This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

Adaptive Distance Relaying for Distribution Lines

Connecting Inverter-Interfaced Solar PV Plant
Priyanka Mishra, Ashok Kumar Pradhan, Senior Member, IEEE and Prabodh Bajpai, Senior Member, IEEE
Abstract—Distribution lines are generally protected by over- scheme is sought for PV connected distribution lines where
current relays. With the integration of inverter-interfaced solar distance relay finds a scope.
photovoltaic (PV) plant having current limiting feature, fault Local voltage and current phasors are being used in dis-
current seen by the relay on PV-side of that feeder becomes
comparable to load current. Conventional overcurrent relaying tance relays for impedance calculation to derive protection
principle is not suitable for distribution line protection. Distance decision. Selectivity and directionality make distance relay
relay may be a viable option for protection of distribution robust to different fault situations [9]. In the case of inverter
lines connecting solar PV plant. However, the positive and interfaced PV plant, the current and voltage being controlled
negative sequence source impedances of PV plant depend on even during a fault, the apparent impedance seen by the relay
inverter controller which results in limited performance of fixed
setting distance relay. This paper proposes an adaptive distance gets influenced by the control strategy [10]–[13]. Therefore,
relay setting to protect distribution line connecting PV plant, a conventional distance relay algorithm finds limitation while
using prefault voltage and current data at relaying point. The protecting a distribution line connecting inverter-interfaced PV
method calculates the positive and negative sequence PV source plant. Different methods have been proposed to calculate the
impedances for boundary setting of distance relay at PV-side. The apparent impedance seen by the fixed setting distance relay
proposed trip boundary is modified adaptively with the change
in prefault conditions of PV plant. Performance of proposed on PV-side. In [14], a method for setting the zero sequence
method is tested for different control strategies and operating compensation factor for the distribution line is proposed.
capacities of PV plant and variation in grid source impedance However, the method needs to vary the forward reactance
on a 34-bus distribution system. Real-time application of the reach setting of each ground distance relay in accordance with
proposed method is validated in hardware-in-loop using OPAL- the operating capacity of PV plant which is difficult to achieve.
RT simulators with IEC 61850 as communication protocol and
found to be accurate. In [15], a distance relay based protection scheme is proposed
for the distribution line connected to PV plant. The method
Index Terms—Renewable integration, solar photovoltaic, line
fault, adaptive relaying, distribution line protection, distance considers the control strategy incorporated in the inverter to
relay, hardware-in-loop, IEC 61850. analyse the fault current supplied by PV plant. The scheme
does not address the problem of high fault resistance. In [16],
I. I NTRODUCTION a new relation is used for calculation of apparent impedance
during a fault in PV connected line which applies positive

T ODAY an increasing number of solar photovoltaic (PV)

plant is being integrated to the distribution systems [1].
Fault ride through (FRT) requirement allows the interaction
sequence reference frame (PSRF) control strategy in PV plant
considering only positive sequence impedance of the plant.
If the control strategy is changed to dual sequence reference
of PV plant in the distribution system even during a fault frame (DSRF), the controller will regulate both positive and
[2]. Such a plant has different control strategies to manage negative sequence impedances of PV plant unlike constant
the system and allows limited fault current for the protection source impedance of synchronous based distributed generator
of inverters [3]. As a result, fault current seen by the relay during fault [17]. This change in source impedance affects the
on PV-side of the line, connecting PV plant to the grid, current from PV plant compared to currents from the remote
becomes comparable to load current. Fault discrimination terminal for a fault in the line interconnecting the PV plant
using conventional overcurrent principle is not possible for to the system. This results in inaccurate apparent impedance
such a line [4]. Further, the intermittency of PV plant, re- calculation. In such approaches, the influence of variation in
sulting in different fault currents, makes the protection of source impedance, control strategy and operating capacity of
PV connecting lines more challenging [5]. To protect such PV plant is not considered which limits the performance of
lines, research has been directed towards developing relay such relays. Such an issue can be mitigated by adjusting the
algorithms using voltage signals along with current signals trip boundary in accordance with the operating conditions of
[6]. In [7], voltage-controlled overcurrent relay is proposed to the system.
identify fault in a zone. This method uses voltage magnitude There are different adaptive distance relaying concepts
to start the protection algorithm. With high fault resistance and available for transmission line protection without considering
voltage control strategy in PV plant, the fault voltage seen by PV plant in the system. In [18], an adaptive relaying scheme
the relay becomes more than the threshold. This limits the is proposed using synchrophasor measurements at different
performance of the relay. To overcome such issues, a voltage- buses in the system. In [19], an adaptive distance relay is
restrained overcurrent relay characteristic is proposed in [8]. proposed using the information from both ends of the line to
However, the method finds limitation for varying prefault load- be protected. These methods require dedicated communication
ing conditions (current, voltage and frequency). Additionally, system and is not advantageous in case of communication
the varying pickup level makes it more difficult to coordinate link failure. To avoid such issues, a local data based adaptive
with other such relays in the system. An alternative protection relay setting is proposed in [20], [21]. However, these methods

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

VDC_ref
VDC_ref RfId+-ωLfIq++VSd
+
2 PV
Limiter + VPV
-+
I d_ref
VDC PI +- PI ∑
dq IPV
Qm -+ PI +- PI ∑ abc

PWMb
I+ Booster
Qref Limiter q_ref
VDC_ref -θPLL
(positive sequence controller) R I++ωL I++V+
f q f d Sq
2
Vdc
+
VSd VSdp Id+ Idp
dq Ia
NF dq VSa NF

