Architecture, Components and Operation of an

Experimental Hybrid Ac/Dc Smart Microgrid

Márcio S. Ortmann , Victor Maryama? , Luiz J. Camurça, Luiz C. Gili, Diego L. Suarez-Solano
Daniela Dantas, Gustavo Finamor, Victor L. da Silva, Lucas Munaretto,
Adriano Ruseler, André G. Andreta, Roberto F. Coelho, and, Marcelo L. Heldwein

Federal University of Santa Catarina (UFSC) – Electrical Engineering Department (EEL)

Power Electronics Institute (INEP) – www.inep.ufsc.br - Phone:+55(48)3721–7464
88040-970 — PO box: 5119 — Florianópolis, SC, BRAZIL
E-mail: heldwein@inep.ufsc.br


Federal Institute of Santa Catarina (IFSC) – Campus Florianópolis
Department of Electrotechnics – www.ifsc.edu.br - Phone:+55(48)3877–9000
88075-010 — Rua 14 de Julho, 150 — Florianópolis, SC, BRAZIL
E-mail: marcio.ortmann@ifsc.edu.br

?
Reference Centers of Innovative Technologies - CERTI
Sustainable Energy Center
88040-970, Florianópolis, Brazil (e-mail: vma@certi.org.br)

Abstract—This work presents the overall architecture and through power electronics (static switch or power converter),
main components of an experimental smart microgrid. These local intermittent energy sources (solar, wind, tidal, among
are part of a research and development project, which includes others), local dispatchable sources (diesel generators, gas
aspects such as the design and implementation of power and
protection equipments (e.g. power electronics converters and microturbines, fuel cells, among others) and energy storage
static switches), the development of energy market strategies, systems (batteries, flywheels, compressed air systems, among
regulatory issues, control and dispatch optimization for the others). This is illustrated in Fig. 1, where it is seen that
distributed energy resources, implementation of high level op- different possibilities exist to implement the electric energy
eration and supervisory algorithms, among others. Details of the distribution bus or buses in case of hybrid configurations.
proposed hybrid electrical (ac and dc) architecture are presented
and each distributed energy resource is described. Experimental Hybrid microgrid configurations, i.e. employing ac and dc
results showing the adopted hybrid droop control strategy are subgrids present the advantage of being suitable for the
presented. connection of various types of DERs and loads [6], [7]. In
addition, hybrid systems might be a good compromise between
I. I NTRODUCTION the current ac paradigm, where the loads are prepared for an
Energy microgrids are fundamental components of future ac supply, and the high efficiency of dc grids. One of the
high performance grids [1]–[4]. According to the USA De- challenges in hybrid ac/dc microgrids is how to properly share
partment of Energy, “A microgrid is a group of interconnected power variations among energy resources located in the ac
loads and distributed energy resources within clearly defined and dc subgrids. Without a explicit control method and using
electrical boundaries that acts as a single controllable entity conventional power sharing methods (e.g. droop strategies,
with respect to the grid. A microgrid can connect and discon- virtual synchronous machine strategies, among others) would
nect from the grid to enable it to operate in both grid-connected lead to a non proportional power sharing in the case of load
or island-mode” [5]. Such definition is employed in this work being varied in only one of the subgrids. Different control
and implies that the distributed energy resources (DER), which and energy management methods have being proposed to
typically are connected to the grid through power electronics overcome this situation [8]–[14].
converters, are able to supply this small grid even when the This work presents the overall architecture and main com-
distribution network is disconnected. ponents of an experimental smart microgrid in the shape of a
Smart microgrids are typically constructed by connecting a flexible laboratory that enables the experimentation of research
data processing system to an energy infrastructure in which level microgrid concepts. The components highlighted in blue
distributed energy resources are used to feed electrical loads at Fig. 1 are the chosen components for the implementation of
according to local rules that should comply with contracted this microgrid. The microgrid currently focuses on experiment-
energy supply main grid rules. Today’s typical microgrid ing the behavior of a hybrid ac/dc active distribution network
installations employ an electrical connection to a larger grid that uses a novel hybrid ac/dc droop strategy as proposed in
978-1-5090-5339-1/17/$31.00
c 2017 IEEE
 MAIN GRID

