Electric Power Systems Research 241 (2025) 111396

Contents lists available at ScienceDirect

Electric Power Systems Research

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

Dual-side virtual inertia control for stability enhancement in islanded

AC-DC hybrid microgrids
Kai Shi , Yang Bai , Peifeng Xu *, Yuxin Sun
School of Electrical and Information Engineering, Jiangsu University, No. 301, Xuefu Road, Jingkou District, Zhenjiang, 212013, Jiangsu, China

A R T I C L E I N F O A B S T R A C T

Keywords: With the increasing penetration of renewable energy sources in modern power systems, the reduction in system
Dual-side Virtual Inertia inertia has emerged as a critical challenge, compromising power smoothing and frequency regulation in hybrid
Virtual Inductors AC-DC microgrids operating in islanded mode. To address these issues, a dual-side virtual inertia control strategy
Virtual Inertia
for bi-directional interface converter (BIC) is proposed, which ensures balanced power exchange between the AC
BIC
Inertial Support
and DC transmission grids through transient response by incorporating a virtual inductor into the virtual syn-
chronous machine (VSM) framework. The proposed method enhances system stability and dynamic performance
by stabilizing both AC frequency and DC voltage under load fluctuations. A small-signal model is developed to
analyze the impact of virtual inertia parameters on system stability, with the root locus method applied to
evaluate performance. Simulations in Simulink validate the effectiveness of the control strategy, demonstrating
significant improvements in transient response and frequency regulation. This approach provides excellent
power sharing and enhanced resilience in hybrid microgrids with a high degree of integration of renewable
energy sources, offering promising applications for future power systems.

1. Introduction In conventional power systems, large synchronous generators (SGs)

provide essential inertia. However, renewable energy sources typically
The increasing integration of renewable energy into the grids helps consist of a large number of power electronics devices with no inertia
address power shortages. As a key component of distributed energy [14], which will lead to grid instability during localized load fluctua-
systems, microgrids support local renewable energy consumption due to tions. Accordingly, virtual inertia and energy storage technologies are
their small size and flexibility [1–5]. Hybrid microgrids, which leverage critical for stabilizing AC frequency and DC voltage [15,16]. An adaptive
both AC and DC grids, provide enhanced power distribution capabilities virtual inertia control scheme in [17] optimizes VSG damping and
[6]. Therefore, as the core interconnect component of between AC and inertia coefficients to maintain frequency stability under varying con-
DC transmission grids, the development of BIC control strategies is ditions, while [18] integrates virtual DC motor (VDCM) technology into
essential for enhancing the stability and optimizing the operating per- energy storage regulation, improving DC voltage stability and miti-
formance of hybrid AC-DC power systems [7]. gating fluctuations.
Extensive research on BIC control has promoted various strategies to However, these VSM controls or VSM-like controls only consider
improve power flow coordination and stability. Among these de- separate subgrids and do not take into account the coupling relationship
velopments, bidirectional droop control strategies regulate the scalar- between the AC and DC subgrids. Considering that both AC and DC sides
ized difference between AC frequency and DC voltage for coordinated require inertial support when impacted [19]. As discussed in [20], the
power flow [8–11]. Additionally, adaptive droop control mechanisms integration of additional energy storage systems can provide inertia
have been introduced to compensate for voltage and frequency de- support on both sides, though it may result in increased costs. Inte-
viations under varying load conditions [12,13]. grating VSM control into the DC-side loop, as in [21], adjusts the DC bus
By adopting these BIC control strategies, hybrid microgrids facilitate voltage to provide inertia. While effective, this approach focuses on the
effective power sharing while significantly improving energy utilization DC side and may not fully support inertia on the AC side, particularly
and system efficiency. during significant AC load disturbances, potentially limiting system

* Corresponding author.
E-mail address: [email protected] (P. Xu).

https://doi.org/10.1016/j.epsr.2024.111396
Received 17 October 2024; Received in revised form 6 December 2024; Accepted 22 December 2024
Available online 2 January 2025
0378-7796/© 2024 Published by Elsevier B.V.
K. Shi et al. Electric Power Systems Research 241 (2025) 111396

