Electrical Power and Energy Systems 103 (2018) 347–359

Contents lists available at ScienceDirect

Electrical Power and Energy Systems

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

Controlled switching technique for minimization of switching surge during T

energization of uncompensated and shunt compensated transmission lines
for circuit breakers having pre-insertion resistors
⁎
Kunal A. Bhatta, Bhavesh R. Bhaljaa, , Urmil Parikhb
a
Department of Electrical Engineering, Indian Institute of Technology Roorkee, Roorkee 247667, India
b
Power Products High Voltage Services (PPHV-S), ABB Ltd, Maneja, Vadodara 390010, India

A R T I C LE I N FO A B S T R A C T

Keywords: Controlled Switching Device (CSD) is considered to be the replacement of Pre-Insertion Resistor (PIR). Moreover,
Controlled switching methodology tie Circuit Breaker (CB) used in One-and-a-Half-Breaker (OHB) scheme as well as in line connected with non-
One-and-half breaker scheme switchable reactor scheme are employed with PIR and CSD. In this paper, a new methodology has been proposed
Pre-insertion resistor using CSD with PIR-CB for further mitigation of the switching surge during energization/re-energization of Un-
Mechanical scatter of circuit breaker
compensated Transmission Line (UCTL) and Shunt reactor compensated Transmission Line (SCTL). Unlike the
Shunt compensated transmission line
existing methodology (both for UCTL/SCTL) which requires line side voltage signatures, the proposed metho-
Switching surges
dology does not necessitate the same for SCTL and requires only the polarity of line side voltages for UCTL. The
performance of the proposed methodology has been evaluated by generating diversified cases on 400 kV UCTL
and SCTL using PSCAD/EMTDC software package. The proposed switching strategy is able to mitigate the
switching surge of the order of 33% for UCTL and 20% for SCTL. Comparative evaluation of the proposed
methodology with usual PIR methodology clearly shows its effectiveness with reference to statistical variations
of CB. The proposed methodology has been validated during effects of various parameters such as weather
conditions, insulation resistance, change in closing phase sequence, and Surge Arrestor (SA).

1. Introduction insertion instant of PIR, the voltage across the PIR at removal instant,
MIT, EIT, surge impedance of the line ZC and value of PIR along with its
In case of Extra High Voltage (EHV) and Ultra High Voltage (UHV) energy dissipation level [5–9].
transmission line having Source to line Impedance Ratio (SIR) less than In last decade, controlled switching is becoming popular in lieu of
0.5, switching surge has gained prime importance [1]. The random PIR to limit switching overvoltages on transmission lines. This is
switching of UCTL without PIR and SA may create switching surge up to achieved by precisely controlling the switching instant with reference
3 p.u. [2,3]. This may even lead to reduction in lifespan of power to the gap voltage across individual poles of CB [10,11]. The ad-
system equipment in extreme circumstances. In worst case, it can not vantages of the application of controlled switching strategy for line
only create external flashover on lines but also produce internal flash- reactors, shunt capacitors and transformers are mentioned in [11,12]. It
over on the equipment which further leads to insulation failure of the is reported in the literature that the energization of transmission line
same. Normally, in order to mitigate switching overvoltages, the con- should be carried out at zero voltage across the individual poles of CB
ventional CBs are equipped with PIR. The switching surge is mitigated which is the difference of line side voltage and source side voltage.
by inserting PIR in series with the line impedance for a few milliseconds Moreover, researchers have also proposed that the energization of shunt
at the time of energization of the transmission line. The duration up to capacitor is performed at bus voltage zero for each phase. This is ne-
which PIR is mechanically inserted in series with the line is known as cessary in order to reduce the magnitude of capacitive charging inrush
“Mechanical Insertion Time” (MIT). It is different from “Electrical In- current. Conversely, the energization of shunt reactor/transformer is
sertion Time” (EIT), which is the duration up to which PIR is elec- carried out around gap voltage peak for individual poles of CB [13,14].
trically present in a circuit [4]. The sudden removal of PIR (after MIT) However, the prime limitation of the strategy used for energization of
from the circuit initiates high voltage switching surge known as the the transformer is that it is essential to memorize the last pole of the CB
secondary surge. The magnitude of this secondary surge depends on to be opened. This is due to the fact that the last pole of the CB should

⁎
Corresponding author.
E-mail address: [email protected] (B.R. Bhalja).

https://doi.org/10.1016/j.ijepes.2018.06.024
Received 13 February 2018; Received in revised form 24 April 2018; Accepted 6 June 2018
Available online 15 June 2018
0142-0615/ © 2018 Elsevier Ltd. All rights reserved.
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Fig. 1. Single line diagram of (a) UCTL and (b) SCTL (reactors are connected at both the ends).

be close first. In addition, the said papers have not considered the effect mitigation results compare to implementation on non-PIR CB [21,22].
of a statistical and systematic deviation in making targets due to scatter Therefore, this paper explores the usage of CSD to further mitigate
in operating time of CBs [15]. switching surge during no-load energization of the transmission line
Very few research articles have been reported on controlled en- with PIR-CB. The proposed controlled switching methodology effec-
ergization of uncompensated and shunt compensated transmission tively mitigates switching overvoltages on UCTL and SCTL having CB
lines. Froehlich et al. [16,17] proposed a controlled switching strategy with PIR without utilizing line side voltage measurement.
for energization of shunt compensated transmission lines. However, the
aforementioned methodologies may generate high level of switching
2. Switching of UCTL and SCTL
surge during re-energization of SCTL particularly when the variation in
mechanical operating time scatter of CB is considered. Afterwards, the
Generally, energization of fully discharged transmission lines and
reclosing strategy for shunt compensated transmission line with varying
re-energization (auto recloser) of transmission lines establish switching
compensation level has been presented by Dantas and his co-workers
surges. In this section, various factors, responsible for the production of
[18]. However, the aforesaid methods require accurate line side voltage
a high level of switching surges along with basic terminology related to
measurement. Thereafter, Parikh et al. [19] proposed an algorithm for
controlled switching in conjunction with PIR, has been explained.
controlled fault interruption of series compensated line using Support
Vector Regression (SVR). Though heuristic search method provides
better accuracy, selection of parameters of SVR, training methodology 2.1. Factors responsible for generation of high level of switching surges
and capability to detect unseen/missing zero patterns are several
drawbacks of the said scheme. Different conditions which result into a high level of switching surge
All the researchers mentioned above worked on the reduction of are as follows:
switching overvoltage either using CSD or employing CB with PIR.
However, none of them have considered the utilization of a hybrid Case 1. Energization of fully discharged UCTL/SCTL.
combination of PIR-CB with CSD to limit the switching over voltage. Case 2. Re-energization of partially discharged UCTL/SCTL.
Furthermore, in order to improve reliability and redundancy of the Case 3. Energization of UCTL/SCTL during out of phase condition∗.
substation, most of the EHV & UHV substations are equipped with OHB Case 4. Closing/reclosing of UCTL/SCTL on terminal fault∗.
schemes [20]. Consequently, the tie CB for OHB scheme is provided
with PIR and CSD. This is due to the fact that in OHB scheme, one of the (∗indicates the extreme type test duties which are considered for
bay is connected to the transmission line and the other bay on the same designing of PIR.)
circuit may be connected to any equipment other than the line which Aforesaid four cases will take care of almost all extreme operating
requires controlled switching. Usually, the controlled energization of conditions for which PIR is supposed to mitigate the switching over-
the line is performed near gap voltage zero to mitigate the overvoltage voltages [4]. Fig. 1(a) and (b) show the single line diagram of UCTL and
below an acceptable level. Subsequently, CBs also have setbacks such as SCTL (with shunt reactor connected at both ends) used for the simu-
scatter in operating time, variation in Rate of Decay of Di-electric lation study. Further, in case of SCTL (shunt reactor connected at one
Strength (RDDS) and aging effect. In addition, external parameters like end of the line (source side)), the reactor (XL2) from Fig. 1(b) is re-
ambient temperature, dc voltage etc. also affect the operating time of moved and based on compensation level; the magnitude of ‘XL1’ has
CBs. Hence, the targeting at gap voltage zero to a very precise level been calculated. The further details of these models are given in Section
becomes challenging [14]. Furthermore, it is very difficult to accurately 3. In this paper, the level of switching surge is measured at remote end
replicate the line side voltages during transient period of line switching of the line (Bus-2).
for which the response of instrument transformers becomes sluggish. For energization of fully discharged UCTL/SCTL (Case-1), the
This happens due to the presence of dc offset in the fault current and normal controlled switching practice is to energize the line at source
non-power frequency components in the gap voltage wave. side voltage zero. However, for remaining cases, the same philosophy
As discussed previously, in case of OHB scheme and line with non- cannot be applied. In context to re-energization of partially discharged
switchable reactors, controlled switching may be employed with PIR- line (Case-2), the conditions such as no-fault, short time temporary fault
CBs. The application of controlled switching with PIR-CB has been re- (fault sustains for very short duration compare to protection dead time
cently applied for switching of capacitor bank and transformer The field of 300 ms), long time temporary fault (fault clears just prior to the end
implementation of this has also demonstrated further improvement in of protection dead time) and permanent fault resulting into lockout are
considered. In case of UCTL, exponentially decaying line side voltages

