International Journal of Computer Applications (0975 – 8887)

Volume 81 – No 19, November 2013

Design, Fabrication and Implementation of

Microcontroller Controlled Static Var Compensator

G.PREMKUMAR B.MUTHUKUMAR
AMRITA SCHOOL OF ENGINEERING AMRITA SCHOOL OF ENGINEERING
AMRITA UNIVERSITY AMRITA UNIVERSITY

ABSTRACT power electronics in which SVC is very simple and common

Power factor correction is vital in present society. The two method for reactive power compensation.
key benefits of power factor correction, which includes
reduced power costs and higher system capacity.
Implementation of Flexible AC Transmission Systems
(FACTS) improves the power factor, power transfer capability
and voltage stability. The shunt type of FACTS controller is
used to either absorb or inject VARs into the system, thereby
compensating reactive power. The design, fabrication and
testing of microcontroller controlled Static VAR Compensator
(SVC) consisting of Thyristor Controlled reactor (TCR) and
Thyristor Switched Capacitor (TSC) are dealt with in this
paper. A microcontroller is used to calculate the load
displacement power factor (DPF) and for executing the fuzzy
logic control scheme for TCR branch. The details of the
design, fabrication, test results of Thyristor Controlled
Reactor and Thyristor Switched Capacitor and their capability Fig. 1. Typical SVC Configuration with TSC and TCR
to improve their power factor performance are the main aim Branches [1].
of this paper.
The focus of this project has been on a particular FACTS
(Flexible AC Transmission System) device – the Static Var
General Terms Compensator (SVC) as shown in Fig 1. The SVC is a proven
Microcontroller controlled Static Var Compensator, Thyristor technology for power factor correction and reactive power
Controlled reactor (TCR), and Thyristor Switched Capacitor compensation. Traditionally, the SVC has been used as a
(TSC). shunt-connected device that offers voltage stability and load
compensation to the power system at particular points such as
Keywords transmission line 4midpoints or near varying loads. Since
Displacement power factor, Firing angle, Fuzzy logic, EPRI’s (Electric Power Research Institute) release of the
Harmonics, Reactive load, microcontroller, Residential load, FACTs strategies in 1987, SVC’s have grown in popularity
Static var compensator, Triac, True power Factor. and are well regarded in the power industry [1].
The SVC comprise capacitor bank controlled by triac called
1. INTRODUCTION Thyrsitor Switched capacitor (TSC) and reactor bank
As the demand for electric power grows the energy efficiency controlled by triac called Thyristor controlled reactor (TCR).
term is very significant. Electric load consuming alternating These compensators draw reactive power from the line there
currents consume both active power and reactive power. The by regulating voltage, improving stability, control overvoltage
active power does useful work and reactive power dissipates and reduce voltage flicker. Since static Var compensators use
no energy in the load and return to the source on current switching for reactive power control, these are also called
cycle. When the inductive loads are connected to the power static VAr (Reactive power) switches or systems.
system network, the voltage at the consumer side reduces due
to high reactive power requirement. This is due to the The goal of this paper is to design, fabricate and implement a
increased load current caused by increased reactive power low cost distributed var controller that is capable of power
consumption due to the lagging power factor of the inductive factor displacement (DPF) correction. The var controller
load. This will increase the system losses and hence reduces design will be based on the static var compensator (SVC),
the efficiency. A conventional method for mitigation is which is a proven FACT’s device for power factor correction
problem is to use capacitors for supplying the required and voltage stability. The microcontroller controlled prototype
reactive power. Use of a set of capacitors may be significant Static Var Compensator (SVC) will be economical and robust.
to compensate the lagging power factor for an inductive load. The microcontroller controlled SVC will contain two
branches, a TSC and a TCR branch that are controlled by an
In a real time scenario, the load will change continuously and open loop fuzzy controller. It is a completely self-contained
hence the required reactive power also changes continuously. device capable of being installed on a variety of VAr loads.
In this condition, a fixed capacitor can sometimes cause The microcontroller controlled SVC is intended for the
overcompensation which will result in over voltage at the load following three audiences: The power utility company, the
side. As the power electronic techniques are improving very commercial and residential consumer, and manufacturers of
fast, the continuous adjustments of reactive power using applications that generate VAr [1].

