Electrical Power and Energy Systems 59 (2014) 204–213

Contents lists available at ScienceDirect

Electrical Power and Energy Systems

journal homepage: www.elsevier.com/locate/ijepes

Optimal power flow considering fault current level constraints and fault
current limiters
Amirhossein Khazali, Mohsen Kalantar ⇑,1
Center of Excellence for Power System Automation and Operation, Iran University of Science and Technology, Iran

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this paper is to present a novel method to dispatch the active generation power properly in the
Received 1 January 2013 power system while incorporating fault current levels as constraints for the optimization problem. Due to
Received in revised form 9 December 2013 the limited capacity of protective devices such as circuit breakers, allocating active power without con-
Accepted 18 February 2014
sidering fault current levels can probably lead to fault currents exceeding the rating of these devices.
Available online 22 March 2014
Hence restricting the fault current levels to an allowable amount while minimizing a specified objective
function seems to be necessary. In a number of cases even the appropriate allocation of active power is
Keywords:
not able to reduce the fault current levels to the permitted amount therefore using fault current limiters
Optimal power flow
Fault current level constrained
(FCL) is unavoidable. In this paper also a planning scheme is presented for the location and sizing of fault
Fault current limiter (FCL) current limiters and the effect of fault current limiters on the objective functions is investigated.
Optimal location Ó 2014 Elsevier Ltd. All rights reserved.
Fault current

1. Introduction the permitted value which is determined by the rating of circuit

The optimal power flow problem is affective on secure and
economical operation of power systems. This problem denotes
optimal settings of control variables such as active power of
generators, tap ratios of transformers and generator bus voltages
to minimize a certain object while satisfying equality and inequal-
ity constraints. Transformer tap settings is a discrete value while
bus voltage magnitudes and active power generation outputs of
generators are continuous variables so the OPF problem can be
modeled using mixed integer nonlinear programming.
The optimal power flow problem has been investigated in many
works. In [1–4] the problem is solved using mathematical ap-
proaches. Also in [5–9] the problem has been solved using different
heuristic approaches assuming different objective functions such
as total fuel cost, active power line loss, voltage stability and volt-
age deviations. Although in [10,11] the optimal power flow prob-
lem is solved with regarding fault current limits, no work is
presented that solves the OPF problem regarding the fault current
levels and effect of fault current limiters. Also it should be noticed
that solving the OPF problem considering fault current levels can
lead to an unfeasible problem. In other words there is no solution
for the OPF problem which can suppress the fault currents to under
⇑ Corresponding author. Tel.: +98 9122193154.
E-mail addresses: amirhossein.khazali@gmail.com (A. Khazali), kalantar@iust.
ac.ir (M. Kalantar).
1
Tel./fax: +98 2173225662.
breakers. In this state using fault current limiters in the system is
unavoidable.
In the last decade considering the increasing demand for power,
electric systems have expanded rapidly. Consequently, the level of
fault currents has increased and can exceed the fault currents of
circuit breakers installed in the system. Hence the fault current le-
vel can create a critical situation in the power system especially
when the highest capacities of circuit breakers are used in the sys-
tem [11].
In the recent years fault current limiters (FCL) have been used as
effective devices to overcome high fault current levels. FCLs are
capable of limiting the fault current at the first peak and also lim-
iting short circuit current at steady state without disturbing the
normal operation. Different types of FCLs are used in power sys-
tems including FCLs using power devices and superconductivity
[12]. Fault current limiters have a very low impedance during their
normal operation. However when a fault occurs these devices in-
crease their impedance [13]. In [14] a hybrid fault current limiter
is presented for distribution systems. The introduced device is
implemented to a 11 kV distribution systems with distributed gen-
eration. In [15] the FCL is utilized by an impedance combined with
bus sectionalizing circuit breakers. The economical observations
show the profit of the combination. [16] Proposes solid state fault
current limiters as cost efficient approach to minimize distributed
generation expansion in the distribution network. Genetic algo-
rithm is used in this paper for determining the optimal number,
location and size of fault current limiters with the object of