PWMinv
VSb NF abc Ib ∑ Inverter
+ NF abc Iq+ Iqp Ic
VSq VSqp VSc
PLL
θPLL -1 -1 I
- Rf
VSd VSdn VSa Id- Idn
dq Ia Filter
NF dq NF Lf
VSb Ib Cf
NF abc NF abc
(a)
-
VSq VSqn VSc Iq- Iqn Ic
IS
VS S
- VDC_ref
VSd -ωCfVSq- +ISd
- RfId- +ωLfI-q+VSd
-
- -
2 Transformer
VSd_ref Id_ref
-+ PI ∑ -
+ PI ∑ dq
- abc VM M
- -+ - PI ∑ + PI ∑
IM
VSq_ref Iq_ref
- -
- +ISq VDC_ref θPLL
VSq ωCfVSd -
RfIq--ωLfId- +VSq 2 To grid
(negative sequence controller)
PI block is proportional-integrator compensator. NF is second-order notch filter.
Fig. 1. Double-stage configuration of PV plant with grid and dual controller: (a) transformation of three-phase (abc) signal to dq form.
consider the constant source impedance during fault which by regulating the current measurements to their respective
is not so in the case of a PV plant. In [22], a fuzzy rule reference command. As a result, the fault current behavior is
based approach is applied for adaptive zone setting of distance controlled by the sequenced current controller and associated
relay. This method requires the knowledge of existing network current limits which results in varying source impedance of
topology and may not be valid for operational changes in PV plant. This influences the apparent impedance seen by
the system in the presence of PV plant. Therefore, alternative the distance relay. Thus, the impact of control algorithms on
method is sought for protection of distribution lines connecting output current and impedance of the PV plant is studied in
PV plant. the context of distance relay. An analysis is carried out in the
This paper proposes an adaptive distance relaying to protect following.
distribution lines with PV plant, using local prefault voltage A. Inverter current limitation
and current data. The proposed method calculates the inter- Fig. 1 shows the positive and negative sequence controllers
nal voltage and associated source impedances of PV plant, of inverter. A positive and negative sequence control strategy
considering the strategy incorporated in positive and negative is used to mitigate the double frequency ripples on the DC-
sequence controllers. A method for calculation of fault current link capacitor and to reduce the voltage stress on the inverter
supplied by PV plant is also developed that includes inverter switches. Zero sequence controller is not considered in the
current limiting function. The proposed method sets an adap- arrangement with inverter being connected to the grid via
tive trip boundary for the distance relay at PV-side of the line a dYg1 transformer with delta winding on PV-side [17].
in order to achieve accurate protection decision. The double Even its consideration will not affect the PV output with
stage configuration of PV plant with DSRF control strategy such transformer configuration. Superscripts ’+’, ’-’ and ’0’
is used while developing the method. The proposed scheme represent positive, negative and zero sequence components,
is tested for different operating conditions of PV plant and respectively. Subscripts d and q refer to direct and quadrature-
change the grid-side source impedance on a 34-bus distribution axis components, respectively.
system. Hardware-in-loop testing using OPAL-RT with IEC Under normal operation, d component of positive sequence
61850 communication protocol validates the application of reference current (Id+ ref ) regulates the DC-link voltage (VDC )
proposed method in real-time. by comparing with reference DC-link voltage (VDC ref ).
II. PV PLANT OPERATION DURING FAULT Id+ ref = KP I d (VDC ref − VDC ) (1)
Fig. 1 shows the double-stage configuration of PV plant
where KP I d is the gain of proportional-integrator (PI) com-
connected to grid [23]. A booster operating at maximum power
pensator. When a fault occurs in the PV connected distribution
point tracking (MPPT) [24] is designed to maximize the power
line, inverter injects reactive current to comply with the reac-
output from panels. The inverter tries to push this power to
tive power requirement. For this, the q component of positive
the grid even under low voltage conditions such as fault,
sequence reference current (Iq+ ref ) compares the measured
satisfying FRT requirements of Indian grid code [2]. For this,
reactive power (Qm ) with rated value (Qref ).
the controllers operating in positive and negative sequence
reference frame generate switching pulses for the inverter Iq+ ref = KP I q (Qref − Qm ) (2)

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

where
3 + + + + − − − −
Qm = (V I − VSd Iq + VSq Id − VSd Iq ) 1 -0.055
2 Sq d 0.8

X+M (pu)
R+M(pu)
0.6
-0.06
where KP I q is the gain of PI compensator. VS is voltage at 0.4
0.2
bus S and I is PV output current. These d and q components of 0 -0.065
10 10
the reference currents are passed through a saturation block to 8
6
4 30
40 8
6
4 30
40
20 20
2 10 2 10
protect the switches [3]. Similarly, negative sequence current x (km)
0 0
RF ( ) x (km)
0 0
RF ( )
references, denoted by Id− ref and Iq− ref , are generated using (a) (b)

negative sequence controller. The negative sequence voltage Fig. 3. Positive sequence impedance of PV plant measured at bus M: (a)
− − equivalent resistance and (b) equivalent reactance.
references, denoted by VSd ref and VSq ref , are taken as
0.4 -0.05
zero to maintain balanced voltage at bus S by eliminating 0.3

R-M (pu)

X-M (pu)
-0.1
the effect of the negative sequence voltage components during 0.2
-0.15
unbalanced operation [23]. 0.1
0 -0.2
The generated sequence reference currents are processed 10
8
10
6 40 40
30 30
in dual controller to generate the switching pulses for the 4
2 10
20 5
10
20
0 0 0 0
x (km) x (km)
inverter. However, PI compensator fails to compensate dq (a)
RF ( )
(b)
RF ( )

