, and hence determines its
probability of appearing in future generation. EP uses
two types of rules, named the selection rule and
combination rule in its search. The selection rule is
used to determine the individuals that will be
represented in the next generation. It includes
competition in which each individual in the combined
population has to compete with some other individuals Fig 7. Flow chart of Evolutionary Programming
to get chance to be transcribed to the next generation. 4) The mutation operator is implemented by
On the other had the combination rule operates on means of normally distributed variations
selected individuals to produce new individuals that based on adoptable step sizes.
appear in the next generation. The selection
mechanism is based on a fitness measure or objective 5) Recombination is not used.
function values, defined on each individual in the
population. The combination rule is used to introduce The Flow chart of the EP algorithm is given in figure
new individuals into the current population or to create 6. Evolutionary programming uses the concept of self-
a new population based on the current population. adaptive strategy parameters in combination in real
valued candidate solutions. Again the strategy
The main properties of evolutionary programming
parameters represent a vector of standard deviations of
as refined in the early 90’s can be summarized as
a normal distribution used for mutating the object
follows:
variables.
1) Individuals are represented as real valued
The basic EP method involves 3 steps (Repeat until a
vectors consisting of object variable vectors
threshold for iteration is exceeded or an adequate
plus some additional information.
solution is obtained):
2) No mating selection is used
• Choose an initial POPULATION of trial solutions
3) Parent and offspring population sizes are at random. The number of solutions in a
identical. Selection is implemented as a population is highly relevant to the speed of
randomized process taking parents and off optimization, but no definite answers are available
springs into account. as to how many solutions are appropriate (other
than >1) and how many solutions are just
wasteful.
• Each solution is replicated into a new population.
Each of these offspring solutions are mutated
according to a distribution of MUTATION
types, ranging from minor to extreme with a
continuum of mutation types between. The

S. K. Satapathy, Dr. S. K. Nayak Page No. 90

 severity of mutation is judged on the basis of the
functional change imposed on the parents.
• Computing its fitness assesses each offspring
solution. Typically, a stochastic tournament is
held to determine N solutions to be retained for
the population of solutions, although this is
occasionally performed deterministically. There is
no requirement that the population size is not held
constant, however, or that only a single offspring
is generated from each parent.
1.8.1 EVOLUTIONARY PROGRAMMING (EP)
AND GENETIC ALGORITHM (GA)
There are two important ways in which EP differs from
GA. First, there is no constraint on the representation.
The typical GA approach involves encoding the problem
solutions as a string of representative tokens, the
GENOME. In EP, the representation follows from the
problem. Second, the mutation operation simply changes
aspects of the solution according to a statistical
distribution which weights minor variations in the
behavior of the offspring as highly probable and
substantial variations as increasingly unlikely. Further,
the severity of mutations is often reduced as the global
optimum is approached.
1.9 CONSTRAINED OPTIMIZATION
Any evolutionary computation technique applied to a
particular problem should address the issue of handling
infeasible individuals. The presence of constraints
significantly affects the performance of optimization
algorithm, including evolutionary search methods. The
general way of handling constraints is by penalizing the
infeasible points. Constraint handling in evolutionary
Computation is more or less problem dependent. The
Fig 7. Flowchart for Constrained Optimization.
S. K. Satapathy, Dr. S. K. Nayak
knowledge about linear constraints can be
incorporated into specific operators or a repair
operator can be designed that projects infeasible points
onto feasible ones. Rejection of infeasible individuals
offers a few simplifications of the algorithm: for
example, there is no need to evaluate infeasible
solutions and to compare them with feasible ones.
Quite often the system can reach the optimum solution
easier if it is possible to ``cross'' an infeasible region. It
seems that the penalty components (being constantly
increased through the growing value of the generation
number) change the objective function in a significant
way. Once the population is trapped in a feasible (or
infeasible) local optimum, it may stay there forever.
The Flow chart for Constrained Optimization is given
in figure 7. Constrained numerical optimization is also
based on repair algorithms wherein individuals are
replaced by their repaired versions. Repair algorithms
enjoy a particular popularity in the evolutionary
computation community: For many optimization
problems it is relatively easy to `repair' an infeasible
individual.
The question of replacing repaired
individuals is related to so-called Lamarckian
evolution, which assumes that an individual improves
during its lifetime and that the resulting improvements
are coded back into the chromosome. In many
optimization problems,
an evolutionary computation technique with a
repair algorithm provides the best results when 5% of
repaired individuals replace their infeasible originals.
The method does not require any special parameters
apart from standard parameters for any evolutionary
system (like population size, probabilities of operators,
etc.).
The basic Constrained Optimization method
follows the following procedure:
1) Initialize the population: Starts with a
feasible solution generated by an iterative
procedure by the process of correcting
constraint violations
2) Calculate the fitness for each individual
3) A new population of offspring is produced
from the existing population by mutation or
crossover in such a way that the constraints
are not violated. (This can be done using the
Page No. 91
 repair algorithm - Constraint satisfaction
method or classical optimization method
when an infeasible candidate is obtained)
4) Loop to step2 until some condition is met.
1.10 CONSTRAINED EVOLUTIONARY
PROGRAMMING
The presence of constraints significantly affects the
performance of optimization algorithm, including
evolutionary search methods. Evolutionary
Programming being one of the efficient paradigms for
solving optimization problems when clubbed with
Constrained Optimization based on constraint
handling techniques has proved to be very robust,
efficient and faster at arriving at the global optimum.
The original strategy of an Evolutionary Programming
is preserved. The population, which obtained at the
end of every generation, is tested for constraint
violation. Tournament selection is used as the
selection method for determining the offspring for the
next generation. The general algorithm is given below:
1) Initialize the population: Starts with a feasible
solution generated by an iterative procedure by the
process of correcting constraint violations
2) Calculate the fitness for each individual
3) A new population of offspring is produced from
the existing population by mutation
4) Tournament Selection is conducted between the
parent population and child population and the best
members are selected.
5) Check for constraint violation. If violated, repair
or Loop to 3, else;
6) Loop to step2 until some condition is met
1.11 ANT COLONY OPTIMIZATION
Ant Colony Optimization (ACO) is a recently
proposed meta-heuristic approach for solving hard
combinatorial optimization problems. The inspiring
source of ACO is the pheromone trail laying and
following behavior of real ants, which use pheromones
as a communication medium. In analogy to the
biological example, ACO is based on the indirect
communication of a colony of simple agents, called
(artificial) ants, mediated by (artificial) pheromone
trails. The pheromone trails in ACO serve as
distributed, numerical information, which the ants use
to probabilistically construct solutions to the problem
being solved, and which the ants adapt during the
S. K. Satapathy, Dr. S. K. Nayak
algorithm’s execution to reflect their search
experience. The first example of such an algorithm is
Ant System (AS), which was proposed using as
example application the well-known Traveling
Salesman Problem (TSP). Despite encouraging initial
results, AS could not compete with state-of-the-art
algorithms for the TSP. Nevertheless, it had the
important role of stimulating further research on
algorithmic variants, which obtain much better
computational performance, as well as on applications
to a large variety of different problems. In fact there
exists now a considerable amount of applications
obtaining world class performance on problems like
the quadratic assignment, vehicle routing, sequential
ordering, scheduling, routing in Internet-like networks,
and so on. Motivated by this success, the ACO meta-
heuristic has been proposed as a common framework
for the existing applications and algorithmic variants.
Ant Based Solution Construction(): An ant
constructively builds a solution to the problem by
moving through nodes of the construction graph G.
Ants move by applying a stochastic local decision
policy that makes use of the pheromone values and the
heuristic values on components and/or connections of
the construction graph. While moving, the ant keeps in
memory the partial solution it has built in terms of the
path it was walking on the construction graph.
PheromoneUpdate(): When adding a component ci to
the current partial solution, an ant can update the
values of the pheromone trails that where used for this
construction step. This kind of pheromone update is
called online step-by-step pheromone update. Once an
ant has built a solution, it can (by using its memory)
retrace the same path backward and update the
pheromone trails of the used components and/or
connections according to the quality of the solution it
has built. This is called online delayed pheromone
update. Another important concept in Ant Colony
Optimization is pheromone evaporation. Pheromone
evaporation is the process by means of which the
pheromone trail intensity on the components decreases
over time. From a practical point of view, pheromone
evaporation is needed to avoid a too rapid convergence
of the algorithm toward a sub-optimal region. It
implements a useful form of forgetting, favoring the
exploration of new areas in the search space.
DaemonActions(): Daemon actions can be used to
implement centralized actions which cannot be
performed by single ants. Examples are the use of a
local search procedure applied to the solutions built by
the ants, or the collection of global information that
Page No. 92
can be used to decide whether it is useful or not to
deposit additional pheromone to bias the search
process from a non-local perspective. As a practical
example, the daemon can observe the path found by
each ant in the colony and choose to deposit extra
pheromone on the components used by the ant that
built the best solution. Pheromone updates performed
by the daemon are called offline pheromone updates.
1.11.1 ANT COLONY BEHAVIOR
Real Ants
In many ants the visual system is very simple. Some
species are completely blind. Communication between
ants and between ants and their environment is often
based on the use of chemical signals. Pheromones are
produced by ants and they deposit them on trials when
walking in search of food. By sensing the pheromone,
the following ants can find food. Inspirational source
of ant colony algorithms is the double bridge
Fig 9. Double Bridge Experiment.
experiment described below.
Double bridge experiment
Fig 1. shows the double bridge experiment to illustrate
the ant behaviour. Ant nest is connected to food source
via two paths of differing length. Initially, ants move
randomly and chose between shorter and longer path
with equal probability. While walking, ants deposited
pheromone. When choosing a path, ants chose with
higher probability the path with the highest pheromone
concentration. Ants choosing the short path will be
first back with food. Therefore, trail on shorter path
grows more quickly
Auto Catalysis:
Positive feedback. It is a Self-reinforcing process If no
limiting mechanism is in place, it leads to explosion.
It is the central mechanism in ant algorithms.
S. K. Satapathy, Dr. S. K. Nayak
Probability of an ant choosing a path increases with
the number of ants that chose the same path.
It is interesting to understand how ants, which are
almost blind animals with very simple individual
capabilities, act together in a colony and find the
shortest route between the ant’s nest and a source of
food. They are also capable of adapting to changes in
the environment, for example, finding a new shortest
path once the old one is no longer feasible due to a
new obstacle. The studies by ethnologists reveal that
such capabilities that the ants have are essentially due
to what is called “pheromone trails” that ants use to
communicate information among individuals
regarding path and decide where to go. Ants deposit a
certain amount of pheromone while walking, and each
ant probabilistically prefers to follow a direction rich
in pheromone rather than a poorer one. In case of an
obstacle in place, these ants that choose, by chance,
the shorter path around the obstacle will more rapidly
reconstitute the interrupted pheromone trail compared
to those that choose the longer path. Hence the shorter
path will receive a higher amount of pheromone in the
time unit and this will in turn cause a higher number of
ants to choose the shorter path (auto catalytic) process,
very soon all the ants will choose the shorter path.
The Abstract Algorithm
● Colony of artificial ants build solutions to a given
problem by moving on the problem’s graph
representation
● Each feasible path represents a solution to the
problem
● They move by employing a probabilistic local
decision rule that exploits pheromone trail values
● Once an ant has built a solution (or while the
solution is being built), the ant evaluates the
quality of the solution, and deposits pheromone
on the components it used
● This directs the search of the ants in the future
I.11.2 A SIMPLE ANT COLONY ALGORITHM
Figure 2 shows the simple ant colony algorithm. The
working of the can be described by means of the
following.
Initialize A(t): The problem parameters are encoded as
a real number. Before each run, the initial population
(Nest) of the colony are generated randomly within
the feasible region which will crawl to different
directions at a radius not greater than R.
Page No. 93
 1) Evaluate A(t): The fitness of all ants are evaluated
based on their objective function.
2) Add_trail: The trail quantity is added to the
particular directions the ants have selected in
proportion to the ants’ fitness.
3) Send_ants A(t): According to the objective
function, their performance will be weighted as a
fitness value which directs influence to the level of
trail quantity adding to the particular directions the
ants have selected. Each ant chooses the next node
to move taking into account two parameters: the
visibility of the node and the trail intensity of the
trail previously laid by other ants. The send_ants
process sends ants by selecting directions using
Tournament selection based on the two parameters.
Start
t=0
Initialize A(t)
Evaluate A(t) Add Trail
No Send Ants
Terminate ? t = t+1
Evaporate
Yes
End
Fig 10. A Simple Ant Colony Algorithm
4) Evaporate: finally, the pheromone trail secreted by
an ant eventually evaporates and the starting
point(nest) is updates with the best tour found.
CONCLUSION
Intelligence Optimization Techniques are widely being
used in many disciplines for the solution of complex
S. K. Satapathy, Dr. S. K. Nayak
non-linear problems. The techniques are generally
independent of the problem being solved. The
intelligent optimization techniques, though
independently different, work on the same principle of
‘search through population of points’. The methods
employ different representation and coding methods.
Each is a potential tool for problem solving. They are
bound to reach the global minimum without getting
stuck at the local minimum solutions. Hybrid methods
are gaining more importance as they encompass the
combination of different potential strategies for the
solution of a problem from different perspectives.
These techniques hold promise in the solution of hard
problems or ‘ill-structured’ problems encountered.
Application of these techniques to power system
optimization problems like economic dispatch and
optimal power flow have demonstrated that each
technique can be formulated in a different fashion for
the same problem and yet achieve the same solution.
However the inference obtained from each method is
different and provides an insight into the problem
solving aspect. This helps in using the technique for
more complex and ill-structured problems in power
system optimization problems.
REFERENCES
1) D. E. Goldberg, Genetic Algorithms in Search,
Optimization and Machine Learning”, 4th Indian
Reprint, Pearson Education Asia, 2001.
2) James Kennedy, Russell Eberhart, “Particle
Swarm Optimization”, Proc. IEEE Int’l Conf.
on Neural Networks, Perth, Australia.
3) M. Dorigo and G. Di Caro, 1999. The Ant
Colony Optimization meta-heuristic. In D.
Corne, M. Dorigo, and F. Glover editors, New
Ideas in Optimization}, pages 11-32. McGraw-
Hill.
4) M. Dorigo, G. Di Caro and L. M. Gambardella,
1999. Ant algorithms for discrete optimization.
Artificial Life, 5, 2, pages 137-172.
Page No. 94
 Challenges of Implementing Smart Grids in India
Ramakrishna Raghutu1, Palla Shailesh Kumar2
1
Electrical Engineering, GMRIT, RAJAM (AP), INDIA, ramakrishnaree@gmail.com
2
Electrical Engineering, GMRIT, RAJAM (AP), INDIA, pallashaileshkumar@gmail.com

ABSTRACT

Smart Grids will play a vital role to help utilities in accomplishing this mission. So, the utilities will need to invest
heavily in new hardware, software, business process development, and staff training. Further there would be high
investment in home area networks and smart appliances by the customers. Achieving the broader view of Smart
Grid will require complex task prioritization and right set of policies and regulations to be in place. Justifying its
implementation however requires a full understanding of the long term benefits it would bring to the customers,
utilities, societies in terms of minimizing the cost, fewer blackouts and improved customer service. In addition to
these benefits it would play important role in addressing global issues like energy security and climate change.

