Renewable and Sustainable Energy Reviews 90 (2018) 97–103

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Towards hybrid AC/DC microgrids: Critical analysis and classification of T

protection strategies
⁎
Sohrab Mirsaeidia, Xinzhou Donga, Dalila Mat Saidb,
a
Department of Electrical Engineering, Tsinghua University, Beijing, PR China
b
Centre of Electrical Energy Systems, Faculty of Electrical Engineering, Universiti Teknologi Malaysia, Johor, Malaysia

A R T I C LE I N FO A B S T R A C T

Keywords: Hybrid AC/DC microgrids are arising as an interesting approach as they combine the prominent features of AC
AC microgrids and DC networks and do not require excessive modifications in the distribution network. However, protection of
DC microgrids such microgrids suffers from several challenges including limited fault current contribution of inverter-based DG
Hybrid AC/DC microgrids sources in the islanded mode, inability of single-setting overcurrent relays in protection of dual-mode micro-
Protection challenges
grids, selection of a proper grounding system, and lack of natural zero-crossing current. The aim of this paper is
Protection schemes
to provide a comprehensive review of the available strategies for protection of hybrid AC/DC microgrids. Apart
from describing the most relevant options presented to date and classifying them into specific groups, a critical
analysis is performed, in which the main challenges of each approach are presented. Finally, some conclusions
and recommendations for the protection of future hybrid AC/DC microgrids are pointed out.

1. Introduction
The recent technological developments and the increasing concerns
for global warming have motivated engineers to search for cleaner and
more efficient systems. One of the most efficient ways for the reduction
of impacts of fossil fuels on the environment is to generate energy from
the cleaner energy sources which are located close to the consumers
[1,2]. These sources which are termed as Distributed Generation (DG)
units may be in different forms such as wind turbines, photovoltaic
panels, and fuel cells [3].
A microgrid is a small-scale power grid consisting of DG units,
Energy Storage Systems (ESSs), and electrical/heat loads which oper-
ates as a single-controllable unit at the distribution voltage level.
Microgrids can operate in grid-connected mode or in isolation from the
rest of the network (islanded mode) [4–8]. The operating philosophy is
that under normal condition the microgrid would operate in the grid-
connected mode but in case of any disturbance in the main grid, it
would quickly disconnect from the main grid and continue to operate as
an island [9].
To date, AC system has been the most popular architecture which is
used for the majority of microgrid research projects. Since the design
and modeling of AC systems are much simpler than DC ones, a large
number of microgrids around the world have been developed based on
this technology [10,11]. However, DC-based microgrids are gaining
interest due to many advantages they provide over their counterpart.
These benefits encompass higher reliability, power quality and trans-
mission capacity, non-complex control as well as direct connection to
some DG sources, loads and ESSs [12–14]. Hybrid AC/DC microgrids
are raising as an optimal approach as they combine the main ad-
vantages of AC and DC microgrids [15–17]. The structure of a hybrid
microgrid is depicted in Fig. 1. As can be seen in the figure, the hybrid
microgrid is connected to the main grid via a Static Transfer Switch
(STS). The STS not only can disconnect the microgrid in case of a fault,
but also can be effective in synchronization of both AC and DC subgrids
through current and voltage measurements. It can be also seen from
Fig. 1 that AC and DC subgrids are linked to each other through an
interlinking converter; In addition, a DC/AC converter is used to con-
nect the DC subgrid to the main AC bus [18].
In spite of many benefits provided by microgrids, there are still
some technical challenges which need to be resolved by power system
researchers and engineers. Protection of microgrids is one of them
which requires more attention. During the grid-connected operation
mode, since the main grid provides a large short-circuit current to the
fault point, the protection can be performed by the existing protective
devices in the distribution networks, but in the islanded mode, fault
currents provided by DG units installed in the microgrid are drastically
less than those in the grid-connected mode; hence, the employment of
traditional overcurrent-based protective devices in AC and DC micro-
grids and subgrids is no longer valid and some alternative solutions
should be devised. In a study by Oudalov and Fidigatti [19], an

⁎
Corresponding author.
E-mail address: dalila@fke.utm.my (D.M. Said).

https://doi.org/10.1016/j.rser.2018.03.046
Received 4 February 2017; Received in revised form 7 February 2018; Accepted 17 March 2018
Available online 30 March 2018
1364-0321/ © 2018 Elsevier Ltd. All rights reserved.
S. Mirsaeidi et al.
Fig. 1. Structure of a hybrid microgrid.
adaptive protection strategy was suggested, applying digital relaying
and advanced communication technique. In the presented technique,
the protection settings were updated periodically by means of micro-
grid central controller in accordance with microgrid operating modes.
However, the proposed strategy required updating or upgrading the
protection devices which are presently applied in the distribution net-
works; moreover, fault calculations were relatively sophisticated for a
microgrid functioning in different modes. In [20], a differential-based
protection strategy was introduced which was able to protect radial and
looped microgrids in both modes of operations. However, the proposed
strategy was only effective for the protection of lines and had not the
ability to protect buses connected to DG units or loads. Nikkhajoei and
Lasseter [21] established an alternative protection method based on
symmetrical components. The authors applied zero- and negative- se-
quence currents to protect microgrids against asymmetrical faults.
However, the proposed technique was ineffective in detection of three-
phase faults; besides, the capability of single-phase tripping had not
been considered. Subsequently, in a research by Zamani et al. [22]
another protection strategy was devised using zero- and negative-se-
quence components which could protect microgrids against different
kinds of faults; moreover, the proposed strategy did not require any
communication system. The main problem associated with the pro-
posed method was that it was dependent on the microgrid configura-
tion, because the method had been designed for only radial microgrids
and was not capable of protecting microgrids containing looped fee-
ders; furthermore, since the strategy was based on zero-sequence cur-
rent measurement, it was not effective for protection of microgrids in-
cluding grounded transformers. Dewadasa and his research group
[23,24] proposed another protection approach for inverter-based mi-
crogrids using an admittance relay with inverse time tripping char-
acteristics. Even though the proposed approach could protect micro-
grids in both grid-connected and stand-alone modes, it was unable to
protect microgrids including rotating-based DERs. Another shortcoming
of the strategy was that it was designed for only radial microgrids. In
reference [25], a pattern recognition-based protection strategy was
developed by applying a time-frequency transform which had the
ability to protect radial and looped microgrids against different types of
faults in both grid-connected and islanded modes. However, the main
challenge in pattern recognition-based schemes is that the system has to
be trained. The training is usually achieved by simulations and not real
cases, and hence, it cannot be practically feasible.
