Low Carbon Economy, 2010, 1, 39-53
doi:10.4236/lce.2010.12006 Published Online December 2010 (http://www.SciRP.org/journal/lce)
Copyright © 2010 SciRes. LCE
39
Analysis of Transient Dynamic Response of Two Nearby
Micro-Grids under Three Different Control Strategies
Rashad M. Kamel, Aymen Chaouachi, Ken Nagasaka
Environmental Energy Engineering, Department of Electronics & Information Engineering, Tokyo University of Agriculture and
Technology, Tokyo, Japan.
Email: r_m_kamel@yahoo.com, a.chaouachi@gmail.com, bahman@cc.tuat.ac.jp
Received October 11th, 2010; revised November 20th, 2010; accepted December 10th, 2010.
ABSTRACT
The need of reducing CO2 emissions in electricity generation field for solving global warming problems have led to
increase the interest in Micro-Grid (MG) especially the one which included renewable sources. MG normally operates
in normal interconnected mode and connects with the main grid. When a large disturbance happens in main grid, MG
transfer to islanding mode. This paper deals with connecting two nearby Micro-Grids to enhance transient dynamic
response of the two MGs after isolated from the main grid. Three cases are investigated. The first case discussed the
dynamic response of the two MGs when there is no tie line connection between the two MGs after islanding. Second
case, studied the dynamic performance of the two Micro-Grids when there is a private line connects the two MGs after
islanding from main grid, while the third case deals with two interconnected MGs (after islanding) and automatic gen-
eration control (AGC) applied upon each MG to return the frequency to its nominal value and control the tie line power
to be with its scheduled value. Results proofed that when two nearby MGs are connected by private line after islanding
from the main grid occurs, dynamic response of the two MGs improved well.
Keywords: Micro-Grid, Islanding, Dynamic Response, Tie Line, Nearby MGs and Automatic Generation Control
1. Introduction
Economic, technology and environmental incentives are
changing the face of electricity generation and transmis-
sion. The need of reducing CO2 emissions in the electric-
ity generation field, recent technological developments in
micro generation domain in addition to electricity busi-
ness restructuring are the main factors responsible for the
growing interest in the use of micro generations [1,2].
Energy investors and utility operators are attracted to the
MG role and associated industry for the following fore-
seen opportunities [3]:
Distributed generators (DGs) installed inside MG
can be fueled by locally available renewable and
alternative mix of fuel sources. Greater independ-
ency from importing petroleum fuel can be
achieved by incorporating MG that is powered by
various fuel sources.
MG can support future increase in demand without
investment in the expansion of existing distribution
network by installing the MG very close to the new
load centre.
MG can be used in reducing intermittent and peak
supply burdens on utilities grid by injecting power
during peak periods.
MG could contribute to decreasing the vulnerabil-
ity of the electric distribution system to external
threats and hidden undetected faults that may cause
wide scale blackout by feeding power to the sensi-
tive infrastructure.
In fact the connection of small generation units (micro
sources) with power rating less than a few tens of kilo-
watts to low voltage (LV) networks potentially increases
the reliability to final consumers and brings additional
benefits for global system operation and planning,
namely, regarding investment reduction for future grid
reinforcement and expansion [4]. In this context, a MG
can be defined as a low voltage network (e.g. a small
urban area, a shopping center, or an industrial park) plus
its loads and several small modular generation systems
connected to it, providing both power and heat to local
loads. The MG is intended to operate in the following
two different operating conditions:
Normal Interconnected Mode: MG is connected to
a main grid (distribution network), either being
supplied by it or injecting some amount of power
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
40
into the main system.
Islanding Mode: MG operates autonomously, in a
similar way to physical islands, when the discon-
nection from the upstream distribution network
occurs.
The development of MG can contribute to the reduc-
tion of emissions and the mitigation of climate changes;
this is because available and currently developing tech-
nologies for distributed generation units are based on
renewable sources and micro sources that are character-
ized by very low emissions [5]. The new micro sources
technologies (e.g. micro gas turbines, fuel cells, photo-
voltaic panels and several kinds of wind turbines) used in
MG are not suitable for supplying energy to MG directly.