43
 International Journal of Computer Applications (0975 – 8887)
Volume 81 – No 19, November 2013

2. HARDWARE DESIGN
2.1 POWER FACTOR MEASUREMENT [11]
The power factor of an AC electric power system is defined as
the ratio of the real power to the apparent power, and is a
number ranging between 0 to 1 inclusive. Real power is the
capacity the circuit performance in a given time. Apparent
power equals to the product of the current and voltage of the
circuit used. Energy when stored in the load and returned to
the source or due to a non-linear load that distorts the wave of
the current drawn from the source, the apparent power equals
to or greater than the real power. Low power factor increases
losses in a power distribution system and thus resulting in
increased cost for electrical energy use.
In purely resistive AC circuit, voltage and current waveforms
are in step, changing the polarity in every cycle. In circuits
having reactive loads, containing capacitors or inductors, the
energy is stored and hence resulting in a difference between
voltage and current waveforms. The stored is not used for
load as it returns to source. When a circuit has a low power Fig 2: Power factor measurement circuit
factor it has higher currents to transfer at a given quantity of
power than a circuit having higher power factor. From the shown Fig 2, the voltage transducer is used to step
down the mains supply voltage to low voltage level. The
Circuits containing only heating elements like filament lamps, voltage level is from 230V AC to 6V AC. Then the output of
strip heaters and cooking stoves etc. have a power factor of the voltage transducer is given to Zero Crossing Detector. The
unity. Circuits containing inductive or capacitive elements current consumed by the load is measured with the help of a
such as lamp ballasts and motors etc. often have a power current transducer. The current transducer will convert the
factor below 1.0. The significance of power factor lies in the load current in to lower values that current output will be
fact that utility companies supplies customers with volt- converted in to voltage with the help of the shunt resistor.
amperes, but in real they bill them for watts. The circuits Then the corresponding the AC voltage is given to zero
having power factors less than unity require a utility to crossing detector. The Zero Crossing Detector is used to
generate more than the minimum volt-amperes necessary to convert the sine wave to square wave signal.
supply the real power (watts). This increases generation and
transmission costs. Good power factor is which is greater than The zero crossing detectors are constructed by the operational
0.85 or 85%. Utilities may charge additional costs to amplifier LM 741. The inverting and non-inverting input
customers who have a power factor below some limit. terminals are connected to the potential transformer and
current transformer terminals respectively. So the input sine
A power factor of 1 or "unity power factor" is the goal of any wave signal is converted in to square wave signals. The
electric utility company since if the power factor is below 1, it square signal is in the range of +12v to -12v level. Then the
has to supply more current to the user for a given amount of square wave signal is given to base of the BC 547 switching
power usage. In doing so, they have more line losses. They transistor in order to convert the TTL voltage 0 to 5v level.
should have larger capacity equipment in place than would be Then the both ZCD’s outputs as in Fig 4 shown are given to
necessary. Resulting in the industrial facility to be charged a logical XOR gate 74LS86 to find the phase angle difference
penalty if its power factor is much varying from unity. between the voltage and current. The XOR gate output is
Industrial facilities mostly have lagging power factor, where given to microcontroller or PC and calculates the power factor
the current lags voltage as in inductor. This primarily results with help of software.
having a lot of electric induction motors - the windings of
motors act as inductors as seen by power supply. Capacitors
have opposite effect and can compensate the inductive
windings. A very few industrial sites have large banks of
capacitors strictly for the purpose of correcting the power
factor back toward one to save on utility company charges.