http://dx.doi.org/10.1016/j.ijepes.2014.02.012
0142-0615/Ó 2014 Elsevier Ltd. All rights reserved.
 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213
minimizing protection costs. In [17] in addition to the limitation of
fault currents the FCL is used for preventing voltage sags. In the
presence of distributed generation the settings of over current re-
lays have to be readjusted. However in [18] fault current limiters
are used for restoring the coordination of over current relays.
Utilizing fault current limiters can facilitate the connection of
independent power producers (IPP) to the system. It can also in-
crease the capacity of lines and totally increase the security and
reliability of the system. However, the impedance and installa-
tion location of fault current limiters have an important effect
on the improvement of the system, therefore developing a meth-
od that can determine the sufficient number of FCLs, there loca-
tion and impedance seems to be necessary. In radial systems the
appropriate place for installing fault currant limiters can be spec-
ified simply. But in loop systems this problem is complicated and
requires a suitable method to determine the location, number
and impedance of FCLs considering the system specifications
[19,20].
In [20] a micro genetic algorithm approach has been used for
solving the optimization problem considering superconducting
fault current limiters. Also [21] utilizes a genetic algorithm and
sensitivity factor calculation method. In [22] a rectifier type super-
conducting fault current limiter is placed in a large scale power
system. In this work a method is implemented for determining
the optimal placement and location of fault current limiters.
In this paper initially the OPF problem is solved regarding the
fault current levels. It is indicated that for a number of specified
cases the proper allocation of power is capable of suppressing
the fault current levels to under the rated amount of protection de-
vices such as circuit breakers. But in other cases even the appropri-
ate settings of control devices are not adequate and using fault
current limiters are inevitable. In the next stage an optimal FCL
programming is represented for specifying the optimal location
and amount of fault current limiters. Also the impacts of using FCLs
in the system for suppressing the fault current levels to under the
permitted amount is investigated on total generation cost, power
loss of the system and voltage deviations. The proposed method
Fig. 1. Thevenin equivalent when line is added between k and j.
205
is implemented to the New England 39-bus system and the results
exhibit the effectiveness of the proposed method.
2. Fault current calculation and fault current limiters
The majority of faults in power systems are unsymmetrical
however the three phase fault is the most intensive type of faults
and is used for specifying the rating of circuit breakers. In the
following the calculation of fault current at each bus and the effect
of three phase faults on the currents flowing in the lines are
described:
2.1. Fault current at bus
For a symmetrical fault at bus i the fault current can be obtained
by (1):
ISC
i ¼ ðEi =Z ii Þ  I b ð1Þ
where ISC
is the fault current at bus i and Ei is the voltage before the
i
fault at bus i which is usually assumed to be 1 p.u. Zii is the diagonal
members of the impedance matrix. Finally Ib is the base current
[21].
By adding the impedance Zb between buses j and k each ele-
ment of the impedance bus is modified as [21]:
Z new new
xy ¼ Z xy  ðZ xj  Z xk Þ  ðZ jy  Z ky Þ=ðZ jj þ Z kk  2Z jk þ Z b Þ ð2Þ
where Z new
is the modified element of the impedance matrix. There-
xy
fore the effect of inserting the impedance Zb series with the trans-
mission line is equivalent to inserting the impedance Zp parallel
with the transmission line which can be obtained by the following
relation:
Z p ¼ ðZ b ÞjjðZ b þ Z FCL Þ ¼ Z b ðZ b þ Z FCL Þ=Z FCL ð3Þ
Fig. 1 shows the Thevenin equivalent from the bus under study
when impedance Zb is added between two buses. Finally Zp is used
to modify the elements of the impedance matrix by the below
relation:
206 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213
ðZ ij  Z ik Þ2
DZ ii ¼  ¼ C 2 =ðC 1 þ Z p Þ
Z jj þ Z kk  2Z jk þ Z p
Hence the amount of Zp required to reduce the fault current from Ii,N
to Ii,F can be calculated from:
Z p ¼ ðIi;F  C 2 Þ=ððIi;N  Ii;F Þ  Z ii Þ  C 1
Finally the impedance of the used fault current limiter is:
Z FCL ¼ Z 2b =Z b þ Z p
Fig. 2 illustrates the Thevenin equivalent with the FCL in the
system.
2.2. Fault current flowing through lines
During a fault in the power system if the capacity of protection
equipment are not sufficient to tolerate the fault currents it can
cause damages to the equipment and also interrupt the operation
of the system.
Usually the analysis of fault currents is focused on their magni-
tude and the magnitude of the fault current flowing from bus i to
bus j caused by a fault at bus f should Ifi;j not exceed the rating
amounts of equipments used in the system (jIfi;j j < jIspec j).
The fault current flowing from bus I to bus j caused by a fault at
bus f can be calculated by the below equation:
Ifi;j ¼ ðV i  V j  FSF fi;j  V f Þ=zi;j
FSF fi;j ¼ ðzi;f  zj;f Þ=zf ;f
where Vi, Vj and Vf are the voltages at buses i, j, and f respectively.
Also za,ba, b = i, j, f are the elements of the impedance matrix. In
the normal state Zbus is constant however in the case of installing
new generations or fault current limiters the impedance matrix
should be modified [10].
Fig. 2. Thevenin equivalent when FCL is activated.
In the recent years the increasing demand for power delivery
ð4Þ has resulted to the increase in fault current levels. These levels
can exceed the capacity of circuit breakers. The exceeded fault cur-
rents caused by a fault can prevent the operation of circuit breakers
and hence lead to a permanent fault. Therefore utilizing fault cur-
ð5Þ rents in the system seems to be necessary.
Fault current limiters are variable impedance devices which
there impedance is almost neglectable in the normal operation of
the system. However, in the case of occurring a fault in the system
ð6Þ
the impedance of these devices increase to a specified amount to re-
strict the occurred fault current in the system to the allow value.
FCLs are developed based on different principles. The most general
fault current limiters are: Superconducting fault current limiters
[23,24], magnetic fault current limiters [25] and solid state fault cur-
rent limiters [26,27]. Fault current limiters can also be divided into
inductive fault current limiters and resistive fault current limiters.
One type of resistive fault current limiters is the resistive type
SCFCL which is based on transforming the superconductivity mate-
rial from its superconductivity state (which is used in the normal
state with a zero resistance) to the normal state (for a fault condi-
tion with a specified resistance) by a quenching process which is
demonstrated in Fig. 3.
Magnetic shield fault current limiters are a type of inductive
fault current limiters. When the induced current exceeds the crit-
ical current of the superconductor the transition of inductance is
triggered. The structure and performance of this kind of FCLs is
ð7Þ illustrated in Figs. 4 and 5.
ð8Þ
3. Optimal power flow formulation
The proposed algorithm is tested and compared with other
conventional algorithms on optimal performance in terms of
minimization of (a) total fuel cost of generation, (b) power losses
in transmission lines and (c) sum of voltage deviations on load
 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213 207