They have to be interfaced with the MG through an in-
verter. Thus, the use of power electronic interfaces in the
MG leads to a series of challenges in the design and op-
eration of the MG [6]. Technical challenges associated
with the operation and control of MG are immense. En-
suring stable operation during network disturbances,
maintaining stability and power quality during the is-
landing mode of operation requires the development of
sophisticated control strategies for MG’s inverters in
order to provide stable frequency and voltage in the
presence of arbitrarily varying loads.
1.1. General Overview
Reference [4] described and evaluates the feasibility of
control strategy to be adopted for the operation of the
MG when it becomes isolated. Reference [5] studied the
MG during both connected and islanded modes of opera-
tions. Reference [7] investigated preplanned switching
events and fault events that lead to islanding of the MG.
The feasibility of the MG islanding mode concept was
laboratory tested in a prototype installed in National
Technical University of Athens (NTUA) which compro-
mises a photovoltaic (PV) panel, battery storage, loads
and a controlled interconnection to LV grid [8]. In [9]
and [10], the behavior of micro sources connected to
distribution networks has been addressed.
All the previous mentioned references besides all other
works available in the literature dealt with dynamic re-
sponse of one MG only. The next few years will see in-
tegration of many MGs inside the same distribution net-
work. At that time, if the nearby MGs are connected with
each other by a private line (following fault or high dis-
turbance occurrence in the distribution network), the
transient dynamic response of all interconnected MGs
will highly improved. This paper dealt with the investi-
gation of transient dynamic response of two intercon-
nected MGs after islanding from the distribution network
occurs. To deals with the proposed study, the following
three issues are described.
1) Investigating the transient dynamic response of two
MGs subsequent islanding occurrence from the main grid
if the two MGs are not interconnected with each other.
2) Investigating the transient dynamic response of the
two MGs subsequent islanding occurrence from the main
grid and the two MGs are connected with each other
through a private line.
3) Third case is close to the second case in addition to
automatic generation control (AGC) is applied inside
each MG to return the frequency to the nominal value
and control the tie line power to its scheduled value.
The rest of the paper is organized as follow. Section 2
describes architecture of the two investigated MGs and
the conditions which may lead to transfer the two MGs to
the islanding mode. Section 3 presents a brief description
of all micro sources installed inside the two MGs. Sec-
tion 4 describes three control scenarios proposed for im-
provement the dynamic performance of adjacent MGs.
Results and discussion are presented by Section 5. Con-
clusions are exist in Section 6.
2. Architecture of the Developed Two MGs
System
Figure 1 shows architecture of the developed two MGs
system. System consists of two MGs, each MG con-
nected to the main grid (distribution network) through
separate transformer (T1 and T2). The two MGs have the
same structure but with different micro sources rating
and loading. Each MG comprises low voltage network,
loads, both controllable and non controllable micro
sources and storage device (flywheel). Controllable mi-
cro sources are the micro sources which can control their
output power like micro turbine and fuel cell, while the
non controllable micro sources are the micro sources
which their output depends on weather conditions like
wind turbines and photovoltaic panels. As shown in Fig-
ure 1 each MG consists of 7 buses. Flywheel (storage
device) is connected to bus 1. Wind generation system is
located at bus 2. Two photovoltaic panels with different
rating are connected to buses 4 and 5. Single Shaft Micro
Turbine (SSMT) is connected to bus 6. Bus 7 is provided
with Solid Oxide Fuel Cell (SOFC). Rating of all micro
sources and loads in two MGs are shown in the figure.
All micro sources in the two MGs are interfaced to the
MG through inverters except the wind generation system
directly coupled. As shown in the figure, in steady state
the two MGs are connected to the distribution network
(main grid). If fault occurs in main grid as shown in the
figure, the two MGs will be isolated from the main grid
as soon as possible by CB1 and CB2, while CB3 connects
the two MGs with each other. As we will see in the results,
this proposed strategy has large effect for enhancement
the transient dynamic response of the two MGs.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
41
Figure 1. Architecture of the developed two MGs system.
