Power System Reliability Enhancement Using Unified Power Flow Controllers

M. FOtuhi-Firuzabad R Billinton S.O. Faried S.Aboreshaid

Power Systems Research Group Department of Electrical Technology
University of Saskatchewan Riyadh College of Technology
Saskatoon, Canada Riyadh, Saudi Arabia

Abshact: A major requirement in the application of FACTS devices Interline Power Flow Controllers (IPFC) [2]. This paper
in power systems is to develop techniques which enable system focuses on the impact on power system reliability of
planners to manage with great confidence the uncertainty associated employing UPFC. The UPFC is employed in the system to
with these devices. This paper presents an approach to evaluate adjust the natural power sharing of two different parallel
transmission system reliability when employing a Unified Power transmission lines and therefore enable the maximum
Flow Controller (UPFC). This paper provides a framework within transmission capacity to be utilized.
which the risk associated with management and the development of
The Unified Power Flow Controller (UPFC) is the most
the transmission network can be quantified. Reliability indices such
versatile FACTS device that has emerged for the control and
as the Loss of Load Probability (LOLP), Loss of Load Expectation
optimization of power flow in electrical power transmission
(LOLE), the Loss of Energy Expectation (LOEE) and the System
systems. It offers major potential advantages for the static
Minutes (SM) are utilized to examine the impact of UPFC on the
and dynamic operation of transmission lines since it combines
transmission system reliability. The technique is illustrated by
the features of both the Static Synchronous Compensator
application to a hypothetical system.
(STATCOM) and the Static Synchronous Series Compensator
Keywords. Flexible AC transmission systems, Power system (SSSC)[3-51.
reliability, Unified power flow controllers.
A major requirement in the application of FACTS devices in
I. INTRODUCTION power systems is to develop techniques which enable system
planners to manage with great confidence the uncertainty
The development and utilization of transmission systems
associated with these devices. This paper presents an
involves the integration of several important considerations,
approach to evaluate transmission system reliability when
some of which can be competitive. Present-day systems and
employing a UPFC. This paper provides a framework within
operational structures have necessitated that increased
which the risk associated with management and the
attention be given to this critical power system sector. Better
development of the transmission network can be quantified.
utilization of existing power system capacities and greater
Reliability indices such as Loss of Load Probability (LOLP),
operating flexibility have become imperatives and must be
Loss of Load Expectation (LOLE), the Loss of Energy
achieved without endangering the security and reliability of
Expectation (LOEE) and the System Minutes (SM) are utilized
the transmission system. The required transmission capacity
to examine the impact of UPFC on the transmission system
for a given power flow pattern must satisfy the transmission
reliability. The technique is illustrated by application to a
security constraints. Transmission capacity requirements can
hypothetical system.
be achieved by either constructing new transmission lines or
increasing the transfer capability associated with the existing II. THE UNIFIED POWER FLOW CONTROLLER
facilities[l].
The UPFC shown in Fig. 1 consists of two switching
Flexible AC Transmission Systems (FACTS) was envisaged converters where each converter is a voltage-sourced inverter
as a means of increasing the transmission transfer capability using gate-turn-off (GTO) thyristor valves. These inverters,
of existing transmission lines, thus deferring the need to build labeled “Inverter 1” and “Inverter 2” in the figure, are
new lines. Because of their unique features relative to the operated from a common dc link provided by a dc storage
presently used control hardware, FACTS devices offer several capacitor. This arrangement functions as an ideal ac to ac
new opportunities for better control of large electric power power converter in which the real power can flow freely in
systems. The most striking feature is the ability to directly either direction between the ac terminals of the two inverters
control transmission line flows by structurally changing and each inverter can independently generate (or absorb)
parameters of the grid, and to implement high-gain type reactive power at its own ac output terminal. Inverter 2
controllers based on fast switching. In the USA and other provides the main function of the UPFC by injecting an ac
developed countries where right-of-the-way to build new voltage Vi with controllable magnitude and phase-angle a at
transmission lines are difficult to secure, FACTS has the power frequency, in series with the line via an insertion
stimulated research and development and in some cases, transformer q . This injected voltage can be considered
prototypes installations of Thyristor Controlled Series
essentially as a synchronous ac voltage source. The
Capacitors (TCSC), Static Synchronous Series Compensators
transmission line current flows through this voltage source
(SSSC), Unified Power Flow Controllers (UPFC) and the
resulting in real and reactive power exchange between it and