Fig 3: Input and Output of ZCD

44
 International Journal of Computer Applications (0975 – 8887)
Volume 81 – No 19, November 2013

2.5 LEVEL SHIFTER

The ADC module of PIC 16F877A accepts only positive
voltages and a maximum of 5V is possible since the
maximum reference voltage to the ADC module is 5V. So the
load voltage and load current sensed by the PT and CT is
shifted by 2.5V DC. The reason for shifting by 2.5V is
because, the maximum peak positive voltage of the PT and
CT is 2.5V and maximum peak negative voltage is -2.5V. So
when shifter by 2.5V DC, we get an output voltage in the
Fig 4: XOR gate Input range 0 to 5V, which is needed for the ADC module of the
PIC 16F877A. The level shifting is done using op-amp adder
2.4 VOLTAGE TRANSDUCER CIRCUIT configuration in non-inverting mode with one input from the
LEM LV-20 P voltage transducers have been used to sense CT and the other is 2.5V DC. The output is limited to 5.1V
inverter output voltage and grid voltage. A current using a zener diode at the output. The circuit diagram of level
proportional to measured voltage is passed through external shifter is shown in Figure 6.
resistor. This resistor has been installed in series with the
primary circuit of the transducer. Transducer’s optimum
accuracy is obtained at the nominal primary current, +15V
and -15V.
Voltage to be measured = 230V
Let the input resistance be 33KΩ
Then the input current for 33KΩ = 230/33KΩ = 6.9mA
Conversion ratio = 2500:1000
Secondary rms current = 6.9*2.5 = 17.25mA
Since the secondary voltage can be maximum of only 2.5v,
Measuring resistance = 2.5/17.25mA = 144.92Ω
The chosen value of secondary resistance 100Ω Fig 6 : Level shifter circuit
2.5 CURRENT TRANSDUCER CIRCUIT
The load current is to be sensed and is given as the input to 2.6 LCD DISPLAY
the microcontroller. LEM module LA 25-NP current A LCD of 2x16 ASCII-text is installed on a PCB board,
transducer is used to sense the load current. The turns ratio having 4 data and 3 command pins as input/output. A 5V DC
selected is 3/1000. The recommended measuring resistance supply is given to the LCD.A 10K pot is used to adjust the
range is from 100 ohm to 320 ohm. contrast in the display. This LCD application is used to
display the load voltage, load current, power factor and the
Peak load current = 10A
capacitor value which is need to compensate the lagging
Secondary voltage should be less than 2.5V, Conversion ratio reactive power.
is 3/1000.
2.7 DC POWER SUPPLY
Secondary resistance R= (2.5*(1000/3))/10) =83.3Ω In Fig 7 shown is a DC power supply circuit to provide supply
Hence the secondary resistance chosen is 100 Ω. to the PIC , current sensor, voltage sensor, level shifter, LCD
display etc, a DC power supply of voltages 2.5V, 5V, 12V
and -12V are designed. A 230/15-0-15/0-9 V, 1/1.5 A,
transformer and IN4007 power diodes are used. In order to
obtain regulated voltages, regulator IC’s 7805, 7812 and 7912
are used. Capacitor filter value 1000µF/25V and 10µF/63V
are used at the input and output side of the regulator IC
respectively.

Fig 5: Voltage and Current sensor hardware and output

results

45
 International Journal of Computer Applications (0975 – 8887)
Volume 81 – No 19, November 2013

2.8.2 THYRISTOR SWITCHED CAPACITOR (TSC):

A shunt connected capacitor whose effective reactance is
varied in a stepwise manner by full or zero conduction
operation of the thyristor valve, TSC (shown in Fig 9) is also
a subset of SVC in which thyristor based ac switches are used
to switch in and out (without firing angle control) shunt
capacitor branches, which will achieve the required step
change in the reactive power supplied to the system. Unlike
shunt reactors, shunt capacitors cannot be switched
continuously with variable firing angle control.

Fig 7: DC power supply circuit

2.8 TSC AND TCR BRANCH DESIGN

2.8.1 Thyristor Controlled Reactor (tcr):
A shunt-connected inductor with effective reactance is varied Fig 9: Circuit diagram of TSC
in a continuous manner by partial-conduction control of the
thyristor valve, TCR (shown in Fig 8) is a subset of SVC in 2.9 Microcontroller and SVC Branch (TSC
which conduction time and hence, current in a shunt reactor is And TCR) Interfacing:
controlled by a thyristor-based ac switch with firing angle The microcontroller controls the TCR and TSC branches by
control. giving the control signal to a triac driver which is called
optocoupler. An optocoupler is essential to control each
individual branch utilized. For example, if a bank of 9
capacitors (TSC) are used and 1 inductor (TCR), then 10 triac
drivers will be needed. Likewise, each branch must be given a
dedicated port on the microcontroller. The need of the
optocoupler (MOC3011) is to isolate the high voltage triac
from lower voltage control. This electrical isolation helps to
ensure that the microprocessor remains isolated from the high
voltage as shown in Fig 10. Since the MOC3011 is a current
activated device, the proper resistor to be placed between the
MOC3011 and the microprocessor must be chosen correctly.
Fig 8: Circuit diagram of TCR In this example, two output microcontroller ports were
utilized to drive the two triac drivers [1].
The controlling element in TCR is the thyristor controller,
shown above are a pair of anti-parallel thyristors which
conduct during alternate half cycles of the supply voltage. If
the thyristors are gated into conduction precisely at the peaks
of supply voltage, when fully conducted resulting in the
reactor, and the current is same as though the thyristor
controller were short-circuited. The current is essentially
reactive, lagging behind the voltage by nearly 90 o. It contains
a small in-phase component due to power loss in the reactor,
which may be of the order of 0.5-2% of the reactive power.
Each of these corresponds to a certain value α, the gating
angle measured from a zero crossing of voltage. Full Fig 10: Interface between the Microcontroller and
conduction is obtained with gating angle of 90 o. Partial Optocoupler
conduction is obtained with gating angles between 90o and
180o. 2.9.1 TRIAC SELECTION
The effect of increasing the gating angle is to reduce the The maximum AC load voltage appearing across the switch is
fundamental harmonic component in the current. Equivalent 240V.The full load peak current is 10A. For the specified AC
to an increase in the inductance in the reactor, reducing the voltage and current, TRIAC is selected as switch. To meet the
reactive power and its current. So far as the fundamental above specification after providing sufficient safety limit and
component of current is concerned, the thyristor-controlled the future use of the inverter for higher ratings TRIAC BTA
reactor is controllable susceptance, and therefore is applied as 16 600B of rating 600V and 16A is selected.
a static compensator.