Protection of DC microgrids and subgrids is not only affected by the
above-mentioned challenges, but also requires to overcome two addi-
tional issues, i.e. grounding and lack of natural zero-crossing current.
The main purpose of grounding is to detect ground faults. In order to
design a grounding system, two contradictory requirements, i.e. mini-
mization of DC stray current and maximization of personnel safety,
98
Renewable and Sustainable Energy Reviews 90 (2018) 97–103
must be taken into account. Consequently, designing an optimum
grounding system is a tough challenge. To resolve this challenge, re-
ferences [26,27] proposed diode-grounded, and thyristor grounded
systems, respectively. However, the advantage of [27] in comparison
with [26] was that it did not require regular maintenance.
In an AC system, mechanical circuit breakers disconnect circuit
when current crosses zero at every half-period; however, in DC systems,
current has a steady value, and hence, it has to be forced to zero by
additional means; accordingly, authors of [28] proposed a strategy
using a Solid State Circuit Breaker (SSCBs) to dissipate power during
the interruption of fault current. However, the main drawbacks of this
strategy were limited maximum operating voltage and current, and
losses during on-state. Also, Corzine et al. in [29] introduced a Z-source
breaker to provide fault current limitation and interruption through a Z-
source LC circuit, but the resonant circuit of the Z-source breaker is
strongly dependent on the fault characteristics as well as the parameters
of upstream and downstream components. In addition, voltage oscil-
lations resulting from resonant circuit may lead to overvoltage on other
network components.
The aim of this paper is to provide a comprehensive review of the
existing approaches for the protection of microgrids. An attempt has
been made to classify these approaches into specific categories. Finally,
a critical analysis has been conducted, in which the main challenges of
each approach are listed.
The remainder of this paper is as follows: In Section 2, the key issues
and challenges in protection of microgrids are discussed. Section 3
highlights the most recent works performed on the microgrid protec-
tion. In Section 4, a critical analysis on the existing strategies is carried
out. Finally, in Section 5 some conclusions and research directions for
the protection of future hybrid AC/DC microgrids are pointed out.
2. Protection challenges in microgrids
2.1. Protection challenges in AC microgrids and subgrids
Most conventional distribution systems operate radially, where the
power flows unidirectionally from large power plants to the customers.
In such systems, since the magnitude of short circuit currents is pro-
portional to the fault location, the protection is done by overcurrent-
based protective devices [30]. Also, the time-graded coordination be-
tween them enables upstream devices to operate as backup for down-
stream ones [31]. In recent years, the emergence of microgrid has
changed the structure of distribution systems from passive networks
into active ones. This change has disturbed the operation of over-
current-based strategies such that they would no longer have the ability
to protect new structures [32,33].
As mentioned earlier, the fault current contribution of inverter-
based DG sources in a microgrid is limited (only two to three times the
maximum load current) due to the low thermal capability of their
power electronic devices. Therefore, protective devices in a microgrid
containing inverter-based DG sources would operate very slowly or may
not be triggered at all for fault events during islanded mode. In addi-
tion, the considerable difference between the magnitude of short circuit
currents in the grid-connected and islanded modes makes single-setting
traditional overcurrent relays unable to protect dual-mode operating
microgrids [34,35]. Therefore, overcurrent protective devices cannot
be used for protection of AC microgrids and subgrids including inverter-
based DG sources, and some new techniques should be devised.
2.2. Protection challenges in DC microgrids and subgrids
In spite of numerous merits provided by DC microgrids and sub-
grids, protection of such systems suffers from several challenges. Some
of these challenges such as limited fault current contribution of in-
verter-based DG sources in the islanded mode and inability of single-
setting overcurrent relays in protection of dual-mode microgrids are
S. Mirsaeidi et al.
common between AC and DC systems. Nevertheless, protection of DC
ones is influenced by two additional issues, i.e. grounding and lack of
natural zero-crossing current.
2.2.1. Grounding
Basically, there are two types of faults which can occur in DC net-
works, i.e. Line-to-Ground (LG) and Line-to-Line (LL). Even though the
latter typically has lower fault impedance and causes more severe da-
mage to the network, the former is the most frequent fault type which is
remarkably affected by the grounding system [36,37]. In selection of a
proper grounding system, several factors including minimization of
stray current (leakage current from the conductor to the soil), max-
imization of personnel safety (by minimization of the touch voltage),
and fault detection should be taken into account [27].
Corrosion, which is defined as chemical or electrochemical de-
gradation of metal due to the reaction with the environment such as
soil, is the main consequence of stray current [38]. The phenomenon
appears at places where current leaks from the conductor into the soil.
Due to the fact that current changes its transmitting medium from an
electronic environment in the conductor into an ionic one in the soil, an
electron to ion transfer is done [39]. This electron producing or oxi-
dation reaction is referred to as corrosion if it happens over a long
period of time.
The grounding system can also influence the level of touch voltage
(potential difference between energized device and the feet of a person
in contact with the device). If the level of touch voltage exceeds a
certain value (typically 60 V), it can endanger the personnel safety. In
other words, the maximization of personnel safety is achieved by
minimizing the touch voltage [40].
It should be noted that the touch voltage and stray current are in-
versely proportional with each other through the grounding resistance
[41]; for example, the stray current and touch voltage in a solidly
grounded system respectively have their highest and lowest values,
whereas in a system with a large grounding resistance, the value of
stray current is about zero and the touch voltage is in its maximum
value. In fact, simultaneous minimization of stray current and touch
voltage values is impossible. However, their best values can be achieved
by designing an optimized grounding system [42,43].
