Challenges and Solutions in the

Protection of Transmission Lines

Connecting Nonconventional Power Sources

Marcelo Bini, EDF Renewables

Ricardo Abboud, Paulo Lima, and Fabio Lollo,

Schweitzer Engineering Laboratories, Inc.

Revised edition released August 2023

Previously revised edition released January 2022

Originally presented at the

48th Annual Western Protective Relay Conference, October 2021
 1

Challenges and Solutions in the

Protection of Transmission Lines
Connecting Nonconventional Power Sources
Marcelo Bini, EDF Renewables
Ricardo Abboud, Paulo Lima, and Fabio Lollo, Schweitzer Engineering Laboratories, Inc.

Abstract—This paper discusses the impact of inverter-based obtaining the negative-sequence component in
resources (IBRs) in traditional digital protection relays applied in electromechanical and solid-state relays requires complex and
the interconnection transmission line between the IBR and bulk
more expensive analog filters [7] [8]. These new functions
power system. Real events involving a photovoltaic (PV) power
plant are used to show the behavior of the fault currents, which is significantly improve the sensitivity, reliability, and security of
different from power systems with synchronous generators, transmission line protective functions and schemes, allowing
especially for negative-sequence components. The paper discusses for the secure detection of challenging line faults.
how to properly handle this kind of source by presenting modern This paper evaluates how nonconventional power sources
protective relays features, time-domain functions, and special affect the security and reliability of phasor-based transmission
settings for traditional protection intelligent electronic devices
line protection, especially the protective functions based on
(IEDs). The data of the real events were used to validate the
solutions proposed in this case study. The lessons learned will be symmetrical components. First, the paper briefly reviews the
applied in a new facility (a wind farm) that is currently in the short circuit contribution from nonconventional power sources,
design stage. focusing on the IBRs. The paper reviews the main transmission
line protective features based on negative-sequence
I. INTRODUCTION components and their benefits in terms of sensitivity, reliability,
Over the last decade, the renewable energy share of the and security.
electricity sector grew quickly. Renewable energy installations, A case study evaluating the performance of phasor-based
especially solar photovoltaic (PV) facilities and wind turbines, protective features in a 138 kV transmission line connecting a
are increasingly deployed today. The contribution of these 420 MW PV solar power plant to the bulk power system is
sources to the power system affects the performance of presented. Data from real fault events in this transmission
conventional protection systems, particularly for faults on system are used to evaluate the line protective functions’
transmission lines that connect such power sources to the bulk performance.
power system. Most of these renewable power sources are The paper evaluates the performance of improved phasor-
inverter-based, and their characteristics change according to based and time-domain protective functions that are designed
manufacturer design and specifications. The fault current to provide reliable performance of the transmission line
contribution from an inverter-based resource (IBR) is limited protection system connecting IBR to the grid. Data captured
by the current capacity of its power electronic components and from real fault events are used to test and assess the
its control functions. Typically, the maximum current capacity effectiveness and performance of the improved phasor-based
of these power components does not exceed 1.5 times the full- and time-domain protective features, and the results are
load current during the fault steady-state. Additionally, the presented and discussed.
IBRs do not have the same amount of rotational inertia as a Finally, the paper proposes protection philosophy
synchronous generator [1]. Usually, the IBR does not deliver combining phasor-based elements and time-domain
reliable negative-sequence and zero-sequence quantities, as the incremental quantities. It is based on the validation made with
synchronous generators would do. The IBR’s atypical behavior field data in this paper and previous papers’ conclusions.
and characteristics challenge the phasor-based protective Possible modifications in the phasor-based function setting to
functions when compared to synchronous generators, because improve security, as shown in [9], are also considered.
the assumptions made in their design are no longer valid,
especially assumptions about the fault current contribution II. IBR BEHAVIOR DURING POWER SYSTEM FAULTS
magnitude and associated negative-sequence behavior [1]. A. Dynamic Behavior of Wind Turbines During
Numerical relay benefits and advantages are well known in Short Circuits
the power industry, and the ease of obtaining the symmetrical Wind turbines are complex systems that transform the
components from the measured phase quantities is one of those kinetic energy of the wind into electrical energy, and wind
advantages. It allows the implementation of several new generation is already responsible for a significant portion of the
functions based on symmetrical components, especially the Brazilian energy matrix [10]. One of the specific features of
negative-sequence component [2] [3] [4] [5] [6], since wind generation is the high variability of wind strength, which
 2

translates into variations in turbine rotations. For a generator, The PV modules are composed of a set of interconnected
varying rotation means a varying generated signal frequency. solar cells that transform energy. The cell connections are
Since there is a tendency for the frequency in the generator to directly related to the voltage and electric current that the
vary and a simultaneous need to keep the frequency of the module needs to supply to comply with the manufacturing
electrical system constant, several solutions have been design [16].
developed to allow coupling between the wind turbine and the PV generation systems consist of the connection of solar
power system. modules in series, in parallel, or as a combination of the two in
Wind turbines are commonly classified into different types, which each string is made up of a certain number of modules to
according to their operation and construction. Naturally, each obtain the desired voltage and current levels. Generally, power
type of generator contributes differently when there are faults is the technical specification most used to classify PV modules,
in the system. There are many studies that seek to find the best and the unit widely used is peak watt (Wp).
way to model the different types of machines to understand The complete system of a solar plant, at the generation level,
their behavior during system faults [11]. Good modeling allows is composed of the following equipment:
users to feel more confident when calculating generator • Solar modules
contributions for short circuit and yard equipment strength • Combiner box
studies. • Inverter
References [12] and [13] present the Type I and Type II • Step-up transformer
wind turbine characteristics and fault contributions, To produce the necessary power, the solar modules are
respectively. The contributions of Type III generators during interconnected and form strings. These strings are brought
faults can be divided into two behaviors, which depend on together in a combiner box, and the outputs of the combiner box
whether or not the crowbar is active. When the crowbar is are connected to the ac/dc inverters, which convert the dc
active, the power electronics do not control the field winding; voltage to ac voltage using a three-phase rectifier. The outputs
therefore, the behavior is similar to that of the induction of the inverters are linked to step-up transformers to connect the
generator. Also, when the generator with the activated crowbar energy at medium voltage level to a collector substation, which
is contributing, some components can have frequencies other then raises the voltage to make the generated energy available
than nominal frequency, since the axis rotation of the machine through the power system.
is not necessarily in sync with the network. The highest short Grid-connected solar power plants contribute to short circuit
circuit levels of the Type III wind turbine are obtained when the currents during a fault, modifying the short circuit characteristic
crowbar is active [14]. in the power system. The contribution to the short circuit
When the crowbar is not active and the network is under current depends on a variety of factors, including the maximum
fault conditions, wind turbine contributions are determined by current level supported by the inverter, inverter control systems,
the power electronics. In this situation, the control network environmental conditions, protection system, location, and type
developed by the manufacturer is largely responsible for of fault.
determining the dynamic response of the machine. We have
seen different responses among different manufacturers and III. FIELD CASE ANALYSIS
different wind turbines from the same manufacturer.
Fig. 1 shows the main features of the system used as a case
Type IV has its short circuit current fully controlled by the
study. The field case study is based on a solar PV power plant
power electronics. These devices are sized to carry a current
located in Brazil, which has an installed capacity of
level slightly above the nominal current, which limits the
approximately 420 MW. The PV is connected to Brazil’s
contribution of these generators to a range of 1.1 to 1.3 pu of
national interconnected system (GRID) through three
the nominal current for most of the time [1] [15]. Another
three-winding step-up transformers, as shown in Fig. 1:
characteristic is the low capacity to supply negative-sequence
• 34.5 kV sector—PV distribution voltage level with
current; the short circuit current of this type of generator tends
resistance-grounded and Y-connected step-up
to be balanced, even for unbalanced faults.
transformer winding. The inverters are connected to
B. Behavior of PV Power Plants During Short Circuits the 34.5 kV bus through 0.38/34.5 kV step-up
Generation of electrical energy through PV panels involves transformers, not shown in Fig. 1.
converting sunlight into electrical energy. This conversion • 138 kV sector—PV transmission voltage level with
occurs due to the semiconductor characteristics of the material solid-grounded and Y-connected transformer winding
(normally silicon), which is used in PV cells due to its for connection to the GRID through the 138 kV
commonality. transmission line with a length of 10.37 km
(6.44 miles).
• Delta (D) tertiary winding for harmonic filtering with
no load connected.
 3