current oscillations at twice the nominal frequency during Fig. 4. Negative sequence impedance of PV plant measured at bus M: (a)
unbalanced faults. Therefore, a second-order notch filter (NF) equivalent resistance and (b) equivalent reactance.
is applied to remove double frequency component, as shown protecting the distribution line connecting PV plant. In such
in Fig. 1(a). This analysis shows the effect of current limit and a situation, distance relay may be a viable option for PV
associated control strategy on PV output current which affects connected distribution system. However, the current limitation
the source impedance of PV plant, as discussed below. in a PV plant increases the apparent impedance seen by
B. Source Impedance Variation such relay. Also, the source impedance of PV plant does
To study the impact of positive and negative sequence not remain constant unlike synchronous based distributed
controller on PV source impedances, a case is simulated generator and varies based on control action as discussed
for a solidly grounded distribution system with PV plant in Section II. Therefore, the current limitation and variation
connected at bus M. Impedance diagram of the test system in source impedance of PV plant restrict the conventional
is shown in Fig. 2. The detailed system data is given in distance relay performance for possible application to PV
Appendix B. The power flow direction is from bus M to integrated distribution system. Therefore, an adaptive distance
bus N (PV plant to the grid). EP V and Eg are the internal relay approach is proposed in this work.
voltages of PV plant and grid respectively and ZM and Zg Fig. 2 shows a distribution line MN protected by distance
are corresponding equivalent impedances. Phase-A-to-ground relay RR , where zone-1 covers up to 80% of the line section.
faults with different fault resistances (RF ) are simulated The performance of impedance and angle based distance
at different fault locations (x) and equivalent positive and relaying approaches are analyzed for balanced and unbalanced
negative sequence source impedances of PV plant at bus M faults.
+ + + − − − An impedance based fixed setting distance relay has quadri-
i.e. ZM = RM + jXM and ZM = RM + jXM respectively
are calculated. lateral characteristic. It computes the apparent impedance seen
+ −
Fig. 3 and Fig. 4 show the variation of ZM and ZM by the relay and generates a trip command when the apparent
+ −
with x and RF . RM and RM increase with x and RF as impedance lies inside the prescribed quadrilateral character-
corresponding fault current supplied by PV plant decreases istic. An angle based relay is memory polarized with mho
+ −
and approaches to the rated value. XM and XM variations are characteristic. It compares the angle difference (α) between
less compared to the resistive components, as the considered the operating quantity and polarizing quantity [26]. It can be
distribution system is resistive and exchange of reactive power expressed as,
between PV and grid is not significant. This depicts that the α = ∠Vop − ∠Vpol (3)
equivalent impedances of PV plant vary for different fault where Vpol is memory-polarized positive sequence voltage and
situations. Vop is operating voltage phasor which is obtained as in [27]
EPV M N for different types of faults. The relay generates a trio signal
Eg
IAMF F when α lies in −90◦ to 90◦ .
ZM xZL (1-x)ZL Zg Issues with conventional impedance based distance relay
IF
with RF = 0 − 40 Ω [25] and angle based distance relay are
PV RF Grid
Plant enumerated below for distribution system applications in the
RR
presence of PV plant.
Fig. 2. Impedance diagram of two bus distribution system with PV plant.
A. During balanced fault
A three-phase bolted fault is simulated at a distance of
III. I NFLUENCE OF PV PLANT ON DISTANCE PROTECTION 60% from bus M at 1.0 s to see distance relay performance.
In a PV plant, fault current is limited by the control strategy Fig. 5a shows the d and q components of positive sequence
which limits the application of conventional techniques while currents of PV plant. Before the inception of fault, the power

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

TABLE I
P ERFORMANCE OF EXISTING RELAYING APPROACHES UNDER VARIOUS FAULT CONDITIONS

Fault RF VAM F IAM F 0

IAM Vpol Vop Relaying approaches
F
types (Ω) (kV) (A) (A) (kV) (kV) Impedance based Angle based
(Zapp in Ω) (α in deg)
3.15 + j 2.10 -180◦
ABC 0 0.60∠9◦ 158.4∠ − 24.7◦ 0 23.2∠10◦ 0.6∠ − 170◦
(correct decision) (incorrect decision)
29 − j 13 -64◦
AG 10 3.25∠ − 41.39◦ 27.89∠34.1◦ 65.64∠ − 68.2◦ 23.2∠10◦ 2.6∠ − 54◦
(incorrect decision) (correct decision)

+ +
system of Fig. 2 was running with PV plant, supplying IM d the case. IM d increases to regulate the DC-link voltage and
+
of 1 p.u. During fault, the inverter operates in FRT mode IM q drops to supply reactive power to the grid without hitting
+ +
and delivers IM d to regulate the DC-link voltage and IM q their limits. Accordingly, the control strategy in the PV inverter
to comply reactive power requirement of the grid [2]. Such modulates the source impedance of PV plant with its control
current limitation results in increase in source impedance of parameters which affects the voltage and current pattern seen
PV plant. Consequently, a very low voltage and limited current by the relay. This results in leading power factor of the current
are seen by the relay at PV-side. Table I presents the voltage (whereas during fault condition it should be highly lagging for
and current seen by the relay during the fault. It is observed conventional system). Table I presents the voltage and current
that the apparent impedance seen by the relay on PV-side phasors seen by the relay during fault. It is observed that the
becomes equal to the line impedance between the relay and reactance part in the apparent impedance seen by the relay
fault point and settles on the impedance line of the trip zone changes significantly and the impedance remains outside the
(Fig. 6a). This leads to a correct decision by conventional characteristic, as shown in Fig. 8a. This leads to malfunction
distance relay with quadrilateral characteristic for three phase of the conventional impedance based approach.
fault. For this fault with fault resistance, operating voltage (Vop )
For this fault condition with negligible fault resistance, is given by,
operating voltage is given by, 0
Vop = VAM F − Zr (IAM F + K0 IAM F) (5)
Vop = VAM F − Zr IAM F (4)
where VAM F and IAM F are phase-A voltage and current
where VAM F and IAM F are phase-A voltage and current during fault. Zr is the reach impedance. K0 is zero sequence
0
during fault. Zr is the reach impedance. Such fault situation compensation factor. IAM F is zero sequence current. In such
causes very low VAM F and limited IAM F . Thus, the phase fault situation, VAM F and IAM F are influenced by the change
angle of Vop in (4) is mainly decided by the current phasor. in source impedance of PV plant with the control parameters.
From Table I, the value of ∠Vop is obtained to be -170◦ , having From Table I, the value of ∠Vop as in (5), is obtained to be
∠Vpol = 10◦ . The angular positions of Vpol and Vop are shown −54◦ , having ∠Vpol = 10◦ . The angular positions of Vpol
in Fig. 5b. The angle difference (α) in (3) is −180◦ , which and Vop are shown in Fig. 7b. The angle difference (α) in (3)
is outside the trip zone (−90◦ to 90◦ ). This indicates that the is −64◦ , which is inside the prescribed zone (Fig. 8b). This
angle (α) based approach fails to detect the bolted fault within indicates that for unbalanced fault with fault resistance, angle
its zone of protection (Fig. 6b). (α) based approach works correctly.
1.5 90°
120° 60°
Currents (pu)

90°
Currents (pu)

