0% found this document useful (0 votes)
10 views

A Reliable Accelerated Protection Scheme For Converter-Dominated Power Networks

Uploaded by

Tevin James yiga
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
10 views

A Reliable Accelerated Protection Scheme For Converter-Dominated Power Networks

Uploaded by

Tevin James yiga
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 60, NO.

4, JULY/AUGUST 2024 5599

A Reliable Accelerated Protection Scheme for


Converter-Dominated Power Networks
Subhadeep Paladhi , Member, IEEE, Qiteng Hong , Senior Member, IEEE, and Campbell Booth , Member, IEEE

Abstract—Rising penetration of converter-based sources with converters to ensure reliable power system operation [3]. Con-
different control operations is introducing severe non-homogeneity verters controlled with grid-forming techniques are increasing
in the power systems, especially during faults. Such non- to maintain stable grid operation with high renewable pene-
homogeneity leads the decision derived by available distance and
directional relays to be unreliable. Performances of the avail- tration [3], [4]. Different grid code requirements also enforce
able communication-assisted tripping schemes being dependent on the control operations to adjust accordingly [5]. Diversity in
these two relay decisions at both ends become a concern for the pro- control schemes compels converter-interfaced renewable en-
tection of converter-dominated power networks. This work demon- ergy sources (CIRES) to respond differently compared to con-
strates the impact of converter-based sources on both distance and ventional synchronous generators during fault. This modulates
directional relays and the possible maloperation of the accelerated
protection schemes. A new transfer trip scheme has been proposed source impedances differently and results in a non-homogeneous
mitigating the issue. Considering the homogeneity present in the situation in the grid [6]. Generation variability and intermittency
negative and zero sequence networks in high voltage transmission associated with renewable sources intensify such complexity by
system even with converter-based sources, two indices are derived varying the source-impedance-ratio (SIR) dynamically [7]. Such
using local voltage and current data to identify the fault direction at situations impel to revisit the performance of available protec-
each end of a protected line. The decisions are transferred to the al-
ternative ends through low-bandwidth communication channels to tions schemes for the power networks dominated by converter-
ensure the protection decision derived for the line to be dependable based sources. This article is an improved and expanded version
as well as secured. The scheme is tested for a 9-bus and a 39-bus of the paper presented at the 2022 22nd National Power Systems
system, even with 100% converter-based sources using PSCAD/ Conference [1].
EMTDC simulation platform and found to be reliable for different
faults and system conditions. Comparative assessment with a few
advanced techniques demonstrates the superiority of the proposed
method. B. Literature Review
Index Terms—Accelerated protection, converter-interfaced With such variable SIR situations and increasing non-
renewable sources, directional relaying, distance relaying, power homogeneity in power systems, available local data-based pro-
system faults, transfer trip scheme. tection schemes fail to derive correct decisions at times [6], [8],
[9]. Line differential scheme employing both end current data
I. INTRODUCTION communicated through dedicated communication channels are
A. Motivation and Incitement recommended for such a situation [10]. Cost associated with
the required high-bandwidth dedicated communication channels
MBITIOUS decarbonization target are enforcing power
A grids for large-scale integration of renewable energy
sources [1], [2]. Integration of such sources necessitates nu-
and the communication latency are two major concerns for the
wide-application of such a scheme for protection of transmission
networks. Limited performance of current differential relays
merous control functions to be employed in the interfacing in renewable-connected lines also discourages for such a high
investment [11]. Therefore, the transfer trip schemes requiring
Manuscript received 1 October 2023; revised 4 January 2024 and 15 March
2024; accepted 1 April 2024. Date of publication 7 May 2024; date of current low-bandwidth communication channel are preferred [12]. Such
version 22 July 2024. Paper 2023-PSEC-1171.R2, presented at the 2022 22nd schemes communicate the trip decision derived by local distance
National Power Systems Conference (NPSC), New Delhi, India, Dec. 17–19, and or directional units to the other end for ensuring secure and
approved for publication in the IEEE TRANSACTIONS ON INDUSTRY APPLICA-
TIONS by the Power Systems Engineering Committee of the IEEE Industry Ap- dependable accelerated protection for the lines. High and vari-
plications Society [DOI: 10.1109/NPSC57038.2022.10069232]. This work was able source impedance situations cause severe underreach and
supported in part by the Engineering and Physical Sciences Research Council, overreach issues with distance relays [13]. Different techniques
U.K., under Grant EP/T021829/1, and in part by IIT Indore, India, through Young
Research Seed Grant Scheme under Grant IITI/YFRSG/2023-24/Phase-III/05. are applied to enhance its performance in such situations, like
(Corresponding author: Subhadeep Paladhi.) multiple setting based approach [14], [15], [16], adaptive trip
Subhadeep Paladhi is with the Department of Electrical Engineering, In- boundary setting [17], [18], application of data-driven tech-
dian Institute of Technology Indore, Indore 453552, India (e-mail: spaladhi@
iiti.ac.in). niques [19], [20], control-based adjustment [21], fault loca-
Qiteng Hong and Campbell Booth are with the Department of Electronic and tion mapping [8], and also by introducing intentional delay
Electrical Engineering, University of Strathclyde, G1 1XQ Glasgow, U.K. in decision [22]. Most of the techniques consider either the
Color versions of one or more figures in this article are available at
https://doi.org/10.1109/TIA.2024.3397640. equivalent impedance associated with different sources to be
Digital Object Identifier 10.1109/TIA.2024.3397640 negligible or the system to be homogeneous. The techniques
0093-9994 © 2024 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See https://www.ieee.org/publications/rights/index.html for more information.