In addition to the values of stray current and touch voltage, fault
detection is affected by the grounding system. International standard
IEC 60364 has specified three families of grounding systems using two-
letter codes TN, TT and IT. The first letter which denotes the connection
type between the source and ground buses can either be T (direct
connection) or I (no point is connected) [44]. The second letter re-
presents the connection type between the ground or network and the
electrical device being supplied, which is either T (ground connection is
supplied by a local direct connection to ground) or N (ground con-
nection is supplied by the electricity supply network, either as a sepa-
rate protective earth conductor or combined with the neutral con-
ductor).
In a TT system which includes multiple grounding points, fault does
not migrate because of the large impedance of the fault loop. However,
difficulties associated with high-voltage stress and circulating current
paths still exist. In a TN grounding system, both exposed metallic parts
and lines are commonly connected to the ground via associated mid-
points, but in an IT one, the exposed metallic parts have a common
connection to the ground, whereas lines are not earthed [45]. As a
result, fault detection in a TN system is simple due to the low grounding
resistance, but personnel safety cannot be ensured since the touch
voltage may exceed its maximum permissible level. Conversely, IT
system is a suitable choice for maximization of personnel safety due to
the low fault currents, but fault detection in such system is challenging.
However, the second ground fault in an IT system leads to a line to line
fault with a large fault current, jeopardizing the personnel safety [46].
Consequently, designing an optimum grounding system, in which both
personnel safety and fault detection requirements are met
99
Renewable and Sustainable Energy Reviews 90 (2018) 97–103
simultaneously is a tough challenge.
2.2.2. Lack of natural zero-crossing current
Although operation of circuit breaker in both AC and DC systems is
accompanied by an arc phenomenon, the mechanism of an AC circuit
breaker, relying on the natural zero-crossing of the AC current, enables
it to distinguish the arc within the half cycle after tripping. However,
due to the lack of natural zero-crossing in the DC current, the inter-
ruption of current in a DC system is a major problem which not only
causes a serious hazard for the personnel safety, but also results in the
contact erosion of circuit breakers, thereby decreasing their lifetime
[47–50].
Currently, fuse and Circuit Breaker (CB) are commercially available
protective devices for DC systems [51,52]. Fuse, which is frequently
used in low impedance systems, operates on the principle of melting
down a metal wire when too much current flows through it. It must be
selected based on the time-current and voltage ratings of system in
which it operates. Fuse can be used in both AC and DC systems.
However, deployment of fuse in DC systems requires accurate calcula-
tion of the network time constant due to its direct influence on the fuse
operation [51]. More precisely, if the network time constant is less than
2.5 ms, the fuse metal wire is quickly melted and the current is inter-
rupted; In contrast, a large network time constant (more than 6 ms)
increases the melting time, and hence, the arc cannot be extinguished
rapidly [53]. Additionally, transient overcurrents in a DC network may
cause fuse malfunction. Consequently, fuse is not a suitable option for
protection of DC microgrids, but it still can be used as a backup pro-
tective device.
Molded-Case Circuit Breaker (MCCB), consisting of a quenching
chamber, contacts, and a tripping device (thermal-magnetic or elec-
tronic), is another choice for interruption of the fault current [54]. It is
designed to operate once its instantaneous trip is exceeded. However,
the final opening of the contacts does not happen unless the peak
current lasts for a specific period of time [55]. In the microgrid, some
loads and sources are interfaced with microgrid through power elec-
tronic devices. These power electronic devices often require line-to-line
or line-to-ground filter capacitors. In case of a DC fault incident, the
capacitors swiftly discharge into the fault point and leads to large peak
currents for a short period of time, and hence the adequate force for
completely opening of contact may not be generated [56]; in particular,
contacts in a highly inductive system may weld closed during the fault
[57]. For this reason, circuit breaker cannot be an ideal solution for
interruption of the fault current as well.
3. Solutions for microgrid protection challenges
3.1. Solutions for protection challenges in AC microgrids and subgrids
As discussed earlier, the traditional overcurrent-based strategies
have not the ability to protect AC microgrids and subgrids due to the
drastic difference between the magnitude of fault currents in the grid-
connected and islanded modes. In order to overcome this challenge, a
number of strategies have recently been proposed in the scientific lit-
erature. In the following subsections, apart from introducing and ca-
tegorizing the most relevant approaches proposed to date, the merits
and demerits of each category are discussed.
3.1.1. Adaptive protection
Adaptive protection is defined as an online system which modifies
the preferred protective response to a change in system conditions
through an externally generated signal [58,59]. Adaptive protection
schemes can be classified into three main categories including over-
current, differential and symmetrical components.
3.1.1.1. Adaptive overcurrent schemes. In adaptive overcurrent schemes
[19,60–62], a central protection unit is used to periodically store and
S. Mirsaeidi et al.
update three distinct tables, i.e. event, fault current and action tables.
Event table lists all possible configurations of the microgrid along with
the respective status of DG sources. Subsequently, in accordance with
each configuration, fault currents measured by the relays for all
possible fault locations are stored in the fault current table. Also, for
each configuration, action table lists the relay settings for each fault
type along with its time delays. Finally, the central protection unit is
able to issue the proper tripping signals to the respective relays based
on the status of these three tables in each period. Moreover, in case a
relay fails to trip, its upstream or downstream relay (based on action
table) operates after a predetermined period of time and provides the
secondary protection. Likewise, if a fault takes place in the main grid,
the closest microgrid relay to the main grid interrupts the fault current
provided by the microgrid DG sources, and then the microgrid is
transferred to the islanded mode [63]. However, adaptive overcurrent
protection strategies suffer from some challenges including: (a)
necessity to consider all possible configurations of a microgrid with
regard to different locations and types of faults. (b) complicated
analysis of short circuit currents in a large microgrid with many
radial and looped feeders. (c) costs associated with installation of a
communication infrastructure.
