ussc.pdf

7/25/2019 ussc.pdf

1/7

1650 IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 20, NO. 2, APRIL 2005

PSCAD/EMTDC Simulation of Unified Series-ShuntCompensator for Power Quality Improvement

M. A. Hannan and Azah Mohamed, Member, IEEE

AbstractThis paper deals with the simulation of a unified se-ries-shunt compensator (USSC) aimed at examining its capabilityin improving power quality in a power distribution systems. TheUSSC simulation model comprises of two 12-pulse inverters whichare connected in series and in shunt to the system. A generalizedsinusoidal pulse width modulation switching technique is devel-oped in the proposed controller design for fast control action of theUSSC. Simulations were carried out using the PSCAD/EMTDCelectromagnetic transient program to validate the performance ofthe USSC model. Simulation results verify the capabilities of theUSSC in performing voltage sag compensation, flicker reduction,voltage unbalance mitigation, UPS mode, power-flow controland harmonics elimination. A comparison of the USSC with

other custom power devices shows that the USSC gives a betterperformance in power-quality mitigation.

Index TermsPower-quality mitigation, unified series-shuntcompensator.

I. INTRODUCTION

AN increasing demand for high quality, reliable electrical

power, and an increasing number of distorting loads

have led an increased awareness of power quality both by

customers and utilities. For power-quality improvement, the

development of power electronic devices such as flexible ac

transmission system (FACTS) and custom power devices haveintroduced an emerging branch of technology providing the

power system with versatile new control capabilities [1], [2].

In general, FACTS devices are used in transmission control

whereas custom power devices are used for distribution control.

Since the introduction of FACTS and custom power concept

[3], devices such as unified power-flow controller (UPFC),

synchronous static compensator (STATCOM), dynamic voltage

restorer (DVR), solid-state transfer switch, and solid-state fault

current limiter are developed for improving power quality and

reliability of a system [4], [5]. Advanced control and improved

semiconductor switching of these devices have achieved a new

era for power-quality mitigation.

Investigations have been carried out to study the effectiveness

of these devices in power-quality mitigation such as sag com-

pensation, harmonics elimination, unbalance compensation, re-

active power compensation, power-flow control, power factor

correction and flicker reduction [6]. These devices have been

developed for mitigating specific power-quality problems. For

Manuscript received September 9, 2003; revised February 19, 2004. Thiswork was supported by IRPA: 02-02-02-0019. Paper no. TPWRD-00462-2003.

The authors are with the Department of Electrical, Electronics, and SystemEngineering, Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor,Malaysia (e-mail: [email protected]; [email protected]).

Digital Object Identifier 10.1109/TPWRD.2004.833875

Fig. 1. Basic configuration of USSC.

example, UPFC works well for power-flow control [7]. DVR,

which acts as a series compensator, is used for voltage sag com-

pensation. STATCOM, which is a shunt compensator, is used for

reactive power and voltage sag compensation. The STATCOM

and DVR are only useful for compensating a particular type

of power-quality problem and therefore, it is necessary to de-

velop a new kind of unified series-shunt compensator (USSC)

which can mitigate a wider range of power-quality problems.

By using a unified approach of series-shunt compensators it ispossible to compensate for a variety of power-quality problems

in a distribution system including sag compensation, flicker re-

duction, unbalance voltage mitigation, and power-flow control.

However, not much work has been carried out in the develop-

ment of a USSC.

The objective of this paper is to explore the capabilities of

a USSC in mitigating power-quality problems. The proposed

model of the USSC considers the use of two 12-pulse inverters.

The modeling and simulation of the USSC has been carried

out using the well-known electromagnetic transient simulation

program PSCAD/EMTDC. Analyzes have been made by com-

paring the performance of the USSC to that of the STATCOM

and the DVR.

II. PRINCIPLEOPERATION OFUSSC

The USSC is a combination of series and shunt voltage

source inverters and its basic configuration is shown in Fig. 1.

The basic components of the USSC are two 12-pulse voltage

source inverters composed of forced commutated power semi-

conductor switches, typically gate turn off (GTO) thyristor

valves. One voltage source inverter is connected in series with

the line through a set of series injection transformers, while

the other is connected in shunt with the line through a set of

shunt transformers. The dc terminals of the two inverters are

0885-8977/$20.00 2005 IEEE

Authorized licensed use limited to: Jawaharlal Nehru Technological University. Downloaded on December 3, 2008 at 00:04 from IEEE Xplore. Restrictions apply.