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
5600 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 60, NO. 4, JULY/AUGUST 2024

applied for converter-based source connecting lines (like in [8])


are derived considering the grid connected at the remote end of
the line to be strong and dominated by conventional synchronous
generator based sources. None of these assumptions is true for
converter-dominated power networks. Angular relation between
voltage and current is also found to be inconsistent for faults in
such a network resulting in maloperation of directional relays at
times [9]. Such issues with the available distance and directional
relaying impel to seek for a new technique to ensure correct
performance of communication assisted tripping schemes for
CIRES dominated power systems.

C. Contribution
In this work, the limited performance of available communi-
cation assisted tripping schemes are demonstrated for converter-
dominated power systems and a novel scheme is proposed
mitigating the issue. A new criteria is defined using local voltage
and current data to identify the fault direction in power net-
works in the presence of converter-based sources. The proposed
scheme uses a low-bandwidth dedicated communication channel
to transfer decisions derived at both ends and issues a trip
command when the fault is detected in forward direction at both
ends of the line. The scheme is tested for a modified WSCC
9-bus system with 100% converter-based sources and a CIRES
integrated 39-bus system using PSCAD/ EMTDC simulation
data. The method is found to be accurate and independent of
Fig. 1. Overview of common communication assisted transfer trip schemes.
fault resistances, fault locations and different control operations
associated with converters-based sources.

II. PROBLEM STATEMENT


This section first presents an overview of the available com-
munication assisted transfer trip schemes used for accelerated
protection of transmission networks and later demonstrates the
limited performance of those schemes for a converter-dominated
power system. Fig. 1(a) represents a two-bus equivalent power
network, where the tripping zones for each end distance relay are
shown in Fig. 1(b). The operating principle of four commonly
used tripping schemes are described below.
r Direct Underreach Transfer Tripping (DUTT): DUTT Fig. 2. Modified WSCC 9-bus system with 100% CIRESs.

scheme, as shown in Fig. 1(c) issues a trip signal for both


end circuit breakers when the fault is detected in Zone-1
by any of the relays (RM and RN ) [12]. Zone-2 by the local end relay and the remote end relay de-
r Permissive Underreach Transfer Tripping (PUTT): PUTT tects the fault outside of zone-3, as shown in Fig. 1(f) [12].
scheme employs Zone-1 decision at any end to trip the Now, the performances of the above mentioned schemes are
local breaker immediately and sends the decision to remote tested for a 230 kV, 60 Hz modified WSCC 9-bus system with
end as a permissive trip command. The permissive signal 100% converter-based sources, as shown in Fig. 2 [23], using
confirm the tripping of remote end breaker only when the PSCAD/ EMTDC simulation platform. The solar plant con-
corresponding relay finds the fault in Zone-2 using local nected at bus 2 is integrated to the grid through grid-following
data, as shown in Fig. 1(d) [12]. converter, whereas the solar plants at bus 1 and 3 are integrated
r Permissive Overreach Transfer Tripping (POTT): POTT through grid-forming converters. The grid-following converter
scheme (shown in Fig. 1(e)) issues a trip signal for both is controlled with balanced current controller, whereas the grid-
end circuit breakers when the fault is detected in Zone-2 forming converters are designed with dual-current controller
by both end relays [12]. mimicking synchronous generator negative sequence impedance
r Directional Comparison Blocking (DCB): In this scheme, angle characteristics [24]. Line 2–7 is considered here as the
a trip command is generated when the fault is detected in protected line and the performance of relay R2 and R7 are

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
PALADHI et al.: RELIABLE ACCELERATED PROTECTION SCHEME FOR CONVERTER-DOMINATED POWER NETWORKS 5601

Fig. 3. Performance of both end distance relays leading to failure of available


accelerated protection schemes.

Fig. 5. Sequence networks of the system in Fig. 1(a) for AG faults.