POINT OF COMMON COUPLING (PCC)

Data processing and
Communication Measurements Static switch
infrastructure
Power converter Protection

DISPATCHABLE SOURCES CONTROLLABLE LOADS

Diesel generator CHP Heating Lighting

Fuel cells Gas microturbine Air conditioning Cooling

INTERMITTENT SOURCES CRITICAL LOADS

Tidal Wind Hospital process Data processing

Small hydro Solar Security systems Industrial process

ENERGY STORAGE SYSTEM Electric energy distribution bus(es)

Batteries Supercapacitors ac (conventional) dc

Thermal storage Compressed air Hybrid (ac/dc) Medium frequency ac

Microgrid

Fig. 1. Exemplary components of smart microgrids. Blue marked components are the ones integrated into the implemented experimental microgrid.

[15]. A. Microgrid Central Controller and Communication Infras-

tructure
II. G ENERAL D ESCRIPTION OF THE P ROPOSED H YBRID The Microgrid Central Controller (see Fig. 3) performs
M ICROGRID two vital functions in the microgrid: Black-start coordination
and energy/power management. The black-start coordination
The designed hybrid ac/dc microgrid is presented in Fig. 2. ensures that the components of the system will start, from a
This shows, both, power interconnections, which is the main black-out or first power on, in the right order and respecting
focus of this work, and its communication infrastructure. The any limits, as soon as the specific conditions are met. The
power losses of integrating the renewable energy sources microgrid can be brought from power off to either connected
and storage are minimized by connecting them to the dc- or islanded mode. The energy management of the microgrid, in
rated turn, aims to define an operational strategy for the distributed
bus that has a rated voltage of vdc = 380 V). These
include: a solar photovoltaic (PV) generation system (20 kW), resources of the system, acting on energy sources as well
a wind energy conversion system (WECS) that includes a as on internal demand control, with the goal of operating
wind turbine emulator (11 kW), a stationary energy storage the system in the most economical way possible following
system (12 kWh) and an adjustable dc load. Four bidirectional a model similar to the one presented in [16]. The operating
inverters (2×30 kVA and 2×4 kVA) can perform the interface strategy to be achieved must take into account a number
between the dc and ac distribution buses. It should be noted of individual restrictions of the sources, loads and other
that all inverters are parallel-connected, but are not directly components of the microgrid, as well as systemic constraints
related to any subsystem in dc or ac side. At this moment only such as power balance, reserve requirements and exchanges
30 kVA inverter is implemented. The ac port of the inverters with the distribution system, for example. It plays the role of
is a 3-phase/-wire 110/190 V − 60 Hz bus in which a gas the grid’s economic dispatcher, tertiary and secondary control
microturbine emulator system is also connected, being the only by defining power set-points and starting and shutting down
distributed energy resource at the ac-bus of the microgrid. A sources or loads, also being responsible for restoring any
transformer elevates the voltage level to 220/380 V (four- voltage/frequency unbalances in the microgrid. The MGCC
wire ac-bus), which is the comercial voltage at the laboratory runs on a DELL Intel Xeon E3-1241 v3 3.50GHz quad-
site. Controllable ac loads with different priority levels are core server, installed in a rack along with ethernet switches,
connected to this leves. A grid interconnection module, which firewall, UPS and another server running the SCADA system.
is detailed in section II-B, interfaces the main grid to the The communications of the microgrid were implemented
microgrid. A microgrid central controller (MGCC) and a low mainly by the modbus/TCP protocol over an ethernet cabled
speed bidirectional communication network that integrates network. Serial devices were interfaced by protocol gateways
information from all components are also important parts of and the PV generation units communicate to one of such
the microgrid. gateways using a local CAN network. The SCADA system
 Internet (E3)

Microgrid Central Controller / SCADA

dc bus
3 kW

380 V
x 4 kWh Dc loads
Weather
station Gas Microturbine
Emulator
10 kW

From the grid

Dc adjustable
power supply

4 kVA
11 kW (peak)
Wind energy
30 kVA conversion system

4 kVA
20 kW (peak)

Priority loads
Ac bus
Dc bus PV solar generation

20 kVA
Data bus units ( 10 x 2 kW )
30 kVA
kWh
20 kW (peak)

30 kVA Li-ion battery

grid 12 kWh
Inverters
380 V
60 Hz 40 kVA

20 kVA
Loads
Static switch

Relays
13,8 kV PQ analyzer
60 Hz Energy measurement
kWh
Interconnection module Local load
controller

Fig. 2. Simplified one-line diagram of the proposed microgrid.