3. Dynamic Modeling of MG’s Components
All MG’s components are modeled in detail using Mat-
lab® Simulink® environment. Detailed standalone mod-
els for inverter with different control strategies, SSMT,
SOFC, wind generation system and photovoltaic panel
models are developed in our previous work and can be
found in references [11,12]. All models developed in
references [11,12] are based on description and equations
presented by references [13-15]. Here we will briefly
describe about the inverter control strategies used in this
paper and storage devices (flywheel) modeling beside
three control strategies adopted to improve the perform-
ance of the two MGs.
3.1. Inverter Modeling
Inverter plays a vital role in the system which interfer-
ence of micro sources with MG. Two kinds of control
strategies are developed in this study to operate the in-
verter.
1) PQ Inverter Control: This type of inverter is used to
inject a certain active and reactive power set-value. This
type of inverter is used to interface SSMT, SOFC and the
two photovoltaic panels. The basic structure of the in-
verter PQ controller is shown in Figure 2. Pref in Figure
2 represents active power produced by the micro source
which interfaced to the MG by this inverter. Qref repre-
sents amount of reactive power injected to or absorbed
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
42
from the MG at inverter’s bus.
2) Voltage Source Inverter (VSI) Control: This in-
verter is controlled to “feed” the load with predefined
values of voltage and frequency. Depending on the load,
the VSI real and reactive power is defined. VSI is used to
interface storage device (flywheel) to the MG and repre-
sent reference bus (slack bus) for each MG during
islanding mode. VSI emulates the behavior of synchro-
nous machine, thus controlling voltage and frequency on
the AC system. VSI acts as a voltage source, with the
magnitude and frequency of the output voltage controlled
through droops, as described in the following equation:
*
*
oP
oQ
f
fkP
VV k Q

 (1)
Where, P and Q are the inverter active and reactive out-
put power, kP and kQ are the droop slopes (positive quan-
tities), and fo and Vo are the idle values of the frequency
and voltages (nominal frequency and nominal voltage).
A three-phase model of a VSI implementing the droop
concepts described by Equation (1) was developed as
shown in Figure 3. In this model amount of active and
reactive power injected to or absorbed from MG will
control the voltage and frequency of the MG (like syn-
chronous generator).
3.2. Storage Devices Modeling
Due to the large time constants of responses of some
micro sources, such as fuel cell and micro turbine, stor-
age device must be able to provide the amount of power
required to balance MG following disturbances and/or
significant load changes. Storage devices, such as fly-
wheel and batteries, are modeled as constant DC voltage
source using power electronic interfaces to be coupled
with the electrical network. The storage device used in
this paper is a flywheel and is connected to the VSI. The
active and reactive power needed to balance generation
and consumption inside the MG which injected to or ab-
sorbed from MG are proportional to frequency and volt-
age deviation (frequency and voltage droops).
4. Control of the Two MGs Subsequent
Islanding from the Main Grid
In the presence of unplanned events like severe faults in
the main grid as shown in Figure 1, MG separation from
the main grid must occurs as fast as possible. If there are
no synchronous machines to balance the demand and
supply (as our case), the inverters must be responsible for
frequency and voltage control during islanding operation.
During connected mode (connected with main grid), all
inverters can be operated in PQ mode, in which voltage
and frequency references (main grid) are available. In
this case, a sudden disconnection from the main grid
would lead to the loss of the MG, because load/gener-
ation balancing and therefore frequency and voltage con-
trol is not possible. However, by using a VSI connected
Figure 2. Basic structure of the PQ inverter control scheme.
Figure 3. Voltage source inverter (VSI) control model.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
43
to the flywheel to provide a reference for voltage and
frequency as explained in Equation (1), it thus possible to
operate the MG in islanding mode.
4.1. Two MGs Transfer to Islanding Mode and
Separated from Each Other
In this case there is no interconnection between the two
MGs when islanding from main grid happens. During
this situation, the voltage source inverter (VSI) con-
nected to the flywheel is responsible for frequency and
voltage control in each MG as described by Equation (1)
and shown in Figure 4. However, those devices (fly-
wheels) with high capabilities for injecting power during
small time intervals have a finite storage capacity.
