Improved Virtual Synchronous Generator Control to Analyse and Enhance the Transient Stability of Microgrid

IET Research Journals
Submission Template for IET Research Journal Papers
Improved Virtual Synchronous Generator
Control to Analyse and Enhance the
Transient Stability of Microgrid
ISSN 1751-8644
doi: 0000000000
www.ietdl.org
Khalid Mehmood Cheema1∗, Kashif Mehmood1
1
School of Electrical Engineering, Southeast University, Nanjing 210096, China
* E-mail: kmcheema@seu.edu.cn
Abstract: In recent years, the integration of renewable energy resource (RES) into the power system is growing rapidly, and
it is necessary to analyse and evaluate the effect of RES on transient stability of the power system. In this paper, centre of
inertia (COI) concept is implemented to analyse and evaluate the integration effects of an auxiliary damping control (ADC) based
virtual synchronous generator (VSG) consisting an improved governor. The impact of VSG integration is divided into synchronous
generator (SG) linked parts and COI associated parts. Due to VSG integration into the power system, the significant elements
which disturb the COI dynamic motion and rotor dynamics of SG are examined in detail. Different cases are considered to evaluate
the effectiveness of the proposed method, i.e., VSG’s different integrating location and different power capacities. It is observed
in simulation results that COI dynamic motion and rotor dynamics of SG are positively affected by VSG integration and transient
stability improves significantly.
1 Introduction
Nowadays, with the development of prediction, operation and con-
trol techniques, the penetration level of RES, such as photovoltaic
and wind power, is steadily growing, which means numerous SGs
are being swapped by power electronic devices [1]. The large-scale
RES units cause to reduce the system inertia and have adverse
effects on power system stability, moreover, RES power generation
is random and their grid connected characteristics are different than
conventional SGs [2, 3]. Therefore, it is of great theoretical and prac-
tical value to analyse the influence of RES integration into power
system stability. Few different methods are used to develop RES
based distributed generators (DGs), and their integration into the
power system came up with the concept of microgrid [4]. A micro-
grid is an important form of connecting DGs to the grid.
Generally, a microgrid consists of DGs, energy storage devices and
utility loads linked as grid connected mode or as islanded mode [5].
Microgrids are highly vulnerable to disturbances, therefore, regulat-
ing the frequency oscillations and transient system stability became
the main point of interest in microgrid function [6, 7]. For instance,
the voltage magnitude and frequency oscillations are processed by
the main grid in grid-connected mode. Whereas, when connected in
islanded mode, voltage magnitude and frequency oscillations must
be upheld within specified limits. Upon any disturbance and abrupt
change in the power system, the traditional SG can supply the poten-
tial kinetic energy stored in its rotating parts to the grid [8]. On
the other hand, the DGs consisting photovoltaic systems, there is no
rotating part of inertial response and it can participate in frequency
support by adding virtual inertia via electronic inverters [9, 10]. The
DGs which are electronically interfaced with grid, demonstrate dif-
ferent characteristics than conventional power generating units. In
electronically interfaced DGs, the power generated at primary side
of DGs is regulated by electronic inverters, but they aren’t able to
supply the required inertia and damping to the power grid.
Consequently, electronically interfaced DGs cannot improve the sys-
tem stability. However, the solution of this problem is developed
by applying the suitable control techniques to the grid-connected
inverter and controlling its switching pattern so that it works as an
SG by mimicking the behaviour of SG. The grid-connected inverters
which mimic the steady-state and transient characteristics of SG are
called VSGs. It is predicted that VSG integrated systems will be the
future of power system network [11]. Therefore, it is utmost impor-
tant to analyse and improve the transient stability of VSG based
power systems.
The basic idea of VSG with different approaches is presented in [12-
14], whereas several techniques are demonstrated in the literature to
improve the stability [15-19]. Mostly, these techniques are restricted
to specified scenarios, i.e. grid-connected mode. For instance, in
[20], active power and reactive power independent controls are dis-
cussed by analysing the power coupling mechanism of VSG in
medium- and low-voltage microgrids, but analysis for high-voltage
microgrid and transient stability are not discussed in it. Additionally,
adaptive inertia strategies for VSG with different control schemes are
presented in [21, 22].
The techniques proposed in [23] decouples the active and reactive
power and suppress the dependency of Q-δ, P-V. A mechanism for
frequency, voltage and active-reactive power flow control of VSG
is illustrated in [24]. Moreover, the proposed scheme in [25, 26]
used the coordinate control technique to supply the virtual iner-
tia for VSG to analyse and enhance the transient and steady state
performance of microgrid. It is discussed in [27] that the transient
response of the system can also be enhanced by using energy stor-
age devices with VSG. However, the stability technique based on
storage devices utilisation is ineffective and expensive for an exten-
sive power system. The two-step control technique is presented in
[8, 28], which suppress frequency and power oscillations by fulfill-
ing the criteria of Lyapunov-stability. In [29], an auxiliary damping
controller for a grid-connected VSG is proposed to minimise the
frequency deviation and improve the system stability. However, the
proposed technique is demonstrated on a single machine infinite bus
scenario whereas results for multi-machine scenario postponed for
future research. The effect of the VSG on the transient response of a
microgrid is presented in [6].
Furthermore, a COI based control technique is proposed in [30] to
analyse and improve the transient stability of the system without
considering any RES unit. In this technique, the rotor angles and
frequencies of all generators track the angle and frequency of the
dynamic COI of the system to keep synchronised under significant
disturbances. The similar COI concept is used to analyse the tran-
sient stability of the power system integrated with the inertia less
wind farm [31], but no mechanism is presented to enhance system
stability.
IET Research Journals, pp. 1–11
c The Institution of Engineering and Technology 2015 1
Auto-generated PDF by ReView IET Renewable Power Generation
IET-Submission-DoubleColumn-Template.pdf MainDocument IET Review Copy Only 2
This article has been accepted for publication in a future issue of this journal, but has not been fully edited.
Content may change prior to final publication in an issue of the journal. To cite the paper please use the doi provided on the Digital Library page.