348
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

(trapped charge) are observed because of energy stored in the stray settings for conditions (i) and (ii) [29,30].
capacitance of the line. On the other hand, for SCTL, the presence of
shunt reactor connected either at one end or at both ends of the line 2.3. Behaviour of PIR during energization of transmission lines
creates L-C oscillations with a frequency less than the fundamental
frequency. In practice, either magnetic Potential Transformer (PT) or PIR consists of a resistor stack and two switches. The full assembly
Capacitive Coupling Voltage Transformer (CCVT) is used for measure- together is termed as PIR throughout in this paper. Depending upon the
ment of line side voltages. However, it is reported in the literature that position of bypass switch with respect to main contacts of CB (inter-
PTs are rarely used in practice for voltages more than 100 kV due to rupter), two types of PIR configurations, namely series PIR and parallel
techno-economic constraints [23]. At the same time, it has been men- PIR are commonly used. However, the series PIR configuration is
tioned in IEC-61869-3 that the accuracy offered by PT is maintained up usually employed due to low cost and less space requirement [4].
to optimal level if the variation in voltages is of the order of 80%–120% In this paper, series PIR configuration is used to perform all simu-
and 25%–100% for measuring class and protective class PT, respec- lation cases. Subsequently, the proposed methodology will be equally
tively. As the oscillating line side voltage crosses the above-mentioned applicable to parallel PIR configuration as the electrical behavior for
limit (usually attains a peak value more than 1.2 p.u), it will be difficult both configurations remains the same. It has been reported in the lit-
to accurately replicate the oscillating line side voltages using PT [24]. erature that the usage of PIR suppresses the switching overvoltage by a
During the protection dead time, the magnetic circuit of PT gets satu- factor of R/(R + ZC) at the remote end of the line [5]. Here, ‘R’ in-
rated due to exponentially decaying line side voltages [25]. Hence, the dicates value of PIR and ‘Zc’ represents surge impedance of the trans-
PT does not replicate the line side voltage accurately. On the other mission line.
hand, in case of SCTL, it is troublesome to replicate the line side vol- For series PIR configuration, the electrical making occurs primarily
tages by PT due to the presence of non-power frequency components. In between interrupter followed by the contacts of PIR switch which by-
case of CCVT, the trapped charge will not decay rapidly due to large passes the resistor after MIT is over. Subsequently, EIT is defined as the
capacitance offered by it. The decaying rate of trapped charge also duration from the instant when prestrike happens between interrupter
depends upon the configuration of line, type of switching duty and till the PIR switch electrically closes. Normally, PIR is designed based
reclosing on permanent/temporary faults. Furthermore, in case of on MIT. However, depending upon the magnitude of gap voltages and
single or double line-to-ground faults, mutual coupling along with RDDS of PIR switch, EIT varies for different operating conditions [4]. It
asymmetric nature of the fault current may cause a high level of un- depends on the magnitude, shape, and polarity of the gap voltage across
symmetrical overvoltages on non-faulty phases [26–28]. Fig. 2(a) and the contacts of CB as well as on the values of RDDS of interrupter and
(b) show the simulation results in terms of the response of CCVT during PIR switch. Fig. 4 demonstrates the dependency of EIT on RDDS of
de-energization of a partially discharged (a) UCTL and (b) SCTL. The interrupter and PIR switch. The typical value of RDDS for interrupter
response of CCVT captured from the real field using Disturbance Re- and PIR switch, considered in this study for a 400 kV breaker, is
corder (DR) during de-energization of a partially discharged UCTL is 100 kV/ms and 130 kV/ms, respectively [4]. Fig. 4 shows simulation
shown in Fig. 2(c). The sequence of de-energization for both the lines is results in terms of 10% change in the value of RDDS for both interrupter
shown in Fig. 2(a) and (b). It has been observed from Fig. 2(a) that the and PIR switch on EIT.
waveform of trapped charge reaches zero value after 2–3 cycles. It is observed from Fig. 4 that EIT varies due to change in prestrike
However, in the actual field, it decays exponentially. As shown in instants while MIT remains constant. Moreover, though there is a 10%
Fig. 2(a), the response of CCVT obtained from simulation study has change in the RDDS of interrupter and PIR switch, the prestrike instants
been mapped with the field results as shown in Fig. 2(c). Comparison of are varied in a narrow band for PIR switch and interrupter. Moreover,
both these figures clearly indicates the exactness of the response of the time laps between electrical closing and mechanical closing is very
CCVT obtained from the simulation study and the practical results. narrow in case of PIR switch (‘Td2’) compare to interrupter (‘Td1’)
Therefore, for cases 2, 3, and 4, the precise measurement and inter- which improves the performance of application of controlled switching.
polation of line side voltages is crucial for accurate evaluation of con- It is reported in the literature that insertion and removal of PIR from
trolled switching targets. the circuit generates primary and secondary surge as defined below.

