Received: 19 August 2022 Revised: 26 January 2023 Accepted: 4 February 2023 IET Generation, Transmission & Distribution

DOI: 10.1049/gtd2.12797

ORIGINAL RESEARCH

Transmission line protection using local information in the

presence of inverter-interfaced renewable energy sources
Nashaat K. Yaseen Heresh Seyedi Mehdi Abapour

Faculty of Electrical and Computer Engineering, Abstract

University of Tabriz, Tabriz, Iran
Although inverter-interfaced renewable energy sources (IIRESs) have great economic and
environmental benefits, their fault current characteristics differ from those of synchronous
Correspondence
Nashaat K. Yaseen, Faculty of Electrical and generators. IIRESs also comply with modern grid code (GC) requirements for injecting
Computer Engineering, University of Tabriz, Tabriz, reactive current to compensate for voltage drop during fault conditions. Moreover, the
Iran.
limits of renewable energy-side fault current contribution and the phase angle difference
Email n.kazem99@[Link]
between the local and infeed fault current from conventional sources lead to incorrect
[Correction added on 28-February-2023, after first reactance estimation by the traditional distance relay at the IIRES substation and nega-
online publication: Author name spelling is tively affect relay dependability. As a result, the IIRES-side distance protection is prone
corrected from Heresh Seydi to Heresh Seyedi in to incorrect operation. Furthermore, an improved relaying scheme is proposed to alle-
this version of paper]
viate the adverse effects of IIRES on the reactance calculation. This research aims to
improve the performance of the ground distance protection scheme in the case of a single-
phase to ground fault in zone one. The proposed method, with more than 99% accuracy,
is solely dependent on the local information and, therefore, no communication link is
required to implement the algorithm. Moreover, the algorithm is compatible with the exist-
ing pilot protection methods used in the industry. The simulation studies are modeled in
the Digsilent Power Factory environment.

1 INTRODUCTION instabilities [6], and the delayed tripping of circuit breakers due
to under reach will force the power systems under stress for
Electric power systems require effective economic planning and a longer time period [7]. This might lead to a blackout, simi-
operation. At the same time, the world’s installed power gener- lar to the Europe blackout in 2006 [8]. The fault-ride-through
ating capacity has increased significantly as a result of the rapid (FRT) requirement, on the other hand, is a significant aspect of
growth in global energy consumption, which is mainly supplied grid codes (GC) [9], since it makes the IIRES subject to specific
by fossil fuels, causing environmental pollution. Therefore, it is regulation during fault conditions, such as European codes that
important to focus on sustainable energy to preserve the envi- give priority to supporting the grid voltage with reactive current
ronment due to the fact that it has lower carbon emissions from the IIRES [10].
than synchronous generators [1]. However, as renewable energy As a result, it is crucial to investigate how the distance relay
penetration increases in the power system, it may cause sev- behaves when the IIRESs are present. In [11], adaptive distance
eral problems for protective devices since IIRESs have different protection based on time delay and zero-sequence impedance
fault characteristics from traditional generators [2, 3]. has been investigated. This approach involves a time-delayed
Due to its satisfactory sensitivity and speed, distance pro- decision to turn off the renewable energy source-side distance
tection has been used to protect transmission lines since the relay. The grid-side relay, which can appropriately detect faults,
beginning of the electrical industry. The distance relay estimates is still operational during this short period. This strategy may
the impedance up to the fault location and compares it to the take a longer time to find the problem, which extends the net-
pre-defined relay characteristics [4]. work stress. The technique in [12] is based on pilot protection,
The integration of IIRES will have an impact on distance which can compute both the fault resistance and its location
relay protection [5]. Meanwhile, if distance protection mal- accurately. The suggested strategy protects the transmission line
functions, the system will be prone to transient and voltage better than conventional distance protection. However, there

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided
the original work is properly cited and is not used for commercial purposes.
© 2023 The Authors. IET Generation, Transmission & Distribution published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

2156 [Link]/iet-gtd IET Gener. Transm. Distrib. 2023;17:2156–2168.

YASEEN ET AL. 2157

is a problem in the event that the communication link fails.