1.5 120° 60° 1

I+Md I+Mq 150° 30°
I+Md I+Mq 150° 30° Vpol
0 °
0
180°
°
=-180 =-64° 0
180° Vpol 0° -1
-1.5 Vop -1.5 Vop
-150° -30°
0.9 0.95 1 1.05 1.1 1.15 -150° -30° 0.9 0.95 1 1.05 1.1 1.15 -120°
°
Time (s) -120° -90° -60° Time (s) -90° -60
(a) (b)
(a) (b)
Fig. 7. (a) d and q components of positive sequence current and (b) angular
Fig. 5. (a) d and q components of positive sequence current and (b) angular
position of voltages during phase-A-to-ground fault.
position of voltages during bolted three-phase fault. 5
5
X( )

0
X( )

2.5 Zapp -5 Fixed zone-1

Zapp
Fixed zone-1 -10
0 0 10 20 30 40 50
0 20 40 60 80 R( )
R( )
(a) (b)
(a) (b) Fig. 8. Performance of distance relay RR for (a) fixed setting and (b)
Fig. 6. Performance of distance relay RR for (a) fixed setting and (b) memory-polarized during phase-A-to-ground fault.
memory-polarized during three-phase bolted fault.
From the above results, it can be concluded that both
B. During unbalanced fault impedance and angle based approaches have limited perfor-
An unbalanced phase-A-to-ground fault with 10 Ω fault mance for line protection with PV plant connected at one
resistance is simulated at 60% of line MN from bus M at end. Impedance based approach correctly identifies the bolted
+ +
1.0 s. Fig. 7a shows the current IM d and IM q of PV plant for balanced fault whereas angle based approach works reliably

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

for unbalanced faults with fault resistance. Such issues ask for A. Determination of positive sequence impedance of PV plant
an alternative scheme for correct protection decisions in the During prefault, inverter injects a constant positive se-
presence of PV plant. An adaptive distance relaying approach quence current IM +
to regulate the DC-link voltage VDC
is proposed in this work for the protection of distribution lines while maintaining a constant voltage at bus M. Therefore,
connecting PV plant. positive sequence circuit of a PV plant can be modeled by
IV. P ROPOSED M ETHOD an equivalent source, as shown in Fig. 10. EP V and ZP+V are
internal voltage and its positive sequence source impedance,
The prefault relation between the internal voltage of PV respectively. Zf+ and Ztr+
are positive sequence impedance of
plant and grid-side voltage for the system of Fig. 2 can be, filter and transformer, respectively.
EP V = ρe−jδ Eg (6) The voltage relation for the circuit (Fig. 10) is
+ + +
VM = EP V − IM ZM (9)
where ρ and δ are the magnitude and phase of the ratio of
internal voltages of PV to grid respectively. The apparent + +
where VM and IM are positive sequence voltage and current
impedance seen by the distance relay at bus M is given by components of prefault phasors available at the relay bus.
(refer to Appendix A), +
ZM = ZP+V + Zf+ + Ztr
+
(10)
Zapp = xZL+ + ∆Z (7)
The parallel impedance of the inverter filter 1/(jωCf ) is
where x is the distance of fault point from relay bus M. It ignored being substantially larger [17]. Separating real and
varies depending on the fault location from the relay. ZL+ is imaginary parts and rearranging, (9) can be rewritten as
positive sequence impedance of line MN and ∆Z is the error 
Re[EP V ]

term due to the fault resistance and is given by  +
  + +

Re[VM ] 1 0 −Re[IM ] Im[IM ] Im[E+P V ]

3RF + = + + 
∆Z = (8) Im[VM 0 1 −Im[IM ] −Re[IM RM 
D + D− + D0 (1 + K0 ) ωL+ M
(11)
where K0 is the zero sequence compensation factor, D− and where RM +
and L+ are the real and imaginary parts of Z +
M M.
D0 are the negative and zero sequence current distribution The simplified form of (11) can be
+ +
factors respectively and D is a function of (IAM F , EP V , ZM ,
−
ZM ) as depicted in (32) Appendix A. [V ] = [I] [B] (12)
+ + −
In a PV plant, IAM F , EP V , ZM and ZM depend on inverter where [V] and [I] are corresponding voltage and current ma-
current constraints and control strategies. These are affected
trices. Matrix [B] contains internal voltages and impedances.
by changes in PV capacity connected to the relay bus and irra-
The unknown parameters in [B] can be computed from the
diation level. With the change in irradiation level, the current
relation,
supplied by the PV plant changes whereas controller tries to −1
[B] = [(I 0 I) I 0 ] [V ] (13)
maintain the rated bus voltage. The decrement in irradiation
will increase the source impedance of PV plant and vice On solving (13), real and imaginary components of EP V
versa. Thus, the trip boundary should be updated adaptively to +
and ZM can be obtained. With the change in PV operating
avoid malfunctioning of the relay with the change in PV plant capacity, positive sequence internal voltage and corresponding
condition. To set the boundary adaptively, prefault voltage impedance of PV plant can be adjusted adaptively.
and current phasors are estimated at the relaying point by + ωL+
RPV PV R+f ωL+f R+tr ωL+tr +
IM
full-cycle discrete Fourier transform. Using the prefault data,
source impedances of PV plant are calculated. Subsequently,
the setting impedance for a trip boundary at a system condition EPV +
ZPV Zf+ Ztr+ +
ZM
+
VM
is calculated and saved for relaying purpose. At the sampling 1/ωCf
frequency of 1200 Hz, the impedance is updated for every PV
new pair of voltage and current phasors. In the following, the Fig. 10. Positive sequence circuit diagram of a PV plant during prefault.
methods to determine different factors as in (8) are given.
B. Determination of negative sequence impedance of PV plant
+
+ F IAF
IAMF As discussed in Section II, the negative sequence impedance
+ + +
ZPV Zf Ztr xZ +
L (1-x)ZL Zg+ + of PV plant depends on negative sequence controller. Thus, the
EPV + + Eg
ZM VAMF voltage equation of the controller can be written as
-
IAMF − − − − − − −
- -
Ztr- 3RF VSd = −(RP V + Rf )Id − ω(LP V + Lf )Iq (14)
ZPV Zf xZL- (1-x)ZL Zg
- -

Z-M
-
− − − − − − −
0
IAMF VSq = −(RP V + Rf )Iq + ω(LP V + Lf )Id (15)
0 0
ZPV Zf Ztr0 xZL
0
(1-x)ZL Zg
0 0
The voltage equation, referring to transformer at bus M side,
can be written as
Fig. 9. Sequence network of the system for phase-A-to-ground fault with − − − − − −
fault resistance. VM d = VSd − Rtr IM d − ω(Ltr )IM q (16)