The transmission line in analysis has protective relays

installed at both terminals, indicated in the diagram as IED_PV
and IED_GRID.

Fig. 1. Simplified single-line diagram of the PV used in the case study.

The applied intelligent electronic device (IED) is a Fig. 2. Oscillography records with current (amperes), voltage (kilovolts),
transmission line protective relay, and the main enabled and digital signals from the PV and GRID terminals for the external AG fault.
functions for circuit breaker (CB) tripping are as follows: The small dip in the A-phase voltage suggests that it was a
• Line differential protection—phase elements (87LA, short circuit in the grid relatively distant (electrically) from the
87LB and 87LC), negative-sequence element (87LQ), transmission line in question. Field personnel reports that the
and zero-sequence element (87LG) are enabled fault occurred approximately 200 km (124 miles) away from
• Phase and ground distance time-delayed elements the transmission line in question.
(Z2T)—Zone 1 is not enabled due to short line length Fig. 3 shows the waveforms for each phase and both
and PV fault contribution behavior terminals separately. This makes it easy to observe that the
• Directional comparison teleprotection scheme with current entering the line through the PV terminal is equal to the
permissive overreach transfer trip (POTT) by current leaving the line through the GRID terminal (the currents
overreaching distance (Z2) and neutral directional from both sides are in opposition of phase). This proves that it
overcurrent (67G2) element was an external fault to the line.
• Direct transfer trip (DTT)
• Emergency overcurrent protection (51E)—only
enabled in case of simultaneous loss-of-potential and
loss of differential channel communications
• Phase overvoltage protection (59P)
There is an AG fault in the grid, outside the 138 kV
interconnection transmission line. At the time of the fault, the
PV is connected generating energy; therefore, there is a
contribution from the PV to the fault.
Fig. 2 shows the currents and voltages during the AG fault
external to the transmission line. Users can conclude that it is
an AG short circuit due to the voltage sag in the A-phase and
the current increase in this phase at the beginning of the short
circuit. Current increases in the other phases and distortions in
the waveforms are a challenge for the protection schemes that
are currently employed.
For this fault, the 138 kV line protection should not cause
the opening of CBs. However, as Fig. 2 shows, the IED_GRID Fig. 3. Currents by phase (amperes) at both line terminals.
trips (2:TRIP). The current flowing through the IEDs at both
line terminals is the contribution of the PV, which has high The rest of this paper provides a more detailed analysis of
distortion and is not consistent with the contribution of a the behavior of the protection functions during this event.
synchronous machine to an AG fault.
 4

A. Fault-Type Identification Logic

In faults involving the ground, the protection IEDs compare
the negative-sequence current (I2) and zero-sequence current
(I0) phasors to identify which type of fault is occurring in the
system. Using the A-phase as a reference for calculating the
symmetrical components for an AG or BCG (FSA) fault, IA2
and IA0 are expected to be in-phase. For a BG or CAG (FSB)
fault, IA2 is expected to lag IA0 by 120 degrees, and for a CG
or ABG (FSC) fault, IA2 is expected to lead IA0 by 120 degrees.
Fig. 4 shows these expected relationships. To identify whether
the fault is single-phase or two-phase to the ground, a
comparison is made between the apparent impedance of the
respective fault loops. The one with the lowest apparent
impedance is the fault loop [6].

Fig. 4. Relationship between I2 and I0 for AG, BG, CG, BCG, CAG, and
ABG faults.