46
 International Journal of Computer Applications (0975 – 8887)
Volume 81 – No 19, November 2013

2.9.2 DRIVER CIRCUIT DESIGN voltage waveform and then immediately record the state of
the current waveform. By comparing the waveform states, the
PIC can determine whether the current is leading or lagging
the voltage. To detect a zero crossing, the PIC continuously
polls a waveform watching for its value to flip. For example,
if the PIC read a HIGH for the voltage, then a loop
continuously polling the voltage would detect when the
voltage input flipped to a LOW. When that occurs, the PIC
has detected a zero crossing. If the voltage waveform was
recorded as HIGH after the zero crossing, and the current
waveform was recorded as LOW, then it is determined that
the current is lagging the voltage. Likewise, if the voltage was
recorded as HIGH, and the current was already HIGH, then
the current would be considered leading the voltage. Because
a voltage zero crossing is needed for synchronization, the PIC
needs half a cycle to determine whether the current waveform
Fig 11: Triac Switching Circuit is leading or lagging. Table 3.1 shows all of the possible
combinations of the current waveform leading or lagging
In the shown Fig 11 the on/off pulses are given from the logic. Note that the voltage waveform is the first waveform to
microcontroller to transistor BC557. This transistor will be polled after the zero-crossing is detected.
conduct only if the pulses received from the microcontroller if Table 12: Logic to find whether current is
it the pulses are received then the signal are given to the Lagging/Leading Voltage [1]
BC547 which acts as a switch and then the pulses are given to
the optocoupler in which the led blinks so that the Current HIGH Current LOW
phototransistor conducts. Due to this the signal, is transferred
to the triac gate so mt1 and mt2 terminal gets shorted so that Voltage HIGH LEADING LAGGING
we get constant output voltage such as 230V in the load. The
threshold gate voltage for the triac is 1.5V.
Voltage LOW LAGGING LEADING

3. SOFTWARE DESIGN [1]

3.1 Microcontroller Software Design and
Implementation 3.1.3 Calculating the Time Difference (ΔT) Between
The PIC microcontroller is responsible for Power Factor Voltage and Current Waveforms
calculation through zero crossing detection, TSC switching Once it has been determined whether the current is leading or
control, fuzzy control of the TCR branch and LCD output. lagging the voltage, the PIC will calculate the time difference
separating the two waveforms. If the current is lagging the
3.1.1 Power Factor Calculation: voltage, the PIC will wait for the voltage to cross zero at the
The displacement power factor , which is indirectly the next half cycle and then start its 16-bit timer. The PIC will
control input for both the TSC and TCR branches, is then wait for the current to cross zero. Once the current
calculated by having the microprocessor determine the time crosses zero, the timer is stopped and its value is stored in a
difference between the reactive load’s voltage and current variable. Likewise, if the current is leading the voltage, the
waveforms. The following list summarizes the functional and PIC will wait for the current to cross zero at the next half
mathematical steps taken by the microprocessor to calculate cycle and then start its 16-bit timer. The PIC will then wait for
the displacement power factor. The list will be expanded into the voltage to cross zero, and when this occurs, the PIC will
subsections to provide specific microprocessor stop its timer and record the value to a variable. Calculating
implementation details. this time difference takes one half cycle because the
waveform zero crossings are needed to synchronize the timer.
1. Determine Whether the Reactive Load is Leading
or Lagging 3.1.4 Converting the Time Difference (ΔT) to a
Phase Angle Difference (θ)
2. Calculate the Time Difference Between Voltage and The PIC’s 16-bit timer was configured via software to tick
Current Waveforms once every 1.6us, which provides for a maximum time count
of 0.1 sec before the timer will roll over. 1.6us was chosen
because it is a multiple of the clock frequency, 20MHZ. The
3. Convert the Time Difference to a Phase Angle 0.1s rollover provides ample time for the cycle-by-cycle
Difference calculations that are performed every 8.33ms and 16.66ms.
The 1.6us ticks also provide a high measurement resolution of
the time difference. For only a 1° (46us) phase difference, the
4. Convert Phase Angle Difference to PF PIC will record 28 clock ticks. Using Eq. 3.1 the timer “ticks”
D
can be converted into a time difference (ΔT). After ΔT is
3.1.2 Determining Whether the Reactive Load is found, the phase angle difference (θ) of the load can be found
by using Eq. 3.2 introduced earlier. If the load is lagging then
leading or Lagging the phase angle difference is given a positive sign. If the load
To determine whether the reactive load is leading or lagging,
is leading then the phase angle is assigned a negative sign.
the PIC will first wait for the voltage waveform to cross zero.
Once it crosses zero the PIC will record the state of the ΔT = Ticks*(20MHz/32)-1 (3.1)

