Munich Personal RePEc Archive

Superimposed Sequence Components for

Microgrid Protection: A Review

Opoku, Kwasi and Pokharel, Subash and Dimitrovski,

Aleksandar

University of Central Florida

24 December 2021

Online at https://mpra.ub.uni-muenchen.de/112027/
MPRA Paper No. 112027, posted 18 Feb 2022 05:53 UTC
Superimposed Sequence Components for Microgrid
Protection: A Review
Kwasi Opoku, Subash Pokharel, and Aleksandar Dimitrovski
Department of Electrical & Computer Engineering
University of Central Florida
Orlando, Florida 32826
Email: [email protected], [email protected], [email protected]

Abstract—The new challenge to protective relaying in distri- issues affect the ability of traditional protection schemes to
bution networks (DNs), due to the integration of distributed achieve required sensitivity and speed of fault detection, as
generation (DG), has become a significant area of focus and well as selectivity and security of fault section location.
research for power engineers. To achieve high penetration of
DGs in the DNs, new methods of achieving desired sensitivity, Many researchers have proposed solutions to solve micro-
selectivity, and security of fault detection and coordination grids’ protection issues, and new approaches continuously
must be adopted. Recent literature shows that superimposed evolve. The methods differ by network complexity, type,
sequence components offer great potential in fault detection and capacity, DG connection, and the available measurements
coordination in such DNs. This paper reviews different proposed
solutions to fault detection and coordination in microgrids using and infrastructure reliability. However, the common goal of
superimposed sequence quantities. It also includes a discussion protection schemes is to achieve fast, selective, and sensitive
on their application methods, unique advantages, and limitations. protection most economically and efficiently.
This way, it contributes to existing reviews on microgrid protec-
tion by presenting the unique considerations necessary for the Some solutions build on conventional protection schemes,
superimposed method, as well the way conventional protection while others apply novel techniques and methods. The con-
schemes can be improved for microgrid protection using this ventional protection schemes have been updated by includ-
method. Results from applying a new approach for detecting ing the detailed model of IIDGs [7] and considering their
faults based on superimposed negative sequence admittance control strategy [8]. Impedance-based protection schemes, for
is presented to demonstrate the application of superimposed
quantities in fault detection. instance [9], [10], detect the faults using new measurement
Index Terms—Active distribution network, distributed gener- quantities and communication systems. The rapid development
ation (DG), microgrid protection, superimposed sequence com- of communication techniques in distribution networks has
ponent, low voltage ride-through (LVRT) made efficient information exchange possible resulting in the
protection schemes like multi-agent-based current-differential
I. I NTRODUCTION protection, wide area protection schemes, etc. [11]. The issue
The global electric power system is going through a tran- of data transmission bandwidth and synchronization makes
sition mainly because of the changing landscape of energy such methods uneconomical for distribution systems. The
policy and the economics of distributed generations (DGs). traveling wave (TW) based protection schemes presented in
This has resulted in the proliferation of renewable energy [12] detect the fault by comparing the time of initial traveling
resources throughout the power system. The active distribution waves. However, this approach is considered not well-suited
networks (ADNs) and microgrids are the cornerstones of for short feeders, and some non-fault events might affect
future smart power grids [1], [2]. An effective protection the TW. With adaptive protection schemes [13], [14], the
system is a prerequisite for any power system operation. An relay settings are automatically readjusted in response to
ideal protection scheme for a microgrid protects it from all system conditions. However, the pre-assessment of all possible
types of faults, works for both modes of operations: grid- configurations is challenging to achieve, and implementation
connected, and islanded and can adapt to the plug-and-play cost is very high. Several voltage-independent fault detection
functionality [3]. methods are presented in [15]–[17], where there is no voltage
The new challenges faced by conventional protective relays measurement. Latest developments in areas of computation
in active distribution networks have been identified and ex- and data analytics, with communication, enabled the innova-
tensively discussed [3]–[5]. These challenges are generally as tive protection schemes based on feed-forward neural networks
a result of the the bi-directional flow of both load and fault [18].
currents, the different types of distributed energy resources A set of quantities that seem to offer great potential for
in the network, and the dynamic nature of microgrids with effective fault detection are the superimposed sequence com-
constantly changing topology and operation modes leading ponents, also known as the incremental fault components,
to variation of fault currents. Also, due to the rating of the which have been employed in transmission systems for many
inverters, fault current contribution of inverter-interfaced DGs decades. They have the advantage of not being influenced
(IIDGs) is limited to about 1.2pu in steady state [6]. These by the load conditions in the pre-fault and faulted states.