In the event analyzed, I2 has unstable behavior, as seen in

Fig. 5, which shows the filtered current and voltages at the
GRID terminal. This same behavior occurs at both transmission
line terminals, since the PV fault current contribution passes
through both terminals, as also reported in [1].
After adopting the A-phase voltage (VA) as the angular
reference, the following I2 behavior is observed during the
event: Fig. 5. Fault current (amperes) and behavior of I2 in relation to I0 during AG
• At the beginning of the fault, I2 is in-phase with I0, a fault in the system, studied at different times during the fault.
characteristic behavior of an AG or BCG fault Reference [6] fully describes a modern implementation of
(Fig. 5a). fault-type identification and how incorrect selection affects the
• At the middle of the fault, I2 leads I0 by approximately safety and reliability of protection functions.
120 degrees, a characteristic behavior of a CG or ABG
fault (Fig. 5b). B. POTT
• At the end of the fault, I2 lags I0 by approximately The GRID terminal circuit breaker is tripped by the POTT
140 degrees, a characteristic behavior of a BG or CAG scheme. Fig. 6 shows the currents of both terminals with
fault (Fig. 5c). indications of the digital elements of both IEDs.
I0 remains stable in respect to VA throughout the fault, with The IED_PV initially sees the fault as forward (1:32GF) and
an angular difference compatible with an AG fault. The transfers the permissive signal to the GRID terminal (1:KEY).
D-winding of the 34.5/138 kV step-up transformers in the PV In relation to the PV terminal, the fault is actually in the forward
provides a low-impedance path for the zero-sequence current, direction.
and this causes the GRID to contribute a portion of the I0 current The IED_GRID receives the permissive signal from the PV
that flows in the direction from the PV to the GRID. Thus, I0 is terminal (2:PT) and, when it sees the fault directionality as
less dependent on the behavior of the inverters and has a more forward (2:32GF), it issues a tripping command via the POTT
predictable behavior in this type of system, given that the GRID scheme (2:TRPRM). For the GRID terminal, the fault should
has mostly conventional sources. Fig. 5 also shows the response be declared as reverse.
of the faulted phase selection logic at the GRID terminal for this Despite initially having declared the fault as forward
event. (1:32GF), the IED_PV starts to define the fault as reverse
This unexpected behavior of the relationship between I2 and (1:32GR) at approximately 830 ms, at the same time that the
I0 leads relays that use this methodology to misidentify the fault IED_GRID defines the fault as forward (2:32GF). We will
type and enable unsuitable fault loops. This can have analyze what caused this behavior with further details. The
consequences, such as underreaching or overreaching distance negative-sequence directional element (32Q) was set to provide
protection, errors in the fault-locating function, and incorrect directional decisions for the residual ground directional
openings in cases when single-pole tripping is applied.
 5

overcurrent element (67G2), and this was the element I2R and V2 indicates a clear reverse fault (shown in Fig. 9b),
responsible for the POTT scheme operation. justifying the change of directionality determined by this relay.
In a traditional negative-sequence directional element, the
angular relationship between the negative-sequence voltage
(V2 and I2) determines the fault direction. For a forward fault,
I2 • 1<Z1ANG (I2R) is expected to be 180 degrees out with V2;
whereas, the reverse fault I2R is expected to be in-phase with V2
[17], as shown in Fig. 7. When setting V2 as the angular
reference, I2R is expected to be in the blue region for forward
faults and in the red region for reverse faults. The region in Fig. 8. Operating principles of element 32Q during a forward fault (a) and
black is where the direction is undefined. during a reverse fault (b).

Fig. 6. Oscillography records with current signals (amperes) and indications Fig. 9. Fault current (amperes) and relationship between I2R and V2 for the
of the protections activated on the PV and GRID terminals for the external IED_PV at different times during the fault.
AG fault.
For the IED_GRID, at the beginning of the fault, the
relationship between I2R and V2 is in an undefined region, and
the IED did not determine the fault to be in either direction
(shown in Fig. 10a). At a later time, the relationship between
I2R and V2 indicates a forward fault condition (shown in
Fig. 10b), justifying the IED’s determination that the direction
of the fault was forward.

Fig. 7. Relationship between V2 and I2R for forward and reverse faults.

The protection IEDs used in the analyzed line have a 32Q

element based on measuring the negative-sequence impedance
[18]. For a forward fault, the relationship between V2 and I2 is
the negative sequence of the source impedance (–ZS), as shown
in Fig. 8a. For a reverse fault, the relationship between V2 and
I2 is the sum of the line impedance and the remote source
impedance (ZL + ZR), as shown in Fig. 8b [1].
Here, we will use only the relationship between V2 and I2R
to simplify the analysis. For the IED_PV, at the beginning of
the fault, the relationship between I2R and V2 is consistent with
a forward fault but very close to the undefined direction (shown
in Fig. 9a). However, at a later time, the relationship between Fig. 10. Fault current (amperes) and relationship between I2R and V2 for the
IED_GRID at different times during the fault.
 6

A similar analysis can be performed to assess how a

directional element would behave based on zero-sequence
current (I0) and zero-sequence voltage (V0). The same diagram
can be used for this (shown in Fig. 7) if V2 is replaced with V0
and I2R is replaced with I0R. The IED in question has a
directional element based on the measurement of the zero-
sequence impedance (32V); its operational principle is similar
to that of the 32Q element (shown in Fig. 8).
Fig. 11 shows that at the PV terminal, the angular
relationship between I0R and V0 is compatible with a forward
fault, and the relationship remains this way throughout the fault.

Fig. 12. Fault current (amperes) and relationship between V0 and I0R for the
IED_GRID at different times during the fault.

The protection of transmission lines in the vicinity of

unconventional sources results in an additional complication
and may leave protection engineers without a reliable
alternative for determining the directionality of faults.
C. Line Differential Protection Element (87L)
The applied IED has a line differential element (87L) with
an alpha plane operating principle [19] and provides the phase
Fig. 11. Fault current (amperes) and relationship between V0 and I0R for the differential elements (87LA, 87LB and 87LC), negative-
IED_PV at different times during the fault.
sequence element (87LQ), and zero-sequence element (87LG).
Fig. 12 shows that at the GRID terminal, the angular Fig. 13 shows the filtered phase, negative-sequence, and
relationship between I0R and V0 is compatible with a fault in the zero-sequence currents measured at each line terminal, as well
reverse direction, and the relationship remains this way as the respective differential current. The currents recorded at
throughout the fault. This stable, predictable behavior of each terminal have the same magnitude and a 180 degrees
zero-sequence quantities shows that in this type of system, these difference, resulting in a differential current that equals zero.
quantities are more appropriate than negative-sequence
quantities. This is mainly due to the connection of the PV
transformers, which allow the system to contribute
zero-sequence through the PV terminal. In other words, the
zero-sequence that mostly passes through the PV terminal
during an external fault is actually contributed by the system
and not by the inverters.
For this particular case, I0R- and V0-based directional
elements are shown to be reliable; however, in some
applications, like in parallel lines, the zero-sequence mutual
coupling can be a challenge when applying the zero-sequence
directional element. The negative-sequence directional element
has always been the natural choice of protection engineers in
these cases [3] [17].
Additionally, the negative-sequence directional element is
the main method for determining the directionality of
unbalanced faults between phases and is used to supervise
phase overcurrent elements and phase distance elements in Fig. 13. Filtered phase, negative-sequence, and zero-sequence currents (in
traditional phasor-based implementations [18]. per unit) measured at each line terminal.
 7