IEEE
0-7803-6338-8/00/$10.00(~)2000 745
 the ac system. The real power exchanged at the ac terminal p ,Q
(i.e., at the terminal of ) is converted by the inverter into dc
power which appears at the dc link as positive or negative real
power demand. The reactive power exchanged at the ac
terminal is generated intemally by the inverter.

Fig. 1. Simplifiedtransmission system and the UPFC model.

The basic function of Inverter 1 is to supply or absorb the real

power demand by Inverter 2 at the common dc link. This dc
link power is converted back to ac and coupled to the
transmission line via a shunt connected transformer T,.
A. Basis of Power Flow Control by a UPFC (b)
Fig. 2. Principle of the UPFC: (a) equivalent circuit, (b) phasor
When analyzing power flow in a transmission system with the diagram.
UPFC shown in Fig. 1, the output of Inverter 2 can be
represented by a voltage source with controllable magnitude III. TEST SYSTEM
and phase-angle as shown in Fig. 2(a). Fig. 2@) illustrates the The UPFC is applied in this paper to a system with two
system phasor diagram. Assuming that the sending and different parallel transmission lines. When two transmission
receiving end voltages are equal in magnitude lines are connected in parallel, the natural power sharing
I I
QVs = lVR = V) and the magnitude of the injected voltage is between them is dictated by their respective impedances. In
IV,l= kV, where k e 1 . The transmission line current is then this case, the UPFC is added to the transmission line with
lower thermal rating. If the two lines are of different
given by:
configurations (and consequently of different thermal
ratings), their impedances could still be very close [6].
I=
v, + <. -v, Therefore, the power transmitted in each line will be similar.
fl The voltage drop in both circuits is identical, and therefore,
V the relationship between the line currents I,, and I,, can be
= -[(ksina-sin p)- j(kcosa+~-cosp)] (1)
X expressed as:

The active and reactive power transmitted by the sending end (4)
system are calculated using Equations 2 and 3.
If overload of the lower thermal rating line, 4 , is to be

V2
Q = Im g b >=-(kcosa+ 1-cos /3)
avoided (i.e., I,, II,,-

a UPFC is added to such a system, both of the transmission

-
), then the full power capacity of
the other line, L, ,will never be reached (i.e., I,, e I,, ). If

(3) lines can operate at their maximum capacity when the

X
appropriate voltage level V, is injected.
It can be seen from Equations 2 and 3 that the real and
reactive power are proportional respectively to the imaginary Fig. 3 shows the single line diagram of the test system. It
and real parts of the injected voltage. If these two consists of local and remote generating stations
components could be controlled individually, then the real interconnected by two transmission lines @,,4) of unequal
and reactive power transmitted could be controlled thermal ratings. Transmission l i n e 4 is a four conductor
independently.

746
bundle (quad) circuit configuration whereas transmission line zero depending on the availability of the three components.
L, has a two conductor bundle (twin) circuit configuration. The transition rates among the states can be determined using
The rating capacity of 4 is therefore two times of that of L,. the frequency and duration technique [7]. Having all the three
components available, L., and L, will have maximum transfer
A UPFC is connected to the twin circuit at the sending end of
the transmission system. capabilities of 2P and 1P respectively where P is the rated
capacity of Line L,. If the UPFC fails, the maximum transfer
The system has a generating capacity of 11050 MW where the
capability of L., will decrease to 1P in order to avoid
remote generating site provides 40.7% of the total system
generation and the local generating site contributes the other overloading of L2. The state associated with this case has a
59.3%. The system peak load is assumed to be 8500 MW. transfer capability of 2P. It is worth noting that if L, fails, the
The thermal rating capacities associated with transmission maximum transfer capability of L, is 2P since the overloading
lines L., and L, are 3000 and 1500 MVA respectively. A
constraint of L, no longer exists. The transmission capacity
UPFC with a 450 MVA rating is connected to the twin circuit
model decreases to a three-state model when the UPFC is not
at the sending end of the transmission system. The system
included in the transmission system. The capacity associated
data are given in the Appendix. The availability of the UPFC is
with the three states are 2P, 1P or zero depending on the
98% based on the data given in the Appendix.
availability of transmission lines 4 and L, . Table 1 shows
the transmission capability model with and without inclusion
of the UPFC. The transmission capability model for the case
with the UPFC is calculated considering that the availability of
the UPFC is 0.98.
Table 1. Transmission line model.