3.1.1.2. Adaptive differential schemes. Differential protection schemes
operate based on comparison between the measured currents by relays
installed at both ends of a protected element (such as busbar, line and
transformer). In case a fault occurs in the protected element, the
difference between these measured currents exceeds a threshold value
and the relays trip to isolate the faulted element from the rest of
network. In addition, backup protection can be provided by setting the
adjacent upstream and downstream relays of the protected element
[61,64].
In [65], a differential-based protection strategy using traditional
overcurrent relays as well as communication links is proposed which is
able to protect medium voltage microgrids including both inverter- and
synchronous- based DG sources. Even though the economic issues have
been considered in the scheme, it is unable to protect microgrids in-
cluding unbalanced loads.
Sortomme et al. designed another differential-based protection
scheme applying digital relays and Phasor Measurement Units (PMUs)
along with communication channels [20]. The scheme provides three
levels of protection including instantaneous and comparative voltage
relays. Additionally, the protection against High- Impedance Faults
(HIFs) is presented in the scheme. Nevertheless, the suggested method
is not economical, since PMU is an expensive device.
In [66], another protection scheme is introduced for microgrids
including both radial and looped feeders. In the scheme, lines and
busbars are protected by only current differential relays, whereas the
protection of DG sources is provided by over- and under-voltage, re-
verse power flow, and synchronism check relays. Although the devel-
oped methodology can provide a robust protection for both grid-con-
nected and islanded modes, it still suffers from problems related to the
unbalanced loads and switching transients.
Generally, the main drawbacks of differential protection schemes
are: (a) need for communication system as a key element, since the
failure of such schemes endangers protection of microgrid (b) inability
in protection of buses (c) deployment of costly synchronized measure-
ment devices (d) difficulties associated with unbalanced loads and
transients during connection or disconnection of DG sources.
3.1.1.3. Adaptive protection schemes based on symmetrical
components. The proposed protection schemes in this category
substantially apply principles of symmetrical components and enable
overcurrent-based strategies to protect microgrids in both grid-
connected and islanded modes. The main proposal in the area is put
forward by Nikkhajoei and Lasseter in 2006 [21]. In their proposal,
they make use of zero- and negative- sequence currents to detect and
100
Renewable and Sustainable Energy Reviews 90 (2018) 97–103
isolate, respectively, single-line-to-ground and line-to-line faults in the
islanded mode of operation. However, their devised solution has not the
ability to protect the microgrid during a high impedance fault.
Furthermore, the operation of such protection scheme requires
communication links.
In [22], a microprocessor-based relay along with a protection
strategy is designed. The strategy which is able to protect low voltage
microgrids against both solid and high-impedance faults, operates by
applying zero- and negative- sequence components. The main feature of
the strategy is that it does not require communication links. However,
the proposed method is not capable of protecting microgrids including
mesh feeders.
The authors of [67] developed another protection scheme based on
only positive-sequence components. In their proposed scheme, they use
a designed Microprocessor-Based Relay (MBR) along with PMUs and a
digital communication system to protect microgrids including both
radial and looped feeders against different types of faults. The designed
MBRs have the ability to update their pickup values after any change in
the structure of microgrid, thereby protecting microgrids against sub-
sequent faults. Even though the proposed protection scheme remedies
the drawbacks of the previous works, it is not economical due to the
high price of PMU.
The main issues related with the implementation of the above-
mentioned schemes are: (a) necessity to extensive communication in-
frastructure in some proposals that may fail at some point, jeopardizing
the whole microgrid protection. (b) inability to provide protection for
looped microgrids (c) high costs associated with deployment of PMUs.
3.1.2. Distance protection
Distance protection scheme which offers a high selectivity is an-
other way to protect AC microgrids and subgrids. The installed distance
relays in the scheme are responsible for calculation of impedance using
the measured voltage and current at their location to detect fault oc-
currences. Prior to fault occurrence, the measured impedance value is
high because it includes the load impedance, while in case of a fault
event on a network line, the value becomes equivalent to only the line
impedance and decreases. As a result, the fault in each zone can be
detected and located by comparison between the measured impedance
values before and after the fault [68,69].
The typical time settings for a three-zone distance protection
scheme are depicted in Fig. 2. According to the figure, Zone 1 protects
80% of the line length of AB without any tripping time delay. Zone 2 is
set to not only protect whole Line AB, but also provide protection for
20% of its adjacent line (Line BC) with tripping time delay t1. Also,
100% of both Lines AB and BC plus 25% of Line CD are protected with
tripping time t2 through Zone 3.
The main study in this category is accomplished by Dewadasa and
his research group in references [23,24]. In their proposed protection
scheme, a new admittance relay is developed based on characteristics of
inverse time tripping. The developed relay has the ability to provide
protection in its forward and inverse directions against different kinds
of faults. However, some shortcomings of this methodology include: (a)
dependency on the microgrid configuration and type of installed DG
units (b) errors resulting from fault resistance in measured impedances
by relays (c) complications associated with impedance measurements in
short lines.
3.1.3. Pattern recognition schemes
In reference [25], a new microgrid protection scheme is developed
by applying a time-frequency transform which has the ability to protect
radial and looped microgrids against different types of faults in both
grid-connected and islanded mode. In the developed scheme, first, S-
transform is used to extract the spectral energy contents of the fault
current signals, measured at both ends of each line. Subsequently, fault
patterns are registered by differential energy computations. Based on
predetermined threshold values (in accordance with each type of fault)
S. Mirsaeidi et al.
Fig. 2. Time settings for a three-zone distance protection scheme.
on differential energy, the protection scheme is able to detect and
isolate the faulted line. With regard to the indicated simulation results,
the differential energy can be a suitable criterion, since it remarkably
varies for a faulty phase in comparison with healthy ones. Moreover,
the developed strategy is immune to the noise and less sensitive to
synchronization errors. However, the main challenge in the pattern
recognition schemes is that the system has to be trained. The training is
usually achieved by simulations and not real cases, and hence, it cannot
be practically feasible.