In the external fault condition, despite the unstable behavior IV. EVALUATION OF NEW PROTECTIVE FEATURES
of the unconventional source, its contribution is a current In this section, the performance of new protection functions
passing through the line; therefore, this does not cause a for events in the system is evaluated. Both phasor-based
differential current in the phase, negative-sequence, and zero- protection functions were modified to better adapt to the
sequence elements. contribution of IBRs to faults, and new time-domain-based
For an internal fault in the protected line, the IED at the functions are evaluated.
GRID terminal measures the current contributed by the system,
while the IED at the PV terminal measures the current A. Incremental-Quantity Directional Element (TD32)
contributed by the PV plant. This is not a problem for the phase Reference [21] demonstrates the operational principle of the
and zero-sequence differential elements, given the TD32 element, which operates based on the incremental
characteristics of the system. For the negative-sequence voltage (∆v) and the incremental replica current (∆iZ).
differential element, it is not a problem at first, since the For a forward fault, the relationship between ∆v and ∆iZ is
contribution of the GRID is significantly greater than the the negative of the absolute value of the source impedance
contribution of the PV. As this paper is being written, no behind the relay (ZS), according to (1).
internal fault events on the line are available for analysis. ∆v =− ZS • ∆i Z (1)
Considering that the contribution to an internal fault at the
GRID terminal is much greater than that at the PV terminal, the For a reverse fault, the relationship between ∆v and ∆iZ is
87LQ element will also operate reliably under unbalanced the absolute value of the sum of the protected line impedance
internal faults. (ZL) with the source impedance of the remote terminal (ZR),
according to (2).
D. Distance Element (21)
There was no distance element pickup for this event because ∆v= ZL + ZR • ∆i Z (2)
the fault was far from the protected line. Fig. 14 shows the Mho The graphical representation of ∆v and ∆iZ during system
Zone 2 element characteristic and AG loop apparent faults helps us understand how the relationship between these
impedance. During the fault, the apparent impedance varies quantities indicates the fault direction (shown in Fig. 15) [21].
greatly.
In the case where the fault is within the overreaching zone,
this variation in apparent impedance during the fault could
cause the apparent impedance to move in and out of the zone,
preventing the element from operating and acting as a backup
for external faults in the system. References [1] and [20] show
how the behavior of unconventional sources affects distance
relays.

Fig. 15. Incremental voltage and incremental replica current for a forward
fault (a) and for a reverse fault (b).

After observing Fig. 15, we can conclude that the

incremental voltage and the incremental replica current have
similar waveforms and that the relationship between their
polarities clearly indicates the fault direction: quantities have
an opposite polarity for forward faults, and quantities are
in-phase for reverse faults. Additionally, the relationship
between their magnitudes is related to the system impedances.
Reference [21] details the implementation of the high-speed
TD32 element, based on the relationship between ∆v and ∆iZ.
The next section evaluates the performance of the TD32
element during the events in the system under study. To do this,
Fig. 14. Fault current (amperes), voltage (kilovolts), and behavior of the
Mho Zone 2 and AG loop apparent impedance for the external AG fault. we will play back the real field events in an IED with the TD32
function implemented [22].
 8

1) Case 1: External AG Fault 2) Case 2: External BG Fault Without Power

This is the same fault assessed in Section III. Fig. 16 shows Generation at PV
the behavior of the incremental voltage and incremental replica This external BG fault represents a condition different from
current at the PV terminal for the external AG fault. As the external AG fault described previously, since it occurred at
expected, at the beginning of the fault, these quantities have a time when the PV had zero generation. Thus, the only current
opposite polarities, and the relay classifies the fault as forward contribution to the fault was the zero-sequence current of the
(TD32F = 1). system through the transformers at the PV terminal.
Additionally, this fault is much closer to the protected line than
the previous fault.
Fig. 18 shows the behavior of the incremental voltage and
incremental replica current at the PV terminal for this fault. As
expected, at the beginning of the fault, these quantities have
opposite polarities, and the relay reliably classifies the fault as
forward (TD32F = 1).

Fig. 16. Fault current (amperes), voltage (kilovolts), incremental voltage

(kilovolts), and incremental replica current (amperes) at the PV terminal for
an external AG fault.

Fig. 17 shows the behavior of the incremental voltage and

incremental replica current at the GRID terminal for the
external AG fault. As expected, at the beginning of the fault,
these quantities have the same polarity, and the relay classifies
the fault as reverse (TD32R = 1).
Fig. 18. Fault current (amperes), voltage (kilovolts), incremental voltage
(kilovolts), and incremental replica current (amperes) at the PV terminal for
an external BG fault.

Fig. 19 shows the behavior of the incremental voltage and

incremental replica current at the GRID terminal for this fault.
As expected, at the beginning of the fault, these quantities are
in-phase, and the relay reliably classifies the fault as reverse
(TD32R = 1).

Fig. 17. Fault current (amperes), voltage (kilovolts), incremental voltage

(kilovolts), and incremental replica current (amperes) at the GRID terminal
for an external AG fault.
 9

C. Enhanced Phasor-Based and Incremental-Quantity Fault-

Type Identification Logics
The IED of [22] implements a faulted phase selection logic
based on incremental quantities. This logic is important for
supervising protection elements based on incremental
quantities, such as the TD32 element and the underreaching
distance element based on incremental quantities (TD21) [23].
For pilot protection with single-pole trip schemes, the
correct identification of the phase under fault is also needed to
trip the correct phase. This logic based on incremental
quantities is used in [22] in pilot protection schemes based on
TD32, TD21, and TW32.
As shown previously, the elements based on incremental
quantities behaved properly during real faults in the system
under study. These functions can be safely applied to perform
fast tripping for internal faults in the protected transmission
line.
However, the incremental quantities are available for a
limited time after the fault begins. This time is determined by
the number of periods used in the calculation of the incremental
Fig. 19. Current (amperes), voltage (kilovolts), incremental voltage quantity [21]. Thus, these functions do not provide backup
(kilovolts), and incremental replica current (amperes) at the GRID terminal protection for external faults in the system, similar to 87L
for an external BG fault.
function. Hence, the protection system must be complemented
The operational situation described in this case is quite with phasor-based functions.
common, because it is a solar plant and there is no energy One of the challenges described in Section III for phasor
generation at night. protection is the correct identification of the phase under fault
Reference [17] shows how this kind of fault with only zero- due to the unexpected behavior of the symmetrical components
sequence current can cause improper operation of directional in the contribution of the IBR.
elements with a traditional quadrature connection; this is more The IED of [22] also implements phasor-based fault-type
critical, specifically in short lines, because the high zero- identification logic (FID) that uses the negative- and zero-
sequence current can reach the phase directional overcurrent sequence voltages in addition to the negative and zero-sequence
element pickup level. Both the TD32 element and the 32V currents. The objective of using sequence voltages is to allow
element shown previously are safe for this type of fault. the FID logic to operate reliably under weak infeed conditions
At the time of writing this paper, we only have faults or when sequence currents are small and unreliable, as in the
involving the ground to evaluate the performance of the TD32 case of IBRs [22]. The logic calculates two quantities, as seen
element; however, [20] considers the use of TD32 reliable for in (3) and (4).
all fault types in this kind of application. S2 =3I 2 •1∠Z1ANG − H 2 • 3V2 (3)
B. Traveling-Wave (TW) Differential (TW87) and TW S0 =3I0 •1∠Z1ANG − H 0 • 3V0 (4)
Directional (TW32) Elements
Reference [21] describes the principles and fundamentals of where:
protection functions based on current and voltage TWs, namely 3I2 is the negative-sequence current
TW32 and TW87. 3I0 is the zero-sequence current
These elements operate based on the high frequency Z1ANG is the positive-sequence line impedance angle
transient signals generated by the fault and are independent of 3V2 is the negative-sequence voltage
the source’s contribution to the fault in the first place. Thus, 3V0 is the zero-sequence voltage
these elements are alternatives for protecting lines where one or S2 is the negative-sequence quantity
more terminals have IBR connected.
S0 is the zero-sequence quantity
This paper does not intend to evaluate the performance of
these elements, since there are no high frequency sampled H2 and H0 are constants
records on the order of MHz available in the system evaluated
here.
 10