Load
Fig. 3. Single line diagram of the test system.
Without the UPFC
Transfer
Capability
I
Probability
I
I
With the UPFC
Transfer
Capability
I
Probability
I
[MW] [MW]
IV. RELIABILllYMODEL 3000 0.99788680 4500 0.97586251
1500 0.00210873 3000 0.02202429
Each major component in a power system has a unique
stochastical behavior that identifies its degree of reliability in 0 0.00000447 1 5 ~ 0.00210873
the overall network. The availability and reliability of a system I n 10.00000447 I
depend on the performance of each element within the
system. The failure rate and repair time of individual v. STUDY RESULTS
components are therefore essential parameters in evaluating The test system shown in Fig. 3 is used to measure the
the system reliability. Depending on the accuracy reliability impact of employing the UPFC in a transmission
requirements and the intent behind the studies, component system. System reliability is usually predicted using one or
behavior is usually modeled using a state-space more indices which quantify the expected system reliability
representation. performance. The most frequently used indices are the Loss
The generating units and transmission lines are represented Of h a d Expectation (LOLE) and the Loss Of Energy
by two-state models in which a component is either in the Expectation (LOEE) [8]. The LOLE is the average number of
operating state or in the down state 171. The UPFC is also days (or hours) in a period during which the load is expected
represented by a two-state model in which it is assumed that it to exceed the available generating capacity. The LOLE,
is either available or it resides in the failed state. It is assumed therefore, is an indication of the expected number of days (or
that the UPFC is bypassed by a fully reliable circuit breaker hours) in which the load, partial or fully, is not satisfied. This
without interrupting transmission line L,when the UPFC fails. risk index does not provide any information as to how severe
or frequent the loss is, however, LOLE is the most widely
Considering a two-state model for each component, the three used criterion at the present time. The LOEE, also designated
components L.,, L, and the UPFC can be represented by an as the Expected Energy Not Supplied (EENS) or the Expected
eight-state model [7] to recognize the available transfer Unsupplied Energy (EUE),is the expected energy that will not
capability associated with the transmission system. Assuming be supplied by the system due to those occasions when the
that the three components of UPFC, L, and L, are load demanded exceeds the available capacity at the load
independent, the eight-state model can be reduced to a four- point. This index has certain advantages, as it incorporates
state equivalent model where the states with identical the severity of the deficiency. The complement of LOEE, the
capability are considered to be the same. The transfer expected energy supplied, is sometimes divided by the total
capability associated with each state could be 3P, 2P, 1P or energy demanded to yield the normalized index known as the

