0% found this document useful (0 votes)
37 views22 pages

2 DC Microgrids-Part II

This document reviews power architectures, applications, and standardization issues related to DC microgrids (MGs). It first discusses typical power hardware topologies used in DC MGs and their suitability for different applications. It then provides an overview of state-of-the-art protection and grounding issues in DC MGs. Finally, it addresses standardization efforts for DC systems, pointing out important future research directions.

Uploaded by

Gopika
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
37 views22 pages

2 DC Microgrids-Part II

This document reviews power architectures, applications, and standardization issues related to DC microgrids (MGs). It first discusses typical power hardware topologies used in DC MGs and their suitability for different applications. It then provides an overview of state-of-the-art protection and grounding issues in DC MGs. Finally, it addresses standardization efforts for DC systems, pointing out important future research directions.

Uploaded by

Gopika
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 22

3528 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO.

5, MAY 2016

DC Microgrids—Part II: A Review of Power


Architectures, Applications, and
Standardization Issues
Tomislav Dragičević, Member, IEEE, Xiaonan Lu, Member, IEEE, Juan C. Vasquez, Senior Member, IEEE,
and Josep M. Guerrero, Fellow, IEEE

Abstract—DC microgrids (MGs) have been gaining a continually [11]–[13], data centers [14], [15], and vehicular electric power
increasing interest over the past couple of years both in academia systems [16], [17]. Accompanied with technological advance-
and industry. The advantages of dc distribution when compared ments in power electronics and improvements in computational
to its ac counterpart are well known. The most important ones in-
clude higher reliability and efficiency, simpler control and natural power of real-time controllers, dc systems of today are capable
interface with renewable energy sources, and electronic loads and of achieving much broader functions than simple voltage reg-
energy storage systems. With rapid emergence of these components ulation. Hence, they are becoming an increasingly popular so-
in modern power systems, the importance of dc in today’s society lution for many types of residential and industrial applications,
is gradually being brought to a whole new level. A broad class of
such as dc-powered homes [18]–[22], fast EV charging stations
traditional dc distribution applications, such as traction, telecom,
vehicular, and distributed power systems can be classified under [23]–[29], hybrid energy storage systems (ESSs) [30]–[36], and
dc MG framework and ongoing development, and expansion of the renewable energy parks [37]–[43].
field is largely influenced by concepts used over there. This paper With regard to traditional ac, this kind of distribution has
aims first to shed light on the practical design aspects of dc MG proved to offer more efficient and reliable energy transfer (see
technology concerning typical power hardware topologies and their [44] for efficiency comparison of various architectures used in
suitability for different emerging smart grid applications. Then, an
overview of the state of the art in dc MG protection and ground- data centers). Calling on to this and other benefits, such as
ing is provided. Owing to the fact that there is no zero-current circumvention of problems with harmonics, unbalances, syn-
crossing, an arc that appears upon breaking dc current cannot be chronization, and reactive power flows, different types of dc
extinguished naturally, making the protection of dc MGs a chal- power distribution systems are experiencing more and more
lenging problem. In relation with this, a comprehensive overview of widespread use. Since extensive low- and medium-voltage dc
protection schemes, which discusses both design of practical pro-
tective devices and their integration into overall protection systems, distribution network infrastructure still does not exist, individual
is provided. Closely coupled with protection, conflicting grounding dc systems are nowadays either operated autonomously or con-
objectives, e.g., minimization of stray current and common-mode nected to ac utility mains. Therefore, it is convenient to classify
voltage, are explained and several practical solutions are presented. them as dc microgrids (MGs) [19]. This research topic has re-
Also, standardization efforts for dc systems are addressed. Finally, cently spurred a great amount of interest in research community
concluding remarks and important future research directions are
pointed out. [20], [45]–[55]. However, although dc MG appears to be a novel
technology, it is essentially stemming from a hybrid mixture of
Index Terms—DC microgrid (MG), power architectures, protec- several existing and ongoing research fields; one group being al-
tion and grounding, standardization.
ready mentioned traditional dc applications and the other being
I. INTRODUCTION the ac MGs area.
ROLIFERATION of renewable energy sources (RES), to- MG concept was originally proposed in 2002, as a building
P gether with ever more electronic loads and electric vehi-
cles (EVs) in modern power networks has prompted an idea
block of future low-voltage distribution systems which were
experiencing revolutionary changes at the time due to market
of considering the application of low-voltage dc distribution liberalization and introduction of distributed generation (DG)
to areas far beyond the traditional dc distributed power sys- [56], [57]. It was based on the idea of aggregating number of
tems (DPSs) [1]–[10], or power supplies for telecom stations microsources and loads into one unique entity, which could be
interpreted as a single dispatchable prosumer from the overhead
Manuscript received March 27, 2015; revised June 14, 2015; accepted July power system viewpoint. AC MG research has been very dy-
28, 2015. Date of publication August 4, 2015; date of current version December namic over the course of following decade for MGs have soon
10, 2015. The work of X. Lu was supported by the U.S. Department of Energy
(DOE) Office of Electricity Delivery and Energy Reliability. Recommended for after their introduction been recognized as a key element for in-
publication by Associate Editor T.-F. Wu. tegrating DG and storage, while allowing power independence
T. Dragičević, J. C. Vasquez, and J. M. Guerrero are with the Department from utility mains at the same time. From power electronics
of Energy Technology, Aalborg University, Aalborg 9220, Denmark (e-mail:
[email protected]; [email protected]; [email protected]). standing point, the roots of their development stem from control
X. Lu is with the Energy Systems Division, Argonne National Laboratory, of paralleled inverters in uninterruptible power supplies (UPSs)
Lemont, IL 60439 USA (e-mail: [email protected]). [58]–[62].
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. Simultaneously with the fast development in ac MG arena, Ito
Digital Object Identifier 10.1109/TPEL.2015.2464277 et al. reported one of the first dc MG experimental prototypes

0885-8993 © 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications standards/publications/rights/index.html for more information.
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3529

of 10 kW in 2004, claiming for simpler control, while superior


efficiency and reliability with regard to its ac counterpart [63].
Later on, several works followed up on it and demonstrated
usage of flexible dc MGs for telecom systems [64], distributed
renewable and storage plants [65], data centers [66], and resi-
dential households [67]. In the next years, accompanied by the
rapid emergence of consumer electronics, integration of ever
more RESs, and expansion of EV market with associated charg-
ing infrastructure, dc MGs have firmly established as a research
field itself.
Driven by practical requirements of specific dc MG appli-
cations, a number of hardware topologies have been proposed.
Principal criteria in that sense were control flexibility, robust-
ness, and reliability. As they are often opposed to each other,
Fig. 1. Single bus dc MG with ESS directly connected to the common bus.
different tradeoffs need to be taken into account before decid-
ing the most suitable configuration for a given application. For
its extreme robustness, direct connection of the battery stack
to the common dc bus has been historically a popular struc- tions such as EMerge Alliance, IEC, and IEEE are engaged in
ture for practical industrial applications. Some typical examples developing voltage and performance standards.
are power supplies for remote telecom stations [68] and rural- This paper is not only motivated by the recent outburst of pub-
based households [69]. Due to high capacitance of the battery lications in this area, but also by an immense amount of their
stack, dynamic stability of these kinds of systems is inherently real-world residential and industrial deployments [69], [85]–
assured [70]. Therefore, control design of converters that are [88]. It is intended to supplement its first part which is more
used to interface different sources is not a critical issue from the focused on control issues, stability analysis, and stabilization
bandwidth point of view, and thus, standardized components techniques for dc MGs. The organization of this paper is as
available in the market can be readily used. On the other hand, follows. In Section II, a configuration of the dc MG system is
by interconnecting battery through power electronics converter presented and various circuit topologies are shown. Section III
interface, flexibility of the system is greatly increased since the points out four practical dc MG applications and presents critical
voltage can then be actively regulated, rather than leaving it de- discussion about particular performance requirements for each
pendent on the battery operating conditions. Picking up on this one of them. Applicability of different architectures described
basic structure, a number of alternatives have been proposed, beforehand and control strategies presented in the first part of
including bipolar type dc bus [20], SST-enabled dc MGs [51], this paper are also evaluated. Section IV sheds light on the cur-
[71], [72], redundant dc bus structures [54], [73], and multiple rent status of protection for dc MGs. For that matter, discussion
dc MG clusters [19], [74]. about possible types of faults is followed by a description of
For safe and reliable operation of dc MG, a well-functioning prominent protection devices and grounding methods, and also
protection system is instrumental in any topology. Its princi- their integration into overall coordinated protection strategies.
pal objective is to minimize the propagation of disturbances by Overview of available dc standards and those under devel-
detecting and isolating faults within the minimum time frame opment is given in Section V. Finally, Section VI presents the
[75], [76]. Protection of dc systems is in general a challeng- concluding remarks.
ing task due to difficulties in extinguishing arc, which on the
contrary happens naturally in ac systems. Accurate short-circuit II. DC MG SYSTEM OVERVIEW
current calculation and fault detection are the most important
Typical dc MG systems are reviewed here. There are a num-
prerequisites for the good design of protection system [45].
ber of topologies that have been reported in the literature and
Moreover, an assessment of the influence of realistic protec-
some of their most prominent representatives are assessed in
tion devices and grounding methods on the total system per-
this section. In general, the topologies of dc MG can be clas-
formance becomes critical when deploying a protection system
sified into three categories, i.e., single-bus topology, multibus
in real-world environment. Up to this point, protection of dc
topology, and reconfigurable topology, respectively.
MGs was designed based on technologies and strategies taken
over from existing matured solutions developed for auxiliary dc
systems in big power plants and traction power systems [77]. Re- A. Single-Bus Topology
cently, however, a number of new coordinated protection strate- Single-bus topology is commonly used for dc MGs, and it can
gies have been proposed [78]–[84]. Apart from protection sys- be regarded as the basis of multibus systems. Fig. 1 shows an ex-
tems, standardization of future dc systems regarding dc voltage emplary schematic diagram of the most common type of dc MG,
levels, installation design, maintenance practices, and instru- which is frequently deployed in practical industrial applications.
mentation and control recommendations are at present moment It is based on the singular dc bus to which ESS, typically an elec-
the most important barriers to wide-scale adoption. Organiza- trochemical battery stack, is directly connected [89]–[92]. The
3530 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

the source voltages between 340, +170, and −170 V. Therefore,


the reliability of the system is increased since in case of failure
of one line, it is still possible to supply power by the other two
lines and an auxiliary converter.

B. Multibus Topology
Besides the single-bus topologies, dc MGs can be extended
to multibus configurations aiming at higher availability and re-
liability. An interesting topological modification for enhancing
the reliability of the system in that sense was proposed by Balog
and Krein [54]. It is based on the automated hot-swap principle
between more buses using the method of auctioneering diodes.
In that particular work, the case of two redundant buses is an-
alyzed. The location of intervention is on the load side, where
critical loads automatically select the bus to be supplied from
based on higher voltage. The structure of this system is shown
Fig. 2. Single bus dc MG with ESS connected to the common dc bus through in Fig. 4. Similar strategy for selecting the most appropriate
dedicated converter.
supply bus is presented in [73], but it uses a game-theoretic
approach. In that case, loads are more flexible to define the
way of selecting desired bus, and several objective functions are
tested for that matter. Besides, simultaneous supply from multi-
number of series battery cells is determined depending on the
ple buses is possible in order to enhance the total efficiency of the
voltage required by loads. Electrical power supply systems for
system.
telecom applications have been historically using this configu-
Multiple dc MG cluster configuration (see Fig. 5) is an alter-
ration operating at 48 V [93]. Many types of converters that can
native redundant solution [19], [74], [104]. In this way, every
be seamlessly connected to the bus operating around this nom-
MG is able to absorb or inject power from its neighboring MGs
inal voltage are available in the market (e.g., [94]). However,
in case of shortage or surplus of power, respectively. Addi-
despite inherent dynamic stability of the system, an uncontrol-
tionally, depending on the configuration in which the MGs are
lable voltage in the common dc bus, which depends mostly on
connected, some corrupted buses can be automatically isolated
the state of charge (SoC) and current of the battery, limits its
from the system in case of failure. Power exchanges between
application only to dense, singular bus systems [70]. Apart from
multiple dc buses are regulated by imposing appropriate local
that, it suffers from the practical problem of unregulated battery
voltage deviations. However, by dint of digital communication
charging as it needs to be performed coordinately by a number
technologies, total average voltage can be regulated to a nominal
of paralleled converters that have inherent imperfections in bus
value [74].
voltage measurement. This causes a circulating current problem
Low-voltage dc distribution systems can be interfaced to
that leads to uneven loading of those converters [4], [95]–[97],
medium-voltage ac utility mains through SST, as shown in
and also to accelerated wear and tear of stationary batteries [70].
Fig. 6. With connection of a number of SSTs to medium-voltage
On the other side, connection of the ESS through dedicated con-
ac network, it is envisioned that the energy management in lower
verter interface to a regulated low-voltage dc bus (see Fig. 2)
voltage levels will be completely in SST domain, largely sim-
allows application of more flexible control and possibility of
plifying the task of system operator in the ac grid above SST
connecting multiple buses for enhancing the reliability of the
[72]. However, although it is widely deemed that SSTs will
system or to supply loads in a wider area.
bring a new revolution to future distribution systems, serving
Moreover, the reliability of this system can be increased by
as the true enablers of dc distribution architecture, they are still
using multiple battery stacks. This topology, operated at low
in the early development stage [105]. Recognizing its poten-
voltage, has by now been the most widely studied in the lit-
tial, many academic and industrial projects around the world,
erature [15], [18], [46], [48]–[50], [55], [98]–[103]. However,
such as FREEDM [106], UNIFLEX [107], and HEART [108]
its attractive features are partially counterbalanced with several
are currently underway, focusing on pushing this technology to
associated technical problems. First, there now appears a need
the next technical readiness level. Speed of its further evolution
for careful design of circuit and control parameters due to a fact
will to a large extent determine the significance of the role of
that equivalent capacitance of dc bus gets much smaller than in
dc distribution in future power systems. It is expected that pre-
case of direct battery connection. Second, there exists only one
vious advances in dc MGs area will play an instrumental role in
dc bus, compelling the consumers to be powered from it.
realizing the SST scenario [51].
As a practical remedy to the aforementioned problem, several
alternative concepts have been examined. Kakigano et al. pro-
posed a bipolar single regulated bus structure (see Fig. 3), that C. Reconfigurable Topology
operates at ±170 V [20]. The main benefit of this configuration Besides the aforementioned architectures of dc MG, recon-
is a fact that load-side dc–dc converters have an option to choose figurable topologies were also proposed for the purpose of
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3531

Fig. 3. Bipolar single bus dc MG.

each dc distribution unit in the zonal architecture is powered by


the utility grid separately, fault can be isolated within each unit
without influence on the operation of other parts of the zonal
system. This topology is similar to the redundant bus system
proposed in [54], but it has better flexibility due to higher number
of switches.
Multiterminal dc (MTDC) system is proposed in [109]–[111],
Fig. 4. Dual-bus, separately fed dc MG with a bus selector at each load. as shown in Fig. 9. It should be noted that the MTDC architec-
ture was originally proposed for high-voltage dc (HVdc) sys-
tems, e.g., off-shore wind farms. In this configuration, the dc
increased flexibility during the faults or periodic equipment distribution network is fed by multiple input terminals. The
maintenance periods. Fig. 7 shows a dc ring bus architecture system usually features mesh configuration. Compared to con-
[81]. Each node and the link between neighboring nodes are ventional radial configuration, power flow in mesh configura-
controlled by intelligent electronic devices (IEDs). The main tion is more complicated. However, it can be also more flexible
merit of this configuration is high reliability and redundant since the system connection provides multiple paths for power
operation. Since the load connected to the common dc bus can transmission.
be fed bidirectionally, alternative path is provided at the ring Table I sums up the hardware topologies that have been
bus in case of fault. When encountering a fault in this dc MG, presented above and gives critical comparisons from different
the fault is isolated by IEDs first, and then the load power turns points of view. The voltage levels of each configuration reported
to be supplied by the sources in the normal alternative path. so far in the literature are listed. Some of the configurations
Another dc MG architecture that features high reliability is can be implemented by using standardized off-the-shelf com-
based on zonal configuration, as shown in Fig. 8 [75]. It is seen ponents, such as industrial inverters. However, for the others,
that this type of dc MG is comprised of multiple dc distribution specialized components are required. For example, dc voltage
units, with each one of them being connected in series to form balancers are needed for the configuration with bipolar regu-
a zonal architecture. Note that each dc distribution unit in the lated bus, while SSTs are required for SST-enabled MGs. IEDs
zonal configuration has two dc buses. These two buses form should be used for reconfigurable dc MG architectures. Consid-
the redundant configuration and the required load power can be ering the impact of ESS, its type of connection to the dc MG
flexibly obtained by either one of them. In particular, assuming is also summarized. In relation with this, it is shown that con-
that the load power is supplied by the upper bus, in case of its figurations which can incorporate directly connected ESS have
failure the switches at the upper side are turned OFF and the inherent stability. Final two columns of the table present two
switches at the lower side are turned ON to change the source bus performance metrics of each configuration, i.e., expandability
and ensure the normal operation of the load. Meanwhile, since to multiple buses and reliability.
3532 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

Fig. 5. Multiple bus dc MG with all components connected through dedicated converter interfaces.

