Fault Detection in a Series Compensated line during
Power Swing using Superimposed Apparent Power
Jitendra Kumar
Department of Electrical Engineering
Indian Institute of Technology, Roorkee
Roorkee, India
jeetusingh61@gmail.com
Abstract— In this paper, detection of fault in a series
compensated line during power swing is considered. Voltage and
current signals contain non-fundamental componens due to fault
in a series compensated line during power swing. Presence of
transients due to MOV and airgap operation, sub synchronous
resonance, voltage and current inversions further make the fault
detection process more complex. There are many conventional
schemes available in literature to indentify the fault during
power swing. During power swing, both active and reactive
powers oscillate within positive and negative regions, whereas
the apparent power oscillates within positive region only. In this
work, a superimposed apparent power based technique is
proposed to detect the fault during power swing in a series
compensated line. A 400 kV power system is simulated using
EMTDC/PSCAD software. The performance of the technique
has been considered for symmetrical fault, asymmetrical fault,
far-end fault, high resistance fault, and single pole tripping
situation during power swing in a series compensated line.
Performance of the proposed technique is satisfactory and
accurate.
Index Terms-Power swing, Series compensation, Fault detection,
Distance relay, Single-pole tripping
I. INTRODUCTION
Series compensation is maintained in a line to avail more
power flow through a line, improve the voltage profile and to
reduce the line loss [1]-[3]. However, inception of fault in a
series compensated line leads to sub-synchronous resonance,
voltage and current inversion situations [4]. Operation of
MOV and air gap further adds transients in voltage and
current waveforms [5]. Mismatch in mechanical and
electrical powers occurs due to line loss, sudden loss of loads,
and generator disconnection. This situations lead to presence
of oscillation in voltage and current signals at the relay end
[6]-[14]. Many relaying schemes are constrained during
power swing present in a series compensated line [10].
Selectivity and dependability issues of distance relay are
major concerns during power swing. Detection of
asymmetrical and symmetrical faults during power swing in a
978-1-4673-8040-9/15/$31.00 ©2015 IEEE
Premalata Jena
Department of Electrical Engineering
Indian Institute of Technology Roorkee
Roorkee, India
pjenafee@iitr.ac.in
series compensated line is challenging as the swing
modulates the voltage and current waveforms. Single-pole
tripping (SPT) is another situation due to which the power is
available in other healthy phases following to clearance of a
line-to-ground fault. But it introduces negative and zero
sequence components in voltage and current signals. Fault
detection during SPT and power swing in a series
compensated line is further more complex task.
Many techniques are available for fault detection during
power swing present in normal transmission line [7]-[15]. In
distance relay, power swing blocking (PSB) element blocks
the relay operation during power swing and maintains the
selectivity property [4]. Out of step trip (OST) element
operates for fault during power swing and maintains the
dependability property [4]. Settings of PSB and OST are
difficult in distance relays specifically for high resistance
fault, weak source and far end fault [7]. In [8], decaying DC
component extracted with Prony method is used to
distinguish the symmetrical fault present during power swing.
In [9], a differential power based technique is considered
using the actual and predicted samples of voltage and current
to detect the symmetrical fault only during power swing,
where an autoregression technique is used to predict the
future fault samples. The frequency component present in
three-phase instantaneous power is utilized to detect the
symmetrical fault during power swing [10]. In [11], low
resistance fault is detected using the angle between the relay
end voltage and the line voltage drop up to the fault point.
This technique fails for high resistance fault. Another
technique is proposed to detect the high resistance fault
during power swing and SPT by comparing the angles of
voltage at the relay end, estimated line voltage drop, and zone
termination voltage [12]. A mathematical morphological
technique is used in [13] to detect the symmetrical faults
during power swing.
Transients present in fault reactance during power swing are
calculated by using two parallel FIR filters and used as an
indication of fault during power swing condition [14]. In [15],
centre of circular locus of the admittance trajectory is
exploited to distinguish the fault from power swing. Phase with swing is increased and oscillates within positive region
angle of superimposed positive sequence current based only.
technique is used to detect the fault during power swing and 200

the technique fails during current and voltage inversion cases. v (V ) 0

Rate of change of resistance is used to detect the fault during -200
(a)
power swing [16]. In [17]-[18], negative sequence component 2.3
20
2.4 2.5 2.6 2.7 2.8 2.9 3

of current is used to detect fault during power swing in a

i (A)
series compensated line. CUSUM of negative sequence
0

current is taken to detect the fault. (b)