identifies the fault in forward direction when ΔV1 lags ΔI1


i.e. the angle difference between ΔV1 and ΔI1 is negative.
Results in Fig. 4 show that the relay R7 finds the fault in
Fig. 4. Performance of both end directional relays leading to failure of avail-
able accelerated protection schemes. forward direction correctly, whereas the relay R2 detects the
fault in reverse direction. This results in maloperation of PUTT,
POTT and DCB schemes, even when the directional principle
tested for the purpose with both distance and directional relaying is applied for the relays. Thus, there is a need for a new scheme
principles, applying one at a time. Distance relays are set with for reliable protection of transmission lines in converter-based
quadrilateral characteristics as in [25] with a fault resistance source dominated power systems.
coverage of 60Ω.
A phase-B-to-phase-C-to-ground (BCG) fault is created in III. PROPOSED METHOD
line 2–7 at a distance of 0.25pu from bus 2 with RF = 20Ω. This section proposes a new technique for reliable and accel-
Performance of the relays at bus 2 and bus 7 (R2 and R7 ), when erated protection of converter-dominated power networks.
incorporated with distance relaying principle, are demonstrated For the two-bus equivalent system in Fig. 1(a), apparent
in Fig. 3. Results show that both the relays fail to identify the fault impedance (Zapp ) calculated by the distance relay RM for an
in corresponding Zone-1, which indicates a clear maloperation AG fault, created at a distance of xpu from bus M and with a
for DUTT and PUTT schemes. It is also observed that the relay fault resistance RF , is given by [28],
R2 finds the fault even outside its Zone-2 boundary. This causes  
incorrect operation even when the PUTT and DCB schemes VAM IAF
Zapp = = xZ1L + RF .
are applied. Such maloperation of distance relay is due to the IAM + K0 I0M IAM + K0 I0M
non-homogeneity present between the fault currents fed through (1)
both ends of the faulted line. In this case, fault current at bus 2 is where, VA and IA are the voltage and current measurements
fed from the solar plant (connected at bus 2 ) interfaced through respectively. ‘M’ in subscript represents the measurement at
grid-following converter, whereas the fault current at bus 7 is bus M and ‘F’ represents the the variables in the faulted path.
fed from two solar plants (connected at bus 1 and 3) interfaced ZL represents the impedance of line MN. Subscript ‘0’ and ‘1’
through grid-following converters. Similar maloperation may indicate the zero and positive sequence components respectively.
also be experienced for faults in any other lines in the system. K0 is the zero sequence compensation factor. Fig. 5 represents
The decisions derived using distance relay Zone-2 setting the sequence network of the system in Fig. 1(a) for AG faults.
can also be derived using directional relaying principles [12]. ‘RN’ in subscript represents the equivalent representation of the
Therefore, the performance of relay R2 and R7 are tested after renewable sources connected to the bus, whereas ‘S’ indicates
employing superimposed component based directional relaying the conventional sources which are considered here to be out
principle, which is applicable for all types of faults [26]. Results of operation. ‘2’ in the position of first subscript indicates
are provided in Fig. 4. ΔV1 and ΔI1 represent the positive negative sequence components. M end is considered to be con-
sequence superimposed voltage and current components, which nected to a renewable source through a grid-following converter.
are obtained by subtracting 2-cycle memorized prefault data Therefore, the positive sequence source-equivalent of this side
from the fault data [27]. According to the principle, the relay is represented by a dependent current source in parallel with

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
5602 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 60, NO. 4, JULY/AUGUST 2024

a variable impedance [8]. The converter is considered to be


controlled with balanced current controller. So it cannot inject
negative sequence current even during asymmetrical faults. On
the other hand, the N end is considered to be connected to a
renewable source through grid-forming converter and controlled
using dual-current controller mimicking synchronous generator
negative sequence impedance angle characteristics. Positive and
negative sequence source-equivalents are represented with a
dependent voltage source with a variable series impedance. A
dependent voltage source can also be transformed to an equiva-
lent current source at each instant [6]. Therefore, the sequence
network presented in Fig. 5 is a generalized representation for a
converter dominated power network.
I1F , I2F and I0F being equal, IAF in (1) can be replaced by Fig. 6. Proposed transfer trip scheme logic implementation.
I0F , as in (2).
 
VAM 3I0F
= xZ1L + RF . (2) in a generalized form in (7).
IAM + K0 I0M IAM + K0 I0M
|VRM |
(2) is rewritten in (3) by expanding the variables with their S1M = sin (α − γ)
complex forms. |IRM |

|VAM | S2M = |Z1L | sin (θ1L − γ + β) . (7)