Fig. 1: Simple VSG Structure
However, in this paper the detail implementation of COI concept in
hybrid power system consisting of multiple SGs and a VSG with
ADC and improved governor is carried out. The VSG with ADC
is implemented in [29]. The VSG is modified in a way that it has
the controlled input power which minimises the power fluctuations
and frequency oscillations relative to the COI frequency. More pre-
cisely, this paper investigates an improved VSG control to achieve
stability during transient state while connected with other SGs. The
proportional and resonant control gain at input of VSG improve the
ability of VSG to stabilize the transient dynamic of power system.
The comparative analysis is performed to validate the proposed mod-
ification and its effect on transient dynamics of power system. The
rest of the paper is organised as follows. Section 2 describes the
basic VSG structure and load sharing between MI-VSG and SGs.
Section 3 explains the centre of inertia concept and section 4 demon-
strate the proposed modification in VSG control. Section 5 discusses
the transient stability effects on the system due to MI-VSG integra-
tion. Section 6 describes the integration scenarios and discuss the
simulation results and section 7 concludes the paper.
2 VSG and Improved governor
The basic structure of a VSG is explained here, followed by the
improved governor of VSG.
2.1 VSG Structure
A basic and rather simple structure of VSG is shown in Figure 1.,
and it can be observed that VSG consist of a DG unit, energy stor-
age device, DC/AC converter, filter circuit and grid. If the power
of the distributed generator and energy storage system is assumed
as the input torque of the prime mover, while DC/AC converter is
assumed as the electromechanical energy transformer between sta-
tor and rotor. Then, the fundamental component of midpoint voltage
is representing the electromotive force of the VSG. Resistance and
inductance of filter unit representing the stator winding impedance.
The simple swing equation of SG is used as a core part of VSG, and
it is stated as:
Pin − Pout = 2I
dωv
dt
− D(ωv − ωg) ωv (1)
where ωg is grid frequency, ωv is virtual angular frequency, Pin
is inverter input power, Pout is output electrical power of inverter,
D is virtual damping coefficient, and I is inertia constant. In order
to keep the VSG speed equal to the grid frequency, virtual damping
coefficient plays an important role.
2.2 Improved Governor Control
The basic and improved governor of VSG control is illustrated in
Figure 3 part (a) and (b) respectively. In comparison to the gover-
nor of conventional SG, the governor unit of typical VSG control
is somewhat different and backward because the absence of iner-
tia in it. The diverse characteristics for power regulation of SG and
VSG causes massive power fluctuation when both generators are
Fig. 2: Governor Control of VSG (a) Simple Governor (b) Improved
Governor
connected in parallel. To compete with the governor unit of the
conventional SG, additional inertia is incorporated into the VSG
governor unit. Mathematically the simple yet improved governor
control can be stated as:
Tm =
kg(ωref − ω)(kpgs + kig)
(s + kt)s
(2)
where kt is the time constant, kg is the gain, kpg is the propor-
tional control gain and kig shows the integral control gain of the
governor unit. The (2) shows that speed control unit and first order
inertia is the part of proposed improved governor control. The pro-
portional and integral controller act as the speed controller and track
the angular frequency reference signal. Adding the supplementary
inertia into the governor block, the VSG with ADC consist two
first order inertia supplies, one is in VSG rotor motion control and
other is in improved governor control. This supplementary inertia
improved the stability of VSG with ADC in comparison to the con-
ventional and simple VSG. The simple transfer function of rotor
motion and governor block of improved VSG with ADC can be
stated as following by considering the inertia link only.
G =
kg
(s + kt)(Is + D)
(3)
Moreover, the simple transfer function of rotor motion and speed
governor of conventional and simple VSG can be stated as:
G =
1
Is + D
(4)
The values of poles can be calculated by (3) and (4). Using suit-
able parameters, the poles values are calculated and concluded that
the poles value of (3) is far less then the poles value of (4). There-
fore, it is determined that the stability of (3) is much better then (4).
3 Centre of Inertia
In a hybrid power system containing VSG with ADC and SGs, the
angle of the system and COI of the frequency are stated as the
aggregated average of VSG with ADC and SGs as the following
equations:
δC =
1
It
Iiδi (5)
ωC =
1
It
Iiωi (6)
where
It = Ii (7)
where ωC is the rotor speed of COI, δC is the rotor angle of COI,
ωi is rotor speed, δi is the rotor angle, and Ii is inertia constant of
2 c The Institution of Engineering and Technology 2015