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

− − − − − −
VM q = VSq − Rtr IM q + ω(Ltr )IM d (17)
Distribution system
Neglecting the parallel impedance of inverter filter, = IM−
d Id−
−
and Iq− = IM q . Substituting (14) in (16) and (15) in (17), we Obtaining voltage and current at relay
get (18) and (19) respectively.
Calculation of positive and negative sequence
− − − − − − − − −
VM d = −(RP V + Rf + Rtr )IM d − ω(LP V + Lf + Ltr )IM q impedances of PV plant using (11) and (18)
(18)
−
VM = −(R −
+ R −
+ R − −
)I + ω(L −
+ L−
+ L− −
)I + using (22)
Calculation of IAF
q PV f tr M q PV f tr M d
(19)
− is
On solving (18) and (19) and substituting ZM = ZP−V +Zf− + Yes
D+IAF
+ +I >I + No
− − M M limit
Ztr , ZM can be calculated as
I+ +
AMF = IM limit
? + = D+ I+
IAMF AF + IM

− V−
ZM =− M (20)
−
IM Calculation of factors D-, Do, D
− − − − − −
where VM = VM d −jVM q and IM = IM d −jIM q
are negative Trip boundary generation using (5)
sequence voltage and current in dq frame at bus M. Zf− and Fig. 11. Flow diagram for trip boundary generation by proposed method.
−
Ztr are negative sequence impedance of filter and transformer,
respectively. Thus, for prefault unbalanced loading condition, phasors are used to calculate the positive and negative source
negative sequence impedance of PV source can be obtained at impedances of PV plant using (13) and (20). The obtained
zero internal voltage. PV source impedance is used to calculate the sequence fault
current through fault point (24). The positive sequence fault
C. Calculation of positive sequence fault current supplied by current supplied by PV plant is calculated. Subsequently, the
PV plant for boundary setting factors D0 , D− and D are calculated. The obtained source
Fig. 1 provides the positive sequence controller of PV plant. impedances and factors are used to calculate the setting
Following fault inception, the inverter injects d component impedance using (7) for adaptive trip boundary.
of positive sequence current to regulate the inverter DC-link
V. R ESULTS
voltage and q component of positive sequence current to
comply with the reactive power requirement of the grid. There- Proposed method is tested for 24.9 kV, 60 Hz, 34-bus distri-
fore, maximum possible positive sequence current IM + bution system, having unbalanced loading and unbalanced line
limit
supplied by the PV plant can be expressed as phase impedance [28]. 1 MW PV plant is connected at bus 32
q with relay RR on PV-side, as shown in Fig. 12. After passing
+ + 2 + 2
IM limit = (IM d limit ) + (IM q limit ) (21) the voltages and currents through anti-aliasing filter, phasors
where IM + + are estimated with full-cycle discrete Fourier transform. The
d limit and IM q limit are d and q components of
+ performances of fixed quadrilateral characteristic and proposed
IM limit . During fault, the positive sequence current supplied
adaptive setting of distance relay, covering 80% of line 31-
by the PV plant can be computed using current division rule
32 from bus 32 with 40 Ω reach are compared for phase-
for a constant PV source impedance. But Fig. 3 shows that
A-to-ground faults in zone-1. The results demonstrate the
source impedance of PV plant changes with fault conditions
performance of the proposed scheme in identifying zone-1
and the dq components of the inverter current may reach their
faults correctly for different control strategies and operating
limits. Thus, the positive sequence fault current seen by relay
capacities of PV plant and variation in grid source impedance.
at bus M, considering the current limits, will be
+ + + +
 To have the equivalence with the system of Fig. 2 as analyzed
+ IM limit IM limit < (D IAF + IM )
IAM = + earlier, bus 32 is considered as M and bus 31 as N.
F D+ IAF + IM Otherwise
29 30 31(N) 32(M)
(22) Grid
where IM is prefault current measured at the relaying point. 1
+
IAF and D+ can be calculated using (24) and (33) respec- 2
RR PV

tively. With the change in size of PV plant, the current limit 28 27 26 25 24 23 22 33

3 34
value should be changed accordingly.
The maximum fault resistance coverage is determined by 4

using the relation in [25]. For minimum current of 1.2 times

of rated value and arc length of 100 mm in open air, the 5 6 7 8 9 10 11 12 13 19 20 21
maximum fault resistance coverage of 40 Ω is considered for
the 24.9 kV system. Following the above mentioned approach, 18 17 16 15 14
the adaptive boundary of a distance relay can be set accurately Fig. 12. The 34-bus distribution system with PV plant connected at bus 32.
for phase-A-to-ground fault. Similar approach can be extended
to other types of faults. Case-A: Performance of the proposed method for different
The flow diagram of proposed method is provided in control strategies
Fig. 11. The phasors are estimated from prefault voltage Different control strategies are used in PV inverter for
and current samples available at relay point. The estimated managing distribution systems. PSRF controller [16] supplies