stability. Bidirectional virtual inertia control in hybrid microgrids is system through appropriate control strategies.
explored in [22] and [23]. While [22] focuses on the DC side, [23]
implements virtual inertia on both the AC and DC sides. Although
effective, the additional control strategies required may complicate the 2.2. Operating modes of AC-DC hybrid microgrid
system. Both [24] and [25] focus on incorporating virtual capacitors
into the DC-side or interface converter control loop to provide inertia Islanded operation in AC-DC hybrid microgrids is inherently more
support for both subgrids. Although these methods effectively address complex and challenging to stabilize compared to grid-connected mode.
the inertia issue. Few studies have explored the role of virtual induc- This is primarily because, in islanded mode, the microgrid must inde-
tance in bi-directional interface converters. By implementing virtual pendently manage power balance, frequency, and voltage without the
inductors to emulate the rotor dynamics, a comparable virtual inertia support of an external grid. To maintain stability, the microgrid relies on
effect can be achieved on the DC side, facilitating efficient power local generation, energy storage, and load regulation, and any fluctua-
transfer and inertia compensation between the AC and DC subgrids. tion in load or energy sources can lead to system instability. In this
In this paper, a dual-side virtual inertia control strategy is proposed mode, droop control is employed to enable power distribution between
for BIC based on VSM control. This strategy is designed to provide in- AC and DC subsystems, while distributed energy resources without
ertial support for both AC and DC sides, addressing transient distur- stable outputs are operated under constant power control, effectively
bances. An improved active power loop is derived using virtual inertia functioning as loads. BIC can operate in two modes, rectification and
and virtual inductance principles, enhancing the system’s response. The inversion.
key contributions and innovations of this study include:
3. Dual-side virtual inertia control of BIC
(1) The control strategy stabilizes AC frequency by using virtual
inertia to absorb and release energy, counteracting fluctuations BICs in hybrid microgrids accurately simulate and regulates the dy-
and improving stability. namic characteristics of a finite-capacity microgrid (MG) through VSM
(2) Virtual inductance on the DC side smooths voltage variations, control, with the active power control [26] presented in Eq. (1):
absorbs power fluctuations, and ensures steady power flow be- dωvsm
tween AC and DC sides. Jvsm ωvsm = Pm − Pe − DJ (ωvsm − ωn ) (1)
dt
(3) An enhanced active power loop integrates virtual inertia and
inductance, stabilizing AC frequency and DC voltage, and where Pm and Pe are mechanical power and electromagnetic power, Jvsm
improving robustness against power disturbances. and DJ are virtual rotational inertia and damping coefficient, and ωvsm
and ωn are the angular frequency and rated angular frequency of the
The paper is organized as follows: a background analysis of the VSM, respectively, JVSM, the inertia, reflects the system ability to reserve
proposed control strategy is provided in Section 1. Section 2 details the adequate frequency regulation margins and prevent instability. And the
AC/DC hybrid microgrid’s system configuration and operating Modes. damping coefficient DJ quantifies the power output response to varia-
Section 3 introduces the dual-side virtual inertia control strategy and its tions in frequency.
mechanism for BIC. Section 4 assesses stability through small-signal On the AC side, the dynamic response is improved through the
modeling and root locus analysis. Section 5 presents simulation results implementation of VSM control in BIC. During load transients that result
validating the control strategy’s effectiveness. Finally, Section 6 sum- in frequency deviations, the VSM plays a pivotal role in maintaining
marizes the findings and concludes the paper. frequency stability, while power transferred from the DC side contrib-
utes to frequency restoration. To achieve this, both the DC transmission
2. Configuration and operating modes of AC-DC hybrid grid and the BICs function as virtual inertia sources, providing essential
microgrid support to stabilize the AC bus frequency. That is:
dωvsm
2.1. Configuration of AC-DC hybrid microgrid Jvsm ωvsm = Pset − Pac − DJ (ωvsm − ωn ) (2)
dt

The topology of the AC-DC hybrid microgrid is illustrated in Fig.1, where Pset and Pac are respectively the active power reference value and
where distributed energy resources (DERs) are connected to both AC and the transmission power of the BIC as the AC bus frequency changes.
DC buses via power conversion devices, achieving flexible energy con- Frequency regulation is managed by emulating the mechanical dy-
version and distribution. As the core interface between the AC and DC namics of SG, with the virtual inertia of the BIC deriving its frequency
transmission grids, the BIC plays a crucial role in ensuring efficient support from the energy stored in the DC transmission grid. By applying
power sharing and maintaining the stable operation of the AC-DC hybrid the properties of derivatives, it can be concluded that:

Fig. 1. Typical topology of hybrid microgrid.