i-th generator, and It is inertia constant of the system. Moreover, the
COI motion equation can be stated as:
σ(ωC) =
1
It
PC (8)
where σ is the differential algorithm and PC is expressed as:
PC = PC,m − PC,e (9)
where PC,m and PC,e is the input mechanical power and output
electrical power of COI respectively. The dynamics of VSG with
ADC or SGs for COI reference frame can be described as following
by considering (5) and (6)
δC
i = δi − δC (10)
ωC
i = ωi − ωC (11)
where the δC
i is the rotor angle and ωC
i rotor speed of i-th gener-
ator based on COI reference frame. The equation of dynamic motion
of individual SG based on COI reference frame can be expressed as
by considering (11)
σ(ωC
i ) = σ(ωi) − σ(ωC) (12)
σ(ωC
i ) =
1
Ii
(Pi,m − Pi,e) −
1
It
(PC,m − PC,e) (13)
where Pi,m and Pi,e are the mechanical power and electrical
power of i-th generator, respectively.
4 Proposed Modification in VSG
The detailed structure of VSG with ADC is illustrated in Figure 3.,
including proposed proportional and resonant controller gain and
improved governor. The output power and frequency of VSG with
ADC is calculated at its output terminals by current and voltage sig-
nals. The governor block is replaced with an improved governor by
adding a proportional and integral controller and supplying addi-
tional inertia in order to adjusts the input power command relative
to the frequency deviation. The electromechanical swing equation
can be solved by numerical integration after evaluating the nec-
essary parameters. Whereas, the mechanical phase δ is generated
by calculating and passing the rotational ω from an integrator for
every control cycle. The Vref controller provides the regulated reac-
tive power and voltage at VSG with ADC output terminal. For the
inverter, the PWM pulses are generated by output voltage angle and
magnitude of VSG with ADC. Moreover, the proposed technique is
used to minimise the VSG with ADC oscillations and power varia-
tions by controlling the input power of VSG with ADC by applying
suitable PR values and improved governor parameters. Hence the
modified input VSG with ADC is called as MI-VSG in this paper
in order to differentiate it from simple VSG. The proposed system
model is shown in a blue dotted block in Figure 3. The Ppr is an
auxiliary input power into VSG with ADC, and it can be stated as
follow:
Ppr = Gp(ω − ωC) + Gr(ω − ωC) (14)
where Gp and Gr is proportional controller gain and resonant
controller gain, respectively and Gr = krωcuts
s2+ωcuts+ω2 . ωC is the
COI frequency, and it can be calculated by by rotor speed equation.
The proportional resonant controller significantly enhances the
damping ratio of the system, whereas, improved governor strength-
ens the inertia of the angular frequency and minimise the difference
between VSG governor and SG’s governor. Moreover, the effect of
PR controller is significantly better than typical proportional integral
controller or proportional integral and derivative controller due to its
dynamic controlling characteristics.
Fig. 3: Detail Structure of Proposed Modified Input VSG
5 Transient Stability Analysis with MI-VSG
Integration
When a disturbance occurs in the power system, traditional SG
accelerates and decelerate according to the difference in input
mechanical power and output electrical power. The difference
between grid frequency and rotor speed of generator causes to
change rotor angle of generator followed by changes in electrical
power. The rotor speed, output electrical power and grid frequency
oscillations can be minimised by SG inertia. However, in MI-VSG
construction, the virtual rotor angle is considered as an angle of
MI-VSG terminal voltage. The armature resistance and synchronous
reactance have no distinct value in MI-VSG. Therefore, during a dis-
turbance in the power system, the output electrical power can be
regulated easily and rapidly. The rotor angle deviation and relative
swing concept cannot be used to describe the relation of MI-VSG
and SG because the power angle feature is not available in integrated
MI-VSG. However, it is proposed that for a specified inertia value,
the MI-VSG unit calculate the angle of COI as the reference which
is comparable to voltage angle deviation in SGs. Considering (5) to
(13), it is comprehended that the COI concept is centred on the power
angle features of the SGs and mutual synchronisation process among
SGs. Consequently, in this paper, the effects of MI-VSG integration
on COI dynamics and SGs regarding COI coordinate are analysed
according to COI.
5.1 MI-VSG Integration Effects on COI Transient Dynamics
The dynamic motion equation of COI can be stated as following after
integrating the MI-VSG in the system.
σ(ωC,V ) =
1
It,V
(PC,m,V − PC,e,V ) (15)
where ωC,V , It,V , PC,m,V and PC,e,V are rotor speed of COI,
inertia constant of the system, mechanical power of COI and electri-
cal power of COI, respectively, while MI-VSG is integrated into the
system.
Considering (15), it is realised that by integrating MI-VSG, the
motions of system COI are affected by variations in mechanical
power, electrical power and COI inertia constant. In order to know
the effects of MI-VSG integration on COI dynamics, it is necessary
to study the MI-VSG effects on mechanical power, electrical power
and COI inertia constant.
5.1.1 Effect on COI Inertia Constant: It is known that MI-
VSG mimics the swing equation of SG and possess inertia, but the
amount of MI-VSG inertia is varied for similar power MI-VSG and