7/25/2019 ussc.pdf

2/7

HANNAN AND MOHAMED: PSCAD/EMTDC SIMULATION OF USSC FOR POWER QUALITY IMPROVEMENT 1651

Fig. 2. Principle operation of USSC.

connected together and their common dc voltage is supported

by a capacitor bank [8].

The USSC is almost similar to the UPFC, but the only differ-

ences are that the UPFC inverters are in shunt-series connection

and it is used in transmission systems whereas the USSC in-

verters in series-shunt connection and it is used in distributionsystems.

The principle operation of a USSC is described by first re-

ferring to the model shown in Fig. 2. As mentioned earlier, the

USSC consists of a shunt connected inverter and a series con-

nected inverter. The series connected inverter injects a voltage

in series with the distribution line, which in turn changes

the voltage across the distribution line reactance , hence

changing the current and the power flow through the distribution

line. The exchange of real power and reactive power

can be written in terms of phase angle (the angle between the

injected voltage and the line current I), the injected voltage

, and the line current I, as

(1)

(2)

The current injected by the shunt inverter has a real or direct

component , which can be in phase or in opposite phase with

the line and a reactive or quadrature component , which is in

quadrature with the line voltage, thereby emulating an induc-

tive or a capacitive reactance at the point of connection with the

distribution line. The reactive current can be independently con-

trolled which in turn will regulate the line voltage.

The USSC behaves as an ideal ac-to-ac inverter, in which the

exchange of real power at the terminal of one inverter to the

terminal of the other inverter is through the common dc link

capacitor. It should be noted that the shunt inverter is controlled

in such a way as to provide precisely the right amount of real

power at its dc terminal to meet the real power needs of the

series inverter and to regulate the dc voltage of the dc bus. Thus,

real power is absorbed from or delivered to the distribution line

through the shunt connected inverter, which injects a current at

the point of connection.

Thus, USSC includes the functions of both series and shunt

connected inverters which generates or absorbs reactive power

to regulate voltage magnitude and current flow at the ac ter-minal, respectively [9].

Fig. 3. Simulation model of USSC using 12-pulse series and shunt inverters.

III. SIMULATION MODEL

The simulation model of the USSC which has been devel-

oped using the electromagnetic transient simulation programme

PSCAD/EMTDC is shown in Fig. 3. The USSC consist of two

12-pulse inverters in which one 12-pulse inverter is connected in

series and the other 12-pulse inverter is connected in shunt. The

series and shunt combination of inverters consists of a two-level,

three-phase, 24 self-commutated GTO switches with anti-par-

allel diodes. This valve combination and its capability to act as

a rectifier or as an inverter with instantaneous currentflows in

positive or negative direction, respectively, is the basic voltage

source converter concept. The shunt connected inverter is con-

nected to the load by means of two sets of three single-phasetransformers which are of Y-Y and Y- configurations to avoid

phase shift of other than the order of harmonics in the

secondary of the transformers, which may result in large circu-

lating current due to common core offlux. The phase to neu-

tral harmonic voltages of Y-Y connected secondary, other than

the order of , i.e., 5th, 7th, 17th, 19th . are oppo-

site to those of the phase to phase harmonic voltages of Y-

connected secondary and with 1/3 times the amplitude. There-

fore, the output voltage of shunt inverter would be a 12-pulse

wave form, with harmonic order of . However, the se-

ries connected inverter is connected to the source by means of

two sets of three single-phase transformers which are of Y-

configuration. This is because, usually Y-connected secondary

windings allow the injection of positive, negative and zero se-

quence voltages, whereas the delta connected secondary wind-

ings allow only the injection of positive and negative sequence

voltages. The delta connection prevents zero sequence currents

entering into the system from the inverter.

The primary windings of all the single-phase transformers

are connected in series in order to avoid harmonic circulating

current. The leakage reactances of the all the transformers are

kept low so as to prevent a large voltage drop. The 22/4.6-kV

step-down transformers with a leakage reactance of 0.01 per unit

are considered. Two consecutive 6-pulse inverters are used to

make up the 12-pulse inverter and the phase shift between thesetwo inverters is calculated and found to be 30 degrees by using


7/25/2019 ussc.pdf

3/7


Fig. 4. Shunt inverter control system.

the phase shift displacement angle formula which is given by

, where m is the number of the 6-pulse inverters used.

The capacitor plays an important role in the USSC operation

by acting as a dc source to provide reactive power to the load

and to regulate the dc voltage. The size of the dc capacitor con-

sidered in this simulation is 3340 .