Fig. 3. Resistive fault current limiters under the normal and fault condition.

where x is the vector of dependent variables, [Vb] is the vector of

load bus voltages, [Qg] is the vector of generator reactive power out-
puts and [Sl] is the transmission line loadings.
The vector of decision variables is presented as below:

uT ¼ ½½V g ; ½T; ½Q c  ð13Þ

[Vg] is the vector of generator bus voltages, [T] is the vector of trans-
former taps and [Qc] is the vector of reactive compensation devices.
The equality constraints are the load flow equations as:
X
PGi  PDi ¼ V i V j ðGij cos hij þ Bij sin hij Þ ð14Þ
j2N B

X
Q Gi  Q Di ¼ V i V j ðBij cos hij  Gij sin hij Þ ð15Þ
Fig. 4. Structure of magnetic shield SFCL. j2N B

where PGi and QGi are the active and reactive powers generated at
bus i. PDi and QDi are the active and reactive loads at bus i. Vi and
buses [28–31]. The function is optimized while satisfying equality
Vj are the voltages at buses i and j. Gij and Bij are the conductivity
and inequality constraints. The first objective is to minimize the
and susceptance of the line between buses i and j. Finally hij = hi  hj
generation fuel cost:
is the difference between bus angles.
Ng
X The inequality constraints are as follows. These constraints in-
F1 ¼ ai P2Gi þ bi PGi þ ci ð9Þ clude the bus voltages and also the line limits.
i¼1

where Ng is the number of generators, ai, bi, ci are the generation

V min
bi 6 V bi 6 V max
bi i ¼ 1; 2; . . . Nb ð16Þ
coefficients and PGi is the active power generated by the ith
Sl 6 Sl;max l ¼ 1; 2; . . . Nl ð17Þ
generator.
The second objective function that is regarded is the system where Vbi is the voltage at bus i and and V min V max
show the
bi bi
power loss: minimum and maximum amount of the voltage at bus i. Nb is
Nl
X the number of buses. Also Sl is the flow of power at line l. Smax is
F 2 ¼ Ploss ¼ PL ð10Þ the maximum amount of power that can flow at line l. Nl is the
L¼1 number of lines.
The constraints for the active power Pgi and reactive power Qgi
where PL is the real power loss at line-L and Nl is the number of
generated by the generator i are as follows:
transmission lines. The third objective function is the voltage devi-
ation at load buses and can be expressed as: Pmin 6 P gi 6 Pmax i ¼ 1; 2; . . . Ng ð18Þ
gi gi
Nb
X
F3 ¼ jV i  V sp
i j ð11Þ Q min 6 Q gi 6 Q max i ¼ 1; 2; . . . Ng ð19Þ
gi gi
i¼1
where Ng is the number of generators.
where Vi is the voltage at load bus-i, which is usually set to 1.0 p.u.
Finally the reactive compensation sources at bus i, Qci and tap
and Nb is the number of load buses.
setting at line l Tl are restricted as:
In all of the problems the dependent vector is considered as:

xT ¼ ½½V b ; ½Q g ; ½Sl  ð12Þ Q min

ci 6 Q ci 6 Q max
ci i ¼ 1; 2; . . . Nb ð20Þ

Fig. 5. Inductive SFCL under normal and fault conditions.