2.2. Controlled switching of transmission lines having PIR breaker (i) Primary surge: It is the voltage observed at the remote end of the
line for the entire duration up to which PIR is inserted in series with
The representation of OHB scheme and line with non-switchable the transmission line.
reactor scheme are shown in Fig. 3(a) and (b). As shown in Fig. 3(a), (ii) Secondary surge: It is the sudden increase in voltage observed at the
CSD has been provided with PIR-CB on tie bay for controlled de-en- remote end of the transmission line post removal of the PIR from
ergization of the reactor to minimize the probability of re-ignition [29]. the circuit [5]
Conversely, following conditions are considered for a transmission line
with non-switchable reactor (Fig. 3(b)). Both of above-mentioned surges are considered as a switching
surge, developed due to PIR
(i) The transmission line can be used as uncompensated line by
opening disconnector of the shunt reactor (UCTL mode). 3. Modelling of critical parameters
(ii) The transmission line can be used with compensation by con-
necting shunt reactor (SCTL mode). 3.1. Modelling of transmission line
(iii) The transmission line is disconnected by opening its disconnector
and the line reactor is used as the bus reactor for reactive power The modeling of a 400 kV, 50 Hz UCTL and SCTL with SIR less than
compensation. 0.5 is done using PSCAD/EMTDC software package [31]. Aforemen-
tioned Fig. 1 (a) and (b) show the single line diagram of UCTL and SCTL
For condition (iii), in order to minimize the possibility of re-ignition with shunt reactor connected at both ends of the line. The sampling
with the PIR-CB of the line, the CSD has been provided for controlled frequency of 4 kHz has been used for a fundamental frequency of 50 Hz
de-energization of the reactor. Conversely, for conditions (i) & (ii), the [32,33]. Moreover, 369 kV switching class SA are employed at both
CSD is usually bypassed. As discussed in the previous section, to miti- ends of the line for suppression of switching overvoltages [34]. The
gate secondary surge further, the controlled energization of the trans- transmission line has been modeled by the frequency dependent phase
mission line with PIR-CB is performed with same CSD with different model, available in PSCAD, as it is commonly used for transient

349
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Fig. 2. Response of CCVT, simulation results during de-energization of partially discharged (a) UCTL (b) SCTL, and (c) field results during partially discharged UCTL.

overvoltage study [31]. R + jωL

ZC =
The parameters of the transmission line and base source are given in G + jωC (3)
Appendix. Furthermore, as per Indian Electricity rules, ground clear-
ance of the conductor and the spacing between phase conductors for where ω is the angular velocity (rad/s) and R, G, L and C represents
rated line voltage (kV) are calculated using (1) and (2), respectively resistance (Ω/km), conductance (mho/km), inductance (H/km) and
[35]. capacitance (F/km) of the line, respectively.
In case of SCTL, the impedance of shunt reactor (Xlr) and the im-
Ground clerance(CLG ) = [5.2 + {(kV −33)/33} × 0.3] (1) pedance (Xn) of Neutral Grounding Reactor (NGR) is calculated using
(4) and (5), respectively.
Spacing between phase conductors(CLP ) = [3.7 + {(kV −33)/33} × 0.3]
(2) 1
Xlr =
η × βc (4)
Moreover, the surge impedance (ZC) of the transmission line is
calculated using (3). where η and βc represents the amount of compensation and the positive

350
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Fig. 3. Representation of (a) one-and-half-breaker scheme and (b) line connected with non-switchable reactor scheme.

sequence susceptance of the transmission line, respectively. (ii) Mechanical Insertion Time (MIT) of the PIR
−1 (iii) Insertion instant of PIR.
1 ⎧ 1 ⎫
Xn = × ω × (C0−C −1) + −X11
3 ⎨
⎩ X11 ⎬
⎭ (5) In order to analyze the effect of an individual parameter on the level
where C0 and C1 is the zero and positive sequence capacitance of the of the switching surge, the variation in the level of switching surge
line, respectively, whereas X11 is the positive sequence impedance of against different values of one of the parameter has been observed,
the shunt reactor [36]. keeping the other two parameters constant. In this manner, the level of
switching surge against different values of PIR has been determined,
3.2. Modelling of PIR maintaining the insertion instant and MIT of PIR constant. Based on
literature reported on PIR-CB, the value of MIT equals to the time of
In order to determine the optimal value of PIR and MIT, a simula- half cycle (i.e. 10 ms considering the fundamental frequency of 50 Hz)
tion study has been carried out on a power transmission network, as is taken [4], [37]. A simulation study during energization of a partially
shown in Fig. 1(a and b), using PSCAD/EMTDC software package [31]. discharged 200-km long UCTL has been carried out. This study has been
In this paper, all the simulations are performed by installing SAs at both carried out using following insertion instants of PIR.
the ends. During energization of a partially discharged UCTL using PIR-
CB, the level of switching surge mainly depends on the following (i) Half of the peak of the supply side voltage
parameters: (ii) Peak value of the supply side voltage

(i) Value of the Pre-Insertion Resistor (PIR) In this study, the variation in the level of the switching surge

Fig. 4. Effect of change in the value of RDDS of interrupter and PIR switch on EIT.

351
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Fig. 5. Variation in level of switching surge against different values of PIR, (b) for insertion instant of PIR at half of the peak value, (b) at the peak value of the supply
side voltage, (c) variation in level of switching surge against different values of MIT.

(primary and secondary) against the different values of PIR (Ratio of 4. Controlled switching of line with PIR
the value of PIR (‘R’) against surge impedance of the transmission line
(‘Zc’)) has been observed. The range of variation in the ratio of ‘R/Zc’ is 4.1. Evaluation of optimal targets
considered as 0.04–1.7. Fig. 5(a) and (b) show the simulation results.
The PIR insertion target for Fig. 5(a) and (b) is half of the peak value The existing controlled switching methodology is based on targeting
and the peak value of the supply voltage considering non-faulty con- the making instant when the gap voltage across the CB contacts is
dition. It has been observed from Fig. 5(a) and (b) that the level of minimum. In this regard, it is essential to accurately replicate the line
primary surge is higher than the secondary surge till inter-section point side voltages. However, in this paper, a new controlled switching
‘a’. At this point, the value of R/Zc is around 0.77. Thereafter, the value methodology is presented taking reference of source side voltages and
of primary surge reduces, whereas the secondary surge remains almost just the polarity of line side voltage at the time of de-energization of the
constant up to point ‘b’ and then increases. Hence, the minimum value line. In order to employ point on wave switching, seven testing targets,
of the secondary surge is obtained when the value of R/Zc = 1. It is to as shown in Fig. 6(a), have been considered on the source side voltage
be noted from Fig. 5(a) and (b) that the minimum value of secondary waveform. These targets are termed as the “reference targets”.
surge is lower in case when the closing target is half of the peak of the By utilizing these targets, the proposed controlled switching meth-
supply voltage compared to the closing target as the peak of the supply odology minimizes the switching overvoltage during energization of
voltage. UCTL and SCTL by optimizing the electrical insertion instant of PIR
In order to further verify the optimal value of MIT, a simulation (electrical closing instant of interrupter). It has been reported in the
study during energization of a partially discharged 200-km long UCTL literature that the level of switching surges, during energization of a
has been carried out considering non-faulty cases using the optimal partially discharged UCTL and SCTL, can be minimized by limiting the
value of PIR (R/Zc = 1). In this study, the variation in the level of value of gap voltage across the contacts of CB at the time of closing
switching surge (either primary or secondary whichever is higher) [11], [29]. In order to determine the optimal closing targets for UCTL
against different values of MIT (36°–288°) using both above-mentioned and SCTL, a simulation study of line side voltages during protection
closing targets, has been obtained. Fig. 5(c) shows simulation results of dead time of a partially discharged UCTL and SCTL has been carried
this study. It has been observed from Fig. 5(c) that the minimum level out. Fig. 6(b) and (c) show the waveform of supply voltage and trapped
of switching surge is obtained, when the value of MIT = 180° for both charge (line side voltage) for 200-long UCTL in case of no-fault and
aforementioned closing targets. temporary short time single-line-to-ground (L-G) fault condition, re-
spectively. It has been observed from Fig. 6(b) and (c) that the mag-
nitude of the trapped charge is almost equal to half of the peak value of
the supply voltage at the end of protection dead time. Utilizing this fact,
half of the peak value of supply voltage is taken as the optimal closing
target for UCTL. These are indicated as reference target ‘1’ and ‘5’