IV. CONTROL SYSTEM OFUSSC

The control system of the USSC can be divided into two

parts, namely a shunt inverter controller and a series inverter

controller, in which they control the shunt current and the series

injected voltage, respectively. When the series and the shunt

connected inverters operate as stand-alone devices, they ex-

change almost exclusively reactive power at their terminals. The

series connected inverter injects a voltage in quadrature with

the line current thereby emulating an inductive or a capacitive

reactance in series with the line. The shunt connected inverter,

however, injects a reactive current, thereby also emulating a

reactance at the point of connection. While operating both

series and shunt-connected inverters together as a USSC, the

series injected voltage can be at any angle with respect to the

line current. The exchange of real power flow can be between

the terminals of series and shunt connected inverters throughthe common dc link capacitor.

A. Shunt Inverter Controller of USSC

The controller of a shunt inverter is used to operate the

voltage source inverter such a way that the phase angle between

the inverter voltage and the line voltage is dramatically adjusted

so that the shunt inverter generates or absorbs reactive power

at the point of connection of the system. Fig. 4 illustrates the

control design of the shunt inverter together with sinusoidal

pulsewidth modulated (SPWM) switching implemented in

PSCAD/EMTDC.

In the voltage control loop, the measured three-phase voltagesare fed to the phase locked loop in order to detect the

phase angles and angular positions of the voltages. The is

responsible for providing the voltage synchronizing signal with

an angle . The measured voltage in per unit and a constant are

fed into a maximum block to calculate the maximum voltage

signal which is then passed through the first order low pass filter

to attenuate the voltage transients. The signal is then compared

with a reference voltage. A voltage error is observed and is fed to

the voltage lag-lead function block, in which the output is

fed to the proportional integral (PI) control block. The output of

the PI controller is the angle order , which gives either a leading

or lagging phase angle which is necessary to adjust the voltageof the capacitor. The angle order represents the shift between the

system voltage and the voltage generated by the shunt inverter.

The angle order combined with voltage synchronizing signal

becomes the voltage modulating signal in which its magnitude

and phase are controlled.

The phase-locked loop also provides a voltage syn-

chronizing signal which is multiplied by a carrier frequency of

1.65 kHz, which is 33 times the system operating frequency

so as to convert the carrier ramp signal into the triangular car-

rier signal whose amplitude is fixed between 1 to 1. In the

SPWM technique, the triangular carrier signal is compared with

the voltage-modulating signal so as to obtain thefiring signalsof the GTOs. The zero crossings of the voltage ramps fire/block

the GTOs, depending on the displacement angle . If ,

the shunt inverter output voltage is said to be in phase with

the ac system voltage. However, if there is an error between

the reference voltage and the system voltage in per unit, that

is, , then the displacement angle and the

shunt inverter voltage lags behind the ac system voltage thus

causing real power flow into the shunt inverter. Consequently,

the dc capacitor voltage will increase, thus causing an increase

in the ac output voltage of the shunt inverter. The increase in

ac output voltage causes a reduction in the error voltage until

. If , then the displacement angle

and the shunt inverter voltage leads the ac voltage thuscausing real power flow into the system. Consequently, the dc


7/25/2019 ussc.pdf

4/7


Fig. 5. Load voltages: (a) without USSC and (b) with USSC.

capacitor voltage will decrease, thus causing a decrease in the ac

output voltage of the shunt inverter and a reduction in the error

voltage until .

B. Series Inverter Controller of USSC

In the series inverter, the SPWM technique is also used to con-

trol the magnitude and phase of the ac voltage by synchronizing

the GTOs switching to the ac system voltages. The control for

the series inverter are almost similar to that of the shunt inverter,

but the only difference is that in the series inverter voltage con-

trol loop, the measured phase currents are input to the in

order to generate the synchronizing signals. The series inverter

injected voltages are kept in quadrature with the line currents to

provide series compensation, whereas in the shunt inverter in-

jected currents are kept in quadrature with the line voltage.

In the series inverter control, the generation of the displace-

ment angle and the generation of the triangular carrier signal

and the voltage modulating signal is similar to that of the shuntinverter. In general, the overall controller function is the same

to that of the shunt inverter controller.

V. SIMULATION RESULTS

The performance of the USSC model is evaluated by means

of simulations using the PSCAD/EMTDC transient simulation

program. The USSC is placed in a 22-kV distribution system

with a static load of 5.2 MVA. There are twelve single-phase

transformers with each rated at 1 MVA, 22/4.16 kV and a

leakage reactance of 0.01 p.u., connecting the USSC to the

distribution system. Simulations were carried out to illustrate

the effectiveness of the USSC as a unified compensator forvoltage regulation, voltage sag compensation, voltage flicker

reduction, and voltage unbalance mitigation.