ej(α–β) = x|Z1L |ejθ1L
|IAM + K0 I0M | where VRM and IRM are the operating voltage and current for
|3I0F | the relay at bus M.
+ ej(γ–β) RF (3) x is always positive for any fault in the forward direction.
|IAM +K0 I0M |
Thus, a new directional criteria for converter-dominated power
α and β are the phase angles of VAM and (IAM + K0 I0M ). θ1L network is proposed in (8).
and γ represent the line impedance angle and the phase angle of 
I0F respectively. (3) is rewritten in (4). S1M > 0; f or f orward f ault
= (8)
S2M < 0; f or reverse f ault
|VAM |
ej(α−γ) = x|Z1L |ej(θ1L −γ+β) Using this criteria, a transfer trip scheme is proposed. The
|IAM + K0 I0M |
scheme issues trip command only when the fault is identified
|3I0F | in forward direction at both ends, which is similar to POTT [12]
+ RF (4)
|IAM +K0 I0M | but applies the proposed directional criteria for deriving the
(5) is derived by comparing the imaginary parts of both sides in decisions. Steps associated with the proposed scheme are shown
(4). in Fig. 6. Currents and voltages at each end are obtained from
local current transformer (CT) and capacitor voltage transformer
|VAM | (CVT) respectively. A combination of undervoltage and over-
sin (α − γ) = x|Z1L | sin (θ1L − γ + β)
|IAM + K0 I0M | current principle is applied for fault detection. A zero sequence
(5) overcurrent check is also used in parallel to enhance the fault
From (5), x can be computed as in (6). detection sensitivity for ground faults with high RF . A local
S1M voltage based techniques, as available in [29], is applied for
x= (6) fault classification.
S2M
where, IV. RESULTS
|VAM | Performance of the proposed protection scheme is tested in
S1M = sin (α − γ)
|IAM + K0 I0M | the 9-bus system of Fig. 2 with 100% converter-based sources
S2M = |Z1L | sin (θ1L − γ + β) . and a CIRES-integrated 39-bus system, as shown in Fig. 7 using
PSCAD/ EMTDC simulation platform for different faults with
Renewable sources at both ends are connected to the grid through variation in some critical parameters. Solar and wind farms
a dYg-type transformer. Therefore, the zero sequence network integrated to the 39-bus system are of 300 MVA each with the
remain homogeneous as considered for a system without re- specifications as provided in [30] and controlled with balanced
newable sources. Thus, γ is the angle of zero sequence current current controller. Type-III and Type-IV wind farm are complied
measured at bus M. For other types of faults, S1M and S2M with standard control schemes as available in [31], [32]. Voltage
used in the new directional criteria proposed in (8) are derived and current phasors are estimated using 1-cycle discrete Fourier
in Appendix. For all types of faults, S1M and S2M are expressed transform at a sampling rate of 64 samples/ cycle.

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
PALADHI et al.: RELIABLE ACCELERATED PROTECTION SCHEME FOR CONVERTER-DOMINATED POWER NETWORKS 5603

Fig. 7. Converter-based source integrated 39-bus system.

A. Performance Evaluation in a 9-Bus System System With


100% Converter-Based Sources
Performance of the proposed method is tested for two faults
(internal and external with respect to the line 2–7 of the system
in Fig. 2). The first case is same as mentioned in Section II
i.e for a BCG fault created in line 2–7 at a distance of 0.25pu
from bus 2 with RF = 20Ω. It has already been demonstrated
that all the communication assisted trip schemes fail to protect
the line correctly when both end relays are incorporated with
conventional distance or directional relaying principles. Results
in Fig. 8 show that relays at both ends of line 2–7 calculates
the directional index as proposed in this work are positive (0.24
Fig. 8. Performance of (a) R2 and (b) R7 with the proposed method for a
and 0.73). Thus, the method identifies the internal fault correctly system with 100% CIRESs.
satisfying the conditions in (8) and issues trip command for both
end circuit breakers.
In the second case, a BCG fault is created in line 7–8 of TABLE I
DIRECTIONAL INDICES CALCULATED AT BOTH LINE ENDS FOR DIFFERENT
the system in Fig. 2 at a distance of 0.5pu from bus 7 with FAULTS WITH VARIATION IN ASSOCIATED PARAMETERS
RF = 15Ω. From the results in Fig. 8, it is observed that the
relay at bus 2 calculates the directional index as positive (=1.40)
using the proposed method, whereas the relay at bus 7 calculates
it as negative (=−0.53). Thus, the proposed scheme maintains
security correctly even for external fault.

B. For Different Faults Created at Different Locations With


Variation in RF
Performance of the proposed method is now tested for differ-
ent faults created in the CIRES-integrated 39-bus test system,
as shown in Fig. 7 varying the fault location (x) and associated
fault resistance (RF ). RF for ground faults is varied up to 100Ω.
Line 28–29 is considered to be out-of-operation for this study index as positive when the fault created at any location in line
and the performances of relays at both ends of line 29–26 are 29–26 and derive the decision for all internal faults correctly.
tested incorporating the proposed protection technique. Results For faults created in line 25–26, relay at bus 29 calculates the
obtained for different cases are provided in Table I. It is observed index as positive and confirm the fault to be in forward direction,
that both relays at bus 29 and bus 26 calculate the directional whereas relay at bus 26 calculates the index as negative for all

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
5604 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 60, NO. 4, JULY/AUGUST 2024

Fig. 10. Reactive current requirement with voltage dip for different grid codes.