INTRODUCTION • Helps utilities in improving their performance and

The Smart Grid is an idea of a better electricity
delivery infrastructure. Smart Grid implementations
will certainly increase the quantity, quality, and use of
information available from advanced sensing,
computing, and communications hardware and
software. As a result, they help utilities address two of
the main issues in today’s world
• Environmental concerns.
• Power delivery limitations and disturbances.
Effective use of Smart Grid technologies helps
utilities in:
• Improved grid usage.
• Improves grid efficiency and security.
• Better match of demand with supply of energy
and grid congestion.
• Enable distributed generation.
• Allow customers to manage their consumption
level and to take benefit of pricing and supply
options.
Smart Grid Background:
What are Smart Grids?
Smart Grid uses computer hardware and software,
sensors, telecommunication equipment and services to:
• Helps the customer to manage consumption and
use electricity wisely.
• Enables customer to respond to utility that help
minimize the period of surpluses, bottlenecks, and
outages.
controlling costs by timely availability of
information.
Thus Smart Grids associates customer to electricity by
an information rich network. And also it provides
utilities with valuable operational information that
helps them to improve efficiency. In a layman's term
Smart Grid is an efficient combination of electrical
infrastructure and information
WHAT IS PROMPTING SMART GRID
DEVELOPMENT
Environmental Impact:
Smart Grid development is happening at a very fast
pace because of the broad interest of policy makers
and utilities in decreasing the adverse effect that
energy usage has on the environment. Smart Grids
uses technology to drive efficiencies in transmission,
distribution, and consumption. As a result, fewer
generating plants, fewer transmission and distribution

Ramakrishna Raghutu, Palla Shailesh Kumar Page No. 95

assets are required in order to cater the growing
demand of electricity. With the possible expectation of
wind farm sprawl, landscape preservation is one of the
evident benefits. Since maximum generation today
results in emission of greenhouse gas, Smart Grids
reduces air pollution and plays a significant role in
combating global climate change issue.
Smart Grids has the capability to accommodate
technical difficulties of integrating renewable
resources like wind and solar to the grid, providing
further reduction in greenhouse gas emissions.
Costs:
The ability to bypass the cost of the plant and grid
development is a major advantage to both the utilities
and customers. And Smart Grids will not reduce funds
expansion, of course; therefore huge investments are
required in order to setup a link between the customers
and the Smart Grid. Further with the aid of Smart
Grids less generating units would be required in order
to full fill the energy demand of the growing
population and cost of setting up more and more plants
can be deferred. At that point of time, more emphasis
will be on overall development of T&D efficiency
based on demand response, load control, and many
other Smart Grid technologies.
Energy efficiency would be the second priority in
order to save cost with reference to the customers.
With timely and detailed information provided by
Smart Grids, customers would be encouraged to limit
waste, adopt energy-efficient building standards, and
invest more and more in energy efficient appliances.
Utility Operations:
Smart Grids can assist the utilities, as the principal
focus of the utilities is to improve business processes.
Many utilities have an extensive list of projects that
they would like to fund in order to improve the
customer service or to ease workforce’s burden of
repetitive work. Calculating Smart Grid benefits by the
cost/benefit analysis it puts emphasis in favour of the
change and can also significantly decrease
settlement/payback periods.
Mobile workforce group and asset management group
work collectively to organize assets and then maintain,
renovate, and replace them. Thus results in increased
productivity and fuel saving from superior methods.
Similarly, Smart Grid provides customers with real
time information and encourages them to do online
payments, thus lowering billing costs. Utilities can
Ramakrishna Raghutu, Palla Shailesh Kumar
include these cost and service improvement in the list
of Smart Grid benefits.
Theft Control:
This is not an issue in developed countries like US, but
in developing countries like India, where people have
a little insight of the grid and higher poverty rate,
power theft is quite common. With development of
Smart Grid, power theft can be controlled to a greater
extent, thereby improving the efficiency of our
distribution system. Thus grids will provide higher
quality and reliable power supply, and there will be
fewer blackouts.
BARRIERS IN IMPLEMENTATION OF SMART
GRID:
Policy and regulation:
The current policy and regulatory frameworks were
typically designed to deal with the existing networks
and utilities .To some extent the existing model has
encouraged competition in generation and supply of
power but is unable to promote clean energy supplies.
With the move towards smart grids, the prevailing
policy and regulatory frameworks must evolve in
order to encourage incentives for investment. The new
frameworks will need to match the interests of the
consumers with the utilities and suppliers to ensure
that the societal goals are achieved at the lowest cost
to the consumers.
Generally, governments set policy whereas regulators
monitor the implementation in order to protect the
consumers and seeks to avoid market exploitation.
Over the last two decades, the trend of liberalized
market structure in various parts of the world has
focused the attention of policy-makers on empowering
competition and consumer choice. The regulatory
models have evolved to become more and more
effective to avoid market abuse and to regulate the
rates of return.
Moving forward, the regulatory model will have to
adopt the policy which focuses much on long term
carbon reduction and security of supply in the defined
outcomes and they need to rebalance the regulatory
incentives to encourage privately financed utilities to
invest at rates of return that are commensurate to the
risk. This may mean creating frameworks that allow
risk to be shared between customers and share holders,
so that risks and rewards are balanced providing least
aggregate cost to the customer.
Business Scenario:
Page No. 96
 The majority of examples results in negative business
cases, undermined by two fundamental challenges:
High capital and operating costs:
Capital and operating costs include large fixed costs
linked to the chronic communications network.
Hardware costs do not cause in significant growths in
economies of scale and software integration possess a
significant delivery and integration risks.
Benefits are constrained by the regulatory frame
work :
When calculating the benefits, organizations tend to
be conservative in what they can gather as cash
benefits to the shareholders. For example in many
cases, line losses are considered to be put on to the
customer and as a result any drop in losses would have
no net impact on the utility shareholder. The smart
grid benefits case may begin on a positive note but, as
misaligned policy and regulatory incentives are
factored in, the investment becomes less attractive.
Therefore regulators are required to place such
policies and regulations in place which could provide
benefits both to the utilities and the consumers.
Therefore the first factor to be considered is to provide
incentives to the utilities in order to remove
inefficiencies from the system. They should be aptly
remunerated for the line losses on their networks.
On the budget side of the calculation, there is no
avoiding the fact that smart technologies are expensive
to implement, and at the present level it is right to
factor in the risk associated with delivery. But the
policy makers and regulators can mitigate that risk by
seeking economies of scale and implementing
advanced digital technologies.
Technology maturity and delivery risk:
Technology is one of the essential constituents of
Smart Grid which include a broad range of hardware,
software, and communication technologies. In some
cases, the technology is well developed; how ever , in
many areas the technologies are still at a very initial
stage of development and are yet to be developed to a
significant level. As the technologies advances, it will
reduce the delivery risk; but till then risk factor have to
be included in the business situation.
On the hardware side, speedy evolution of technology
is seen from vendors all over the world. Many recently
evolved companies have become more sceptical to the
communications solutions and have focused on
operating within a suite of hardware and software
solutions. Moreover the policy makers, regulators, and
Ramakrishna Raghutu, Palla Shailesh Kumar
utilities look upon well-established hardware providers
for Smart Grid implementation .And this trend is
expected to continue with increasing competition from
Asian manufacturers and, as a consequence, standards
will naturally form and equipment costs will drop as
economies of scale arises and competition increases.
On the software and data management side, the major
challenge is to overcome the integration of the entire
hardware system and to manage high volume of data.
With multiple software providers come multiple data
formats and the need for complex data models. In
addition, the proliferation of data puts stresses on the
data management architecture that are much similar to
the telecommunications industry than the utilities
industry. Many of these issues are currently being
addressed in pilots such as Smart Grid task force and,
as a consequence, the delivery risk will reduce as
standards will be set up.
Lack of awareness:
Consumer’s level of understanding about how power
is delivered to their homes is often low. So before
going forward and implementing Smart Grid concepts,
they should be made aware about what Smart Grids
are? How Smart Grids can contribute to low carbon
economy? What benefits they can drive from Smart
Grids? Therefore:
• Consumers should be made aware about their
energy consumption pattern at home, offices...etc.
• Policy makers and regulators must be very clear
about the future prospects of Smart Grids.
• Utilities need to focus on the overall capabilities
of Smart Grids rather than mere implementation
of smart meters. They need to consider a more
holistic view.
Access to affordable capital:
Funds are one of the major roadblocks in
implementation of Smart Grid. Policy makers and
regulators have to make more conducive rules and
regulations in order to attract more and more private
players. Furthermore the risk associated with Smart
Grid is more; but in long run it is expected that risk-
return profile will be closer to the current situation as
new policy framework will be in place and risk will be
optimally shared across the value chain.
In addition to this, the hardware manufacturers are
expected to invest more and more on mass production
and R&D activities so that technology obsolescence
Page No. 97
risk can be minimized and access to the capital
required for this transition is at reasonable cost.
Skills and knowledge:
As the utilities will move towards Smart Grid, there
will be a demand for a new skill sets to bridge the gap
and to have to develop new skills in analytics, data
management and decision support. To address this
issue, a cadre of engineers and managers will need to
be trained to manage the transition. This transition will
require investment of both time and money from both
government and private players to support education
programs that will help in building managers and
engineers for tomorrow. To bring such a change
utilities have to think hard about how they can manage
the transition in order to avoid over burdening of staff
with change.
Cyber security and data privacy:
With the transition from analogous to digital
electricity infrastructure comes the challenge of
communication security and data management; as
digital networks are more prone to malicious attacks
from software hackers, security becomes the key issue
to be addressed.
In addition to this; concerns on invasion of privacy
and security of personal consumption data arises. The
data collected from the consumption information could
provide a significant insight of consumer’s behaviour
and preferences. This valuable information could be
abused if correct protocols and security measures are
not adhered to.
If above two issues are not addressed in a transparent
manner, it may create a negative impact on customer’s
perception and will prove to be a barrier for adoption.
SOLUTIONS TO OVERCOME THE BARRIERS
IN IMPLEMENTATION:
Despite the challenges mentioned above, there are a
number of steps that can be taken to speed up the
implementation of smart grid technologies. Foremost
step that is required to be taken is that policy-makers
and regulators need to restructure the economic
incentives and align risk and reward across the value
chain. By building the right economic environment for
the private sector investment and focusing more
broadly about the way that social value cases are
created and presented implementation would become
much easier. By analyzing these solutions in bigger
environments i.e. in cities, the entire industry will
learn what it takes to implement smart grids
Ramakrishna Raghutu, Palla Shailesh Kumar
successfully and will result in developing an industry
that is set to boom in the coming periods.
Forming Political and Economic Frameworks:
Policy makers and regulators have to implement a
framework which optimally spread the risk over the
whole value chain i.e. to guard the investors from risk
and to yield the result at lower cost to the customers.
They have to form a robust incentive model in order to
attract more and more private investment. Also rate of
return should be based on the output generated.
Rewards and penalty mechanism should be considered
in order to monitor the performance of the utilities and
to encourage them to deliver the outcomes in the most
efficient manner.
Technological and delivery risk associated with Smart
Grid are significant. And this can be overcome over a
due course of time as more issues arise and are
addressed. Risks associated with Smart Grid have to
be shared by every member across the value chain.
While making the framework regulators must consider
how much of that risk a utility can pass on to the
contractors, suppliers and consumers. By maintaining
the proper balance, there will be an improved
alignment of the incentives. And further they have to
tackle numerous policy disputes and recommend
potential solutions.
Moving Towards a Societal Value System:
The major challenge for the transition from analogous
to digital infrastructure will be to move from utility-
centric investment decision to societal-level decisions
which determine wider scopes of the Smart Grid. This
would help in the accelerated adoption of Smart Grid
Technology by the society.
Achieving greater efficiency in energy delivery:
Smart Grid Technology should consider building
greater efficiency into the energy system which would
result in reduction of losses, peak load demand and
thereby decreasing generation as well as consumption
of energy. New regulatory framework which
incentivizes utilities for reducing the technical losses
would help utilities to perform more efficiently.
Enabling distributed generation and storage:
Smart grids will change where, when and how energy
is produced. Each household and business will be
empowered to become a micro-generator. Onsite
photovoltaic panels and small-scale wind turbines are
the predominant examples; developing resources
Page No. 98
consist of geothermal, biomass, hydrogen fuel cells,
plug-in hybrid electric vehicles and batteries. As the
cost of traditional energy sources continues to rise and
the cost of distributed generation technologies falls,
the economic situation for this evolution will build.
Increasing Awareness on Smart Grids:
There is an imperative need to make the society and
the policy makers aware about the capabilities of a
Smart Grid. The main step is to form a perfect,
universal description on the common principles of a
smart grid .Beyond agreement on a characterization,
the matter also needs to be debated more holistically
as a true enabler to the low-carbon economy, rather
than as an investment decision to be taken within the
meeting room of distinct utilities. The importance of
consumer education is not to be underestimated. The
formation of user-friendly and state-of-the-art products
and services will play a significant role in convincing
the society about Smart Grids. Also the utilities are
required to scrutinize the major challenges in
implementation of Smart Grid and their impact on
their business model and operations.
Creating a Fresh Pool of Skills and Knowledge:
Successful implementation of the smart grid will
require a large number of highly skilled engineers and
managers mainly those who are trained to work on
transmission and distribution networks. As a result to
on-job training and employees development will be
vital across the industry. Simultaneously, there is a
requirement for investment in the development of
relevant undergraduate, postgraduate and vocational
training to make sure the availability of a suitable
workforce for the future. The investment in T&D
should not be limited and neither in research and
knowledge development, which would be essential for
the development of this sector.
Addressing Cyber security Risks and Data Privacy
Issues:
Smart Grid success depends on the successful
handling of two major IT issues:
• Security
• Integration and data handling
With increase in computers and communication
networks comes the increased threat of cyber-attack.
The Government should look into this matter because
consumer’s consumption data can be misused by the
utilities and the third party. Utilities have to give
assurance to the consumers that their valuable
information is handled by authorized party in ethical
Ramakrishna Raghutu, Palla Shailesh Kumar
manner. The government has to adopt high standard
level in order to withstand cyber-attacks.
ON-GOING SMART GRID ACTIVITIES:
• APDRP, R-APDRP initiative for distribution
reform (AT&C focus)
• DRUM India – Distribution Reform Upgrade,
Management
• Four pilot sites (North Delhi, Bangalore, Gujarat,
Maharashtra)
• Smart Grid Vision for India
• Smart Grid Task Force – Headed by Sam Pitroda
• BESCOM project – Bangalore – Integration of
renewable and distributed energy resources into
the grid
• KEPCO project in Kerala India - $10 Billion
initiative for Smart-Grid
• L&T and Telvent project – Maharashtra –
Distribution Management System roll-out
• Distributed generation via roof-top solar for 40%
in a micro-grid
RECOMMENDATIONS:
Regulators:
Create a regulatory framework which aligns incentives
of each member in the value chain.
• Allocate risk and reward efficiently.
• Consider both utilities and customer while making
policies.
• Adopt output based regulatory system
(Reward/Penalties) which stresses on utilities to
perform better.
Utilities
• Adopt more holistic approach about Smart Grids,
so that they can convey its future benefits to the
customers.
• Reduce the risk of technology obsolescence by
R&D activities.
• Undertake large scale pilot projects and analyze
the benefits.
• Transformation from utility-centric investment
decision to societal-level decisions.
Vendors
• Required to play important role in policy making
process
Page No. 99
• To help utilities to adopt flexible design and
compatibility of Smart Grid fast.
• To convince customers about the acceptance of
changing trend by product and service offering.
Customers:
• Plays critical role by demanding for more flexible
service.
• To encourage more players to enter in this field
and in order to make the market competitive
• To help utilities and regulators to set goals and
make conducive policies.
• To increase the awareness in society.
CONCLUSION
In this paper an attempt has been made to analyze the
key challenges in implementing the Smart Grid
concept in India. In most of the advanced countries
Utilities have made major achievements in terms of
productivity, reliability, and efficiency through the use
of Smart Grid technology. Indian utilities are still
lagging far behind when compared to other countries.
Today their main focus is on providing energy at
reasonable price but soon the day will come when the
Ramakrishna Raghutu, Palla Shailesh Kumar
utilities will be focusing on encompassing sustainable
use and environmental improvement into their
agendas.
REFERENCE:
1. .http://www.projectsmonitor.com/ELECTRICITY/i
ndia-is-gearing-up-for-smart-grid-technology
2. http://gigaom.com/cleantech/5-reasons-why-
developing-countries-need-smart-grids-too/
3. http://www.metering.com/node/17642
4. http://panchabuta.wordpress.com/2010/11/17/first
-pilot-project-of-smart-grid-technology-in-
electronic-city-bangalore-india/
5. http://www.energyandcapital.com/articles/smart-
grid-developments/1257
6. http://spectrum.ieee.org/tag/smart+grids
7. http://www.drumindia.org/smartgrid/agenda.asp
8. http://en.wikipedia.org/wiki/Smart_grid James O.
McSpadden, “ Wireless Power Transmission
Demonstration”, Texas A&M University, June,
1997.
Page No. 100
 Hybrid Model of Linear Dynamic Systems using
Eigen Spectrum Analysis