Hence, both the magnitude and angle of such incremental
This material is based upon work supported by the U.S. Department of quantities can be applied in various schemes to provide a fast,
Energy’s Office of Energy Efficiency and Renewable Energy (EERE) under sensitive, and selective method of detecting faults and fault
the Solar Energy Technologies Office Award Number DE-EE0002243-2144.
The views expressed herein do not necessarily represent the views of the locations in microgrids. This paper reviews the schemes that
U.S. Department of Energy or the United States Government. employ superimposed fault quantities and assesses their unique

978-1-6654-7902-8/22/$31.00 ©2022 IEEE

advantages and limitations.
The remainder of this article is organized as follows. Section
II introduces the theory of superimposed quantities. It is
followed in Section III with solutions that have been explored
for microgrid protection by applying superimposed quantities.
A case study to demonstrate how the technique is applied in
a negative sequence admittance detection method is presented
in Section IV. Section V discusses some disadvantages of the Fig. 3. Pure Fault Network
superimposed quantities, and Section VI concludes the paper.
II. T HEORY OF S UPERIMPOSED Q UANTITIES
to be pre-fault phase currents in phases A, B and C, and
When a fault occurs in a linear network, a pure fault Ia ′ , Ib ′ , Ic ′ to be the corresponding faulted quantities. Then
circuit can be obtained from the pre-fault circuit and the the pre-fault positive sequence quantities can be obtained as,
faulted circuit using the principle of superposition. The current
and voltage values obtained in the pure fault circuit are the 1
superimposed quantities. I1 = (Ia + a2 Ib + aIc ) (3)
3
Consider a basic network, with line MN with impedance
ZL which connects the grid side (with impedance ZM ) at bus Also, the fault-state positive sequence component I1 ′ , can be
M, and a DG (with impedance ZN ) at bus N. A fault, f1 obtained from the phase currents Ia ′ , Ib ′ , Ic ′ :
occurs on the line, at a distance x from bus M, as shown on 1 ′
the equivalent circuit in Fig.1. I1′ =(I + a2 Ib′ + aIc′ ) (4)
Using the superposition theorem, the voltages and currents 3 a
in the faulted network (Fig.2) can be obtained as a sum of the From (2), superimposed positive sequence current ∆I1 , can
pre-fault circuit (Fig.1) and a pure fault circuit (Fig.3). In this be obtained from the phase quantities by:
case, the pure fault circuit is obtained by replacing the voltage 1
sources in the pre-fault circuit with a short circuit and their ∆I1 = I1′ − I1 = (∆Ia + a2 ∆Ib + a∆Ic ) (5)
internal impedance [19]. The fault voltage source Vf is also 3
represented by a fault impedance Zf . The magnitude of Vf Similar analysis could be used to obtain superimposed nega-
is equal to the voltage at the fault point before the incidence tive sequence current ∆I2 , positive sequence voltage ∆V1 and
of fault. Thus, the superimposed quantities obtained from the negative sequence voltage ∆V2 . The superimposed sequence
pure fault circuit are assumed to be driven only by the fault, impedance (∆Z) and admittance (∆Y ) could be found using
i.e., changes in the pre-fault condition. This is different from ∆V and ∆I.
the faulted circuit which includes both the prefault condition Modern microprocessor-based relays are able to derive
and the change that occurs due to the fault. these quantities with the use of memory, together with faster
sampling rates.
III. A PPLICATION OF S UPERIMPOSED S EQUENCE
Q UANTITIES
Many solutions for microgrid protection involve modifica-
tion of conventional schemes, considering the unique char-
acteristics of such microgrids. One such modification is the
application of superimposed quantities as the quantities used
Fig. 1. Pre-fault Network to determine protection criteria. A description of various
applications incorporating superimposed quantities methods in
conventional schemes are presented in this section.
A. Adaptive Protection Schemes
A common method employed to deal with the unique
challenges of microgrid protection is adaptive protection.
The adaptive protection method is an early approach ini-
tially used in transmission networks to modify protection
settings in response to changes in the network conditions
Fig. 2. Faulted Network [20]. This approach is therefore extensively applied in mi-
crogrids considering the changing conditions in microgrids.
The superimposed quantities, voltage (∆V ) and current This way, the schemes are applicable in both islanded and
(∆I), are therefore incremental quantities obtained by com- grid-connected modes and with changing network topology.
paring the pre-fault and faulted quantities, as given by (1) and Superimposed components have been deployed together with
(2): adaptive techniques in various schemes to provide distinct
Vf ault = Vpre−f ault + ∆V (1) measuring quantities to help achieve the required selectivity
If ault = Ipre−f ault + ∆I (2) and sensitivity of the protection system.