code introduces a proportionality constant (k) to determine the


required reactive current in relation to the voltage deviation. k
varies in the range of 0−10 based on the agreement with network
operators, with a default value of 2. Variation of reactive current
(Iq ) generation with respect to the voltage drop ( ΔV
Vn ) during fault
for four different European grid codes are shown in Fig. 10 [33].
Here, In and Vn are the nominal current and voltage respectively.
Such variation in Iq modulates the phase angle associated with
fault current significantly, and affects relay operations at times.
Performance of the proposed protection scheme is tested for AG
Fig. 9. Performance of the proposed method with variation in generation of
CIRES connected to the line. faults created in line 29–26 at a distance of 0.6pu from bus 29
with RF = 35Ω, when the converter interfacing the solar plant
connected to bus 29 is controlled to satisfy North American
grid-code (NAGC) and European Union grid code (EUGC),
the cases and maintains security for such external faults. Results
one at a time. The converter operates within a power factor
validate the method to be applicable for all types of faults and
range of 0.95 lag to 0.95 lead while complied with NAGC,
also to be independent of fault locations and fault resistances.
whereas it prioritizes reactive current generation when complied
with EUGC [5], [21]. Results in Fig. 11(a) show the variation
C. For Generation Variation in the Renewable Plant in S1 and S2 computed by the relays at bus 29 and bus 26
Connected to the Protected Line with the change in grid code requirements as imposed on the
Variation in renewable plant generation influences the equiv- connected CIRES interfacing converter. Such variations in S1
alent source impedance significantly. Such a situation degrades and S2 cannot affect the directional index proposed in this work.
the performance of protective relays. Therefore, it is necessary to From the results shown in Fig. 11(b), it is observed that the
evaluate the performance of the proposed protection technique index computed for both the situations are almost same and
under similar situations. For this purpose, BCG faults are created positive. Thus, the proposed scheme performs accurately even
in line 29–26 of the system in Fig. 7 at a distance of 0.4pu from with the variation in grid code requirements imposed for the
bus 29 with RF = 30Ω, while the generation of the solar plant converter-control operation.
connected at bus 29 is varied from 25% to 100%. Line 28–29 is
considered to be out-of-operation. Similar values for S1 and S2 E. In the Presence of Different CIRESs Connected to the
calculated by the relays at bus 29 and bus 26 for all the cases (as Protected Line
shown in Fig. 9(a)) demonstrate these variables to be adaptive to
Different types of CIRESs are integrated to the grid with
the generation variation of the renewable plant. Results shown
different converter arrangements modulating the voltage and
in Fig. 9(b) verify that the generation variation cannot affect the
current signals differently during faults. This may affect relay
directional index proposed in this work. Positive and consistent
operations significantly. Performance of the proposed method
values of the indices confirm correct operation of the proposed
is tested for such a situation, when BCG faults are created in
method for all the cases with variation in CIRES generation.
line 29–26 at a distance of 0.4pu from bus 29 with RF = 30Ω
and the solar plant connected to bus 29 is replaced by Type-
D. With CIRESs Satisfying Different Grid Code Requirements III and Type-IV wind farms of same capacity, one at a time.
CIRESs need to satisfy the grid code requirements to remain Variations in S1 and S2 computed by the relays at both ends
connected to the grid, especially during faults. Variation in grid of the line 29-26 for all situations are shown in Fig. 12(a).
code influences the converter-control operation to act accord- The directional index ( SS21 ) calculated at both ends, as shown
ingly, especially the reactive current injection for supporting in Fig. 12(b) are found to be consistent and positive for all
the low voltage situations. For an example, the German grid the situations. Thus, the proposed method derives protection

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
PALADHI et al.: RELIABLE ACCELERATED PROTECTION SCHEME FOR CONVERTER-DOMINATED POWER NETWORKS 5605

Fig. 11. Performance of the proposed method with CIRESs complied with Fig. 12. Performance of the proposed method for different types of CIRESs
different grid code requirements. connected to the line.

V. CONCLUSION
TABLE II
COMPARATIVE ANALYSIS WITH RECENT AVAILABLE METHODS Power grid is experiencing a noticeable change in fault char-
acteristics with growing penetration of converter-based renew-
able sources. Conventional distance and directional relaying
decisions derived in such a new grid scenario is found to be
unreliable. This work demonstrates the impact of CIRESs on
communication assisted accelerated protection schemes, where
the trip decisions are generated and supervised by conventional
distance and/ or directional relays. The work proposes a new cri-
teria to identify the fault direction in converter-dominated power
networks using local voltage and current data. The directional
decisions derived at both ends are transferred mutually through
a low-bandwidth dedicated communication channel to generate
a secure and dependable trip command for the line end breakers.
The method is derived using the homogeneity present in negative
decisions correctly satisfying the criteria in (8). This declares and zero sequence networks in high voltage transmission grids,
the method to be independent of the converter-control operation even with CIRESs. This generalizes the method for application
with different renewable sources. in any transmission network with or without CIRESs. The
method is tested for a 9-bus system with 100% converter-based
sources and a CIRES integrated 39-bus system. The method
is found to be reliable for faults created at different locations,
F. Comparative Assessment with different fault resistances, in the presence of CIRESs with
Different techniques are being tested to prevent the growing different control operations satisfying grid code requirements.
maloperation issues with conventional distance and directional The proposed protection scheme addresses two important con-
relays in the presence of CIRESs. Features of the proposed cerns for the present power systems. It performs correctly being
method is now compared with some of those techniques, avail- independent of the behavior of the sources feeding the fault
able in [8], [22], [34], [35], [36]. The comparative assessment current. In addition, accelerated protection decisions from both
in Table II clearly reveals the novelty and necessity of the ends help in reliable and quick isolation of the faulted part which
proposed method for the present power systems moving towards helps in maintaining system stability in low-inertia situations as
decarbonization. being experience with growing penetration of converter-based

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
5606 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 60, NO. 4, JULY/AUGUST 2024

Fig. 13. Sequence networks of the system in Fig. 1(a) for (a) BCG, (b) BC, and (c) 3-phase faults.