The single converter to link the dc MG and ac grid is usually


used in small-scale systems. In [112], a PV-wind-battery hybrid
MG is taken as an example to represent the ac and dc hybrid MG.
The PV panel and battery form the dc bus, and the wind turbine
and ac utility grid form the ac bus. Since the power capacity of
the above system is typically not so large, a single bidirectional
Fig. 6. SST-enabled dc MG. converter is used to link the dc subgrid and ac subgrid. In [113],
an integrated configuration is employed with one ac–dc–ac con-
verter as the energy hub to connect both ac and dc subgrids. In
D. Interconnection Between DC MG and AC Grids
particular, the ac MG and ac utility grid are connected to the two
In order to ensure the power availability and reliability, dc ac sides of the ac–dc–ac converter, respectively. Meanwhile, the
MGs can be interconnected with existing ac infrastructures dc link of the converter is connected to the dc MG. Systematic
through dc–ac converter interface. In that way, either an ac power flow management among the ac utility grid, ac MG, and
grid connection or a hybrid ac/dc MG can be established. In dc MG is designed to achieve the coordinated control.
any case, dc link is typically used for efficient local coupling of In order to increase the exchanging power capability between
dc-type components. The interface toward ac grid is reflected dc MGs and ac grids, parallel interface converters can be used
in most architectures presented in previous sections and can be between the ac and dc bus. In [114], parallel converters between
achieved by unidirectional or bidirectional dc–ac converters. In ac and dc bus are analyzed in detail. The model of the parallel
case of a hybrid MG, dc MG can be regulated to absorb power converters is derived and the control system is designed to sup-
from the ac MG when it has surplus of power. On the other hand, press the zero-sequence circulation current. In [115], parallel
when load requirements exceed the available power, dc MG can bidirectional interlinking inverters are used to connect the dc
deliver it back to the ac side (only if bidirectional interface is subgrids and ac subgrids. An improved droop control method
used). based on per-unit frequency and amplitude is employed to en-
Focusing on the power architectures of the interconnecting hance the power sharing among different converters. Mean-
energy conversion systems between dc MGs and ac grids, some while, it is pointed out that the proposed method also applies for
work has been done in the existing literature. In general, the different architectures with both ac and dc subgrids. In [116],
interconnecting system can be classified into three categories, different operation modes are studied and the one with ac to dc
i.e., single, parallel, and aggregated converter system. power flow is discussed in detail. A hierarchical control system
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3533

Fig. 7. DC MG based on ring bus.

for ac and dc hybrid MGs. In [117], a hierarchical architec-


ture is proposed to aggregate the dc MG with ac MG and ac
utility grid. In Level 1, which is a microsource level, dc MGs
are formed to integrate the DGs and energy storage units. In
Level 2, combo-source level, each dc MG is connected to an
ac link by using dc–ac bidirectional inverters. In Level 3, MG
level, several ac-coupled combo sources are aggregated to form
a MG and to achieve the functions of optimization, synchro-
nization, dispatching, protection, etc. Finally, each MG in Level
3 is connected to the ac utility grid, respectively. In [118], the
sources and loads in an ac and dc hybrid MGs are categorized
into three groups, i.e., DG park, dc MG, isolated ac MG, and
Fig. 8. Zonal dc MG. ac utility interface. An admittance-based model is derived and
the overall system is thereby studied from the dynamic stability
point of view. Considering the power architectures reviewed in
Sections II-A–II-C, their capability of interconnecting with ac
utility grid or ac MG depends primarily on the availability of
high enough dc-link voltage level. For that matter, only in the
single-bus dc MG with directly connected battery, there is no
possibility of direct dc-link voltage control. However, it may still
be connected to utility mains if an adequate number of battery
cells are connected in series.
As a final remark, it should be noted that although the power
architectures were presented in this section within the MG
Fig. 9. MTDC MG.
framework, they have a quite general structure. In that sense,
the vast majority of them can be used in a very broad voltage
and power range, given the proper selection of semiconductor
is used to ensure the reliable operation of the whole ac and dc devices and/or their arrangement within the multilevel converter
hybrid MG. structures. For that matter, a resemblance with typical architec-
Apart from the configurations with single and parallel inter- tures used in HVdc systems can be readily observed (see, i.e.,
face converters, the aggregated configuration is also employed [119]). In the respective reference, singular bus, bipolar bus,
3534 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

TABLE I
OVERVIEW OF HARDWARE TOPOLOGIES FOR DC MGS

DC Bus Reported Voltage Standardized Direct ESS Inherent Expandability to Reliability


Configuration Levels (V) Components Connection Stability Multiple Buses

Direct battery connection 12, 24, 48 Yes Yes Yes No High


Single unipolar regulated bus 24, 48, 380 Yes No No Yes Medium
Bipolar regulated bus ±170, 340 No No No Yes Medium
Multiple regulated buses structure 48, 380 Yes No No Yes Medium
SST-enabled MG 380 and higher No Possible If ESS directly connected Yes Medium
DC ring bus Lab tests made at 24 V but suitable for higher No No No Yes High
Zonal DC MG 380 or higher Yes Possible If ESS directly connected Possible High
MTDC MG 380 or higher, and commonly used for HVdc systems Yes No No Yes High

multiple bus, multiterminal, and ring bus structures have been


shown to be used in HVdc systems up to levels of 350 kV. The
exception is the topology presented in Fig. 1, which has a direct
connection of the battery stack that is not typical for HVdc sys-
tems. Therefore, that topology is voltage limited by the battery,
and the required voltage in that case is achieved by stacking
up the number of low-voltage battery cells in series. However,
it is not viable to increase the number of cells indefinitely due
to problems associated with balancing of individual cells and
reliability (if one cell fails, the whole stack will fail). To the
best knowledge of the authors, the highest reported voltage of a
battery stack by now is 6 kV [120].
The thorough discussion about the applications of each topol-
ogy specifically for four emerging dc MG applications is pro-
vided in the following section. It presents the typical perfor-
mance requirements and applicability of corresponding power
architectures, together with the summary of suitable control
strategies in that they can achieve the imposed objectives. More- Fig. 10. Applications of dc MGs.
over, the typical voltage and power limitation limits for each
application are given.

A. Performance Requirements for DC MG Applications


III. APPLICATIONS OF DC MGS TO FUTURE SGS Desirable performance characteristics and functionalities of
Taking into account the advantages of dc MGs, their usage has MGs in general were presented in [75], [121], and [122]. For
been proposed for a number of different applications in future completeness of this paper, they are summarized here by con-
SG, with each one of them having its own specific require- sidering both the control and architectural perspectives. From
ments. In this section, a critical view on performance require- control point of view, MG should be operated in a stable way and
ments of four modern dc MG applications is presented. They provide sufficient power quality at the point of common cou-
include high-efficiency households, renewable energy parks, hy- pling (PCC). Moreover, it should provide functionalities, such
brid ESSs, and EV fast charging stations, as shown in Fig. 10. as intelligent coordination of its internal units, as well as grid
While it should be acknowledged that the considered list is by support capability when it is in grid-connected mode. These can
no means complete, we have limited our discussion only to these be summarized as follows:
novel applications of dc MGs. More details about some well- 1) stability: sufficient stability margins during normal and
known and matured examples, such as DPSs [1]–[10], telecom- transient operating modes;
munication [11]–[13], data centers [14], [15], vehicular electric 2) power quality: satisfactorily voltage waveform in the com-
power systems [16], [17], can be found in the existing literature. mon dc bus;
Considering the specific characteristics of various power ar- 3) coordinated control: supervisory energy management sys-
chitectures that were reviewed in Section II of this paper together tem responsible for carrying out complex functions, such
with control strategies that were reviewed in the first part of these as economic dispatch, unit commitment, mode chang-
two articles, a critical judgment in selecting the most suitable ing, efficiency optimization, and power flow control
ones for every given application is provided below. Moreover, between internal clusters;
typical voltage and power limitations, as reported in the litera- 4) grid support capability: ability to provide ancillary ser-
ture, are presented. vices to the overhead ac power system.
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3535

TABLE II
TYPICAL PERFORMANCE REQUIREMENTS FOR VARIOUS DC MG APPLICATIONS

DC MG Application Hardware Architecture Control Strategy

Redundant Power Flexibility Scalability Coordinated Control Grid Support


Supply Capability

High-Efficiency Importance Level Low Medium High High Medium


Households
Description Leads to higher Flexible clustering of Should continue Efficiency and Households are
installation cost. households is a working with new economic conceived to support
promising future appliances integrated optimization in both the grid indirectly—
concept. into household. single household and according to the
cluster should be price signal.
achieved.
Renewable Energy Importance Level Medium Critical Medium High Critical
Parks
Description Individual faulty unit Flexibility in terms Number of units Required when there Required by grid
can be bypassed so of bypassing faulty within a branch can is a need to codes for renewable
redundant power elements should be be defined coordinately increase energy systems.
supply is not highly guaranteed for fault considering the or reduce power
required. ride through and required voltage production of
occasional level but more multiple generators.
maintenance of RES branches can be
generators. added if the power
rating of central
inverter permits.
Hybrid ESS Importance Level Medium Low High High High
Description Would not give too Small and dense Plug and play of Degradation of ESS Important for grid
much benefit due to a systems formed additional sources and grid price signal support in some
fact that all units are around a common dc for expanding energy can be integrated into cases such as
typically connected bus. Probability of and/or power cost function. renewable energy
to the same point. fault on dc bus is capacity is highly parks or EV charging
extremely low. desirable. stations.
EV Fast Charging Importance Level Medium Medium Medium High High
Stations
Description Too costly and not In case of chargers Plug and play Optimization of Grid codes are still
competitive with aggregated in capability of charging strategies, not adopted, but will
present commercial parking lots, additional ESS will charging scheduling, become important
solutions. reconfiguration and be a desirable feature energy management issue when they do.
fault ride could when grid codes start between grid and
increase reliability. to be imposed. ESS.

On the other hand, from architecture point of view, function- either by inserting passive elements within the converter filter
alities that can be achieved through appropriate power architec- or adding control loops within the local control, respectively,
ture include redundancy, flexibility, and scalability. They can be [123], [124]. On the other hand, power quality is related to the
summarized below: waveform of the voltage in common dc bus, and influence over
1) redundancy: redundant power supply for the loads with it can also be achieved by proper combination of low-level ar-
minimized reliance on communication; chitecture and control. One example of such a manipulation is
2) flexibility: possibility of reconfiguring the system online interleaving number of dc–dc converters to reduce the voltage
during faults and/or to form MG clusters; ripple and improve the transient response [125].
3) scalability: possibility to seamlessly add or remove units Here, we attempt to assess the extent to which the remain-
within MG. ing five benchmark functionalities, i.e., redundancy, flexibility,
Among the items given above, stability and power quality in scalability, coordinated control, and grid support, are relevant
control strategy category can be interpreted as characteristics of for viable performance of every addressed application. For that
the system. Although they have been treated in unitary fashion matter, we designate each one of them with one of the four levels
in [75], [121], and [122], here we draw the line between them of importance, i.e., low, medium, high, and critical.
and the other functionalities. The reason behind this is a fact that Detailed elaboration about particular applications and impor-
stability and sufficient power quality are mandatory objectives tance of specific functions for their proper operation are given
for every dc MG without exception. Also, the interventions to in following sections. The resulting evaluation is summarized
enhance stability margins and power quality mainly rely on local in Table II. Moreover, through each of the sections, practical
control diagram with the assistance of system level coordinated hints on how to meet specific objectives by choosing adequate
control to limited extent. The function of stability enhancement combination of hardware architecture and control strategy are
is programmed at the converter architecture and/or lowest con- presented. Summary of that particular assessment is presented
trol level. In particular, it can be improved by incorporating in Table III. In addition, Table III shows the typical voltage and
passive and active damping techniques, which are implemented power limitations reported for each particular application.
3536 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

TABLE III
SUITABLE CONTROL STRATEGIES AND ARCHITECTURE

Application Control strategies Architectures Typical voltage limitation Typical power limitation

Future households Appliance scheduling—centralized, Single bus dc MG For safety reasons, the maximum Up to several tens of kilowatt for
decentralized or distributed Bipolar single bus dc MG voltage for this application is individual households.
Adaptive droop control—for the cases Redundant bus system—for commonly selected as 400 V.
when ESS are installed within the mission critical applications
households Multiple dc MG clusters
Renewable energy parks MPPT Single bus dc MG—for smaller Medium-voltage dc buses have Largest dc onshore and offshore
Grid support functionalities scale parks been deployed to avoid multiple renewable grids are in range of
Coordinated control (when reduction Ring bus dc MG etc., with step-up transformers. A solution several hundred megawatts [156],
of available production is enhanced fault-tolerant capability utilizing 7.5 kV has been reported [157].
required)—typically centralized Multiple dc MG clusters with in [155].
fault isolation Highest analyzed voltages for
capability—virtually equivalent to offshore wind parks achieved by
series–parallel collector grids using series dc configuration were
SST-enabled dc MG—for larger up to 300 kV [156].
parks with connection to MV ac
networks
Hybrid ESS SoC estimation Single bus dc MG with direct The vast majority of hybrid ESSs Total power capacity depends on
Charging algorithms connection of battery reported in the literature is application but will in general
Adaptive droop control—equalizing Single bus electronically designed for 24, 48, and 380 V dc never exceed the nominal power
SoC of multiple ESS of same type regulated bus dc MG buses. However, higher common of a given application. Commonly,
DC bus signaling—coordination dc bus voltages such as 750 V it is sized at approximately half of
between ESS of different types have been used in some the nominal power (e.g., for
applications [158]. For a sole smoothing power fluctuation of
battery storage, 6-kV prototype 2-MW wind turbine, optimum of
has been developed [120]. 1-MW hybrid ESS is computed
[159] while in a generic dc MG of
10 kW, 6 kW is used [160]).
EV charging stations DC bus signaling—current sharing Single electronically regulated dc 5-kV MV dc substation for Maximum individual dc type
between grid-tied converter and ESS bus for single charging station powering EVs was reported in chargers are rated at 240 kW
converter Multiple dc MG clusters or ring [161]. However, much more [163]. However, more of these
Adaptive droop control—sharing bus dc MG in case of multiple common solution is to use step may be connected in parallel
between multiple ESS parking lots down transformer and establish around the common dc bus.
SoC estimation SST-enabled dc MG for larger lower voltage dc bus through
Consensus algorithms—alternative for stations with MV ac connection rectifier interface [24], [162].
coordination between grid tied and Ring bus dc MG etc., with
ESS converter enhanced fault-tolerant capability
Charging algorithms—on site
chargers typically have preinstalled a
number of algorithms for different
HEV batteries
Grid support functionalities