352
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Fig. 6. Reference targets, (b) effect of trapped charge decay for 200 km long UCTL in case of non-faulty, (c) short time single-line-to-ground fault situations, (d)
conceptual representation to determine the optimal closing target for SCTL.

Table 1 within the limit. Fig. 6(d) shows the simulation results for the following
Level of gap voltage for different closing targets of SCTL. two extreme cases of line side voltages.
Sr. No Scenario at the instant of closing Gap voltage Level of gap
(p.u.) voltage (p.u.) • When the line side voltage attains negative peak (point ‘p’)
Supply side
voltage (p.u.)
Line side voltages
(p.u.)
• When the line side voltage attains positive peak (point ‘q’)
1 Vmax Negative peak Vg1p 2.22 In this case, the gap voltage is the difference between supply side
(point ‘p’) and line side voltages. With reference to point ‘p’, it has been observed
2 Vmax/2 Negative peak Vg2p 1.72 from Fig. 6(d) that the gap voltage for the aforementioned three closing
(point ‘p’) targets (Vmax, Vmax/2, and Vzero) is Vg1p, Vg2p, and Vg3p, respec-
3 Vzero Negative peak Vg3p 1.22
tively. Similarly, for the second extreme case of line side voltage (point
(point ‘p’)
4 Vmax Positive peak Vg1q 0.22 ‘q’), the gap voltage for the same three reference targets is Vg1q, Vg2q,
(point ‘q’) and Vg3q, respectively. It is clear from Fig. 6(d) that the level of gap
5 Vmax/2 Positive peak Vg2q 0.72 voltage changes widely for two closing targets (Vg1p & Vg1q and Vg2p
(point ‘q’)
& Vg2q), whereas the level of gap voltage, developed considering the
6 Vzero Positive peak Vg3q 1.22
(point ‘q’)
supply voltage zero as a closing target (Vg3p and Vg3q), remains un-
changed irrespective of change in polarity of the line side voltage. It can
be well appreciated that the supply voltage zero should be the best
(considering both the polarity of the trapped charge) in Fig. 6(a). In suitable target which maintains the gap voltage within moderate limit.
order to determine the optimal closing targets for SCTL, a simulation Table 1 shows the level of the gap voltages for different closing targets.
study has been carried out considering oscillating nature of line side It is to be noted from Table 1 that the level of gap voltage changes
voltages. Utilizing this fact, three supply side closing targets, as given widely for closing targets Vg1p & Vg1q and Vg2p & Vg2q, whereas it
below, have been evaluated. stays constant for Vg3p & Vg3q. In the proposed methodology, the
supply voltage zero is taken as an optimal closing target for SCTL
(i) Positive peak of the supply voltage (Vmax). (target ‘0’ in Fig. 6(a)). Hence, the proposed methodology does not
(ii) Half of the positive peak of the supply voltage (Vmax/2). depend on the effect of the polarity of the oscillating line side voltage.
(iii) Supply voltage zero (Vzero).
4.2. Proposed methodology
Out of the above-mentioned targets, the optimal closing target is
that target for which the gap voltage does not attain the extreme level Fig. 7(a) shows the flowchart of the proposed methodology. The
for any polarity of the line side voltages. Hence, in order to determine proposed methodology does not require precise measurement of line
the level of switching overvoltages, a simulation study during en- side voltage during protection dead time. It utilizes rising tip side half of
ergization of a partially discharged SCTL considering non-faulty and the peak value of the supply voltage as a closing target for UCTL and the
different faulty conditions has been performed with aforesaid targets. It supply voltage zero as a closing target for SCTL. In Fig. 7(a), the half of
has been observed that the switching overvoltages are maintained well the peak value of the supply voltage on rising tip side is indicated as

353
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Start

Input- Type of line

configuration

No Breaker status
changeover
detected?

Yes
Does Input
No Input Random
auto reclosing from busbar Trip
From load closing
attempted? side command
side command
and line side information
CT (1-Ø)
Yes CCVT

Store the polarity of trapped charge by CSD

I-V converter
No Is fault
detected?
ADC
Yes

Identification of faulty phase

Data storage module
Yes

No L-L or No Data evaluation and polarity detection

L-G L-L-L-G
L-L-G module
fault? fault?
fault?
Yes Yes
Yes Controlled command
A for UCTL B
A/B
B for SCTL
CSD
PIR inserted in other phases after 1200 , one after other in R-Y-B sequence

Closing Compensating
PIR switch for each phase is closed after MIT coil (3-Ø) sensors

End PIR-CB

(a) (b)
Fig. 7. (a) Flow chart of the proposed methodology (b) conceptual lay out of CSD.

target “A”, whereas the supply voltage zero is indicated as target “B”. As When the random closing command is raised either by manual
shown in Fig. 7(a), the configuration of the transmission line (un- mode or through Supervisory Control and Data Acquisition (SCADA), it
compensated (UCTL)/compensated (SCTL)) is given as input to the is converted into controlled closing command by introducing inten-
proposed methodology. The process starts upon detection of breaker tional delay. This delay is calculated based on voltage & current data
status changeover from close to open. At the time of de-energization, acquired from the actual field and type of load to be switched.
the polarity of the individual phase of line side voltage has been stored Moreover, based on the type of load to be switched such as transmission
by CSD. This is required for UCTL as the direction of the trapped charge line, transformer, shunt-reactor, and shunt capacitor, evaluation of the
remains unaltered during the protection dead time. Hence, in the pro- optimum switching targets will be carried out by the data evaluation
posed methodology, the reclosing operation for UCTL has been per- and polarity detection module of CSD. For this purpose, different set-
formed on the same polarity of the source voltage. tings are provided. Various tasks performed by data evaluation and
Fig. 7(b) shows conceptual layout of CSD for the storage of polarity. polarity detection module are as under.
The CSD is capable to acquire current and voltage data from the actual
field through CT and CCVT. However, the proposed scheme requires (i) It analyses the voltage and current signals acquired from the real
voltage data only. Hence, as shown in Fig. 7(b), the CSD acquires line field.
side voltage data through instrument transformers continuously. It is to (ii) It evaluates mechanical operating time of the CB for last operation
be noted that the voltage signal is fed directly to the data storage and by collecting the feedback from an auxiliary contact of the CB.
evaluation module of CSD using Analog to Digital Converter (ADC), (iii) It detects zero crossing of busbar side and line side voltages for
whereas current signal fed to this module through current-to-voltage (I- individual phases. This will be used for evaluating optimum output
V converter) converter and ADC. The data storage module of CSD stores command for individual poles of the CB after considering operating
the acquired data using non-volatile memory. time of CB. This is known as adaptive operating time functionality