B. Grid Interconnection Module

The proposed microgrid operates in, both, connected and
islanded modes. In this sense, a proper grid interconnection
module (see Fig. 4) is highly important with respect to reliable
connection/disconnection procedures, circuit protection and
electrical safety. The developed interconnection system is
composed of a 3-phase thyristors-based static switch, voltage
and current measurement circuits and a central digital signal
processing and communication unit based on a BeagleBone
Black platform. The interconnection module realizes several
electrical measurements, and, according to the microgrid state
communicated by the Microgrid Central Controller (MGCC),
makes decisions and interacts with the DER power converters
and the MGCC through a low speed communication bus. Its
main attributions are:
• Grid disconnection triggered by fault events in the main
grid. The events were defined according to the standard
Fig. 3. Microgrid Central Controller functions. IEEE 1547.
• Grid disconnection for programmed events.
• Grid-microgrid synchronization and reconnection process
is built on the Elipse E3 platform and it is responsible for through local controllers actuating at the positive se-
data acquisition, database storage and real-time monitoring of quence phase and amplitude of the voltages in the micro-
the microgrid. It also serves as a graphical interface for config- grid with references at the main grid quantities. The volt-
uring other microgrid components, including the optimization age and phase errors are compensated by proportional-
parameters of the central controller. integral controllers and the compensated signals are
Signal conditioning Voltage sensors Thyristors (a) vpv MPPT vbus

PV modules Temp Voc

10×200 W dc-bus
200..350 V 360..420 V
D

01
Cpv C bus vbus

dc-dc converter
10 0..2.2 kW

(b) (c)

Processor board Current sensors

(a)

Fig. 5. (a) General scheme of the MPPT algorithm employed in the

photovoltaic solar subsystem for one string and respective dc-dc converter.
(b) Installation of the PV modules. (c) Built 2.2 kW dc-dc converter.

converter limits the power limit under maximum power point

5 A/div tracking (MPPT) conditions according to

10 ms/div  0 if vbus ≤ Vmin

PM P P T if Vmin ≥ vbus ≤ Vupp

PP V (vbus ) = ,
(b)  α(vbus ) · PM P P T if Vupp ≥ vbus ≤ Vmax

0 if vbus ≥ Vmax

Fig. 4. Interconnection module: (a) module assembled in the microgrid main (1)
distribution cabinet; and, (b) currents during an intentional disconnection
procedure.
where α(vbus ) = (vbus − Vmax )/(Vupp − Vmax ), Vmin and Vmax
are, respectively, the minimum and maximum dc-bus voltages
at which the converter should work and Vupp is the dc-bus
communicated to the microgrid DER power electronics voltage level at which the converter changes from MPPT mode
converters, changing their droop reference values. to droop mode.
• Power quality measurements in the point of common A temperature-based MPPT technique is employed accord-
coupling (PCC) including harmonics. ing to [17], where the PV module temperature is indirectly
• Communication of the microgrid status. i.e. islanded, estimated by measuring its open circuit voltage. This presents
connected or reconnection modes, to the MGCC. a great advantage that is no precise current measurement is
required and, thus, leads to a lower cost converter system.
III. M ICROGRID E NERGY R ESOURCES Current measurements can then be only used for protection
This section presents details of all generation and energy purposes. Fig. 5 presents a general scheme of one of the PV
storage subsystems of the proposed microgrid. generation system strings.
A. Solar Photovoltaic Generation System B. Wind Energy Conversion System
The overall photovoltaic (PV) system provides a peak power The Wind Energy Conversion System (WECS) consists
capacity of 20 kW at rated conditions. The array is composed of a wind turbine emulation system (WTES) that drives a
of ten PV strings and each string employs ten 200 W (peak) permanent magnet synchronous generator (PMSG) to which a
(Kyocera KC200GT) series-connected modules. Each string high perfomance SiC-based three-level rectifier system (ηrect ≥
feeds a high performance boost interleaved dc-dc converter 98%) is connected. The rectifier controls the power flow from
that performs the power flow control and conditioning to the the PMSG according to a maximum efficiency flux oriented
microgrid dc-bus. The converter switches at 250 kHz each leg control scheme as presented in [18].
and achieves an efficiency η ≥ 98% at rated power (2.2 kW). As shown in Fig. 6, the WTES is composed of a permanent
Proper control strategies allow maximum extraction of the magnet synchronous motor (PMSM) mechanically coupled to
power available in the PV array (PM P P T ) as long as the ac- the PMSG, a commercially available inverter (WEG CFW11),
bus voltage is not over a given limit, from which the dc-dc a torque/speed sensor (Interface T25-100-F6A) and a personal
 Wind Turbine Emulation System Power Power
Grid Source Source µgrid
Grid CFW11 PMSM PMSG Rectifier μ grid
UCC
Ucc