Therefore, correcting permanent frequency deviations
during any islanded operation conditions should then be
considered as one of the key objectives for any control
strategy. In order to promote adequate secondary control
aiming to restore frequency to nominal value after dis-
turbance, local frequency control by using a PI controller
at each controllable micro source (SSMT and SOFC) is
used to control active power outputs of the primary en-
ergy sources based on the frequency deviation error as
shown in Figure 4. As indicated in Figure 4, the fly-
wheel with its VSI acts as primary frequency control
(acts as the inertia of synchronous machine in bulk power
system) and the controllable micro sources (SSMT and
SOFC) acts to balance the load and generation inside
each MG. In addition, the VSI connected to the flywheel
controls voltage of the MG (by controlling reactive pow-
er injected to or absorbed from MG) and acts like auto-
matic voltage control in conventional bulk power plant as
indicated in Figure 4.
4.2. Two MGs Transfer to Islanding Mode and
Connected to Each Other by a Private Line
In this case, there is a private tie line connected the two
MGs immediately subsequent islanding from main grid.
The value of power flowing through the tie line depends
on the difference between the two MGs frequencies
(f1-f2) as shown in Figure 5. Also, as discussed in case
A, flywheel with its VSI will control voltage and fre-
quency (primary frequency control beside voltage control)
and controllable micro sources (SSMT and SOFC) acts
as secondary frequency control as indicated by Figure 5.
In this work, at the instant of islanding from main grid,
the first MG has heavy loads and little generation, so that
MG1 was imported certain amount of power (about 15
kW) from the main grid before transfer to islanding mode,
while the second MG has lightly loads and more genera-
tion which force MG2 to export some power to the main
grid (9 kW) before islanding occurs.
4.3. Two MGs Transferred to Islanding Mode
and Connected to Each Other and Applied
Automatic Generation Control upon Each
MG
This case is close to the second case in addition to fre-
quency bias tie line control (discussed later) is applied
inside each MG as shown in Figure 6. Frequency bias tie
line controller will acts on the reference power of the
controllable micro sources inside each MG (SSMT and
SOFC) to correct the frequency deviation (return the
frequency to its nominal value) and also back the tie line
power to its scheduled value. In this study, the scheduled
value of tie line power is zero. This means that the two
MGs exchanges powers (active and reactive) during
transient state only. After transient period finished, each
MG will feeds its loads and the tie line power (active and
reactive) returns to its value before transient occurrence.
4.4. Frequency Bias Tie Line Control
The basic objectives of frequency bias tie line control are
to restore balance between each MG’s loads and genera-
tions. This is met when the control action maintains:
Frequency at the nominal value.
Net interchange power with neighboring Micro-
grids at scheduled value.
The supplementary control (frequency bias tie line
control) in a given MG should ideally correct only for
changes in that MG. In other words, if there is a change
in MG1’s load, there should be supplementary control
action only in MG1 and not in MG2 (after transient state
finished). To execute that supplementary control inside
each MG, a control signal made up of tie line flow devia-
tions (Ptie12) added to frequency deviation weighted by
a bias factor would accomplish the desired objectives.
This control signal is known as Area Control Error (ACE)
in bulk conventional power system [16]. The same tech-
nique is used in our study for controlling the tie line
power exchanges between the two MGs.
Based on above discussions, Area Control Error of
MG1 (ACE1) is given by the following equations:
11211tie
A
CEPB f
 (2)
B1 is the bias factor for MG1, and given by the follow-
ing equation:
111P
BK D
(3)
where, KP1 is frequency droop gain of MG1 given in equ-
ation (1) used for controlling flywheel installed in that
MG. D1 represents percentage of load change inside MG1
due to frequency deviation from its nominal value.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
44
Figure 4. Control scheme of the two MGs (no private line connected the two MGs).
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
45
Figure 5. Control of two connected MGs without AGC.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
46
Figure 6. Control scheme of two connected MGs with AGC.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
47
Similarly for MG2, Area Control Error (ACE2) is given
by:
22122 1222tie tie
A
CEPB fPB f  (4)
B2 is the bias factor for MG2, and is given by the fol-
lowing equation:
222P
BK D (5)
where, KP2 is the frequency droop constant of MG2 used
for controlling flywheel installed in that MG. and D2
represents percentage of load variation in MG2 due to
frequency deviation from the nominal value subsequent
islanding occurrence from main grid.