The technique in [13] is based on the high-frequency compo-
nents of fault signals. This method is capable of successfully
dealing with different values of fault resistance. However, using
high-frequency components to aid in adaptive distance protec-
tion will result in increased sensitivity to high-frequency noise.
Therefore, it reduces the effectiveness of the recommended
procedures. The challenges of protecting lines connected to
FIGURE 1 Schematic of the test system
IIRESs against remote and intermediate infeed current are
investigated in [14]. In [15], a method has been proposed that
is dependent on the zero- sequence compensation factor and close to an offshore wind farm. The main contributions of the
does not rely on the host operating system’s grid code. The new scheme are: (1) it is only dependent on local information
results indicate that the mentioned approach is accurate and and intrinsically does not require any communication link. (2)
fast. However, since the method is dependent on the zero- It may be integrated with any of the existing pilot protection
sequence circuit, it cannot be applied to the double phase and approaches. (3) The proposed technique is not sensitive to the
three-phase faults. The technique in [16] aims to emulate the value of fault resistance since it works properly even in the case
symmetrical fault currents that influence distance relays. The of high fault resistance. (4) The proposed method is immune
suggested method is cost-effective because it uses local data to to changes in the injected power by the IIRES, grid topol-
control the fault current angle from converter-interfaced renew- ogy, evolving faults, and simultaneous faults. (5) Based on real
able plants, imitating the features of a synchronous generator. data, the proposed method’s fault location is close to the fault
Since there are so many types of renewable plants and control location as determined by an advanced commercial protection
methods, this strategy is difficult to generalize. In [7], an adap- device. The Digsilent Power Factory software is used to model
tive two-terminal distance protection has been proposed. After the simulation studies.
estimating the fault location, voltage and resistance can be deter-
mined. Although this method is simple, it is not sufficiently fast
and accurate. An adaptive relaying scheme has been proposed 2 IMPACT OF IIRES ON DISTANCE
in [17] to cope with the intermittent nature of wind energy due PROTECTION
to fluctuations in penetration, wind conditions, and distributed
generation structure. Since a communication link is needed to Figure 1 shows the connection of an offshore wind farm to the
get information on the number of plants that are taking part and grid through a transmission line. The distance relay (R) is located
the wind condition, the method might be complicated in prac- close to the IIRES. When an SL-G fault occurs on the transmis-
tice and also not sufficiently fast. In [18], a technique has been sion line through a resistance fault, the conventional distance
proposed for a sequence-component-based current differential relay estimates the impedance as follows:
protection for the transmission line connected to the IIRES,
which is regulated by the positive-sequence control method ( )
IF
under unbalanced faults. Compared to the traditional method, Zseen−AG = Z f + RF ∗
the protection’s sensitivity has been enhanced. Additionally, the IL + K 0 ∗IL0
scheme is well-resistant to various fault resistances and fault ( )
locations. However, in the event that the communication link = Z f + SI Supplement impedance , (1)
breaks down, the algorithm will stop working.
Another issue related to transmission line protection in the where Zseen−AG is the impedance seen by the relay in the event of
presence of renewable sources is the frequency variations that a SL-G fault. IF and IL are the fault and local currents, respec-
take place in the systems with IIRESs. These frequency vari- tively. K 0 = (ZL0 − ZL1 )∕ZL1 is the zero-sequence compensation
ations may affect the phasor estimation part of the protective factor. ZL0 and ZL1 are the zero and positive sequence line
relays. In order to avoid these effects, two different approaches impedances, respectively. IL0 is the zero sequence local current.
may be used: protective relays and PMUs. In the first approach, According to (1), there will be supplementary impedance (SI)
the system frequency is calculated by the relay at the beginning added to the actual impedance (Z f ), estimated by the ground
of the phasor estimation process, and the sampling frequency is distance relay, only in the presence of fault resistance and remote
adapted to the actual frequency. In the second approach, phasor infeed current.
and frequency are both calculated, then based on the estimated The phase difference between the corresponding pre-fault
frequency, the calculated phasor is compensated to remove the voltages determines the angle formed by the line’s fault cur-
effect of frequency variations [19]. rents at either end when the system is exclusively supplied by
This research presents a distance relay protection scheme synchronous generators [20]. Because high phase discrepancies
for single-line to ground faults (SL-G), which are the most between a line’s two terminal voltages limit the stability mar-
common transmission system faults, taking into consideration gin, the voltage angles are very close. As a result, the angles
the grid codes that have been used in practice in this field of of the supplemental impedance’s numerator and denominator
study. The proposed method is applied to a transmission line are close, rendering SI mostly resistive. If the relay is at the
2158 YASEEN ET AL.

FIGURE 2 German GC dynamic’s reactive current requires during

voltage fluctuations

power–sending or power–receiving end of the line, the small

fictional reactance added to the (Z f ) is handled by slightly FIGURE 3 Sequence network diagram of the system for a single line to
tilting the reactance element clockwise or counterclockwise ground fault
[21, 22].
In the event that the IIRESs are present, the current angle
is controlled by the converter’s control system, which ensures
that the grid code GC of the host system is met. Different 3 PROPOSED ADAPTIVE RELAYING
GCs have different requirements for confirming that the IIRES METHOD
generates sufficient reactive current during faults. The GC in
Germany, for example, requires that wind farms support grid This section describes a new protection scheme for adaptive
voltage by supplying more reactive current when voltage drops, distance relays in the presence of IIRES during SL-G faults
as shown in Figure 2, and less reactive current when voltage in order to prevent the issues discussed in Section 2. Figure 3
increases. After a failure occurs, the voltage must be modified depicts the sequence networks of the power system arrange-
within 20 ms, or one cycle, by adding more reactive current to ment for a SL-G fault. It consists of a voltage source converter
the low-voltage side of the wind farm transformer. This cur- connected to the grid through a transmission line and a delta-
rent must be at least 2% of the rated current for every 1% star winding transformer. To protect the transmission line, the
of the voltage drop. If required, the reactive output current distance relay R is positioned at bus 1 (local bus), and the dis-
must be capable of reaching at least 100% of the rated cur- tance relay B is located at bus 2 (remote bus). In both positive
rent. All of the above is true outside the dead band of 10% and negative sequence networks, the voltage source converter
around the nominal voltage [23, 24]. According to the North is depicted as a controlled current source [28]. Since the trans-
American grid code, utilities frequently allow IIRES with a unity former has a delta-star winding, the converter is unable to
power factor (active current generator) to ride through faults supply the fault current’s zero-sequence component [8].
[Link] 0
[25, 26]. In Figure 3, Xcon and [Link] 0 represent the positive,
Regardless of whether the IIRESs comply with GC during negative, and zero-sequence reactance’s of the converter reactor
a fault, the angle of the current close to a relay at an IIRES’ and transformer. [Link] 0 , [Link] 0 and [Link] 0 are the positive,
substation will be significantly different from the other-end negative, and zero-sequence components of the local bus cur-
current. As a result, enormous imaginary components can be rent, remote bus current, and fault current, respectively. [Link] 0
included in the SI. Consequently, the influence of fault resis- and [Link] 0 are the positive, negative, and zero-sequence
tance and infeed current on distance relay impedance seems to components of the line impedance, and the equivalent grid
be no longer primarily influenced by the resistance axis. The impedance, respectively. Vg 1 is the positive sequence equivalent
limitation of 1.1–1.2 p.u [27] on the IIRES-side fault current grid voltage. [Link] 0 , [Link] 0 and [Link] 0 are the positive,
makes the problem worse for a relay at an IIRES substation negative, and zero-sequence components of the local bus volt-
because it limits the denominator of SI. With an adaptive dis- age, remote bus voltage, and fault location voltage, respectively.
tance protection scheme, relays can adjust to changes in the m indicates the fault location and RF is the fault resistance.
impedance of the line and work based on the actual value of the According to Figure 3, for a 2 SL-G fault, the sequence
impedance. components of the local bus voltage VR are calculated by
YASEEN ET AL. 2159

substituting the sequence components of VF by the sequence The impedance seen by relay R is obtained by dividing VR by
components of (IF ∗RF ) to the following equations. IL− . Therefore, (13) can be obtained.