sources. Comparative assessment with a few advanced protec- where,


tion techniques reveals the novelty of the proposed method α = arg (V2M − V0M ) , β = arg (I2M − K0 I0M ) ,
and justifies its requirement in the new power grid scenario.
The work can be further extended for applying in low voltage γ = arg(−I0F ) = arg(I0M ) − π
distribution grids, where a significant non-homogeneity may
present between negative and zero sequence networks. B. For BC Faults
Applying KVL in the faulted loop of the sequence network in
APPENDIX Fig. 13(b) and rearranging the variables, Zapp calculated by the
distance relay RM for a BC fault, created at a distance of xpu
Sequence networks of the system in Fig. 1(a) for BCG, BC
from bus M and with a fault resistance RF , can be expressed as,
and 3-phase faults are shown in Fig. 13.
V1M − V2M 2I2F
= xZ1L − Rph . (11)
A. For BCG Faults I1M − I2M I1M − I2M
Comparing (11) with (2), S1M and S2M for BC faults are
Applying Kirchhoff’s Voltage Law (KVL) in the faulted loop
expressed in (12).
of the sequence network in Fig. 13(a) and rearranging the
variables, Zapp calculated by the distance relay RM for a BCG |V1M − V2M |
S1M = sin (α − γ)
fault, created at a distance of xpu from bus M and with a fault |I1M − I2M |
resistance RF , can be expressed as,
S2M = |Z1L | sin (θ1L − γ + β) . (12)
V2M –V0M I0F where,
 = xZ1L –  I
I2M –K0 I0M I2M –K0L 0M
α = arg (V1M − V2M ) , β = arg (I1M − I2M ) ,
×((1–d)Rph +3RF ). (9)
γ = arg(–I2F )
 Z0L I2F
where, =
K0L and d =
Z1L . Arcing resistance Rph is neg-
I0F Bus M being connected to a grid-following converter-interfaced
ligible to RF for high resistance faults. Homogeneity in the renewable plant controlled with balanced current controller, no
equivalent negative and zero sequence networks results in d to negative sequence current is measured at M. CIRES connected
be a real term for solid faults with low RF . Comparing (9) with at bus N being controlled with dual current controller mimicking
(2), S1M and S2M for BCG faults can be expressed as in (10). synchronous generator’s negative sequence impedance, Z2S2
can be expressed as, Z2S2 = mZ2L , where m is a real-valued
|V2M –V0M | multiplier. For such a situation, I2F can be obtained as in (13).
S1M = sin (α − γ)
|I2M –K0 I0M |
V2F V2M
I2F = − ≈− (13)
S2M = |Z1L | sin (θ1L − γ + β) . (10) (1 − x)Z2L + Z2S2 (1 − x + m)Z2L

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
PALADHI et al.: RELIABLE ACCELERATED PROTECTION SCHEME FOR CONVERTER-DOMINATED POWER NETWORKS 5607