SG. Therefore, when MI-VSG is installed after removing the SG, the
COI inertia constant can be stated as:
It,V = It − IR (16)
where IR is the inertia constant of SG which has been removed.
5.1.2 Effect on Mechanical Power: the governor in SG is
incapable of controlling the mechanical power of an SG during
transient state. However, the improved governor in a MI-VSG act
quickly and regulate the mechanical power rapidly because it does
not possess any rotating mechanism. The mechanical power of
MI-VSG can be described as:
PC,m − PC,ref =
ω − ωref
kp
(17)
where PC,m is the mechanical power of COI, PC,ref is the
reference power of COI and kp is the droop coefficient.
5.1.3 Effect on Electrical Power: In an electrical system, the
output power of an SG is time variant and depends on its working
state. However, the aggregated sum of output electric power from
all generators is equal to the dynamic power consumption. There-
fore, COI electric power and electrical power of COI with MI-VSG
integration can be expressed as:
PC,e = P (18)
PC,e,V = PV − PV,e (19)
where P, PC,e,V , PV and PV,e are the total power consumed,
electrical power of COI with MI-VSG integration, total consumed
power with MI-VSG integration and electrical power of MI-VSG,
respectively.
The dynamic power consumption is depending on voltage stability
in the system and after MI-VSG integration, if the effect of dynamic
transient voltage is minor and negligible, then the dynamic power
consumption considered similar to the pre-MI-VSG integrated state.
It can be described as:
PV = P (20)
Considering (18)-(19), it is recognized that the electrical output
of integrated MI-VSG is the main element to affect the dynamics
of electrical power of COI with MI-VSG integration. Due to the
different excitation system, the dynamic response of MI-VSG’s elec-
trical power is varying from SG’s dynamic response. The dynamic
response of integrated MI-VSG electric output will be used to find
the difference between electrical power of COI and MI-VSG inte-
grated electrical power of COI when it is known that the MI-VSG
integration does not affect the dynamic response of the system.
Moreover, in the transient state, the mechanical power of COI
remains the same. Meanwhile, the inertia constant of COI remains
unaffected during dynamic operation because it is an inherent char-
acteristic of the power system. However, the electrical power of COI
affected during the transient state. The specific MI-VSG integrating
location and condition determines the effects on mechanical power,
electrical power and COI inertia constant as well as COI dynamics.
5.2 MI-VSG Integration Effect on SG
The transient trajectories of every SG, as well as rotor angle states
and rotor motion dynamics, are affected by MI-VSG integration.
Considering the COI reference frame, the dynamic motion rep-
resents the kinetic energy of rotor regarding the overall system
variation, while rotor angle of each SG represents the variations in
potential energy of rotor regarding the dynamic centre of the system.
Rendering COI with MI-VSG integration, the dynamic motion of
i-th generator is expressed as:
σ(ωC
i,V ) =
1
Ii
(Pi,m,V − Pi,e,V ) −
1
It,V
(PC,m,V − PC,e,V )
(21)
The variation in i-th generator dynamic motion can be expressed
as follows by considering (11) and (21)
σ(ωC
i,V ) − σ(ωC
i ) =
1
Ii
[(Pi,m − Pi,e) − (Pi,m,V − Pi,e,V )]
+ [
1
It
(PC,m − PC,e) −
1
It,V
(PC,m,V − PC,e,V )]
(22)
Considering (22), the variations in transient dynamics of SG and
COI’s motion are used to measure the effect of MI-VSG integra-
tion on the dynamics of individual SG regarding COI. The variations
in input mechanical power and output electrical power affect the
dynamic motion of SG. The power flow of the system is changed
with MI-VSG integration because the pre-fault input mechanical
power and output electrical power of SG is different from input
mechanical power and output electrical power of SG with MI-VSG
integration. Due to MI-VSG integration into the power system, the
pre-fault input mechanical power and output electrical power of SG
shifted to new operation position to fulfil the required power bal-
ance. During a fault, the dynamic response of output electrical power
of each SG is very different, because the dynamic response of inte-
grated MI-VSG is different from SG.
The rotor angle of each SG, relative to COI with integrated MI-VSG
is stated as:
δC
i,V = δi,V − δC,V (23)
The SG operating states can be realized by the changes in rotor
angle positions of each SG. Generally, at pre-fault state, an SG pos-
sess the lesser value of rotor angle, if it’s output electrical power is
shared with integrated MI-VSG. During a fault, the rotor angle of i-
th generator changes according to the inertia constant of SG, power
angle and dynamic response of output electrical power of SG.
The dynamic response of rotor angle and rotor speed with integrated
MI-VSG are affected by the change in accelerating power of each
SG. Therefore, to validate the influence of integrated MI-VSG on the
transient dynamics of each SG is variations regarding COI dynamic
motion, simulations are performed.
6 Simulation Analysis and its Review
For simulation analysis, the IEEE 9 bus power system is imple-
mented as illustrated in Figure 4. The implemented SG model
consists of a governor unit and excitation controller for each SG.
To suppress the effects of the load’s dynamic feature, the fixed value
impedances are employed as power loads. Three cases are estab-
lished to carry out the analysis of the proposed system of Figure 4
and parameters data is given in Table 1 and Table 2 for MI-VSGs and
SGs, respectively. Moreover, the comprehended variables for estab-
lished cases are selected as the capacity of MI-VSG, access location
and removal of a specified SG.
Moreover, the total power capacity of the system remains the
same with and without MI-VSG integration. Furthermore, the fault
is occurred at bus 7 at time 10 second and continue till 10.1 seconds.
Table 1 Parameters for 3 Generators
Parameter SG1 SG2 SG3
Rated power 8.5 MW 10 MW 33.5 MW
Rated Voltage 13.8 kV 16 kV 18 kV
Inertia 1.8 2.1 3.2
Rated pf 0.85 0.9 0.87