3.2. Solutions for protection challenges in DC microgrids and subgrids
Although the majority of the proposed protection schemes for AC
microgrids and subgrids can be designed compatible with DC ones to
overcome the common challenges, provision of a robust scheme for DC
ones also requires addressing the challenges associated with grounding
and lack of natural zero-crossing current. The following subsections
review the main proposed approaches, attempting to resolve these
challenges.
3.2.1. Reconfigurable grounding systems
As mentioned in Subsection 2.2.1, personnel safety and fault de-
tection are two contradictory requirements which are affected by the
type of grounding system. In fact, best personnel safety and fault de-
tection cannot be achieved in a certain grounding system. Hence, some
alternative solutions have recently appeared in the technical literature
which try to ponder both of these requirements by applying re-
configurable grounding systems. More precisely, in such grounding
systems, the network normally operates in ungrounded mode to mini-
mize corrosion phenomenon resulting from high stray currents, but in
case of sensing an unacceptable level of touch voltage, it automatically
transfers to the grounded mode. However, it switches back to the un-
grounded mode after clearance of abnormal operating condition.
The most basic structure of a reconfigurable system, referred to as
diode grounded system, contains a direct metallic connection of the
negative bus to the earth by means of a diode circuit [26]. In case a
certain threshold voltage is reached, the current is allowed to flow
through diode circuit to get dissipated in order that personnel safety is
ensured. However, due to the fact that corrosion cannot be entirely
obviated in a diode grounded system, it requires regular maintenance.
In order to possess an active control over the grounding instances,
thyristor grounded system was developed in [27]. In the proposed
system, an overvoltage relay (R1) continuously monitors the difference
between negative bus and ground voltage magnitudes and triggers the
thyristor gate once it exceeded a predetermined value. Furthermore, the
system is equipped with a current sensor (R2) in order to check the
status of the flowing current. If the level of sensed current was lowered,
the system can be switched back to the ungrounded mode. Otherwise, a
positive to ground fault event is the most probable reason that DC
breakers must be swiftly opened. Moreover, the status of system can be
101
Renewable and Sustainable Energy Reviews 90 (2018) 97–103
reported to the environment via an alarm signal.
The salient feature of the thyristor grounded system in comparison
with diode one is that it sustains the system ungrounded, unless a
dangerous voltage is sensed; therefore, the thyrsitor grounded system
considerably minimizes the stray current and its negative consequences.
3.2.2. DC current interruption approaches
As discussed before, protection of DC microgrid using fuses or cir-
cuit breakers has some performance restrictions due to their inherent
large time constants and time delays, respectively. In order to overcome
the limitations, Tang and his colleague presented a new current inter-
ruption approach for Multi-Terminal DC (MTDC) grids and navy ship-
board DC Zonal Electric Distribution (DCZED) systems by means of
electro-mechanical switches. In their proposed approach, they split the
network into several zones and make use of no-load switches to cease
the fault currents [70,71]. More precisely, once a fault was recognized
in a zone, converters supplying the network de-energize the bus(s), and
subsequently the faulted zone is isolated by no-load switches. Finally,
the rest of network is re-energized to continue its operation. The main
problem with the proposed approach is that it entirely shuts down the
network after the fault detection which may not be necessary.
An alternative approach was proposed using SSCBs at DC terminals
of Voltage Source Converters (VSCs) or on the downstream side of DC/
DC converters [28]. The approach can be implemented by different
solid state switches such as Gate Turn-Off (GTO) thyristors, Insulated-
Gate Bipolar Transistors (IGBTs), and Insulated-Gate Commutated
Thyristors (IGCTs). However, employment of each of the switch
topologies has its own merits and demerits [28]. SSCBs are also
equipped with a parallel combination of a snubber circuit and Me-
tal–Oxide Varistors (MOVs) to dissipate power during the interruption
of fault currents. Notwithstanding advantages of the SSCB, some of its
demerits has made it disputable. Contrary to mechanical contacts, the
maximum operating voltage and current of the SSCB is limited in order
to protect its switching devices. Overrating of SSCB also leads to ex-
ponential increase of costs. Furthermore, the resistance of SSCB is much
larger than that of the mechanical CB, leading to more losses during the
on-state and thus, the reduction of the overall system efficiency.
In [29], a new type of solid state breaker was introduced, termed as
Z-source breaker. The proposed breaker is able to automatically com-
mutate a main-path Silicon-Controlled Rectifier (SCR) during a fault
through a Z-source LC circuit. In spite of swift operation of the Z-source
circuit breaker, its resonant circuit is strongly dependent on the fault
characteristics as well as parameters of upstream and downstream
components. In addition, voltage oscillations resulting from resonant
circuit may lead to overvoltage on other network components.
4. Critical analysis
After a critical analysis on the available protection strategies for
protection of microgrids, it seems that an adaptive protection is
S. Mirsaeidi et al.
Table 1
Examples and disadvantages of the available protection strategies for AC and DC microgrids and subgrids.
Microgrid/subgrid type Protection strategy Examples Disadvantages
AC Adaptive overcurrent [19,60–62]
Adaptive differential [20,65,66]
Adaptive symmetrical [21,22,67]
components
Distance [23,24]
Pattern recognition [25]
DC Reconfigurable grounding [26,27]
DC current interruption [28,29,70]
necessarily required. Most of the proposed adaptive protection methods
are completely dependent on exchange or transfer of data/information
in the shape of measured system parameters (voltage, current, phase
angle, etc.) and direction and interlocking signals between different
protection devices via some kind of communication link. Therefore, to
have an effective adaptive protection scheme, the reliability of com-
munication link is critical. The factors like high risks of communication
link failures and cyber security threats as well as the high costs involved
to avoid them are the major challenges for implementation of eco-
nomical adaptive protection schemes. These challenges should be ad-
dressed properly in order to fully exploit the benefits of adaptive pro-
tection schemes. Considering the above factors, the future protection
schemes for AC, DC, and hybrid AC/DC microgrids, in general, are most
likely to be hybrid in nature comprised of optimum combination of
traditional effective protection schemes. The choice will depend on the
type of microgrid structure, reliability needs of customers and the type
and control of DG units. Additionally, redundant protection and com-
munication systems will be required for microgrid, providing higher
level of reliability. Table 1 lists examples and disadvantages of the
available protection strategies for AC and DC microgrids and subgrids.