For instance, in [21], an adaptive overcurrent scheme was
Similarly, the symmetrical components of the superimposed developed for protection coordination in a microgrid with
phase quantities can be obtained. Let us consider Ia , Ib , Ic both IIDGs and rotation type DGs (RTDGs.) Using |∆I1 |
and |∆I2 |, an impact factor is developed depending on the component to develop criteria limited to balanced faults. A
microgrid mode of operation and the type of DGs connected negative sequence-based scheme was proposed for all other
in the network at a particular time. This impact factor is then faults.
used to compute an adaptive fault current. A limitation to this The direction of fault is determined in [22] by observing
solution is that it uses a simple signaling method to determine the phase difference between the pre-fault negative sequence
topology by using existing breaker status indications in the current, I2pre , and ∆I2 . The negative sequence currents have
distribution network. However, for distribution networks with the advantage of being present in all asymmetric faults; thus,
high penetration of DGs and continuously varying network an additional condition would be required for balanced faults.
conditions, additional methods such as directional scheme, as However, in [29], the amplitude change between I1pre and
in [22] (discussed in part C below), should be included to ∆I1 is deployed in a directional scheme to determine the fault
improve protection selectivity [23]. direction. It is, therefore, applicable for all fault types.
As an improvement to the traditional T32Q element imple-
B. Wide Area Protection Schemes mented in [30], the solution in [31], proposed a new directional
Many of the solutions proposed for microgrid protection element for unbalanced fault direction detection using both
take advantage of communication infrastructure in the distri- magnitude and angle of ∆Z2 . For symmetrical faults, ∆Z1
bution network for information sharing among various agents was proposed for direction detection. This way, the challenges
(intelligent electronic devices – IEDs, computing devices, and of the T32Q element (security of the protection when reactive
switching devices). This, therefore, enables the implementa- current is generated by the inverter and sensitivity issues
tion of wide-area protection schemes in ADNs. during the forward unbalanced faults) were eliminated. As
For instance, to identify the faulted section in a microgrid, discussed in [32], the performance of this scheme is affected
the solution proposed in [24] divided the network into inten- for faults that lead to the absence of power frequency and
tional islands bounded by intelligent circuit breakers (CBs). subsequent weak output of the power generator.
These CBs can communicate the ∆If flow with each other Although directional schemes may be implemented without
and to a central protection and control system. Thus the effect communication, some methods take advantage of communi-
of the fault current quantities used are unaffected by the loads cation infrastructure to improve the schemes. In [33], |∆I1 |
in different sections of the wide area. A determination is then is observed to be highest in the faulted feeder of a micro-
made for the fault region by comparing the magnitudes and grid. With the help of communication between relays, fault
phase difference of ∆If in two adjacent CBs. The protection direction is established when both local and remote relays of
center uses different criteria to determine bus and transformer the protected line determine that the higher |∆I1 | is in the
faults and faults outside or inside an intentional island. forward direction. This particular technique is simple but is
Similarly, an integrated wide area protection scheme was not applicable for medium to high impedance faults (MIFs
proposed in [11]. The distribution network is divided into and HIFs) since fault component currents are significantly
integrated protection units, which communicate directly with impacted under these conditions.
the main protection center. The protection center analyzes the The solution proposed in [34] for detecting faults in micro-
phase difference of ∆I1 between the main and slave feeders grids, monitors ∆I1 at the two terminals of the protected line.
to identify the faulted section. A curve, each with a minimum limit threshold, is developed
An effective use of superimposed quantities to improve based on the variation of ∆I1 along the line, as observed
the sensitivity and dependability of fault detection in such by each smart terminal unit (STU). Using peer-to-peer com-
wide-area schemes is as a starting criterion. This helps pre- munication, a blocking pilot scheme is developed such that
vent unwanted action as a result of non-fault conditions in a fault is determined to be in the forward direction when
the wide network. For example, the scheme in [11] uses at least one of the STUs communicates an action signal to
(∆I2 + ∆I0 )/∆I1 as a starting criterion for the protection the other. Otherwise, a blocking signal is continuously sent.