B. High-Efficiency Households future grids with high penetration of renewables [130], [131].
For that matter, real-time price signal is predicted to be the main
Since the majority of modern household energy sources and
driving factor that will motivate local households to rethink their
loads, e.g., PV panels, batteries, LEDs, and consumer electron-
ics operate naturally at dc, it is more efficient to connect them distribution of energy.
In order to achieve flexibility, it is important to establish co-
around the dc bus to form a dc MG [126]. There are several
demo projects of dc high-efficiency houses all over the world ordinated control of appliances within the households. It should
shown in the existing publications and project websites, such as be noted that integration of local ESSs with controllable appli-
ances can also greatly aid to this functionality. Finally, all these
USA [21], [22], Japan [20], Korea [127], and Denmark [128].
In contrast to the traditional ac system, dc system has higher units can be grouped into a number of clusters, where each one
efficiency. The energy saving opportunities by changing infras- of them could absorb from or inject energy to other neighboring
cluster(s), depending on the conditions of the system. In this
tructure from ac to dc can be up to 15% [129].
Except of benefits from the efficiency point of view, the con- way, even higher degree of flexibility and, consequently, larger
trol simplicity in case of dc makes it a much better option for potential to obtain concrete economic benefit can be obtained.
While clustering of households is still at notional level, there are
integration of energy management systems. For this reason, dc
MGs are excellent candidates for households that could greatly several published works that help in understanding the benefits
of this concept [19], [74], [132].
facilitate the achievement of objectives imposed by a number of
The most common objectives in control of future house-
present SG roadmaps, which state that increase in flexibility1 of
household consumers is a key for sustainable operation of the holds are reduction of the energy cost and maximization of
customer comfort [132]–[137]. These objectives are most com-
monly achieved by scheduling the appliances within the house-
1 Flexibility of consumers metioned here is related to controllable power
holds through various unit commitment strategies. In that sense,
extraction from the grid and is different from the hardware flexiblity adressed
in Section III-A.
the respective appliances are typically coordinated on a higher
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3537

level in ON/OFF fashion and are not responsible for the voltage developed. This kind of converter can supply both ac and dc
in the common dc bus. Therefore, the local control strategies components, simultaneously, which makes it possible to behave
do not apply since the voltage is regulated to a constant value as the interface to the local dc loads and external ac grid. In
by a grid-tied converter. Coordination based on the scheduling [43], a solar converter with high voltage gain is proposed to har-
of appliances using decentralized [132], distributed [133], and vest maximum power from solar panels. From the system level
centralized [134]–[137] control methods have been proposed. point of view, the energy management system for renewable
Higher level of flexibility can be achieved if complimentary energy parks has also been studied. The cost and stability of a
ESSs are used within the households. In that case, some of dc-coupled renewable energy park with hybrid sources is ana-
the control techniques presented in the first part of this paper lyzed in [145], where the forecasting model for different sources
can deployed. For instance, in [138], the control method for a and loads is also developed. In [146], the optimal allocation and
PV-based building with coordinated operation of batteries has economic analysis for the energy storage units for renewable
been proposed to avoid the permanent active power injection to energy parks is studied. In [147], the detailed power loss model
the utility grid. In [20] and [139], the bipolar configuration of is derived to optimize the system efficiency.
dc MG has been proposed, which is useful to further enhance Regarding typical hardware configuration of large collector
the dc power quality. Meanwhile, fuzzy droop control has been grids, a number of wind turbines or PV generators are connected
employed to reach the dynamic balancing of the SoC of each in series to form a branch, which operates at a nominal dc
energy storage unit. In [140], batteries and heat pumps have voltage. Then, number of branches are connected in parallel to
been used as controllable loads in order to eliminate the power form a complete collector grid [148]. A critical function here
fluctuation by PV plant at the PCC. An optimized search algo- is a fault ride through since stop of production presents a big
rithm was deployed to minimize the difference between highest economic loss. In particular, when one unit in a branch fails, it
and lowest power flow from the utility mains. is necessary to bypass it in order to maintain the healthy units
Regarding suitable hardware architectures, the single bus sys- operational. On the other hand, branches are typically designed
tem shown in Fig. 2 is the most suitable configuration. In special for a specific voltage, and if the power rating of the central
cases, where redundant power supply is needed, configurations inverter permits, additional branches can be added to expand the
from Figs. 3 or 4 can be used. Finally, structure from Fig. 5 power rating of the whole system. In relation to architectures
is appropriate for the application of multiple household cluster presented in Section II, single bus system of Fig. 2 can be used
concepts. for smaller scale applications. On the other hand, for larger
parks, connection to MV ac network can be achieved through
SST, as shown in Fig. 6. Finally, as series–parallel collector
C. Renewable Energy Parks grids resemble the structure of multiple clusters of Fig. 5, this
A number of RESs can be clustered around the common dc configuration can be used to achieve desired dc-link voltage by
bus to form an energy park. In that way, the system efficiency connecting number of generators in series. It should be noted
can be increased and issues such as reactive power circulation that for the sake of maintenance and fault ride through capability,
and harmonics avoided [126]. These kinds of dc systems are it is mandatory to incorporate possibility of bypassing any given
sometimes also referred to as the collector grids and are used generator. In order to enhance the fault-tolerant capability, dc
for both photovoltaic and wind power applications [141], [142]. ring bus configuration in Fig. 7 and other configurations with
Compared to households, they present a different perspective, higher reliability can be considered.
since they inject large amounts of power into the conventional
ac power system and have more direct implications on its op-
eration. Therefore, they need to be able to provide a number of D. Hybrid ESSs
ancillary services defined by well-established grid codes [143], ESS technologies are characterized by several important per-
[144]. As compliance with grid codes often implies reduction of formance indicators, such as energy efficiency, energy and
active power injection to the grid, coordination between multiple power density, cycle life, and rate of self-discharge [149]. In
renewable generators, which should achieve this task simulta- that sense, for situations when a single ESS is not able to meet
neously, is mandatory. Several applications of dc collector grids the prescribed requirements of specific application, it is advis-
are presented in the following paragraph. able to combine two or more technologies that complement each
A highly integrated dc bus module is developed in [39]. It other. Since dc distribution offers a natural interface for practi-
provides the functionality to connect parallel interfaces for RES cally all types of ESS, these kinds of hybrid ESS are frequently
which share the same dc bus and have the same dc output value. deployed around the common dc bus.
In [40], a dynamic adaptor is proposed in a PV-based dc MG. A large number of hybrid ESS applications have been re-
The adaptor is connected at the output side of the solar field to ported in the literature and only several selected ones are shown
cope with the randomness in the electrical production. Mean- here. From architectural point of view, the majority of applica-
while, it can provide auxiliary function when the local dc MG is tions are small and dense systems gathered around the common
connected to the utility ac grid. In [41], in order to maximize the dc bus, which can be established either through direct connec-
power utilization, the design and implementation of the multiple tion of the battery stack or through regulated power electronic
MPPTs is presented. The coordinated operation among differ- converters. Since the bus is small, the most common faults in
ent trackers is studied. In [42], a switched boost converter is these kinds of systems are directly on storage units and design
3538 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

of fault ride through capability based on bypassing the faulted signal, consideration about degradation of particular ESS tech-
part of the common dc bus would unnecessarily complicate the nology, and grid friendly operation. Additionally, this type of
system. On the other hand, plug and play capability is a highly multiobjective control can be achieved in myriad of ways. In the
desirable feature which can be exploited either for extending following paragraph, several approaches used in fast dc charging
the power or energy capacity of the system [55]. Optimization applications are provided.
of charge/discharge processes among different technologies is In [27], the traditional power quality problems caused by
another important aspect. Objective function in this regard usu- the integration of multiple EVs have been analyzed. In [29], the
ally incorporates terms related to degradation of respective ESS power flow related to the dc charging station is optimized online
technologies according to imposed mission profiles and can be to reduce the power line losses. In [154], the economic impact
executed online [150] or offline for capacity optimization [151]. to the cost and benefits for installing the EV charging station is
In grid-connected applications, variable grid price signal can be studied. Distributed bus signaling strategy was used in [23] to
incorporated as well within the objective function. In the remain- coordinate operation of grid and ESS converters. Flywheel based
der of this section, several hybrid ESS applications are reviewed. on induction machine was used in this particular application
The most common representative of hybrid ESS is secondary since it does not introduce significant performance degradation
battery in combination with supercaps. In [33], a model predic- problems with cycling as batteries.
tive control has been employed to enforce super-caps to absorb In order to reach an efficient power conversion in EV charging
only fast power changes, prolonging the lifetime of the battery stations, different topologies of power electronic converters can
in that way. Superconductive magnetic energy storage (SMES) be used. In [28], the topologies of nonisolated bidirectional dc–
was combined with battery in [34]. In this case, fuzzy control dc converters are reviewed to derive the suitable solution for the
was used to make SMES act as a power buffer, while the battery EV charging station with multiple parking decks. In [105], SST-
is responsible for slower variations. A single synchronous buck based fast EV chargers are developed to reduce the size of the in-
converter was used as an interface between fuel cell and bat- terfacing converters. It can be noted that configurations like sin-
tery in [35]. Associated control was designed with the objective gle bus dc MG in Fig. 2, multiple dc MG clusters in Fig. 5, SST-
to share the pulsed current with respect to permitted voltage enabled dc MG in Fig. 6, and fault-tolerant configuration such
ranges of the two ESS technologies. Similar, but more reason- as ring bus dc MG in Fig. 7 can be used in case of parking lots.
able approach which uses supercaps instead of batteries, and Based on the above discussion for each specific application of
bidirectional isolated dc–dc converter, was presented in [36]. dc MG, the suitable control strategies and architectures for each
With the aim to operate compressed air energy storage at its of them are summarized in Table III, and the typical voltage and
highest efficiency during every operating condition, its supple- power limitations for each application are also shown in this
mentation with super-caps was investigated in [152]. Finally, table.
as larger architectures in case of hybrid ESSs would bring un-
necessary losses and complexity, the two most common power IV. GROUNDING AND PROTECTION ISSUES IN DC MGS
architectures are those shown in Figs. 1 and 2.
As opposed to many advantages that dc systems bring in
relation to ac, design of their protection represents a consider-
E. EV Fast Charging Stations able challenge. The most important aggravating circumstance
in that view is problem with extinguishment of dc arc which
EVs have gained increasing attention due to potential of re-
inevitably appears upon the tripping of protective devices. On
ducing the greenhouse gas emissions and conventional fuel con-
the other hand, as worldwide industry has just recently started
sumption. However, it has been shown that their large-scale in-
to conceive dc MGs as serious actors in future SGs, there is
tegration could have a significant adverse effect on the electric
still a general lack of understanding and experience with opera-
grid [26] and dc MGs have been conceived as a promising solu-
tional issues of these systems [75]. Therefore, widely accepted
tion to support the operation of the power system by aggregating
protection standards and guidelines are yet to be defined.
multiple EVs [153]. Fleets of PHEV chargers represent a spe-
The aim of this section is to update the current status in dc MG
cific group of consumers with an outstanding potential from the
protection area. For that matter, we address four main aspects:
increase of flexibility standpoint.
1) types of faults; 2) grounding options; 3) protective devices;
It is deemed that grid codes similar to those imposed for
and 4) design of protection systems.
RES will be developed for fast dc charging stations, especially
when the level of their penetration becomes more significant.
In contrast to vehicle to grid where energy stored in vehicle’s A. Types of Faults
battery is used for grid support, charging stations with supple- Fig. 11 shows two basic types of faults that can exist in dc
mental ESS have been proposed not to disrupt the recommended power systems, i.e., line to ground and line to line faults [82]. It
charging pattern of the vehicle’s battery. In that sense, different should be noted that although less serious, line to ground faults
types of energy storages can be employed in the dc EV charg- are the ones most frequently appearing in industrial distribution
ing stations, such as flywheel, battery, and hybrid ESSs [153]. systems [164]. Line to ground fault can have either low or
Internal coordination between the ESS and grid is an impor- high fault impedance, while line to line fault typically has a
tant but challenging task, since number of objectives need to low impedance [45]. Moreover, a fault can occur at various
be simultaneously met, i.e., tracking of the variable grid price points in the system, and each one of them will have different
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3539

optional and exist only in systems which include direct interface


of the battery to the common dc bus (e.g., system shown in
Fig. 1). On the other hand, the impact of line to ground faults on
system performance depends heavily on the grounding method.
More details about the grounding methods are given in the next
section.

B. Grounding
The most important objectives of the system grounding are to
Fig. 11. Two fault types in dc systems. (a) Line to ground fault. (b) Line to facilitate fault detection, minimize stray currents (leakage cur-
line fault. rent from conductor to the ground), and to ensure the personnel
safety by minimizing the common-mode voltage (potential dif-
implications. For instance, unlike feeder faults that can be ference between device conductor and ground) [167].
rapidly isolated from the main bus, faults at the main bus itself Corrosion is the main problem associated with stray current.
present a greater danger since all the feeders connected to it get The phenomenon of stray-current-induced corrosion was first
affected [45]. observed at the end of 19th century, in dc-powered electrified
When designing the protection system, it is important to iden- rail transit systems [168]. The problem has been initially at-
tify all possible fault locations and to calculate the fault current tributed to the chemical structure of the soil surrounding the
for every given location [165]. All active sources within the rails, but it was soon realized that the soil cannot be the sole
system participate in the formation of the total fault current and contributor to it since corrosion has been observed in all metallic
their actual contributions depend on the effective impedances structures around the rail base. Therefore, it has been concluded
between the respective source and the fault locations. For in- that the leakage or stray current is the main culprit for sig-
stance, in case of bus fault, the contribution of a battery which nificant amplification of this undesirable effect. Stray currents
is directly connected to the common dc bus via cable can be cause corrosion at the locations of current leakage from conduc-
expressed as follows: tor into the soil. Since the current flow in conductor is primarily
ubat   electronic, and ionic in the soil (i.e., ground), an electron to ion
ibat (t) = 1 − e−t/τ e f f (1) transfer is inevitable as current changes its transmitting medium.
Rbat + Rcable In particular, this electron producing or oxidation reaction can
where ubat is the battery open-circuit voltage at the time of the be represented by the following [169]:
fault, Rbat and Rcable are battery and connecting cable internal
resistances, respectively, while τeff = (Lbat + Lcable )/(Rbat + Fe → Fe2+ + 2e− . (4)
Rcable ) is the time constant of the effective battery/cable system. The reaction shown above is known as corrosion if it takes
On the other hand, if a battery (or any other type of source) place over a prolonged period of time. In a broad sense, cor-
is connected to the bus through power electronics converter rosion of metallic objects will occur always at places where
interface, current resulting from quick discharge of its filter current transfers from them to the surrounding environment. By
capacitor plays a pivotal role in the fault current undertaking the measures to reduce the stray current, the corro-
udc −t/τ c sion rate will be reduced as well and severe structural damages
ic (t) = e (2)
Rc to conductors avoided. Apart for traction applications, mod-
where udc is the common dc bus voltage at the time of the eling, assessment, and stray-current control methods have been
fault, RC is the capacitor series resistance and τc = Cdc Rc is recently also proposed for large photovoltaic plants [170], [171].
the time constant of the effective converter/common dc bus RC Another important metric of grounding system performance
system. If there exists a cable of considerable length between is the level of common-mode voltage. It is referred to a phe-
a converter and common dc bus, corresponding parameters of nomenon where the negative pole of the bus fluctuates as the
the cable should be included in calculation as well, resulting neutral of the ac side that supplies dc bus through converter in-
in a higher order dynamic system [164]. Similarly, faults that terfaces [172]. In traction applications, this voltage is commonly
appear on feeders can be treated just as special cases of bus referred to as the touch potential [167]. Although the potential
faults and their current contribution can be obtained by looking difference between the positive and negative pole will be con-
into effective impedances [45], [164], [166]. trolled to a constant value, this kind of fluctuating phenomenon
Total fault current can be calculated as a summation of current presents a challenge for the operation of the system. In partic-
contributions from the battery, which is directly connected to the ular, it gets mapped to the neutral points of all the loads and
common bus and all other source converter currents [82] generators, contributing to the loop currents between the inter-
connected converters. More importantly, common-mode voltage