208 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213

T min
l 6 T l 6 T max
l l ¼ 1; 2; . . . Nl
4. Optimal power flow incorporating fault current levels and
fault current limiters
In this section the optimal power flow problem is solved regard-
ing the fault current levels in the system. The fault current level
constraints are added to the OPF problem as the following:
Isc max
k;j;f 6 Ik;j
where Isck;j;f indicates the maximum fault current at line k–j for a fault
at bus f. Also Imax
k;j is the maximum permitted fault current according
to the capacity of protection devices such as circuit breakers at a
specified line. However computing all of the fault current levels is
time consuming regarding the combination of bus numbers, line
numbers and circuit breakers that are installed at the end of lines
[10]. Hence to reduce the computational burdens for large systems
initially a fault current level study is done at each line for a fault at
each bus. According to the ratings of circuit breakers which are in-
stalled at each line if the fault current level for a specific line vio-
lates the rating of installed CBs for a fault at any bus the line will
enter a fault current level list. This list specifies the lines which
there maximum fault current levels should be involved as con-
straints such as relation (22) for the OPF problem. After determining
the initial fault current level list the OPF problem is solved consid-
ering this list. After solving the OPF problem a fault current level
analysis is done on the system considering the new amounts of con-
trol variables obtained by the OPF problem. For any line if the fault
current level amount exceeds the rating of the installed circuit
breakers for that line for a fault at any bus the line enters the fault
current level list and the OPF problem is solved again. Fig. 6 illus-
trates the algorithm for making the fault current level list.
ð21Þ However as including fault current level constraints in the opti-
mization problem this leads to exacerbate the operation of the sys-
tem including the increase of total generation cost, power loss and
also voltage deviations. Therefore to prevent the replacement of
protection devices and improving the quality of power system
operation utilizing fault current limiters are suggested. Also for a
number of lines the appropriate allocation of active power is not
adequate for suppressing the fault currents to beyond the desired
levels and therefore using replacing circuit breakers or using fault
current limiters become inevitable. Fig. 7 demonstrates the men-
ð22Þ tioned two states. In case (a) the fault current for a specified line
decreases to under the desired level by a proper setting of control
variables while in case (b) using fault current limiters is necessary
and the OPF problem is not able to decrease the fault currents to
under the permitted amount and is unfeasible. Fig. 6 shows this
stage that leads to the solving the FCL optimal sizing and place-
ment problem and then the solving the conventional OPF problem.
5. FCL optimal placement and sizing formulation
In this paper the goal is to reduce the fault current at each bus
and the fault current flowing in lines to a specified amount. This
goal is obtained by the installation of adequate fault current limit-
ers in the system. However the total impedance used for the fault
current limiters should be minimized to reduce the financial costs.
Hence the problem can be formulated as following:
X
N FCL
minJ 1 ¼ Z i;FCL ð23Þ
i¼1
Z min max
i;FCL 6 Z i;FCL 6 Z i;FCL i ¼ 1; 2; . . . NFCL ð24Þ
Isc sc;max
f 6 If f ¼ 1; 2; . . . ; Nb ð25Þ

Perform fault current level analysis for

initial settings of control variables and for
each line

Enter the lines that there maximum fault

current level exceeds ratings of CBs to the
fault current level list
Add the
line to
the fault
current Perform the OPF
level list
Perform the fault current level analysis by
the new amounts of control variables
obtained from OPF

Yes
No Solve the FCL
Is the OPF problem optimal sizing and
feasible? location problem
Yes

For a new line does the fault current level

exceed the rating of circuit breakers? Solve the
conventional OPF
problem neglecting
No
fault current levels

Print the results from the OPF problem as

the final solution

Fig. 6. Flowchart of the proposed OPF considering fault current levels.

 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213 209

Decreasing
Fault current level by fault current
placing and activating level by OPF
fault current limiter

Desired fault
current level

Neglecting fault
current levels in the
OPF problem

Fault current level

Case a

Decreasing
Fault current level by Desired fault fault current
placing and activating current level level by OPF
fault current limiter

Neglecting fault
current levels in the
OPF problem

Fault current level

Case b
Fig. 7. Two states for limiting fault currents using OPF and FCL.