-20
2.3 2.4 2.5 2.6 2.7 2.8 2.9 3

In this paper, a superimposed apparent power based Time(s)

technique is proposed to detect the asymmetrical and Fig. 2. Voltage and current waveforms during power swing. (a) The three-
phase voltage waveforms. (b) The three-phase current waveforms.
symmetrical faults present in a series compensated line 300
during power swing. The difference between the pre and 200
Total apparent Power

during fault apparent power is considered for detection 100

process. The performance of the technique is tested for

S(VA),
P(W), 0

voltage and current inversion cases. Other situations like high Q (VAr)
-100

resistance fault, close-in fault, and single pole tripping cases -200
Total Real Power
are also considered. The performance of the proposed
Total Reactive Power
-300
2.7 2.8 2.9 3 3.1 3.2 3.3

technique is found to be accurate. Time(s)

Fig. 3. The Apparent power (S), Real power P (W), Reactive power Q (VAr)
II. APPARENT POWER DURING POWER SWING AND SPT IN at the relay R located at bus A during power swing in line-1.
A SERIES COMPENSATED LINE 1500
Apparent power during fault

A. Apparent Power during Power Swing in a Series 1000

Compensated Line S (VA)

500
A 400 kV, 50 Hz system as shown in Fig. 1 is considered Pre-fault apparent power

for evaluating the performance of the proposed technique. 0

The detail system data are provided in Appendix A. In Fig. 1,

2.5 3 3.5
Time(s)

line-1 and line-2 are 70% compensated. Relay located at bus-
A is considered for evaluating the performance of the
technique. A three-phase fault is created in line-2 at 1.3 s.
After 0.2 s the breakers B3 and B4 are opened simultaneously
to clear the fault. As a result swing is observed in voltage and
current signals of relay (protecting line-1) located at bus-A.
The voltage and current waveforms are shown in Fig. 2. The
voltage and current samples are stored at a sampling rate of 1
kHz. The voltage and current phasors are estimated using one
cycle least square technique. The apparent power at this
situation is calculated using the expression provided in (1).
S = Va I a + Vb I b + Vc I c (1)
The real, reactive, and apparent powers a-re shown in Fig.
3. It is observed that the real and reactive powers are
modulated with other frequency components and the sample
values are oscillating within both positive and negative
regions. At the same time, the apparent power is also
oscillating but the magnitude varies within the positive region
only. By exploiting the pattern of apparent power, both
asymmetrical and symmetrical faults can be detected during
power swing present in a series compensated line.
A B1 B2
Line1
R 2000
Fig. 4. The pre-fault and during fault apparent power (S) at the relay R.
B. Apparent Power during SPT and Power Swing in a Series
Compensated Line
Line to ground fault of ag-type is created in line 1 at 1 s
and the duration of fault is 0.1 s. The phase-a pole of B1 and
B2 are opened at 1.1 s simultaneously. After 1.1 s, the SPT
condition has been maintained. The swing is present in relay
(R) voltage and current by creating the conditions as
explained in section II and subsection-A. Under such
situation, the apparent power is estimated using (1) and
shown in Fig. 5. It is observed that the apparent power during
power swing is less in comparison to apparent power
estimated during power swing and SPT condition. It is
because there is a phase change in voltage and current during
SPT with swing in comparison to swing situation only. As a
result, the apparent power is greater in case of SPT and swing
situation than only swing situation. During SPT and swing,
the apparent power oscillates within positive region only.
This feature can be utilized in detecting fault during power
swing with SPT condition in a series compensated line.
B
2500
Apparent Power during power swing
Apparent Power during SPT power swing