747
 Energy Index of Reliability (EIR). This index can be used to Table 2. Reliability indices without the UPFC.
compare the adequacy of systems that differ in size. Other PK LOLP LOLE LOEE UPM EIR
normalized values of LOEE, such as System Minutes (SM) [MW] [hdyr] [GWhr/yr] I
and Units Per Million (UPM) are also used by some power 7500 0.001333 11.65 2.249 34.327 10.999966
utilities. The load model is assumed to be the straight load
8500 0.101648 888 233.73 3147.6 0.996852
duration curve (LDC). The equivalent capacity model at the
load point is convolved with its load model in order to assess 9000 0.465155 4063 1317.7 16760 0.983239
the system risk [8]. 10000 1.0 8736 8663.6 99171 0.900829
Fig. 4 shows the variation in the LOLE and LOEE as a function
of the system peak load. The system load factor is assumed to
be 100% indicating that the LDC is a straight line at the
system peak load throughout the year. It can be seen that for
a given peak load, the risk indices (LOLE and LOEE) decrease
significantly when the UPFC is included. The inclusion of the
UPFC allows the capacity of transmission line L, to be
extended to its thermal limit and, therefore, transfer more
available capacity from the remote generation site to the load
point. The UPFC effect becomes more significant as the peak case in a practical power system, as the system load profile
load increases. The reason for this is that for lower system varies throughout the year. The variation in the system load
peak loads, the local generation has sufficient capacity to profile can be represented by the system load factor. Fig. 5
prevent load interruption. Tables 2 and 3 show the numerical shows variations in the LOLE, LOEE and SM indices as a
values of the system indices for four different load levels. The function of the system load factor. The annual system peak
system risk (LOLP) is one at the load level of 10000 MW when load is assumed to be 8500 MW. It can be seen from the
UPFC is not employed. The reason for this is that the results that the reliability indices increase as the system load
maximum capacity available at the load point is 9550 MW and factor increases.
therefore there is always some load curtailment throughout
In order to provide a set of specific numerical indices and a
the year. datum for comparing alternate configurations, the reliability
In the previous studies, the reliability analysis of the system indices are given in Tables 4 and 5 . Table 4 shows the basic
was conducted using an annualized system load curve where reliability results while the impact on the indices of
the analysis was conducted for a single load level and incorporating the UPFC in the transmission system is
expressed on a base of one year. This, however, is not the presented in Table 5.

1000 1
It can be seen by comparing the results shown in Table 3 with
-loooo without UPFC those shown in Table 4 that the system reliability improves

7500 7750 8000 8250 8500 8750 9000 9250 9500

Load [MW]

10000 7

0.1 ! 1

1500 7750 8000 8250 8500 8750 9000 9250 9500

Load [MW]

Fig. 4. Effect of employing the UPFC on the LOLE and LOEE

indices

748
 IOoo 1 J
c(
30 1
3 25
2 20

!3 -- 15
U

s
IO
t
B
10

. . 1 . . , . . . . . . 5
50 55 60 65 70 75 80 85 90 95 100 0
Load factor [%I
7000 7500 8000 8500 9000 9500 10000
Peak load W W ]
1000 7
Fig. 6. Percentage changes in the LOLE at various load levels.

system peak load and the associated load factor are assumed
to be 8250 MW and 85% respectively. It can be seen from the
Y

ru
results that the LOLE and LOEE increase as the availability of
the UPFC decreases. The rate of system reliability
improvement increases up to a specific load level and then
decreases. The reason for this is that as the system peak load
50 55 60 65 70 75 80 85 90 95 100 increases, there is a point at which deficiencies are due to
Load factor [%I generating capacity rather than transmission problems. Fig. 6
shows the variation in the percentage rate of decrease in the
10000 1 LOLE when the availability of the UPFC increases from 92% to
98%.
VI. CONCLUSIONS

This paper presents an approach to evaluate transmission

system reliability when employing a UPFC. This technique
provides a h e w o r k within which the risk associated with
management and the development of the transmission
network can be quantified. A set of reliability indices such as
50 55 60 65 70 75 80 85 90 95 100 Loss of Load Probability (LOLP), Loss of Load Expectation
Load factor [%]
(LOLE), the Loss of Energy Expectation (LOEE) and the
Fig. 5: Impact of load factor variation on the reliability indices. System Minutes (SM) are presented to measure the impact of
the UPFC on the transmission system reliability. The
considerably with the inclusion of the UPFC in the analysis. technique is illustrated by application to a hypothetical test
The results presented in both tables also show that for a system where the UPFC is used to adjust the natural power
given peak load the system risk increases as the system load sharing of two different parallel transmission lines and,
factor increases. therefore, enable the maximum transmission capacity to be
utilized. The results presented indicate that the system
In the previous analyses it was assumed that the UPFC has an
reliability is considerably improved by using a UPFC. The
availability of 98%. Table 6 shows the impact on the system
effects on the transmission system reliability of variation on
reliability indices of variation inthe UPFC availability. The
the system load factor and the availability of the UPFC are
Table 6. Effect of the UPFC availability on the reliability indices. also presented.