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

only positive sequence current even during a fault whereas due to intermittency. Such a situation may affect the distance
DSRF controller [23] manages both positive and negative relay protecting the distribution line connecting the PV plant.
sequence currents. Such differences in currents affect the Two cases are simulated for PV plant with DSRF control
apparent impedance seen by the relay. Cases are simulated strategies at the same system loading condition. In case-1,
for PSRF and DSRF control strategies in 1 MW PV plant, the operating capacity is 1 MW at 1000 W/m2 irradiance
operating at the same system loading condition. Table II level, whereas in case-2, the operating capacity is reduced
provides the influence of control strategies on PV source to 200 kW at 200 W/m2 irradiance level and thereby the
+ −
impedances ZM and ZM . It is observed that with PSRF control PV source impedance becomes significantly high (Table III).
−
strategy, ZM is significantly higher than DSRF control strategy Fig. 14 shows the advantage of proposed method, where, the
in PV inverter and thereby the ideal trip boundary should be trip boundary with reduced generation is substantially larger
modified adaptively to prevent maloperation of the distance than the rated capacity, shown in Fig. 13b.
relay. In such a situation, proposed method modifies the Phase-A-to-ground faults with 30 Ω resistance are created
boundary adaptively without knowing the control strategies, in line 31-32 at a distance of 60% from bus 32 for both 1 MW
as shown in Fig. 13a and Fig. 13b. and 200 kW operating capacities. The result for 200 kW PV
Phase-A-to-ground faults with 10 Ω and 30 Ω fault resis- capacity is shown in Fig. 14. It is observed that the apparent
tances are created at 60% of the line 31-32 from bus 32 for impedance settles inside the adaptively set boundary and is
both PSRF and DSRF control strategies in PV inverter. The outside the fixed setting. Similar result is obtained for 1 MW
result for PSRF control strategy is presented in Fig. 13a. It PV capacity, as shown in Fig. 13b. It shows the application
shows that the conventional approach is able to identify 10 of proposed method for line connecting PV plant where the
Ω fault resistance but fails in case of 30 Ω fault resistance operating capacities of PV plant may vary.
because negative sequence current is negligible in PSRF
strategy and reach of the relay depends on the magnitude TABLE III
C OMPARISON OF SYSTEM PARAMETERS FOR C ASE -B
and angle of the local positive sequence current. On the other
hand, proposed method correctly identifies the fault with the Operating capacity
+ −
adaptive boundary. Similar result is obtained for PV plant of PV plant ZM (Ω) ZM (Ω) ρ δ
with DSRF control, as shown in Fig. 13b. It confirms that the 1 MW 347∠ − 139.5◦ 130 ∠135◦ 0.984 0.04◦
proposed method identifies the fault in the prescribed zone and 200 kW 971∠ − 100◦ 1136∠38.92◦ 0.99 -0.4◦
is accurate even for different control options in PV inverter.
TABLE II 5
C OMPARISON IN PV SOURCE IMPEDANCES FOR C ASE -A
+ −
Control strategy ZM (Ω) ZM (Ω) 0
X( )

PSRF 331∠ − 151.72◦ 21956∠ − 63◦

DSRF 347∠ − 139.5◦ 130 ∠135◦ -5 Fixed zone-1 RF=30
Adaptive zone-1(200 kW, DSRF)
5 0 20 60 40 80 100
RF=10
R( )
Fig. 14. Performance of proposed adaptive method for 200 kW operating
X( )

capacity of PV plant.
0
Case-C: Performance of the proposed method for variation in
Fixed zone-1
Adaptive zone-1 (1MW, PSRF) RF=30 grid source impedance
-5
0 10 20 30 40 50 60 As in Appendix A, the denominator of the apparent
R( )
impedance Zapp in (31) depends on factors D, D0 and D−
(a) which are functions of grid source impedance Zg , as in (32)
5
and (27) respectively. For normal value of Zg (Appendix B)
0
RF=10 and 10 Ω fault resistance in the middle of line, the denominator
X( )

-5 RF=30 of (31) becomes 2.41∠14.67◦ and for two times of the normal
-10
value of Zg , it is 2.42∠16.9◦ , considering 1 MW capacity
Fixed zone-1
Adaptive zone-1 (1 MW, DSRF) of PV plant with DSRF control strategy. This indicates that
-15 for possible variation of Zg , Zapp does not change much. In
0 10 20 30 40 50 60 70
R( ) the proposed approach, therefore, Zg is considered as a fixed
(b) value. For the two values of Zg corresponding trip boundaries
Fig. 13. Performance of proposed adaptive method for (a) PSRF control are plotted and shown in Fig. 15. The plots show negligible
strategy and (b) DSRF control strategy in PV inverter.
difference for the two values of grid source impedances which
is in agreement with [20].
Case-B: Performance of the proposed method for different
operating capacities of PV plant VI. H ARDWARE - IN - LOOP VALIDATION
A PV plant consists of several units connected in parallel. The proposed method is validated using OP4510 OPAL-
At an instant, the operating capacity of PV plant may reduce RT real-time simulator [29]. The experimental setup for real-

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

5 delay and the latency encountered by the IEC 61850 protocol

0 was 32.7 ms. In actual practice, by incorporating proposed
algorithm using local measurements, it will take 19.3 ms to
X( )

-5
Fixed zone-1 generate a trip signal. On the contrary, fixed setting approach
Adaptive zone-1 (Z g)
-10 fails to identify the fault within its own boundary. This real-
Adaptive zone-1 (2 Z g)
-15 time hardware-in-loop result clearly shows the advantage of
0 10 20
40 50 30 60 70 proposed adaptive setting compared to fixed setting distance
R( )
Fig. 15. Trip boundaries for change in grid side source impedance. relay while protecting a distribution line connected to PV
plant.
time simulation is shown in Fig. 16. For this purpose, the
power system model of Fig. 12 is simulated in simulator-1 VII. C ONCLUSION
and proposed adaptive distance relaying algorithm is realized
in simulator-2. The simulated three-phase voltage and current For satisfying FRT requirement, PV plant remains con-
data as sample values (SV) are transferred from simulator- nected to the grid during a fault. The voltage and current
1 to simulator-2 through IEC 61850 communication protocol in a line connecting to the grid are influenced by the fault
[30]. In case of a fault in the zone-1, the proposed relay behaviour of PV plant. It causes protection challenges to
logic implemented in simulator-2 generates a trip signal and conventional distance relaying scheme protecting such lines.
a generic object oriented substation event (GOOSE) message This work considers dual control scheme for PV plant with
is sent back to the simulator-1 that opens the circuit breaker FRT ability. It is seen that inverter current limitation and PV
in the power system model. source impedance variation during fault are two important
reasons for incorrect operation of the distance relay on PV-
side of the connecting line. An adaptive setting is proposed
for reliable operation of distance relay. For this, the sequence
impedances of PV plant are calculated using prefault data. A
comparative analysis of proposed adaptive setting for zone-
1 with conventional fixed setting reveals the advantage of
the proposed method for variations in control strategies and
operating capacities of PV plant and change in grid source
impedance. Result with hardware-in-loop testing also confirms
the strength of the proposed method for possible use in
distribution systems.
Fig. 16. Experimental setup for real-time simulation.
A PPENDIX A
120
IM with proposed adaptive approach 1 The prefault voltage, VM at F (Fig. 2) can be expressed as,
IM with fixed setting approach
Trip signal