5. Conclusion
Based on the analysis of the wide range of technical publications
presented in the previous sections, the main conclusions and re-
commendations for the protection of future hybrid AC/DC microgrids
can be summarized as follows: (a) There are still relatively small
number of references that properly address the protection challenges in
AC and DC microgrids and subgrids. In many cases, the information
given in these references is short or incomplete, barely including ver-
ification simulations or tests, most of them to be more of an idea than a
thoroughly studied solution. (b) It seems to be clear that there is an
increasing need or even necessity to upgrade many protection devices
currently used in power systems, mainly in Low-Voltage (LV) systems,
in order to provide new capabilities for the application of new un-
conventional protection systems in microgrids. (c) Regardless of the
Renewable and Sustainable Energy Reviews 90 (2018) 97–103
– Necessity to upgrade available overcurrent relays
– Necessity to know all possible configurations
– Complicated calculations for large microgrids
– Communication system is required
– Need for communication system as a key element
– Inability in protection of buses [20]
– Deployment of costly synchronized measurement devices [20]
– Difficulties resulting from unbalanced loads and transients
– Need for communication system as a key element [21,67]
– Inability in detection of three-phase faults [21]
– Lack of single-phase tripping capability [21]
– Dependency on the microgrid configuration [21,22]
– Need for application of grounded transformers [21,22]
– High costs associated with deployment of PMUs [67]
– Dependency on the type of installed DG units
– Dependency on the microgrid configuration
– Measurement errors resulting from fault resistance
– Complications associated with impedance measurements in short lines.
– The training is achieved by simulations and not real cases, and hence, it cannot be practically
feasible.
– In diode grounded systems corrosion cannot be entirely obviated, and hence they require regular
maintenance [26]
– Network is entirely shut down after the fault detection [70]
– Limited maximum operating voltage and current in SSCBs [28]
– Losses during the on-state in SSCBs [28]
– Resonant circuit Z-source breaker is strongly dependent on the fault characteristics and parameters
of upstream and downstream components [29]
– Overvoltage on other network components due to voltage oscillations resulting from resonant
circuit in Z-source breakers [29]
protection methodology, it seems likely that some kind of commu-
nication is going to be necessary, either centrally operated or dis-
tributed. Also, economical analysis of high-data-rate and high-coverage
communication technologies is of great importance. (d) In order to deal
satisfactorily with the protection problems associated with a bidirec-
tional power flow, the need for a directional feature is clear. (e) There
are some specific problems that are hardly analyzed in cited references,
such as HIFs and protection of meshed networks. (f) Similarly to other
research areas in power system protection, it is likely that, in order to
get an optimal protection system for AC and DC microgrids and sub-
grids, combination of different protection techniques will be necessary.
Acknowledgements
This work was supported in part by the National Natural Science
Foundation of China (Grant No. 51120175001), and in part by the
National Key Research and Development Plan of China (Grant No.
2016YFB0900600).
References
[1] Das K, Nitsas A, Altin M, Hansen AD, Sørensen PE. Improved load-shedding scheme
considering distributed generation. IEEE Trans Power Del 2017;32(1):515–24.
[2] Liao H, Milanović JV. Methodology for the analysis of voltage unbalance in net-
works with single-phase distributed generation. IET Gener Transm Distrib
2017;11(2):550–9.
[3] Pereira BR, da Costa GRM, Contreras J, Mantovani JRS. Optimal distributed gen-
eration and reactive power allocation in electrical distribution systems. IEEE Trans
Sustain Energy 2016;7(3):975–84.
[4] Lotfi H, Khodaei A. Hybrid AC / DC microgrid planning. IEEE Trans Power Syst
2016;8(1):296–304.
[5] Gururani A, Mohanty SR, Mohanta JC. Microgrid protection using Hilbert–Huang
transform based-differential scheme. IET Gener Transm Distrib
2016;10(15):3707–16.
[6] Mishra DP, Samantaray SR, Joos G. A combined wavelet and data-mining based
intelligent protection scheme for microgrid. IEEE Trans Smart Grid
2016;7(5):2295–304.
[7] Cintuglu MH, Ma T, Mohammed OA. , “Protection of autonomous microgrids using
agent-based distributed communication,”. IEEE Trans Power Deliv
2016;32(1):351–60.
102
S. Mirsaeidi et al.
[8] Lai K, Illindala MS, Haj-Ahmed MA. Comprehensive protection strategy for an is-
landed microgrid using intelligent relays. IEEE Trans Ind Appl 2016;8(99):47–55.
[9] Ragaini E, Tironi E, Grillo S, Piegari L, Carminati M. Ground fault analysis of low
voltage DC micro-grids with active front-end converter, In: Proceedings of the 3rd
Renew. Power Gener. Conference (RPG 2014), Naples, Italy, p. 1–6, 2014.
[10] Justo JJ, Mwasilu F, Lee J, Jung JW. AC-microgrids versus DC-microgrids with
distributed energy resources: a review. Renew Sustain Energy Rev
2013;24:387–405.
[11] Jiayi H, Chuanwen J, Rong X. A review on distributed energy resources and mi-
crogrid. Renew Sustain Energy Rev 2008;12(9):2465–76.
[12] Planas J Andreu, Gárate JI, Martínez De Alegría I, Ibarra E. AC and DC technology
in microgrids: a review. Renew Sustain Energy Rev 2015;43:726–49.
[13] Papadimitriou CN, Zountouridou EI, Hatziargyriou ND. Review of hierarchical
control in DC microgrids. Electr Power Syst Res 2015;122:159–67.
[14] Fei Gao SB, Alessandro C, Chintan P, Pat W, Christopher IH, Greg A. Comparative
stability analysis of droop control approaches in voltage source converters-based DC
microgrids. IEEE Trans Power Electron 2016;32(3):2395–415.