354
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Table 2
Summary of simulation cases generated for UCTL & SCTL.
Length of line Level of compensation Type of fault Polarity of Fault resistance Duration of Load Fault distance (% Fault inception Total
(km) (%) trapped charge (Ω) fault (ms) angle (δ) of line length) angle (degree) cases

UCTL
200 (1) Not Applicable R-G, R-Y, R-Y- + ve, −ve (2) 0.0001, 100 (2) 100, 250 (2) −30° (1) 0, 50, 100 (3) 0°, 90°, 135° (3) 288 cases
G, R-Y-B-G
(4)

SCTL
450 (1) 30, 50, 70 (3) R-G, R-Y, R-Y- + ve, −ve (2) 0.0001, 100 (2) 100, 250 (2) −30° (1) 0, 50, 100 (3) 0°, 90°, 135° (3) 864 cases
G, R-Y-B-G
(4)

and it is based on the auto compensating curve fitting technique. energization of partially discharged UCTL and SCTL suggests that the
(iv) Variation in different parameters such as ambient temperature, dc level of switching surge generated in case of stagger pole closing of CB
supply voltage, and hydraulic/air pressure of the CB are re- is found lower than simultaneous pole closing of CB [12].
sponsible for change in the MOT of the CB. This else otherwise, In staggered closing, each phase of the breaker is closed one after
would result into incorrect determination of the actual switching another, generally in R-Y-B sequence. The stagger pole operation of CB
targets. Required compensation factor in operating time of in- reduces the level of switching surges by preventing the interaction
dividual poles of CB (both close and open) has been evaluated among the phases, where switching surge on one phase induces addi-
based on data collected from compensating sensors. Hence, this tional surge at the wrong time on another phase. In case of simulta-
module calculates an intentional time delay after the initiation of neous three-phase reclosing, these additional induced surges create the
input close command considering all the above mentioned para- highest overvoltage [40]. Hence, in proposed methodology, the concept
meters. of staggered pole operation of the CB has been utilized.
(v) The instant, when the tripping command has been given to CB, the
trip command information module activates the data evaluation 4.3. Simulation cases
and polarity detection module. Thereafter, the data evaluation and
polarity detection module starts acquiring the line side voltage In order to evaluate performance of the proposed algorithm, wide
data from data storage module for a time period of 2–4 cycles. In variation in system conditions, operating parameters and type of fault
case of opening operation of the CB, the change in line side voltage have been considered. The simulations also include critical cases like
has been identified by data evaluation and polarity detection terminal fault (zero km from bus-1) and out of phase closing. Here, out
module using abrupt change detection algorithm [38]. There are of phase closing is carried out by inserting PIR at an instant when the
numbers of abrupt change detection algorithms developed by polarity of the supply-side voltage is opposite to the polarity of line side
manufacturers [39]. Discussion of these algorithms is out of scope voltage. This leads to generation of the maximum level of the secondary
of this paper. Subsequently, the polarity of the line side voltage for surge. Table 2 shows summary of simulation cases generated with
individual phases have been identified and the optimal closing variation in operating and system parameters. In this context, line
targets for next energization operation are also determined [39]. length (1) × type of fault (4) × polarity of trapped charge (2) × fault
The non-auto-recloser is as good as energization of the fully dis- resistance (2) × duration of fault (2) × load angle (1) × fault distance
charged transmission line. Hence, the PIR should be inserted at (3) × fault inception angle (3) = 288 simulation cases have been gen-
supply voltage zero (“B”) for both types of configurations of the erated. Afterwards, these cases are tested for all seven reference targets
transmission line (UCTL/SCTL). Generally, auto-reclosing is per- as specified in Fig. 5(a). Therefore, an overall dataset of 2016 (288 × 7)
formed only for single-line-to-ground (L-G) fault. However, in fault cases have been investigated for UCTL in this study. Similarly,
order to verify the effectiveness of the proposed methodology, utilizing all the above-mentioned cases along with three different va-
faults other than L-G fault are also considered. It is to be noted that lues of the compensation level (30%, 50%, and 70%), 6048 (2016 × 3)
the proposed methodology for non-faulty condition utilizes the cases have been generated for SCTL. Moreover, two types of SCTL
same closing targets which are used for faulty conditions. In case of configurations are considered (i) shunt reactors are connected at both
non-faulty condition, the first phase to re-close is the phase which ends of the line and (ii) shunt reactor is connected at one end of the
attains reference target A or B, prior to other two phases after the line. Therefore, a total of 12,096 (6048 × 2) simulation cases have been
protection dead time. Upon detection of a fault, phase/phases in- simulated for SCTL.
volved in the fault will be identified by the faulted phase classifier
algorithm [38].
5. Results and discussions

After re-closing the first phase, the trapped charge of reclosed phase
5.1. Performance of proposed controlled switching methodology for UCTL
as well as other phase/phases shifts towards the polarity of first re-
and SCTL
closing phase. In case of L-G fault, among two non-faulty phases, the
phase which attains closing target “A/B” earlier shall be closed first. In
In real scenario, there may be a possibility of nuisance tripping
case of line-to-line (L-L)/double-line-to-ground (L-L-G) fault, the
which is considered as no-fault condition in this paper. This effect is
healthy phase should be re-closed first at the target “A/B” in the pro-
considered in simulations by varying the polarity of trapped charge
posed flowchart. The targeting in case of triple-line (L-L-L)/triple-line-
(positive and negative) with constant line length (200 km) and load
to-ground (L-L-L-G) faults on UCTL/SCTL is performed in the same
angle (30°). Furthermore, these (2 × 1 × 1) cases are tested on 7 re-
manner as it is used during energization of a fully discharged line (non-
ference targets in context to insertion instant of PIR. Therefore, a total
auto recloser case denoted by target “B” in the flowchart). Moreover,
of 14 cases (2 × 7), involving auto recloser during no-fault condition,
for all cases (including faulty or non-faulty condition with or without
have been investigated in this study for UCTL. Similarly, 14 × 3 (three
recloser), the remaining two phases should be energized in R-Y-B phase
compensation levels) = 42 cases of auto recloser during no-fault con-
sequence after 120°, one after other. The literature reported on the
dition have been studied for SCTL. In order to evaluate performance of

355
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Fig. 8. Over voltages along the length of (a) 200 km long UCTL and 450 km long SCTL with (b) 30%, (c) 50%, and (d) 70% level of compensation.