algorithm. The method fails, however, to deal with single-line-to-ground
faults, considering the significant impact of fault resistance on
C. Directional Protection Schemes the characteristic curve for ∆I1 observed by the STUs.
Another widely used method employed to overcome the In the grid-connected mode of operation, the main grid
challenge of bi-directional fault current flow is directional with its higher short circuit power contributes to the fault
protection. An evaluation of directional protection algorithms current, rendering it much higher than a similar fault in
for ADNs using fault-state and superimposed sequence quan- islanded microgrid mode. This is a major influence on the
tities was carried out in [25]. The results revealed that the protection schemes in the grid-connected mode [35]- [36]. To
superimposed quantity applications provide the most effective deal with the loss of selectivity, particularly in grid-connected
applicable measuring quantities and thus have the best per- mode, a directional scheme was developed in [37]. It compares
formance in detecting fault direction for all types of faults the phase difference between ∆I1 , one period just before
accurately. As in transmission networks [26]–[28], these su- and one period after the fault. The phase difference value is
perimposed sequence quantity applications can use both the significantly influenced by the main grid’s fault current con-
phase difference and magnitude of an evaluated quantity to tribution. This method employs the conventional overcurrent
determine fault incidence and direction. protection and the new directional method described. Hence,
One advantage of directional protection is that it can pro- it has the practical advantage of being applicable to some
vide a non-communication-assisted method to determine the existing technology.
direction of a fault. The solution proposed in [25] only used
the phase difference between ∆I1 and ∆V1 to develop a D. Differential Schemes
criterion for detecting symmetrical fault direction. This simple Differential protection schemes are known to achieve good
application takes advantage of the absence of loads in the fault sensitivity results. Also, as a unit protection scheme, it is
inherently selective in identifying faulted zones. Superimposed due to the fact that no negative sequence ride-through current
quantities are applied in differential protection schemes to pro- is contributed by the main grid or the IIDG. On the other hand,
vide a quantity that limits the challenges faced by traditional in [44], by including the characteristics of the LVRT current
differential schemes. of the IIDG, the ratio of ∆I1 at the grid and DG side is used
To determine fault direction, a differential method is intro- to develop a criterion for fault location. In addition, the phase
duced in [38], using ∆I1 . For a fault within the protected line difference of this ratio is used to set a threshold that ensures
with local terminal M and remote terminal N , the operate the sensitivity of the scheme.
quantity is |∆I1M + I1N | ≥ Iset , where ∆I1M and ∆I1N are A summary of the different superimposed quantities pro-
the superimposed positive sequence currents measured at M posed in the various schemes has been provided in Table I.
and N, and Iset is the pre-determined threshold. The criterion
also includes a restraint coefficient to account for errors that TABLE I
may occur due to external faults. A PPLICATIONS OF S UPERIMPOSED Q UANTITIES : S UMMARY
The solution in [39] uses the phase difference of the ∆I1
Operation Superimposed
flowing into a main substation bus and the remote feeder bus Applications
Proposed
Mode: Quantities
Comm:
Solution Y, N
to develop a criterion for fault location. This improves the GC or B Used
Adaptive [21], [22] B ∆I1 , ∆I2 Y
reliability during external faults since the current phases at the Wide Area [24] B ∆I1 Y
main bus and the remote bus are the same only for the faulted [11] B ∆I0 , ∆I1 , ∆I2 Y
feeder. Similarly, the phase difference is used to distinguish Directional [22] B ∆I1 , ∆I2 Y
[25] B ∆I1 , ∆V1 Y
between bus faults and feeder faults. [33], [34] B ∆I1 Y
[29] B ∆I1 N
Superimposed quantities also improve sensitivity of starting [31] B ∆Z1 , ∆Z2 N
elements in differential schemes. For instance, an impedance [37] GC ∆I1 N
[8] GC ∆I0 , ∆I1 , ∆I2 , ∆V1 Y
differential method proposed in [40], uses the ∆I1 to deter- [43] GC ∆I1 , ∆I2 Y
mine fault incidence before beginning fault detection algo- [44] GC ∆I1 Y
Differential [10], [38], [39] B ∆I1 Y
rithm. Furthermore, the main criterion for fault location uses [40] B ∆I1 , ∆V1 , ∆Z Y
∆V1 and ∆I1 values to calculate the differential impedance GC = Grid-Connected, B = Both; Comm = Communication; Y = Yes, N =
∆Zdif f and restraint impedance ∆Zres of the line. An internal No
fault occurs when |∆Zdif f | ≤ 0.9 |∆Zres |.