ACE for each MG represents the required change in
MG generation. The block diagram in Figure 6 illus-
trates how Automatic Generation Control (AGC) imple-
mented inside each MG using ACE signal which applied
to the reference power of the controllable micro sources
installed inside each MG (SOFC and SSMT).
5. Results and Discussions
Disconnection from the upstream main grid was simu-
lated in order to understand the dynamic behavior of each
MG in the three studied cases. As we discussed before,
SSMT and SOFC are supposed to be controllable micro
sources used for secondary frequency control in the first
two studied cases as shown in Figures 4 and 5, and used
for AGC in third case as shown in Figure 6. In all three
studied cases, wind speed and solar irradiance are as-
sumed varying continuously. Wind speed and solar ir-
radiance data are available in reference [4]. The simula-
tion results are presented for the main quantities (Fre-
quency, voltage, power, tie line power… etc).
From Figures 7-15, the following points can be sum-
marized.
5.1. For the First Case (Two MGs Separated)
Before islanding occurs, the two MGs are in their
steady state. The frequency of the two MGs is at
the nominal value (50 Hz). Any load change inside
each MG can be compensated by the main grid and
no need for secondary frequency control applied to
the controllable micro sources inside each MG.
When islanding occurred at t = 60 sec., the two
MGs islanded from main grid.
For this studied case (no interconnection between
the two MGs), each MG performs alone and the
VSI connected to the flywheel in each MG acts to
control the voltage and frequency of this MG.
Frequency deviation in each MGs acts on reference
power set points of controllable micro sources and
try to adjust the generation and load inside each
MG. Due to power deficit in MG1, frequency
dropped to about 49.45 Hz as shown in Figure 7,
while due to power surplus in MG2, frequency
raised to about 50.15 Hz as shown in Figure 8.
Frequency deviation acts to increase the power of
SSMT and SOFC of MG1, while it acts to decrease
the power of SSMT and SOFC of MG2 as shown in
Figures 12 and 13, respectively.
The voltage of MG1 buses shows high drop (Fig-
ure 14) due to deficit of power (active and reac-
tive), while voltage of MG2 buses shows small drop
Figure 15 due to lost some reactive power which
was supplied by the main grid before islanding.
In order the two MGs can keep their stability sub-
sequent islanding occurrence from the main grid,
the flywheel of MG1 must inject about 15 kW and
the flywheel of MG2 must absorb about 9 kW as
shown in Figures 9 and 10, respectively.
Controllable micro sources continue in adapting
their generation until balance between load and
generation inside each MG occurs (Figures 12 and
13) and the frequency of each MG back to its no-
minal value as shown in Figures 7 and 8.
Fluctuation of powers generated by the renewable
source (wind turbine and photovoltaic panels) due
to change of wind speed and solar irradiance cause
fluctuations in the frequency of the two MGs Fig-
ures 7 and 8 and the VSI connected to the flywheel
compensate those fluctuations (Figures 9 and 10)
until the controllable micro sources can balance the
load and generation inside each MG.
5.2. For the Second Studied Case (Two MGs
Connected with Each Other without AGC)
For this case, the two MGs have the same condition
of the first case, but the two MGs connected with
each other subsequent islanding occurrence from
the main grid. In this case due to existence of the
tie line between the two MGs, power surplus in
MG2 will support the power deficit in MG1 and the
frequency deviation of the two MGs shows small
deviation compared with the first case. As shown
in Figure 7, frequency f1 dropped only to 49.93 Hz
compared with 49.75 Hz in the first case, while the
frequency f2 will drop to about 49.93 Hz compared
with 50.15 Hz in the first case as shown in Figure
8.
Amount of active power required to be injected by
flywheel of MG1 about 4 kW compared with 15
kW in the first case (Figure 9), while the active
power required from flywheel of MG2 about 3 kW
compared with – 9 kW in the first case (Figure 10).