47
 International Journal of Computer Applications (0975 – 8887)
Volume 81 – No 19, November 2013

3.1.5 Converting the Phase Angle Difference (θ)

to Displacement Power Factor
The Displacement Power Factor (DPF) can be calculated
easily by taking the COSINE of θ. By performing this
calculation is optional for the PIC because, θ is used as the
control variable for the TSC and TCR branches. If a slower
clock frequency was chosen this calculation could be omitted
to save processing time.

3.2 TSC Switching Control [1]

The optimum capacitor configuration for the individual TSC
branches can be found by sizing the capacitors in binary steps.
For the SVC prototype, the binary steps of 1uF, 2uF, 4uF,
8uF, 16uF, 32uF and 64uF were chosen to provide a total of
128uF in discrete steps of 1uF with only using 7 capacitors.
This flexibility allows for a tightly configured displacement
power factor compensation that can be easily programmed to
be either leading or lagging. as shown in Fig 12 illustrates the
TSC branches configured in binary steps. Fig 12 TSC Branches Configured in Binary Steps [1]

The binary configuration is important to the software design

because the PIC must be able to intelligently switch in and out
the appropriate TSC branch at the appropriate time. Every 5
seconds the PIC is programmed to examine the current phase
angle difference (θ) and switch in or out the next capacitor
size in binary order. Generally, if θ positive (lagging
Displacement Power Factor (DPF)) the PIC will switch in
TSC branches in binary order (increments of 1uF) until θ is no
greater than –20° (leading DPF). Likewise, if θ is negative
(leading DPF) the PIC will switch out TSC branches until θ is
approximately –20° (leading DPF).
This overcompensation leading by approximately 20° is
needed so the TCR branch can fine-tune the DPF to a value
closer to unity.
As presented earlier, the only time the TSC branch should be
switched is during the zero crossing of the current so
transients are avoided. The PIC is programmed is examine θ
every 5 seconds, but this time value is arbitrary and should be
configured to best match the specific application. It is
recommended that a time value not less than 5 seconds be
used so that the PIC is not continuously updating the TSC
branches. When the time interval expires the PIC will wait for Fig 13 : Steps for switch in and out TSC branch. [1]
the next current zero crossing.
This is accomplished by having the PIC poll the current 3.3 TCR Fuzzy Logic Control Strategy
waveform input looking for a state change. For example, if the In [2] a fuzzy logic control scheme was designed and
current waveform is initially read as HIGH, then when the simulated for a power system SVC. This particular fuzzy logic
waveform switches to a LOW, a zero crossing has occurred. control strategy was simulated on a fixed capacitor – variable
Having already made its TSC switching decision based upon inductor type SVC, but the fuzzy control strategy is
θ, the PIC will either switch in or out a single TSC branch at completely independent of the SVC type used. The goal of
this exact instant. Fig. 13 summarizes the steps taken by the this fuzzy controller was to provide maximum damping and
PIC during TSC branch switching. stability to the system. This particular fuzzy controller only
used the real power flow as a raw input, and the output is the
firing angle. The power flow signal is stabilized and filtered
so that the measures of acceleration, time deviation and speed
deviation can be extracted. These measurements combine to
determine the current “state” of the system. This system state
is used as a fuzzy input, to a set of rules where the defuzzified
output will determine whether the inductor susceptance
should be capacitive (to increase real power flow at the SVC)
or inductive (to decrease real power flow at the SVC). Fig. 14
summarizes the steps taken by the PIC during TSC branch
switching.