A major challenge faced by the differential schemes but IV. N EGATIVE S EQUENCE A DMITTANCE BASED
not considered in the solutions above is data synchronization. S UPERIMPOSED Q UANTITY
Many modern solutions rely on sophisticated data synchro-
nization infrastructure that increases the cost. To deal with A. Methodology
data synchronization issues in communication networks used A superimposed negative sequence admittance method,
for differential protection, the impedance differential solution ∆Y2 , proposed in [45], is considered in this paper to demon-
in [10] uses the superimposed positive sequence current as strate the application of such quantities for microgrid protec-
a fault-instance-based synchronization method. Here, the in- tion. Negative sequence quantities are known to be present
stance of the fault, identified by ∆I1 , provides a time reference in significant quantities in unbalanced faults which form
for the exchanged data. This is a simple alternative to using about 95% of all faults in power systems [46]. This element
sophisticated data synchronization infrastructure. shows high magnitude to fault incidence during forward faults
and provides a good distinction from normal operation. The
E. Applications Based on LVRT Considerations direction of the fault is determined by the phase angle. The
IIDGs are required to have a Low Voltage Ride-Through element is evaluated by the equation
(LVRT) capability and remain connected to the grid when the
voltage sags to a certain level. This is achieved by providing I2f ault − I2pre ∆I2
∆Y2 = = . (6)
a reactive output current to support the voltage during voltage V2f ault − V2pre ∆V2
sag [41]. This has an influence on the quantities used in fault ∆I2
detection. In [8], it was demonstrated that fault characteristics The phase angle, arg{ ∆V 2
}, for forward faults is deter-
of IIDGs with LVRT are not to be based on equivalent constant mined to be (7) and for reverse faults to be (8)
current or power source. Similarly, studies conducted in [32]
0◦ + φ < arg{∆Y2 } < 180◦ + φ (7)
and [42] reveal that the magnitude and angle of the equivalent
positive and negative superimposed impedances of IIDGs are 180◦ + φ < arg{∆Y2 } < 360◦ + φ (8)
affected by the control strategies of the inverter to achieve
LVRT. where φ is the compensation for the angle of the protected
By analyzing LVRT characteristics of IIDGs, a fault detec- line.
tion method was proposed in [8] using the phase difference To discriminate fault conditions from normal imbalance in
between ∆V1 of the bus and ∆I1 of the feeders. Also, the network, a ratio factor α = |I2 /I1 | can be used as a starting
(∆I2 + ∆I0 )/∆I1 is compared to imbalance due to regular criterion for fault detection as employed in other methods,
sequence components, and used as a starting criterion for [31]. α is often set to about 0.1 as minimum threshold for
the fault point detection algorithm. This way, the method is unbalanced shunt fault. Additionally, to improve the sensitivity
independent of fault resistance and applicable to high and low of the element during high impedance faults, another factor,
impedance faults. β = |I0 /I2 | can be implemented in the starting criterion.
Similarly, to eliminate the effect of the LVRT current during Since it is expected that negative sequence and zero sequence
fault location, a negative sequence current direction scheme currents measured at the fault point will be close in value, β
was developed in [43] to detect asymmetrical faults. This is will be close to 1.
 B2 0.2+j0.1 Ω
B1 Load
Utility Grid Δ/Yg Z2 F1
B
0.3+j0.157 Ω
F3 F2 DG
10 kV 10/0.4 kV A
Z1

Fig. 4. Test system

B. Simulation Results
Simulation studies were carried out on the test system in
Fig. 4 to demonstrate the performance of the ∆Y2 method
for detecting usually elusive open-phase faults. The system is
a simple 10/0.4kV microgrid that connects a 13.2 + j3.14Ω Fig. 5. ∆Y2 evaluated by Relay A
load and a 0.01 + j0.01MVA IIDG to the main grid. The DG
response is simulated using PQ control as modeled in [47].
Relays A and B are positioned at Buses B1 and B2. Open-
phase faults are applied at F1 (load side), F2 (Bus B2) and
F3 (Bus B1) at t=0.08s in each case. Considering the normal
operation for this network |∆Y2 | setting can be set to be not
less than 30 for a faulted case. This allows a significant margin
for normal operation. To compensate for the the protected line,
φ is set at 14◦ for relay B and 27.5◦ for relay A.
A commonly used method of identifying an open-phase is
the amount of current imbalance [48]. It is shown in [46], by
analyzing the sequence networks of a one-phase open-fault,
that for a system grounded at both sides of the open-phase,
|I2 | will be about 50% to 100% of |I1 |, depending on the
Fig. 6. ∆Y2 evaluated by Relay B
zero sequence impedance, Z0 . In the case where one side of
the open-phase is ungrounded, |I2 | will be almost equal to |I1 |
since I0 will be zero. Thus, for this demonstration α = |I2 /I1 | not have a reliable constant value as a threshold for protection
is the starting criterion for the open-phase detection and it has elements [49], [50].
a setting of 0.2. The angular characteristics of superimposed fault compo-
Fig.5 and Fig.6 show the operation of the relays A and B nents were also studied in [32]. It was observed that the phase
respectively in response to the series faults. As it can be seen, angles of superimposed impedances vary with fault positions,
both relays correctly detect the direction of faults in all cases. affecting the performance of superimposed component-based
|∆Y2 | is about 50.6 for forward faults. But reverse faults has a directional solutions.