This means that when the two MGs connected with
each other following islanding occurrence, the rating
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
48
Figure 7. Frequency of MG1 for the three proposed controller schemes.
Figure 8. Frequency of MG2 for the three proposed controller schemes.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
49
Figure 9. Active power injected by VSI connected to flywheel of MG1.
Figure 10. Active power injected by VSI connected to flywheel of MG2.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
50
Figure 11. Tie line active power exchanges between the two MGs (Ptie12).
Figure 12. Active power injected by SSMT installed in MG1.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
51
Figure 13. Active power injected by SSMT installed in MG2.
Figure 14. Voltage at bus # 6 of MG1.
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
52
Figure 15. Voltage at bus #6 of MG2.
of the required storage devices (flywheels) and ac-
companied inverters (VSI) is small compared with
separated MGs.
Voltage of MG1 has less drop (0.97 p.u) in the
second case compared with the first case (0.955 p.u)
due to power (active and reactive) flows from MG2
through the tie line.
As shown in the Figures 7-15, the dynamic re-
sponse of the two MGs can be highly improved if
the nearby MGs installed in the same distribution
network are connected with each other by a private
line subsequent islanding occurrence from main
grid. Also, this interconnection between the adja-
cent MGs will be very important if one dominant
micro source fails inside any MGs. For instance, if
VSI accompanied with flywheel fails in any MG
(after islanding from main grid), that MG will
transfer to blackout unless the interconnection be-
tween the MGs exist. Also, this interconnection
between the nearby MGs may be very useful if the
controllable micro sources inside any MG reached
their nominal power and can not produced any ad-
ditional power. During those conditions, intercon-
nection between adjacent MGs is necessary to keep
the stability of the heavy loaded MG and keeps it
far from blackout.
5.3. For Third Case (Two MGs Connected with
Each Other with AGC)
This case has performance almost close to the per-
formance of the second case, however, AGC inside
each MG acts on controllable micro sources to re-
turn the tie line power to its scheduled value (zero
in our case) besides returns frequency to its nomi-
nal value as shown in Figures 7, 8, and 11.
For this case, voltage of MG2 is less than the two
previous cases (Figure 15). This is because, when
AGC applied in MG2, this MG must reduce its
power (active and reactive) to return the tie line
power to zero value. Reducing power generation
inside MG2 leads to some drop in its voltages.
Applying AGC inside each MG requires high rat-
ing of the controllable micro sources (Figure 12) to
enable the MGs feed their loads locally and re-
duced the tie line power to zero as shown in Figure
11.
AGC is necessary when contract between the own-
ers of the two MGs states that the two MGs only
support each other during transient state. After the
transient state finished, each MG must feed its
loads. In this case, tie line improved reliability and
security of the two MGs. Also, reducing the tie line
Analysis of Transient Dynamic Response of Two Nearby Micro-Grids under Three Different Control Strategies
Copyright © 2010 SciRes. LCE
53
power to zero after the transient state finished, will
reduce the losses inside the two MGs. This is be-
cause each MG feeds its loads locally and no power
circulating from one MG to the other.
6. Conclusions
This paper dealt with improvement the dynamic response
of the two nearby MGs when the two MGs connected
through a private line subsequent islanding from the
main grid occurs. It is noted from the results that the dy-
namic performance will gain high improvement by con-
nected two nearby MGs. Frequency deviations is highly
reduced. It is noted from the results that instead of the
frequency of MG1 drops to 49.75 Hz without intercon-
nection between the two MGs, it will only drop to 49.93
Hz. For MG2, frequency dropped to about 49.93 Hz in-
stead of increasing to 50.15 Hz. Also, amount of active
power required to be injected by each flywheel highly
reduced. AGC is applied to return the frequency of the
two MGs to its nominal value and control the tie line
power to its scheduled value according to the contract
signed between the owners of the two MGs. Tie line
connected the two MGs will be necessary in some emer-
gencies situations like failure of a dominant micro source
inside any MG or reaching the nominal power of con-
trollable micro sources inside one of the MGs. Such con-
ditions will force one of the MGs to blackout if tie line
absent.