VR0 = IL0 ∗mZL0 + IF0 ∗RF , (2) VR VR ∗ mZ 1L

ZseenR = = . (13)
I−
L VR − I F ∗ RF
VR1 = IL1 ∗mZL1 + IF1 ∗RF , (3)
Furthermore, the fault location (m) is calculated using (14),
VR2 = IL2 ∗mZL2 + IF2 ∗RF . (4) which is derived from (13).

Equation (5) can be calculated by adding (2), (3), and (4) as. ZseenR ∗VR − Z seen R ∗I F ∗RF
m= . (14)
V R ∗ZL1
VR0 − mZL0 ∗IL0 − IF0 ∗RF + VR1 − mZL1 ∗IL1 − IF1 ∗RF + VR2

− mZL2 ∗IL2 − IF2 ∗ RF = 0. (5) However, in (14), there is also an unknown variable RF . As a
result, the following procedures are required to obtain another
Afterwards, the terms (+ mZL1 ∗IL0 ) and (−mZL1 ∗IL0 ) are added equation.
to (5), and Equation (6) can be obtained. On the basis of Figure 3, the sequence components of the
local bus voltage VR can also be obtained through substituting
VR0 − mZL0 ∗IL0 − IF0 ∗RF + VR1 − mZL1 ∗IL1 − IF1 ∗RF + VR2 the sequence components of VF by the sequence components
of VB − [(1 − m)ZL (IF − IL )] as in the following equations.
− mZL2 ∗IL2 − IF2 ∗RF − mZL1 ∗IL0 + mZL1 ∗IL0 = 0. (6)
[ ( )]
VR0 + (1 − m ) ZL0 IF0 − IL0 − mZL0 ∗ IL0 = VB0 , (15)
Equation (7) is the result of rearranging Equation (6), where
VR = VR0 + VR1 + VR2 , IF = IF0 + IF1 + IF2 , and ZL2 = ZL1 . [ ( )]
VR1 + (1 − m ) ZL1 IF1 − IL1 − mZL1 ∗ IL1 = VB1 , (16)
( )
VR − RF ∗ IF0 + IF1 + IF2 − mZL0 ∗IL0 − mZL1 ∗IL1 − mZL1 ∗IL2 [ ( )]
VR2 + (1 − m ) ZL2 IF2 − IL2 − mZL2 ∗ IL2 = VB2 . (17)
− mZL1 ∗IL0 + mZL1 ∗ IL0 = 0. (7)
Therefore, (18) is formed by summing (15), (16), and (17) as.
Rewriting Equation (7) results in Equation (8), where IL =
IL0 + IL1 + IL2 . [ ( )] [ ( )]
VR + (1 − m ) ZL0 IF0 − IL0 + (1 − m ) ZL1 IF1 − IL1
( ) [ ( )] ( )
VR − IF ∗ RF − mZL1 IL0 + IL1 + IL2 + (1 − m ) ZL2 IF2 − IL2 − mZL1 ∗ IL1 + IL2

− mZL0 ∗ IL0 + mZL1 ∗ IL0 = 0. (8) − mZL0 ∗ IL0 = VB . (18)

After simplifying (8), Equation (10) can be derived. After simplifying (18), Equation (21) can be obtained.
( )
VR − IF ∗RF − mZL1 ∗IL − m ZL0 − ZL1 ∗ IL0 = 0, (9) [ ( )]
VR + (1 − m ) ZL1 IF − IF0 − IL + IL0
[ ( )] ( )
+ (1 − m ) ZL0 IF0 − IL0 − mZL1 ∗ IL − IL0
[ ( ) ]
VR − IF ∗RF − m ZL1 ∗IL + ZL0 − ZL1 ∗IL0 = 0. (10)
− mZL0 ∗ IL0 = VB , (19)
The compensated fault current is defined as: [ ( )] [ ( )]
VR + (1 − m ) IF0 ZL0 − ZL1 − (1 − m ) IL0 ZL0 − ZL1
( 0 )
ZL − ZL1
IL− = IL + IL0 → ZL1 ∗ IL− + (1 − m ) ZL1 ∗IF − (1 − m ) ZL1 ∗ IL
ZL1
( ) ( )
= ZL1 ∗ IL + ZL0 − ZL1 ∗ IL0 . (11) − mZL1 ∗IL − m ZL0 − ZL1 IL0 = VB , (20)
[ ( ) ( )
Equation (12) can be derived by substituting (11) into VR + (1 − m ) ∗ IF0 ZL0 − ZL1 − IL0 ZL0 − ZL1
Equation (10). ]
+ ZL1 ∗ IF − ZL1 ∗ IL
VR − IF ∗ RF
IF ∗RF + mZL1 ∗ IL− = VR → IL− = . (12) ( )
mZL1 − mZL1 ∗IL − m ZL0 − ZL1 IL0 = VB . (21)
2160 YASEEN ET AL.

For IF = 3IF0 , Equation (22) can be obtained. the equation and rewrite it as below.
1
2Z 1 + ZL0 [ ( ) ] IF = 3∗IL0 , (29)
(1 − m )
(1 − m ) L IF − m ZL1 ∗ IL + ZL0 − ZL1 IL0
3 1
where IL0 = (ILA + ILB + ILC ).
= VB − VR . (22) 3