N
presents a personnel safety concern if it gets exceeds the value
ifault (t) = ibat (t) + ic,j (t) (3) of 60 V, which is generally considered as a safety threshold
j =1
[167].
with N being the number of source converters connected to the It is important to note that the common-mode voltage and the
common bus. It should be noted that ibat (t) in equation above is level of stray current are related to each other by the grounding
3540 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

resistance. More precisely, for a very high grounding resistance,


there will essentially be no stray current, but the common-mode
voltage will take its maximum value. On the contrary, if the
system is solidly grounded, there will be virtually no common-
mode voltage, but stray current will be the highest. Therefore,
the objectives to minimize both of them at the same time are
conflicting with each other. However, with a proper grounding
strategy, this tradeoff can be explicitly taken into account.
Different grounding alternatives have been proposed for dc
power systems. Some references suggest ungrounded operation,
especially for low-voltage applications [75], [173], [174]. The
reason for that is a fact that the common-mode voltage will typi-
cally not be on such a high level to present a danger for personnel Fig. 12. Block diagrams of reconfigurable grounding systems. (a) Diode
safety, and will offer a possibility of continuing operation even grounded system. (b) Thyristor grounded system.
in case of a single phase-to-ground fault. Still, it is instrumental
then to discover and correct a possible second ground fault or it TABLE IV
will create a line-to-line fault [175]. SUMMARY OF COMMON-MODE VOLTAGE VERSUS STRAY CURRENT
In grounded mode, either a solid, low- or high-resistance RELATIONSHIPS WITH DIFFERENT GROUNDING METHODS

grounding can be applied [45]. Solid grounding is rarely used


in modern systems because of the pronounced stray-current- System Grounding Method Common-Mode Voltage Stray Current Level
induced corrosion [167]. There are a number of options re- Ungrounded System High Low
garding selection of the point in the system to be grounded; it Solidly Grounded Low High
can be either a positive pole, midpoint, or negative pole of the Diode Grounded Moderate/Low Moderate/High
Thyristor Grounded Moderate/High Moderate/Low
common dc link [164]. For that matter, Jacobson and Walker
[176] suggest to use midpoint grounding in order to minimize
the ground potential and enhance safety. Also, the circulating
current can be greatly reduced in this way in relation to ac-side stray-current-induced corrosion, but to automatically ground the
neutral grounded system. IEC 60364 standard [177] defines system upon sensing dangerously high common-mode voltage.
three grounding strategies which give specifications on how the After the clearance of abnormal operating condition is achieved,
source bus and conductive parts of the system are connected the system is automatically switched back to ungrounded mode.
to earth. The respective methods are denoted by double letter Block diagrams of two examples of reconfigurable grounding
abbreviations, i.e., TN, IT, and TT. Here, first letter reveals the systems, i.e., diode and thyristor grounding, are depicted in
type of connection of source bus to the ground, whereas the sec- Fig. 12.
ond one gives the conductive part type of ground connection. T, Diode grounding, as shown in Fig. 12(a), is comprised of a
N, and I indicate direct connection to earth, direct connection to direct metallic connection of the negative portion of dc bus to
neutral, and isolation, respectively. the ground, but via a diode circuit. In that way, current is allow to
In the case of TN, power line and exposed conducting parts flow from the ground toward the negative bus when the voltage
are commonly earthed through respective midpoints, while in across diode exceeds some threshold value. Once it happens, it
IT, the power line is not earthed while conducting parts are gets dissipated. Since corrosion is not completely eliminated in
commonly earthed either on positive or negative pole. On the these kinds of systems, they require a periodical maintenance
other hand, there are multiple grounding points in TT systems. In [178].
general, it can be said that TN systems require low-grounding Contrary to the diode scheme, bidirectional thyristor scheme
impedance and that it is easy to detect a line fault when this offers active control over the grounding instances [see
grounding strategy is applied. However, the touch voltage may Fig. 12(b)]. This is achieved by triggering the thyristor gate only
get unacceptably high depending on the voltage level at the when the negative-to-ground voltage exceeds a preset value.
time when fault occurs. IT systems do not need low-grounding Moreover, a current sensor is installed in series with the ground-
impedance, and have very low fault current if there is a single ing conductor. If the decay of current is observed, it means that
fault. However, another ground fault is likely to induce much the system can be returned to ungrounded mode. On the con-
higher current [82]. Moreover, it is complex to detect the line trary, if the current would maintain its amplitude, then a positive
fault. It should be noted that positive pole is grounded by IT to ground fault is the most likely reason. In that case, dc feeder
method in telecom systems in order to minimize the corrosion breakers are tripped in that case. The status of the system can be
effect [45]. Installation of grounding system is straightforward communicated with environment through the alarm signal. The
in case of TT grounding, and the faults do not migrate [82]. basic advantage of the thyristor grounding strategy is that the
Nevertheless, the high-voltage stress problem still exists and system maintains operation in ungrounded mode in every other
circulating current paths are likely to occur. instance, keeping the stray current minimized [167].
Unlike methods presented above, Paul [167] suggests several Table IV compares the characteristics of previously described
reconfigurable grounding options with a possibility to operate diode and thyristor grounding schemes with respect to their
the system normally in ungrounded mode in order to minimize ungrounded and solidly grounded counterparts [167].
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3541

C. Protective Devices
Protection devices that are presently commercially available
for dc systems include fuses and circuit breakers (CBs) [45].
However, they inherently introduce large time constants and
time delays before activation, respectively. In addition, inter-
ruption of the current in both cases is accompanied by the ap-
pearance of the arc. While arc gets extinguished naturally in ac
systems within the half cycle after tripping by first crossing of
the current through zero, it presents a challenge in dc systems
since the current has a steady value. Arc occurrence presents
a dangerous condition not only from the safety point of view,
but also causes contact erosion in CBs and consequently a short
lifetime and high maintenance costs [75], [179].
Fuses operate on a principle of melting down the fuse link in
a heat absorbing material. Their ratings are specified in terms
of RMS values of voltages and currents and are, hence, equally
applicable to ac and dc systems. However, it needs to be ensured
that the time constant of current increase during the fault is below
certain limits since slow increase of temperature may prevent
the heat-absorbing material to extinguish the arc [180].
Standard molded-case circuit breakers (MCCB) that consist
of a contactor, a quenching chamber, and a tripping device can be
used as well. They can have either a thermal–magnetic or elec-
tronic tripping device. In either case, sufficient voltage blocking
capability can be achieved by connecting contactors in series
[45]. One potential problem with MCCBs in systems domi-
nated by power electronic converters, as is the typical case in
dc MGs, is that short-circuit currents are supplied mainly by fil-
ter capacitors. Their quick discharge following the fault causes Fig. 13. Fault current limiting and interrupting device [182]. (a) Topology. (b)
currents of very high magnitude, but which last for a very short Fault current waveform.
periods of time. Therefore, the force generated by these currents
may be insufficient for opening the contacts of MCCB [75]. Exploiting the capability of fast switching, this dedicated
Due to inherent tripping time delays and other problems as- power dissipation element can be readily used for current lim-
sociated with fuses and CBs, protections utilizing only these iting as well. In that sense, SSCBs can actively regulate the
devices have inherent performance limitations. In order to cope fault current to a selected level, rather than leaving it dependent
with that, usage of power electronics-based protection devices on characteristics of supplying sources. Limited fault current
has been proposed. They are commonly referred to as the provides less thermal and magnetic stress on the surrounding
solid-state circuit breakers (SSCBs) [181], [182]. Several semi- components resulting in a lower installation cost, as well as
conductor devices, such as gate turn-off thyristors (GTOs), safer operation of the system in general [179], [186], [187].
insulated-gate bipolar transistors (IGBTs), and insulated-gate Moreover, current limiting circuits installed at strategic places,
commutated thyristors (IGCTs), have been used in this context, e.g., radial line supplying a high power density load center, al-
with each one of them having some specific advantages and low usage of commercially available protection devices with
drawbacks. For that matter, Meyer et al. [181] summarizes and low power rating in the entire downstream network. It should be
compares a number of different switch topologies under tech- noted that current limitation functionality can also be achieved
nical and economic aspects. It is demonstrated that GTOs and by superconducting elements [188]. However, they need to be
IGCTs have much lower onlosses than IGBTs. Also, it is esti- actively cooled and the current limiting capacity with present
mated that IGCT in rectifier configuration has by far the highest technology is inferior to SSCB solutions [187].
reliability. A typical example of IGBT-based SSCB with current limiting
In order to be able to interrupt the current, there needs to exist capability is depicted in Fig. 13(a) [182]. When the semicon-
an element to which SSCB dissipates energy during interrup- ductor device (IGBT) switches OFF, the current flowing toward
tion. Typical elements installed for that matter are metal–oxide the fault location decays since the power gets dissipated in par-
varistors (MOVs), capacitors, and switched resistors or their allel combination of a snubber circuit and MOV. On the other
combinations [182]–[184]. This brings a different perspective hand, when the IGBT turns on, the current raises up again. The
of current interruption when compared to conventional breakers hysteresis-based controllers with preselected upper and lower
where the electric power during tripping the device is mani- thresholds for the fault current are the most commonly deployed
fested in an arc that needs to be extinguished using an arc chute to achieve current limiting. An idealized waveform of a current
[183], [185]. propagation using this kind of controller is shown in Fig. 13(b).
3542 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

D. Design of Protection Systems V. STANDARDS AND REGULATION ISSUES ASSOCIATED


WITH DC MGS
The objective of protection system is to detect, locate, and
isolate any fault so that the dc MG can continue operating its Although dc MG has been intensively studied in recent years,
uncorrupted part. In order to achieve this goal, knowledge about it is still an emerging technology which needs to face number of
possible faults should be incorporated to select the most appro- challenges before widespread acceptance in industrial and com-
priate protection devices and grounding methods into a complete mercial applications. Apart from protection issues discussed in
protection system. More specifically, desirable characteristics the previous section, standardization is another barrier to pro-
of such a system are reliability, selectivity, speed, performance, liferation of dc MGs. In order to promote dc MG, several orga-
economics, and simplicity [75]. In the remainder of this section, nizations dedicated themselves to develop practical standards.
several prominent approaches that try to meet these features are Here, the existing standards and those under development are
summarized. reviewed and discussed.
An insightful example of selecting appropriate protection de- European standard ETSI EN 300 132-3-1 discusses the low-
vices for different components of a dc MG was presented in voltage dc systems [190]. It is mainly designed for data/telecom
[45]. According to specific fault withstanding capability of each equipment with voltage levels of up to 400 V. In normal
associated unit, it was concluded that converter protection must condition, considering that long power cables may be used,
be very fast in order to limit the fault current through IGBT the reference voltage at the input terminal of the equipment is
diodes, while battery protection can allow slower reaction time allowed to vary within specified bounds, i.e., 365 ± 15 V. In
as it can withstand considerably higher fault current without abnormal condition, voltage variations, voltage dips, short in-
damage. For that matter, ultrafast hybrid CB was proposed for terruptions, and voltage surges are taken into account. For a
converter protection and regular CBs for battery protection. On certain type of fault, different time durations are specified for
the other hand, fuses and MCCBs were used for feeder protec- fault clearance. Meanwhile, for protection purpose, maximum
tion. In particular, it was claimed that it is better to use MCCBs steady-state current and inrush current are considered. In order
closer to the loads since they can simultaneously trip both poles, to quantify the current limits, pulse time duration T50 is defined,
while fuses are more appropriate for installation closer to the bus which indicates the time duration when the pulse current ex-
due to their magnetic sensing that gives good selectivity. DC- ceeds 50% of the peak value. By considering different values of
link voltage information together with fault current was used for T50 , the limits of pulse current are identified and shown by the
accurate fault detection. ratio between peak pulse current and nominal terminal current.
A handshake method was proposed in [77] to split the dc sys- Besides the aspects mentioned above, grounding and electro-
tem into three possible zones and isolate the possible fault on magnetic compatibility (EMC) requirements are also indicated
any of the buses. The drawback of this method is that it needs in this standard.
to completely shut down the system following the fault, which EMerge Alliance is an open industry association founded re-
may be unnecessary in most conditions. Alternative method that cently to promote dc MGs in energy efficient buildings. The
avoids the shutdown was suggested in [82], where a loop-type vision of EMerge Alliance is to deploy advanced architectures
dc bus was used together with protection that is able to detect and control systems based on dc MGs in occupied spaces,
the fault and separate only the faulted part, so that the rest of data/telecom centers, building exteriors, and building services
the system keeps operating. The method was realized by using [191]. Except of promoting usage of dc MGs in industrial appli-
communication system where one master unit receives current cations, EMerge Alliance is also working on dc MG standards.
measurements from two slave units and gives commands to Although still under development, the preliminary standards for
control the bus switches depending on the difference between occupied spaces and data/telecom centers have been released
the two measured currents. A protection scheme that relies on [191]. For occupied spaces, the available power is harvested
solid-state converter with fast-acting current-limiting capabil- from conventional ac grid and optional onsite generation such
ity was proposed in [189]. The respective converter uses two as PV. The main power consumers in the occupied spaces are
emitter turn-off devices in antiparallel configuration to real- defined as air conditioners, lightning, wireless internet devices,
ize both switching and protection. Furthermore, a relay-based etc. The dc buses are laid in the ceiling to form the power
coordination and backup protection schemes were addressed. grid, and daylight sensors are optimally located to rationally
Coordination was realized in a fully decentralized fashion, by adjust the lightning system for energy saving purpose. Apart
using appropriate devices such as fuses and diodes. from occupied spaces, EMerge Alliance also works on stan-
In [78], consideration of the dynamic characteristics of fault dardization of data/telecom centers [191]. Since data/telecom
impedances is addressed in detail, giving a motivation to employ centers as information hubs play an important role in mod-
a unit protection of dc MGs. For that matter, a flexible design ern society, it is necessary to improve their performance and
framework for unit protection of dc MGs, which achieves high efficiency. DC MGs can significantly shorten the energy con-
sensitivity while taking into account a total cost of the system by version chain from the input side of ac grid to dc end-user
using commercial of the shelf technologies, is proposed. More- and thereby reduce the system power loss. Hence, they be-
over, practical limitations of communication technology have come the promising candidates for modern data/telecom cen-
been discussed, and it has been shown that better discrimination ters. AC power is rectified to 380 V dc power to supply vari-
of faults can be achieved by proposed method. ous components in data/telecom center, e.g., racks and servers,
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3543