Manmadha kumar.B1, Srinivasa Acharya2, G.T.Chandrasekhar3

Department of E.E.E, AITAM, Tekkali, Andhra Pradesh1,2.
Department of E.E.E Sri Sivani College of Engineering,Chilakapalem, Andhra Pradesh3
#
boddmann@yahoo.co.in, *sriacharya85@gmail.com, #gtchsekhar@gmail.com

ABSTRACT
This paper suggest a model reduction procedure for higher order continuous time systems. The proposed
reduction method generates the reduced order denominator of model by using Eigen Spectrum analysis,
where as the reduced order numerator polynomial is obtained by using Formal Approach analysis. This
proposed method is used for the stability analysis and the design of the compensator for the higher order
continuous tine systems.
Keywords- Eigen Spectrum analysis- Formal Approach analysis- Pole centroid- Tuning- PID Controller.

ÊË
”Í s”š Ës”r Ër s …….. s”Îfš ËÎfš
ÌË
’Í s’š Ës’r Ër s ……… s’Îfš ËÎfš sËÎ
INTRODUCTION G(S)= = (1)
The most control systems found in practice are of
orders higher than two, it would be useful to establish Where, -λ1< -λ2 <-λ3…………. < –λn are poles of
guidelines on approximation of high-order systems by high order system (HOS).
lower-order ones in so far as the transient response is
concerned. Almost all the practical systems are of Step1: Fixing of the Eigen spectrum zone (ESZ) of the
very higher order. Application of the available HOS as shown in Fig.1 If poles are (i =1,…,n) are
methods for performance analysis and the design located at Real + Img (i = 1,…,p) within the ESZ, then
of compensators or controllers to improve the the two lines passing through the nearest (Reλ1) and
performance of these high order systems becomes farthest (Re λp) real poles when cut by two lines
tedious involving much computational effort. To passing through the farthest imaginary pole pairs
overcome this problem one of the important and (±Im(max)) form the Eigen spectrum zone(ESZ).
useful approach suggested in literature is the
application of order reduction technique. Step2: Pole centroid is defined as the mean of real
parts of the poles and is expressed :
λm Ï
∑iÐ1 Reλi
p
PROPOSED METHOD
(2)
p
In this reduction both the pole centroid and system
stiffness of original order system and reduced order System stiffness is defined as the ratio of the nearest to
systems are kept exactly same to obtain the reduced the farthest pole of a system in terms of real parts only
order system poles, while the zeros are synthesized by and is put as:
ÑÂ
λS Ï
Formal approach method between the transient λp
responses of original and reduced order systems using
(3)
ÑÂ
pertaining to a unit step input.
λ1
Step3: Determination of Eigen spectral points of LOS:

ÑÂ
(A)Eigen spectrum method:
λ1'
Let, the transfer function of high order system (HOS) λs = = λs
ÑÂ λp
of order 'n' is: '

Manmadha kumar.B, Srinivasa Acharya, G.T.Chandrasekhar Page No. 101

 Fig. 2. PID controller with plant

ÒÓ λ1 sÒÓ λ2 s7…………..s ÒÓ λp
λ'm = (4), and Example:
1
p Considered fourth order system is

× Ø

2 Ù s52 r s2s
2 Ú sA2 Ù s42 r s4A2s
(7)
Where λ'i (i=1,………….,r) are the poles of LOS
' '
located at -(Re λi +Im λi ) . Now if,
i1,2,3........p By using Eigen spectrum analisis the considered
fourth order
Ñ λ'p › ÒÓ λ1' system has reduced as

× Ø

= M (5)
A.ÛA52 s 
p ›
'
2 r s 42 s 
(8)
' ' ' '
i.e. Re λ 1 + M = Re λ
2 , Re λ
2 + M= Re λ 3

Fig. 3 Time response of reduced second order model

Tuning:

constant values of œÜ , œ$ , œ are 5, 1, 4. Now the

For the considered fourth order model the actual
Fig. 1. Eigen spectrum Zone (ESZ) of HOS
LOS transfer function with controller is
A.AÛ2 Ù s.5Û52 r st.5Ý2sA
« λS p ' ®
ª M 1
Q ­ .AÛ2 Ù st.5Û52 r s.5Ý2sA
Z
D|˜
(9)
λ'p F G K
ª
ª
­
' ­
0 After tuning the constant values of œÜ ,œ$ , œ are
Ö
= (6)
1 M λS 1Mp
© ¬ 4.55, 0.5, 2.5. Now the HOS transfer function with
controller is
A.42 Þ sÝ.2 Ú sÛ.42 Ù sÝ.52 r sÛt.2sÛA
Tuning of PID controllers:
.42 Þ sÝ.2 Ú sÝ.42 Ù sÝÝ.52 r st.2sÛA
PID control of a plant is shown in Fig.. If a (10)
mathematical model of the plant can be derived, then it
is possible to apply various design techniques for outputs:
determining parameters of controller that will meet the The Eigen spectrum analysis algorithm has been
transient and steady state specifications of the closed implemented in MATLAB and results are obtained
loop system. for the considered fourth order system.
Fig. 4. Shows the time response for the transfer
function of LOS given in equation (9).

Manmadha kumar.B, Srinivasa Acharya, G.T.Chandrasekhar Page No. 102

 CONCLUSIONS
An algorithm for the order reduction of linear dynamic
systems has been given based on the Eigen spectrum
analysis. The algorithm has been illustrated with the
help of a numerical example having real poles only. It
is being extended for the systems with complex poles.
The same reduction can be done by using formal
approach where the zeros are determined with this
method.

REFERENCES
Fig. 4 Time response of lower order system
i) G. Parmar, Life Member SSI, AMIE, S.
Mukherjee, FIE, and R. Prasad, "Reduced Order
Fig. 5. Shows the timeresponse for the transfer
Modelling of Linear Dynamic Systems using
function of HOS given in equation (10). Particle Swarm Optimized Eigen Spectrum
Analysis", International Journal of
Computational and Mathematical Sciences 1;1;
2007.
ii) M. Jamshidi," Large Scale Systems Modelling
and Control Series", Vol. 9, North Holland,
Amsterdam, Oxford, 1983
iii) C.B Vishwakarma and R. Prasad, ''System
Reduction Using Modified Pole Clustering And
Pade Approximation'', XXXII National Systems
Conference, NSC 2008, December 17-19, 2008.
iv) Shiv Kumar Tomar and Rajendra Prasad,"
Linear Model Reduction Using Mihailov
Stability Criterion And Continued Fraction
Expansions", XXXII National Systems
Fig. 5 Time response of higher order system Conference, NSC 2008, December 17-19, 2008.
v) T.Manigandan, N.Devarajan, S.N.Sivanadam",
Design of PID controller using reduced order
model", Academic Open Internet Journal,
Volume 15, 2005.

Manmadha kumar.B, Srinivasa Acharya, G.T.Chandrasekhar Page No. 103

 Modelling of Photovoltaic Cell Using Matlab/Simulink
Bideha Ranjan Mahunta1, Ashiya Iliyash2, Balaji Prasad Padhi3
Students, Dept. Of E.E.E, Gandhi Institute for Technology and Management, Bhubaneswar,
bidehamahunta@gmail.com,ashiya.iliyash010@gmail.com1,2
Professor, Dept. of E.E.E,Gandhi Institute for Technology and Management, Bhubaneswar,balajippadhi@gietbbsr.com

ABSTRACT

This paper presents the implementation of a generalized photovoltaic model using Matlab/Simulink software
package, which can be representative of PV cell, module, and array for easy use on simulation platform. The
proposed model is designed with a user-friendly icon and a dialog box like Simulink block libraries. This makes the
generalized PV model easily simulated and analyzed in conjunction with power electronics for a maximum power
point tracker. Taking the effect of sunlight irradiance and cell temperature into consideration, the output current and
power characteristics of PV model are simulated and optimized using the proposed model. This enables the
dynamics of PV power system to be easily simulated, analyzed, and optimized.

Keywords- Generalized model, photovoltaic module, Matlab / Simulink.

I. INTRODUCTION maximum power point tracking (MPPT) for PV

system applications. The mathematical PV models
With increasing concerns about fossil fuel
used in computer simulation have been built for over
deficit,
the pass
Sky rocketing oil prices, global warming, and damage
to environment and ecosystem, the promising
four decades [1]-[3]. Almost all well-developed PV
incentives to develop alternative energy resources with
models describe the output characteristics mainly
high efficiency and low emission are of great
affected by the solar insolation, cell temperature, and
importance. Among the renewable energy resources,
load voltage. However, the equivalent circuit models
the energy through the photovoltaic (PV) effect can be
are implemented on simulation platforms of power
considered the most essential and prerequisite
electronics, such as SPICE. Recently, a number of
sustainable resource because of the ubiquity,
powerful component-based electronics simulation
abundance, and sustainability of solar radiant energy.
software package have become popular in the design
Regardless of the intermittency of sunlight, solar
and development of power electronics applications.
energy is widely available and completely free of cost.
However, the Sim Power System tool in Matlab
Recently, photovoltaic array system is likely
/Simulink package offers wind turbine models but no
recognized and widely utilized to the forefront in
PV model to integrate with current electronics
electric power applications. It can generate direct
simulation technology. Thus, it is difficult to simulate
current electricity without environmental impact and
and analyze in the generic modeling of PV power
contamination when is exposed to solar radiation.
system. This motivates me to develop a generalized
Being a semiconductor device, the PV system is static,
model for PV cell, module, and array using
quite, and free of moving parts, and these make it have
Matlab/Simulink.The main contribution of this paper
little operation and maintenance costs. Even though
is the implementation of a generalized PV model in
the PV system is posed to its high capital fabrication
the form of masked block, which has a user-friendly
cost and low conversion efficiency, the skyrocketing
icon and dialog in the same way of Matlab/Simulink
oil prices make solar energy naturally viable energy
block libraries or other component-based electronics
supply with potentially long-term benefits. PV module
simulation software packages, such as Caspoc. The
represents the fundamental power conversion unit of a
remainder of this paper is organized as follows. For
PV generator system. The output characteristics of PV
easy presentation, the traditional PV models are
module depends on the solar insolation, the cell
addressed in Section II. And the nonlinearity of PV
temperature and output voltage of PV module. Since
current versus voltage (I-V) and power versus voltage
PV module has nonlinear characteristics, it is
(P-V) characteristics are shown as well. Section III
necessary to model it for the design and simulation of

Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi PAGE NO. 104
demonstrates the implementation and simulation
results for the proposed model using Matlab/ Simulink
software package. A different MPPT strategy is found
and some interested issues are also discussed. Finally,
brief conclusions are drawn in Section IV.
II. LITERATURE REVIEW
2.1 Introduction
This chapter contains the studies on the conventional
photovoltaic and the process of photovoltaic
Generation and Advantages and Disadvantages of
photovoltaic cell.
2.2 Process of PV Generation
A Photovoltaic PV generation is the whole assembly
of solar cells, connections, Protective parts, supports
etc. In the present modeling, the focus is only on
Cell/module/array .Solar cells consists of a p-n
junction fabrication in a thin wafer or layer of
semiconductor, which is usually silicon. In the dark,
the I-V output characteristic of a Solar cell has
an exponential characteristic c similar to that of a
diode. When solar energy or photons hits the solar
cell, with energy greater than band gap
energy of the Semiconductor, electrons are
knocked loose from the atoms In the material,
creating Electron-hole pairs. These carriers are
swept apart under the influence of the internal
electric field of the p-n junction and create a
current proportional to the incident radiation.
When the cell is short circuited, this current
flows in the external circuit; when open circuited,
this current is shunted internally by the
intrinsic p-n junction diode. The characteristic of this
diode therefore set the open circuit voltage
characteristics of the cell. (Francisco 2005)
2.3 ADVANTAGES AND DISADVANTAGES:
2.3.1 ADVANTAGES:
The 89 Pet watt of sunlight reaching the Earth's
surface is plentiful - almost 6,000 times more than the
15 terawatt of average electrical power consumed by
humans Additionally, solar electric generation has the
highest power density (global mean of 170 W/m²)
among renewable energies. Solar power is pollution-
free during use. Production end-wastes and emissions
are manageable using existing pollution controls. End-
of-use recycling technologies are under development
PV installations can operate for many years with
little maintenance or intervention after their initial
set-up, so after the initial capital costs of building
Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi
any solar power plant, operating costs are extremely
low compared to existing power technologies.
Solar electric generation is economically superior
where grid connection or fuel transport is difficult,
costly or impossible. Long-standing examples include
satellites, island communities, remote locations and
ocean vessels.
When grid-connected, solar electric generation
replaces some or all of the highest-cost electricity used
during times of peak demand (in most climatic
regions). This can reduce grid loading, and can
eliminate the need for local battery power to provide
for use in times of darkness. These features are
enabled by net metering. Time-of-use net metering can
be highly favourable, but requires newer electronic
metering, which may still be impractical for some
users. Grid-connected solar electricity can be used
locally thus reducing transmission/distribution losses
(transmission losses in the US were approximately
7.2% in 1995).Compared to fossil and nuclear
energy sources, very little research money has been
invested in the development of solar cells, so there
is considerable room for improvement. Nevertheless,
experimental high efficiency solar cells already have
efficiencies of over 40% in case of concentrating
photovoltaic cells and efficiencies are rapidly rising
while mass-production costs are rapidly falling.
2.3.2 Disadvantages:
Photovoltaic’s are costly to install. While the modules
are often warranted for upwards of 20 years, much
of the investment in a home-mounted system may be
lost if the home-owner moves and the buyer puts less
value on the system than the seller. The city of
Berkeley has come up with an innovative financing
method to remove this limitation, by adding a tax
assessment that is transferred with the home to pay for
the solar panels. Nine U.S. states have duplicated this
solution.
Solar electricity is seen to be expensive. Once a PV
system is installed it will produce electricity for no
further cost until the inverter needs replacing. Current
utility rates have increased every year for the past 20
years and with the increasing pressure on carbon
reduction the rate will increase more aggressively.
This increase will (in the long run) easily offset the
increased cost at installation but the timetable for
payback is too long for most.
Solar electricity is not available at night and is less
available in cloudy weather conditions from
conventional photovoltaic technologies. Therefore, a
Page No. 105
storage and complementary power system is required.
This is why many buildings with photovoltaic arrays
are tied into the power grid; the grid absorbs excess
electricity generated throughout the day, and provides
electricity in the evening.
Apart from their own efficiency figures, PV systems
work within the limited power density of their
location's insulation. Average daily insulation (output
of a flaplate collector at latitude tilt) in the contiguous
US is 3-7 kilowatt·h/m² and on average lower in
Europe.
Solar cells produce DC which must be converted to
AC (using a grid tie inverter) when used in current
existing distribution grids. This incurs an energy loss
of 4-12%
III. METHODOLOGY
3.1 PHOTOVOLTAIC MODEL:
Solar cell is basically a p-n junction fabricated in a
thin wafer or layer of semiconductor. The
electromagnetic radiation of solar energy can be
directly converted electricity through photovoltaic
effect. Being exposed to the sunlight, photons with
energy greater then the band-gap energy of the
semiconductor are absorbed and create some electron-
hole pairs proportional to the incident irradiation.
Under the influence of the internal electric fields of the
p-n junction, these carriers are swept apart and create a
photo current which is directly proportional to solar
isolation. PV system naturally exhibits a nonlinear I-V
and P-V characteristics which vary with the radiant
intensity and cell temperature
(a)
(b)
Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi
Fig.3.1 One-diode equivalent circuit model for a PV
cell
(a) Five parameters model
(b) Simplified four parameters model
3.1 (1) SOLAR CELL MODEL:
A general mathematical description of I-V output
characteristics for a PV cell has been studied for over
the past four decades [1]-[3]. Such an equivalent
circuit-based models mainly used for the MPPT
technologies. The equivalent circuit of the general
model which consists of a photo current, a diode, a
parallel resistor expressing a leakage current, and a
series resistor describing a internal resistance to the
current flow, is shown in Fig. The voltage-current
characteristic equation of a solar cell is given as
I = IPH − IS exp q(V + IRS) / kTCA −1 − (V + IRS) / RSH
(1)
Where
IPH is a light-generated current or photocurrent,
IS is the cell saturation of dark current,
q (= 1.6 ×10−19C) is an electron charge,
k (= 1.38 ×10−23J/K) is a Boltzmann’s constant,
TC is the cell’s working temperature,
A is an ideal factor,
RSH is a shunt resistance, and RS is a series resistance.
The photocurrent mainly depends on the solar
isolation and cell’s working temperature, which is
described as
I PH = [ISC + KI (TC −TRef)] λ ……..(2)
where ISC is the cell’s short-circuit current at a
25°C and 1kW/m2, KI is the cell’s short-circuit
current temperature coefficient, TRef is the cell’s
reference temperature, and λ is the solar isolation in
kW/m2. On the other hand, the cell’s saturation
current varies with the cell temperature, which is
described as[4]
IS=IRS (TC/TRrf) 3 exp [q EG (1/TRef) ….(3)
Where IRS is the cell’s reverse saturation current at a
reference temperature and a solar radiation EG is the
bang-gap energy of the semiconductor used in the cell.
The ideal factor A is dependent on PV technology [4]
and is listed in Table I.An even more exact
mathematical description of a solar cell, which is
called the double exponential model as shown in Fig.
1(b) [5], is derived from the physical behavior of solar
cell constructed from polycrystalline silicon. This
model is composed of a light-generated current source,
Page No. 106
two diodes, a series resistance and a parallel
resistance. However, there are some limitations to
develop expressions for the V-I curve parameters
subject to the implicit and nonlinear nature of the
model. Therefore, this model is rarely used in the
subsequent literatures and is not taken into
consideration for the generalized PV model. The shunt
resistance RSH is inversely related with shunt leakage
current to the ground. In general, the PV efficiency is
insensitive to variation in RSH and the shunt-leakage
resistance can be assumed to approach infinity without
leakage current to ground. On the other hand, a small
variation in RS will significantly affect the PV output
power. The appropriate model of PV solar cell with
suitable complexity is shown in Fig. 1(c) [4]. Equation
(1) can be rewritten to be[1]-[6][7]
I = I PH − IS [exp (q (V + IRS) / kTC A) -1] …. (4)
For an ideal PV cell, there is no series loss and no
leakage to ground,
i.e., RS = 0 and RSH = ∞. The above equation can be
simplified can be rewritten as
I=IPH-IS[exp(qV/KTCA)-1] ……..(5)
3.2 SOLAR MODULE AND ARRAY MODEL:
Since a typical PV cell produces less than 2W at 0.5V
approximately, the cells must be connected in series-
parallel configuration on a module to produce enough
high power. A PV array is a group of several PV
modules which are electrically connected in series and
parallel circuits to generate the required current and
voltage. The equivalent circuit for the solar module
arranged in NP parallel and NS series is shown in Fig.
2(a). The terminal equation for the current and voltage
of the array becomes as follows [8]-[11]
I=NPIPH-NPIS[exp(q(V/NS+IRS/NP)/KTCA)-1]
(NPV/NS+IRS)/RSH (6)
In fact, the PV efficiency is sensitive to small change
in RS but insensitive to variation in RSH . For a PV
module or array, the series resistance becomes
apparently important and the shunt down resistance
approaches infinity which is assumed to be open. In
most commercial PV products, PV cells are generally
connected in series configuration to form a PV module
in order to obtain adequate working voltage. PV
modules are then arranged in series-parallel structure
to achieve desired power output. An appropriate
equivalent circuit for all PV cell, module, and array is
generalized and Expressed in Fig. 2(b). It can be
shown that NS = NP = 1 for a PV cell, NP =1 and NS :
series number of cells for a PV module, and NS and NP
Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi
: series-parallel number for a PV array. The
mathematical equation of generalized model can be
described as
I = NP IPH − NP IS[exp(q(V / NS + IRS / NP ) / TCA)−1]
……(7)
The most simplified model [6]-[12]of generalized PV
module is depicted in Fig. 2(c). The equivalent circuit
is described on the following equation
I = NP I PH − NP IS [exp (qV/NSkTC A) −1 ….(8)
3.3 DETERMINATION OF MODEL PARAMETERS:
All of the model parameters can be determined by
examining the manufacturer’s specifications of PV
products. The most important parameters widely used
for describing the cell electrical performance is the
open-circuit voltage VOC and the short-circuit current
ISC. The aforementioned equations are implicit and
nonlinear; therefore, it is difficult to arrive at an
analytical solution for a set of model parameters at a
specific temperature and irradiance. Since
Normally IPH >> IS and ignoring the small diode and
ground-leakage currents under zero-terminal voltage,
the short-circuit current ISC is approximately equal to
the photo current IPH, i.e., IPH=ISC ….(9)
On the other hand, the VOC parameter is obtained by
assuming the output current is zero. Given the P open-
circuit voltage VOC at reference temperature and
ignoring the shunt-leakage current, the reverse
saturation current at reference temperature can be
approximately
Obtained as
IRS=ISC/[exp(qvoc/NSKATC)-1] ……(10)
In addition, the maximum power can be expressed as
Pmax =Vmax Imax =γVOCISC ……..(11)
Where Vmax and Imax are terminal voltage and output
current of PV module at maximum power point
(MPP), and γ is the cell fill factor which is a measure
of cell quality
3.4 SIMULATION RESULTS OF PV CELL AND
MODULE
For a PV cell with an ideal I-V characteristic, its open
circuit voltage and short-circuit current are given as
VOC = 0.596V and ISC = 2.0A, respectively. In
addition, NS = NP = 1 for a PV cell. Both I-V and P-V
output characteristics of generalized PV model for a
cell are shown in Figs. The nonlinear nature of PV cell
is apparent as shown in the figures, i.e., the output
Page No. 107
current and power of PV cell depend on the cell’s
terminal operating voltage and temperature, and solar
isolation as well. We find from Figs. that with
increase of working temperature, the short-circuit
current of the PV cell increases, whereas the
maximum power output decreases. In as much as the
increase in the output current is much less than the
decrease in the voltage, the net power decreases at
high temperatures. On the other hand, we observe
from Figs. And that with increase of solar isolation,
the short-circuit current of the PV module increases,
and the maximum power output increases as well.
The reason is the open-circuit voltage is
logarithmically dependent on the solar irradiance, yet
the short-circuit current is directly proportional to the
radiant intensity. By the same token, the Solarex MSX
60 PV module is taken for example. The key
specifications are listed in Table II in which the
nominal operating cell temperature (NOCT) is the
temperature that the cells will reach when they are
Operated at open circuit in an ambient temperature of
200C under AM 1.5 irradiance conditions with λ=0.8
kW/m2 and a wind speed less than 1 m/s. The
electrical characteristics of PV module are
generally represented by the current versus voltage
and power versus voltage curves. Both I-V and P-V
output characteristics of PV module at various
isolation and temperatures are carried out and the
results are shown in Figs. We also see from Figs. And
that with increase of working temperature, the short-
circuit current of the PV module Increases, whereas
the maximum power output decreases. The increase in
the short-circuit current is much less than the decrease
in the open-circuit voltage, and the effect makes
maximum power decreasing by about 0.45%/0C at
high temperatures. On the other hand, from, we also
observe that with increase of solar isolation, the
short-circuit current and the maximum power output
of the PV module increase as shown in Figs. There
as on is the open circuit voltage is logarithmically
dependent on the solar irradiance, yet the short-circuit
current is directly proportional to the radiant intensity.
3.5 STATISTICAL SIMULATION
For easy simulation, the solar radiation intensity for a
sample day is assumed to be a function of Gaussian
function which is defined as
λ (t) = λmaxexp [ − (t − tC )2 / 2σ2 ] …..(12)
where λmax is the maximal radiation intensity at a
given time, tC is the center time, and σ is the standard
deviation of Gaussian function. Fig. 9 shows a plot of
the Gaussian function for the solar radiation intensity
Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi
for a sample day with the conditions: λmax =1kW/m2, tC
=12, and σ = 0.5. The peak of sunlight intensity occurs
at noon. The cell temperature for a sample day is
assumed to be at a fixed temperature of NOCT by
ignoring the effect of the solar irradiation. Given
sunlight irradiance for a sample day, the output current
is governed by its output voltage which is determined
by the ensuing load. Without taking cell temperature
variations into consideration, the maximum power is
determined by both radiation intensity and output
voltage. For different output voltage, the output
current and power are shown in Figs. 10(a) and (b) for
a sample day. For easy comparisons, the root mean
square (R.M.S.) values of output power for various
output voltage for a sample day are calculated an
depicted in Fig. The maximum R.M.S. power during
the course of a sample day occurs with the output
voltage in the range of 15-16V.
IV. SIMULATIO RESULTS AND DISCUSSION
4.1 INTRODUCTION:
In the section of PV cell/module, a few graphs had
been obtain in MATLAB/SIMULINK to show the
output current, output voltage and output power by
varying solar irradiance and the cell temperature.
4.2 MODEL PERFORMANCE OF PV CELL/MODULE:
Model performance of Different Solar irradiance
The model I-V characteristic curves under different
solar irradiances at 250c are shown in fig. from the
graph, it is noted that the higher is the irradiance, the
larger are the short circuit current(ISC) and the open
circuit voltage(VOC) .And obviously, the large will be
the maximum power (PMax) is shown in fig . Vice
versa, a smaller short circuit current (ISC) and open
circuit voltage (VOC) is obtained if the solar irradiance
is lower.
Table4.1 Solarex MSX 60 Specifications (1KW/M2,
25OC)
Characteristics Specifications
Short –circuit current(ISC) 3.8A
Open-circuit voltage(VOC) 21.1V
Temperature coefficient of short-circuit
3MA/OC
current(K1)
Normal operating cell
49OC
temperature(NOCT)
Si-Poly(A-value) 1.3
Typical peak power(PP) 60W
Temp coefficient of open-circuit voltage -73MV/OC
Page No. 108
 10

0 0.8kw/m2

Current-------------------------------->
0.6kw/m2

-10 0.4kw/m2

0.2kw/m2
-20

-30

-40

-50
0 10 20 30 40 50 60
Voltage------------------------------------>

Fig. 4.3 P-V

V Characteristics curves of the PV model
under different values of isolation.

Fig. 4.1 MATLAB/SIMULINK model

90

80
Fig.4.4 MATLAB/SIMULINK model
70
1000

60
900
0.8kw/m2
50 800 Current(amp)
0.6kw/m2
700
40 0.4kw/m2
600

30 0.2kw/m2
500

400
20
300
10
200

0 100
0 5 10 15 20 25 30 35 40 45
Voltage------------------------
------------------------> 0
0 10 20 30 40 50 60
Time(hours)-------------------------------
------------------------------->
Fig. 4.2 I-V
V Characteristics curves of the PV model
under different values of isolation Fig .4.5 solar radiation intensity for a sample
day in the form of Gaussian function

Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi Page No. 109
 70

60

50

Power(w)40

Fig.4.3(c) Appropriate model

30

20

10

0
0 10 20 30 40 50 60
Voltage(V)

Fig. 4.6 P-V output characteristics with different TC

7
Fig.4.3(d) Simplified model

6
4.3 Different models of PV cells:
5

1
Fig. 4.3 (a) General model
0
0 10 20 30 40 50 60
Voltage(V)

Fig. 4.7 I-V output characteristics with different TC

Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi Page No. 110

Fig.4.3(b) Double exponential model
The above figures 4.3.(a), (b) ,(c), (d) are
shows the different models of PV cells .in this Fig. (a)
is the general model of the PV cell to understand the
electronic behavior of a solar cell , it is useful to create
a model which is electrically equivalent and is based
on discrete electrical components whose behavior is
well known. An ideal solar cell may be modeled by a
current source in parallel with a diode; in practice no
solar cell is ideal, so shunt resistance and a series
resistance component are added to the model. In this
fig (b) two diodes are connected in parallel with
current source in this practice no solar cell is ideal, so
shunt and series resistances are connected to this
model. in this fig (c) is the appropriate model in this
one diode is connected in parallel with the current
source , in practice no solar cell is ideal ,so that in this
model only series resistance is connected .in this
model appropriate voltage and current values are
appears. The fig (d) shows the simplified method of
solar cell in this the diode is connected in parallel with
the current source.
Finally concluded that the above all methods the
general method of PV cell is used, in this no solar cell
is ideal , so shunt resistance and series resistance are
connected .and the diode is connected in parallel with
the current source.
Bideha Ranjan Mahunta, Ashiya Iliyash, Balaji Prasad Padhi
V. CONCLUSION:
A generalized PV model which is representative of
the all PV cell, module, and array has been developed
with Matlab/Simulink and been verified with a PV cell
and a commercial module. The proposed model takes
sunlight irradiance and cell temperature as input
parameters and outputs the I-V and P-V characteristics
under various conditions. This model has also been
designed in the form of Simulink block libraries. The
masked icon makes the block model more user-
friendly and a dialog box lets the used easily configure
the PV model. Such a generalized PV model is easy to
be used for the implementation on Matlab/Simulink
modeling and simulation platform. Especially, in the
context of the Sim Power System tool, there is now a
generalized PV model which can be used for the
model and analysis in the filed of solar PV power
conversion system.
VI. REFERENCES
[1] S. W. Angrist, , Direct Energy Conversion,
Allyn and Bacon, Inc., 4 edition, 1982, pp. 177-
227.
[2] O. Wasynczuk, “Dynamic behavior of a class of
photovoltaic power systems,” IEEE Transactions
on Power Apparatus and Systems, vol. PAS-102,
no. 9, 1983, pp. 3031-3037.
[3] J. C. H. Phang, D. S. H. Chan, and J. R. Philips,
“Accurate analytical method for the extraction of
solar cell model parameters,” Electronics Letters,
vol. 20, no. 10, 1984, pp.406-408.
[4] C. C. Hua and C. M. Shen, “Study of maximum
power tracking techniques and control of dc-dc
converters for photovoltaic power system,”
Proceedings of 29 annual IEEE Power
Electronics Specialists Conference, vol. 1, 1998,
pp. 86-93.
5] J. A. Gow and C. D. Manning, “Development of
a photovoltaic array model for use in power-
electronics simulation studies,” IEE
Proceedings- Electric Power Applications, vol.
146, no. 2, 1999, pp.193-199.
Page No. 111
 Enhancement of Power Quality by SVC with ANN
and Fuzzy Logic Technique
Mr.B.V.S.Acharyulu1,Dr. K. B. Madhu Sahu2
Student, Department of EEE, Aditya Institute of Technology and Management, Tekkali, acharyulu201@yahoo.com1
Dean of Academics, Aditya Institute of Technology and Management, Tekkali.2