[15] Tah A, Das D. An enhanced droop control method for accurate load sharing and
voltage improvement of isolated and interconnected DC microgrids. IEEE Trans
Sustain Energy 2016;7(3):1194–204.
[16] Meng L, Dragicevic T, Roldan-Perez J, Vasquez JC, Guerrero JM. Modeling and
sensitivity study of consensus algorithm-based distributed hierarchical control for
DC microgrids. IEEE Trans Smart Grid 2016;7(3):1504–15.
[17] Feng W, Jet TK, Kaiyuan L, Goh YH, Si B, Chew H. Harmonic mitigation of hybrid
AC/ DC micro-grids with PMSG, Renewable Power Generation (RPG) 2016, In:
Proceedings of the 5th IET International Conference on, London, U.K., p. 1–5, 2016.
[18] Unamuno E, Barrena JA. Hybrid ac/dc microgrids - Part I: review and classification
of topologies. Renew Sustain Energy Rev 2015;52:1251–9.
[19] Oudalov A, Fidigatti A. Adaptive network protection in micro-grids. Int J Distrib
Energy Resour 2009;4(3):201–5.
[20] Sortomme E, Venkata SS, Mitra J. Microgrid protection using communication-as-
sisted digital relays. IEEE Trans Power Deliv 2010;25(4):2789–96.
[21] Nikkhajoei H, Lasseter RH. Microgrid fault protection based on symmetrical and
differential current components. Consort Electr Reliab Technol Solut 2006.
[Contract No. 500-03-024].
[22] Zamani MA, Sidhu TS, Yazdani A. A protection strategy and microprocessor-based
relay for low-voltage microgrids. IEEE Trans Power Deliv 2011;26(3):1873–83.
[23] Dewadasa M. Protection for distributed generation interfaced networks [[Ph.D.
Thesis]. Faculty of Built Environment and Engineering, Queensland University of
Technology; 2010.
[24] Dewadasa M, Majumder R, Ghosh A, Ledwich G. Control and protection of a mi-
crogrid with converter interfaced micro sources, In: Proceedings of the 2009
International Conference on Power Systems, Kharagpur, India, p. 1–6, 2009.
[25] Kar S, Samantaray SR. Time-frequency transform-based differential scheme for
microgrid protection. IET Gener Transm Distrib 2014;8(2):310–20.
[26] Lee CH, Lu CJ. Assessment of grounding schemes on rail potential and stray currents
in a DC transit system. IEEE Trans Power Deliv 2006;21(4):1941–7.
[27] Paul D. DC traction power system grounding, 2001 IEEE Industry Applications
Conference, In: Proceedings of the 36th IAS Annual Meeting, Chicago, USA, p.
2133–39, 2001.
[28] Meyer C, Schroder S, De Doncker RW. Solid-state circuit breakers and current
limiters for medium-voltage systems having distributed power systems. IEEE Trans
Power Electron 2004;19(5):1333–40.
[29] Corzine KA, Ashton RW. A new Z-source DC circuit breaker. IEEE Trans Power
Electron 2012;27(6):2796–804.
[30] Piesciorovsky EC, Schulz NN. Fuse relay adaptive overcurrent protection scheme for
microgrid with distributed generators. IET Gener Transm Distrib 2017;11(2):540–9.
[31] Liu Z, Hoidalen HK, Saha MM. An intelligent coordinated protection and control
strategy for distribution network with wind generation integration. CSEE J Power
Energy Syst 2016;2(4):23–30.
[32] Sahoo AK. Protection of microgrid through coordinated directional over-current
relays, In: Proceedings of the 2014 IEEE Global Humanitarian Technology
Conference - South Asia Satellite (GHTC-SAS), Trivandrum, India, pp. 129–34,
2014.
[33] Choudhary NK, Mohanty SR, Singh RK. A review on microgrid protection, In:
Proceedings of the 2014 International Electrical Engineering Congress (iEECON),
Chonburi, Thailand, p. 1–4, 2014.
[34] Bui DM, Chen SL, Lien KY, Jiang JL. Fault protection solutions appropriately used
for ungrounded low-voltage AC microgrids, In: Proceedings 2015 IEEE Innov. Smart
Grid Technol. - Asia, ISGT ASIA 2015, Bangkok, Thailand, p. 4066–71, 2016.
[35] Hussain A, Aslam M, Arif SM. N-version programming-based protection scheme for
microgrids: a multi-agent system based approach. Sustain Energy Grids Netw
2016;6:35–45.
[36] Meghwani A, Srivastava S, Chakrabarti S. A non-unit protection scheme for DC
microgrid based on local measurements. IEEE Trans Power Deliv
2016;32(1):172–81.
[37] Monadi M, Gavriluta C, Luna A, Candela I, Rodriguez P. Centralized protection
strategy for medium voltage DC microgrids. IEEE Trans Power Deliv
2016;32(1):430–40.
[38] Memon SA, Fromme P. Stray current corrosion and mitigation: a synopsis of the
technical methods used in dc transit systems. IEEE Electrification Mag
2014;2(3):22–31.
[39] Cotton I, Charalambous C, Aylott P, Ernst P. Stray current control in DC mass transit
systems. IEEE Trans Veh Technol 2005;54(2):722–30.
[40] Ibrahem A, Elrayyah A, Sozer Y, De Abreu A. DC railway system emulator for stray
current and touch voltage prediction. IEEE Trans Ind Appl 2016;53(1):439–46.
103
Renewable and Sustainable Energy Reviews 90 (2018) 97–103
[41] Niasati M, Gholami A. Overview of stray current control in DC railway systems, In:
Proceedings of the 2008 International Conference on Railway Engineering -
Challenges for Railway Transportation in Information Age, Hong Kong, China, p.
1–6, 2008.
[42] Mitolo M, Liu H. Touch voltage analysis in low-voltage power systems studies. IEEE
Trans Ind Appl 2016;52(1):556–9.