80 Trip signal
VM = EP V − IM (xZL+ + ZM
+
IM (A)

No trip signal
following ) (23)
0.5
40 CB opening
trip signal initiation by
32.79 A proposed method with For phase-A-to-ground fault, the sequence fault currents in
Fault inception communication latency
0 0 fault resistance RF becomes
0.95 1 1.052 1.061 1.1 VM
+ − 0
Time (s) IAF = IAF = IAF = (24)
Fig. 17. Phase-A current at bus 32 and trip signal generated in hardware-in- Zsum + 3RF
loop testing.
At 1 MW operating capacity of PV plant with DSRF control where Zsum is equivalent impedance seen by relay at bus M.
strategy, the system is simulated for same loading condition +
(ZM + xZL+ )(Zg+ + (1 − x)ZL+ )
and accordingly the adaptive setting was adjusted in simulator- Zsum = +
2. Phase-A-to-ground fault with 10 Ω fault resistance is ZM + ZL+ + Zg+
−
simulated at 60% of the line 31-32 from bus 32 at 1.0 s. (ZM + xZL− )(Zg− + (1 − x)ZL− )
+ −
(25)
Fig. 17 presents phase-A current magnitude at bus 32 and ZM + ZL− + Zg−
corresponding trip command generated. At t=1.0 s, the current (xZL0 + Ztr
0
)(Zg0 + (1 − x)ZL0 )
starts increasing, resulting in change in status of timer. For the +
ZL0 + Zg0 + Ztr0
proposed method, the timer reaches 4 counts with apparent
impedance inside the adaptive trip boundary and publishes a The sequence currents during fault from relay end M will be
trip signal as a GOOSE message at t = 1.052 s that opens the
− − − 0 0 0
circuit breaker at t=1.061 s to disconnect the PV plant from IAM F = D IAF and IAM F = D IAF (26)
the system. The time difference between the fault inception
where
in simulator-1 and trip signal received by simulator-1 is 52
ms in which phasor computation using full-cycle discrete Zg− + (1 − x)ZL− Zg0 + (1 − x)ZL0
Fourier transform took 16 ms, counter introduced 3.3 ms D− = and D0 = (27)
Zg− + ZL− + ZM
−
Zg0 + ZL0 + Ztr
0