Equation (23) can be derived by substituting (11) into (22). 3.2 Estimation of the remote bus voltage VB
2ZL1 + ZL0 In the first stage, the sequence components of the fault loca-
(1 − m ) IF − mZL1 ∗ IL− = VB − VR , (23)
3 tion voltage VF are determined using the following equations,
( ) according to Figure 3.
(1 − m ) 2ZL1 + ZL0 IF − 3 (VB − VR ) = 3∗mZL1 ∗IL− . (24)
V0F = VR0 − IL0 ∗mZL0 , (30)
Equation (25) can be derived by obtaining the fault current
IF from (12) and substituting it into (24). V1F = VR1 − IL1 ∗mZL1 , (31)
[ ( )]
3∗RF + (1 − m ) 2ZL1 + ZL0 VR − 3∗RF ∗VB
IL− = ( ) . (25) V2F = VR2 − IL2 ∗mZL2 . (32)
3∗ZL1 ∗RF + mZL1 (1 − m ) 2ZL1 + ZL0
Adding (30), (31), and (32) yields (33).
VR
Now, by re-arranging (25) and substituting ZseenR = → VF = VR1 − IL1 ∗mZL1 + VR2 − IL2 ∗mZL2 + VR0 − IL0 ∗mZL0 .
IL −
IL− =
VR
, the fault resistance is calculated by (26) in terms of (33)
ZseenR
fault location (m):
Obtaining (34) for the fault location voltage VF is possible
( ) ( ) after simplification of (33).
VR 2ZL1 + ZL 0 (1 − m ) mZL1 − ZseenR
RF = . (26) ( )
I ( 1 )
3 ∗ ZseenR (VR − VB ) − 3 ∗ ZL1 ∗ VR VF = VR − m L 2ZL + ZL0 . (34)
3
Equations (14) and (26) can be solved using the Gauss-Seidel Afterwards, the sequence components of the remote bus
method to determine the fault location and fault resistance. voltage VB , illustrated in Figure 3, will be defined in terms of
However, before solving these equations,VB and IF should be the sequence components of the fault location voltage VF as
determined. Since these signals are dependent on the remote follows:
bus information, an innovative approach is required to estimate
them in terms of local signals. V0B + (1 − m ) ZL0 ∗IB0 + VF0 + VB1 + (1 − m ) ZL1 ∗IB1
The procedure for estimating the fault current IF and remote
voltage bus VB based on the local data bus is provided below. + VF1 + VB2 + (1 − m ) ZL2 ∗IB2 + VF2 = 0. (35)

Equation (36) for the remote bus voltage VB in terms of the

3.1 Estimation of the fault current IF fault location voltage VF can be found after simplifying (35).
( )
I ( 1 )
As shown in Figure 3, the zero-sequence local current in the VB = (1 − m ) ∗ B 2ZL + ZL0 + VF . (36)
zero-sequence circuit can be calculated using the zero-sequence 3
fault current [29] as follows: Since IF = IL + IB , the remote bus voltage VB is as follows:
(1 − m ) ZL0 + Zg0 ( )
IF − IL ( 1 )
IL0 = IF0 ∗ . (27) VB = (1 − m ) 2ZL + ZL0 + VF . (37)
ZL0 + Xtr0 + Zg0 3

By utilizing (34) and (37), the remote bus voltage VB will be

For IF = 3IF0 , the fault current in (28) can thus be deter-
derived in terms of the local bus voltage VR .
mined.

ZL0 + Xtr0 + Zg0

IF = 3∗IL0 . (28) 3.3 Flowchart of the proposed relaying
(1 − m ) ZL0 + Zg0 scheme
In general, the magnitudes of Xtr0 and Zg0 are negligible in The flowchart for the proposed method, which comprises five
comparison to ZL0 . As a result, it is safe to remove them from steps, is shown in Figure 4. The initial estimation of fault
YASEEN ET AL. 2161

TABLE 1 VB and IF calculated for RF = 5 Ω

Error% Error%

AFL (%) IF _Act IF_cal VB _Act VB _cal

30 4.612 4.479 196.99 187.53

2.88 4.80
50 4.920 4.838 192.72 173.32
1.67 10.07
60 5.403 5.358 188.91 168.93
0.83 10.58
70 6.283 6.309 184.90 164.35
0.41 11.11
80 7.963 8.161 173.65 157.55
2.49 9.27

single-circuit transmission lines, two external grids, and an

offshore wind farm that is connected to the bus R (local
bus). Different fault locations and fault resistance values are
investigated in the numerical analysis.
The performance of the proposed technique for distance
relay R is evaluated at various fault locations and fault resis-
tance values, with changes in the system specifications and
parameters, evolving and simultaneous faults, and with real data.
The accuracy of the recommended method is illustrated by a
comparison with conventional distance protection.

4.1 Performance evaluation for estimation

of the VB and IF

In the first stage, performance of the proposed technique for

calculating the VB and IF based on local information (VB_cal
and IF _cal ) is evaluated. Afterwards, the estimated values are
compared with the actual values (VB_Act and IF _Act ) for vari-
ous actual fault locations (AFL) with fault resistances of 5, 10,
FIGURE 4 Flowchart for the proposed relaying scheme
15 and 50 Ω.
Based on the results of Tables 1–4:
location is calculated in the first stage using traditional distance
protection. The fault current IF is then calculated using (29) ∙ Estimation of IF is sufficiently accurate, with a 1.65% average
based on local bus data. The next step is to estimate the remote error.
voltage bus in (34) and (37) based on local information. After- ∙ In some occasions, the estimation of VB is not accurate
wards, the fault resistance RF is calculated using (26). Finally, enough, with an average error of 5.45%. However, in the next
the fault location (m) is recalculated using (14) in accordance subsection, it will be demonstrated that the inaccuracy of VB
with the preceding procedures. This way, the initial estimation estimation only slightly affects the performance of the overall
(m) is improved by using iterative procedures that depend on protection scheme. This is due to the small sensitivity of the
the second estimation (m) through the utilization of an accurate fault location, m, to the value of VB .
algorithm based on local information.

4.2 Performance evaluation for fault

4 PERFORMANCE EVALUATION location errors

The test system is depicted in Figure 1, with data given The performance of the new scheme is verified by estimating
in Table A1 (Appendix) [12]. It consists of two 380KV the fault location errors using the Equation (38) [30] below.
2162 YASEEN ET AL.