the proposed methodology, a simulation study has been carried out performance of PIR-CB with non-PIR-CB.
during energization of a partially discharged UCTL and SCTL using PIR- The MOT of CB changes due to statistical and systematic variations.
CB. This simulation study has been performed on SCTL considering The systematic variations in MOT occur due to external parameters
aforementioned three levels of compensation for non-faulty and dif- such as variation in the ambient temperature, dc battery supply voltage,
ferent faulty conditions. Fig. 8(a) shows the overvoltage in terms of the and aging effect (after the time of installation). Based on appropriate
level of secondary surge for 200-km long UCTL. Moreover, Fig. 8(b)–(d) testing in laboratories, it has been observed that the controlled
shows the same effect for 450-km long SCTL with 30%, 50%, and 70% switching relays can compensate aforementioned systematic variation
compensation level, respectively. in operating time using suitable compensation curves. On the other
This comparison is in terms of the maximum secondary surge gen- hand, the statistical variations in operating time of the CB cannot be
erated at the remote end (Bus-2) of UCTL and SCTL with SA. It is to be compensated. In this paper, an overall variation of ± 1 ms due to the
noted from Fig. 8 that the level of switching surge increases from statistical variation is considered. A simulation study during energiza-
sending end to receiving end and becomes maximum at the mid-point tion of a partially discharged 200-km long UCTL and 450-km long SCTL
of the line. Thereafter, it decreases slowly towards the receiving end. In has been carried out for the proposed methodology (which contains
case of energization of a partially discharged UCTL (Fig. 8(a)), the both CSD and PIR-CB) and the usual PIR methodology. In this study,
highest level of switching surge (1.47p.u.) is observed during short time various conditions such as no-fault, L-G fault, L-L-G fault, L-L fault, and
temporary (100 ms) L-G fault compared to non-faulty and different L-L-L-G fault have been analysed. The Mechanical Operating Time
faulty situations. On other hand, it is clear from Fig. 8(b)–(d) that the (MOT) scatter due to statistical variations of the CB of the order of ±
highest level of the secondary surge is observed during no-fault situa- 1 ms has been considered. During this study, three different levels of
tion with 30% compensation level compared to other faulty conditions. compensation for SCTL (30%, 50%, and 70%) have been considered.
Moreover, the level of secondary surge is always higher in case of L-G Moreover, the level of switching surge is monitored at the far end of the
fault compared to other faults. In case of L-G fault, the level of sec- transmission line. The simulation results for UCTL are shown in
ondary surge is highest (1.62 p.u.) for 30% compensation level. Fig. 9(a), whereas Fig. 9(b)–(d) show the simulation results for SCTL. It
is to be noted from Fig. 9(a) that the maximum level of switching surge
given by the usual PIR methodology and the proposed methodology is
5.2. Effect of RDDS and mechanical operating time scatter of the order of 1.9 p.u. and 1.34 p.u., respectively, during no-fault
condition. However, during faulty (L-G fault) condition, the maximum
The latest generation of CBs have mechanical and dielectric char- level of switching surge is around 1.76 p.u. and 1.54 p.u. for the usual
acteristics which can attain the making target with the accuracy of ± PIR methodology and the proposed methodology, respectively. Con-
1 ms [40]. In order to evaluate performance of the proposed metho- versely, as observed from Fig. 9(b)–(d), the maximum level of switching
dology, a simulation study has been carried out. During this study, surge given by the proposed methodology is always lower than the
energization of a partially discharged UCTL and SCTL is performed usual PIR methodology for faulty/no-fault situations. The above dis-
considering the combined effect of the Mechanical Operating Time cussion clearly indicates superiority of the proposed scheme over the
(MOT) and RDDS scatter of the CB. Various switching scenarios are usual PIR methodology in terms of reduction in the level of switching
used in order to compare the switching overvoltage mitigation

356
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Fig. 9. Comparative evaluation of the proposed methodology and usual PIR (stand-alone) methodology during energization of a partially discharged (a) UCTL, (b)
30%, (c) 50%, and (d) 70% compensated SCTL.

surge. In order to evaluate performance of the proposed methodology, a si-

mulation study during energization of a partially discharged 200-km
5.3. Effect of strong and weak source long UCTL has been carried out considering variation in the decaying
rate of the trapped charges in no-fault condition. In this study, at the
In order to evaluate performance of the proposed algorithm in time of energization of a partially charged UCTL, two extreme levels
context to grid strength, three different sources having different para- (i.e. 0.8 p.u. and 0.2 p.u.) of trapped charges at the end of protection
meters, as shown in Table 3, have been considered. In order to evaluate dead time are considered. The simulation results of the study are shown
the performance of the proposed methodology, a simulation study in Table 4. Hence, the level of switching surge remains immune to the
during energization of a partially discharged 200-km long UCTL and variation in the magnitude of the trapped charge due to weather con-
450-km long SCTL has been carried out for the proposed methodology ditions and line insulation.
(which contains both CSD and PIR-CB) and the usual PIR methodology.
For all these cases, the same numbers of faulty and non-faulty cases, 5.5. Effect of SA
as mentioned in previous sub-section, have been generated. In this
study, the dataset of 288 faulty cases and 02 non-faulty cases for UCTL In order to evaluate performance of the proposed methodology
and 864 faulty cases and 06 non-faulty cases for SCTL have been in- considering present and absence of SA at both the ends of the line, a
vestigated for all three sources. It has been observed from Table 3 that simulation study during energization of a partially discharged 200-km
the variation in secondary surge is negligible with change in short long UCTL and 450-km long SCTL has been carried out without con-
circuit level or grid strength. sidering effect of SA. Data sets are generated as shown in Table 2 for
UCTL and SCTL. The simulation results of this study are shown in
5.4. Effect of weather conditions and line insulation Table 5. The switching surge is measured at the remote end of the line.
It has been observed from Table 5 that the change in the level of
It has been reported in literature that the magnitude of trapped switching surge, obtained during energization of a partially discharged
charges changes due to effect of weather conditions and line insulation. UCTL and SCTL considering presence/absence of SA, is negligible.

Table 3
Performance of proposed methodology against the short circuit capacity of the source.
Parameters Length of line (km) S1 (Strong source) S2 (Weak source-1) S3 (Weak source-2)

Short circuit level (kA) – 40 28 11

Positive sequence impedance (Ω) – 0.4115 + j 3.546 1.5 × S1 4 × S1
Zero sequence impedance (Ω) – 2.822 + j 10.08 1.5 × S1 4 × S1
Level of switching surge (p.u.) for UCTL 200 1.47 1.47 1.46
Level of switching surge (p.u.) for SCTL 450 1.62 1.6 1.58

357
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

Table 4 presence/absence of SA.