ABSTRACT
Electrical distribution system suffers from various problems like reactive power burden, unbalanced loading, voltage
regulation and harmonic distortion. Though DSTATCOMS are ideal solutions for such systems, they are not popular
because of the cost and complexity of control involved. In this paper, Static Var Compensators (SVCs) remain ideal
choice for such loads in practice due to low cost and simple control strategy. These SVCs, while correcting power
factor, inject harmonics into the lines causing serious concerns about quality of the distribution line supplies at PCC.
This paper proposes to minimize the harmonics injected into the distribution systems by the operation of TSC-TCR
type SVC used in conjunction with fast changing loads at LV distribution level. Fuzzy logic system and ANN is
used to solve this nonlinear problem, giving optimum triggering delay angles used to trigger switches in TCR. The
scheme with Artificial Neural Network (ANN) and Fuzzy logic is attractive and can be used at distribution level
where load harmonics are within limits. This project sets the quality of power that is to be provided at the point of
common coupling (PCC). The interface between sources and loads is described as the point of common coupling;
and observance of the design goals will minimize total harmonic distortion in the power system considered.
KeyWords:
Static Var Compensators(SVC), Artificial Neural Network (ANN), Fuzzy logic technique

I. INTRODUCTION
Modern power system is a complex network
comprising of numerous generators, transmission lines,
variety of loads and transformers. As a consequence of
increasing power demand, some transmission lines are
more loaded than was planned when they were built.
With the increased loading of long transmission lines,
the problem of transient stability after a major fault can
become a transmission limiting factor [1].
FACTS controllers are capable of controlling the
network condition in a very fact manner and this feature
of FACTS can be exploited to improve the voltage
stability, and steady state and transient stabilities of a
complex power system [3]-[8]. This allows increased
utilization of existing network closer to its thermal
loading capacity, and thus avoiding the need to
construct new transmission lines.
[9]. SVCs have been used for high performance steady
state and transient voltage control compared with
classical shunt compensation. SVCs are also used to
dampen power swings, improve transient stability, and
reduce system losses by optimized reactive power
control [10]-[11].
Thyristor Controlled Series Capacitor 9TCSC) is of the
important members of FACTS family that is
increasingly applied with long transmission lines by the
utilities in modern power systems. It can have various
roles in the operation and control of power systems,
such as scheduling power flow; decreasing
unsymmetrical components; reducing net loss;
providing voltage support; limiting short-circuit
currents; mitigating sub synchronous resonance (SSR);
damping the power oscillation; and enhancing transient
stability [12]-[14].

Static VAR Compensator (SVC) is a first generation
FACTS device that can control voltage at the required
bus thereby improving the voltage profile of the system.
The primary task of an SVC is to maintain the voltage at
particular bus by means of reactive power compensation
(obtained by varying the firing angle of the thyristors)
A Static Synchronous Series Compensator (SSSC) is a
member of FACTS family which is connected in series
with a power system. It consists of a solid state voltage
source converter which generates a controllable
alternating current voltage at fundamental frequency.

Mr. B. V. S. Acharyulu, Dr. K. B. Madhu Sahu Page No. 112

When the injected voltage is kept in quadrature with the
line current, it can emulate as inductive or capacitive
reactance so as to influence the power flow through the
transmission line [15]. While the primary purpose of a
SSSC is to control power flow in steady state, it can also
improve transient stability of a power system. Among
the available FACTS devices an algorithm is
proposed for on-line control of SVCs
compensating varying unbalanced load by
incorporating ANN to choose the optimum
combination of firing angles of TCR. The resulting
controller is expected to control the SVC so that it
balances the reactive power drawn by the supply,
minimize the reactive power drawn from the
supply and minimize the harmonics injected
into the system in an acceptable time. The inner-
area power system has special characteristic of stability
behavior [19].
II. FACTS CONTROLLERS
2.1 Static Var Compensator (SVC)
Static var systems are applied by utilities in
transmission applications for several purposes. The
primary purpose is usually for rapid control of voltage
at weak points in a network. Installations may be at the
mid point of transmission interconnections or at the line
ends. Static Var Compensator are shunt connected
static generators / absorbers whose outputs are varied so
as to control voltage of the electric power systems. In
its simple form, SVC is connected as Fixed Capacitor-
Thyristor Controlled Reactor (FC-TCR) configuration
as shown in Fig. 1.
The SVC is connected to a coupling transformer that is
connected directly to the ac bus whose voltage is to be
regulated. The effective reactance of the FC-TCR is
varied by firing angle control of the anti parallel
thyristor. The firing angle can be controlled through a
PI (Proportional + Integral) controller in such a way that
the voltage of the bus, where the SVC is connected, is
maintained at the reference value.
Fig. 1 Configuration of SVC
Mr. B. V. S. Acharyulu, Dr. K. B. Madhu Sahu
Static Var Compensator(SVC)is a recently proposed
series connected solid state device that injects voltage
into the system in order to regulate the load side voltage.
The SVC was first installed in 1996. It is normally
installed in a distribution system between the supply
and the critical load feeder. Its primary function is to
rapidly boost up the load-side voltage in the event of a
disturbance in order to avoid any power disruption to that
load.
There are various circuit topologies and control schemes
that can be used to implement a SVC. In addition to
voltage harmonics, voltage sags and swells
compensation, SVC can also perform other tasks such
as: line voltage harmonics compensation, reduction
of transients in voltage and fault current limitations.
The general configuration of the SVC consists of an
Injection/Booster transformer, a Harmonic filter, a
Voltage Source Converter (VSC), DC charging circuit
and a Control and Protection system as shown in Fig. 1.
III. CONTROL STRATEGIES
3.1 Control Strategy of SVC
Fuzzy control systems are rule-based systems in which a
set of fuzzy rules represents a control decision
mechanism to adjust the effects of certain system
stimuli. The aim of fuzzy control systems is normally to
replace a skilled human operator with a fuzzy rule based
system. The fuzzy logic controller provides an
algorithm, which can convert the linguistic control
strategy based on expert knowledge into an automatic
control strategy. Figure 1 illustrates the basic
configuration of a fuzzy logic controller, which consists
of a fuzzification interface, a knowledge base, a
decision-making logic, and a defuzzification interface
[1].
SVC consists of series compensator compensator. The
shunt compensator is controlled by a PWM current
control algorithm, while the series converter is
controlled by a PWM voltage control algorithm.
According to the adopted control scheme, these parts of
SVC have different functions as follows:
3.2 Static Shunt Compensator of SVC
The instantaneous current of the nonlinear load iL is
expanded into 3 terms. The first term iLjp is the load
functions sent from PLL (Phase Locked Loop) in
accordance with equation.(3)
ILdq0=Tabcdq0 iLabc (3)
Page No. 113
By this transform, the fundamental positive sequence
components are transformed into dc Quantities in d and

Figure.3 Network Topology of ANN

The training algorithm used is Levenberg Marquardt
Figure.2 Control of the shunt Converter of the SVC back propagation (LMBP).
The compensator output depends on input and its
q axes, which can easily be extracted by low-pass filter. evolution. The chosen configuration has seven inputs
All harmonic components are transformed into ac three each for reference load voltage and source current
quantities with a fundamental frequency shift. respectively, and one for output of error (PI) controller.
The neural network trained for outputting fundamental
ILq= îLq+ ïLq (4) reference currents [20]. The signals thus obtained are
Since, iL = is + ic (5) compared in a hystersis band current controller to give
switching signals.
This means there is no harmonics and reactive
components in the system currents. The switching loss
can cause the dc link capacitor voltage to decrease.
Other disturbances, such as unbalances and sudden
variations of loads can also cause this voltage to
fluctuate. In order to avoid this, in Figure 2. A PI
controller is used. The input of the PI controller is the
error between the actual capacitor voltage and the
desired value, its output then added to the reference
current component in the d-axis to form a new
IV. DESIGNING & TRAINING OF ANN
An ANN is essentially a cluster of suitably
interconnected non-linear elements of very simple form
that possess the ability of learning and adaptation. These Figure.4 Block diagram of ANN-based compensator
networks are characterized by their topology, the way in
which they communicate with their environment, the 4.1Central Control with fuzzy controller
manner in which they are trained and their ability to The central control generates the refrence signals for
process information [18]. Their ease of use, inherent both the shunt and series converters of the DPFC. It is
reliability and fault tolerance has made ANNs a viable focused on the DPFC tasks at the power-system level,
medium for control. An alternative to fuzzy controllers such as power-flow control, low-frequency power
in many cases, neural controllers share the need to oscillation damping, and balancing of asymmetrical
replace hard controllers with intelligent controllers in components.According to the system requirement, the
order to increase control quality [19]. A feed forward central control gives correspnding voltage-reference
neural network works as compensation signal generator. signals for the series converters and reactive current
This network is designed with three layers. The input signal for the shunt converter. All the reference signals
layer with seven neurons, the hidden layer with 21 and generated by the central control are at the fundamental
the output layer with 3 neurons. Activation functions frequency.
chosen are tan sigmoidal and pure linear in the hidden
and output layers respectively.

Mr. B. V. S. Acharyulu, Dr. K. B. Madhu Sahu Page No. 114

4.2 Fuzzy Logic Controllers:
Fuzzy set comprises from a membership function which
could be defines by parameters. The value between 0 and 1
reveals a degree of membership to the fuzzy set. The
process of converting the crisp input to a fuzzy value is
called as “fuzzificaton.” The output of the Fuzzier module
is interfaced with the rules.The basic operation of FLC is
constructed from fuzzy control rules utilizing the values of
fuzzy sets in general for the error
and the change of error and control action The results are
combined to give a crisp output controlling the output
variable and this process is called as
“DEFUZZIFICATION.”
Fuzzy logic is a form of multi valued logic that can be
taken from a fuzzy set theory. Fuzzy logic variables can
have a trth logic between 0 and 1. Fuzzy logic is a good
mean to control a system where there is no specific relation
between input and output quantities. It has a simple rule
based if-then approach to solve a control problem rather
than modeling total system. This logic depends on
operator’s experience not on modeling of the system.
Because of its usefulness in reducing complexity in
mathematical models it is gaining more attention. In power
system area fuzzy logic is used in stability studies, unit
commitment, and reactive power control in distribution
systems etc. fuzzy logic is made from fuzzification,
knowledge rule base and defuzzification.
Steps to design fuzzy based VSC:
1) Choose the inputs to FLC: two inputs generator
speed variation and speed derivative deviation us
2) Choose membership functions for inputs in fuzzy set
notation. Triangular membership functions are used.
Defining knowledge rule base.
Figure. 5(a) Grid voltage deviation (∆e)
Figure.5(b) Grid voltage derivative deviation (∆e)
Fig. 5(c) Fuzzy output
Figure.5(a,b&c)Membership functions of inputs & outputs
Mr. B. V. S. Acharyulu, Dr. K. B. Madhu Sahu
V. SIMULATION
In order to show the performance of the SVC in
harmonics, voltage sags and swells mitigation, a simple
distribution network was simulated using MATLAB
(Fig. 2). A SVC was connected to the system through a
series transformer with a capability to insert a
maximum voltage of 50% of the phase to ground system
nominal voltage. In this simulation the In-Phase
Compensation (IPC) method was used. The load
considered in the study is a 5.5 MVA capacity with
0.92 p.f, lagging.
In order to understand the performance of the SVC under
unbalanced conditions, Single-phase voltage sag was
simulated and the results are shown in Fig. 6. The
supply voltage with one phase voltage dropped down to
50% is shown in Fig. 6a. The SVC injected voltage and
the load voltage are shown in Fig. 6b and c,
respectively. As can be seen from the results, the SVC
was able to produce the required voltage component
rapidly and helped to maintain a balanced and constant
load voltage at 1.00 p .u.
Voltage swells: As can be seen from the The
performance of the SVC with an unbalanced voltage
swell is shown in Fig. 7. In this case, two of the three
phases are higher by 25% than the third phase as shown
in Fig. 7a. The injected voltage that was produced
by SVC in order to correct the load voltage and the
load voltage are shown in Fig. 7b and c, respectively.
Notice the constant and balanced voltage at the load
throughout the simulation, including during the
unbalanced voltage swell event.
VI. CONCLUSION
Static Var Compensators (SVCs) remain ideal
choice for fast changing loads due to low cost
and simple control strategy. DSTATCOM
being ideal solution suffers from serious
limitation of high cost and complex control
strategy. The SVCs, while correcting power
factor, inject harmonics in distribution lines.
The operation of thyristor- controlled
compensators at various conduction angles can
be used advantageously to meet the
unbalanced reactive power demands in a
fluctuating load environment. The proposed
ANN based approach can be effectively used to
reduce and balance the reactive power drawn
from the source under unbalanced loadings
while keeping the harmonic injection into the
power system low. The case study proves that the
percentage THD under optimized condition is
Page No. 115
 Order of WITHOUT WITHOUT SVC with SVC with UPQC with UPQC with
harmonics SVC utility SVC utility FUZZY FUZZY ANN ANN controller
side voltage Load side controller controller controller Utility side
current utility side utility side utility side current
voltage current voltage
3rd &5th 4.2 24.2 2.99 2.99 1.2 1.2
5th & 7th 4.2 24.6 3.42 3.42 1.19 1.19
7th &9th 4.2 24.6 2.18 2.18 1.3 1.3
Table.1 Voltage and current harmonics (THD‟s) of SVC
much less than the percentage THD under unity
power factor condition (unoptimised). The
computational time required was found to be
satisfactory for the system considered. The scheme
can be effectively used at distribution level
where the load harmonics is not a major problem.
The simulation results showed clearly the performance
of the SVC in mitigating voltage sags and swells.
The closed loop control schemes of direct current control,
for the proposed SVC have been described. A suitable
mathematical have been described which establishes the
fact that in both the cases the compensation is done but the Figure 7. Three-phase voltage swell; (a): Source
response of ANN and FUZZY controller is faster and the voltages; (b): Injected voltages; (c): Load voltages
THD is minimum for the both the voltage and current
which is evident from the plots and comparison Table.1.
Proposed model for the SVC is to compensate input
VII. REFERENCES
voltage harmonics and current harmonics caused by non-
linear load. The work can be extended to compensate the
supply voltage and load current imperfections such as sags,
[1] Hamid Reza Mohammadi, Ali Yazdian Varjani, and
swells, interruptions, voltage imbalance, flicker, and
Hossein Mokhtari,“Multiconverter Unified Power-
current unbalance. Proposed SVC can be implemented
using simple analog hardware, because it is having PLL Quality Conditioning System: SVC” IEEE
and Hysteresis blocks. RANSACTIONS ON POWER DELIVERY, VOL.
24,NO. 3, JULY 2009.
[2] R.Rezaeipour and A.Kazemi, “Review of Novel
control strategies for UPFC” Internal Journal of
Electric and power Engineering 2(4) 241-247, 2008.
[3] S. Ravi Kumar and S.Siva Nagaraju“Simulation of
DSTATCOM and DVR in power systems” Vol. 2, No.
3, June 2007 ISSN 1819-6608 ARPN Journal of
Engineering and Applied Sciences.
[4] M.V.Kasuni Perera” Control of a Dynamic Voltage
Restorer to compensate single phase voltage sags”
Master of Science Thesis, Stockholm, Sweden 2007.
[5] M. Basu, S. P. Das, and G. K. Dubey, “Comparative
evaluation of two models of SVC for suitable interface
to enhance power quality,” Elect.Power Syst. Res., pp.
821–830, 2007.
Figure 6. Three-phase voltage sags; (a): Source
voltages; (b): Injected voltages; (c): Load voltages