A reference case is developed, by integrating a 10 MW VSG at bus
5 without replacing any SG, to compare the results of MI-VSG.
6.1 Effects on COI’s Transient Stability
In this section, the effects on COI’s transient stability are analysed
after MI-VSG integration into the power system.
6.1.1 Case 1: For this case, the proposed MI-VSG is integrated
at bus 5 parallel to SG2 without replacing any SG. The capacity
of MI-VSG is adjusted at 10 MW and 35 MW. Considering prior
explained integrated MI-VSG structure, the variation in dynamic
voltage stability of the system and changes in the structure of the
network is according to (20). Moreover, the changes in COI sys-
tem inertia, COI mechanical power and COI output electrical power
while MI-VSG is integrated into the network, are expressed as:
It,V = It
PC,m,V = PC,m − PV,m
PC,e,V = PC,e − PV,e
(24)
With the MI-VSG integration, the variations in accelerating speed
of COI can be stated as:
σ(ωC,V ) − σ(ωC) =
(PC,m,V − PC,e,V )
It,V
−
(PC,m − PC,e)
It
=
(PV,e − PV,m)
It
(25)
Considering (25), when MI-VSG is integrated without remov-
ing any SG, the transient power response of integrated MI-VSG
inﬂuences the dynamic motion of COI. The accelerating power and
electrical power response of MI-VSG are illustrated in Figure 5.
according to their capacities. In Figure 5., it is depicted that when
Fig. 4: 9 Bus Power System with integrated VSG
Table 2 Parameters for MI-VSG with Improved Governor
Parameters 10 MW MI-VSG 35 MW MI-VSG
Voltage 20 kV 20 kV
Resistance 0.02 p.u 0.06 p.u
Impedance 0.36 p.u. 0.57 p.u
Inertia 0.28 0.71
Proportional gain 23 30
Resonant gain 13 21
Governor Gp 14.5 19
Governor Gi 7 13
the capacity of MI-VSG increases, then absolute power increases
too during and after the fault.
Fig. 5: Transient Dynamics (a) MI-VSG Active Power (b) MI-VSG
Accelerating Power
Whereas, the absolute power of COI for integrated MI-VSG
decreases with increase in capacity and it is illustrated in Figure
6. Considering (25), it is observed that with the large capacity
of integrated MI-VSG, the rotor speed of COI shows slow speed
acceleration and deacceleration during and after fault conditions,
respectively, as depicted in Figure 5(a).
6.1.2 Case 2: In order to determine the effects on the dynamic
motion of COI, the MI-VSG is integrated onto different locations
in the power system. For instance, the 10 MW MI-VSG is inte-
grated at bus 4, 5, 9 without removing any SG. The accelerating
power and output voltage of integrated MI-VSG are shown in Figure
7. It is illustrated in Figure 7(a). that during fault and experiencing
the same system disturbances, the dynamic response of output volt-
age of integrated MI-VSG changes according to integrating location
of MI-VSG into the system. Moreover, Figure 7(b). depicts that the
dynamic output voltage affects the output electrical power.
The transient dynamic response of rotor speed of COI and accel-
erating power of COI are shown in Figure 8(a). and Figure 8(b). The
MI-VSG’s accelerating output power dynamics changes due to the
dynamic response of the accelerating power of COI. However, the
rotor speed of COI changes with variations in the accelerating power
of COI.
6.1.3 Case 3: The SG behaviour is examined to understand the
effects on transient dynamics of COI by realising the COI inertia
constant variations. In this case, SG2 is exchanged by MI-VSG
having equivalent active and reactive power. According to this new