Line 2
Generator Transformer S(VA) 1500
B3 B4 1000
Fig. 1. The power system (400 kV, 50 Hz).
500
A line-to-ground fault of ag-type is created in line-1 during
swing and the apparent power is estimated using the
0
2.7 2.8 2.9 3 3.1 3.2 3.3
Time(s)
expression provided in (1). The pre and during fault apparent
Fig. 5. Apparent power (S) during power swing and SPT condition at the
powers are shown in Fig. 4. The apparent power during fault relay R.
 III.PROPOSED SCHEME FOR FAULT DETECTION DURING merged with the fundamental one. By increasing power angle
POWER SWING IN A SERIES COMPENSATED LINE between the generator and the infinite bus (approximately
Voltage and current data are stored at the relay location 400), the fast swing is achieved. Different case studies are
and each phase voltage and current phasors are estimated provided in the following subsections to evaluate the
using one cycle least square technique. The apparent powers performance of the technique.
estimated during power swing and fault (swing present) are A. Single Line to Ground Fault
provided in (2) and (3). The system as shown in Fig. 1 is considered for evaluating
S pre = Vapre I apre + Vbpre I bpre + Vcpre I cpre (2) the performance of the proposed technique. Swing is present
Sfault = Vasf I asf + Vbsf I bsf + Vcsf I csf (3) in voltage and current measured at relay located at bus-A
protecting line-1. At this situation, single-line to ground fault
where, Vapre ,Vbpre ,and Vcpre are the one cycle voltage phasors
of ag-type is created at t=3 s at 100 km from the relay
of phase-a, phase-b, and phase-c respectively before fault. location with a fault resistance of 20 Ω. The voltage and
I apre , I bpre ,and I cpre are the one cycle current phasors of current signals are stored at a sampling frequency of 1 kHz
phase-a, phase-b, and phase-c respectively before fault. and phasors are estimated using one cycle least square
Vaf ,Vbf ,and Vcf are one cycle voltage phasors of phase-a, technique. The pre and fault apparent powers are estimated
phase-b, and phase-c respectively during fault. using the expressions provided in (2) and (3). The
I af , I bf ,and I cf are the one cycle current phasors of phase-a, superimposed apparent power is estimated using (4). The
phase-b, and phase-c respectively during fault. superimposed power (S), index (J), and trip signal are shown
Magnitude of superimposed apparent power is defined as in Fig. 6. It is found that the superimposed apparent power
(4) oscillates within positive region but at certain time, the
ΔS = Sfault − Spre
magnitude may touch the zero value. That is why, a
where, Sfault is apparent power during fault condition and Spre summation index (J) is used to identify the fault during power
is pre-fault apparent power. swing. The corresponding trip signal is generated within 2-3
Summation of superimposed apparent power over a period ms. The dynamic plot of trip signal shows the correctness and
(summation of successive samples of superimposed apparent consistency of the proposed technique.
power) is provided in the following expression.
p
Jk = ∑ [( J
k=1
k-1 + ΔSk ) / k ] (5)
ΔS(VA)
1500
1000
500
where, Jk-1 is the previous value of J index and it is initialized (a)
0
2.7 2.8 2.9 3 3.1 3.2 3.3
with zero. ‘p’ is total number of samples of apparent power 80
60
and 'k' is the instant at which the apparent power magnitude Index 40
20
available . (b) 0
2.7 2.8 2.9 3 3.1 3.2 3.3
Fault detection criterion is given as Trip
1

J ≥g/k (6) Signal 0.5

0
In the above expression as provided in (6), ‘g/k’ is the (c) 2.7 2.8 2.9 3 3.1 3.2 3.3
threshold value for fault detection, where ‘g’ is considered as Time(s)