48
 International Journal of Computer Applications (0975 – 8887)
Volume 81 – No 19, November 2013

4. IMPLEMENTATION OF CONTROL
CIRCUIT FOR TSC AND TCR BRANCH

Fig 16: Control circuit for SVC (TSC –TCR)

Fig 14: Steps for controlling the TCR branch [1]

3.4 Software Implementation

The software for the PIC was developed using the C
programming language and a specific PIC C compiler, known
as “PICC”, which is distributed by CCSInfo [5]. This
compiler compiles the C source code and generates the
equivalent assembly instructions. A software PIC programmer
known as MPLAB, distributed by Microchip, was chosen as
the tool that would program the PIC microcontroller once the
assembly code had been generated [6]. MPLAB transfers the
microcontroller instructions from the PC to the PIC via a
serial cable. Once the PIC has been programmed, the PC,
serial cable, and developer software is no longer needed. The
PIC is capable of executing completely autonomously. Fig. 15
summarizes the functionality of the microcontroller.

Fig 17 : Binary ratios of capacitor bank ( TSC) and TCR

5. CONCLUSION
In this paper, microcontroller controlled SVC is designed,
fabricated, tested manually and controlled automatically to get
unity power factor by supplying variable reactive power to the
system shown in Fig 15, Fig 16. The hardware and software
of this SVC control system is developed based on
Microcontroller PIC 16f877A chip, one of the low cost
industrial controller. The power circuit of single phase
2KVAR TSC-TCR type SVC have been design and tested
experimentally using a test system.
Fig 15 : Functional Block Diagram of Microcontroller [1] The future scope of this project is to design filter to reduce the
harmonics in the system, control the SVC through LABVIEW
and the values like voltage, current, power factor and reactive
power can transmit wirelessly through ZIG-BEE or using
ARDUINO.

6. REFERENCES
[1] S. Zemerick, P. Klinkhachorn, A. Feliachi. “ Design of a
microprocessor controlled personal static var
compensator (PSVC)”. In proc. Of the IEEE summer
PES meeting, Chicago, PP. 1468-1473, July 21-25 2002.
[2] T. Hiyama, W. Hubbi, T. Ortmeyer, “Fuzzy Logic
Control Scheme with Variable Gain for Static Var

49
 International Journal of Computer Applications (0975 – 8887)
Volume 81 – No 19, November 2013

Compensator to Enhance Power System Stability”, IEEE [7] P. Mehta, M. Darwish. “Active reactive-power
Transactions on Power Systems, Vol. 14, No. 1, Feb. controller.” in Proc. IEE Electric Power Applications,
1999 pp. 186-191 Vol. 142, Issue.6, pp. 405-409, November 1995.
[3] A. Gelen, T. Yalcinoz. “Simulation of TSC on voltage [8] Q. Gu, A. Pandey, S. K. Starrett. “Fuzzy logic control
regulation for static and dynamic load models using schemes for static var compensator to control system
MATLAB.in Proc. of the IEEE 38th North American damping using global signal.” Electrical Power System
Power Symposium (NAPS), Illinois, USA, September Research, Vol. 67, Issue. 2, pp. 115-122, November
17-19, 2006. 2003.
[4] N. G. Hingorani, L. Gyugyi. Understanding FACTS: [9] T. Ahmed, K. Nishida, K. Soushin, M. Nakaoka. “Static
Concepts and Technology Flexible AC Transmission var compensator-based voltage control implementation
Systems. New York: IEEE Press, 1999. of single-phase self-excited induction generator.” IEE
Proc. Generation Transmission & Distribution, Vol. 152,
[5] R. M. Mathur, R. K. Varma. Thyristor-Based FACTS Issue. 2, pp. 145-156, March 2005.
Controllers for Electrical Transmission Systems. New
York:IEEE Press, 2002. [10] S. K. M. Kodsi, C. A. Canizares, M. Kazerani. “Reactive
current control through SVC for load power factor
[6] B. Endres, G. Thiele, I. Bonfanti, G. Testi. “Design and correction.” Electric Power Systems Research, Vol. 76,
operational testing on thyristor modules for the SVC Issue. 9-10, pp. 701-708, June 2006.
Kemps Creek.” IEEE Trans. on Power Delivery, pp.
1321-1326, November 1989.

IJCATM: www.ijcaonline.org 50