Thus, γ for BC fault is derived for relay RM as follows. [8] S. Paladhi and A. K. Pradhan, “Adaptive distance protection for lines
  connecting converter-interfaced renewable plants,” IEEE J. Emerg. Sel.
V2M Topics Power Electron., vol. 9, no. 6, pp. 7088–7098, Dec. 2021.
γ = arg(−I2F ) = arg (14) [9] K. Jia, Z. Yang, Y. Fang, T. Bi, and M. Sumner, “Influence of inverter-
Z2L interfaced renewable energy generators on directional relay and an
improved scheme,” IEEE Trans. Power Electron., vol. 34, no. 12,
On the other hand, I2F flows only through the bus N. Therefore pp. 11843–11855, Dec. 2019.
for relay RN , γ for BC fault is obtained as in (15). [10] Y. Xue, B. Kasztenny, D. Taylor, and Y. Xia, “Series compensation,
power swings, and inverter-based sources and their impact on line current
f
γ = arg(−2I2F ) = arg(I2N )−π (15) differential protection,” in Proc. 66th Annu. Conf. Protective Relay Eng.,
2013, pp. 80–91.
[11] A. Chowdhury, S. Paladhi, and A. K. Pradhan, “Adaptive unit protection
A3 For 3-Phase Faults for lines connecting large solar plants using incremental current ratio,”
IEEE Syst. J., vol. 16, no. 2, pp. 3272–3283, Jun. 2022.
Only positive sequence components are presents for 3-phase [12] H. J. A. Ferrer and E. O. Schweitzer III, Modern Solutions for Protection,
faults, as shown in Fig. 13(c). Zapp calculated by the distance Control, and Monitoring of Electric Power Systems. Pullman, WA, USA:
Schweitzer Eng. Lab., 2010, pp. 79–83.
relay RM for a 3-phase fault, created at a distance of xpu from [13] S. Paladhi and A. K. Pradhan, “Adaptive zone-1 setting following structural
bus M and with a fault resistance RF , can be expressed as, and operational changes in power system,” IEEE Trans. Power Del.,
vol. 33, no. 2, pp. 560–569, Apr. 2018.
V1M I1F [14] M. Thompson and A. Somani, “A tutorial on calculating source impedance
= xZ1L + RF . (16) ratios for determining line length,” in Proc. 68th Annu. Conf. Protective
I1M I1M
Relay Eng., 2015, pp. 833–841.
For 3-phase faults, RF is considered to be very small [15] Siemens, “Distance protection relay for transmission lines,” Tech.
(< 1Ω) [37]. Thus, the value of II1M1F
RF in (16) becomes neg- Rep., 1999. [Online]. Available: https://www.automation-berlin.com/
downloads/siemens/energy_ptd/7SA522_Catalog_SIP_E6.pdf
ligible, especially when the imaginary parts of both sides are [16] Alstom, “Network protection and automation guide,” Alstom Grid, Tech.
compared (as done in (5) for obtaining S1M and S2M ). There- Rep. 978-0-9568678-0-3, May 2011.
[17] K. E. Arroudi and G. Joos, “Performance of interconnection protection
fore, (16) can be simplified as in (17). based on distance relaying for wind power distributed generation,” IEEE
Trans. Power Del., vol. 33, no. 2, pp. 620–629, Apr. 2018.
V1M
= xZ1L . (17) [18] A. K. Pradhan and G. Joos, “Adaptive distance relay setting for lines
I1M connecting wind farms,” IEEE Trans. Energy Convers., vol. 22, no. 1,
pp. 206–213, Mar. 2007.
Comparing (17) with (2), S1M and S2M for 3-phase faults can [19] K. Li, L. Lai, and A. David, “Stand alone intelligent digital dis-
be expressed as in (18). tance relay,” IEEE Trans. Power Syst., vol. 15, no. 1, pp. 137–142,
Feb. 2000.
|V1M | [20] J. Upendar, C. Gupta, and G. Singh, “Comprehensive adaptive distance
S1M = sin (α − β) relaying scheme for parallel transmission lines,” IEEE Trans. Power Del.,
|I1M | vol. 26, no. 2, pp. 1039–1052, Apr. 2011.
[21] A. Banaiemoqadam, A. Hooshyar, and M. A. Azzouz, “A control-based
S2M = |Z1L | sin (θ1L ) . (18) solution for distance protection of lines connected to converter-interfaced
sources during asymmetrical faults,” IEEE Trans. Power Del., vol. 35,
where, no. 3, pp. 1455–1466, Jun. 2020.
[22] Y. Fang, K. Jia, Z. Yang, Y. Li, and T. Bi, “Impact of inverter-interfaced
α = arg (V1M ) and β = arg (I1M ) renewable energy generators on distance protection and an improved
scheme,” IEEE Trans. Ind. Electron., vol. 66, no. 9, pp. 7078–7088,
Sep. 2019.
[23] R. W. Kenyon, A. Sajadi, A. Hoke, and B.-M. Hodge, “Open-source
REFERENCES PSCAD grid-following and grid-forming inverters and a benchmark for
[1] S. Paladhi, Q. Hong, and C. Booth, “A reliable accelerated protection zero-inertia power system simulations,” in Proc. IEEE Kansas Power
scheme for converter-dominated power networks,” in Proc. 22nd Nat. Energy Conf., 2021, pp. 1–6.
Power Syst. Conf., 2022, pp. 654–657. [24] M. Nagpal, M. Jensen, and M. Higginson, “Protection challenges and
[2] “Renewables integration in India,” NITI Ayog, Int. Energy practices for interconnecting inverter based resources to utility transmis-
Agency, Tech. Rep., 2021. [Online]. Available: https://iea.blob. sion systems,” Working Group C32, Power System Relaying and Con-
core.windows.net/assets/7b6bf9e6-4d69-466c-8069-bdd26b3e9ed1/ trol Committee, Tech. Rep., 2020. [Online]. Available: https://www.pes-
RenewablesIntegrationinIndia2021.pdf psrc.org/kb/report/109.pdf
[3] ESIG, “Grid-forming technology in energy systems integration,” Energy [25] Y. Liang, Z. Lu, W. Li, W. Zha, and Y. Huo, “A novel fault impedance
Systems Integration Group, Reston, VA, USA, Tech. Rep., 2022. [Online]. calculation method for distance protection against fault resistance,” IEEE
Available: https://www.esig.energy/wp-content/uploads/2022/03/ESIG- Trans. Power Del., vol. 35, no. 1, pp. 396–407, Feb. 2020.
GFM-report-2022.pdf [26] M. Biswal, B. B. Pati, and A. K. Pradhan, “Directional relaying for double
[4] W. Du et al., “Modeling of grid-forming and grid-following inverters circuit line with series compensation,” IET Gener., Transmiss., Distrib.,
for dynamic simulation of large-scale distribution systems,” IEEE Trans. vol. 7, no. 4, pp. 405–413, 2013.
Power Del., vol. 36, no. 4, pp. 2035–2045, Aug. 2021. [27] S. Paladhi, A. K. Pradhan, and J. G. Rao, “Accurate superim-
[5] IRENA, “Grid codes for renewable powered systems,” Int. Renewable posed component estimation for improved relay performance dur-
Energy Agency, Abu Dhabi, UAE, Tech. Rep., 2022. [Online]. Available: ing power swing,” IEEE Syst. J., vol. 16, no. 4, pp. 6119–6129,
https://www.irena.org/-/media/Files/IRENA/Agency/Publication/ Dec. 2022.
2022/Apr/IRENA_Grid_Codes_Renewable_Systems_2022.pdf?rev= [28] G. Ziegler, Numerical Distance Protection: Principles and Applications.
986f108cbe5e47b98d17fca93eee6c86 Hoboken, NJ, USA: Wiley, 2011.
[6] S. Paladhi and A. K. Pradhan, “Adaptive fault type classification for [29] A. Hooshyar, E. F. El-Saadany, and M. Sanaye-Pasand, “Fault type clas-
transmission network connecting converter-interfaced renewable plants,” sification in microgrids including photovoltaic DGs,” IEEE Trans. Smart
IEEE Syst. J., vol. 15, no. 3, pp. 4025–4036, Sep. 2021. Grid, vol. 7, no. 5, pp. 2218–2229, Sep. 2016.
[7] S. Paladhi, “Distance protection for lines connecting converter-interfaced [30] S. Paladhi, J. R. Kurre, and A. K. Pradhan, “Source-independent zone-1
renewable sources,” Ph.D. dissertation, IIT Kharagpur, Kharagpur, India, protection for converter-dominated power networks,” IEEE Trans. Power
2021. Del., vol. 39, no. 1, pp. 341–351, Feb. 2024.