specified fault current of bus f. Nb is the number of buses. Finally

Table 1
Fuel cost coefficients and variable limits. Ifk;j is the fault current flowing from bus k to bus j caused by a fault
at bus f and Ispec is the maximum fault current flowing in the lines.
Generator bus number a b c Pmax Pmin
Nl is the total number of lines in the system.
30 0.01 9.5 633 1200 200
31 0.02 10.3 145 1200 200
32 0.05 12.3 615 1200 200 6. Simulations and results
33 0.01 32.3 739 1200 200
34 0.07 5.5 56 1200 200
In order to verify the proposed approach the method is imple-
35 0.12 40 220 1200 200
36 0.01 10 621 1200 200 mented to the New England 39 bus system. The system consists
37 0.04 2.5 621 1200 200 of 39 buses, 46 lines and 10 generators. Also 15 lines are under
38 0.016 8.8 100 1200 200 tap setting transformers. Therefore the OPF optimization problem
39 0.006 12.5 440 1200 200
consists of 34 control variables including 15 tap changing trans-
formers, 10 voltage magnitude of voltage buses and also 9 genera-
V min
G
V max
G V min
L
V max
L
Tmin Tmax
tion output of generators. Bus 31 is the slack bus. Table 1
Limits of tap settings and voltages for load and generator buses
demonstrates the fuel cost coefficients, limits of PV bus voltages
0.9 1.1 0.95 1.05 0.95 1.05
and tap settings of transformers.
For the system at its initial condition a fault current analysis is
performed. The results are demonstrated in Table 2. The lines
Table 2
Lines which there fault current exceeds for the initial fault current level analysis.
which their fault current exceeds 2.2 p.u. and the ratings of circuit
breakers are under 2.2 p.u. enter the fault current level list For sup-
Line 3 Line 9 Line 22 Line 46 pressing the fault current level of these lines to under the permit-
Faulted bus 37 31 34 34 ted value specified by breakers at these lines the fault current level
Fault current (p.u.) 2.8807 2.3651 2.237 2.3951 of lines in Table 2 are added as constraints to the optimal power
flow problem. In the next section The OPF problem is executed
considering these constraints.
Ifk;j 6 Ispec f ¼ 1; 2; . . . ; Nl ð26Þ
6.1. OPF incorporating fault current levels
where Zi,FCL shows the impedance of the ith fault current limiter,
NFCL is the number of installed FCLs, Z min max
i;FCL and Z i;FCL are the minimum In this section the OPF problem is solved considering three
and maximum permitted impedance of the fault current limiters. objective functions including total fuel cost, power loss and voltage
Also Isc sc;max
f and I f are the fault current at bus f and the maximum deviations of the system. Due to the economic aspects it is desired
210 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213