0.05. Another factor ‘k’ is used for scaling down the values of
‘J’ and ‘g’ all together. The reason of scaling down the value
of ‘g’ is to match its value with J. Initially, superimposed
apparent power and J are zero up to the fault inception point.
When J crosses the threshold the value, then the fault is going
to be registered. This index provides the trip signal. Status of
trip signal is '1' indicates occurrence of fault and '0' indicates
no fault situation.
IV. PERFORMANCE TESTING OF THE PROPOSED SCHEME
FOR FAULT DETECTION DURING POWER SWING
Performance of the proposed technique is tested for
different types of balanced and unbalanced faults during
power swing condition. The 400 kV system as shown in Fig.
1 is simulated using PSCAD/EMTDC software. Slow swing
is obtained by reducing the power angle between the
generator and the infinite bus. During slow swing, 1-2 Hz
frequency component is present along with the fundamental
and in case of fast swing 5-8 Hz frequency component is
Fig. 6. Performance of the proposed scheme for ag- type fault during power
swing. (a) Superimposed apparent power (∆S) (b) Index (J). (c) Trip signal.
B. High Resistance Fault
In order to test the performance of the algorithm, ag-type
fault with fault resistance of 100 Ω is created at t=3 s in line 1
(with slip frequency = 5 Hz and δ = 1000) at 100 km from
relay location. In case of high resistance fault, current
magnitude is reduced and it becomes difficult to detect the
fault using current magnitude only. In most of the cases,
MOV does not operate for high resistance fault. At this
situation, summation of superimposed apparent power
calculated from (4), index (J), and trip signal are estimated
and shown in Fig. 7. The trip signal is ‘0’ for no fault
condition and ‘1’ for fault condition. The trip signal is
consistent throughout and detects the fault clearly.
 1500 2000
1000 ΔS(VA) 1000
ΔS(VA)
500 0
0 (a) 2.7 2.8 2.9 3 3.1 3.2 3.3
(a) 2.7 2.8 2.9 3 3.1 3.2 3.3 80
60 60
Index 40
40 20
Index 20 0
(b)
2.7 2.8 2.9 3 3.1 3.2 3.3
0
(b) 2.7 2.8 2.9 3 3.1 3.2 3.3 1
Trip
1 Signal
0.5
Trip 0
0.5 (c) 2.7 2.8 2.9 3 3.1 3.2 3.3
Signal
0 Time(s)
(c) 2.7 2.8 2.9 3 3.1 3.2 3.3 Fig. 9. Performance of the proposed scheme during power swing for an ag-
Time(s)
type fault (fault distance = 2km, RF =20 Ω). (a) ∆S (b) Index, g (c) Trip
Fig. 7. Performance of the proposed technique during power swing for ag- signal.
fault (RF =100 Ω). (a) ∆S (b) Index (J) (c) Trip signal.
E. Three-phase Fault
C. Far-end Fault
Power swing and three phase fault both are symmetrical in
During power swing with slip frequency of 5Hz and
nature. It is always challenging to distinguish the three-phase
δ=1000, an ag-type fault is created in line-1 at 3 s with fault
fault during power swing. Many methods are proposed to
distance of 278 km and RF =100 Ω. For Far end fault along
detect the three-phase fault during power swing and every
with high resistance fault condition, current magnitude
method has its own limitations. A three phase fault is created
further reduces. So, current based scheme may fail to detect
at a distance of 100 km from the relay location with slip
the fault in such a situation. Reduction of phase current in
frequency=5Hz and δ=1000. With help of the proposed
such condition is a reason of the apparent power reduction,
technique, the '∆S ' is calculated and it is shown in Fig 10(a).
which is shown in Fig. 8. But proposed technique has
The value of index increases above the threshold as shown in
sufficient value of apparent power and successfully able to
Fig. 10(b). It is observed that the trip signal is '1' throughout
detect the fault during power swing condition for series
and the fault is detected properly.
compensated line with far end high resistance fault.
Superimposed apparent power, index, and corresponding trip ΔS(VA)
1000
500
signal are shown in Fig. 8. The trip signal is ‘1’ and correctly 0
(a)
identifies the fault. 2.7
20
2.8 2.9 3 3.1 3.2 3.3