In addition to S2 and S0, the logic also uses the positive- Given the applications and operating principle of this logic,
sequence voltage (V1) in determining the fault type when there it should work for faults in the forward direction.
is no ground involved and also to determine whether the fault We evaluated the performance of the improved
is, for example, CG or ABG. Fig. 20 shows the expected phasor-based fault-type identification logic and the
angular relationship between these quantities for each fault type incremental-quantity-based faulted phase selection logic for the
[22]. In Fig. 20, the fault is BCG, confirmed both by the events in Case 1 and Case 2, presented above.
relationship between S0 and S2 and by the relationship between In Fig. 21, events marked with “1” are from the GRID
S2 and V1. terminal and those marked with “2” are from the PV terminal.
The IED of the PV terminal correctly identifies the fault as AG
by using the improved phasor-based element (2:FIDAG) and
using the element based on incremental quantities (2:FSAG).
The IED of the GRID terminal correctly identifies the fault
using the element based on incremental quantities (1:FSAG),
and there is no indication of using the improved phasor element
(1:FIDAG), as expected, since the fault is in the reverse
direction.

Fig. 21. Fault current (amperes), voltage (kilovolts), performance of the

Fig. 20. The relationship between S0 and S2 (a) and the relationship between improved phasor-based faulted phase selection logic (1:FIDAG and
S2 and V1 (b) during unbalanced faults. 2:FIDAG), and the incremental-quantity-based faulted phase selection logic
(1:FSAG and 2: FSAG) for Case 1.
In [22], this logic is used for improved identification of
phase under fault and is used in the following applications: For Case 2, the PV terminal IED correctly identifies the fault
• To allow single-pole tripping in pilot protection as BG by using the improved phasor-based element (2:FIDBG)
schemes with 67G and 67Q and using the incremental-quantity-based element (2:FSBG), as
• To allow single-pole tripping in the pilot protection shown in Fig. 22. The IED of the GRID terminal correctly
scheme using weak infeed logic with 59G and 59Q identifies the fault using the incremental-quantity-based
• To improve fault-type signaling element (1:FSBG), and there was no indication of using the
• To improve fault-type selection for the impedance- improved phasor-based element (1:FIDBG), as expected, as the
based fault-locating function fault is in the reverse direction for this line end.
• To improve single-pole tripping in DTT schemes
without phase-separated bits
 11

V. LINE PROTECTION SYSTEM CONFIGURATION

FOR THE NEW WIND FARM ENTERPRISE
A new wind farm is being built, which will connect to the
national interconnected system (GRID) through a single-
circuit, 500 kV overhead transmission line with a length of
27 km (16.8 mi).
This section discusses how the new protection system for
this transmission line is being designed. Given the experience
protecting the 138 kV transmission line at the PV and the
performance available using the new protection elements
described in Section IV, a relay that combines time-domain
with phasor-based features was chosen for the Main 1
protection and a phasor-based line differential relay was chosen
for the Main 2 protection of this new transmission line.
Fig. 24 shows the simplified single-line diagram of the wind
farm and its connection with GRID.

Fig. 22. Fault current (amperes), voltage (kilovolts), performance of the

improved phasor-based faulted phase selection logic (1:FIDBG and
2:FIDBG), and the incremental-quantity-based faulted phase selection logic
(1:FSBG and 2: FSBG) for Case 2.

Fig. 23 shows the performance of the faulted phase selection Fig. 24. Simplified single-line diagram of new EDF Renewables wind farm
and its connection with GRID.
logic for a CAG fault external to the protected line. For this
case, only the PV event is available. Both logic options A. Pilot Protection Scheme
correctly identify the phases under fault. Additionally, the The transmission lines connecting the wind farm with GRID
correct behavior of the TD32 element is shown. have an optical ground wire (OPGW). The POTT pilot
protection scheme will be adopted.
The following directional elements are available:
• Main 1 protection IED
− Directional element based on incremental
quantities (TD32)
− Directional element based on TWs (TW32)
− Negative-sequence impedance directional
element (32Q)
− Zero-sequence impedance directional
element (32V)
− Phase directional element (32P-1)
• Main 2 protection IED
− Negative-sequence impedance directional
element (32Q)
− Zero-sequence impedance directional
element (32V)
− Phase directional element (32P-2)
The phase directional elements (32P) in each IED are
classified with different codes as they have different operating
principles:
• 32P-1—phase directional element polarized with
Fig. 23. Fault current (amperes), voltage (kilovolts), performance of the positive-sequence voltage memory
improved phasor-based faulted phase selection logic (FIDCAG), incremental-
quantity-based faulted phase selection logic (FSCA), and TD32F for an
• 32P-2—phase directional element based on the
external CAG fault. positive-sequence impedance angle, obtained from the
relationship between the positive-sequence voltage
The improved method described was proven to be safer and
and positive-sequence current
more robust than the traditional method based only on
symmetrical component currents in the events evaluated.
 12