TABLE 2 VB and IF calculated for RF = 10 Ω TABLE 4 VB and IF calculated for RF = 50 Ω

Error% Error% Error% Error%

AFL (%) IF _Act IF_cal VB _Act VB _cal AFL (%) IF _Act IF_cal VB _Act VB _cal

30 4.398 4.277 197.93 198.87 30 2.815 2.791 211.88 234.83

2.75 0.47 0.85 10.83
50 4.703 4.633 194.01 184.74 50 2.967 3.004 209.08 224.90
1.49 4.78 1.25 7.57
60 5.146 5.115 189.97 180.76 60 3.118 3.203 207.32 221.94
0.60 4.85 2.73 7.05
70 5.931 5.968 186.62 177.31 70 3.330 3.467 206.21 219.69
0.62 4.99 4.11 6.54
80 7.370 7.530 179.83 173.24 80 3.613 3.654 204.55 217.59
2.17 3.66 1.13 6.37

TABLE 5 Fault location errors for RF = 0 Ω

TABLE 3 VB and IF calculated for RF = 15 Ω
AFL Conv. Error Proposed Error
Error% Error%
(%) method (%) method (%)
AFL (%) IF _Act IF_cal VB _Act VB _cal 30 0.29980 0.07 0.29979 0.07
30 4.174 4.066 198.95 208.08 50 0.49944 0.11 0.49942 0.12
2.59 4.59 60 0.59921 0.13 0.59918 0.14
50 4.465 4.408 194.46 194.36 70 0.69893 0.15 0.69890 0.16
1.27 0.05 80 0.79865 0.17 0.79861 0.17
60 4.862 4.846 192.65 190.78
0.33 0.97
70 5.544 5.592 188.81 188.19 TABLE 6 Fault location errors for RF = 5 Ω
0.87 0.33 AFL Conv. Error Proposed Error
80 6.730 6.858 185.84 185.92 (%) method (%) method (%)
1.90 0.04 30 0.29473 1.76 0.30058 0.19
50 0.49171 1.66 0.50028 0.06
60 0.58877 1.87 0.59996 0.01
The results are compared with conventional distance protection
70 0.68342 2.37 0.69934 0.09
(Conv.) with various fault resistance values.
80 0.77137 3.58 0.79788 0.27
Actual location − Estimated location
Error ( % ) = ∗ 100. (38)
Actual location
Based on the results of Tables 5–10: 4.3 Estimation of computation time
∙ When the value of fault resistance is zero, both conventional It is critical to evaluate the computation time (Comp.) of the
and proposed methods are sufficiently accurate. However, proposed algorithm. Therefore, computation time is estimated
it should be noted that fault resistance is almost present in in various scenarios of fault resistance and fault location, as
practical short-circuit faults. depicted in Table 11 below.
∙ As the value of fault resistance increases, the accuracy of The simulation results show the time required to complete
conventional distance protection is significantly affected. the iterative procedure is typically 5 ms, even when the fault
However, the proposed method remains accurate even in the resistance values are increased at different fault locations, as
case of 50 Ω and 100 Ω fault resistances. Since, in practice, the illustrated in Table 11.
value of fault resistance rarely reaches 50 Ω, the performance It should be noted that in the proposed method, the itera-
of the proposed method in the presence of fault resistance is tive procedure is initialized by the approximate fault location
satisfactory. obtained by the conventional distance method, as shown in
∙ The inaccuracy of the VB estimation does not affect the fault Figure 4. In practice, this appropriate initialization leads to con-
location estimation. This indicates that the sensitivity of the vergence in all cases. However, in the case of unreasonable
fault location to the VB voltage is small. initialization, convergence is not guaranteed.
YASEEN ET AL. 2163

TABLE 7 Fault location errors for RF = 10 Ω TABLE 11 Estimation of computation time for RF = 0, 5, and 15 Ω

AFL Conv. Error Proposed Error AFL Comp. time Comp. time Comp. time
(%) method (%) method (%) (%) (RF = 0 𝛀) (RF = 5 𝛀) (RF = 15 𝛀)

30 0.28965 3.45 0.30153 0.51 30 5 ms 5 ms 5 ms

50 0.48397 3.21 0.50129 0.26 50 6 ms 5 ms 5 ms
60 0.57833 3.61 0.60089 0.15 60 5 ms 5 ms 6 ms
70 0.66790 4.59 0.69993 0.01 70 5 ms 5 ms 4 ms
80 0.74410 6.99 0.79724 0.35 80 5 ms 4 ms 5 ms

TABLE 8 Fault location errors for RF = 15 Ω TABLE 12 Calculation of the fault location for RF = 5 Ω (Scenario One)
AFL Conv. Error Proposed Error AFL Proposed Error
(%) method (%) method (%) (%) method (%)
30 0.28459 5.14 0.30264 0.88 30 0.30058 0.19
50 0.47624 4.75 0.50244 0.49 50 0.50028 0.06
60 0.56790 5.35 0.60197 0.33 60 0.59996 0.01
70 0.65239 6.80 0.70067 0.10 70 0.69934 0.09
80 0.71684 10.40 0.79669 0.41 80 0.79788 0.27

TABLE 9 Fault location errors for RF = 50 Ω

TABLE 13 Calculation of the fault location for RF = 5 Ω (Scenario Two)
AFL Conv. Error Proposed Error
AFL Proposed Error
(%) method (%) method (%)
(%) method (%)
30 0.24913 16.96 0.30590 1.97
30 0.30040 0.13
50 0.42217 15.57 0.50864 1.73
50 0.50015 0.03
60 0.49498 17.50 0.61010 1.68
60 0.59987 0.02
70 0.54399 22.29 0.70998 1.42
70 0.69933 0.10
80 0.52634 34.21 0.79518 0.60
80 0.79808 0.24

4.4 Performance assessment based on

ity. This indicates the robustness of the new technique for
renewable generation uncertainty, changes in
renewable generation fluctuation.
system condition ∙ Scenario Two: Table 13 shows that the proposed method’s
accuracy, which has a maximum error of 0.24%, is not
It is necessary to evaluate the robustness of the provided
affected by doubling the short-circuit capacity of the two
technique against fluctuations in renewable production output
external grids.
power due to maintenance or variability in weather conditions,
∙ Scenario Three: This step is the opposite of scenario two, in
as well as changes in the short-circuit capacity of the external
which the short-circuit capacity of the external grid is reduced
grid and system conditions. The following simulation tests have
to half of its initial value. Based on the results in Table 14, it
been performed with various scenarios.