while dc UPS is deployed to guarantee reliable power genera- tions and scopes of the standards mentioned in this section [193],
tion. [195], [197].
IEEE Standard 946 provides the recommended practice for dc Finally, it should be noted that the IEEE Standard 1547 can
auxiliary power systems [192]. Rather than being customized for also be used to normalize the operation of dc MGs. Although
dc MG, it was originally designed for battery-based dc auxiliary it is designed for ac distribution system with RESs, some of
power systems for generation stations [193]. However, some of the concepts in it can be employed in dc systems and thereby
the content of this standard also applies to dc MGs nowadays. used for expansion into standards for dc MGs. This standard
It normalizes the design and operation of lead acid batteries in covers different operation modes of MGs, i.e., grid-connected
dc power systems. Sizing of batteries, determination of battery and islanding operation modes. Meanwhile, requirements for
duty cycles, maintenance, testing, and replacement are included. both normal and abnormal conditions are discussed. Section 4
Meanwhile, technical specifications of battery chargers are also in the given standard introduces the interconnection technical
discussed. For example, in order to ensure safe operation of the specifications and requirements, and is being regarded as the
battery, the output voltage ripple is limited to 2% of the nominal fundamental standardization guideline for MGs [197], [198]. It
value without additional filtering. Typical system architecture normalizes not only the operation in normal condition, e.g., volt-
of a dc distribution system is also shown in this standard, and age regulation and synchronization, but also the responses when
the detailed voltage rating of each component is listed. distributed sources encounter faults. Detailed requirements of
IEC SG4 has established a dedicated technical committee in acceptable voltage and frequency ranges are given to quantify
charge of delivering standards for LVDC distribution systems the response of protective devices. Power quality issues are also
of up to 1500 V [194]. The objective is to coordinate the stan- within the scope, including dc injection, harmonics, etc.
dardization of different areas where LVDC distribution systems Among the standardization topics from different organiza-
are used, e.g., data centers, commercial buildings, electricity tions, a common concern is the general consensus about the
storage for mobile products, EVs, etc. IEC SG4 also focuses on optimal voltage level for dc MG. At present, the most likely op-
market needs to effectively promote LVDC systems into prac- tion seems to be 380 V as the rated voltage level, especially for
tical applications. By considering these requirements, energy household and data/telecom center applications. On the other
efficiency, reduction of EMC issues, and reduced usage of nat- hand, 48- or 24-V dc voltages can be commonly selected as
ural resources are taken into account. Meanwhile, measuring auxiliary dc voltage levels for low-voltage rating appliances.
methods, protection, and grounding issues are discussed. Fi- For instance, in dc nanogrid proposed by the Virginia Polytech-
nally, a dc-powered green home is proposed where PV or wind nic University [18], hybrid voltage levels of 380 and 48 V coexist
power are used as the energy input, and the heating ventilation to meet the requirements of different appliances in the house.
air conditioning system is energized by dc power. However, different voltage levels have also been selected. For
Recently formed IEEE DC@Home committee sponsored by example, a 1500-V dc MG is built up in Finland [199], while a
the IEEE PES Intelligent Grid Coordinating Committee has bipolar dc MG proposed in Japan, uses ±170 V [20].
also gathered some of the leading experts in the dc MG area to
cooperate in developing the standards and defining the roadmap
for research and commercialization of dc MGs [195]. The goal VI. CONCLUSION AND FUTURE TRENDS
of DC@Home committee is to 1) create a business case for dc This paper has summarized the state of the art in hardware
by determining the actual energy losses; 2) identify research topologies and applications of dc MGs, together with practical
required to advance the state of the art; 3) make preliminary concerns regarding their wide-scale deployment. Several types
recommendations on how dc would be delivered into the homes; of prominent hardware topologies are presented before continu-
and 4) determine the standards that should be followed. ing with assessment of their suitability for practical applications,
REbus is an open standard that proposes the normalized oper- such as dc households, EV charging stations, hybrid ESSs, and
ation of dc MG for both residential and commercial applications RES parks. Regarding the hardware topologies, the common in-
[85]. REbus dc MG operates alongside existing ac utility grid, dustrial approach based on direct connection of battery stack to
and coordinates the operation of hybrid onsite renewable en- the common dc bus can be replaced by more flexible structures
ergy generation. Communication layers are included to enable which include regulated bus.
system level energy management and monitor the status of each As an extension to singular bus structure, redundant bus
unit in dc MG. Based on REbus standard, the nominal dc bus topologies have been proposed as attractive solutions in case
voltage is 380 V. However, for decentralized coordination of when highly reliable distribution is required. More recently, SST
purposes of primary source units, loads and ESSs, this voltage has been recognized as a key enabling technology not only for
level can be changed within the specified range. dc MGs, but also for future dc distribution systems in general.
Apart from the aforementioned standards, some other works However, more development related to this technology needs to
can be also used for standardizing the operation of dc MG. For be done before it enters the SG.
example, National Electrical Code (NEC) includes articles and Practical performance requirements of four typical dc MG
clauses introducing the utilization of dc technology [193]. For applications, i.e., efficient households, renewable energy parks,
instance, military standard MIL-STD-1399 includes sections hybrid ESSs, and fast dc EV charging stations, have been
defining the requirements of dc equipment in shipboard power discussed. In that sense, a critical opinion about the usefulness
systems [193], [196]. Table V summarizes the general descrip- of functionalities such as redundancy, flexibility, scalability,
3544 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

TABLE V
GENERAL DESCRIPTIONS AND SCOPES OF DC STANDARDS

Standard Description Scope

1 ETSI EN 300 132-3-1 Power supply interface at the input to data/telecom equipment • Designed for data/telecom equipment
Subpart 1 of Part 3: Direct current source up to 400 V
• For voltage levels of up to 400 V
• Consider voltage level in normal operation and requirements
for different types of abnormal operation
• Limit of fault current
• Grounding and EMC
2 EMerge Alliance dc MG Standards for occupied spaces and data center • Recommended architecture and control system in dc MGs
3 IEEE 946 Recommended practice for the design of dc auxiliary power • Recommended practice of lead-acid-battery-based dc auxiliary
systems for generating stations power systems for generation stations
• Normalize the operation regarding batteries, e.g., sizing,
determination of duty cycle, maintenance, etc.
• Technical specifications of battery chargers
• Typical architecture of dc systems
• Voltage ratings of dc equipment

4 IEC SG4 LVDC distribution system up to 1500V • Coordinate the standardization of different areas, e.g., data
centers, commercial buildings, etc.
• Energy efficiency, EMC, reduction of natural resources
• 100% dc or hybrid ac and dc architecture
• Life cycle of equipment, protection and grounding
5 IEEE DC@Home DC-powered house • Standard and roadmap of applying LVDC MG in residential
houses
• Evaluation of losses using dc power at home
6 REbus Open standard for clean power distribution relying on dc • Operate alongside with existing ac system
• Hybrid coordination of on-site renewable energy generation
• 380 V dc common bus with acceptable variation based on the
status of source, load and energy storage
7 NEC Legal codes including introduction of dc technology • Article 393, 625, 690, 692 introduce dc technology
8 MIL-STD-1399 Electrical interface characteristics for shipboard equipment • Includes sections defining the requirements of dc equipment in
shipboard power systems
9 IEEE 1547 Requirements for interconnecting distributed resources with • Designed for ac system and some of its contents can be used as
electric power systems a reference when designing the standards for dc systems
• Islanding and grid-connected operation modes
• Normal and abnormal operation
• Requirements and practice for distributed sources

coordinated control, and grid support capability has been given. Given the increasing penetration of RESs and electronically
Moreover, elaboration about possibility to achieve the proposed regulated loads in modern distribution networks, it is foreseeable
performance specification within the framework of available that dc MGs will continue to be rapidly developed in the follow-
architectures and control strategies has been presented together ing years. Referencing to the topics covered in this manuscript,
with several examples for each application. several key lines of future research can be pointed out. First of
Finally, the state of the art in protection and standardization all, a lot of attention will need to be devoted to still unresolved
for dc MGs has been presented. Regarding protection, basic practical issues associated with dc systems, i.e., protection and
protection principles and grounding methods have been ex- grounding schemes. For that matter, design of fast acting dc
plained within the framework of available devices. This has been breakers and grounding devices based on solid-state technology
followed by a critical evaluation of several newly proposed pro- with current limiting capability will continue to be a very impor-
tection systems. Regarding regulation issues, it has been pointed tant research topic. As is customary, objective will be to achieve
out that the key prerequisite for true large-scale integration of extended functionalities with more compact design and higher
low-voltage dc distribution networks into the SG is adoption of reliability. Moreover, coordinated integration of these kinds of
corresponding standards. In that sense, IEEE 1547 is the suitable devices into overall protection systems is envisioned to be a
standard for ac MGs, but only parts of it can be used for dc MGs. fertile ground for researchers in the nearest future. On the other
Therefore, dedicated standards for normalizing the operation of hand, new and improved converter topologies will be devel-
dc MGs are highly required. For that matter, standards such as oped for interfacing different types of sources or ac grids with
EMerge Alliance, IEC SG4, and IEEE DC@Home and REbus dc MGs. Finally, the applications of dc MGs will be extended
are under development. This paper has addressed the current to a large extent. In particular, penetration of medium voltage
status and future development directions for each one of them. dc distribution systems into practical industrial applications is
On the other hand, ETSI EN 300 132-3-1, IEEE 946, NEC, expected to experience rapid growth in near future, both for
and MIL-STD-1399 are already established and the number of autonomous and grid-connected applications. This expansion
concepts used in them can be already directly used. will naturally lead to proposals of new hardware architectures
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3545

and alternative arrangements of the common dc bus for higher [23] T. Dragicevic, S. Sucic, J. C. Vasquez, and J. M. Guerrero, “Flywheel-
reliability and redundancy. based distributed bus signaling strategy for the public fast charging sta-
tion,” IEEE Trans. Smart Grid, vol. 5, no. 6, pp. 2825–2835, Nov. 2014.
[24] S. Bai and S. M. Lukic, “Unified Active filter and energy storage sys-
tem for an MW electric vehicle charging station,” IEEE Trans. Power
REFERENCES Electron., vol. 28, no. 12, pp. 5793–5803, Dec. 2013.
[25] S. Rivera, B. Wu, S. Kouro, V. Yaramasu, and J. Wang, “Electric vehicle
[1] W. A. Tabisz, M. Jovanovic, and F. C. Lee, “Present and future of dis- charging station using a neutral point clamped converter with bipolar DC
tributed power systems,” in Proc. Appl. Power Electron. Conf. Expo., bus,” IEEE Trans. Ind. Electron., vol. 62, no. 4, pp. 1999–2009, Apr.
1992, pp. 11–18. 2015.
[2] S. Schulz, B. H. Cho, and F. C. Lee, “Design considerations for a dis- [26] K. Clement-Nyns, E. Haesen, and J. Driesen, “The impact of charging
tributed power system,” in Proc. IEEE Annu. Power Electron. Spec. plug-in hybrid electric vehicles on a residential distribution grid,” IEEE
Conf., 1990, pp. 611–617. Trans. Power Syst., vol. 25, no. 1, pp. 371–380, Feb. 2010.
[3] B. K. Johnson, R. H. Lasseter, F. L. Alvarado, and R. Adapa, “Expandable [27] K. Kim, T. Yoon, G. Byeon, H. Jung, H. Kim, and G. Jang, “Power
multiterminal DC system based on voltage droop,” IEEE Trans. Power demand and power quality analysis of EV charging station using BESS
Del., vol. 8, no. 4, pp. 1926–1932, Oct. 1993. in MicroGrid,” in Proc. IEEE Veh. Power Propul. Conf., 2012, pp. 996–
[4] J. S. Glaser and A. F. Witulski, “Output plane analysis of load-sharing in 1001.
multiple-module converter systems,” IEEE Trans. Power Electron., vol. [28] Y. Du, X. Zhou, S. Bai, S. Lukic, and A. Huang, “Review of non-
9, no. 1, pp. 43–50, Jan. 1994. isolated bi-directional DC-DC converters for plug-in hybrid electric ve-
[5] S. Luo, Z. Ye, R.-L. Lin, and F. C. Lee, “A classification and evaluation hicle charge station application at municipal parking decks,” in Proc.
of paralleling methods for power supply modules,” in Proc. IEEE Annu. IEEE Annu. Appl. Power Electron. Conf. Expo., 2010, pp. 1145–1151.
Power Electron. Spec. Conf., 1999, vol. 2, pp. 901–908. [29] S. Jung, H. Lee, C. S. Song, J.-H. Han, W.-K. Han, and G. Jang, “Optimal
[6] I. Batarseh, K. Siri, and H. Lee, “Investigation of the output droop charac- operation plan of the online electric vehicle system through establishment
teristics of parallel-connected DC–DC converters,” in Proc. IEEE Annu. of a DC distribution system,” IEEE Trans. Power Electron., vol. 28, no.
Power Electron. Spec. Conf., 1994, pp. 1342–1351. 12, pp. 5878–5889, Dec. 2013.
[7] R. D. Middlebrook, “Input filter considerations in design and application [30] J. Cao and A. Emadi, “A new battery/ultracapacitor hybrid energy storage
of switching regulators,” in Proc. IEEE Ind. Appl. Soc. Annu. Meet., system for electric, hybrid, and plug-in hybrid electric vehicles,” IEEE
1976, pp. 366–382. Trans. Power Electron., vol. 27, no. 1, pp. 122–132, Jan. 2012.
[8] G. S. Thandi, R. Zhang, K. Xing, F. C. Lee, and D. Boroyevich, “Mod- [31] H. Zhou, T. Bhattacharya, D. Tran, T. S. T. Siew, and A. M. Kham-
eling, control and stability analysis of a PEBB based DC DPS,” IEEE badkone, “Composite energy storage system involving battery and ultra-
Trans. Power Del., vol. 14, no. 2, pp. 497–505, Apr. 1999. capacitor with dynamic energy management in microgrid appl,” IEEE
[9] C. M. Wildrick, F. C. Lee, B. H. Cho, and B. Choi, “A method of defining Trans. Power Electron., vol. 26, no. 3, pp. 923–930, Mar. 2011.
the load impedance specification for a stable distributed power system,” [32] Z. Ding, C. Yang, Z. Zhang, C. Wang, and S. Xie, “A novel soft-switching
IEEE Trans. Power Electron., vol. 10, no. 3, pp. 280–285, May 1995. multiport bidirectional DC-DC converter for hybrid energy storage sys-
[10] M. Belkhayat, R. Cooley, and A. Witulski, “Large signal stability criteria tem,” IEEE Trans. Power Electron., vol. 29, no. 4, pp. 1595–1609, Apr.
for distributed systems with constant power loads,” in Proc. IEEE Annu. 2014.
Power Electron. Spec. Conf., 1995, vol. 2, pp. 1333–1338. [33] B. Hredzak, V. G. Agelidis, and M. Jang, “A model predictive control
[11] K. M. Mistry, E. Silverman, T. Taylor, and R. Willis, “Telecommunica- system for a hybrid battery-ultracapacitor power source,” IEEE Trans.
tions power architectures: Distributed or centralized,” presented at the Power Electron., vol. 29, no. 3, pp. 1469–1479, Mar. 2014.
Int. Telecommunication Energy Conf., Florence, Italy, 1989. [34] T. Ise, M. Kita, and A. Taguchi, “A hybrid energy storage with a SMES
[12] T. M. Gruzs and J. Hall, “AC, DC or hybrid power solutions for today’s and secondary battery,” IEEE Trans. Appl. Supercond., vol. 15, no. 2,
telecommunications facilities,” in Proc. Int. Telecommun. Energy Conf., pp. 1915–1918, Jun. 2005.
2000, pp. 361–368. [35] Z. Jiang and R. A. Dougal, “A compact digitally controlled fuel
[13] F. Bodi, “DC-grade reliability for UPS in telecommunications data cen- cell/battery hybrid power source,” IEEE Trans. Ind. Electron., vol. 53,
ters,” in Proc. Int. Telecommun. Energy Conf., 2007, pp. 595–602. no. 4, pp. 1094–1104, Jun. 2006.
[14] A. Kwasinski, “Quantitative evaluation of DC microgrids availability: [36] Z. Zhang, Z. Ouyang, O. C. Thomsen, and M. A. E. Andersen, “Analysis
Effects of system architecture and converter topology design choices,” and design of a bidirectional isolated DC-DC converter for fuel cells and
IEEE Trans. Power Electron., vol. 26, no. 3, pp. 835–851, Mar. 2011. supercapacitors hybrid system,” IEEE Trans. Power Electron., vol. 27,
[15] G. AlLee and W. Tschudi, “Edison redux: 380 Vdc brings reliability no. 2, pp. 848–859, Feb. 2012.
and efficiency to sustainable data centers,” IEEE Power Energy Mag., [37] C. Meyer, M. Hoing, A. Peterson, and R. W. De Doncker, “Control and
vol. 10, no. 6, pp. 50–59, Nov./Dec. 2012. design of DC grids for offshore wind farms,” IEEE Trans. Ind. Appl.,
[16] A. Emadi, M. Ehsani, and J. M. Miller, Vehicular Electric Power System: vol. 43, no. 6, pp. 1475–1482, Nov./Dec. 2007.
Land, Sea, Air, and Space Vehicles. New York, NY, USA: Taylor & [38] J. Robinson, D. Jovcic, and G. Joos, “Analysis and design of an offshore
Francis, 2010. wind farm using a MV DC grid,” IEEE Trans. Power Del., vol. 25,
[17] A. Emadi, Y. J. Lee, and K. Rajashekara, “Power electron and motor no. 4, pp. 2164–2173, Oct. 2010.
drives in electric, hybrid electric, and plug-in hybrid electric vehicles,” [39] M. M. Amin and O. A. Mohammed, “Design and implementation of DC-
IEEE Trans. Ind. Electron., vol. 55, no. 6, pp. 2237–2245, Jun. 2008. bus system module for parallel integrated sustainable energy conversion
[18] D. Boroyevich, I. Cvetkovic, D. Dong, R. Burgos, F. Wang, and F. Lee, systems,” in Proc. IEEE Power Energy Soc. Gen. Meet., 2011, pp. 1–8.
“Future electronic power distribution systems: A contemplative view,” [40] H. Valderrama-Blavi, J. M. Bosque, F. Guinjoan, L. Marroyo, and
in Proc. Int. Conf. Opt. Electr. Electron. Equip., 2010, pp. 1369–1380. L. Martinez-Salamero, “Power adaptor device for domestic DC micro-
[19] T. Dragicevic, J. C. Vasquez, J. M. Guerrero, and D. Skrlec, “Advanced grids based on commercial MPPT inverters,” IEEE Trans. Ind. Electron.,
LVDC electrical power architectures and microgrids: A step toward a vol. 60, no. 3, pp. 1191–1203, Mar. 2013.
new generation of power distribution networks,” IEEE Electrif. Mag., [41] T.-F. Wu, J. Yang, L.-C. Lin, and Y. Chen, “Design and implementation
vol. 2, no. 1, pp. 54–65, Mar. 2014. of multiple maximum power point trackers for multi-string PV panels
[20] H. Kakigano, Y. Miura, and T. Ise, “Low-voltage bipolar-type DC micro- in DC distribution systems,” in Proc. IEEE Annu. Appl. Power Electron.
grid for super high quality distribution,” IEEE Trans. Power Electron., Conf. Expo., 2012, pp. 563–567.
vol. 25, no. 12, pp. 3066–3075, Dec. 2010. [42] R. Adda, O. Ray, S. K. Mishra, and A. Joshi, “Synchronous-reference-
[21] D. Dong, I. Cvetkovic, D. Boroyevich, W. Zhang, R. Wang, and frame-based control of switched boost inverter for standalone DC
P. Mattavelli, “Grid-interface bidirectional converter for residential DC nanogrid applications,” IEEE Trans. Power Electron., vol. 28, no. 3,
distribution systems-part one: High-density two-stage topology,” IEEE pp. 1219–1233, Mar. 2013.
Trans. Power Electron., vol. 28, no. 4, pp. 1655–1666, Apr. 2013. [43] S.-M. Chen, T.-J. Liang, and K.-R. Hu, “Design, analysis, and imple-
[22] D. Dong, F. Luo, X. Zhang, D. Boroyevich, and P. Mattavelli, “Grid- mentation of solar power optimizer for DC distribution system,” IEEE
interface bidirectional converter for residential DC distribution systems- Trans. Power Electron., vol. 28, no. 4, pp. 1764–1772, Apr. 2013.
part two: AC and DC interface design with passive components mini- [44] A. Pratt, P. Kumar, and T. V Aldridge, “Evaluation of 400V DC distri-
mization,” IEEE Trans. Power Electron., vol. 28, no. 4, pp. 1667–1679, bution in telco and data centers to improve energy efficiency,” in Proc.
Apr. 2013. Int. Telecommun. Energy Conf., 2007, pp. 32–39.
3546 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