Index 10
400
ΔS(VA) 0
200 (b)
2.7 2.8 2.9 3 3.1 3.2 3.3
0
(a) 2.7 2.8 2.9 3 3.1 3.2 3.3 1
Trip
30 0.5
20 Signal
Index 0
10 (c) 2.7 2.8 2.9 3 3.1 3.2 3.3
0 Time(s)
(b)
2.7 2.8 2.9 3 3.1 3.2 3.3
Trip
1 Fig. 10. Performance of the proposed scheme during power swing for three-
Signal 0.5 phase fault (fault distance = 100 km). (a) ∆S (b) Index (c) Trip signal.
0
(c) 2.7 2.8 2.9 3 3.1 3.2 3.3 F. For Single-pole Tripping Situation
Time(s)

Fig. 8. Performance of the proposed scheme during power swing for ag- type The performance of the proposed technique has been
far end fault (fault distance = 278 km, RF =100 Ω). (a) ∆S (b) Index, g (c) tested for SPT condition. The SPT is maintained in line-1 by
Trip signal. opening the phase-a pole of breakers B1 and B2
D. Close-in Fault simultaneously at t=1.1 s after the occurrence of an ag-fault in
Three-phase close-in fault leads to collapse in voltage line-1 at t = 1 s. During SPT and power swing, bg-type fault
measured at the relay location. Such a situation also has been created in line-1 at t= 3 s. Superimposed apparent
introduces subsidence transient in voltage signal. At the same power, index and trip signal are shown in Fig. 11. The trip
time, current magnitude increases significantly and MOV signal is '1' and it declares the fault correctly.
operates as the energy level crosses the threshold value. Due 2000
ΔS(VA)
to the presence of transients, voltage based schemes may not 1000
0
be able to detect the fault correctly. To test the performance (a)
2.7 2.8 2.9 3 3.1 3.2 3.3

of the proposed technique, an ag-type fault is created in line- Index 40

20
1, at a distance of 2 km with Rf=20 Ω during power swing (b) 0
condition (with slip frequency=5 Hz and δ=1000). During 1
2.4 2.6 2.8 3 3.2

close-in fault condition, relay finds more fault current. Trip

Signal
0.5
Superimposed apparent power calculated from (4), index and (c) 0
2.4 2.6 2.8 3 3.2
trip signal for close-in fault case are shown in Fig. 9. The trip Time(s)