A LOLP LOLE LOEE UPM EIR w. RE”m

[hr/yr] GWhr/yr [l] C. J. Ray, A. R Crathome, “Transmission Capacity Planning
80% 0.001381 12.062 3.943 64.374 0.999936 in an Open System”, Proceedings of the W Symposium of
84% 0.001239 10.827 3.695 60.326 0.999940 Specialists in Electric Operational and Expansion Planning
I 88% I 0.001098 I 9.593 I 3.447 I 56.277 I 0.999944 I (SEPOPE), Curitiba Brazil, May 21-26,2000.
92% I I
0.000957 8.358 I 3.199 I 52.229 I 0.999948 [2] B. Mwinywiwa, et. al., ‘‘MUlti-Tenninal UPFC for Power
98% I 0.000745 I 6.507 I 2.827 I 46.16 I 0.999954 System Deregulation”, Proceedings of the IEEE Winter
A UPFC availability
Meeting, Singapore,2000.

749
 [3] IEEE FACTS Working Group, IEEE Transmission and and M.Sc. and Ph.D. Degrees in Electrical Engineering from the
Distribution Committee, “FACTS Applications”, IEEE University of Saskatchewan where he is presently an Associate
Professor of Electrical Engineering. His research interest includes
Publication No.96TP 116-0.
power system dynamics and electrical machines.
[4] L. Gyugyi, “A Unified Power-Flow Control Concept for
Flexible AC Transmission Systems.” IEE hoceedings-C, Vol. Saleh Aborerbnid (M’1997) obtained B.Sc. Degree from King Saud
139, 1992,pp. 323-331. University in 1990 and M.Sc. and Ph.D. Degrees in Electrical
Engineering From the University of Saskatchewan in 1993 and 1997
[SI C.D. Schauder, L. Gyugyi, M.R. Lund, D.M. Hamai, T.R
respectively. Presently Head of the Electrical Engineering Department
Rietman, D.R Torgerson and A. Edris, “Operation of the at Riyadh College of Technology. His resewh interest includes power
Unified Power Flow Controller (UPFC) Under Practical system reliability and dynamics.
Constraints,” IEEE Transactions on Power Delivery, Vol. 13.
1998,pp. 630-636.
[6] P.M. Anderson, Analysis of Faulted Power Systems, IEEE
Press, 1995.
[7] R Billinton, R N. Allan, Reliability Evaluation of Engineering
Systems, Plenum Press, 1994.
[8] R Billinton, R N. Allan, Reliability Evaluation of Power
Systems, Plenum Press, 1996.

VIII. APPENDIX

Table I. System Data

Cap. Lineimpedance,p.u.h 1 r
Line
MVA (100 MVA. 500 kV Base) [fh] [hrs]
L, 3000 0.0012+j0.016 1.85 10

Lz 1500 0.0024+j0.019 1.85 10

K BIOGRAPHIES
Mabmud Fotuhi-Firuzabad (S’1994, SM’1999) obtained B.Sc. and
M.Sc. Degrees in Electrical Engineering from Sharif University of
Technology and Tehran University in 1986 and 1989 respectively and
M.Sc. and Ph.D. Degrees in Electrical Engineering from the University
of Saskatchewan in 1993 and 1997 where he is presently a Post-
doctoral Fellow.
Roy Billinton (F’1978)obtained B.Sc. and M.Sc. Degrees from the
University of Manitoba and Ph.D. and D.Sc. Degrees from the
University of Saskatchewan. Worked for Manitoba Hydro in the
System Planning and Production Divisions. Joined the University of
Saskatchewan in 1964. Formerly Head of the Electrical Engineering
Department. Presently C.J. Mackenzie Professor of Engineering and
Associate Dean, Graduate Studies, Research and Extension of the
College of Engineering.
Author of papers on Power System Analysis, Stability, Economic
System Operation and Reliability. Author or co-author of eight books
on reliability. Fellow of the IEEE, the EIC and the Royal Society of
Canada and a Professional Engineer in the Province of Saskatchewan.
Sberif Omar Faried (M’1993,SM’2000) obtained B.Sc. and MSc.
Degrees in Electrical Engineering from Ain Shams University, Egypt

750