[45] D. Salomonsson, L. Soder, and A. Sannino, “Protection of low-voltage [68] Autonomous power supply system—Končar Hybrid Box. (2014).
DC microgrids,” IEEE Trans. Power Del., vol. 24, no. 3, pp. 1045–1053, [Online]. Available: http://www.koncar-institut.com/solutions/
Jul. 2009. autonomous_hybrid_ power_supply.
[46] V. Nasirian, S. Moayedi, A. Davoudi, and F. Lewis, “Distributed cooper- [69] Schneider-Electric: Solar rural electrification for communities.
ative control of DC microgrids,” IEEE Trans. Power Electron., vol. 30, (2015). [Online]. Available: http://solar.schneider-electric.com/solution/
no. 4, pp. 2288–2303, Apr. 2015. residential-community-electrification/.
[47] C. Wang, X. Li, L. Guo, and Y. W. Li, “A nonlinear-disturbance-observer- [70] T. Dragicevic, “Hierarchical control of a direct current microgrid with
based DC-bus voltage control for a hybrid AC/DC microgrid,” IEEE energy storage systems in a distributed topology,” Ph.D. Dissertation,
Trans. Power Electron., vol. 29, no. 11, pp. 6162–6177, Nov. 2014. Faculty Elect. Eng. Comput., Zagreb Univ., Zagreb, Croatia, 2013.
[48] X. Lu, J. M. Guerrero, K. Sun, and J. C. Vasquez, “An improved droop [71] S. P. Engel, N. Soltau, H. Stagge, and R. W. De Doncker, “Dynamic
control method for DC microgrids based on low bandwidth communi- and balanced control of three-phase high-power dual-active bridge DC-
cation with DC bus voltage restoration and enhanced current sharing DC converters in DC-grid applications,” IEEE Trans. Power Electron.,
accuracy,” IEEE Trans. Power Electron., vol. 29, no. 4, pp. 1800–1812, vol. 28, no. 4, pp. 1880–1889, Apr. 2013.
Apr. 2014. [72] R. W. De Doncker, “Power electronic technologies for flexible DC dis-
[49] N. L. Diaz, T. Dragicevic, J. C. Vasquez, and J. M. Guerrero, “Intelli- tribution grids,” in Proc. Int. Power Electron. Conf., 2014, pp. 736–743.
gent distributed generation and storage units for DC microgrids—A new [73] N. C. Ekneligoda and W. W. Weaver, “A game theoretic bus selection
concept on cooperative control without communications beyond droop method for loads in multibus DC power systems,” IEEE Trans. Ind.
control,” IEEE Trans. Smart Grid, vol. 5, no. 5, pp. 2476–2485, Sep. Electron., vol. 61, no. 4, pp. 1669–1678, Apr. 2014.
2014. [74] Q. Shafiee, T. Dragicevic, J. C. Vasquez, and J. M. Guerrero, “Hierar-
[50] S. Anand, B. G. Fernandes, and J. M. Guerrero, “Distributed control chical control for multiple DC-microgrids clusters,” IEEE Trans. Energy
to ensure proportional load sharing and improve voltage regulation in Convers., vol. 29, no. 4, pp. 922–933, Dec. 2014.
low-voltage DC microgrids,” IEEE Trans. Power Electron., vol. 28, no. [75] R. M. Cuzner and G. Venkataramanan, “The status of DC micro-
4, pp. 1900–1913, Apr. 2013. grid protection,” in Proc. IEEE Ind. Appl. Soc. Annu. Meet., 2008,
[51] X. She, A. Q. Huang, S. Lukic, and M. E. Baran, “On integration of pp. 1–8.
solid-state transformer with zonal DC microgrid,” IEEE Trans. Smart [76] M. Saeedifard, M. Graovac, R. F. Dias, and R. Iravani, “DC power
Grid, vol. 3, no. 2, pp. 975–985, Jun. 2012. systems: Challenges and opportunities,” in Proc. IEEE Power Energy
[52] A. A. A. Radwan and Y. A.-R. I. Mohamed, “Linear active stabiliza- Soc. Gen. Meet., 2010, pp. 1–7.
tion of converter-dominated DC microgrids,” IEEE Trans. Smart Grid, [77] L. Tang and B.-T. Ooi, “Locating and isolating DC faults in multi-
vol. 3, no. 1, pp. 203–216, Mar. 2012. terminal DC systems,” IEEE Trans. Power Del., vol. 22, no. 3, pp. 1877–
[53] L. Meng, J. C. Vasquez, J. M. Guerrero, and T. Dragicevic, “Agent- 1884, Jul. 2007.
based distributed hierarchical control of DC microgrid systems,” in Proc. [78] S. D. A. Fletcher, P. J. Norman, S. J. Galloway, P. Crolla, and G. M. Burt,
ElectrIMACS Conf., 2014, pp. 281–286. “Optimizing the roles of unit and non-unit protection methods within DC
[54] R. S. Balog and P. T. Krein, “Bus selection in multibus DC micro- microgrids,” IEEE Trans. Smart Grid, vol. 3, no. 4, pp. 2079–2087, Dec.
grids,” IEEE Trans. Power Electron., vol. 26, no. 3, pp. 860–867, Mar. 2012.
2011. [79] K. Corzine, “DC microgrid protection with the Z-source breaker,” in
[55] T. Dragicevic, J. M. Guerrero, J. C. Vasquez, and D. Skrlec, “Supervi- Proc. IEEE Annu. Conf. Ind. Electron. Soc., 2013, pp. 2197–2204.
sory control of an adaptive-droop regulated DC microgrid with battery [80] S. R. B. Vanteddu, A. Mohamed, and O. Mohammed, “Protection design
management capability,” IEEE Trans. Power Electron., vol. 29, no. 2, and coordination of DC distributed power systems architectures,” in Proc.
pp. 695–706, Feb. 2014. IEEE Power Energy Soc. Gen. Meet., 2013, pp. 1–5.
[56] R. H. Lasseter, “MicroGrids,” in Proc. IEEE Power Eng. Soc. Winter [81] J.-D. Park, J. Candelaria, L. Ma, and K. Dunn, “DC ring-bus microgrid
Meet. Conf., 2002, vol. 1, pp. 305–308. fault protection and identification of fault location,” IEEE Trans. Power
[57] T. Ackermann, G. Andersson, and L. Söder, “Distributed generation: A Del., vol. 28, no. 4, pp. 2574–2584, Oct. 2013.
definition,” Electr. Power Syst. Res., vol. 57, no. 3, pp. 195–204, 2001. [82] J.-D. Park and J. Candelaria, “Fault detection and isolation in low-voltage
[58] J.-F. Chen and C.-L. Chu, “Combination voltage-controlled and current- DC-bus microgrid system,” IEEE Trans. Power Del., vol. 28, no. 2,
controlled PWM inverters for UPS parallel operation,” IEEE Trans. pp. 779–787, Apr. 2013.
Power Electron., vol. 10, no. 5, pp. 547–558, Sep. 1995. [83] S. D. A. Fletcher, P. J. Norman, K. Fong, S. J. Galloway, and G. M. Burt,
[59] E. A. A. Coelho, P. C. Cortizo, and P. F. D. Garcia, “Small-signal stability “High-speed differential protection for smart DC distribution systems,”
for parallel-connected inverters in standalone AC supply system,” IEEE IEEE Trans. Smart Grid, vol. 5, no. 5, pp. 2610–2617, Sep. 2014.
Trans. Ind. Appl., vol. 38, no. 2, pp. 533–542, Mar./Apr. 2002. [84] A. A. S. Emhemed and G. M. Burt, “An advanced protection scheme for
[60] J. M. Guerrero, L. Hang, and J. Uceda, “Control of distributed unin- enabling an LVDC last mile distribution network,” IEEE Trans. Smart
terruptible power supply system,” IEEE Trans. Ind. Electron., vol. 55, Grid, vol. 5, no. 5, pp. 2602–2609, Sep. 2014.
no. 8, pp. 2845–2859, Aug. 2008. [85] REbus DC microgrid. (2013). [Online]. Available: http://rebuspower.com
[61] K. De Brabandere, B. Bolsens, J. den Keybus, A. Woyte, J. Driesen, and /technical.shtml
R. Belmans, “A voltage and frequency droop control method for parallel [86] Arcola energy DC microgrid. (2015). [Online]. Available: http://
inverters,” IEEE Trans. Power Electron., vol. 22, no. 4, pp. 1107–1115, arcolaenergy.com/pages/dc-microgrid
Jul. 2007. [87] Elecyr DC microgrids. (2008). [Online]. Available: http://www.hybrid
[62] J. M. Guerrero, L. G. De Vicuña, J. Matas, M. Castilla, and J. Miret, “A propulsion.com/
wireless controller to enhance dynamic performance of parallel inverters [88] SPIDERS: The smart power infrastructure demonstration for en-
in distributed generation systems,” IEEE Trans. Power Electron, vol. 19, ergy reliability and security. (2012). [Online]. Available: http://
no. 5, pp. 1205–1213, Sep. 2004. energy.sandia.gov/wp/ wp-content/gallery/uploads/SPIDERS_Fact_
[63] Y. Ito, Z. Yang, and H. Akagi, “DC microgrid based distribution power Sheet_2012-1431P.pdf
generation system,” in Proc. Power Electron. Motion Control Conf., [89] F. Valenciaga and P. F. Puleston, “Supervisor control for a stand-
2004, vol. 3, pp. 1740–1745. alone hybrid generation system using wind and photovoltaic en-
[64] A. Kwasinski and P. T. Krein, “A microgrid-based telecom power system ergy,” IEEE Trans. Energy Convers., vol. 20, no. 2, pp. 398–405,
using modular multiple-input DC-DC converters,” in Proc. Int. Telecom- Jun. 2005.
mun. Conf., 2005, pp. 515–520. [90] A. M. Knight and G. E. Peters, “Simple wind energy controller for an
[65] H. Kakigano, Y. Miura, T. Ise, and R. Uchida, “DC Micro-grid for su- expanded operating range,” IEEE Trans. Energy Convers., vol. 20, no. 2,
per high quality distribution—System configuration and control of dis- pp. 459–466, Jun. 2005.
tributed generations and energy storage devices,” in Proc. IEEE Power [91] F. Valenciaga and P. F. Puleston, “High-order sliding control for a wind
Electron. Spec. Conf., 2006, pp. 1–7. energy convers. System based on a permanent magnet synchronous gen-
[66] D. Salomonsson, L. Soder, and A. Sannino, “An adaptive control system erator,” IEEE Trans. Energy Convers., vol. 23, no. 3, pp. 860–867, Sep.
for a DC microgrid for data centers,” IEEE Trans. Ind. Appl., vol. 44, 2008.
no. 6, pp. 1910–1917, Nov./Dec. 2008. [92] S. J. Chiang, H.-J. Shieh, and M.-C. Chen, “Modeling and control of
[67] P. Biczel, “Power electronic converters in DC microgrid,” in Proc. Com- PV charger system with SEPIC converter,” IEEE Trans. Ind. Electron.,
pat. Power Electron., 2007, pp. 1–6. vol. 56, no. 11, pp. 4344–4353, Nov. 2009.
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3547