(a) i) Turbine dynamics

emulation
ω, T, Pmec Bus control mode LabVIEW ii) Communication
P1 PC
PC Pel iii) Power source
Setpoints
Droop parameters iv) Fuel consumption
P,Q emulation and
estimation

µgrid communication
(b) (c) network

Fig. 7. Power hardware-in-the-loop scheme for gas microturbine emulation.

Fig. 6. (a) General scheme of the wind energy conversion system emulation
in Fig. 7, where two back-to-back connected commercial
scheme. The personal computer (PC) runs the mechanical model of a wind programmable power sources (Supplier – FCAT 3000-22-
turbine in LabVIEW and interacts with the variable speed drive (CFW11) to 6N4545) are employed. The microgrid side inverter operates
control the PMSM that creates the appropriate torque-speed characteristics at
the mechanical axis. (b) Machines used in the WTES. (c) SiC-based three-
in droop control mode, and receives set-points from a personal
level delta-switch rectifier and two-phase interleaved buck converter. computer running a LabVIEW real-time model of the gas
microturbine. Besides the dynamic model the LabVIEW model
also features fuel consumption according to a Capstone C30
computer running a LabVIEW application where the mechan- documentation efficiency levels. It is also responsible for the
ical system discrete model of a WECS is solved in real-time communication with the MGCC.
and sets the torque versus speed characteristics of the WTES
mechanical axis. The behavior of a given wind turbine is D. Stationary Energy Storage System
emulated considering its constructive dimensions, which yields The stationary energy storage system employs ten series-
power coefficient curves Cp (λ, β), where λ is the tip-speed connected Li-ion modules. The rated parameters of each
ratio and β is the blade pitch angle. Thus, from a given wind module are: 42 Ah capacity, safe failure modes, lifetime
velocity vw and a rotor radius R, the rotor power Pmec and of 3000 cycles (80% DOD), integrated battery management
torque Tmec are respectively calculated [19] with system (BMS) with CAN network, internal fuse, among other
features. The overall battery bank presents a nominal volt-
Pmec = 21 ρπR2 Cp (λ, β) vW
3
(2) age of 240 V and approximately 12 kWh storage capacity.
ρπR2 Cp (λ, β) vW
3 The battery is interfaced with the 380 V dc-bus through a
1
Tmec = 2 . (3) bidirectional dc-dc converter operating in droop mode. The
ω
instantaneous peak power is nearly 30 kW, but the maximum
continuous power is approximately 12 kW. The battery can
Pmec = 12 ρπR2 Cp (λ, β) vW3
(4) be used to smooth intermittent generation, optimize power
2 3
ρπR Cp (λ, β) vW consumption/generation and, maintain power balance at all
Tmec = 12 . (5) times, specially during a microgrid disconnection event, since
ω
the gas microturbine takes nearly 30 s to be operating at rated
The WECS operates in two different modes: constant speed
power.
mode for very high wind speeds and MPPT mode. In both
cases the control action is realized through the electrical E. Dc adjustable load
torque reference. Finally, the personal computer running the The adjustable dc load is built with a three phase buck
LabVIEW model communicates with the inverter, which drives interleaved dc-dc converter. Its function is to track a power
the motor. reference which is determined by the MGCC to emulate
different load levels. For this, the converter modifies the
C. Microturbine Emulation System output voltage that feeds a resistive load. This component
A power hardware-in-the-loop system has been conceived has, among other features, input voltage of 380 V, maximum
in order to emulate a gas microturbine behavior. This system output power of 3 kW and a Modbus serial channel, which
emulates the electrical characteristics (voltage and current) allow the communication with others microgrid components.
of commercial solutions, such as the Capstone C30 model. This converter is also responsible to serve as a over-voltage
The employed model was proposed ini [20] and the con- protection system in case the dc-bus starts to ramp up its
trol is according to [21]. The general scheme is presented voltage beyond a designed maximum level.
 Fuse and contactor