1) Ground Faults strategy presented above, 32P-2 is also available for unbalanced
For faults involving ground, the TD32 and 32V elements phase-to-phase faults.
ensure reliable indication of the fault direction. These elements For three-phase faults (3P), the TD32, 32P-1 and 32P-2
have been shown to be safe for the events evaluated and have elements are available. Reference [1] shows that 32P may lose
high sensitivity for detecting high-impedance ground faults. security for 3P GRID faults and [9] suggests an enhanced 32P
The TD32 element is directly associated with the directional that can be implemented by current supervision and simple
comparison teleprotection scheme for sending the permissive logic. This enhanced 32P element is used to supervise all phase
signal. The POTT scheme uses a current-supervised TD32 elements.
element (TD67) for tripping. Likewise, the POTT scheme uses The POTT scheme uses current-supervised 32Q (67Q) and
a zero-sequence current-supervised 32V element (67G) to send 32P-1 or 32P-2 elements (67P) to send the permissive signal
permissive signal and for tripping. and for tripping. At the PV terminal, these elements do not
Even given the low fault contribution of the wind farm, the operate for a line fault with the 50FP setting strategic presented
connection of the wind farm’s step-up transformer above and trip at this terminal is based on weak infeed and
(34.5/500 kV) allows it to contribute with zero-sequence echo-converted-to-trip (ECTT) logic. In the Main 1 relay, the
current through the wind farm terminal for line faults. weak infeed logic can be triggered by phase undervoltage
(27P), negative-sequence overvoltage (59Q), or zero-sequence
2) Phase-to-Phase and Three-Phase Faults
For phase-to-phase faults (2P) without ground involvement, overvoltage (59G). In the Main 2 relay, phase-to-phase (27PP)
the TD32 element will also be employed to indicate the and 59G are available
direction of the faults. For both 2P and 3P elements, the TD32 element has the
As shown in Section III, the use of the negative sequence sensitivity to quickly identify and declare forward faults, even
can result in safety and reliability problems given the with the low magnitude of the IBR contribution, as discussed in
contribution of the IBR terminal, but its use is desirable for [20]. The way TD32 is implemented in the POTT scheme in
identification of the direction for faults between phases without [22] requires current supervision for TD32 (TD67) only for
ground involvement, since the 32V element is unavailable to tripping and not for sending the permissive signal. In the case
assist in this case. of an internal fault, both terminals will send the permissive
Reference [9] suggests a strategy for applying the 32Q signal; however, only the GRID side will have enough current
function in this type of system. This strategy consists of to trip by the POTT. To ensure a fast trip at the IBR side, it is
defining a setting for the forward and reverse fault supervision necessary to use a DTT.
overcurrent elements (50FP and 50RP, respectively) of the 32Q B. Line Current Differential Protection
function with a threshold high enough to guarantee the security The Main 2 relay has a line differential function based on
of these elements. the alpha plane, providing phase differential elements (87LA,
In this application, at the IBR terminal, 50FP and 50RP are 87LB, and 87LC), a negative-sequence differential element
set so that 32Q is enabled only for reverse faults. At the GRID (87LQ), and a zero-sequence differential element (87LG).
terminal, 50FP and 50RP are set so that 32Q is enabled for line As shown in Section III, all these elements remain safe for
faults. These settings allow the elements to be enabled by the external faults. Even with the unexpected behavior of I2 from
GRID contribution current only, where the relationship the IBR terminal, the same current passes through both line
between V2 and I2 is coherent with the fault direction. Since the terminals in the event of external faults. Thus, all differential
IBR contribution is limited, this strategy allows the function to elements can be enabled. Reference [9] also evaluates the
have the sensitivity to detect reverse unbalanced faults at the security and reliability of 87L function for IBR connecting
IBR terminal, increasing the security of the hybrid POTT transmission lines.
scheme that uses weak infeed and echo logic.
For the Main 2 relay, when 50FP is not activated, the fault- C. Distance and Step Distance Protection
type identification function stops comparing the phasors of I0 As mentioned previously, the application of the distance
and I2 and switches to a voltage-based faulted phase function at the wind farm terminal leads to safety and reliability
identification logic if the weak infeed logic is enabled, which is issues. An instantaneous Zone 1 element may overreach for
reliable for terminals with a weak source [9]. Therefore, proper external faults, and a time delay Zone 2 element may not be
setting of the 50FP is doubly beneficial in this IED. In addition able to provide backup for the system due to the variation in the
to fault location, fault-type signaling, and single-pole trip, FID apparent impedance.
logic in this IED also supervises distance elements. The enhanced 32P element can supervise phase distance
Additionally, the 32P-1 element works independently and is elements at both line terminals. Additionally, to prevent Zone 1
always available for phase-to-phase faults. As with the phase overreaching, the phase Zone 1 element can be limited using
directional element with a quadrature connection, there can be the distance element overcurrent supervision or adding a time
a conflict between 32P-1 elements of different phases in faults delay; these two conditions can also be combined in an OR gate.
like Case 2 of Section III. Adding 67G and 67Q supervision in Adding a dropout delay in the Zone 2 element can avoid
the pilot protection scheme improves security in this case. element dropout, providing dependable backup. All these
In the Main 2 IED, the 32Q element has priority over the solutions are suggested and detailed in [9].
32P-2 element. When 32Q is not enabled, per the setting
 13