Fig. 6: Transient Dynamics (a) COI Accelerating Power (b) COI
Rotor Speed
setup, the total dynamic power consumption remains the same. Con-
sidering the (24) is it concluded that inertia constant of COI changes
only and it can be stated as:
It,V = It, − ISG2
(26)
Variations in accelerating power of COI and inertia constant of
COI disturbs the dynamic motion of COI, and it can be described as:
σ(ωC,V ) =
1
It − ISG2
(PC,m − PC,e − ∆PV ) (27)
The dynamic response of the accelerating power of COI is illus-
trated in Figure 9(a)., which shows that during and after fault, the
total value of the accelerating power of COI deviates significantly, if
comparing with the reference case. Therefore, the rotor speed of COI
response is shown in Figure 9(b)., by comparing with the reference
case, it is observed that during and after fault, rotor speed of COI
changes slightly. The possible reason for this minor effect is lower
inertia constant of COI.
6.1.4 Review: Considering the above simulations, it indicates
that MI-VSG shared the total dynamic output electrical power, which
causes the accelerating power of COI to deviate toward downside
during and after the fault. However, the critical reasons for this
deviation are MI-VSG’s capacity, its integrating location and char-
acteristics of power response. Moreover, dynamic motion of COI is
affected by accelerating power of COI when MI-VSG is integrated
without removing any SG, whereas inertia constant of COI affected
the dynamic motion of COI and accelerating power of COI when
MI-VSG is integrated by replacing SG2.
Fig. 7: Transient Dynamics (a) MI-VSG Output Voltage (b) MI-
VSG Accelerating Power
6.2 Effects on SG Transient Stability
6.2.1 Case 1: In this case, the effects on the transient stability
of individual SG is observed. The comparison of each SG with the
rotor speed of COI of 10 MW and 35 MW MI-VSG is represented
in Figure 10. As previously observed in Figure 6(b). that by integrat-
ing MI-VSG into the system, the accelerating speed of rotor of COI
drops. However, the impact on rotor speed of COI for each SG is
not positive entirely. Comparing with the reference case of 10 MW
VSG without proposed modification, the rotor speed of SG1, SG2
and SG3 experience significantly lower fluctuations regarding COI.
But by comparing each SG with other, the SG1 experience quite low
fluctuations, whereas the rotor speed fluctuations of SG2 and SG3
are significantly greater as illustrated in Figure 10.
6.2.2 Case 2: In this case, the MI-VSG is integrated at three
different buses, and the dynamics of rotor speed of individual SG
regarding COI are observed as illustrated in Figure 11. Observing
Figure 8(b)., it is noted that different integrating locations of VSGs
cause the variations in accelerating speed of COI, however, the effect
on rotor speed of each SG regarding COI is not even.
In Figure 11., in comparison with the reference case, when MI-VSG
is integrated at bus 4, it experiences fewer variations for SG1, SG2,
and SG3 regarding COI. However, when MI-VSG is integrated at
bus 5 it goes through the higher variations for all SG1, SG2, and
SG3 regarding COI. Moreover, when MI-VSG is placed at bus 9, in
comparison to the reference case, it experiences lower variations in
SG1 and SG3 while higher variation in SG2 regarding COI.
6.2.3 Case 3: In this case, MI-VSG is integrated by remov-
ing SG2 and dynamics of rotor speed regarding COI of SG1 and

Rotor Speed
SG3 are observed by comparing these with the reference case, as
illustrated in Figure 12. It is observed in Figure 12(a). that rotor
speed of SG1 regarding COI fluctuates higher in comparison with
the reference case. However, the rotor speed of SG3 regarding COI
fluctuates significantly lower in comparison with the reference case.
Moreover, it can be observed that the settling time for oscillations of
SG1 and SG3 are lower than the reference case which depicts that
power system stabilizes soon after experiencing the disturbance.
6.2.4 Review: Analysing the above three cases, it is observed
that in SG1, SG2, and SG3, the impact on rotor dynamics are not
equal. For example, if the rotor speed of COI drops, then it is not
mandatory that rotor speed of each SG regarding COI drops too.
Considering the effect of VSG’s integrating location, the dynamic
transient trajectories of SG1, SG2, and SG3 affect slightly because
the rotor dynamic of SGs is reasonably subtle to variations in power-
sharing of the entire system. Though the effect on the electrical
power of COI is insignificant, it shows that distribution of rotor angle
values and relative swings among SGs are dropped when the loca-
tion of VSG integration is close to the SG that work under the steady
state rotor angle.
6.3 Simulation Analysis for Complex System
To analyze and validate the effects of proposed modification on the
individual SG with respect to COI in complex systems, the simula-
tions are performed for IEEE 30 bus two area system. The IEEE 30
bus two area system is shown in following Figure 13. This system
consists of a total six SGs, three SGs in area 1 and three SGs in area
2. The parameters of SG1 and SG4, SG2 and SG6, SG3 and SG5
of IEEE 30 bus two area system are similar to SG1, SG2 and SG3
Rotor Speed
of Table 1, respectively. The reference case is developed in each area
by integrating a 10 MW VSG at bus 22 for area 1 and at bus 23 for
area 2, without replacing any SG, in order to compare the results of
MI-VSG integration in area 1 and area 2, respectively.
For area 1, MI-VSG is integrated at bus 22 by removing the SG3
and analyze the effects on SG1 and SG2 of area 1 and SG4, SG5
and SG6 of area 2. Subsequently, for area 2, MI-VSG is integrated
on bus 23 by replacing the SG5 and observe the effects on SG4
and SG6 of area 2 and SG1, SG2 and SG3 of area 1. For area 1,
when replacing SG3 with MI-VSG and comparing the result with
reference case, it is observed that the oscillations of rotor speed of
SG1 and SG2 are significantly lower than reference case as shown
in Figure 14., whereas the positive effect on MI-VSG integration on
the oscillations of rotor speed of SG4, SG5 and SG6 in area 2 is
relatively less than as in area 1, as illustrated in Figure 15.
For area 2, the SG5 is replaced with MI-VSG and compared the
result with reference case, it is observed that the oscillations of rotor
speed of SG4 and SG6 is significantly lower than the reference case
as depicted in Figure 16., while the oscillations of rotor speed of
SG1, SG2 and SG3 are marginally less than the reference case
as shown in Figure 17. In other words, the positive effect of MI-
VSG integration on the rotor speed of SG4 and SG6 in area 2 is
significantly improved while for area 2 it is slightly better than the
reference case.
6.3.1 Review: Considering the simulation results of the 30 bus
two area system, it is observed that individual SG affect positively
by MI-VSG integration. When MI-VSG is integrated into area 1 by
replacing SG1 at bus 22, the rotor speed oscillations of remaining
SGs in the same area are significantly lower in comparison with the
reference case, whereas, in area 2, the rotor speed oscillations are