BMS
Communication board

Power connections

CAN data bus

Battery modules

Fig. 10. Developed inverter and its experimental waveforms.

Communication to utilized [22] for power sharing, as well voltage and current
MGCC
other devices strategies [23] for proper operation in terms of power quality.
dc-bus
380 V Charge A. Hybrid Ac/Dc Power Sharing Operation
+ −
controller BMS
In ac microgrids, droop techniques have been frequently
employed for power sharing among inverters [24], [25].
Basically, such scheme controls the frequency and voltage
Fuse 160 A amplitude references, in order to guarantee active proper levels
Contactor 200 A
of active and reactive powers in each converter. However,
droop schemes are sensible to the line impedance, which
dc-dc converter Battery pack is particularly critical in low voltage grids, and thus virtual
10×8224S impedances are commonly employed.
Fig. 8. Energy storage system based on lithium-ion battery modules with
It should be noted that the proposed microgrid is a hybrid
integrated BMS electronics available from Beckett Energy Systems. grid, with DERs and loads in both ac- and dc-side. In this
sense, it is desirable that all generation units share the load,
DSC (control/protection) according to a predefined rate, irrespective of to which subgrid
Dc-bus filter Power modules and cooling unit the loads and sources are connected to. A interesting concept
is proposed in [15], where the frequency in ac side is related
Internal power supply to the dc-bus voltage, and vice versa. Thus, the overall system
operates with a single information source, which is utilized
Communication link for all generation sources for power sharing.
V. E XPERIMENTAL R ESULTS
Fig. 9. Adjustable dc-load unit. This section presents experimental results from the micro-
grid operation. The physical layout of the microgrid laboratory
is seen in Fig. 11.
IV. P OWER I NVERTERS C ONTROL The 4 kVA power inverter unit is presented in Fig. 10,
Inverter units interface the three-phase ac-bus to the dc-bus. where its main components are highlighted. Fig. 10 also shows
In this sense, the overall microgrid performance is directly experimental waveforms of the grid connected inverter during
related to the inverters control strategies. An inverter unit is a current reference step change, where a fast response is
composed of a two-level voltage source inverter (VSI) with observed.
an LCL filter, DSP-based control, internal power supply and Fig. 12 shows the operation of one of the PV generation
auxiliary circuits. All units are able to operate in parallel system 2-kW subsystems, where the PV string voltage vpv
during, both, connected and island microgrid operation modes. presents voltage increases from time to time when the PV
Thus, a proper power sharing algorithm must be employed, string current ipv is interrupted in order to measure the PV
since there are no high speed communications among the DER string open-circuit voltage. The frequency of these measure-
and inverter units. In this work a droop control scheme is ments can be changed according to the expected thermal
 ac loads dc loads batteries PV dc-dc converters battery dc-dc converter interconnection module

inverter

MGCC
μTurbine
emulator

Fig. 11. Microgrid installation layout.