smaller magnitude of about 0.82. The phase angle, arg {∆Y2 }
provides an even clearer distinction between the forward and VI. C ONCLUSION
reverse faults, with about 154◦ in the forward direction and The goals of achieving large scale integration of DGs and
−13◦ in the reverse. The starting criteria, α, operates correctly reliable operation of microgrids require protection systems that
in all cases with a value of between 0.47 and 0.54. can adequately address the new challenges DGs pose to distri-
To observe the influence of the infeed by the DG, the results bution network protection. A review of different proposed pro-
obtained when the DG is disconnected were also considered. tection schemes based on superimposed quantities is carried
Simulation results show values of 50.6 and 153.4◦ for |∆Y2 | out in this paper. It has been established that these quantities
and arg {∆Y2 } respectively. Thus it can be realized from the can be applied to both conventional and novel schemes to
results in the previous paragraph, that in this case, the infeed improve selectivity, sensitivity and security of the protection.
does not affect the performance of the element. These applications have varying limitations in terms of specific
fault types or modes of operation. Ongoing research also
V. D ISADVANTAGES OF S UPERIMPOSED S EQUENCE
shows that the efficacy of these quantities are significantly
Q UANTITIES
affected by the inverter control strategies of IIDGs.
Although different superimposed quantities applications re-
veal great potential for resolving microgrid protection chal- R EFERENCES
lenges, some literature also shows limitations when these [1] O. Dharmapandit, R. Patnaik, and P. Dash, “A fast time-frequency
quantities are applied. These limitations are mainly due to the response based differential spectral energy protection of ac microgrids
including fault location,” Protection and Control of Modern Power
effects of the control strategy of IIDGs. Systems, vol. 2, no. 1, p. 30, 2017.
As described in Section III, IIDGs provide both active and [2] H. J. Laaksonen, “Protection principles for future microgrids,” IEEE
reactive current to support voltage sag during fault. Analysis Transactions on Power Electronics, vol. 25, no. 12, pp. 2910–2918,
2010.
conducted in [42] shows that this requirement leads to non- [3] H. Nikkhajoei and R. H. Lasseter, “Microgrid protection,” in 2007 IEEE
linearity in the output current during faults, and this is influ- Power Engineering Society General Meeting. IEEE, 2007, pp. 1–6.
enced by the level of the voltage sag and the fault resistance. [4] M. R. Miveh, M. Gandomkar, S. Mirsaeidi, and M. R. Gharibdoost, “A
review on protection challenges in microgrids,” in 2012 Proceedings of
Hence, solutions whose criteria depend on magnitude ratio 17th Conference on Electrical Power Distribution. IEEE, 2012, pp.
or phase difference of superimposed voltage and current may 1–5.
 [5] Y. Bansal and R. Sodhi, “Microgrid fault detection methods: Re- [28] A. Sirisha and S. Bhide, “Incremental quantities based relays,” in 2014
views, issues and future trends,” in 2018 IEEE Innovative Smart Grid International Conference on Power, Automation and Communication
Technologies-Asia (ISGT Asia). IEEE, 2018, pp. 401–406. (INPAC). IEEE, 2014, pp. 27–32.
[6] “Ieee standard for interconnection and interoperability of distributed [29] H. Muda and P. Jena, “Sequence currents based adaptive protection
energy resources with associated electric power systems interfaces,” approach for dns with distributed energy resources,” IET Generation,
IEEE Std 1547-2018 (Revision of IEEE Std 1547-2003), pp. 1–138, Transmission Distribution, vol. 11, no. 1, pp. 154–165, 2017.
2018. [30] SEL-221-16 Distance Relay/Fault Locator, Schweitzer Engineering Lab-
[7] M. E. Baran and I. El-Markaby, “Fault analysis on distribution feeders oratories, Pullman WA, USA. 2021. Available at https://selinc.com/.
with distributed generators,” IEEE Transactions on Power Systems, [31] A. Hooshyar and R. Iravani, “A new directional element for microgrid
vol. 20, no. 4, pp. 1757–1764, 2005. protection,” IEEE Transactions on Smart Grid, vol. 9, no. 6, pp. 6862–
[8] F. Zhang and L. Mu, “A fault detection method of microgrids with 6876, 2018.
grid-connected inverter interfaced distributed generators based on the [32] K. Jia, Z. Yang, Y. Fang, T. Bi, and M. Sumner, “Influence of inverter-
pq control strategy,” IEEE Transactions on Smart Grid, vol. 10, no. 5, interfaced renewable energy generators on directional relay and an
pp. 4816–4826, 2019. improved scheme,” IEEE Transactions on Power Electronics, vol. 34,
[9] S. Mirsaeidi, D. M. Said, M. W. Mustafa, and M. H. Habibuddin, “A no. 12, pp. 11 843–11 855, 2019.
protection strategy for micro-grids based on positive-sequence compo- [33] W. Huang, N. Tai, Ke Li, X. Zheng, and Shi Chen, “Protection scheme
nent,” IET Renewable Power Generation, vol. 9, no. 6, pp. 600–609, for active distribution networks using positive-sequence components,” in
2015. 2015 IEEE Power Energy Society General Meeting, 2015, pp. 1–5.