Performance of the proposed methodology considering change in decaying rate
of trapped charges. Appendix A
Sr. No Level of trapped charge (p.u.) Maximum level of switching surge (p.u.)
System Parameters
1 0.2 1.18 Source parameters (both ends):
2 0.8 1.27
Positive sequence impedance = 0.4115 + j 3.546 Ω
Zero sequence impedance = 2.822 + j 10.08 Ω
Table 5
Transmission-line parameters:
Performance of the proposed methodology considering presence/absence of SA. Positive sequence impedance = 0.017272 + j 0.3135 Ω/km
Zero sequence impedance = 0.277 + j 1.04 Ω/km
Type of the The maximum level of switching surge (p.u.) considering no-fault and
Positive sequence capacitance = 14.88 nF/km
line different faulty conditions
Zero sequence capacitance = 9.8 nF/km
With SA No-SA Difference Ground Clearance = 9.39 m
Spacing between two phase conductors = 7.74 m
UCTL 1.53 1.57 0.04
SCTL 1.57 1.59 0.02
Appendix B. Supplementary data

Table 6 Supplementary data associated with this article can be found, in the
Performance of the proposed methodology against change in closing sequence online version, at http://dx.doi.org/10.1016/j.ijepes.2018.06.024.
of the CB.
References
Type of the The maximum level of switching surge (p.u.) considering no-fault and
line different faulty conditions
[1] IEEE Guide for Protective Relay Applications to Transmission Lines. IEEE Std
Closing sequence R-Y- Closing sequence R-B- Difference C37113 2015; 2015. p. 1–141. doi:10.1109/IEEESTD.2016.7502047.
B Y [2] Wagner C, Bankoske J. Evaluation of surge suppression resistors in high-voltage
circuit breakers. IEEE Trans Power Appar Syst 1967;PAS-86:698–707. http://dx.doi.
UCTL 1.54 1.61 0.07 org/10.1109/TPAS.1967.291881.
[3] Report IC. Switching surges, part III: field and analyzer results for transmission lines
SCTL 1.63 1.69 0.06
past. IEEE Trans Power Appar Syst 1970;PAS-89:698–707. http://dx.doi.org/10.
1109/TPAS.1970.292750.
[4] Heiermeier H, Raysaha RB. Power testing of preinsertion resistors: limitations and
5.6. Effect of change in closing sequence solution. IEEE Trans Power Delivery 2017;32:1688–95. http://dx.doi.org/10.1109/
TPWRD.2016.2519604.
[5] Greenwood A. Electrical Transients in Power Systems. 2nd ed. New York, USA: John
For staggered pole operation of CB, the closing phase sequence R-B- Wiley & Sons. INC.; 1991.
Y has been reported in the literature [12] instead of proposed closing [6] Thoren HB. Reduction of switching overvoltages in EHV and UHV systems. IEEE
sequence (i.e. R-Y-B). Comparative evaluation of both the aforemen- Trans Power Appar Syst. 1971;PAS-90:1321–6. http://dx.doi.org/10.1109/TPAS.
1971.292935.
tioned phase sequences in terms of level of switching surge has been [7] Thomas C, Hedin R. Switching surges on transmission lines studied by differential
carried out. In this study, energization of a partially discharged 200-km analyzer simulation. IEEE Trans Power Appar Syst 1969;PAS-88:636–45. http://dx.
long UCTL and 450-km long SCTL has been carried out. Data sets are doi.org/10.1109/TPAS.1969.292352.
[8] Esmeraldo PCV, Amon FJ, Carvalho FMS, Carvalho ACC, Morais SA. Circuit-breaker
generated as shown in Table 2 for UCTL and SCTL. The simulation re- requirements for alternative configurations of a 500 kV transmission system. IEEE
sults are shown in Table 6. The switching surge is measured at the re- Trans Power Delivery 1999;14:169–75. http://dx.doi.org/10.1109/61.736710.
mote end of the line. It has been observed from Table 6 that the change [9] Faria Da Silva F, Bak CL, Gudmundsdóttir US, Wiechowski W, Knardrupgård MR.
Use of a pre-insertion resistor to minimize zero-missing phenomenon and switching
in the level of switching surge, obtained during energization of a par- overvoltages. 2009 In: IEEE Power Energy Soc. Gen. Meet. PES ’09; 2009. doi:10.
tially discharged UCTL and SCTL considering change in closing se- 1109/PES.2009.5275992.
quence of the CB, is negligible. [10] Legate AC, Brunke JH, Ray JJ, Yasuda EJ. Elimination of closing resistors on EHV
circuit breakers. IEEE Trans Power Delivery 1988;3:223–31. http://dx.doi.org/10.
1109/61.4249.
6. Conclusion [11] CIGRE Task Force 13.00.1. Controlled switching A state of the art survey – part I.
ELECTRA; 1995. p. 1–34.
[12] CIGRE Task Force 13.00.1. Controlled switching A state of the art survey – part II,
This paper presents a new methodology which effectively mitigates ELECTRA; 1996. p. 1–26.
the switching surge during energization/re-energization of UCTL and [13] ABB High Voltage Products. Controlled Switching Buyer’s and Application Guide.
SCTL with the help of CSD with PIR-CB. Compared to the existing Ludvika, Sweden; 2013.
[14] Ito H. Controlled switching technologies, state-of-the-art. IEEE/PES Transm.
controlled switching strategy (both for UCTL/SCTL) which requires line
Distrib. Conf. Exhib., vol. 2. Yokohama, Japan: IEEE; 2002. p. 1455–60. doi:10.
side voltage signatures, the proposed methodology does not necessitate 1109/TDC.2002.1177695.
the same for SCTL and requires only the polarity of line side voltages at [15] Fernandez PC, Zani CR. Use of Controlled switching systems in power system to
the time of de-energization for UCTL. The performance of the proposed mitigate switching transients.Trends & benefits –Brazilian Experience. CIGRE WG
13–207, vol. 13–207. Rio de Janeiro, Brazil: CIGRE session; 2002. p. 3–8.
methodology has been evaluated by generating diversified cases on [16] Froehlich K, Hoelzl C, Stanek M, Carvalho AC, Hofbauer W, Hoegg P, et al.
400 kV UCTL and SCTL using PSCAD/EMTDC software package. These Controlled closing on shunt reactor compensated transmission lines I closing control
cases were generated by varying parameters such as polarity of trapped device development. IEEE Trans Power Delivery 1997;12:734–40. http://dx.doi.
org/10.1109/61.584359.
charge, fault location, type of fault, fault resistance, and fault inception [17] Froehlich K, Hoelzl C, Stanek M, Hofbauer W, Hoegg P, Peelo DF, et al. Controlled
angle. The simulation results clearly indicate that the proposed meth- closing on shunt reactor compensated transmission lines part 2: application of
odology is able to further mitigate the switching surge of the order of closing control device for high-speed autoreclosing on BC hydro 500 kV transmis-
sion line K. IEEE Trans Power Delivery 1997;12:734–40. http://dx.doi.org/10.
33% and 20% for UCTL and SCTL, respectively. The comparative eva- 1109/61.584360.
luation of the proposed methodology with the usual PIR methodology [18] Dantas KMC, Neves WLA, Fernandes D. An approach for controlled reclosing of
indicates its effectiveness during statistical variations of CB of the order shunt-compensated transmission lines. IEEE Trans Power Delivery
2014;29:1203–11. http://dx.doi.org/10.1109/TPWRD.2013.2289394.
of ± 1 ms. The response of the proposed methodology remains immune [19] Parikh UB, Bhalja BR. SVR-based current zero estimation technique for controlled
to the effect of various parameters such as type of source, weather fault interruption in the series-compensated transmission line. IEEE Trans Power
conditions, insulation resistance, closing phase sequence (R-B-Y), and Delivery 2013;28:1364–72. http://dx.doi.org/10.1109/TPWRD.2013.2247425.