∙ Scenario One: The results presented in Table 12 illustrate that TABLE 14 Calculation of the fault location for RF = 5 Ω (Scenario
the proposed technique has a maximum inaccuracy of 0.27% Three)
while using double the amount of renewable energy capac- AFL Proposed Error
(%) method (%)
TABLE 10 Fault location errors for RF = 100 Ω 30 0.30063 0.21
AFL Conv. Error Proposed Error 50 0.50041 0.08
(%) method (%) method (%) 60 0.60020 0.03
70 0.38960 44.34 0.68995 1.44 70 0.70071 0.10
80 0.25511 68.11 0.79464 0.67 80 0.80247 0.31
2164 YASEEN ET AL.

TABLE 15 Calculation of the fault location (Scenario Four) TABLE 18 Calculation of the fault location (Scenario Seven)

AFL Fault Proposed Error T.L. length AFL Proposed Error

(%) resistance method (%) (km) (%) method (%)

30 10 Ω 0.30153 0.51 100 70 0.70080 0.11

70 15 Ω 0.70067 0.09 50 80 0.79676 0.41

TABLE 16 Calculation of the fault location (Scenario Five) TABLE 19 Calculation of the fault location (Scenario Eight)

Impedance Fault AFL Proposed Error Line

of T.L. resistance (%) method (%) parameters AFL RF Proposed Error
error (%) (Ω) method (%)
200% 15 Ω 60 0.60092 0.15
Increasing 10% 60 10 0.60069 0.12
50 Ω 70 0.70547 0.78
Decreasing 10% 80 15 0.79599 0.50

TABLE 17 Calculation of the fault location (Scenario Six)

in Table 18. The maximum inaccuracy was 0.11% and 0.41%
AFL Fault Proposed Error for transmission lines that were 100 and 50 km long, respec-
(%) resistance method (%) tively, for different fault locations. These results show that
30 10 Ω 0.30204 0.68 the proposed method is still accurate for shorter transmission
70 15 Ω 0.70168 0.24 lines.
∙ Scenario Eight: Table 19 illustrates the simulation of the
proposed technique with the line parameters set to slightly
different values from their actual values in two different
can be seen that the proposed method can withstand consid- scenarios (increasing and decreasing by 10%). The results
erable variation in the external grid short-circuit capacity with demonstrate the effectiveness of the proposed method
a maximum error of 0.31%. against errors in the line parameters, with maximum errors
∙ Scenario Four: Table 15 demonstrates that reducing the of 0.12% and 0.50% for fault resistances 10 Ω and 15 Ω,
injected power of the converter-based source to half of respectively, for different fault locations.
its original value has no impact on the proposed method’s
accuracy, which has a maximum error of 0.51% and
0.09% for different fault resistances of 10 Ω and 15 Ω, 4.5 Performance evaluation based on
respectively. simultaneous and evolving faults
∙ Scenario Five: Table 16 shows the effectiveness of the pre-
sented method when the transmission line impedance is The validity of the proposed scheme is assessed in this sec-
doubled. In this case, the maximum fault location errors were tion by testing for simultaneous and evolving faults. Due to
0.15% and 0.78% for fault resistances of 15 Ω and 50 Ω, the fact that simultaneous faults involve faults in two differ-
respectively. As a result, the proposed scheme is robust to ent locations in different phases and evolving faults include
topological changes, and Equation (29) accurately calculates a primary fault and a secondary fault, they are more compli-
the fault current in terms of local information. In addition, cated in their nature than frequently occurring shunt faults and
according to Table 12, variations in the IIRES generation have a negative effect on the efficiency of the distance relaying
will not affect the proposed method either. Therefore, the scheme. Various scenarios were used in the simulation tests that
proposed method is robust against changes in the system follow.
specifications and parameters.
∙ Scenario Six: When the external grid 1 in Figure 1 is removed, ∙ Scenario One: Table 20 displays the results of the fault loca-
the transmission line contains conventional generation at one tion calculation for simultaneous faults occurring at phases a
terminal and IIRES at the other, making the system more and b in line 1 with different fault locations applied to the
realistic and allowing the new technique to be evaluated test system. The proposed method is clearly accurate and
without sharing any contribution of fault currents from con- can function effectively even when these faults are present.
ventional sources. Table 17 shows that the proposed method Obviously, since the proposed method is based on local infor-
is still accurate, with maximum errors of 0.68% and 0.24% for mation, in the case of simultaneous faults, the estimated fault
fault resistances of 10 Ω and 15 Ω, respectively, for different location is that of the fault nearest to the local relay.
fault locations. ∙ Scenario Two: The primary and secondary faults are con-
∙ Scenario Seven: The evaluation of the presented technique sidered to be at the same location but occurring at separate
for various transmission line lengths at RF = 5 Ω is shown times in this scenario of evolving faults. Consider an SL-G
YASEEN ET AL. 2165

TABLE 20 Calculation of the fault location for RF = 5 Ω (Scenario One) TABLE 24 Actual data 132 KV, single-circuit T.L

AFL Simultaneous Proposed Error Case

(%) faults method (%)
Actual data One Two
30 AG 0.30212 0.70
RF 0.26 Ω 0.23 Ω
50 BG
AFL% 43.7 41.6
50 AG 0.50464 0.92
iL1a 9.5743 KA 8.7935 KA
70 BG
iL2a 5.5492 A 484.56 A
iL3a 6.6143 A 1.1073 KA
uL1b 26.253 KV 23.875 KV
TABLE 21 Calculation of the fault location for RF = 5 Ω (Scenario Two)
uL2b 70.657 KV 69.088 KV
AFL Evolving Proposed Error
uL3b 68.755 KV 68.208 KV
(%) faults method (%)
a
50 AG 0.49593 0.81 iL1, 2, and 3: The fault current is in phases a, b, and c, respectively.
b
uL1, 2, and 3: The fault voltage is in phases a, b, and c, respectively.
BG