[10] W. Huang, T. Nengling, X. Zheng, C. Fan, X. Yang, and B. J. Kirby, [34] C. Zhou, G. Zou, J. Yang, and X. Lu, “Principle of pilot protection
“An impedance protection scheme for feeders of active distribution based on positive sequence fault component in distribution networks
networks,” IEEE Transactions on Power Delivery, vol. 29, no. 4, pp. with inverter-interfaced distributed generators,” in 2019 IEEE PES GTD
1591–1602, 2014. Grand International Conference and Exposition Asia (GTD Asia), 2019,
[11] F. Zhang, L. Mu, and W. Guo, “An integrated wide-area protection pp. 998–1003.
scheme for active distribution networks based on fault components [35] L. Che, M. E. Khodayar, and M. Shahidehpour, “Adaptive protection
principle,” IEEE Transactions on Smart Grid, vol. 10, no. 1, pp. 392– system for microgrids: Protection practices of a functional microgrid
402, 2019. system.” IEEE Electrification Magazine, vol. 2, no. 1, pp. 66–80, 2014.
[12] X. Li, A. Dyśko, and G. M. Burt, “Traveling wave-based protection [36] P. Mahat, Z. Chen, B. Bak-Jensen, and C. L. Bak, “A simple adaptive
scheme for inverter-dominated microgrid using mathematical morphol- overcurrent protection of distribution systems with distributed genera-
ogy,” IEEE Transactions on Smart Grid, vol. 5, no. 5, pp. 2211–2218, tion,” IEEE Transactions on Smart Grid, vol. 2, no. 3, pp. 428–437,
2014. 2011.
[13] S. M. Brahma and A. A. Girgis, “Development of adaptive protection [37] S. Tian, Y. Liu, and G. Mei, “New theory of directional current
scheme for distribution systems with high penetration of distributed protection for the distribution network containing dg based on fault
generation,” IEEE Transactions on Power Delivery, vol. 19, no. 1, pp. component,” in 2011 International Conference on Advanced Power
56–63, 2004. System Automation and Protection, vol. 1, 2011, pp. 558–563.
[14] S. I. Gkavanoudis, D. Tampakis, K. N. D. Malamaki, G. C. Kryonidis, [38] H. Gao, J. Li, and B. Xu, “Principle and implementation of current
E. O. Kontis, K. O. Oureilidis, J. M. Maza-Ortega, and C. S. Demou- differential protection in distribution networks with high penetration of
lias, “Protection philosophy in low short-circuit capacity distribution dgs,” IEEE Transactions on Power Delivery, vol. 32, no. 1, pp. 565–574,
grids with high penetration of converter-interfaced distributed renewable 2017.
energy sources,” IET Generation, Transmission Distribution, vol. 14, [39] G. Yang, W. Liu, D. Li, B. Liang, and Y. Liu, “Microgrid protection
no. 22, pp. 4978–4988, 2020. based on positive sequence fault component current,” in 2019 4th
[15] K. Liu and Y. Li, “Study on solutions for active distribution grid International Conference on Power and Renewable Energy (ICPRE),
protection,” in Proceedings of the CSEE, vol. 34, no. 16, 2014, pp. 2019, pp. 1–6.
2584–2590. [40] G. Chen, Y. Liu, and Q. Yang, “Impedance differential protection for
[16] A. Ukil, B. Deck, and V. H. Shah, “Current-only directional overcurrent active distribution network,” IEEE Transactions on Power Delivery,
relay,” IEEE sensors journal, vol. 11, no. 6, pp. 1403–1404, 2010. vol. 35, no. 1, pp. 25–36, 2020.
[17] A. Ukil, B. Deck, and V. H. Shah, “Current-only directional overcurrent [41] Z. Dai, C. Li, and X. Chen, “Fault model of iidg considering lvrt and
protection for distribution automation: Challenges and solutions,” IEEE its application in fault analysis of active distribution networks,” in 2016
Transactions on Smart Grid, vol. 3, no. 4, pp. 1687–1694, 2012. IEEE Region 10 Conference (TENCON), 2016, pp. 3831–3834.