(a) 280
vdc
260 1.5
v pv
240

Voltage [V]
i pv 220 vac
(a)
200
180
100 V/div 0 100 200 300 400 500 600 600 800
0.5 A/div Time [s]
20 s/div
(b) 40
Fig. 12. Experimental results from the microgrid operation: PV generation 30 iac
idc
system waveforms showing the PV string voltage and current under low
Current [A]

radiance levels achieved on a cloudy day. 20

10
variations in the PV modules. The acquisitions in Fig. 12 were
performed in a overcast day. The variations in ipv are due to 0
sudden variations in the microgrid dc-bus voltage vdc . 0 100 200 300 400 500 600 600 800
Islanded mode operation is observed in Fig. 13, where Time [s]
the measurements were performed with a Yokogawa WT1800
Power Analyzer. The active power in the dc-bus pdc is varied (c) 10
in steps (see Fig. 13(c)) in order to verify what occurs in, both, 8 pdc
ac and dc subgrids. The dc-bus current is seen in Fig. 13(b)
and varies at nearly the same way as the dc-bus power. 6
Power [kW]

However, the ac-bus phase a current iac does not follow the
4
same behavior. However, the observed proportionality between
ac- and dc-bus voltages in Fig. 13(a) shows that the hybrid 2
droop control is operating as expected since it guarantees the
0
the voltage amplitude in both subgrids follow one another 0 100 200 300 400 500 600 600 800
irrespective of the behavior of the currents. Time [s]
VI. C ONCLUSIONS
Fig. 13. Experimental results from the microgrid operation under islanded
Microgrids are a promising concept for future electrical mode: (a) ac-bus phase a rms voltage vac and scaled dc-bus voltage vdc ·2/3;
systems. In this paper, the overall architecture and the DERs (b) ac-bus phase a rms current iac and dc-bus current idc ; and, (c) average
dc-bus power pdc .
of an experimental microgrid have been presented. The main
specifications are introduced and the more relevant aspects
related to technology and operation of the smart microgrid
components were discussed. Concerning the proposed hybrid and the presence of Power-HIL emulation systems allow a
electrical architecture (ac/dc), high efficiency and versatility great variety of future studies in several areas, e.g. control,
are possible due to variety of energy resources and loads. From protection, dispatch optimization, communication, among oth-
the laboratory point of view, the flexibility of the arrangement ers. Finally, experimental results were presented that illustrate
certain aspects of the hybrid ac/dc microgrid operation. [13] A. Mohamed, V. Salehi, and O. Mohammed, “Real-time energy man-
agement algorithm for mitigation of pulse loads in hybrid microgrids,”
VII. ACKNOWLEDGEMENT IEEE Transactions on Smart Grid, vol. 3, no. 4, pp. 1911–1922, Dec
2012.
This work was developed within the scope of Engie Brasil [14] N. Eghtedarpour and E. Farjah, “Power control and management in a
Energia’s R&D Program, which is regulated by the Brazilian hybrid ac/dc microgrid,” IEEE Transactions on Smart Grid, vol. 5, no. 3,
National Electric Energy Agency (ANEEL). pp. 1494–1505, May 2014.
[15] V. Maryama, V. Zeni, C. Q. Pica, M. S. Ortmann, and M. L. Heldwein,
R EFERENCES “Unified hybrid (ac/dc) active distribution networks droop-based load-
sharing strategy,” in Innovative Smart Grid Technologies Conference
[1] R. H. Lasseter, “Microgrids,” in Power Engineering Society Winter Europe (ISGT-Europe), 2014 IEEE PES, pp. 1–6.
Meeting, 2002. IEEE, vol. 1, pp. 305–308 vol.1. [16] D. Tenfen and E. C. Finardi, “A mixed integer linear programming model
[2] J. A. Peas Lopes, C. L. Moreira, and A. G. Madureira, “Defining for the energy management problem of microgrids,” Electric Power
control strategies for microgrids islanded operation,” Power Systems, Systems Research, vol. 122, pp. 19–28, 2015.
IEEE Transactions on, vol. 21, no. 2, pp. 916–924, 2006. [17] R. F. Coelho, F. M. Concer, and D. C. Martins, “A mppt approach based