Since the connection between the wind farm and the GRID However, the termination of the line at the IBR is through a
is radial in this system, a distance element with an extended transformer, which means that ZT has a much higher value
Zone 1 can be applied to the GRID terminal relay, with a reach compared to ZC. This causes iR and iI to have opposite polarities
beyond the 500 kV bus of the wind farm, as shown in Fig. 25. and causes iTW to have a magnitude smaller than the incident
This increases instantaneous coverage for line faults, especially wave, which makes it difficult to detect.
in the event of communications channel failures. Additionally, Reference [25] shows actual records of current TWs in a
at the GRID terminal, the TD21 element can be applied with a transformer termination made by a TW-based fault-locating
reach near to the remote bus. device in a configuration similar to the system studied here. The
experience of this real system shows that, despite the
attenuation in the magnitude, detection is still possible. Since
the magnitude of the TW signal generated in the fault depends
on the voltage incidence angle, the signal attenuation at the
wind farm terminal reduces the sensitivity of TW-based
functions, especially for faults that occur far from the voltage
peak.
In the system in analysis, it will be applied using the
Fig. 25. Extended Zone 1 improves instantaneous coverage for line faults. following TW-based functions:
This strategy cannot be applied if more equipment • TW-based directional element (TW32)
(transmission lines or transformers) gets connected to the • TW-based line differential element (TW87)
500 kV bus of the wind farm terminal in the future. • TW-based fault location with data from two terminals
For these faults, the current that passes through the line (DETWFL) [25]
terminals is the current coming from the system, which is • Predictive line monitoring based on TWs [26]
mostly composed of synchronous machines; therefore, the In cases when the IBR terminal is connected to more
distance functions can operate normally. transmission lines, the low impedance increases the reliability
of the TW functions.
D. TW-Based Features
Faults in the transmission line generate voltage and current E. Line Protection System Reliability
transients that propagate toward the line terminals. When an Most of the protection strategies described are dependent on
incident wave (iI) reaches the line terminal, part is transmitted the availability of the communications channel between the line
(iT) and part is reflected (iR), as shown in Fig. 26 [24]. terminals, excluding the distance elements. Therefore, to ensure
the availability of the protection system, the means of
communication must be highly available and redundant.
This becomes even more critical in applications when other
transmission lines are connected to the IBR bus or more than
one transformer, since the strategy of extending the reach of
Zone 1 is not fully applicable.
The National Electric System Operator (ONS) of Brazilian
Fig. 26. Representation of incident (iI), reflected (iR), and transmitted (iT) grid requires, at a minimum, the total redundancy of the
waves at a transmission line terminal.
communications channel for all facilities with a voltage level
The portion of the wave that is reflected and transmitted equal to or greater than 230 kV.
depends on the ratio between the line’s characteristic Usually, transmission lines, especially newer ones, have an
impedance (ZC) and the equivalent terminal impedance (ZT), OPGW cable, which provides a reliable communications
which can be determined using (5) and (6). channel for both 87L protection and pilot protection schemes,
ZC − ZT either via direct fiber or through multiplexers.
iR = iI (5) To provide redundancy, an alternative route through nearby
ZC + ZT
substations is an option. In this case, for radially connected
2  ZC IBRs, alternative routes are unlikely to be available.
iT = iI (6) Communication redundancy can be obtained through a second
ZC + ZT
OPGW cable, power line carrier (PLC), or spread-spectrum
An IED that monitors the current signal at the line terminal radios for short lines. These last two means of communication
sees the sum of the incident and reflected currents (iTW) do not support or ideally meet the requirements of the
according to (7). 87L function [27], making it difficult to adopt the 87L function
i TW= i I + i R (7) with full redundancy.
Thus, pilot protection using a directional comparison
At the GRID terminal, two other transmission lines are
scheme is the natural alternative for these communications
connected, so ZT is smaller than ZC. This causes iR to have the
channels. The use of the TD32 function in the pilot protection
same polarity as iI and causes iTW to have a magnitude greater
than the incident wave, which makes it easier to detect.
 14

scheme, as detailed previously, increases dependability for fast VII. ACKNOWLEDGMENT

elimination of internal faults in this type of system. The authors acknowledge Ritwik Chowdhury and Ryan
For GRID terminal bus faults and adjacent line faults, it is McDaniel of Schweitzer Engineering Laboratories, Inc. (SEL)
more challenging to provide remote backup with the wind farm for their time spent reviewing this paper and having technical
terminal IED. The use of redundant local protection methods, discussions.
including bus and circuit breaker failure protection, should be
considered. The ONS’s minimum requirement is redundancy VIII. REFERENCES
for all these protection schemes in facilities with a voltage [1] R. Chowdhury and N. Fischer, “Transmission Line Protection for
greater than or equal to 230 kV. Systems With Inverter-Based Resources – Part I: Problems,” IEEE
In the event of a total failure of communication with the Transactions on Power Delivery, August 2020. Available:
GRID terminal, it might not be possible to eliminate the fault at doi.org/10.1109/TPWRD.2020.3019990.
the wind farm terminal. In this case, [9] suggests the use of a [2] B. Kasztenny, N. Fischer, and H. Altuve, “Negative-Sequence
Differential Protection – Principles, Sensitivity, and Security,”
long-time undervoltage element for backup. Enabling the 27P proceedings of the 41st Annual Western Protective Relay Conference,
element or even a distance function only when the two College Station, TX, October 2014.
communications channels are lost is an alternative that should [3] K. Zimmerman and J. Mooney, “Comparing Ground Directional
be evaluated. Element Performance Using Field Data,” proceedings of the 8th Annual
Conference for Fault and Disturbance Analysis, College Station, TX,
April 1993.
VI. CONCLUSION
[4] R. Abboud, P. Lima, and A. Pontizelli, “Utilização de IEC 61850
This paper evaluated how an inverter-based source’s Valores Amostrados (SV) e Sincrofasores para Localização de Faltas em
atypical behavior and characteristics challenge the actual Tempo Real,” proceedings of the Symposium of Experts in Operation
Planning and Electric Expansion, Foz do Iguaçu, Paraná, Brazil,
phasor-based protective functions, including symmetrical-
May 2014.
components-based directional elements, fault-type [5] N. Fischer, D. Finney, and D. Taylor, “How to Determine the
identification logic, and distance elements. Field data from a Effectiveness of Generator Differential Protection,” proceedings of the
138 kV transmission line connecting a 420 MW PV to the 40th Annual Western Protective Relay Conference, College Station, TX,
GRID were evaluated, and the issues related to IBR’s fault October 2013.
current contribution behavior were discussed. [6] D. Costello and K. Zimmerman, “Determining the Faulted Phase,”
proceedings of the 36th Annual Western Protective Relay Conference,
Additionally, the data from the events were used to evaluate College Station, TX, October 2009.
the performance of improved phasor-based and new time- [7] F. Calero, “Rebirth of Negative-Sequence Quantities in Protective
domain protective features presented in modern protection Relaying With Microprocessor-Based Relays,” proceedings of the
relays. These elements proved to be more secure and reliable 30th Annual Western Protective Relay Conference, College Station, TX,
for these systems. October 2003.
The TD32 element was reliable for the evaluated field cases. [8] W. Elmore, Protective Relaying Theory and Applications,
Marcel Dekker, Inc., New York, NY, 1994.
The use of this element in a directional comparison pilot
[9] R. Chowdhury and N. Fischer, “Transmission Line Protection for
protection scheme could improve the protection reliability. Systems With Inverter-Based Resources – Part II: Solutions,”
The 87L function was secure for the evaluated events and proceedings of the IEEE Transactions on Power Delivery, Virtual
was also presented as an excellent protection scheme for the Format, October 2020. Available: ieeexplore.ieee.org/document/
9233925.
IBR connecting transmission lines in previous papers; however,
[10] Ministry of Mines and Energy, “Resenha Energética Brasileira,”
two redundant and independent communication channels March 2020. Available: antigo.mme.gov.br/web/guest/secretarias/
suitable for 87L are not always available. A directional planejamento-e-desenvolvimento-energetico/publicacoes/resenha-
comparison pilot protection scheme was more flexible in terms energetica-brasileira.
of the communication channel requirements and allowed the [11] E. Senger and F. Filho, “Plantas eólicas: modelagem para estudos de
deployment of a fully redundant protection system for these curto-circuito e critérios de ajustes das proteções,” proceedings of the
Protection and Control Technical Seminar, Rio de Janeiro, Brazil, 2014.
systems when applied with TD32.
[12] M. Nunes, “Avaliação do comportamento de aerogeradores de
A protection philosophy combining phasor-based elements velocidade fixa e variável integrados em redes elétricas fracas,”
and time-domain incremental quantities was proposed, based PhD dissertation, UFSC Institutional Repository, 2003. Available:
on the validation made with field data in the conclusion of this repositorio.ufsc.br/xmlui/handle/123456789/85795.
paper and previous papers. This protection philosophy will be [13] E. Pinheiro, “Análise do Comportamento Dinâmico de Usinas Eólicas a
applied in the transmission line connecting the new wind farm Velocidade Variável Utilizando ATPDraw,” PhD dissertation, Federal
University of Minas Gerais, 2004.
that is currently in the design stage.
[14] J. Marques, “Turbinas eólicas: modelo, análise e controle do gerador de
Considerations about transmission line protection system indução com dupla alimentação,” PhD disseration, Federal University of
and system backup protection reliability were discussed. Santa Maria, 2004.
Brazilian grid protection system minimum requirements were [15] B. Kasztenny, V. Mynam, and N. Fischer “Sequence Component
also accounted for. Applications in Protective Relays – Advantages, Limitations, and
Solutions,” proceedings of the 46th Annual Western Protective Relay
Finally, the use of TW-based protection features to protect
Conference, Spokane, WA, October 2019.
the transmission lines connecting IBR is in the authors’ interest [16] ABB, “Technical Application Papers No. 10: Photovoltaic Plants,”
and will be evaluated for the new wind farm enterprise in Available: abb.com.
Brazil.
 15