Fig. 10: Rotor Speed Dynamic Regarding COI of (a) SG1 (b) SG2 and (c) SG3
Fig. 11: Rotor Speed Dynamic Regarding COI of (a) SG1 (b) SG2 and (c) SG3
Fig. 12: Rotor Speed Dynamic Regarding COI of (a) SG1 (b) SG3
marginally better than the reference case. This case is vice versa
when MI-VSG integration into area 2 at bus 23 by replacing SG2.
7 Conclusion
The effects of MI-VSG integration on COI’s dynamic motion,
dynamics of rotor angle and rotor speed of each SG regarding COI
are examined qualitatively. An investigation based on comparison
using the MI-VSG capacity, its integrating location and replacement
of SG with MI-VSG is carried out in simple IEEE 9 bus system and
a complex IEEE 30 bus two area system to validate the analysis. Fur-
thermore, it is concluded in this research that
1. Different integrating locations of MI-VSG suppress the irregular
power distribution of the system, which minimise the effect on rotor
speed and rotor angle of each SG regarding COI.
2. When MI-VSG is integrated into system without removing any
SG, the acceleration power of COI is the only factor which affects the
dynamic motion of COI and with suitable parameters adjustments it
implies to improve stability.
3. Power system experience relatively smooth transition with respect
to reference case when an SG replace with MI-VSG. By this prac-
tice, the transient dynamics of COI are influenced and remaining
SG’s experience relatively lower fluctuations than reference case.
4. In complex two area system, the positive effect of MI-VSG inte-
gration is significant in the area where it is integrated, however, the
effect is marginally improved in the other area while comparing with
reference case.
However, the parameter optimization and optimal integrating loca-
tion of MI-VSG will be carried out in future research.

Fig. 13: 30 Bus Two Area Power System
Fig. 14: Area 1, Rotor Speed Dynamic Regarding COI of (a) SG1 (b) SG2
Fig. 15: Area 2, Rotor Speed Dynamic Regarding COI of (a) SG4 (b) SG5 and (c) SG6

Fig. 16: Area 2, Rotor Speed Dynamic Regarding COI of (a) SG4 (b) SG6
Fig. 17: Area 1, Rotor Speed Dynamic Regarding COI of (a) SG1 (b) SG2 and (c) SG3
References
[1] S. Wen, Y. Wang, Y. Tang, Y. Xu, and P. Li.: ’Proactive fre-
quency control based on ultra-short-term power ﬂuctuation fore-
casting for high renewables penetrated power systems’, IET Renew.
Power Gener., 2019, 13, pp. 2166-2173
[2] W. Liu, R. Ge, Q. Lv, H. Li, and J. Ge.: ’Research on a small
signal stability region boundary model of the interconnected power
system with large-scale wind power’, Energies, 2015, 8, pp. 2312-
2336
[3] R. Shah, N. Mithulananthan, R. Bansal, and V. K. Ramachan-
daramurthy.: ’A review of key power system stability challenges for
large-scale PV integration’, Renew. Sustain. Energy Rev., 2015, 41,
pp. 1423-1436
[4] A. Hirsch, Y. Parag, and J. Guerrero.: ’Microgrids: A review of
technologies, key drivers, and outstanding issues’, Renew. Sustain.
Energy Rev., 2018 90, pp. 402-411
[5] R. Majumder.: ’Some aspects of stability in microgrids’, IEEE
Trans.Power Syst., 2013, 28, pp. 3243-3252
[6] N. Soni, S. Doolla, and M. C. Chandorkar.: ’Improvement of
transient response in microgrids using virtual inertia’, IEEE Trans.
Power Deliv., 2013, 28, pp. 1830-1838
[7] R. Majumder, A. Ghosh, G. Ledwich, and F. Zare.: ’Power man-
agement and power ﬂow control with back-to-back converters in a
utility connected microgrid’, IEEE Trans. Power Syst., 2009, 25, pp.
821-834
[8] J. Alipoor, Y. Miura, and T. Ise.: ’Power system stabilization
using virtual synchronous generator with alternating moment of iner-
tia’, IEEE J. Emerg. Select. Topics Power Electron., 2014 3, pp.
451-458