Table 3 Table 4
Control variables for different objective functions regarding fault current levels. Buses with fault currents more than the permitted amount for a symmetrical fault at
the specified bus.
Total fuel Power loss Voltage deviation
cost ($) (MW) (p.u.) Bus number 31 32 33 34 36 38
Bus fault current 21.8 20.53 23.07 38.14 25.02 26.61
T1 0.9790 0.95 1.05
T2 0.95 1.05 0.9687
T3 0.9893 1.05 1.05
T4 0.9503 1.05 0.9955
T5 1.05 1.0498 1.0361
T6 0.95 1.0061 1.05
levels of the specified lines are suppressed to under the rating
T7 1.0405 1.0013 1.0468 amounts of circuit breakers while minimizing the total fuel cost.
T8 1.05 1.05 1.0482 Also the results for system power loss and also voltage deviations
T9 0.95 0.9838 1.0135 are illustrated in Table 3. Fig. 8 shows a comparison between the
T10 1.05 1.05 1.027
maximum fault currents of the studied lines before and after opti-
T11 0.9715 0.9956 0.9869
T12 1.0036 1.0275 0.9818 mization. It is obvious that the fault currents are limited to under
V1 0.9211 0.9254 0.9736 the allowed amount.
V2 0.9 0.9 1.0425
V3 1.1 1.0960 1.1
6.2. FCL optimal sizing and placement
V4 1.0659 1.1 1.0982
V5 0.9 0.9003 0.9008
V6 1.1 1.1 1.0571 In this section the placement of fault current limiters are pro-
V7 1.0983 0.9694 1.1 grammed to suppress the fault currents obtained by (1) and (7)
V8 0.9233 0.9747 0.9677
at buses and lines that have critical fault currents exceeding the
V9 0.9 1.0081 1.0158
V10 1.1 1.0263 1.0096 rating of protective devices. To start the optimization procedure
P1 200 200.8270 200 initially fault currents have been calculated at each bus (assuming
P2 872.5906 737.1923 56.9804 a symmetrical fault). Also the fault currents flowing through the
P3 355.8121 709.4084 1088.5 lines caused by faults at other buses that are not permitted were
P4 846.8513 678.4576 407.2304
mentioned in the previous section. For this system the fault cur-
P5 342.8564 697.9022 313.4271
P6 200 553.5478 1186 rents generated by symmetrical faults exceed the determined limit
P7 1200 653.5292 1200 (20 p.u.) for the following buses illustrated at Table 4. Six buses ex-
P8 288.1815 200 586.8395 ist that according to the impedance of the system exceed the deter-
P9 714.0144 577.1717 774.7046
mined limit of fault currents. Also in the cases noticed in Table 2
P10 1200 1170.2 439.3149
Amount of objective 167390 49.5326 0.2879
the fault current flowing in lines is more than the specified limit.
function So the fault current limiters should be sized and placed to have
the minimum impedance (for economic concerns) in addition of
limiting the fault current to the allowed limit. The fault currents
flowing in lines caused by faults at other buses are dependent on
the voltages of the system and therefore affect the optimization
problem. But the fault currents at buses caused by symmetrical
faults only depend on the impedance matrix and do not affect
the OPF problem. For the fault currents at lines the FCL optimiza-
tion problem is solved considering the initial state of the system.
Also it is assumed that inductive fault current limiters are used
in this work.
The optimization problem is solved in three cases. In the first
case only the fault current of the buses are reduced to the allow-
able limit by placing fault current limiters. In the second case the
amount and location of fault current limiters are determined in
the way to restrict the fault current flowing in lines to the permit-
ted limit. Finally in the third case the programming of fault current
limiters is implemented to the system considering both types of
fault currents.
For the first case the results are obtained and illustrated in
Table 5 by solving the optimization problem at (23)–(25). The
Fig. 8. Fault currents before and after solving the OPF problem.
results are obtained for solving the optimization problem with
three and four fault current limiters. By using two FCLs in the
system it is not possible to satisfy the constraints of (23)–(25). It
to use the minimal number of fault currents. In other worlds the is observed that in the case of using four fault current limiters
least number of fault current limiters are desired to cover the en- the total reactance that should be installed in the system is
tire system and suppressing the fault current at any location of 0.0455 p.u. while using three fault current limiters in the system
the system to under the allowed amount. In this paper for the first requires a total 0.1404 p.u. reactance. Fig. 9 shows the bus fault
case using one or two FCLs was not adequate to suppress the fault currents for three states before and after installing three and four
currents in all of the system. Therefore for the first and second case fault current limiters in the system for buses in Table 4.
of the optimal placement and sizing problem three FCLs were cho- In the second case the fault current limiters are used to restrict
sen. However in the third case even using three FCLs is not ade- the fault current flowing in the lines to an acceptable level. The
quate and using four FCLs is proposed. results in Table 6 are acquired after solving the optimization
The results for solving the optimization problem for each of problem assuming using three and four fault current limiters
these objective functions are demonstrated at Table 3. The fault (for this case it is assumed that programming the FCLs amount
 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213 211

Table 5
Optimal location and amount of fault current limiters for limiting bus fault currents.

Using four fault current limiters Line 19 1 45 46

Reactance (p.u.) 0.0028 0.0117 0.0031 0.0279
Using three fault current limiters Line 19 12 46
Reactance (p.u.) 0.0023 0.0381 0.1

Fig. 9. Bus Fault currents before and after installing fault current limiters for case 3.

Table 6
Optimal location and amount of fault current limiters for limiting branch fault currents.

Using four fault current limiters Line 46 3 45 11

Reactance (p.u.) 0.0083 0.0095 0.00423 0.095
Using three fault current limiters Line 3 46 1
Reactance (p.u.) 0.065 0.0941 0.076

Fig. 10. Bus Fault currents before and after installing fault current limiters for case 3.

and placement is done regarding the base case of the system). In
case of using four fault current limiters the total impedance that
should be installed in the system is 0.11703 p.u., also using three
FCLs increases the total used reactance to 0.2351 p.u. Similar to
the previous case using two FCLs is not enough for the entire
system.
In the last case both type of fault currents are restricted. For this
case at least four fault current limiters should be used to decrease
the fault current at each bus to less than 20 p.u. and the fault cur-
rent flowing in lines to under 2.2 p.u. The results attained for this
case are:
 A fault current limiter with an impedance 0.0105 p.u. should
be placed at line 19.
 A fault current limiter with an impedance 0.4194 p.u. is
placed at line 1.
 A fault current limiter with an impedance 0.020 p.u. is
placed at line 41.
 A fault current limiter with an impedance 0.0034 p.u. is
placed at line 30.
Figs. 10 and 11 indicate the line and bus fault currents before
and after placing the fault current limiters.
212 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213

Fig. 11. Line fault currents before and after installing fault current limiters for case 3.

Table 7
Control variables obtained from the conventional OPF.