A. USSC for Voltage Sag Compensation

To illustrate the use of the USSC in compensating voltage

sags, a voltage sag condition is simulated by creating a balanced

three-phase fault using a three-phase generator. The simulation

results showing the use of USSC for compensating voltage sags

are shown in Fig. 5 in terms of the load voltages in per unit. For

the system without the USSC, the load voltage drops from 1.0 to

0.50 p.u., as shown in Fig. 5(a). This is a voltage sag condition

which is due to a three-phase fault created at time fora duration of 0.75 s. For the system with the USSC connected,

TABLE ICOMPARISON OF VOLTAGESAGCOMPENSATION CAPABILITY OFUSSC,

DVRAND D-STATCOM

the load voltage increases from 0.50 to 1.0 p.u., as shown in

Fig. 5(b). The load voltage returns to its rated voltage due to the

voltage sag compensation capability of the USSC.

Comparing the voltage sag compensation capability of

the USSC, D-STATCOM and DVR, the results are shown in

Table I in terms of the minimum, maximum and steady-state

voltage values. The minimum and maximum voltage values

correspond to the voltage at the starting and ending of voltage

sags, respectively. For the voltage sag compensated by theUSSC, the maximum, minimum and steady-state load voltages

are 1.09, 0.93, and 1.0 p.u., respectively. In the case of voltage

sag compensated by DVR, the load voltage is at a steady-state

value of 0.97 p.u. with no minimum and maximum voltages.

In addition, the voltage sag compensated by the D-STATCOM,

shows that the maximum, minimum and steady-state load

voltages are 1.14, 0.87, and 0.98 p.u., respectively. It can be

concluded from the simulation results that the USSC show a

better voltage sag compensation capability as compared to the

DVR and the D-STATCOM in terms of the voltage magnitudes.

B. USSC for Voltage Flicker Reduction

Voltageflicker which is a phenomenon of annoying light in-

tensityfluctuation caused by variable electric loads and arc fur-

naces have been a major power-quality concern. To illustrate

the use of the USSC in reducing voltage flicker, simulations

were carried out by first connecting a variable electric load of

5.2 MVA, 22 kV as the source of voltage flicker. Fig. 6(a) shows

theflicker effect of a phase rms voltage for the system without

the USSC connected. By connecting the USSC, it can be seen

that the rms voltage of phase A is flicker free, as shown in

Fig. 6(b). The results shown here are for the phase A root-mean-

square (rms) voltage but however the responses are similar forthe phase B and C voltages. Fig. 6(c) shows the rms voltage of

phase A using D-STATCOM in the system.

Many techniques have been proposed for evaluating the

voltageflicker levels. In this simulation, the fast Fourier trans-

form (FFT) technique is further used to calculate the voltage

flicker index. The simulation results shows that without the

USSC connected, the effect of the variable electric load results

in a voltage flicker index of 0.40 in which the value exceeds

its IEEE standard limit of 0.07. However, with the USSC

connected, it is noted that the calculated voltage flicker index

is reduced to 0.002. It is also observed that the D-STATCOM

can reduce the voltageflicker to 0.002. These results prove that

both the USSC and the D-STATCOM can be used for reducingvoltageflicker.


7/25/2019 ussc.pdf

5/7


Fig. 6. Rms Voltageflicker: (a) without USSC, (b) with USSC connected, and(c) with D-STATCOM connected.

C. USSC for Voltage Unbalance Mitigation

The effect of voltage unbalance is detrimental as it causes

heating in motors, thus requiring them to be derated. Unbalance

can also affect sensitive single phase loads because it creates

undervoltage in one or more of the lines. Therefore, it is impor-

tant to investigate on whether the USSC can mitigate voltage

unbalance. In this simulation, initially an unbalanced voltage

condition is created by applying two single phase to groundfaults on the phase A and phase C at time for a

fault duration of 100 ms. Fig. 7(a) shows the simulation results

of the three-phase unbalanced voltages for the system without

the USSC connected. It can be seen that during the fault con-

dition, the maximum phase voltages are ,

, and . The percentage of voltage unbal-

ance is calculated and found to be 28.7%. This value indicates

that the voltage unbalance is severe during the fault because the

limit of the voltage unbalance is specified as 2%.