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

Using phase-A voltage and current at relay location during [9] S. A. Lavand, and S. A. Soman, “Predictive analytic to supervise zone
fault VAM F and IAM F , the apparent impedance seen by the 1 of distance relay using synchrophasors,” IEEE Trans. Power Del.,
vol. 31, no. 4, pp. 1844–1854, Aug. 2016.
distance relay at bus M can be expressed as [10] M. Nagpal, and C. Henville, “Impact of power-electronic sources on
VAM F transmission line ground fault protection,” IEEE Trans. Power Del.,
Zapp = 0 (28) vol. 33, no. 1, pp. 62–70, Feb. 2018.
IAM F + K0 IAM F [11] A. Hooshyar, M. A. Azzouz, and E. F. El-Saadany, “Distance protection
of lines emanating from full-scale converter-interfaced renewable energy
where power plants - part I: Problem statement,” IEEE Trans. Power Del.,
ZL0 − ZL+ vol. 30, no. 4, pp. 1770–1780, Aug. 2015.
K0 = (29)
ZL+ [12] Y. Fang, K. Jia, Z. Yang, Y. Li, and T. Bi, “Impact of inverter-
interfaced renewable energy generators on distance protection and an
IAM F is phase-A fault current seen by the relay and can be improved scheme,” IEEE Trans. Ind. Electron., vol. 66, no. 6, pp. 7078–
obtained as 7088, Sep. 2019.
[13] J. Jia, G. Yang, A. H. Nielsen, and P. Rnne-Hansen, “Impact of
+ − 0 VSC control strategies and incorporation of synchronous condensers on
IAM F = IAM F + I0AM F + IAM F (30)
distance protection under unbalanced faults,” IEEE Trans. Ind. Electron.,
vol. 66, no. 2, pp. 1108–1118, Feb. 2019.
Substituting sequence components and prefault voltages in
[14] A. M. Tsimtsios, and V. C. Nikolaidis, “Setting zero-sequence compen-
(28), Zapp can be expressed as sation factor in distance relays protecting distribution systems,” IEEE
Trans. Power Del., vol. 33, no. 3, pp. 1236–1246, June 2018.
3RF
Zapp = xZL+ + (31) [15] K. Jia, C. Gu, Z. Xuan, L. Li, and Y. Lin, “Fault characteristics analysis
D+ D− + D0 (1 + K0 ) and line protection design within a large-scale photovoltaic power plant,”
IEEE Trans. Smart Grid, vol. 9, no. 5, pp. 4099–4108, Sep. 2018.
where + [16] A. Hooshyar, M. A. Azzouz, and E. F. El-Saadany, “Distance protection
IAM F (Zsum + 3RF ) of lines emanating from full-scale converter-interfaced renewable energy
D= (32)
Z + +xZ + power plants-part II: Solution description and evaluation,” IEEE Trans.
EP V (D+ + ρe−jδ Z + M L
+Z + +Z +
) Power Del., vol. 30, no. 4, pp. 1781–1791, Aug. 2015.
M L g
[17] M. A. Azzouz, A. Hooshyar, and E. F. El-Saadany, “Resilience en-
+
The current distribution factor D involved in (32) and (22) hancement of microgrids with inverter-interfaced DGs by enabling faulty
is defined as phase selection,” IEEE Trans. Smart Grid, vol. 9, no. 6, pp. 6578–6589,
Zg+ + (1 − x)ZL+ Nov. 2018.
D+ = + (33) [18] D. Pal, B. Mallikarjuna, R. J. Reddy, M. J. B. Reddy, and D. K. Mohanta,
ZM + ZL+ + Zg+ “Synchrophasor assisted adaptive relaying methodology to prevent zone-
3 mal-operation during load encroachment,” IEEE Sensors J., vol. 17,
A PPENDIX B no. 23, pp. 7713–7722, Dec. 2017.
The following data are for 24.9 kV, 60 Hz power system. [19] Y. Q. Xia, K. K. Li, and A. K. David, “Adaptive relay setting for
standalone digital distance protection,” IEEE Trans. Power Del., vol. 9,
Grid impedances: Zg+ =Zg− =125∠60◦ Ω, Zg0 =160∠62◦ Ω. no. 1, pp. 480– 491, Jan. 1994.
Line impedances: ZL+ =ZL− =9.5∠34◦ Ω, ZL0 =20.9∠55◦ Ω. [20] A. K. Pradhan and G. Jos, “Adaptive distance relay setting for lines
connecting wind farms,” IEEE Trans. Energy Convers., vol. 22, no. 1,
ACKNOWLEDGEMENT pp. 206–213, March 2007.
The authors are thankful to DST, India and EPSRC, UK for [21] R. Dubey, S. R. Samantaray, and B. K. Panigrahi, “Adaptive distance
protection scheme for shunt-facts compensated line connecting wind
sponsoring the Indo-UK joint project, UK-India Clean Energy farm,” IET Gener. Transmiss. Distrib., vol. 10, no. 1, pp. 247–256, 2016.
Research Institute (UKICERI)- DST/RCU/JVCCE/2015/02 [22] M. Singh, and V. Telukunta, “Adaptive distance relaying scheme to
(C), through which the research was conducted. tackle the underreach problem due renewable energy,” in Proc. IEEE
18th National Power System Conf. (NPSC), Dec. 2014.
R EFERENCES [23] P. Mishra, A. K. Pradhan, and P. Bajpai, “Voltage control of PV inverter
connected to unbalanced distribution system,” IET Renew. Power Gener.,
[1] R. Yan, T. K. Saha, P. Meredith, A. Ananth, and M. I. Hossain, vol. 13, no. 9, pp. 1587–1594, July 2019.
“Megawatt-scale solar variability study: An experience from a 1.2 MWp [24] D. Das, S. Madichetty, B. Singh, and S. Mishra, “Luenberger ob-
photovoltaic system in Australia over three years,” IET Renew. Power server based current estimated boost converter for PV maximum power
Gener., vol. 10, no. 8, pp. 1229–1236, 2016. extraction-a current sensorless approach,” IEEE J. Photovol., vol. 9,
[2] Central Electricity Regulatory Commission, “Indian electricity no. 1, pp. 278–286, Jan 2019.
grid code,” Technical Report, 2010. [Online]. Available: [25] V. V. Terzija, and H. J. Koglin, “On the modeling of long arc in still air
http://cercind.gov.in/2010/ORDER/February2010/IEGC Review Propo and arc resistance calculation,” IEEE Trans. Power Del., vol. 19, no. 3,
sal.pdf pp. 1012–1017, July 2004.
[3] A. K. Singh, I. Hussain, and B. Singh, “Double-stage three-phase grid- [26] K. El-Arroudi, and G. Jos, “Performance of interconnection protection
integrated solar PV system with fast zero attracting normalized least based on distance relaying for wind power distributed generation,” IEEE
mean fourth based adaptive control,” IEEE Trans. Ind. Electron., vol. 65, Trans. Power Del., vol. 33, no. 2, pp. 620–629, April 2018.
no. 5, pp. 3921–3931, May 2018. [27] F. Calero, and H. J. Altuve, “Identifying the proper impedance plane and
[4] H. Hooshyar, and M. E. Baran, “Fault analysis on distribution feeders fault trajectories in distance protection analysis,” in 66th Ann. Georgia
with high penetration of PV systems,” IEEE Trans. Power Syst., vol. 28, Tech Protective Relaying Conf., April 2012.
no. 5, pp. 2890–2896, Aug. 2013. [28] K. P. Schneider, B. A. Mather, B. C. Pal, C. W. Ten, G. J. Shirek,
[5] V. Telukunta, J. Pradhan, A. Agrawal, M. Singh, and S. G. Srivani, H. Zhu, J. C. Fuller, J. L. R. Pereira, L. F. Ochoa, L. R. de Araujo,
“Protection challenges under bulk penetration of renewable energy R. C. Dugan, S. Matthias, S. Paudyal, T. E. McDermott, and W. Kersting
resources in power systems: A review,” CSEE J. Power Energy Syst., , “Analytic considerations and design basis for the IEEE distribution test
vol. 3, no. 4, pp. 365–379, Dec. 2017. feeders,” IEEE Trans. Power Syst., vol. 33, no.3, pp. 3181–3188, May
[6] O. Janke, “Directional reactive power undervoltage protection a protec- 2018.
tion concept for connecting decentralized renewable energy sources,” in [29] P. Mishra, A. K. Pradhan, and P. Bajpai, “Adaptive relay setting for
Proc. 11th IET Int. Conf. Dev. Power Syst. Prot. (DPSP), April 2012. protection of distribution system with solar PV,” in Proc. IEEE 20th
[7] D. Ventruella, and P. Steciuk, “A second look at generator 51-V relays,” National Power System Conf. (NPSC), Dec. 2018.
IEEE Trans. Ind. Appl., vol. 33, no. 3, pp. 848–856, May 1997. [30] X. Jin, R. Gokaraju, R. Wierckx, and O. Nayak, “High speed digital
[8] J. Gers, and E. J. Holmes, Protection of electricity distribution networks, relaying scheme using FPGA and IEC 61850,” IEEE Trans. Smart Grid,
2nd edition, Jan. 2011. vol. 9, no. 5, pp. 4383–4393, Sep. 2018.

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIE.2020.2975462, IEEE
Transactions on Industrial Electronics

B IOGRAPHIES Prabodh Bajpai (SM’16) is currently an Associate

Professor with the Department of Electrical Engi-
neering, Indian Institute of Technology Kharagpur,
India. His research interests include power system
restructuring, renewable energy systems, and power
Priyanka Mishra is currently working toward her system optimization.
Ph.D. in Energy Science and Engineering at Indian
Institute of Technology Kharagpur, India. Her re-
search interests include renewable energy integra-
tion, protection and control of distribution system.

Ashok Kumar Pradhan (M’94–SM’10) is currently

a Professor with the Department of Electrical Engi-
neering, Indian Institute of Technology Kharagpur,
India. He is a Fellow of Indian National Academy
of Engineering (INAE) and National Academy of
Sciences (NASI), Allahabad, India. His research
interests include Power System Relaying and Mon-
itoring.

0278-0046 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Canberra. Downloaded on May 03,2020 at 15:21:52 UTC from IEEE Xplore. Restrictions apply.