[18] N. Rezaei and M.-R. Haghifam, “Protection scheme for a distribution [42] Z. Yang, K. Jia, Z. Li, H. Zhao, Y. Fang, T. Feng, and B. Liu,
system with distributed generation using neural networks,” International “Adaptability analysis of the directional relay for the system with
Journal of Electrical Power & Energy Systems, vol. 30, no. 4, pp. 235– inverter-interfaced renewable energy generators,” in 2019 IEEE 8th
241, 2008. International Conference on Advanced Power System Automation and
[19] M. El Khatib, J. Hernandez-Alvidrez, and A. Ellis, “Fault analysis Protection (APAP), 2019, pp. 1611–1616.
and detection in microgrids with high pv penetration,” Sandia National
Laboratories (SNL-NM), Albuquerque, NM (United States), 2017. [43] Y. Li, G. Zou, J. Yang, and H. Sui, “Faulty section location scheme
[20] G. D. Rockefeller, C. L. Wagner, J. R. Linders, K. L. Hicks, and for distribution grid with inverter interfaced distributed generation,” in
D. T. Rizy, “Adaptive transmission relaying concepts for improved 2017 IEEE PES Innovative Smart Grid Technologies Conference Europe
performance,” IEEE Transactions on Power Delivery, vol. 3, no. 4, pp. (ISGT-Europe), 2017, pp. 1–6.
1446–1458, 1988. [44] J. Yang, C. Zhou, and G. Zou, “A protection scheme based on positive
[21] H. Muda and P. Jena, “Real time simulation of new adaptive overcurrent sequence fault component for active distribution networks,” in 2018 2nd
technique for microgrid protection,” in 2016 National Power Systems IEEE Conference on Energy Internet and Energy System Integration
Conference (NPSC), 2016, pp. 1–6. (EI2), 2018, pp. 1–5.
[22] H. Muda and P. Jena, “Superimposed adaptive sequence current based [45] K. Opoku, S. Pokharel, and A. Dimitrovski, “A negative sequence
microgrid protection: a new technique,” IEEE Transactions on Power element for fault detection in inverter-interfaced microgrid,” Submitted
Delivery, vol. 32, no. 2, pp. 757–767, 2016. for publication, 2021.
[23] D. Sampath Kumar, D. Srinivasan, A. Sharma, and T. Reindl, “Adaptive [46] J. L. Blackburn and T. J. Domin, Protective relaying: principles and
directional overcurrent relaying scheme for meshed distribution net- applications. CRC press, 2006.
works,” IET Generation, Transmission Distribution, vol. 12, no. 13, pp. [47] Z. Shuai, C. Shen, X. Yin, X. Liu, and Z. J. Shen, “Fault analysis of
3212–3220, 2018. inverter-interfaced distributed generators with different control schemes,”
[24] M. Xu, T. Meng, G. Zou, J. Zhang, X. Lin, and J. Yang, “A centralized IEEE Transactions on Power Delivery, vol. 33, no. 3, pp. 1223–1235,
protection and control scheme for microgrid,” in 2015 IEEE PES Asia- 2018.
Pacific Power and Energy Engineering Conference (APPEEC), 2015, [48] T. Smith and B. Graves, “Detection of loss of voltage phase,” in
pp. 1–5. Conference Record of 2013 Annual IEEE Pulp and Paper Industry
[25] D. Yuan, X. Dong, S. Chen, Z. Q. Bo, B. R. J. Caunce, and A. Klimek, Technical Conference (PPIC), 2013, pp. 165–169.
“An new directional comparison scheme for distribution line protection,” [49] Z. Wu, G. Wang, H. Li, G. Pan, and X. Gao, “Analysis on the distribution
in Proceedings of the 41st International Universities Power Engineering network with distributed generators under phase-to-phase short-circuit
Conference, vol. 3, 2006, pp. 881–885. faults,” in Zhongguo Dianji Gongcheng Xuebao(Proceedings of the
[26] H. Gao and P. A. Crossley, “Directional relay for ehv transmission lines Chinese Society of Electrical Engineering), vol. 33, no. 1. Chinese
using positive sequence fault components,” in 2005 IEEE Russia Power Society for Electrical Engineering, 2013, pp. 130–136.
Tech. IEEE, 2005, pp. 1–5. [50] M. Xu, G. Zou, C. Xu, W. Sun, and S. Mu, “Positive sequence
[27] G. Benmouyal and J. Roberts, “Superimposed quantities: Their true differential impedance protection for distribution network with ibdgs,”
nature and application in relays,” in Proceedings of the 26th Annual in 2016 IEEE International Conference on Power System Technology
Western Protective Relay Conference, Spokane, WA, 1999. (POWERCON), 2016, pp. 1–5.