With the application of USSC, the three-phase load voltages

are recorded as shown in Fig. 7(b). It is evident from Fig. 7(b)

that in the presence of the USSC, the load voltage profile has

improved in which the phase A and C voltages are increasedand the phase B voltage is reduced, thus making the three phase

voltages more balanced. The percentage of voltage unbalance

decreases from 28.7% to 1.6%. For cases with the DVR and

D-STATCOM connected, the simulation results show that the

percentages of voltage unbalance are reduced to 5.02% and

2.2%, respectively, as shown in Fig. 7(c) and 7(d).

D. USSC Acting in UPS Mode

To show that the USSC can operate in uninterruptible power

supply (UPS) mode, an outage isfirst created at time

for a duration of 0.75 s using a three-phase fault generator. The

outage simulation result is as shown in Fig. 8(a). When theUSSC is connected in the system, the USSC recovers the load

Fig. 7. Three-phase load voltages: (a) without USSC, (b) with USSC, (c) withDVR, and (d) with D-STATCOM.

Fig. 8. Load voltages:(a) short outage,(b) with USSC, (c) with D-STATCOM,and (d) with DVR.

voltage from 0.0 to 1.00 p.u. within a short time as shown inFig. 8(b). In the UPS mode, both the series and shunt inverters


7/25/2019 ussc.pdf

6/7


Fig. 9. Active and reactive power flows: (a)without USSC and (b) with USSC.

of the USSC operate in parallel and support a load of the sum

of the inverters rating.

Fig. 8(c) and 8(d) show the results of the operation of the

D-STATCOM and the DVR in UPS mode in which the per unit

voltages are recovered to 0.98 and 0.96 p.u, respectively. In

the UPS mode, the DVR works better than that of the USSCand the D-STATCOM because compensated load voltage by

DVR is a steady-state value with no minimum and maximum

voltage at the starting and the ending of voltage sag. Though

the compensation capability is 0.96 p.u., which is less than that

of D-STATCOM and USSC.

E. USSC for Power-Flow Control

Theflow of instantaneous active and reactive powers into or

out of the USSC are investigated using the transient simulation.

When a fault occurs at time for a duration 0.75 s,

the active and reactive powers into the system are as shown in

Fig. 9(a). The simulation result indicates that during the fault pe-riod, both the active and reactive powers of the system increase.

From the USSC operation point view, the increase in the re-

active power requires an increase in the active power injected

by the series inverter of the USSC and an increase in the active

power supplied by the shunt inverter of the USSC. The exchange

of real power can be made in either direction between the series

and shunt inverters of the USSC. With the USSC connected in

the system, the reactive power of the system is reduced from

1.2 MVAr to zero in order to achieve a steady-state value of ac-

tive power, as shown in Fig. 9(b). Thus, the active and reactive

powerflows are controlled and maintain at a pre-fault levels.

F. Harmonic Elimination

Simulation results in Fig. 10 show that the USSC inverters

generate a voltage total harmonic distortion (THD) of 78%. Due

to high frequency switching losses, the inverters have generated

a THD which is higher than the acceptable level of 5%. There-

fore,filtering is indispensable so as to eliminate the harmonics

generated by the USSC. Several methods can be used for re-

ducing the harmonics produced by the USSC. In this simulation,

a passivefilter is connected at the load side of the distribution

system.

To illustrate the effect of using an inductancecapacitance

(LC) passivefilter, simulations were carried out and the THD of

the system without and with the filter inserted into the system arerecorded as shown in Fig. 10(a) and (b), respectively. From the

Fig. 10. Total harmonic distortion: (a) withoutfilter and (b) withfilter.

TABLE IIPQ MITIGATION OFUSSC VS. D-STATCOM AND DVR

simulation results, it can be seen that with thefilter connected,

the harmonics are suppressed and the THD of the system is re-

duced from 78% to 1.88%. The THD of 1.88% is far below the

value of the IEEE standard THD limit of 5%. The size of the

passive filter considered in the simulation are and

.

G. Capabilities of USSC Versus D-STATCOM and DVR

Usually individual custom power devices such as

D-STATCOM and DVR are concentrated on solving defi-

nite power-quality problems in a distribution system, as shown

in Table II. However, by using USSC, it is possible to com-

pensate a variety of power-quality problems as compared to

D-STATCOM and DVR as stated in Table II.

It is noted that in the UPS mode, mitigated load voltage by the

DVR is a steady state value, where as in the D-STATCOM and

the USSC have minimum and maximum values at the starting

and the ending of the fault.