Mr. B. V. S. Acharyulu, Dr. K. B. Madhu Sahu Page No. 116

 Electricity through Wireless Transmission
Development and Possibility
Ramakrishna Raghutu1 , K.Perindevi2, D.Rohini3
Electrical Engine
Engineering, GMRIT, RAJAM (AP), INDIA
ramakrishnaree@gmail.com ,perinkoppala@yahoo.in, Rohinid.205@gmail.com

ABSTRACT
In the present
esent paper the various technologies available so far for wireless transmission of electricity and the need for
a Wireless System of Energy Transmission is being discussed to find its possibility in actual practices, their
advantages, disadvantages and economical
nomical consideration. This paper is mainly concentrated on : i) The most popular
concept known as Tesla Theory, ii) The microwave power transmission (MPT) called Solar power satellite, and iii)
The highly efficient fibre lasers for wireless power transmis
transmission.
sion. Many concepts, research papers, patents are
available on wireless transmission of electricity but the commercial technologies are yet to be materialized. The
paper also discusses the possible ways to get useful and practical results out of all resear
research
ch carried out so far
elsewhere.
An electric toothbrush's base and handle contain coils
I. INTRODUCTION that allow the battery to recharge.
In our present electricity generation system we waste
The action
on of an electrical transformer is the simplest
more than half of its resources. Especially the
instance of wireless energy transfer. The primary and
transmission and distribution losses are the main
secondary circuits of a transformer are not directly
concern of the present power technology. Much of thi this
connected. The transfer of energy takes place by
power is wasted during transmission from power plant
electromagnetic coupling through a process known as
generators to the consumer. The resistance of the wire
mutual induction. (An added benefit is the capability
used in the electrical grid distribution system causes a
to step the primary voltage either up or down.) The
loss of 26-30%
30% of the energy generated. This loss
battery charger of an electric toothbrush is an example
implies that our present system of elec electrical
of how this principle can be used. The main drawback
distribution is only 70-74%
74% efficient. We have to think
to induction, however, is the short range.
range The receiver
of alternate state - of - art technology to transmit and
must be very close to the transmitter or induction unit
distribute the electricity. Now- a- days global scenario
in order to inductively couple with it.
has been changed a lot and there are tremendous
development in every field. If we don’t keep pace with
the development of new power technology we have to
face a decreasing trend in the development of power
sector. The transmission of power without wires may
be one noble alternative for electricity transmission.
II. THE TECHNOLOGIES AVAILABLE
VAILABLE - Near field
These are wireless transmission techniques over
distances comparable to, or a few times the diameter
of the device(s).
1. Induction
A Splash power mat uses induction to recharge
multiple devices simultaneously.

2. Resonant induction

In 2006, Marin Soljačić and other researchers at the

Massachusetts Institute of Technology applied the near
field behaviour well known in electromagnetic theory

Ramakrishna Raghutu, K.Perindevi, D.Rohini Page No. 117

to a wireless power transfer concept based on coupled
resonators. In a theoretical analysis, they demonstrate
that by sending electromagnetic waves around in a
highly angular waveguide, evanescent waves are
produced which carry no energy. If a proper resonant
waveguide is brought near the transmitter, the
evanescent waves can allow the energy to tunnel
(specifically evanescent wave coupling, the
electromagnetic equivalent of tunneling) to the power
drawing waveguide, where they can be rectified into
DC power. Since the electromagnetic waves would
tunnel, they would not propagate through the air to be
absorbed or dissipated, and would not disrupt
electronic devices or cause physical injury like
microwave or radio wave transmission might.
Researchers anticipate up to 5 meters of range for the
initial device, and are currently working on a
functional prototype.
On June 7, 2007, it was reported that a prototype
system had been implemented. The MIT researchers
successfully demonstrated the ability to power a 60-
watt light bulb from a power source that was seven
feet (2 meters) away at roughly 40% efficiency.
"Resonant inductive coupling" has key implications in
solving the two main problems associated with non-
resonant inductive coupling and electromagnetic
radiation, one of which is caused by the other; distance
and efficiency. Electromagnetic induction works on
the principle of a primary coil generating a
predominantly magnetic field and a secondary coil
being within that field so a current is induced within
its coils. This causes the relatively short range due to
the amount of power required to produce an
electromagnetic field. Over greater distances the non-
resonant induction method is inefficient and wastes
much of the transmitted energy just to increase range.
This is where the resonance comes in and helps
efficiency dramatically by "tunneling" the magnetic
field to a receiver coil that resonates at the same
frequency. Unlike the multiple-layer secondary of a
non-resonant transformer, such receiving coils are
single layer solenoids with closely spaced capacitor
plates on each end, which in combination allow the
coil to be tuned to the transmitter frequency thereby
eliminating the wide energy wasting "wave problem"
and allowing the energy used to focus in on a specific
frequency increasing the range.
Beginning in the early 1960s resonant inductive
wireless energy transfer was used successfully in
implantable medical devices including such devices as
Ramakrishna Raghutu, K.Perindevi, D.Rohini
pacemakers and artificial hearts. While the early
systems used a resonant receiver coil later systems
implemented resonant transmitter coils as well. These
medical devices are designed for high efficiency using
low power electronics while efficiently
accommodating some misalignment and dynamic
twisting of the coils. The separation between the coils
in implantable applications is commonly less than 20
cm. Today resonant inductive energy transfer is
regularly used for providing electric power in many
commercially available medical implantable devices.
Wireless electric energy transfer for experimentally
powering electric automobiles and buses is a higher
power application (>10kW) of resonant inductive
energy transfer. High power levels are required for
rapid recharging and high energy transfer efficiency is
required both for operational economy and to avoid
negative environmental impact of the system. An
experimental electrified roadway test track built circa
1990 achieved 80% energy efficiency while
recharging the battery of a prototype bus at a specially
equipped bus stop. The bus could be outfitted with a
retractable receiving coil for greater coil clearance
when moving. The gap between the transmit and
receive coils was designed to be less than 10 cm when
powered. In addition to buses the use of wireless
transfer has been investigated for recharging electric
automobiles in parking spots and garages as
well.Some of these wireless resonant inductive devices
operate at low milli Watt power levels and are battery
powered. Others operate at higher kilowatt power
levels. Current implantable medical and road
electrification device designs achieve more than 75%
transfer efficiency at an operating distance between
the transmit and receive coils of less than 10 cm.
A. Far field
Means for long conductors of electricity forming part
of an electric circuit and electrically connecting said
ionized beam to an electric circuit.
These methods achieve longer ranges, often multiple
kilometre ranges, where the distance is much greater
than the diameter of the device(s).
1) Radio and microwave
The earliest work in the area of wireless transmission
via radio waves was performed by Heinrich Rudolf
Hertz in 1888. A later Guglielmo Marconi worked
with a modified form of Hertz's transmitter. Nikola
Tesla also investigated radio transmission and
reception.
Japanese researcher Hidetsugu Yagi also investigated
wireless energy transmission using a directional array
antenna that he designed. In February 1926, Yagi and
Page No. 118
Uda published their first paper on the tuned high
high-gain satellite to be built in high earth orbit to collect
directional array now known as the Yagi antenna. sunlight and convert that energy into microwaves, then
While it did not prove to be particularly useful for beamed to a very large antenna on earth, the
power transmission, this beam antenna tenna has been microwaves would be converted into conventional
widely adopted throughout the broadcasting and electrical power.
wireless telecommunications industries due to its
excellent performance characteristics.
High power
In this remarkable discovery of the "True Wireless"
and the principles upon which transmissio
transmission and
reception, even in the present day systems, are based,
Dr. Nikola Tesla shows us that he is indeed the "Father
of the Wireless." The most well known and famous
Wardenclyffe Tower (Tesla Tower) was designed and
constructed mainly for wireless transmis
transmission of
electrical power, rather than telegraphy. The most
popular concept known is Tesla Theory in which it
was firmly believed that Wardenclyffe (Fig.1) would
permit wireless transmission and reception across
large distances with negligible losses. In sp
spite of this
he had made numerous experiments of high quality to Fig.2. The basis for Tesla’s system for the wireless
validate his claim of possibility of wireless transmission of electrical power.
transmission of electricity (Fig.2). But this was an
William C. Brown, the leading authority on wireless
unfortunate incidence that people of that century was
power transmission technology, has loaned this
not in a position to recognise his sple splendid work
demonstration unit to the Texas Space Grant
otherwise today we may transmit electricity wirelessly
Consortium to show how w power can be transferred
and will convert our mother earth a wonderful adobe
through free space by microwaves. A block diagram of
full of electricity.
the demonstration components is shown below.
The primary components include a microwave source,
a transmitting antenna, and a receiving rectenna.

Fig.1. The 187-foot

foot Wardenclyffe Tower (Tesla
Tower) in 1903.
This was to be the first broadcasting system in the
world.
rld. Tesla wanted to transmit electricity from this
Tower to the whole globe without wires using the
Ionosphere. The source of the transmitted electricity
was to be the Niagara Falls power plant.
The modern ideas are dominated by microwave power
transmission
ion (MPT, Figure 3) called Solar power
Fig.3 .Microwave power transmission.
trans
The microwave source consists of a microwave oven
magnetron with electronics to control the output
power. The output microwave power ranges from 50
W to 200 W at 2.45 GHz. A coaxial cable connects the
output of the microwave source to a coax-to-
coax
waveguide
veguide adapter. This adapter is connected to a
waveguide ferrite circulator which protects the
microwave source from reflected power. The

Ramakrishna Raghutu, K.Perindevi, D.Rohini Page No. 119

circulator is connected to a tuning waveguide section 3. Coupling to the ionosphere using propagation
to match the waveguide impedance to the antenna through electrified gases(Fig.4&5).
input impedance.
The slotted waveguide antenna consists of 8
waveguide sections with 8 slots on each section. These
64 slots radiate the power uniformly through free
space to the rectenna. The slotted waveguide antenna
is ideal for power transmission because of its high
aperture efficiency (> 95%) and high power handling
capability.

A rectifying antenna called a rectenna receives the
transmitted power and converts the microwave power
to direct current (DC) power. This demonstration
rectenna consists of 6 rows of dipoles antennas where
8 dipoles belong to each row. Each row is connected
to a rectifying circuit which consists of low pass filters
and a rectifier. The rectifier is a Ga As Schottky
barrier diode that is impedance matched to the dipoles
by a low pass filter. The 6 rectifying diodes are
connected to light bulbs for indicating that the power
is received. The light bulbs also dissipated the
received power. This rectenna has a 25% collection
and conversion efficiency, but rectennas have been
tested with greater than 90% efficiency at 2.45 GHz.
The transmission of power without wires is not a
theory or a mere possibility, it is now a reality. The
electrical energy can be economically transmitted
without wires to any terrestrial distance, many
researchers have established in numerous
observations, experiments and measurements,
qualitative and quantitative. These have demonstrated
that it is practicable to distribute power from a central
plant in unlimited amounts, with a loss not exceeding
a small fraction of one per cent, in the transmission,
even to the greatest distance, twelve thousand miles -
to the opposite end of the globe. This seemingly
impossible feat can now be readily performed by
electrical researchers familiar with the design and
construction of my "high-potential magnifying
transmitter," There were three popular theories present
in the literature of the late 1800's and early 1900's.
They were:
1. Transmission through or along the Earth,
Fig.4. Diagram showing the transmitting & receiving
circuit for the transmission & reception of electric
power by wireless.
Fig.5. Two optical forms of wireless antennae formed
of search light beam-ionised atmospheric streams.
Low power
A new company, Powercast introduced wireless power
transfer technology using RF energy at the 2007
Consumer Electronics Show, winning best Emerging
Technology. The Powercast system is applicable for a
number of devices with low power requirements. This
could include LEDs, computer peripherals, wireless
sensors, and medical implants. Currently, it achieves a
maximum output of 6 volts for a little over one meter.
It is expected for arrival late 2007.
A different low-power wireless power technology has
been proposed by Landis.
2) Laser
With a laser beam centered on its panel of
photovoltaic cells, a lightweight model plane makes
the first flight of an aircraft powered by a laser beam
inside a building at NASA Marshall Space Flight
Center.