[93] P. Lindman and L. Thorsell, “Applying distributed power modules in [115] P. C. Loh, D. Li, Y. K. Chai, and F. Blaabjerg, “Autonomous operation
telecom systems,” IEEE Trans. Power Electron., vol. 11, no. 2, pp. 365– of hybrid microgrid with AC and DC subgrids,” IEEE Trans. Power
373, Mar. 1996. Electron., vol. 28, no. 5, pp. 2214–2223, May 2013.
[94] “TS-MPPT-60, Solar System Controller,” 2013. [116] X. Lu, J. M. Guerrero, K. Sun, J. C. Vasquez, R. Teodorescu, and
[95] J.-W. Kim, H.-S. Choi, and B. H. Cho, “A novel droop method for L. Huang, “Hierarchical control of parallel AC-DC converter interfaces
converter parallel operation,” IEEE Trans. Power Electron., vol. 17, for hybrid microgrids,” IEEE Trans. Smart Grid, vol. 5, no. 2, pp. 683–
no. 1, pp. 25–32, Jan. 2002. 692, Feb. 2014.
[96] R. Ayyanar, R. Giri, and N. Mohan, “Active input-voltage and load- [117] Z. Jiang and X. Yu, “Hybrid DC- and AC-linked microgrids: Towards
current sharing in input-series and output-parallel connected modu- integration of distributed energy resources,” in Proc. IEEE Energy 2030
lar DC-DC converters using dynamic input-voltage reference scheme,” Conf., 2008, pp. 1–8.
IEEE Trans. Power Electron., vol. 19, no. 6, pp. 1462–1473, Nov. [118] A. A. A. Radwan and Y. A. I. Mohamed, “Assessment and mitiga-
2004. tion of interaction dynamics in hybrid AC/DC distribution generation
[97] D. Sha, Z. Guo, and X. Liao, “Cross-feedback output-current-sharing systems,” IEEE Trans. Smart Grid, vol. 3, no. 3, pp. 1382–1393, Sep.
control for input-series-output-parallel modular DC-DC converters,” 2012.
IEEE Trans. Power Electron., vol. 25, no. 11, pp. 2762–2771, Nov. [119] N. Flourentzou, V. G. Agelidis, and G. D. Demetriades, “VSC-based
2010. HVDC power transmission systems: An overview,” IEEE Trans. Power
[98] P. Liutanakul, A.-B. Awan, S. Pierfederici, B. Nahid-Mobarakeh, and Electron., vol. 24, no. 3, pp. 592–602, Mar. 2009.
F. Meibody-Tabar, “Linear stabilization of a DC bus supplying a constant [120] M. Bragard, N. Soltau, S. Thomas, and R. W. De Doncker, “The balance
power load: A general design approach,” IEEE Trans. Power Electron., of renewable sources and user demands in grids: power electronics for
vol. 25, no. 2, pp. 475–488, Feb. 2010. modular battery energy storage systems,” IEEE Trans. Power Electron.,
[99] A. Kwasinski and C. N. Onwuchekwa, “Dynamic behavior and stabiliza- vol. 25, no. 12, pp. 3049–3056, Dec. 2010.
tion of DC microgrids with instantaneous constant-power loads,” IEEE [121] N. Hatziargyriou, H. Asano, R. Iravani, and C. Marnay, “Microgrids,”
Trans. Power Electron., vol. 26, no. 3, pp. 822–834, Mar. 2011. IEEE Power Energy Mag., vol. 5, no. 4, pp. 78–94, Jul./Aug. 2007.
[100] P. Karlsson and J. Svensson, “DC bus voltage control for a distributed [122] G. Venkataramanan and C. Marnay, “A larger role for microgrids,” IEEE
power system,” IEEE Trans. Power Electron., vol. 18, no. 6, pp. 1405– Power Energy Mag., vol. 6, no. 3, pp. 78–82, May/Jun. 2008.
1412, Nov. 2003. [123] M. Cespedes, L. Xing, and J. Sun, “Constant-power load system stabi-
[101] J. Schonberger, R. Duke, and S. D. Round, “DC-bus signaling: A dis- lization by passive damping,” IEEE Trans. Power Electron., vol. 26, no.
tributed control strategy for a hybrid renewable nanogrid,” IEEE Trans. 7, pp. 1832–1836, Jul. 2011.
Ind. Electron., vol. 53, no. 5, pp. 1453–1460, Oct. 2006. [124] A. M. Rahimi and A. Emadi, “Active damping in DC/DC power elec-
[102] K. Sun, L. Zhang, Y. Xing, and J. M. Guerrero, “A distributed control tronic converters: A novel method to overcome the problems of constant
strategy based on DC bus signaling for modular photovoltaic genera- power loads,” IEEE Trans. Ind. Electron., vol. 56, no. 5, pp. 1428–1439,
tion systems with battery energy storage,” IEEE Trans. Power Electron., May 2009.
vol. 26, no. 10, pp. 3032–3045, Oct. 2011. [125] O. Garcia, P. Zumel, A. de Castro, and J. A. Cobos, “Automotive DC-DC
[103] T. Dragicevic, J. M. Guerrero, and J. C. Vasquez, “A distributed con- bidirectional converter made with many interleaved buck stages,” IEEE
trol strategy for coordination of an autonomous lvdc microgrid based Trans. Power Electron., vol. 21, no. 3, pp. 578–586, May 2006.
on power-line signaling,” IEEE Trans. Ind. Electron., vol. 61, no. 7, [126] J. Johnston, J. Counsell, G. Banks, and M. J. Stewart, “Beyond power
pp. 3313–3326, Jan. 2014. over Ethernet: The development of digital energy networks for buildings,”
[104] Q. Shafiee, T. Dragicevic, J. C. Vasquez, and J. M. Guerrero, “Model- presented at the CIBSE Technical Symp., London, U.K., 2012.
ing, stability analysis and active stabilization of multiple DC microgrid [127] M.-H. Ryu, H.-S. Kim, J.-H. Kim, J.-W. Baek, and J.-H. Jung, “Test bed
clusters,” in Proc. IEEE Int. Energy Conf., 2014, pp. 1284–1290. implementation of 380V DC distribution system using isolated bidirec-
[105] X. She, A. Q. Huang, and R. Burgos, “Review of solid-state transformer tional power converters,” in Proc. IEEE Energy Convers. Congr. Expo.,
technologies and their application in power distribution systems,” IEEE 2013, pp. 2948–2954.
J. Emerg. Sel. Topics Power Electron., vol. 1, no. 3, pp. 186–198, Sep. [128] Intelligent DC microgrid living lab. (2014). [Online]. Avail-
2013. able: http://www.et.aau.dk/research-programmes/microgrids/activities/
[106] A. Q. Huang, M. L. Crow, G. T. Heydt, J. P. Zheng, and S. J. Dale, “The intelligent-dc-microgrid-living-lab/
future renewable electric energy delivery and management (FREEDM) [129] D. J. Becker and B. J. Sonnenberg, “DC microgrids in buildings and
system: The energy internet,” Proc. IEEE, vol. 99, no. 1, pp. 133–148, data centers,” in Proc. IEEE Int. Telecommun. Energy Conf., 2011,
Jan. 2011. pp. 1–7.
[107] S. Bifaretti, P. Zanchetta, A. Watson, L. Tarisciotti, and J. C. Clare, [130] Smart grid strategy: The intelligent energy system of the future, Danish
“Advanced power electronic conversion and control system for universal ministry of climate, energy building. (2013). [Online]. Available:
and flexible power management,” IEEE Trans. Smart Grid, vol. 2, no. 2, http://www.ens.dk/sites/ens.dk/files/undergrund-forsyning/el-naturgas-
pp. 231–243, Jun. 2011. varmeforsyning/Smart-grid/smartgridstrategy_uk_web.pdf
[108] M. Liserre. HEART—The highly efficient and reliable smart trans- [131] (2009). Smart grid: An introduction, the U.S. department of en-
former. (2015). [Online]. Available: http://www.pe.tf.uni-kiel.de/en/ ergy. [Online]. Available: http://energy.gov/oe/downloads/smart-grid-
staff/professors/prof.-dr.-ing.-marco-liserre/presentation_sonderborg introduction-0
[109] N. R. Chaudhuri, R. Majumder, and B. Chaudhuri, “System frequency [132] A.-H. Mohsenian-Rad, V. W. S. Wong, J. Jatskevich, R. Schober, and
support through multi-terminal DC (MTDC) Grids,” IEEE Trans. Power A. Leon-Garcia, “Autonomous demand-side management based on
Syst., vol. 28, no. 1, pp. 347–356, Feb. 2013. game-theoretic energy consumption scheduling for the future smart grid,”
[110] R. T. Pinto, P. Bauer, S. F. Rodrigues, E. J. Wiggelinkhuizen, J. Pierik, and IEEE Trans. Smart Grid, vol. 1, no. 3, pp. 320–331, Dec. 2010.
B. Ferreira, “A novel distributed direct-voltage control strategy for grid [133] D. Guezgouz, D. E. Chariag, Y. Raingeaud, and J.-C. Le Bunetel, “Mod-
integration of offshore wind energy systems through MTDC Network,” eling of electromagnetic interference and PLC transmission for loads
IEEE Trans. Ind. Electron., vol. 60, no. 6, pp. 2429–2441, Jun. 2013. shedding in a microgrid,” IEEE Trans. Power Electron., vol. 26, no. 3,
[111] C. Gavriluta, I. J. Candela, C. Citro, J. Rocabert, and P. Rodriguez, pp. 747–754, Mar. 2011.
“Decentralized primary control of MTDC networks with energy stor- [134] A.-H. Mohsenian-Rad and A. Leon-Garcia, “Optimal residential load
age and distributed generation,” IEEE Trans. Ind. Appl., vol. 50, no. 6, control with price prediction in real-time electricity pricing environ-
pp. 4122–4131, Nov./Dec. 2014. ments,” IEEE Trans. Smart Grid, vol. 1, no. 2, pp. 120–133, Sep.
[112] X. Liu, P. Wang, and P. C. Loh, “A hybrid AC/DC microgrid and its 2010.
coordination control,” IEEE Trans. Smart Grid, vol. 2, no. 2, pp. 278– [135] P. Du and N. Lu, “Appliance commitment for household load schedul-
286, Jun. 2011. ing,” IEEE Trans. Smart Grid, vol. 2, no. 2, pp. 411–419, Jun. 2011.
[113] R. Majumder, “A Hybrid microgrid with DC connection at back to back [136] H. Saele and O. S. Grande, “Demand response from household cus-
converters,” IEEE Trans. Smart Grid, vol. 5, no. 1, pp. 251–259, Jan. tomers: Experiences from a pilot study in Norway,” IEEE Trans. Smart
2014. Grid, vol. 2, no. 1, pp. 102–109, Mar. 2011.
[114] Z. Ye, “Modeling and control of parallel three-phase PWM converters,” [137] X. Chen, T. Wei, and S. Hu, “Uncertainty-aware household appliance
Ph.D. dissertation, Electr. Comput. Eng. Dept., Virginia Polytechnic In- scheduling considering dynamic electricity pricing in smart home,” IEEE
stitute, Blacksburg, VA, USA, 2000. Trans. Smart Grid, vol. 4, no. 2, pp. 932–941, Jun. 2013.
3548 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 31, NO. 5, MAY 2016