VI. CONCLUSION

A unified approach to the mitigation of multiple power-

quality problems has been investigated by using USSC. The

two level, USSC incorporating 12-pulse series and shunt con-

nected inverters has been modeled in PSCAD/EMTDC program

and a new SPWM-based control scheme has been implemented

to control the GTOs of the inverters. Simulations have been

carried out to evaluate the performance of the USSC under

various operating conditions and power-quality disturbances.

Simulation results prove that the USSC can perform voltage sag

compensation,flicker reduction, voltage unbalance mitigation,UPS mode and power-flow control. It was also observed that


7/25/2019 ussc.pdf

7/7


harmonics generated by the USSC can be significantly reduced

by inserting a passive filter into the system. A comparison is

made between the USSC and the other custom power devices

such as D-STATCOM and DVR in terms of their capabilities

in power-quality mitigation. It is shown that the USSC gives

a better performance in power-quality mitigation, especially

in voltage sag compensation and power-flow control, andalso provide more power-quality solutions as compared to the

D-STATCOM and DVR.

REFERENCES

[1] J. R. Enslin, Unified approach to power quality mitigation, in Proc.IEEE Int. Symp. Industrial Electronics (ISIE 98), vol.1, 1998, pp. 820.

[2] B. Singh, K. Al-Haddad, and A. Chandra,A review of activefilters forpower quality improvement,IEEE Trans. Ind. Electron., vol. 46, no. 5,pp. 960971, 1999.

[3] M.-C. Wong, Y.-D. Han, and L.-B. Zhao,Study of distribution systemunified conditioner (DS-UniCon),in Proc. 3rd Int. Power Electronicsand Motion Control Conf. (PIEMC), vol. 3, 2000, pp. 13651370.

[4] A. L. Olimpo and E. Acha,Modeling and analysis of custom powersystems by PSCAD/EMTDC,IEEE Trans. Power Delivery, vol. 17, no.

1, pp. 266272, Jan. 2002.[5] P. Pohjanheimo and E. Lakervi, Steady state modeling of custom power

components in power distribution networks,in Proc. IEEE Power En-gineering Society Winter Meeting, vol. 4, Jan. 2000, pp. 29492954.

[6] N. G. Hingorani and L. Gyugyi, Understanding FACTS Concept andTechnology of Flexible AC Transmission System. New York: IEEE

Press, 2000.[7] R. L. V. Arnez and L. C. Zanetta, Unified power flow controller

(UPFC): its versatility in handling power flow and interaction with thenetwork, in Proc. IEEE/PES Asia Pacific Transmission and Distribu-tion Conf. and Exhib., vol. 2, Oct. 2002, pp. 13381343.

[8] D. E. Soto-Sanchez and T. C. Green,Voltage balance and control in amulti-level unified powerflow controller,IEEE Trans. Power Delivery,vol. 16, no. 4, pp. 732738, Oct. 2001.

[9] C. Schauder,The unified powerflow controller-a concept becomes re-ality, in IEE Colloq. Flexible AC Transmission SystemsThe FACTS,1998, pp. 7/17/6.

[10] A. D. Le Roux and H. T. Mouton, A series-shunt compensator withcombined UPS operation, in Proc. IEEE Int. Symp. Industrial Elec-tronics, vol. 3, 2001, pp. 20382043.

[11] C. Su and G. Joos,Series and shunt active power conditioners for com-pensating distribution system faults,in Canadian Conf. Electrical and

Computer Engineering, vol. 2, Mar. 2000, pp. 11821186.

M. A. Hannan receivedthe B.Sc. degree in electricaland electronic engineering from Bangladesh Institute

of Technology (BIT), Chittagong, Bangladesh, in1990. He is currently pursuing the M.Sc. degree in

the area of custom power devices at the UniversitiKebangsaan Malaysia, Selangor, Malaysia.

He was with Bangladesh Power DevelopmentBoard from 1990 to 2000. His research interests arein custom power devices and power electronics.

Azah Mohamed (M90) received the B.Sc. degreefrom the University of London, London, U.K., in1978, and the M.Sc. and Ph.D. degrees from theUniversity of Malaya in 1988 and 1995, respectively.

Currently, she is a Professor in the Departmentof Electrical, Electronics, and Systems Engi-neering, Universiti Kebangsaan Malaysia, Selangor,Malaysia. Her research interests are in power quality,especially harmonics, voltage sag studies, andcustom power devices.

ussc.pdf

Documents