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.
5608 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 60, NO. 4, JULY/AUGUST 2024

[31] PSCAD, “Type 3 wind turbine model,” PSCAD, Manitoba Hydro Inter- Qiteng Hong (Senior Member, IEEE) received the
national Ltd., 211 Commerce Drive, Winnipeg, Manitoba, Canada, Tech. B.Eng. (Hons.) degree and the Ph.D. degree in elec-
Rep. R3P 1A3, Nov. 2018. tronic and electrical engineering from the University
[32] PSCAD, “Type 4 wind turbine model,” PSCAD, Manitoba Hydro Inter- of Strathclyde, Glasgow, U.K., in 2011 and 2015,
national Ltd., 211 Commerce Drive, Winnipeg, Manitoba, Canada, Tech. respectively. He is currently a Reader with the Uni-
Rep. R3P 1A3, Dec. 2018. versity of Strathclyde, Glasgow, U.K. His research
[33] C. Sourkounis and P. Tourou, “Grid code requirements for wind interests include power system protection and control
power integration in Europe,” in Proc. Conf. Papers Energy, 2013, in future networks with high penetration of renew-
pp. 1–9. ables. Dr. Hong is a member of the IEEE Working
[34] Y. Liang, W. Li, and Y. Huo, “Zone I distance relaying scheme of lines Group P2004 and IEEE Task force on Cloud-Based
connected to MMC-HVDC stations during asymmetrical faults: Prob- Control and Co-Simulation of Multi-Party Resources
lems, challenges, and solutions,” IEEE Trans. Power Del., vol. 36, no. 5, in Energy Internet. He was also a Regular Member of the completed CIGRE
pp. 2929–2941, Oct. 2021. WG B5.50.
[35] K. Ma, H. K. Høidalen, Z. Chen, and C. L. Bak, “Improved zone 1 top-line
tilting scheme for polygonal distance protection in the outgoing line of
type-4 wind parks,” CSEE J. Power Energy Syst., vol. 9, no. 1, pp. 172–184,
Jan. 2023.
[36] C. Chao, X. Zheng, Y. Weng, Y. Liu, P. Gao, and T. Nengling, “Adaptive dis- Campbell Booth (Member, IEEE) received the
tance protection based on the analytical model of additional impedance for B.Eng. and Ph.D. degrees in electrical and electronic
inverter-interfaced renewable power plants during asymmetrical faults,” engineering from the University of Strathclyde, Glas-
IEEE Trans. Power Del., vol. 37, no. 5, pp. 3823–3834, Oct. 2022. gow, U.K., in 1991 and 1996, respectively. He is cur-
[37] V. Makwana and B. Bhalja, “A new digital distance relaying scheme for rently a Professor with the Department of Electronic
compensation of high-resistance faults on transmission line,” IEEE Trans. and Electrical Engineering, University of Strathclyde.
Power Del., vol. 27, no. 4, pp. 2133–2140, Oct. 2012. His research interests include power system protec-
tion; plant condition monitoring and intelligent as-
set management, applications of intelligent system
techniques to power system monitoring, protection,
and control, knowledge management, and decision
support systems.

Subhadeep Paladhi (Member, IEEE) received the


Ph.D. degree in electrical engineering from the Indian
Institute of Technology, Kharagpur, India, in 2021.
From 2021 to 2023, he was with the Department of
Electronic and Electrical Engineering, University of
Strathclyde, Glasgow, U.K., as a Research Associate.
Since 2023, he has been with the Department of Elec-
trical Engineering, Indian Institute of Technology In-
dore, Indore, India, as an Assistant Professor. His re-
search interests include power system protection and
monitoring in the presence of converter-interfaced
renewable sources.

Authorized licensed use limited to: Makerere University Library. Downloaded on October 25,2024 at 16:14:57 UTC from IEEE Xplore. Restrictions apply.

You might also like