TABLE 25 Parameters for actual data 132 KV, single-circuit T.L

Length 11 Km
TABLE 22 Fault location errors for real data (Case One)
Positive-sequence resistance 0.0759 Ω/ Km
AFL Conv. Error Proposed Error
(%) method (%) method (%) Positive-sequence reactance 0.4794 Ω/ Km
Zero-sequence resistance 0.8237 Ω/ Km
0.437 0.43083 1.41 0.43781 0.19
Zero-sequence reactance 1.5227 Ω/ Km
Distance relay
SIEMENS-SIPROTEC (7SA522) Quadrilateral Characteristic
fault for phase (a) with a 50% fault location that occurred
at t = 70 msec. At t = 75 msec, an SL-G fault for
phase (b) happened at the same location. The result in
Table 21 demonstrates the efficacy of the proposed scheme As a result of the mentioned fault conditions, the con-
for evolving faults, with a maximum inaccuracy of 0.81% at ventional method’s maximum error for cases one and two
RF = 5 Ω. was 1.41% and 1.39%, respectively. However, the presented
method’s maximum error, which is more accurate, is 0.19%
for case one and 0.17% for case two. Furthermore, as men-
4.6 Performance assessment based on real tioned in Section 4.2, when an IIRES is connected to a nearby
data local substation, the inaccuracy of the conventional method will
increase as the value of the fault resistance rises, but the pro-
The performance of the proposed method is evaluated in two posed method still remains accurate. Tables 22 and 23 illustrate
actual cases of faults on the transmission line connecting the that the proposed scheme’s fault location is close to the real fault
132 KV Basra substation to the 132 KV Al-Shuaiba substation. location as determined by an advanced commercial protection
The faults are in Zone one of the distance protection system device.
located at the 132 KV Basra substation. The information in
Tables 24 and 25 were obtained from the distance relay of the
SIEMENS (SIPROTEC 7SA522) to protect the 132 KV single- 5 CONCLUSION
circuit overhead transmission line from an SL-G fault at fault
locations of 43.7% and 41.6% at fault resistances of 0.26 Ω and This article provides a comprehensive analysis to illustrate
0.23 Ω, respectively. the influence of the IIRES on distance protection based on
Tables 22 and 23 compare the results of the proposed practical grid code for a single-phase to ground fault. It is evi-
technique’s application to those of the traditional distance relay. dent that the limitation of amplitude and phase offset of the
Based on the results in Tables 22 and 23: short-circuit current produced by IIRES causes the traditional
distance protection to malfunction.
An improved relay scheme based on local information is
TABLE 23 Fault location errors for real data (Case Two) proposed, which does not require any communication links.
AFL Conv. Error Proposed Error Moreover, the algorithm is compatible with the existing pilot
(%) method (%) method (%) protection methods used in the industry.
0.416 0.41022 1.39 0.41669 0.17 Furthermore, the simulation results show that the system can
effectively determine the fault location with more than 99%
2166 YASEEN ET AL.