[3] N. Hatziargyriou, H. Asano, R. Iravani, and C. Marnay, “Microgrids,” on temperature measurements applied in pv systems,” in Industry Ap-
Power and Energy Magazine, IEEE, vol. 5, no. 4, pp. 78–94, 2007. plications (INDUSCON), 2010 9th IEEE/IAS International Conference
[4] J. M. Guerrero, J. C. Vasquez, J. Matas, L. G. de Vicuna, and M. Castilla, on, pp. 1–6.
“Hierarchical control of droop-controlled ac and dc microgrids - a gen- [18] L. J. Camurça, J. Lago, and M. L. Heldwein, “High efficiency wind
eral approach toward standardization,” IEEE Transactions on Industrial energy conversion system based on the three-level delta-switch t-
Electronics, vol. 58, no. 1, pp. 158–172, 2011. type converter and pmsg model-based loss minimization,” in Power
[5] DoE, “Summary report: 2012 doe microgrid workshop doe eere,,” Electronics Conference and 1st Southern Power Electronics Conference
Department of Energy - Office of Electricity Delivery and Energy (COBEP/SPEC), 2015 IEEE 13th Brazilian. IEEE, 2015, pp. 1–6.
Reliability Smart Grid R&D Program, Tech. Rep., 2012. [19] A. Murdoch, J. Winkelman, S. Javid, and R. Barton, “Control design
[6] Z. Jiang and X. Yu, “Hybrid dc- and ac-linked microgrids: Towards and performance analysis of a 6 mw wind turbine-generator,” IEEE
integration of distributed energy resources,” in 2008 IEEE Energy 2030 transactions on power apparatus and systems, no. 5, pp. 1340–1347,
Conference, Nov 2008, pp. 1–8. 1983.
[7] J. M. Guerrero, P. C. Loh, T. L. Lee, and M. Chandorkar, “Advanced [20] W. I. Rowen, “Simplified mathematical representations of heavy-duty
control architectures for intelligent microgrids - part ii: Power quality, gas turbines,” Journal of engineering for power, vol. 105, no. 4, pp.
energy storage, and ac/dc microgrids,” IEEE Transactions on Industrial 865–869, 1983.
Electronics, vol. 60, no. 4, pp. 1263–1270, April 2013. [21] G. Li, G. Li, W. Yue, M. Zhou, and K. Lo, “Modeling and simulation
[8] X. Liu, P. Wang, and P. C. Loh, “A hybrid ac/dc microgrid and its of a microturbine generation system based on pscad/emtdc,” in Critical
coordination control,” IEEE Transactions on Smart Grid, vol. 2, no. 2, Infrastructure (CRIS), 2010 5th International Conference on. IEEE,
pp. 278–286, June 2011. 2010, pp. 1–6.
[9] P. C. Loh, D. Li, Y. K. Chai, and F. Blaabjerg, “Hybrid ac/dc micro- [22] J. Rocabert, A. Luna, F. Blaabjerg, Rodri, x, and P. guez, “Control
grids with energy storages and progressive energy flow tuning,” IEEE of power converters in ac microgrids,” IEEE Transactions on Power
Transactions on Power Electronics, vol. 28, no. 4, pp. 1533–1543, April Electronics, vol. 27, no. 11, pp. 4734–4749, 2012.
2013. [23] A. G. Andreta, M. S. Ortmann, and M. L. Heldwein, “Hybrid current
[10] A. A. A. Radwan and Y. A. R. I. Mohamed, “Networked control and control technique applied to grid connected inverters,” 2015.
power management of ac/dc hybrid microgrids,” IEEE Systems Journal, [24] Y. Wei, C. Min, J. Matas, J. M. Guerrero, and Q. Zhao-Ming, “Design
vol. PP, no. 99, pp. 1–12, 2014. and analysis of the droop control method for parallel inverters con-
[11] F. Nejabatkhah and Y. W. Li, “Overview of power management strategies sidering the impact of the complex impedance on the power sharing,”
of hybrid ac/dc microgrid,” IEEE Transactions on Power Electronics, Industrial Electronics, IEEE Transactions on, vol. 58, no. 2, pp. 576–
vol. 30, no. 12, pp. 7072–7089, Dec 2015. 588, 2011.
[12] P. C. Loh, D. Li, Y. K. Chai, and F. Blaabjerg, “Autonomous control of [25] E. A. A. Coelho, P. C. Cortizo, and P. F. D. Garcia, “Small-signal sta-
interlinking converter with energy storage in hybrid ac/dc microgrid,” bility for parallel-connected inverters in stand-alone ac supply systems,”
IEEE Transactions on Industry Applications, vol. 49, no. 3, pp. 1374– Industry Applications, IEEE Transactions on, vol. 38, no. 2, pp. 533–
1382, May 2013. 542, 2002.