358
K.A. Bhatt et al. Electrical Power and Energy Systems 103 (2018) 347–359

[20] Bhalja BR, Maheshwari RP, Chothani, Chothani N. Protection and Switch Gear. 2nd. [37] Begamudre Rakosh Das, Begamudre. Extra high voltage AC transmission
New Delhi: India Oxford University Press; 2011. Engineering. 4th ed. New age international; 2011. http://dx.doi.org/10.1017/
[21] Smith Z, Collette L, Yonezawa T. Controlled Switching for Circuit Breakers with Pre- CBO9781107415324.004.
insertion Resistors Energizing Shunt Capacitor Banks. CIGRE -IEC Colloq., [38] Ukil A, Ivanovi R. Application of abrupt change detection in power systems dis-
Montreal, Canada: CIGRE-IEC Colloquium; May 9–11, 2016. p. 1–7. turbance analysis and relay performance monitoring. IEEE Trans Power Delivery
[22] Hu J, Bisewski B, Sherry R, Dalzell M, Zavahir M, Bartzsch C. Converter 2007;22:59–66. http://dx.doi.org/10.1109/TPWRD.2006.887096.
Transformer Inrush Control Using Hybrid Pre-insertion Resistors and Point-on- [39] Stanek M. Analysis of circuit breaker controlled switching operations – From
Wave Switching in the New Zealand HVDC System. CIGRE Sess. Paris, France; 2016, manual to automatic. In: Proc. Univ. Power Eng. Conf., vol. 2015–Novem; 2015.
p. 1–8. doi:10.1109/UPEC.2015.7339842.
[23] Seyedi H, Tanhaeidilmaghani S. New controlled switching approach for limitation [40] Goldsworthy D, Roseburg T, Tziouvaras D, Pope J. Controlled Switching of HVAC
of transmission line switching overvoltages. IET Gener Transm Distrib Circuit Breakers: Application Examples and Benefits. In: 61st Annu. Conf. Prot.
2013;7:218–25. http://dx.doi.org/10.1049/iet-gtd.2012.0285. Relay Eng., Texas, USA: IEEE; 2008. p. 520–35. doi:10.1109/CPRE.2008.4515078.
[24] IEC 61869–3. “Instrument transformers – part 3. Geneva, Switzerland; 2011.
[25] Seyedi H, Golabi S, Abam Z. Limitation of transmission line switching overvoltages. Kunal A. Bhatt received the B.E. in Electrical Engineering from L. D. Engineering
2010 In: IEEE Int. Conf. Power Energy, Paris, France: IEEE; 2010, p. 363–8. doi:10. College, Gujarat University, in 2002 and M. E. in power system engineering from Gujarat
1109/PECON.2010.5697610. Technological University in 2013, Ahmedabad, Gujarat, India. Currently, he is perusing
[26] Dantas KMC, Neves WLA, Cardoso GA, Fonseca LC. Real Time Implementation of the Ph. D. in Electrical Engineering, Indian Institute of Technology Roorkee, Uttarakhand,
Transmission Line Controlled Switching. Int. Conf. Power Syst. Transients, Texas, India.
USA; June 14–17 2011, p. 1–6.
[27] Kezunovic M, Kojovic L, Skendzic V, Fromen CW, Sevcik DR, Nilsson SL. Digital
models of coupling capacitor voltage transformers for protective relay transient Bhavesh R. Bhalja (M’2007-SM’ 2010) received B. E. (Electrical Engineering) and M. E.
studies. IEEE Trans Power Delivery 1992;7:1927–35. http://dx.doi.org/10.1109/ degree in power system engineering from B.V.M. Engineering College, Sardar Patel
61.156996. University, Vallabh Vidyanagar, India in 1999 and 2001, respectively. He obtained Ph. D
[28] Ove US, Slguerdidjane H, Boudaoud F, Bastard P, Dupraz JP, Collet M, et al. On from Indian Institute of Technology Roorkee in 2007. Currently, he is working as an
Associate Professor, Department of Electrical Engineering, Indian Institute of Technology
controlled switching of high voltage unloaded transmission lines I. E I
Elektrotechnik Und Informationstechnik Springer 2002;119:415–21. Roorkee, Uttrakhand, India. He has a teaching experience of more than 15 years. He has
published more than 90 papers in journals and conferences at international and national
[29] ABB substation automation products. Switchsync ™ PWC600 Product Guide.
Vesteras, Sweden; 2013. levels. He got Young Engineers Award and Merit Award by Institution of Engineers, India
[30] ABB High Voltage Products. Live Tank Circuit Breakers Buyer’s Guide Table of in 2009 and 2007, respectively. He has also written a book on Protection and Switchgear
contents. Ludvika, Sweden; 2014. published by Oxford University Press, 2011, India. His research interests include power
[31] Manitoba-HVDC Research Center. PSCAD/EMTDC User’s Manual Version 4.2; system protection and automation, distributed generation, power quality, system mod-
eling and simulation and artificial intelligence.
2005. p. 1–492.
[32] Bornard P, Tesseron J, Bastide J, Nourris M. Field experience of digital fault re-
corders and distance relay in EHV substations. IEEE Trans Power Appar Syst. Urmil B. Parikh is working with Power Products High Voltage Services (PPHV-S), ABB
1984;PAS-103:133–9. http://dx.doi.org/10.1109/TPAS.1984.318586. India Ltd, Vadodara, India. He received the Bachelor of Electrical Engineering from
[33] Youssef OAS. Combined fuzzy-logic wavelet-based fault classification technique for Gujarat University in 1998. He has also been part of Technology Center prior to working
power system relaying. IEEE Trans Power Delivery 2004;19:582–9. http://dx.doi. with Services for a span of eight years. Post to his graduation, he worked as a faculty in
org/10.1109/TPWRD.2004.826386. the department of Electrical Engineering at SVM Institute of Technology, Bharuch, India
[34] De Carvalho AV, Freire ARF, De Oliveira HM. Transient interaction between cou- during the period of 1999 to 2005. Then, he pursued his masters from Indian Institute of
pling capacitors voltage transformers and transmission lines. In: IEEE Power Energy Technology, Roorkee, India in 2007, with Power Systems specialization. He has also
Soc. Gen. Meet. PES ’09, Calgury-AB, CANADA; 2009. p. 1–8. doi:10.1109/PES. worked for Areva T & D India for a short span of one and a half years. His main areas of
2009.5275646. interest are power system studies, controlled switching, power system protection and GIS.
[35] Section-37. the Indian Electricity Rules, 1956; 2000. p. 45–8. He is also a contributive member of Cigre WG A3.35 “Guidelines and best practices for the
[36] Nashawati E, Le B, Taylor D. Impacts of shunt reactor on transmission line pro- commissioning of controlled switching projects”.
tection; 2011.

359