Finally we can conclude that connecting two nearby
MGs with a private tie line when transfer to islanding
mode from main grid is very effective method to increase
the reliability and security of both MGs. Of course more
improvement can be achieved if more nearby MGs con-
nected with each others (three, four …etc). Connecting
more than MGs in islanding mode produces more strong
MGs system when the MGs isolated from main grid due
to large disturbance (fault) occurs in the main grid.
REFERENCES
[1] R. Strzelecki, “Power Electronic Control in Smart Elec-
trical Networks,” Springer-Verlag, London, 2008.
[2] R. Lasseter, et al., “White Paper on Integration of Distrib-
uted Energy Resources - The CERTS Micro Grid Con-
cept,” 2002. http://certs.lbl.gov/pdf/LBNL-50829.pdf
[3] N. Hatziargyriou, “MICROGRIDS-Large Scale Integra-
tion of Micro-Generation to Low Voltage Grids,” EU
contract ENK5-CT-2002-00610, Technical annex, May
2002. http://microgrids.power.ece.ntua.gr/
[4] J. Lopes, et al., “Defining Control Strategies for Micro
Grids Islanded Operation,” IEEE Transactions on Power
Systems, Vol. 21, No. 2, May 2006, pp. 916-924.
[5] F. Kanellos, et al., “Micro-Grid Simulation during Grid-
Connected and Islanded Modes of Operation,” Intern-
tional Conference on Power Systems Transients
(IPST’05), Montereal, No. IPST05-113, June 19-23,
2005.
[6] S. Barsali, et al., “Control Techniques of Dispersed Gen-
erators to Improve the Continuity of Electricity Supply,”
Proceedings of Power Engineering Society Winter Meet-
ing, Vol. 2, 2002, pp. 789-794.
[7] F. Katiraei, et al., “Micro-grid Autonomous Operation
during and Subsequent to Islanding Process,” IEEE
Transactions on Power Delivery, Vol. 20, No. 1, January
2005, pp. 248-257.
[8] D. Georgakis, S. Papathanassiou, N. Hatziargyriou, A.
Engler and C. Hardt, “Operation of a Prototype Microgrid
System Based on Micro-sources Equipped with Fast act-
ing Power Electronics Interfaces,” Proceedings of IEEE
Power Electronics Specialists Conference, Vol. 4, Aachen,
2004, pp. 2521-2526.
[9] T. Tran-Quoc, et al., “Dynamic Analysis of an Insulated
Distribution Network,” Proceedings of IEEE Power Sys-
tems Conference and Exposition, Vol. 2, October 10-13,
2004, pp. 815-821.
[10] R. Caldon and R. Turri, “Analysis of Dynamic Perform-
ance of Dispersed Generation connected through Inverters
to Distribution Networks,” Proceedings of 17th Interna-
tional Conference on Electricity Distribution (CI-
RED), Barcelona, No. 87, May 2003.
[11] R. M. Kamel and K. Nagasaka, “Design and Implementa-
tion of Various Inverter Controls to Interface Distributed
Generators (DGs) in Micro Grids,” Japan Society of En-
ergy and Resources, Session 6.1, Tokyo, 2009, pp. 61-64.
[12] R. M. Kamel and K. Nagasaka, “Micro-Grid Dynamic
Performance Subsequent to Islanding Process,” Japan
Society of Energy and Resources, Session 6.1, Tokyo,
2009, pp. 65-68.
[13] Y. Zhu and K. Tomsovic, “Development of Models for
Analyzing the Load-following Performance of Microtur-
bines and Fuel Cells,” Electric Power Systems Research,
Vol. 62, No. 1, May 2002, pp. 1-11.
[14] J. Padullés, G. W. Ault and J. R. McDonald, “An Inte-
grated SOFC Plant Dynamic Model for Power Systems
Simulation,” Journal of Power Sources, Vol. 86, No. 1-2,
March 2000, pp. 495- 500.
[15] N. Hatziagyriou, et al., “Modeling of Micro-sources for
Security Studies,” Proceedings of the CIGRÉ 2004 Con-
ference, Paris, August 30-September 3, 2004.
[16] P. Kundur, “Power Systems Stability and Control,”
McGraw-Hill, New York, 1994.