2. Propagation as a result of terrestrial resonances,

Ramakrishna Raghutu, K.Perindevi, D.Rohini Page No. 120

In the case of light, power can be transmitted by
converting electricity into a laser beam that is then
fired at a solar cell receiver. This is generally known
as "powerbeaming". Its drawbacks are:
1. Conversion to light, such as with a laser, is
moderately inefficient (although quantum
cascade lasers improve this)
2. Conversion back into electricity is moderately
inefficient, with photovoltaic cells achieving
40%-50% efficiency. (Note that conversion
efficiency is rather higher with monochromatic
light than with insolation of solar panels).
3. Atmospheric absorption causes losses.
4. As with microwave beaming, this method
requires a direct line of sight with the target.
NASA has demonstrated flight of a lightweight model
plane powered by a laser beam.It has been proven that
electrical energy can be propagated around the world
between the surface of the Earth and the ionosphere at
extreme low frequencies in what is known as the
Schumann Cavity. Knowing that a resonant cavity can
be excited and that power can be delivered to that
cavity similar to the methods used in microwave ovens
for home use, it should be possible to resonate and
deliver power via the Schumann Cavity to any point
on Earth. This will result in practical wireless
transmission of electrical power.
The intent of the experiments and the laboratory Tesla
had constructed was to prove that wireless
transmission of electrical power was possible.
Although Tesla was not able to commercially market a
system to transmit power around the globe, modern
scientific theory and mathematical calculations
support his contention that the wireless propagation of
electrical power is possible and a feasible alternative
to the extensive and costly grid of electrical
transmission lines used today for electrical power
distribution.
Power transmission system using directional
ultrasound for power transmission includes a
transmitting device and a receiving device. The
transmitting device has a set of ultrasound transducers
forming an ultrasound transducer array, wherein the
array is a set of spaced individual transducers placed
in the X-Y plane disposed to generate an ultrasound
beam in the Z direction (Fig.6).
Another possibility is to use highly efficient fibre
lasers for wireless power transmission where the
possibilities are similar to microwaves concept but
Ramakrishna Raghutu, K.Perindevi, D.Rohini
lasers emit energy at frequencies much higher that
microwaves. For several years NASA, ENTECH, and
UAH have been working on various aspects of
collection of the laser radiation and conversion to
electrical power for laser wireless power transmission.
Fig.6. System and method for wireless electrical
power Transmission (directional ultrasound for power
Transmission).
III. MERITS, DEMERITS & ECONOMICS OF
WIRELESS TECHNOLOGIES
Merits
An electrical distribution system, based on this method
would eliminate the need for an inefficient, costly, and
capital intensive grid of cables, towers, and
substations. The system would reduce the cost of
electrical energy used by the consumer and rid the
landscape of wires, cables, and transmission towers.
There are areas of the world where the need for
electrical power exists, yet there is no method for
delivering power. Africa is in need of power to run
pumps to tap into the vast resources of water under the
Sahara Desert. Rural areas, such as those in China,
require the electrical power necessary to bring them
into the 20th century and to equal standing with
western nations. The wireless transmission will solve
many of these problems.
The electrical energy can be economically transmitted
without wires to any terrestrial distance, so there will
be no transmission and distribution loss. More
efficient energy distribution systems and sources are
needed by both developed and under developed
nations. In regards to the new systems, the market for
wireless power transmission is enormous. It has the
potential to become a multi-billion dollar per year
market. The increasing demand for electrical energy in
industrial nations is well documented. If we include
the demand of third world nations, pushed by their
increasing rate of growth, we could expect an even
faster rise in the demand for electrical power in the
near future. These systems can only meet these
requirements with 90–94 % efficient transmission.
Page No. 121
High Transmission Integrity and Low Loss: - To
transmit wireless power to any distance without limit.
It makes no difference what the distance is. The
efficiency of the transmission can be as high as 96 or
97 per cent, and there are practically no losses.
Demerits
Biological Impact: - One common criticism of the
Tesla wireless power system is regarding its possible
biological effects. Calculating the circulating reactive
power, it was found that the frequency is very small
and such a frequency is very biologically compatible.
Economic Impact
The concept looks to be costly initially. The
investment cost of Tesla Tower was $150,000 (1905).
In terms of economic theory, many countries will
benefit from this service. Only private, dispersed
receiving stations will be needed. Just like television
and radio, a single resonant energy receiver is
required, which may eventually be built into
appliances, so no power cord will be necessary!
Monthly electric utility bills from old-fashioned,
fossil-fuelled, loss prone electrified wire-grid delivery
services will be optional, much like “cable TV” of
today. In the 21st century, “Direct TV” is the rage,
which is an exact parallel of Tesla’s “Direct
Electricity.”
IV. ADDITIONAL REMARKS
Many concepts, research papers, patents are available
on wireless transmission of electricity but most
research work were carried out in isolation, so it needs
a joint collaborative efforts to get a very useful results
on this advanced technology on power transmission
for the benefit of mankind globally in future.
Whatever the future may bring, the universal
application of these great principles is fully assured,
though it may be long in coming. With the opening of
the first power plant, incredulity will give way to
wonderment, and this to ingratitude, as ever before.
The world is still not able to achieve the benefit of the
God gifted potential of Dr N.Tesla. People neglected
him and his good work. He deserved much better
treatment from the tycoons of his age, than to spend
the last 40 years of his life in abject poverty. However,
he was too much of a gentleman to hold a grudge.
Instead, regarding the magnifying transmitter, Tesla
wrote in his autobiography, “I am unwilling to accord
to some small-minded and jealous individuals the
satisfaction of having thwarted my efforts. These men
are to me nothing more than microbes of a nasty
Ramakrishna Raghutu, K.Perindevi, D.Rohini
disease. My project was retarded by laws of nature.
The world was not prepared for it. It was too far ahead
of time. But the same laws will prevail in the end and
make it a triumphal success”. If this has had not been
happened, then today we will be in a wonder world of
plenty of power using the technology of wireless
transmission of electricity.
V. CONCLUSION
The transmission of power without wires is not a
theory or a mere possibility, it is now a reality. The
electrical energy can be economically transmitted
without wires to any terrestrial distance. Many
researchers have established in numerous
observations, experiments and measurements,
qualitative and quantitative. Wireless transmission of
electricity have tremendous merits like high
transmission integrity and Low Loss (90–97%
efficient) and can be transmitted to any where in the
globe and eliminate the need for an inefficient, costly,
and capital intensive grid of cables, towers, and
substations. The system would reduce the cost of
electrical energy used by the consumer and get rid of
the landscape of wires, cables, and transmission
towers. It has negligible demerits like reactive power
which was found insignificant and biologically
compatible.
It has a tremendous economic impact to human
society. Many countries will benefit from this service.
Monthly electric utility bills from old-fashioned,
fossil-fuelled, loss prone electrified wire-grid delivery
services will be optional, much like “cable TV” of
today.
VI. REFERENCES
1. Nikola Tesla, “The Transmission of Electrical
Energy Without Wires as a Means for Furthering
Peace,” Electrical World and Engineer. Jan. 7, p.
21, 1905.
2. Nikola Tesla, My Inventions, Ben Johnston, Ed.,
Austin, Hart Brothers, p. 91,1982.
3. Thomas F. Valone, “ Tesla’s Wireless Energy...
For the 21st Century!!! One Step Beyond Direct
TV!!!” Extra Ordinary Technology, 1, no. 4, Oct /
Nov / Dec 2003.
4. James O. McSpadden, “Wireless Power
Transmission Demonstration”, Texas A&M
University, June, 1997.
Page No. 122
 High Voltage Direct Current Transmission Lines
Abinash Dash, Suraj Maharana
Students, Dept. of E.E.E, Gandhi Institute For Education and Technology, Bhubaneswar,
Ab.sonu2009@gmail.com,skmspeed@gmail.com

ABSTRACT

Rapid developments in the field of power electronic devices with turn off capability like insulated gate
bipolar transistors (IGBT) and gate turn off transistors (GTO), makes the voltage source converters (VSC) getting
more and more attractive for High voltage direct current transmission (HVDC).This new innovative technology
provides substantial technical and economical advantages for direct applications compared to conventional HVDC
transmission systems based on thyristor technology. VSC Application for HVDC systems of high power rating (up
to 200MW) which are currently in discussion for several projects are mentioned. The underlying technology of VSC
based HVDC systems, its Characteristics and the working principle of VSC based HVDC system are also presented.
This paper concludes with a brief set of guidelines for choosing VSC based HVDC systems in today’s electricity
system development.

Gas Insulated Substation

Prakash Bari, Debasish Sahoo
Students, Dept. of E.E.E, Gandhi Institute For Education and Technology, Bhubaneswar,
prakash.bari@yahoo.in, debasishsahoo2021@gmail.com

ABSTRACT

This paper presents about “Gas Insulated Substation”. Conventional substation requires large
installation size, protection against atmospheric pollution and moisture, noiseless operation, non-
explosive frame resistant, reduced maintenance, minimum radio interference, and totally enclosed
substation. Gas insulated substations are particularly suitable for underground construction beneath
public buildings. Gas insulated substation mainly used for a power transmission system or a sub-
station system, of which outgoing bus bar is shortened to reduce consumption of the outgoing bus
bar Sulfur Hexafluoride. SF6 acts as insulation between live parts & the earthed metal closure. Gas
insulated switchgear is used in industrial areas to fulfil high-energy demands by space saving design
with a minimum of cost.

Page No. 123

 Flexible AC Transmission Systems
(FACTS)
Sumit Mahato,Saurav Kumar
Students, Dept. of E.E.E, Gandhi Institute For Education and Technology, Bhubaneswar
Sumitmahato2011@gmail.com,rahulsaurav56@gmail.com

ABSTRACT

This paper presents about “Flexible AC transmission Systems”. Every power industry requires their power
transmission system to be highly stable with maximum thermal limits. Flexible ac transmission system is a new
integrated concept based on power electronic switching converters and dynamic controllers to enhance the system
utilization and power transfer capacity as well as the stability, security, reliability and power quality of ac system
interconnections.

Brushless DC Motor
Deepak Kumar Mohapatra, Pijush Banerjee
Students, Dept. of E.E.E
Gandhi Institute For Education and Technology, Bhubaneswar
deepak.lord@yahoo.co.in

ABSTRACT:

Permanent magnet (PM) brushless DC motors (BLDCM) are generated by virtually inverting the stator and
rotor of PM DC motors. These motors are actually fed by rectangular AC waveform. The advantage is the removal
of brushes, leading to the elimination of many problems associated with brushes. Another advantage is the ability to
produce a larger torque because of the rectangular interaction between current and flux. The Simulation is done by
widely used MATLAB SIMULINK software.

Page No. 124

 Key Note-03

Coordinated Design of PSS and TCSC using PSO

Technique for Power System Stability Enhancement
Dr. Sidhartha Panda

Department of Electrical Engineering, Veer Surendra Sai University of Technology (VSSUT),

Burla-768018, Odisha, India, E-mail:panda_sidhartha@rediffmail.com

ABSTRACT: Power system stability improvement by coordinated design of a Power System Stabilizers (PSS) and
a Thyristor Controlled Series Compensator (TCSC) controller is addressed in this paper. Particle Swarm
Optimisation (PSO) technique is employed for optimization of the parameter-constrained nonlinear optimization
problem implemented in a simulation environment. The proposed controllers are tested on a weakly connected
power system. The eigenvalue analysis and simulation results show the effectiveness and robustness of proposed
controllers to improve the stability performance of power system by efficient damping of low frequency oscillations
under severe disturbance.
KEYWORDS: power system stability, PSS, TCSC, co-ordinated design, particle swarm optimization.

1. INTRODUCTION that required linearization of the system involved.

Low frequency oscillations are observed when large
power systems are interconnected by relatively weak
tie lines. These oscillations may sustain and grow to
cause system separation if no adequate damping is
available [1-3]. Power system stabilizers (PSS) are
now routinely used in the industry to damp out power
system oscillations. However, during some operating
conditions, this device may not produce adequate
damping, and other effective alternatives are needed in
addition to PSS. Recent development of power
electronics introduces the use of Flexible AC
Transmission System (FACTS) controllers in power
systems. FACTS controllers are capable of controlling
the network condition in a very fast manner and this
unique feature of FACTS can be exploited to improve
the stability of a power system [4]. Thyristor
Controlled Series Compensator (TCSC) is one of the
important members of FACTS family that is
increasingly applied by the utilities in modern power
systems with long transmission lines. It can have
various roles in the operation and control of power
systems, such as scheduling power flow; decreasing
unsymmetrical components; reducing net loss;
providing voltage support; limiting short-circuit
currents; mitigating subsynchronous resonance (SSR);
damping the power oscillation; and enhancing
transient stability. Most of these proposals presented
in literature are based on small disturbance analysis
However, linear methods cannot properly capture
complex dynamics of the system, especially during
major disturbances. This presents difficulties for
tuning the FACTS controllers in that the controllers
tuned to provide desired performance at small signal
condition do not guarantee acceptable performance in
the event of major disturbances.
The problem of PSS parameter tuning in the presence
of FACTS controllers is a complex exercise as
uncoordinated local control of FACTS devices and
PSS may cause destabilizing interactions. A number of
conventional techniques have been reported in the
literature pertaining to design problems of
conventional power system stabilizers namely: the
eigenvalue assignment, mathematical programming,
gradient procedure for optimization and also the
modern control theory. Unfortunately, the
conventional techniques are time consuming as they
are iterative and require heavy computation burden
and slow convergence. In addition, the search process
is susceptible to be trapped in local minima and the
solution obtained may not be optimal. Further, the
controllers should provide some degree of robustness
to the variations in system parameters, loading
conditions, and configurations as the machine
parameters change with operating conditions. To
improve overall system performance, many researches
proposed the coordination between PSS and FACTS
Power Oscillation Damping (POD) controllers [5, 6].

Dr. Sidhartha Panda Page No. 125

 Key Note-03

Some of these methods are based on the linearized Where,

power system models and others on complex nonlinear X C = Nominal reactance of the fixed capacitor C
simulations, based on the small signals.
X P = Inductive reactance of inductor L connected in
Recently, Particle Swarm Optimization (PSO) parallel with C
technique appeared as a promising algorithm for σ = 2( π − α ) = Conduction angle of TCSC
handling the optimization problems. PSO is a
Controller
population based stochastic optimization technique,
inspired by social behaviour of bird flocking or fish XC
k= = Compensation ratio
schooling [7]. PSO shares many similarities with XP
Genetic Algorithm (GA); like initialization of
population of random solutions and search for the
optimal by updating generations. However, unlike GA, Since the relationship between α and the
PSO has no evolution operators such as crossover and equivalent fundamental frequency reactance offered
mutation. One of the most promising advantages of by TCSC, XTCSC(α), is a unique-valued function, the
PSO over GA is its algorithmic simplicity as it uses a
TCSC is modelled here as a variable capacitive
few parameters and easy to implement. In PSO, the
reactance within the operating region defined by the
potential solutions, called particles, fly through the
problem space by following the current optimum limits imposed by α. Thus XTCSC min ≤ XTCSC ≤ XTCSC max,
particles [7]. In view of the above, PSO is employed in with XTCSC max = XTCSC (α min) and XTCSC min = XTCSC
the present work to simultaneously tune the
(1800) = XC. In this paper, the TCSC controller is
parameters of the PSS and TCSC controllers.
assumed to operate only in the capacitive region, i.e.,
In this paper, PSO based optimal tuning algorithm is α min > α r where α r corresponds to the resonant point,
used to coordinate the PSS and TCSC controller
simultaneously. By minimizing the objective function as the inductive region associated with 900 < α <α r
in which the influences of both PSS and TCSC induces high harmonics that cannot be properly
controllers are considered, interactions among these modelled in stability studies.
controllers are improved. The proposed controllers
have been applied and tested on a weakly connected 3. POWER SYSTEM UNDER STUDY
power system under wide range of loading conditions
and severe disturbances. In addition, the proposed Tr. Line-1
VT VB
controllers are also tested under different small G
disturbances. The eigenvalue analysis and non-linear XT Z=R+jX
TCSC
simulation results are presented to demonstrate the
effectiveness and robustness of the proposed Tr. Line-2
controllers to enhance power system stability. Y=G+jB

2. MODELLING THE TCSC DYNAMICS

Basically, the TCSC consists of three components:
Figure 1. Single-line diagram of the study system.
capacitor banks, bypass inductor and bidirectional
thyristors. The control of the TCSC is achieved by the
The SMIB power system with TCSC (shown in Figure
firing angle signal α, which changes the fundamental 1), is considered in this study. The generator has a
frequency reactance of the compensator. There exists a local load of admittance Y = G + jB and a double
steady-state relationship between the firing angle α circuit transmission line of total impedance Z = R +
and the reactance XTCSC( α
). This relationship can be jX. In the figure VT and VB are the generator terminal
described in the following equation [8]: and infinite bus voltage respectively and XT is the
reactance of the transformer.
XC2 σ + sinσ
XTCSC(α ) = XC − The generator is represented by the third-order model
( XC − X P ) π (1)
comprising of the electromechanical swing equation
4XC2 cos2(σ / 2) (k tan( kσ / 2) − tan(σ / 2)) and the generator internal voltage equation. The state
+
( XC − X P ) ( k 2 −1) π equations may be written as [9]:

Dr. Sidhartha Panda Page No. 126

 Key Note-03

• particles strive to improve themselves by imitating traits

1 from their successful peers. Further, each particle has a
ω = [ PM − Pe − D(ω − 1)] (2)
M memory and hence it is capable of remembering the best
position in the search space ever visited by it. The
•
position corresponding to the best fitness is known as
δ = ωb (ω −1) (3) pbest and the overall best out of all the particles in the
population is called gbest [7, 10].
where, PM and Pe are the input and output powers of The modified velocity and position of each particle
the generator respectively; M and D are the inertia can be calculated using the current velocity and the
distance from the pbestj,g to gbestg as shown in the
constant and damping coefficient respectively; ωb is following formulas [7 10].:
the synchronous speed, δ and ω are the rotor angle v(jt,+g1) = w*v(jt,)g +c1*r1( )*( pbest j,g −x(jt,)g) +c2 * r2( )*(gbestg − x(jt,)g)
and speed respectively.
• (8)
The generator power Pe, the internal voltage E ' q and
the terminal voltage VT can be expressed as:
x(jt,+g1) = x(jt,)g +v(jt,+g1) (9)

Pe = E 'q iq + ( X q − X 'd )id iq (4) With j = 1, 2,...., n and g = 1, 2,...., m

• 1
E'q = [ E fd − E'q − ( X d − X 'd )id ] (5)
'
Tdo
VT = [( X qiq )2 + (E'q − X 'd id )2 ] (6)
In the above equation, E fd is the field voltage;
' Xd
Tdo is the open circuit field time constant; and
X d' are the d-axis reactance and the d-axis transient
reactance of the generator respectively.
A simplified IEEE Type-ST1 excitation system is
considered in this work. It can be described as:
• K A (Vref − VT + VS ) − E fd
E fd = (7)
TA
Where, KA and TA are the gain and time constant of the
excitation system; Vref is the reference voltage and
VS is the output stabilizing signal of PSS.
4. OVERVIEW OF PARTICLE SWARM
OPTIMIZATION TECHNIQUE
The PSO method is a member of wide category of
swarm intelligence methods for solving the
optimization problems. It is a population based search
algorithm where each individual is referred to as
particle and represents a candidate solution. Each
particle in PSO flies through the search space with an
adaptable velocity that is dynamically modified
according to its own flying experience and also the
flying experience of the other particles. In PSO each
Where,
n =number of particles in a group;
m = number of members in a particle;
t = number of iterations (generations);
v (jt,)g = velocity of particle j at iteration t,
V gmin ≤ v (jt,)g ≤ V gmax ;
w = inertia weight factor;
c1 , c2 = cognitive and social acceleration factors
respectively;
r1 , r2 = random numbers uniformly distributed in the
range (0, 1);
x(jt,)g = current position of j at iteration t;
pbest j = pbest of particle j;
gbest = gbest of the group.
The j-th particle in the swarm is represented by a g-
dimensional vector xj = (xj,1, xj,2, ……,xj,g) and its rate of
position change (velocity) is denoted by another g-
dimensional vector vj = (vj,1, vj,2, ……, vj,g). The best
previous position of the j-th particle is represented as
pbestj = (pbestj,1, pbestj,2, ……, pbestj,g). The index of best
particle among all of the particles in the group is
represented by the gbestg. In PSO, each particle moves in
the search space with a velocity according to its own
previous best solution and its group’s previous best
solution. The velocity update in a PSO consists of three
parts; namely momentum, cognitive and social parts. The
balance among these parts determines the performance of
a PSO algorithm. The parameters c1 & c2 determine the
relative pull of pbest and gbest and the parameters r1 & r2
help in stochastically varying these pulls. In the above
equations, superscripts denote the iteration number.

Dr. Sidhartha Panda Page No. 127

 Key Note-03

5. PROBLEM FORMULATION X + XT ), and the capacitive

the total reactance (
5.1 PSS and TCSC controller structure
reactance XTCSC (α) . So X Eff is given by:
The commonly used lead–lag structure is chosen in this
study as PSS and TCSC controller. The structures of the
X E f f = X + X T − X T C S C (α ) (10)
PSS and TCSC controller are shown in Figures 2 & 3
respectively. Each structure consists of: a gain block
(with gains KP and KT for PSS and TCSC respectively); a As the network parameters change due to the fault
signal washout block and two-stage phase compensation and its subsequent clearing, the updated values of X Eff
block. The phase compensation block provides the
appropriate phase-lead characteristics to compensate for is used in the simulation.
the phase lag between input and the output signals. The
signal washout block serves as a high-pass filter which ∆ω sTW 1 + sT1P 1 + sT3P
VS
KP
allows signals associated with oscillations in input signal Input 1 + sTW 1 + sT2 P 1 + sT4 P Output
to pass unchanged. Without it steady changes in input Gain block Washout Two-stage
would modify the output. From the viewpoint of the
block lead-lag block
washout function the value of washout time constant is
Figure 2. Block diagram of power system stabilizer.
not critical and may be in the range 1 to 20 seconds [1].
In these structures, TW is the washout time constant of Max.

PSS and TCSC controller. A wash time constant of 10 X TCSC (α )

∆ω sTW 1 + sT1T 1 + sT3T ∆σ 1
sec. is used in the present study. Also two similar phase Input KT
1 + sTW 1 + sT4T
∑
1 + sT2T σ0 + ∆σ 1 + sTTCSC
+ Output
compensator blocks are considered so that T1 P= T3S , T2P + TCSC
Gain
= T4P, T1T = T3T and T2T = T4T. The gains KP, KT and the block Washout Two-stage σ0
Min.
block lead-lag block
time constants T1P , T2P , T1T and T2T are remained to be
determined. In case of PSS, the input signal is the speed Figure 3. Block diagram of TCSC controller
deviation ( ∆ ω ), and the output signal is the stabilizing 5.2 Optimization problem
signal VS, which is added to the excitation system It is worth mentioning that the PSS and TCSC controller
reference voltage VR. The input signal of the proposed are designed to minimize the power angle deviation after
TCSC controller is the speed deviation ∆ ω , and the a large disturbance so as to improve the power system
stability.
output is TCSC reactance, XTCSC (α) . The desired Therefore the objective can be formulated as the
minimization of a non-linear programming problem
value of line reactance is obtained according to the
expressed as follows:
change in the conduction angle ∆ σ . In the Figure 4, t
J = ∑ ∫01 t[| ∆ω (t , x ) |]dt (11)
σ0indicates the initial conduction angle as desired by Where, ∆ ω (t , x ) denotes the speed deviation for a set
the power flow control loop. The power low control loop
of controller parameters x (note that here x represents KT,
acts quit slowly in practice and hence σ0 is assumed to T1T, T2T, KP, T1P, T2P: the parameters of both PSS and
TCSC controller), and t1 is the time range of the
remain constant during large-disturbance transient period. simulation. With the variation of the parameters x, the
The value of α is changed according to the change in ∆ ω (t , x ) will also be changed. For objective function
conduction angle ∆ σ , as α = π −σ / 2 and calculation, the time-domain simulation of the nonlinear
power system model is carried out for the simulation
σ =σ0 +∆σ . This signal is put through a first order period. It is aimed to minimize this objective function in
lag representing the natural response of the controller and order to improve the system response in terms of the
the delay introduced by the internal control, which yields settling time and overshoots.

the reactance offered by the TCSC, XTCSC (α) . The 6. RESULTS AND DISCUSSION

effective transfer reactance ( X Eff ) between the 6.1 Application of particle swarm optimization
technique
generator and the infinite bus is the difference between In order to tune the parameters of the PSS and TCSC
controller, the non-linear model of the example power
system shown in Figure 1, is developed. The relevant
parameters of the power system are given in appendix.

Dr. Sidhartha Panda Page No. 128

 Key Note-03

For the purpose of optimization of equation (12), PSO is Heavy loading (Pe=1.1 0.0963 ± -1.3909 ± 0.1199i
employed. The objective function is evaluated for each pu, δ0 =60.10) 8.0749i
individual by simulating the example power system under
a severe disturbance. The most severe situation where a
three phase short-circuit fault occurs at the generator The behavior of the proposed controllers under transient
busbar terminal is considered in the present study for conditions is verified by applying a 10-cycle three-phase
objective function calculation. While applying PSO, a fault at the generator terminal busbar at t = 1 sec. The
number of parameters are required to be specified. An fault is cleared by tripping of one of the parallel
appropriate choice of the parameters affects the speed of transmission lines. The system response under this severe
convergence of the algorithm. Table I shows the specified disturbance is shown in Figure 4. In the Figure 4 the
parameters for the PSO algorithm. Optimization is response without control (no control), response with PSS
terminated by the prespecified number of generations. only and response with coordinated application of both
Table II shows the optimal values of PSS and TCSC PSS & TCSC controller are shown with dotted line,
controller parameters obtained by the PSO algorithm. dashed line and thick solid line respectively. It is clear
Table I: Parameters used for PSO algorithm from the Figures that, the system is unstable without
Parameters Value control under this severe disturbance.
150
no control
Swarm size 20 140
w ith PSS only
130 w ith PSS & TCSC
Max. Generations 100
120
c 1, c 2 2.0, 2.0 110

δ (deg.)
wstart, wend 0.9, 0.4 100

90

Table II: Optimal parameter settings of the proposed controllers 80

70

Param KT T1T=T T2T=T KP T1P=T T2P= 60

e-ters 3T 4T 3P T4P 50
0 1 2 3 4 5 6
Value 35.21 0.809 0.751 39.2 0.599 0.651 Time (sec)
s 07 04 06 63 97 35 (a)
-3
6.2 Simulation results x 10
The system electromechanical mode eigenvalues without 20 no control
and with the proposed controllers is shown in Table III. It w ith PSS only
is clear from the table that, the open loop system is 15
w ith PSS & TCSC
unstable at heavy loading because of negative damping of
ω (pu)

electromechanical mode (s = 0.0963). It is also clear that, 10

at nominal and light loading conditions, although the
system is stable it is highly oscillatory. The proposed 5
controllers shift substantially the electromechanical mode 0
eigenvalue to the left of the line (s = -1.3858, -1.3913, -
1.3909 for nominal, light and heavy loading condition
1 2 3 4
respectively) in the s-plane. This enhances greatly the
system stability and improves the damping characteristics Time (sec)
of electromechanical mode. 2
no control
Table III: The system electromechanical eigenvalues
1.5 w ith PSS only
without and with Control.
w ith PSS & TCSC
Pe (pu)

1
Type of control/ No With PSO
loading conditions Control Optimized PSS 0.5
& TCSC Controller
Nominal loading -0.0795 ± -1.3858± 0.0883i 0
(Pe=0.9 pu, δ0 =51.80) 7.6831i 1 2 3 4
Light loading (Pe=0.5 -0.2618± -1.3913± 0.0330i Time (sec)
pu, δ0 =31.50) 6.8644i (b)
(c)

Dr. Sidhartha Panda Page No. 129

 Key Note-03

eigenvalue analysis and non-linear simulation results

show the effectiveness and robustness of the proposed
0 controllers to enhance the power system stability.
∆ VT (pu)

-0.2
8. REFERENCES
no control
[1] Kundur, P. (1994). Power System Stability and
-0.4
Control. McGraw-Hill.
w ith PSS & TCSC
[2] P.W. Sauer, M.A. Pai, Power System Dynamics and
-0.6
Stability, Prentice-Hall, Englewood Cliffs, NJ, 1998.
1 2 3 4 [3] Padiyar, K.R. (1996) ‘Power System Dynamics:
Time (sec) Stability and Control’, Wiley.
[4] Hingorani, N. G.; Gyugyi, L. (2000). Understanding
0.5 FACTS: Concepts and Technology of Flexible AC
no control Transmission Systems, IEEE Press, New York.
0.45 w ith PSS & TCSC [5] Y. Wang, Y.L. Tan and G. Guo “Robust nonlinear
co-ordinated control for power systems
XTCSC (pu)

excitation and TCSC control for power systems”

0.4 IEE Proc,-Gener. Tmnsm. Distrib., Vol. 149, No. 3,
May 2002, pp.367-372.
0.35 [6] Cai, Li-Jun, and Erlich, I. (2005) ‘Simultaneous
coordinated tuning of PSS and FACTS damping
controller in a large power system’, IEEE Trans on
1 2 3 4 Power Syst, Vol. 20, No.1, pp. 294–300.
Time (sec) [7] J. Kennedy, R. C. Eberhart, Particle Swarm
(d) Optimization, Proc. IEEE Int'l. Conf. on Neural
(e) Networks, IV, 1942-1948. Piscataway, NJ, IEEE
Figure 4. System response for a three-phase fault and Service Center, 1995.
subsequent line tripping disturbance at nominal loading [8] Mathur, R. M.; Varma, R. K.; (2002). Thyristor-
condition. (a) power angle δ (b) relative speed ω (c) based FACTS Controllers for Electrical
electrical power Pe (d) terminal voltage deviation ∆VT (f) Transmission Systems. IEEE Press, Piscataway.
reactance offered by TCSC XTCSC [9] Yu, Yao-Nan. (1983). Power System Dynamics.
Academic press, Inc., London.
In addition, the first swing in the rotor angle is [10] Z.L. Gaing, A particle swarm optimization approach
significantly suppressed by coordinated application of for optimum design of PID controller in AVR
PSS & TCSC controller. It can be seen that the response system, IEEE Trans. Energy Conv. 9 (2) (2004) 384-
with coordinated application of PSS & TCSC controller 391.
provides good damping characteristics to low frequency
oscillations and stabilizes the system much faster 9. APPENDIX
System data: All data are in pu unless specified
compared to the case where only PSS is acting. Hence,
otherwise.
coordinated design of PSS & TCSC controller extends
the power system stability limit and the power transfer '
VT =1.0; H = 4.0 s.; Tdo
Generator: = 5.044; D =
capability.

7. CONCLUSIONS 4.4; X d =1.0; X q =0.8; X d' = 0.3; Ra=0;

Exciter :( Simplified IEEE Type ST1): KA=200; TA=0.01
In this paper, power system stability enhancement by
s.
coordinated design of PSS & TCSC controller is
Transmission line, Transformer and Load: R + jX = 0 +
presented. The minimization of the rotor speed deviation
j0.6; XT = 0.1; G + jB = 0 +j0.
following a severe disturbance is formulated as an
Power System Stabilizer (PSS): T1P = T3P; T2P = T4P; TW =
optimization problem and particle swarm optimization
10 s.
technique is employed to optimally tune the parameters
of the proposed controllers. The proposed controllers are TCSC Controller: α 0 = 158 0 ; X T C S C 0 = 0. 3367;
applied and tested on a weakly connected power system k=2; K C = 0.5 X ; T1T = T3T; T2T = T4T; TW = 10 s.;
under different loading conditions, and fault clearing
sequences. The performance of the proposed controllers KP = 0 .2 5 X C ; TT C SC =15 ms;
is also tested under different small disturbances. The X T C S C M ax . = 0 .8 X ; X TC SC M in . = 0

Dr. Sidhartha Panda Page No. 130