[9] M. A. El-Hameed, M. M. Elkholy, and A. A. El-Fergany.: ’Efﬁ-
cient frequency regulation in highly penetrated power systems by
renewable energy sources using stochastic fractal optimiser’, IET
Renew. Power Gen., 2019
[10] S. Liao, J. Xu, Y. Sun, Y. Bao, and B. Tang.: ’Wide-area
measurement system-based online calculation method of PV sys-
tems de-loaded margin for frequency regulation in isolated power
systems’, IET Renew. Power Gen., 2017, 12, pp. 335-341
[11] M. Kezunovic, V. Vittal, S. Meliopoulos, and T. Mount.: ’The
big picture: Smart research for large-scale integrated smart grid
solutions’, IEEE Power Energy Mag., 2012, 10, pp. 22-34
[12] Q.C Zhong, G. Weiss.: ’Synchronverters: Inverters that mimic
synchronous generators’, IEEE Trans. Ind. Electron., 2011, 58, pp.
1259-1267
[13] H. Wu, X. Ruan, D. Yang, X. Chen, W. Zhao, Z. Lv, et
al.: ’Small-signal modeling and parameters design for virtual syn-
chronous generators’, IEEE Trans. Ind. Elec., 2016, 63, pp. 4292-
4303
[14] T. Shintai, Y. Miura, and T. Ise.: ’Oscillation damping of a
distributed generator using a virtual synchronous generator’, IEEE
Trans. Power Deliv., 2014, 29, pp. 668-676
[15] A. K. Alaboudy, H. H. Zeineldin, and J. Kirtley.: ’Micro-
grid stability characterization subsequent to fault-triggered islanding
incidents’, IEEE Trans. Power Deliv., 2012, 27, pp. 658-669
[16] A. M. Azmy and I. Erlich.: ’Impact of distributed generation
on the stability of electrical power system’. IEEE Power Engineering
Society General Meeting, 2005, pp. 1056-1063
[17] F. Katiraei, M. R. Iravani, and P. W. Lehn.: ’Micro-grid
autonomous operation during and subsequent to islanding process’,
IEEE Trans. Power Deliv., 2005, 20, pp. 248-257
[18] L. Meegahapola and D. Flynn.: ’Impact on transient and fre-
quency stability for a power system at very high wind penetra-
tion’,IEEE PES General Meeting, 2010, pp. 1-8
[19] A. K. Srivastava, A. A. Kumar, and N. N. Schulz.: ’Impact of
distributed generations with energy storage devices on the electric
grid’, IEEE Syst. J., 2012, 6, pp. 110-117
[20] B. Li, L. Zhou, X. Yu, C. Zheng, and J. Liu.: ’Improved power
decoupling control strategy based on virtual synchronous generator’,
IET Power Electron., 2016, 10, pp. 462-470
[21] J. Li, B. Wen, and H. Wang.: ’Adaptive virtual inertia con-
trol strategy of VSG for micro-grid based on improved bang-bang
control strategy’, IEEE Access, 2019, 7, pp. 39509-39514
[22] T. Kerdphol, M. Watanabe, K. Hongesombut, and Y. Mitani.:
’Self-adaptive virtual inertia control-based fuzzy logic to improve
frequency stability of microgrid with high renewable penetration’,
IEEE Access, 2019, 7, pp. 76071-76083
[23] S. D’Arco, J. A. Suul, and O. B. Fosso.: ’A virtual synchronous
machine implementation for distributed control of power converters
in smartgrids’, Electric Power Syst. Research, 2015, 122, pp. 180-
197
[24] Y. Xiang-Zhen, S. Jian-hui, D. Ming, L. Jin-wei, and D. Yan.:
’Control strategy for virtual synchronous generator in microgrid’,
4th International Conference on Electric Utility Deregulation and
Restructuring and Power Technologies (DRPT), 2011, pp. 1633-
1637
[25] K. Shi, H. Ye, W. Song, and G. Zhou.: ’Virtual inertia con-
trol strategy in microgrid based on virtual synchronous generator
technology’, IEEE Access, 2018, 6, pp. 27949-27957
[26] G. Magdy, G. Shabib, A. A. Elbaset, and Y. Mitani.: ’A novel
coordination scheme of virtual inertia control and digital protection
for microgrid dynamic security considering high renewable energy
penetration’, IET Renew. Power Gen., 2018, 13, pp. 462-474
[27] E. Hammad, A. Farraj, and D. Kundur.: ’On effective virtual
inertia of storage-based distributed control for transient stability’,
IEEE Trans. Smart Grid, 2017
[28] P. Hu, H. Chen, K. Cao, Y. Hu, D. Kai, L. Chen, et al.:
’Coordinated control of multiple virtual synchronous generators in
mitigating power oscillation’, Energies, 2018, 11, p. 2788
[29] B. Gao, C. Xia, N. Chen, K. Cheema, L. Yang, and C. Li.: ’Vir-
tual synchronous generator based auxiliary damping control design
for the power system with renewable generation’, Energies, 2017,
10, p. 1146
[30] Z. Lan, D. Gan, L. Shi, and Y. Ni.: ’A study on the control of
AC/DC power systems based on system dynamic COI’, IEEE Power
Engineering Society General Meeting, 2007, pp. 1-5
[31] S. Liu, G. Li, and M. Zhou.: ’Power system transient stability
analysis with integration of DFIGs based on center of inertia’, CSEE
J. of Power Energy Syst., 2016, 2, pp. 20-29

Improved Virtual Synchronous Generator Control to Analyse and Enhance the Transient Stability of Microgrid

Recommended

More Related Content

What's hot (19)

Similar to Improved Virtual Synchronous Generator Control to Analyse and Enhance the Transient Stability of Microgrid (20)

More from Kashif Mehmood (11)

Recently uploaded (20)

Improved Virtual Synchronous Generator Control to Analyse and Enhance the Transient Stability of Microgrid