[138] M. Sechilariu, B. Wang, and F. Locment, “Building-integrated microgrid: [160] P. Wang, J. Xiao, and L. Setyawan, “Hierarchical control of hybrid energy
Advanced local energy management for forthcoming smart power grid storage system in DC microgrids,” IEEE Trans. Ind. Electron., vol. 62,
communication,” Energy Build., vol. 59, pp. 236–243, 2013. no. 8, pp. 4915–4924, Aug. 2015.
[139] H. Kakigano, Y. Miura, and T. Ise, “Distribution voltage control for [161] G. F. Reed, B. M. Grainger, A. R. Sparacino, R. J. Kerestes, and
DC microgrids using fuzzy control and gain-scheduling technique,” M. J. Korytowski, “Advancements in medium voltage DC architecture
IEEE Trans. Power Electron., vol. 28, no. 5, pp. 2246–2258, May development with applications for powering electric vehicle charging
2013. stations,” in Proc. IEEE Energytech, 2012, pp. 1–8.
[140] K. Tanaka, K. Uchida, M. Oshiro, T. Goya, T. Senjyu, and A. Yona, [162] M. Vasiladiotis and A. Rufer, “A modular multiport power electronic
“Optimal operation for DC smart houses considering forecasted error,” transformer with integrated split battery energy storage for versatile ul-
in Proc. Int. Power Electron. Conf., 2010, pp. 722–727. trafast EV charging stations,” IEEE Trans. Ind. Electron., vol. 62, no. 5,
[141] H. A. B. Siddique, S. M. Ali, and R. W. De Doncker, “DC collector grid pp. 3213–3222, May 2015.
configurations for large photovoltaic parks,” in Proc. Eur. Conf. Power [163] L. Tan, B. Wu, S. Rivera, and V. Yaramasu, “Comprehensive DC power
Electron. Appl., 2013, pp. 1–10. balance management in high-power three-level DC-DC converter for
[142] S. Chuangpishit, A. Tabesh, Z. Moradi-Sharbabk, and M. Saeedifard, electric vehicle fast charging,” IEEE Trans. Power Electron., 2015, to be
“Topology design for collector systems of offshore wind farms with published.
pure DC power systems,” IEEE Trans. Ind. Electron., vol. 61, no. 1, [164] J. Yang, J. E. Fletcher, and J. O’Reilly, “Short-circuit and ground fault
pp. 320–328, Jan. 2014. analyses and location in VSC-based DC network cables,” IEEE Trans.
[143] B.-I. Craciun, T. Kerekes, D. Sera, and R. Teodorescu, “Overview of Ind. Electron., vol. 59, no. 10, pp. 3827–3837, Oct. 2012.
recent grid codes for PV power integration,” in Proc. Int. Conf. Opt. [165] B. Ram, Power System Protection and Switchgear. New York, NY, USA:
Electr. Electron. Equip., 2012, pp. 959–965. McGraw-Hill, 2011.
[144] M. Tsili and S. Papathanassiou, “A review of grid code technical re- [166] A. Berizzi, A. Silvestri, D. Zaninelli, and S. Massucco, “Short-circuit
quirements for wind farms,” IET Renew. Power Gener., vol. 3, no. 3, current calculations for DC systems,” IEEE Trans. Ind. Appl., vol. 32,
pp. 308–332, 2009. no. 5, pp. 990–997, Sep./Oct. 1996.
[145] A. Mohamed and O. Mohammed, “Real-time energy management [167] D. Paul, “DC traction power system grounding,” IEEE Trans. Ind. Appl.,
scheme for hybrid renewable energy systems in smart grid applications,” vol. 38, no. 3, pp. 818–824, Jun. 2002.
Electr. Power Syst. Res., vol. 96, pp. 133–143, 2013. [168] S. A. Memon and P. Fromme, “Stray current corrosion and mitigation:
[146] C. Chen, S. Duan, T. Cai, B. Liu, and G. Hu, “Optimal allocation and A synopsis of the technical methods used in dc transit systems.,” IEEE
economic analysis of energy storage system in microgrids,” IEEE Trans. Electrif. Mag., vol. 2, no. 3, pp. 22–31, Sep. 2014.
Power Electron., vol. 26, no. 10, pp. 2762–2773, Oct. 2011. [169] I. Cotton, C. Charalambous, P. Aylott, and P. Ernst, “Stray current control
[147] T.-F. Wu, C.-H. Chang, L.-C. Lin, and Y.-L. Lin, “Power loss analysis of in DC mass transit systems,” IEEE Trans. Veh. Technol., vol. 54, no. 2,
photovoltaic DC distributed generation systems with grid-connection,” pp. 722–730, Mar. 2005.
in Proc. IEEE Int. Conf. Power Electron., 2011, pp. 2344–2351. [170] J.-M. Shen, H.-L. Jou, and J.-C. Wu, “Novel transformerless grid-
[148] S. Lundberg, “Wind farm configuration and energy efficiency studies- connected power converter with negative grounding for photovoltaic
series DC versus AC layouts,” Ph.D. dissertation, Dept. Energy Environ., generation system,” IEEE Trans. Power Electron., vol. 27, no. 4,
Chalmers Univ. Technol.,Göteborg, Sweden, 2006. pp. 1818–1829, Apr. 2012.
[149] S. Vazquez, S. M. Lukic, E. Galvan, L. G. Franquelo, and J. M. Carrasco, [171] C. Charalambous, N. Kokkinos, and A. Demetriou, “Impact of photo-
“Energy storage system for transport and grid application,” IEEE Trans. voltaic oriented DC stray current corrosion on large scale solar farms’
Ind. Electron., vol. 57, no. 12, pp. 3881–3895, Dec. 2010. grounding and third-party infrastructure: modelling and assessment,”
[150] Y. Wang, X. Lin, Q. Xie, N. Chang, and M. Pedram, “Minimizing state- IEEE Trans. Ind. Appl., 2015, to be published.
of-health degradation in hybrid electrical energy storage systems with [172] O. Lopez, F. D. Freijedo, A. G. Yepes, P. Fernandez-Comesaa, J. Malvar,
arbitrary source and load profiles,” in Proc. Des., Autom. Test Eur. Conf. R. Teodorescu, and J. Doval-Gandoy, “Eliminating ground current in a
Exhib., 2014, pp. 1–4. transformerless photovoltaic application,” IEEE Trans. Energy Convers.,
[151] T. Dragicevic, H. Pandzic, D. Skrlec, I. Kuzle, J. Guerrero, and vol. 25, no. 1, pp. 140–147, Mar. 2010.
D. Kirschen, “Capacity optimization of renewable energy sources and [173] M. E. Baran and N. R. Mahajan, “DC distribution for industrial systems:
battery storage in an autonomous telecommunication facility,” IEEE Opportunities and challenges,” IEEE Trans. Ind. Appl., vol. 39, no. 6,
Trans. Sustain. Energy, vol. 5, no. 4, pp. 1367–1378, Oct. 2014. pp. 1596–1601, Nov./Dec. 2003.
[152] S. Lemofouet and A. Rufer, “A hybrid energy storage system based [174] J. Park, “Ground fault detection and location for ungrounded DC traction
on compressed air and supercapacitors with maximum efficiency point power systems,” IEEE Trans. Veh. Technol., 2015, to be published.
tracking (MEPT),” IEEE Trans. Ind. Electron., vol. 53, no. 4, pp. 1105– [175] M. Valdes, T. Papallo, and B. Premerlani, “Finding fault—Locating a
1115, Jun. 2006. ground fault in low-voltage, high-resistance grounded systems via the
[153] A. Ravichandran, P. Malysz, S. Sirouspour, and A. Emadi, “The critical single-processor concept for circuit protection,” IEEE Ind. Appl. Mag.,
role of microgrids in transition to a smarter grid: A technical review,” in vol. 13, no. 5, pp. 24–30, Sep. 2007.
Proc. IEEE Transp. Electrif. Conf. Expo., 2013, pp. 1–7. [176] B. Jacobson and J. Walker, “Grounding considerations for DC and mixed
[154] P. J. Tulpule, V. Marano, S. Yurkovich, and G. Rizzoni, “Economic DC and AC power systems,” Nav. Eng. J., vol. 119, no. 2, pp. 49–62,
and environmental impacts of a PV powered workplace parking garage 2007.
charging station,” Appl. Energy, vol. 108, pp. 323–332, 2013. [177] IEC 60364-1 Low-Voltage Electrical Installations—Part 1: Fundamen-
[155] H. Choi, M. Ciobotaru, M. Jang, and V. G. Agelidis, “Performance of tal Principles, Assessment of General Characteristics, Definitions, IEC
medium-voltage DC-bus PV system architecture utilizing high-gain DC- 60364-1, 2005.
DC converter,” IEEE Trans. Sustain. Energy, vol. 6, no. 2, pp. 464–473, [178] C.-H. Lee and C.-J. Lu, “Assessment of grounding schemes on rail
Apr. 2015. potential and stray currents in a DC transit system,” IEEE Trans. Power
[156] V. Yaramasu, B. Wu, P. C. Sen, S. Kouro, and M. Narimani, “High- Del., vol. 21, no. 4, pp. 1941–1947, Oct. 2006.
power wind energy conversion systems: State-of-the-art and emerg- [179] J.-M. Meyer and A. Rufer, “A DC hybrid circuit breaker with ultra-fast
ing technologies,” Proc. IEEE, vol. 103, no. 5, pp. 740–788, May contact opening and integrated gate-commutated thyristors (IGCTs),”
2015. IEEE Trans. Power Del., vol. 21, no. 2, pp. 646–651, Apr. 2006.
[157] T. Kerekes, E. Koutroulis, D. Sera, R. Teodorescu, and M. Katsanevakis, [180] IEEE Guide for the Protection of Stationary Battery Systems, IEEE Std
“An optimization method for designing large PV plants,” IEEE J. Pho- 1375-1998, 1998.
tovoltaics, vol. 3, no. 2, pp. 814–822, Apr. 2013. [181] C. Meyer, S. Schroder, and R. W. De Doncker, “Solid-state circuit break-
[158] P. Garcia, L. M. Fernandez, C. A. Garcia, and F. Jurado, “Energy man- ers and current limiters for medium-voltage systems having distributed
agement system of fuel-cell-battery hybrid tramway,” IEEE Trans. Ind. power systems,” IEEE Trans. Power Electron., vol. 19, no. 5, pp. 1333–
Electron., vol. 57, no. 12, pp. 4013–4023, Dec. 2010. 1340, Sep. 2004.
[159] X. Wang, M. Yue, E. Muljadi, and W. Gao, “Probabilistic approach [182] D. Lawes, L. Ran, and Z. Xu, “Design of a solid-state DC circuit breaker
for power capacity specification of wind energy storage systems,” IEEE for light rail transit power supply network,” in Proc. IEEE Energy Con-
Trans. Ind. Appl., vol. 50, no. 2, pp. 1215–1224, Mar. 2014. vers. Congr. Expo., 2014, pp. 350–357.
DRAGIČEVIĆ et al.: DC MICROGRIDS—PART II: A REVIEW OF POWER ARCHITECTURES, APPLICATIONS, AND STANDARDIZATION ISSUES 3549

[183] S. Krstic, E. L. Wellner, A. R. Bendre, and B. Semenov, “Circuit breaker Xiaonan Lu (S’11–M’14) was born in Tianjin,
technologies for advanced ship power systems,” in Proc. IEEE Electr. China, in 1985. He received the B.E. and Ph.D. de-
Ship Technol. Symp., 2007, pp. 201–208. grees in electrical engineering from Tsinghua Univer-
[184] Z. J. Shen, G. Sabui, Z. Miao, and Z. Shuai, “Wide-bandgap solid-state sity, Beijing, China, in 2008 and 2013, respectively.
circuit breakers for DC power systems: Device and circuit considera- From September 2010 to August 2011, he was
tions,” IEEE Trans. Electron Devices, vol. 62, no. 2, pp. 294–300, Feb. a Guest Ph.D. Student at the Department of Energy
2015. Technology, Aalborg University, Aalborg, Denmark.
[185] J. Jebramcik and F. Berger, “Observations on switching characteristics of From October 2013 to December 2014, he was a Post-
arc chutes in DC contactors,” in Proc. Int. Conf. Electr. Contacts, 2014, doctoral Researcher with the Department of Electri-
pp. 1–6. cal Engineering and Computer Science, University
[186] C. Jin and R. Dougal, “Current limiting technique based protection strat- of Tennessee, Knoxville, TN, USA. In January 2015,
egy for an industrial DC distribution system,” in Proc. IEEE Int. Symp. he joined the Energy Systems Division, Argonne National Laboratory, Lemont,
Ind. Electron., 2006, vol. 2, pp. 820–825. IL, USA, where he is currently a Postdoctoral Appointee. His research inter-
[187] F. Luo, J. Chen, X. Lin, Y. Kang, and S. Duan, “A novel solid state fault ests include modeling and control of power electronic converters in renewable
current limiter for DC power distribution network,” in Proc. IEEE Appl. energy systems and microgrids, hardware-in-the-loop real-time test, multilevel
Power Electron. Conf. Expo., 2008, pp. 1284–1289. converters, matrix converters, etc.
[188] K. Yasuda, A. Ichinose, A. Kimura, K. Inoue, H. Morii, Y. Tokunaga, Dr. Lu received the Outstanding Reviewer Award for the IEEE TRANSACTION
S. Torii, T. Yazawa, S. Hahakura, K. Shimohata, and H. Kubota, “Re- ON POWER ELECTRONICS in 2014. He is a Member of the IEEE PELS IAS, and
search development of superconducting fault current limiter in japan,” PES Societies.
IEEE Trans. Appl. Supercond., vol. 15, no. 2, pp. 1978–1981, Jun. 2005.
[189] M. E. Baran and N. R. Mahajan, “Overcurrent protection on voltage-
source-converter-based multiterminal DC distribution systems,” IEEE Juan C. Vasquez (M’12–SM’14) received the
Trans. Power Del., vol. 22, no. 1, pp. 406–412, Jan. 2007. B.S. degree in electronics engineering from the
[190] ETSI EN 300 132-3-1. (2011). [Online]. Available: http://www.etsi.org/ Autonomous University of Manizales, Manizales,
deliver/etsi_en/300100_300199/3001320301/02.01.01_40/en_3001320 Colombia, in 2004, and the Ph.D. degree in auto-
301v020101o.pdf matic control, robotics, and computer vision from the
[191] EMerge Alliance Standards. (2008). [Online]. Available: http://www Technical University of Catalonia, Barcelona, Spain,
.emergealliance.org/Standards/OurStandards.aspx in 2009.
[192] IEEE Recommended Practice for the Design of DC Auxiliary Power He was with the Autonomous University of Man-
Systems for Generating Stations, IEEE Std 946-1992, pp. 1–35, Feb. izales, where he taught courses on digital circuits,
1993. servo systems, and flexible manufacturing systems.
[193] A. T. Elsayed, A. A. Mohamed, and O. A. Mohammed, “DC microgrids He was also with the Technical University of Catalo-
and distribution systems: An overview,” Electr. Power Syst. Res., vol. nia, as a Postdoctoral Assistant, teaching courses based on renewable energy
119, pp. 407–417, 2015. systems. In 2011, he was an Assistant Professor in microgrids. From February
[194] IEC—Standardization management board—SG4 LVDC distribu- 2015 to April 2015, he was a Visiting Scholar at the Center of Power Electronics
tion systems up to 1500V DC. (2009). [Online]. Available: Systems at Virginia Tech. He is currently working as an Associate Professor at
http://www.iec.ch/dyn/www /f?p = 103:85:0::::FSP_ORG_ID:6019 the Department of Energy Technology, Aalborg University, Aalborg, Denmark.
[195] DC in the home. (2013). [Online]. Available: https://standards.ieee.org He is coresponsible of the Research Program in Microgrids.
/about/sasb/iccom/IC13-005-02_DC_in_the_Home.pdf His current research interests include operation, power management, hierar-
[196] MIL-STD-1399. (1992). [Online]. Available: https://www.wbdg. chical control, optimization, and power quality applied to distributed generation
org/ccb/ FEDMIL/std1399_300b.pdf and ac/dc microgrids. He is currently a Member of the IEC System Evaluation
[197] E. Planas, J. Andreu, J. I. Gárate, I. M. de Alegrı́a, and E. Ibarra, “AC Group SEG4 on LVDC Distribution and Safety for use in Developed and De-
and DC technology in microgrids: A review,” Renew. Sustain. Energy veloping Economies and the Renewable Energy Systems Technical Committee
Rev., vol. 43, pp. 726–749, 2015. TC-RES in the IEEE Industrial Electronics Society.
[198] IEEE Application Guide for IEEE Std 1547(TM), IEEE Standard for In-
terconnecting Distributed Resources With Electric Power Systems, IEEE
Std 1547.2-2008, pp. 1–217, Apr. 2009.
[199] J. Rekola, DC distribution and power electronics applications in smart Josep M. Guerrero (S’01–M’04–SM’08–FM’15)
grids. (2012). [Online]. Available: http://webhotel2.tut.fi/units/set/opetus received the B.S. degree in telecommunications engi-
/kurssit/SET_1520/Materiaalit-2012/Jenni_DC-distribution.pdf neering, the M.S. degree in electronics engineering,
and the Ph.D. degree in power electronics from the
Technical University of Catalonia, Barcelona, Spain,
in 1997, 2000, and 2003, respectively.
Tomislav Dragičević (S’09–M’13) received the Since 2011, he has been a Full Professor with the
M.E.E. and the industrial Ph.D. degrees from the Department of Energy Technology, Aalborg Univer-
Faculty of Electrical Engineering, Zagreb, Croatia, sity, Aalborg, Denmark, where he is responsible for
in 2009 and 2013, respectively. the Microgrid Research Program, since 2012, he has
He is currently a Postdoctoral Research Associate been a Guest Professor at the Chinese Academy of
at the Institute of Energy Technology, Aalborg Uni- Science, Beijing, China, and the Nanjing University of Aeronautics and As-
versity, Aalborg, Denmark. His Ph.D. dissertation has tronautics, Nanjing, China; since 2014, he has been the Chair Professor at
been carried out in close cooperation with industry Shandong University, Shandong, China; and since 2015, he is a distinguished
and he has received the highest honours for it. He Guest Professor in Hunan University, Hunan, China.
has authored and coauthored more than 60 technical His research interests include different microgrid aspects, including power
papers in his domain of interest, and 18 of them are electronics, distributed energy storage systems, hierarchical and cooperative
published in international journals. His current research interests include over- controls, energy management systems, and optimization of microgrids and is-
all system design of autonomous and grid-connected dc and ac microgrids, and landed minigrids.
industrial application of advanced modeling, control and protection concepts to Prof. Guerrero is an Associate Editor for the IEEE TRANSACTIONS ON POWER
shipboard power systems, remote telecom stations, domestic and commercial ELECTRONICS, the IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, and the
facilities, and electric vehicle charging stations. IEEE INDUSTRIAL ELECTRONICS MAGAZINE, and an Editor for the IEEE TRANS-
Dr. Dragičević is a Member of the IEEE Power Electronics and IEEE Power ACTIONS ON SMART GRID and the IEEE TRANSACTIONS ON ENERGY CONVER-
Systems Societies. He has served in Scientific Committee Boards in several SION. He was the Chair of the Renewable Energy Systems Technical Committee
IEEE conferences and has been invited for guest lectures and tutorials in a num- of the IEEE Industrial Electronics Society. In 2014, he received the Highly Cited
ber of universities and companies around the world. Researcher Award by Thomson Reuters.

You might also like