[17] J. Roberts and A. Guzmán, “Directional Element Design and Paulo Lima received his BSEE in electrical engineering from Federal
Evaluation,” proceedings of the 21st Annual Western Protective Relay University of Itajubá, Brazil, in 2012. In 2013, he joined Schweitzer
Conference, Spokane, WA, October 1994. Engineering Laboratories, Inc. (SEL) as a protection application engineer in
[18] B. Fleming, “Negative-Sequence Impedance Directional Element,” Brazil. In 2018, he became an application engineering group coordinator, and
proceedings of the 10th Annual ProTest User Group Meeting, he has been the regional technical manager for Brazil since 2020. He has
experience in application, training, integration, and testing of digital protective
Pasadena, CA, February 1998.
relays. He also provides technical writing and training associated with SEL
[19] J. Roberts, D. Tziouvaras, G. Benmouyal, and H. Altuve, “The Effect of products and SEL University.
Multiprinciple Line Protection on Dependability and Security,”
proceedings of the 55th Annual Georgia Tech Protective Relaying
Fabio Lollo received his BSEE in electrical engineering from São Paulo State
Conference, Atlanta, GA, May 2001. University (Bauru campus), Brazil, in 1997. In 1998, he joined BMG
[20] B. Kasztenny, “Distance Elements for Line Protection Applications Engineering as an electrical engineer, working with protection and control
Near Unconventional Sources,” June 2021. Available: selinc.com. design and protection studies. In 2013, he joined Schweitzer Engineering
[21] E. O. Schweitzer, III, B. Kasztenny, A. Guzmán, V. Skendzic, and Laboratories, Inc. (SEL) as senior protection engineer in Brazil. In 2016, he
V. Mynam, “Speed of Line Protection – Can We Break Free of Phasor became design and studies group coordinator. Currently, he is a senior
Limitations?” proceedings of the 41st Annual Western Protective Relay protection engineer and supervises a group of project engineers with SEL
Conference, Spokane, WA, October 2014. Engineering Services, Inc. (SEL ES). He has experience in short circuit
calculation and protection settings for industrial and utility systems.
[22] SEL-T401L Ultra-High-Speed Line Relay Instruction Manual.
Available: selinc.com.
[23] E. O. Schweitzer, III, and B. Kasztenny, “Distance Protection: Why
Have We Started With a Circle, Does It Matter, and What Else Is Out
There?” proceedings of the 71st Annual Conference for Protective Relay
Engineers, College Station, TX, March 2018.
[24] A. Guzmán, B. Kasztenny, Y. Tong, and V. Mynam, “Accurate and
Economical Traveling-Wave Fault Locating Without
Communications,” proceedings of the 71st Annual Conference for
Protective Relay Engineers, College Station, TX, March 2018.
[25] E. O. Schweitzer, III, A. Guzmán, V. Mynam, V. Skendzic, B.
Kasztenny, and S. Marx, “Locating Faults by the Traveling Waves They
Launch,” proceedings of the 67th Annual Conference for Protective
Relay Engineers, College Station, TX, April 2014.
[26] B. Kasztenny, V. Mynam, T. Joshi, and D. Holmbo, “Preventing Line
Faults With Continuous Monitoring Based on Current Traveling
Waves,” proceedings of the 15th International Conference on
Developments in Power System Protection Liverpool, United Kingdom,
March 2020.
[27] B. Kasztenny, N. Fischer, K. Fodero, and A. Zvarych, “Communications
and Data Synchronization for Line Current Differential Schemes,”
proceedings of the 2nd Annual Protection, Automation, and Control
World Conference, Dublin, Ireland, June 2011.

IX. BIOGRAPHIES
Marcelo Bini received his BSEE in electrical engineering from Santa Catarina
State University (Joinville campus), Brazil, in 2006. In 2006, he joined Engevix
Engineering as an electrical engineer, working with hydroelectric power plant
projects with emphasis on protection, control, and measurement systems. In
2017, he joined Araxá Solar as a substation design engineer. In 2019, he joined
EDF Renewables as an electrical engineer, working with protection and control
design and other electrical projects involved in substations, transmission lines,
and medium voltage networks.

Ricardo Abboud received his BSEE degree in electrical engineering from

Federal University of Uberlândia, Brazil, in 1992. In 1993, he joined CPFL
Energia as a protection engineer. In 2000, he joined Schweitzer Engineering
Laboratories, Inc. (SEL) as a field application engineer in Brazil, assisting
customers in substation protection, automation, and control systems. In 2005,
he became the field engineering manager, leading and mentoring the
application engineering group, and in 2014, he became the engineering services
manager. In 2016, he transferred to SEL headquarters in Pullman, Washington,
as an international technical manager, providing advanced technical support
and consultancy about new technologies to international field offices. In 2019,
he joined SEL University as a professor. Currently, he is a principal engineer
with SEL Engineering Services, Inc. (SEL ES).

© 2021, 2022, 2023 by EDF Renewables and

Schweitzer Engineering Laboratories, Inc.
All rights reserved.
20230815 • TP7011-01