signal is 1 and it declares a fault during power swing Fig. 11. Performance of the proposed scheme during power swing and SPT.
correctly. .
 Performance of the proposed technique is compared with Positive sequence reactance=0.88Ω/km
conventional techniques. The conventional techniques such as Positive sequence capacitive reactance=0.487MΩ.km
variation in resistance, variation in swing centre voltage Zero sequence resistance=0.309 Ω/km
(SCV), and variation in the positive sequence superimposed Zero sequence reactance=1.297 Ω/km
current are considered for comparison purpose. For Zero sequence capacitive reactance=0.41934MΩ.km
performance evaluation, a line-to ground fault of ag-type is
REFERENCES
created in line-1 at a distance of 100 km from the relay
[1]. B. Kasztenny, “Distance protection of series-compensated lines:
location with a fault resistance of 100 Ω at t= 3 s. The values problems and solutions,” 28th Annu. West. Prot. Relay Conf., no. GER-
of resistance, SCV, and superimposed positive sequence 3998, pp. 1–34, 2001.
current are calculated and shown in Fig. 12. It is observed [2]. H. J. Altuve, J. B. Mooney, and G. E. Alexander, “Advances in series-
compensated line protection,” in 2009 62nd Annual Conference for
that all the three conventional techniques fail for detecting the Protective Relay Engineers, pp. 263–275, 2009.
fault during power swing, whereas the proposed technique [3]. J. Mahseredjianl, P. J. Lagace, S. Lefebvre and A. chartrend,
discriminates the fault correctly. “Superposition technique for MOV-protected series capacitors in short-
circuit calculations,” IEEE Trans. Energy Convers., vol. 10, no. 3, pp.
2000 1394–1400, 1995.
R 0 [4]. P. Jena, and A. K. Pradhan, “A positive-sequence directional relaying
(ohm)
(a)
-2000
2.7 2.8 2.9 3 3.1 3.2 3.3
algorithm for series-compensated line,” IEEE Trans. Power Deliv., vol.
200 25, no. 4, pp. 2288–2298, 2010.
SCV 0 [5]. R. J. Marttila, “Performance of distance relay mho elements on mov-
(b) -2002.7
protected series-compensated transmissions,” IEEE Trans. Power
20
2.8 2.9 3 3.1 3.2 3.3
Deliv., vol. 7, no. 3, pp. 1167–1178, 1992.
ΔI1 10
[6]. Power Syst. Relay Committee, “Power swing and out of step
0
considerations on transmission lines,” IEEE PSRC WG D6, pp. 1–59,
(c) 2.7 2.8 2.9 3 3.1 3.2 3.3 2005.
60
40 [7]. J. Mooney, P. E. and N. Fischer, “Application guidelines for power
Index 20
0
swing detection on transmission systems,” in 59th Annual Conference
(d) 2.7 2.8 2.9 3 3.1 3.2 3.3 for Protective Relay Engineers, pp. 289–298, 2006.
Time (s) [8]. S. Lotfifard, J. Faiz, and M. Kezunovic, “Detection of symmetrical
Fig. 12. Comparison of performance of different conventional techniques faults by distance relays during power swings,” IEEE Trans. Power
with proposed technique for ag-type fault during power swing (a) R seen by Deliv., vol. 25, no. 1, pp. 81–87, 2010.
relay (b) SCV (c) ΔI1 (b) Index (J) of the proposed technique. [9]. J. G. Rao, and A. K. Pradhan, “Differential power-based symmetrical
fault detection during power swing,” IEEE Trans. Power Deliv., vol.
27, no. 3, pp. 1557–1564, 2012.
V. CONCLUSION [10]. B. Mahamedi, and J. G. Zhu, “A novel approach to detect symmetrical
In this paper, an alternate fault detection technique for a faults occurring during power swings by using frequency components
of instantaneous three phase active power,” vol. 27, no. 3, pp. 1368–
series compensated line during power swing with and without 1376, 2012.
SPT condition is proposed. The summation of superimposed [11]. A. Machraoui and D. W. P. Thomas, “A new blocking principle with
apparent power over a period is used to identify the fault phase and earth fault protection during fast power swings for distance
protection,” IEEE Trans. on Power Deliv., vol. 10, no. 3, pp. 1242–
during power swing and SPT situations. The proposed 1248, 1995.
technique clearly detects both symmetrical and asymmetrical [12]. A. Machraoui and D. W. P. Thomas, “A new principle for high
faults during power swing and SPT situations. The technique resistance earth fault detection during fast power swings for distance
has been evaluated for different test cases like close-in fault, protection,” IEEE Trans. on Power Deliv., vol. 12, no. 4, pp. 1452–
1457, 1997.
high resistance fault and far end fault and the results are [13]. S. Gautam and S. M. Brahma, “Out-of-step blocking function in
satisfactory. A comparative study with the existing distance relay using mathematical morphology,” IEE Procc Gener.
techniques has been carried out and it found that the proposed Transm. Distrib., vol. 6, no. 4, pp. 313–319, 2012.
[14]. P. J. Moore, and A. T. Johns, “New method of power swing blocking
technique clearly detect the faults during power swing. for digital distance protection,” IEE Procc Genev. Tvansm. Distrib.,
APPENDIX A vol. 143, no. 1, pp. 19–26, 1996.
[15]. R. Jafari, N. Moaddabi, M. E. Nasab, G. B. Gharehpetian, and M. S.
System data Naderi, “A novel power swing detection scheme independent of the
Generator:600MVA, 22kV, 50Hz, 4.4MW/MVA rate of change of power system parameters,” IEEE Trans. Power
Deliv., vol. 29, no. 3, pp. 1192–1202, 2014.
X d = 1.81p.u. , X d' = 0.3p.u. , X d" = 0.23p.u. , Td0' = 8s , [16]. A. P. Apostolov, D. Tholomier, and S. H. Richards, “Superimposed
components based sub-cycle protection of transmission lines,” in Proc.
T = 0.03s , X q = 1.76p.u. , X = 1.76p.u. , T = 0.03s
''
d0
''
q
''
q0 IEEE Power Eng. Soc. Power Syst. Conf. Expo, vol. 1, pp.592–597,
2004.
Potier reactance ( X p )=0.15p.u. [17]. P. K. Nayak, A. K. Pradhan, and P. Bajpai, “A fault detection
Transformer: 600MVA, 22/400kV, 50 Hz, ∆/Y, X=0.163p.u. technique for the series compensated line during Power Swing,” IEEE
Trans. Power Deliv., vol. 28, no. 2, pp. 714–722, 2013.
X core = 0.33p.u. , Copper losses=0.00177p.u. [18]. P. K. Nayak, A. K. Pradhan, and P. Bajpai, “Detecting fault during
Transmission line: power swing for a series compensated line,” IEEE Energy, Autom.
Length of line-1 and line-2 are of 280km Signal (ICEAS), 2011 Int. Conf., pp. 1–6, 2011.
Data of each line
Positive sequence resistance=0.12 Ω/km
 corresponding results are depicted in Fig. R1. It is clear that
the index clearly detects the fault.
Thanks for your valuable comments.
Reviewer 1:
800
Comments: 600
ΔS (VA)400
1. Explain what is the criterion for choosing "g", in the paper 200
is 0.05.? 0
(a) 2.7 2.8 2.9 3 3.1 3.2 3.3
Ans: The pre-fault superimposed power is zero. In this work, 40
Index 20
‘g’ is taken as 0.05 to maintain the threshold as close to the
prefault superimposed power. (b)
0
2.7 2.8 2.9 3 3.1 3.2 3.3
2. For all the subsection ( A,B,...) remove the spaces at 1
the titles Trip Signal 0.5
Ans: It has been taken care. 0
(c)
2.7 2.8 2.9 3 3.1 3.2 3.3
3. At F subsection, Tripping situation, change SVC to Time (s)
SCV
Ans: It has been taken care. Fig. R2. Performance of the proposed scheme for uncompensated line during
power swing for ag- type high resistance fault (fault distance = 100 km, RF
=120 Ω) (a) ∆S (b) Index g (c) Trip signal.
Reviewer 2:

Comments: there are minor grammatical errors, mostly in

the introduction, but they do not deter the reader from
understanding the authors thouhgts. there is an error in
Section II F "Tripping Situation". the third line refers to
breakers B2 and B3, i beleive it should refer to B1 and B2.
On page 5 (SVC) should be SCV).

Ans: All the points raised by the reviewer are taken care. In
the manuscript, breaker names should be B3 and B4 instead
of B2 and B3. It has been taken care in the submitted
manuscript.

Reviewer 3:
Comments: The paper describes a new method to detect
all types of faults during power swings on series compensated
lines. The method is fast and dependable, even during single
pole tripping. It is less clear that the method is also immune
to detecting and operating for external faults during swings;
perhaps this might be handled by scheme logic? It would
also be interesting to know how the method would apply to
non series compensated lines, though this part of the subject
seems to be excluded by the paper title, and may be hard to fit
into the 5-page limit. But that means there can still be
interesting developments at a later time.
While the language is generally good, there are several areas
where non-standard English is used.

Ans: The proposed technique will also detect the external

fault as the relay located at bus A will see the fault current
contributed by source-2. But the fault direction estimation
logic will discriminate the direction of fault. Performance of
the proposed technique is also tested for normal transmission
line ( without series compensation). To verify the case, an ag-
type fault is created in line-1 at a distance of 120 km from the
relay location with a fault resistance of 100 ohm and the