accuracy and is not affected by the value of fault resistance. The IEEE 7th International Power Engineering and Optimization Conference
proposed algorithm is proven to have a fast computation time (PEOCO), pp. 213–217. IEEE, Piscataway (2013). [Link]
1109/PEOCO.2013.6564545
and to be immune to renewable generation uncertainty, changes
8. Alam, M.M., Leite, H., Liang, J., da Silva Carvalho, A.: Effects of VSC
in the system specifications and parameters, and simultaneous based HVDC system on distance protection of transmission lines. Int. J.
and evolving faults. Based on actual data, the proposed tech- Electr. Power Energy Syst. 92, 245–260 (2017)
nique’s fault location is close to the fault location as determined 9. Cabrera-Tobar, A., Bullich-Massagué, E., Aragüés-Peñalba, M., Gomis-
by an advanced commercial protection device. Bellmunt, O.: Review of advanced grid requirements for the integration of
large scale photovoltaic power plants in the transmission system. Renew.
Sustain. Energy Rev. 62, 971–987 (2016). doi: [Link]
rser.2016.05.044
CONFLICT OF INTEREST STATEMENT 10. Brussels, B.: ENTSO-E network code for requirements
The authors certify that they have No Conflict of Interest to for grid connection applica_ble to all generators. Available:
disclose. [Link]
130308_Final_Version_NC_RfG1.pdf
11. Fang, Y., Jia, K., Yang, Z., Li, Y., Bi, T.: Impact of inverter-interfaced renew-
FUNDING INFORMATION able energy generators on distance protection and an improved scheme.
IEEE Trans. Ind. Electron. 66(9), 7078–7088 (2019). [Link]
This research received no specific grant from any finding agency 1109/TIE.2018.2873521
in the public, commercial, or not-for-profit sectors. 12. Chakeri, V., Seyedi, H., Tarafdar Hagh, M.: A new approach to transmis-
sion line pilot protection in the presence of inverter-interfaced distributed
generators. IEEE Syst. J. 15(4), 5383–5392 (2021). [Link]
CREDIT CONTRIBUTION STATEMENT 1109/JSYST.2020.3041203
All persons who meet authorship criteria are listed as authors, 13. Yang, Z., Jia, K., Fang, Y., Zhu, Z., Yang, B., Bi, T.: High-frequency fault
component-based distance protection for large renewable power plants.
and all authors certify that they have participated sufficiently in
IEEE Trans. Power Electron. 35(10), 10352–10362 (2020). [Link]
the work to take public responsibility for the content, including org/10.1109/TPEL.2020.2978266
participation in the concept, design, analysis, writing, or revision 14. Hooshyar, A., Azzouz, M.A., El-Saadany, E.F.: Distance protec-
of the manuscript. Furthermore, each author certifies that this tion of lines emanating from full-scale converter-interfaced renewable
or similar material has not been submitted to or published in energy power plants—Part I: Problem statement. IEEE Trans. Power
Deliv. 30(4), 1770–1780 (2015). [Link]
any other publication prior to its appearance in IET Generation,
2369479
Transmission, and Distribution. 15. Hooshyar, A., Azzouz, M.A., El-Saadany, E.F.: Distance protection of lines
emanating from full-scale converter-interfaced renewable energy power
plants—Part II: Solution description and evaluation. IEEE Trans. Power
DATA AVAILABILITY STATEMENT Deliv. 30(4), 1781–1791 (2015). [Link]
The data that support the findings of this study are available 2369480
from the corresponding author upon reasonable request. 16. Banaiemoqadam, A., Hooshyar, A., Azzouz, M.A.: A control-based
solution for distance protection of lines connected to converter-
interfaced sources during asymmetrical faults. IEEE Trans. Power
REFERENCES Deliv. 35(3), 1455–1466 (2020). [Link]
1. Brouwer, A.S., van den Broek, M., Seebregts, A., Faaij, A.: Operational 2946757
flexibility and economics of power plants in future low-carbon power sys- 17. El-Arroudi, K., Joós, G.: Performance of interconnection protection based
tems. Appl. Energy 156, 107–128 (2015). doi: [Link] on distance relaying for wind power distributed generation. IEEE Trans.
apenergy.2015.06.065 Power Deliv. 33(2), 620–629 (2017)
2. Nimpitiwan, N., Heydt, G.T., Ayyanar, R., Suryanarayanan, S.: Fault current 18. Chen, S., Tai, N., Fan, C., Liu, J., Hong, S.: Sequence-component-based
contribution from synchronous machine and inverter based distributed current differential protection for transmission lines connected with IIGs.
generators. IEEE Trans. Power Deliv. 22(1), 634–641 (2007). [Link] IET Gener. Transm. Distrib. 12(12), 3086–3096 (2018). doi: [Link]
org/10.1109/TPWRD.2006.881440 org/10.1049/iet-gtd.2017.1507
3. Hooshyar, H., Baran, M.E.: Fault analysis on distribution feeders with high 19. IEEE Standard C37.118.1-2011. IEEE Standard for Synchrophasor
penetration of PV systems. IEEE Trans. Power Syst. 28(3), 2890–2896 Measurements for Power Systems
(2013). [Link] 20. Ziegler, G.: Numerical Distance Protection: Principles and Applications.
4. Bains, T.P.S., Sidhu, T.S., Xu, Z., Voloh, I., Zadeh, M.R.D.: Impedance- John Wiley & Sons, Hoboken, NJ (2011)
based fault location algorithm for ground faults in series-capacitor- 21. Technical data: SiPROTEC 5 distance protection, line differential pro-
compensated transmission lines. IEEE Trans. Power Deliv. 33(1), 189–199 tection and breaker management for 1-Pole and 3-Pole tripping V7.50
(2018). [Link] and higher. Siemens, Germany (2017). Available: [Link]
5. Telukunta, V., Pradhan, J., Agrawal, A., Singh, M., Srivani, S.G.: Protection [Link]/download-center/downl
challenges under bulk penetration of renewable energy resources in power 22. Technical manual: MiCOMho P446 – Distance relay, Alstom Grid.
systems: A review. CSEE J. Power Energy Syst. 3(4), 365–379 (2017). doi: Stafford, UK (2015). Available: [Link]
10.17775/CSEEJPES.2017.00030 [Link]?prod=P443&type=3
6. Mohammedsaeed, E.K., Jia, K.: Comparison of some low voltage ride- 23. Tsili, M., Papathanassiou, S.: A review of grid code technical requirements
through strategies. In: IECON 2017 - 43rd Annual Conference of the for wind farms. IET Renew. Power Gener. 3(3), 308–332 (2009)
IEEE Industrial Electronics Society, pp. 152–157. IEEE, Piscataway 24. Karbouj, H., Rather, Z.H.: A comparative study on the impact of grid
(2017). [Link] code regulations on stability of wind integrated power systems. In: 2019
7. Idris, M.H., Ahmad, M.S., Abdullah, A.Z., Hardi, S.: Adaptive Mho type 1st Global Power, Energy and Communication Conference (GPECOM),
distance relaying scheme with fault resistance compensation. In: 2013 pp. 342–347. IEEE, Piscataway (2019)
YASEEN ET AL. 2167

25. LeBlanc, M.P., Evans, L., Gardner, P., Scott, N.C.: Canadian grid code for 30. IEEE Guide for Determining Fault Location on AC Transmission
wind development: review and recommendations (2006) and Distribution Lines, IEEE Std C37.114-2014 (Revision of IEEE
26. Ackermann, T., et al.: Code shift: Grid specifications and dynamic wind Std C37.114-2004), pp. 1–76 (2015). [Link]
turbine models. IEEE Power Energy Mag. 11(6), 72–82 (2013) 2015.7024095
27. European Network of Transmission System Operators for Electricity:
Short circuit contribution of new generating units connected with power
electronics and protection behaviour (2019)
28. Howard, D.F., Liang, J., Harley, R.G.: Short-circuit modeling of DFIGs How to cite this article: Yaseen, N.K., Seyedi, H.,
with uninterrupted control. IEEE J. Emerg. Sel. Top. Power Electron. 2(1), Abapour, M.: Transmission line protection using local
47–57 (2013) information in the presence of inverter-interfaced
29. Makwana, V.H., Bhalja, B.R.: A new digital distance relaying scheme for
compensation of high-resistance faults on transmission line. IEEE Trans.
renewable energy sources. IET Gener. Transm. Distrib.
Power Deliv. 27(4), 2133–2140 (2012). [Link] 17, 2156–2168 (2023).
2012.2202922 [Link]
2168 YASEEN ET AL.

APPENDIX

TABLE A1 Parameters for the simulation mode

Devices Parameters Values

VSC-wind farms Offshore wind farms 2*240 MVA

The “HVDC linked Offshore Wind Farm, Digsilent Power Factory examples” design and parameters for offshore wind farms are readily accessible.
External grids Equivalent positive-sequence impedance, seen from bus B 0.5413+7.741i Ω
Equivalent positive-sequence impedance, seen from bus C 0.5413+7.741i Ω
Single-circuit transmission line parameters Length 200 Km
Positives-sequence resistance 0.032 Ω/ Km
Positive-sequence reactance 0.3184 Ω/ Km
Positive-sequence susceptance 3.653672 µS/Km
Zero-sequence resistance 0.256 Ω/ Km
Zero-sequence reactance 1.174 Ω/ Km
Zero-sequence susceptance 2.411612 µS/Km
Distance relay CT and VT CT 1000/1 A
VT 380 KV/100 V
Distance relay Mho relay