Total fuel Power loss Voltage deviation

cost ($) (MW) (p.u.)
T1 0.95 1.05 0.9839
T2 0.95 1.05 1.05
T3 0.95 1.0476 1.05
T4 1.05 1.0186 1.0324
T5 1.023 1.05 0.9968
T6 1.05 0.95 1.0241
T7 1.05 1.0479 0.9502
T8 0.95 1.05 1.0457
T9 0.9927 0.95 0.95
T10 1.0093 1.0116 1.05
T11 0.9588 0.95 0.9991
T12 1.05 1.0052 1.05
V1 1.0788 1.0369 0.9972
V2 1.1 1.0867 1.0997
V3 1.0467 1.0032 0.9933
V4 1.0175 1.0064 1.0137
V5 0.9504 1.1 1.0513
Fig. 12. Comparison of voltage for two cases regarding fault current levels and fault
V6 1.1 1.0274 1.1
current limiters.
V7 0.9 1.0149 0.9557
V8 1.0632 1.1 1.1
V9 1.1 1.1 0.9925
V10 1.0341 1.0336 0.9838 $/h obtained for the fault current constrained OPF the results
P1 1200 465.7705 200 show that installing fault current limiters have a salient effect
P2 622.2099 1116.2 476.9509 on decreasing the total fuel cost. The control variables are
P3 200 321.0027 918.1990
P4 200 416.5056 1200
shown in Table 7.
P5 200 534.86 200 2. Power loss When the optimization problem is solved consider-
P6 200 837.7819 200 ing the system power loss as the objective function decreases
P7 1200 281.9031 760.9882 to 26.66 MW regarding the fault current limiters in the system.
P8 398.7322 412.4604 1200
The difference between the total power losses of the system for
P9 791.4027 590.5969 307.5289
P10 1200 1200 768.1623 these two states shows the impact of FCLs on the operation of
Amount of objective 146540 26.66 0.1690 the system. The results for this state are also illustrated in
function Table 7.
3. Voltage deviation In this state the total voltage deviations
decreases to 0.1690 p.u. which shows the influence of fault
current limiters on the voltage profile. The obtained control
6.3. OPF considering fault current limiters variables by solving this optimization problem are indicated
in Table 7. Also the voltage profile is illustrated for two cases
In this case the fault current limiters are placed. Therefore the considering fault current limits and FCLs in Fig. 12.
fault current level constraints that had to be regarded in the previ-
ous section are not incorporated in the optimization problem any- 7. Conclusion
more. The results for this section are presented in Table 7. For this
state the results are as follows: In this paper a novel framework is presented for the optimal
power flow problem. The fault current limits are considered as
1. Total Fuel Cost For this state a conventional OPF problem is additional constraints in the OPF problem and the aim is to de-
solved to minimize the total fuel cost. For this state the objec- crease the fault currents in addition of other constraints of the
tive function decreases to 146540 $/h. Regarding the 167,690 OPF problem. The fault current limit constraints are defined
 A. Khazali, M. Kalantar / Electrical Power and Energy Systems 59 (2014) 204–213
according to rating of Circuit breakers in the system and if the fault
currents exceed the rating of these switching devices they are re-
garded in the OPF problem as additional constraints. The solution
of the OPF problem has to suppress the levels of fault current to be-
neath the allowed amount. However there is a probability that the
OPF problem considering fault current limits can become infeasi-
ble. In this case utilizing fault current limiters is proposed as a
solution of decreasing fault current levels. In this work a method
is presented to find the optimal placement and values of fault cur-
rent limiters and the optimal placement and sizing problem for
FCLs is solved for three different cases.
The fault current limiters are placed with the goal to restrict the
fault currents at buses and the fault currents flowing through lines
caused by faults at other buses. In this paper the optimization
problem is defined in order to minimize the used impedance while
satisfying the defined constraints. The proposed method is imple-
mented to the 39 bus IEEE system. The results show that the fault
currents are suppressed to the admissible levels using the least
impedance.
Finally the OPF problem is solved neglecting the fault current
level constraints (due to the FCL placements in the system). Com-
paring the results for two cases indicates that placing fault current
limiters in the system can have a salient impact on the economical
and secure operation of the system.
References
[1] Lee K, Park Y, Ortiz J. A united approach to optimal real and reactive power
dispatch. IEEE Trans Power Ap Syst 1985:1147–53.
[2] Momoh JA, El-Hawary ME, Adapa R. A review of selected optimal power flow
literature to 1993. I. Nonlinear and quadratic programming approaches. IEEE
Trans Power Syst 1999;14(1):96–104.