[43] Augusto F, León H, Manuel J, Quevedo P. Design and construction of a dynamic
system for step and touch voltage measurements for grounding systems, In:
Proceedings of the 2011 International Symposium on Lightning Protection,
Fortaleza, Brazil, p. 278–283, 2011.
[44] IEC 60364-1 Low-voltage electrical installations—Part 1: Fundamental principles,
assessment of general characteristics, Definitions, IEC 60364-1, 2005.
[45] Dragičević T, Lu X, Vasquez JC, Guerrero JM. DC microgrids — Part II: a review of
power architectures, applications, and standardization issues. IEEE Trans Power
Electron 2016;31(5):3528–49.
[46] Madingou G, Zarghami M, Vaziri M. Fault detection and isolation in a DC microgrid
using a central processing unit, In: Proceedings of the 2015 IEEE Power Energy Soc.
Innov. Smart Grid Technol. Conference, Washington, DC, U.S., p. 1–5, 2015.
[47] Becker DJ, Sonnenberg BJ. DC microgrids in buildings and data centers, In:
Proceedings of the 2011 IEEE 33rd International Telecommunications Energy
Conference (INTELEC), Amsterdam, Netherlands, p. 1–7, 2011.
[48] Do Park J, Candelaria J, Ma L, Dunn K. DC ring-bus microgrid fault protection and
identification of fault location. IEEE Trans Power Deliv 2013;28(2):779–87.
[49] Yuan C, Haj-ahmed M a, Illindala M. Protection strategies for medium voltage direct
current microgrid at a remote area mine site. IEEE Trans Ind Appl
2015;51(4):2846–53.
[50] Sakagami T. Simulation to optimize a DC microgrid in Okinawa, In: Proceedings of
the 2016 IEEE International Conference on Sustainable Energy Technologies
(ICSET), Hanoi, Vietnam, p. 214–219, 2016.
[51] Brozek JP. DC overcurrent protection - Where we stand, In: Proceedings of the
Conference Record of the 1992 IEEE Industry Applications Society Annual Meeting,
Houston, TX, USA, p. 1306–10, 1992.
[52] Gregory GD. Applying low voltage circuit breakers in direct current systems, In:
Proceedings of the 1994 IEEE Industry Applications Society Annual Meeting,
Denver, CO, U.S., p. 2293–2302, 1994.
[53] IEEE guide for the protection of stationary battery systems. IEEE Standard p.
1375–1998, 1998.
[54] Liu Y, Chen D, Yuan H, Ji L, Ma Z. Research of dynamic optimization for the cam
design structure of MCCB. IEEE Trans Compon, Packag Manuf Technol
2016;6(3):390–9.
[55] Ji L, Chen D, Liu Y, Li X. Analysis and improvement of linkage transfer position for
the operating mechanism of MCCB. IEEE Trans Power Deliv 2011;26(1):222–7.
[56] Cuzner RM, Venkataramanan G. The status of DC micro-grid protection, In:
Proceedings of the 2008 IEEE Industry Applications Society Annual Meeting,
Edmonton, Alta., Canada, p. 1–8, 2008.
[57] Sun H, Rong M, Chen Z, Hou C, Sun Y. Investigation on the arc phenomenon of air
DC circuit breaker. IEEE Trans Plasma Sci 2014;42(10):2706–7.
[58] Zheng KH, Xia MC. Impacts of microgrid on protection of distribution networks and
protection strategy of microgrid, In: Proceedings of the 2011 International
Conference on Advanced Power System Automation and Protection, Beijing, China,
p. 356–59, 2011.
[59] Teimourzadeh S, Aminifar F, Davarpanah M, Shahidehpour M. Adaptive protection
for preserving microgrid security. IEEE Trans Smart Grid 2017;1(12):132–40.
[60] Ustun ST. Design and development of a communication assisted microgrid protec-
tion system [[Ph.D. Thesis]. School of Engineering and Science, Faculty of Health,
Engineering and Science, Victoria University; 2013.
[61] Ustun TS, Ozansoy C, Zayegh A. Fault current coefficient and time delay assignment
for microgrid protection system with central protection unit. IEEE Trans Power Syst
2013;28(2):598–606.
[62] Etemadi AH, Iravani R. Overcurrent and overload protection of directly voltage-
controlled distributed resources in a microgrid. IEEE Trans Ind Electron
2013;60(12):5629–38.
[63] Laaksonen H, Ishchenko D, Oudalov A. Adaptive protection and microgrid control
design for Hailuoto island. IEEE Trans Smart Grid 2014;5(3):1486–93.
[64] Ustun TS, Ozansoy C, Zayegh A. Modeling of a centralized microgrid protection
system and distributed energy resources according to IEC 61850-7-420. IEEE Trans
Power Syst 2012;27(3):1560–7.
[65] Conti S, Raffa L, Vagliasindi U. Innovative solutions for protection schemes in au-
tonomous MV micro-grids, 2009 International Conference on Clean Electrical
Power, Capri, Italy, p. 647–54, 2009.
[66] Dewadasa M. Protection of microgrids using differential relays, In: Proceedings of
the 21th Aus. Univ. Power Eng. Conference (AUPEC), Brisbane, Australia, p. 1–6,
2011.
[67] Mirsaeidi S, Mat Said D, Mustafa MW, Hafiz Habibuddin M. A protection strategy
for micro-grids based on positive-sequence component. IET Renew Power Gener
2015;9(6):600–9.
[68] Singh AR, Dambhare SS. Adaptive distance protection of transmission line in pre-
sence of SVC. Int J Electr Power Energy Syst 2013;53(1):78–84.
[69] Brearley BJ, Prabu RR. A review on issues and approaches for microgrid protection.
Renew Sustain Energy Rev 2017;67:988–97.
[70] Tang L, Ooi BT. Locating and isolating DC faults in multi-terminal DC systems. IEEE
Trans Power Deliv 2007;22(3):1877–84.
[71] Ciezki JG, Ashton RW. Selection and stability issues associated with a navy ship-
board DC zonal electric distribution system. IEEE Trans Power Deliv
2000;15(2):665–9.