[3] Momoh JA, El-Hawary ME, Adapa R. A review of selected optimal power flow
literature to 1993. II. Newton, linear programming and interior point methods.
IEEE Trans Power Syst 1999;14(1):105–11.
[4] AlRashidi M, El-Hawary M. Applications of computational intelligence
techniques for solving the revived optimal power flow problem. Electric
Power Syst Res 2009;79(4):694–702.
[5] Abido M. Optimal power flow using particle swarm optimization. Int J Electr
Power Energy Syst 2002;24:563–71.
[6] Lai L, Ma J, Yokohoma R, Zhao M. Improved genetic algorithm for optimal
power flow under both normal and contingent operation states. Electr Power
Energy Syst 1997;19:287–91.
[7] Abou E AA, Abido MA, Spea SR. Optimal power flow using differential evolution
algorithm. Electr Power Syst Res 2010;80(7):878–85.
[8] Kumari MS, Maheswarapu S. Enhanced genetic algorithm based computation
technique for multi-objective optimal power flow solution. Int J Electr Power
Energy Syst 2010;32(6):736–42.
213
[9] Khazali A, Kalantar M. Optimal reactive power dispatch by a harmony search
algorithm. Electr Power Energy Syst 2011;33:684–92.
[10] Vovos P, Harrison G, Wallace A, Bialek J. Optimal power flow as a tool for fault
level-constrained network capacity analysis. IEEE Trans Power Syst
2005;20(2).
[11] Vovos P, Bialek J. Direct incorporation of fault level constraints in optimal
power flow as a tool for network capacity analysis. IEEE Trans Power Syst
November 2005;20(4).
[12] Neumann C. Superconducting fault current limiter (SFCL) in the medium and
high voltage grid. Power Eng Soc Gen Meet 2006(2):1423–5.
[13] Noe M, Oswald B. Technical and economical benefits of superconducting fault
current limiters in power systems. IEEE Trans Appl Supercond June
1999;9(2):1347–50.
[14] Elmitwally A. Proposed hybrid superconducting fault current limiter for
distribution systems. Int J Electr Power Energy Syst 2009;31(10).
[15] Javadi H. Fault current limiter using a series impedance combined with bus
sectionalizing circuit breaker. Int J Electr Power Energy Syst
2011;33(3):731–6.
[16] Shahriyari S, Yazdani A, Haghifam M. Cost reduction of distribution network
protection in presence of distributed generation using optimized fault current
limiter allocation. Int J Electr Power Energy Syst 2012;43(1):1453–9.
[17] Firouzi M, Gharehpetian G, Pishvaei M. A dual-functional bridge type FCL to
restore PCC voltage. Int J Electr Power Energy Syst 2013;46:49–55.
[18] Javadi H, Mousavi S, Khederzadeh M. A novel approach to increase FCL
application in preservation of over-current relays coordination in presence of
asynchronous DGs. Int J Electr Power Energy Syst 2013;44(1):810–5.
[19] Fault Current Limiter Technical Committee, Technical Report of the IEE of
Japan, vol. 709, 1999.
[20] Hongesombut K, Furusawa K, Mitani Y, Tsuji K. Allocation and circuit
parameter design of superconducting fault current limiters in loop power
system by a genetic algorithm. Trans Inst Electr Eng Jpn 2003;123(9):1054–63.
[21] Teng J, Lu C. Optimum fault current limiter placement with search space
reduction technique. IET Gener Transm Distrib 2010;4(4):485–94.
[22] Nagata M, Tanaka K, Taniguchi H. FCL location selection in large scale power
system. IEEE Trans Appl Supercond 2001;11(1):2489–94.
[23] Cointe Y, Tixador P, Villard C. FCL: a solution to fault current problems in DC
networks. J Phys: Conf Ser 2008;97.
[24] Kozak S, Janowski T. Physical and numerical models of superconducting fault
current limiters. IEEE Trans Appl Supercond 2003;13(2):2068–71.
[25] Osorio M, Cabo L, Veira J, Vidal F. Inductive fault current limiter based on
multiple superconducting rings of small diameter. Supercond Sci Technol
2004;17:98–102.
[26] Guk-Hyun M, Wi Y, Lee K, Joo S. Fault current constrained decentralized
optimal power flow incorporating superconducting fault current limiter
(SFCL). IEEE Trans Appl Supercond June 2011;21(3).
[27] Xie Y, Tekletsadik K, Hazelton D, Selvamanickam V. Second generation high-
temperature superconducting wires for fault current limiter applications. IEEE
Trans Appl Supercond 2007;17(2):1981–5.
[28] Abido M. Optimal power flow using particle swarm optimization. Int J Electr
Power Energy Syst 2002;24(October (7)):563–71.
[29] Osman M, Abo-Sinna M. A solution to the optimal power flow using genetic
algorithm. Elsevier Inc.; 2003.
[30] Sivasubramani S, Swarup K. Multi-objective harmony search algorithm for
optimal power flow problem. Electr Power Energy Syst 2011;33:745–52.
[31] Vaisakh K, Srinivas L. Evolving ant direction differential evolution for OPF with
non-smooth cost functions. Eng Appl Artif Intell 2011;24:426–36.