2
K. Shi et al. Electric Power Systems Research 241 (2025) 111396

( )
d ωvsm − ω2n
2 4. Closed-Loop system stability analysis
Jvsm
Pac = Pset − − DJ (ωvsm − ωn ) (3)
2 dt 4.1. Small-signal modeling
Similarly, in the DC transmission network, adequate inertia supple-
ment can mitigate the transient voltage fluctuation effect caused by load Small perturbations are introduced to both the input and output of
changes on the DC bus. The small capacity of capacitors connected in the active power loop, followed by a small-signal decomposition of Eqs.
parallel on the DC side typically fails to provide a significant source of (3), (6), and (7).
inertia. Therefore, virtual inductance introduced on the DC side of the [( ) ( ∧ )]
BIC simulates the damping effect of inductive elements that counteract dω dIdc
∧
∧ ∧
Pref = Gp Jω + DJ ω − LIdc + DL Idc (9)
∧

sudden current changes in AC systems, with energy stored as: dt dt

1 By applying the superposition theorem and the Laplace transform,
WL = Ldc Idcn
2
(4)
2 the closed-loop transfer function for the active power loop is derived, as
shown in Eq. (10).
where WL is the energy stored in the virtual inductor Ldc at the rated DC
⎧
current Idcn. As the inductance increases, the stored energy WL grows, ⎪
⎪
reducing the impact of load disturbances on the DC current. During ⎪
⎪
⎨ ∧ ∧
Pref Pref
sudden load changes, the inertia required for stabilizing the DC current = G p (J ωs + DJ ) = − Gp (LIdc s + DL ) (10)
⎪ ∧ ∧
comes from the BIC, which transmits energy stored in the AC. Thus, both ⎪ ω
⎪
⎪
⎩ Idc
the AC transmission grid and BIC can be regarded as VSMs that support
the DC bus current. Compared to the virtual inertia in the AC as shown in The Laplace-transformed mathematical model of the BIC in the
Eq. (2), the virtual inductance equation for the DC bus current can be synchronously rotating d-q reference frame is:
expressed as: [ ] [ ][ ] [ ]
ed Ls + R − ωL id v
dIdc = + d (11)
Ldc Idc = Pset − Pdc − DL (Idc − Idcn ) (5) eq ωL Ls + R iq vq
dt
ωc 3 ( )
where Ldc and DL are respectively the virtual inductance and damping P= ed id + eq iq (12)
s + ωc 2
factor, and Pdc is the active power output as fluctuations occur in the DC
bus current. By controlling the DC side of the BIC using the virtual where the subscripts d and q respectively represent the d-axis and q-axis
inductance equation, the abrupt changes in DC voltage during load components, s is the differential operator, and ωc is the rotor angular
transients in the DC transmission grid can be indirectly minimized. the frequency of the low-pass filter. Assuming that the d-axis in the dq-
subsequent simplification result of (5) is showed: coordinate system coincides with the a-axis in the three-phase station-
( 2 ) ary coordinate frame, then |ed| = |ea|. According to voltage-oriented
Ldc d Idc − Idcn
2
Pdc = Pset − − DL (Idc − Idcn ) (6) control, the q-axis component of the electromotive force vector eq is
2 dt
set to 0.
For an AC-DC hybrid microgrid, the power transferred by the BIC The coupling relationship between the d-axis and q-axis is shown in
from the AC to the DC must equal the power transferred from the DC to Eq. (11). By facilitating feedforward decoupling control to minimize
the AC side. This implies the following conditions must be met: coupling effects, and employing a PI controller to precisely track the
ΔPac = ΔPdc (7) current, the control equations for vd and vq are given as follows:
( )
From Eq. (7), the expression for power balance in the AC-DC subnet
⎧ KiI ( )
⎪
⎪ v d = − KiP + idref − id + ωLiq + ed
under steady-state conditions is derived as: ⎨ s
( ) (13)
( ) ( 2 ) ⎪ KiI ( )
J d ωac 2 − ω2acn L d Idc − Idcn
2 ⎪
⎩ vq = − KiP + iqref − iq − ωLid + eq
+ DJ (ωvsm − ωn ) = + DL (Idc − Idcn ) (8) s
2 dt 2 dt
Control block diagram of BIC can be derived from Eq. (8), as shown where KiP and KiI are the proportional and integral gains of the current
in Fig.2. inner loop, and idref and iqref are respectively the command values for id
and iq. Substitute Eq. (13) into Eq. (11):

Fig. 2. dual-side virtual inertia control of BIC.

3
K. Shi et al. Electric Power Systems Research 241 (2025) 111396

{ ( )
Gi (s)( idref − id ) = (R + Ls)id axis, whereas the black line indicates that the negative real root is
(14)
Gi (s) iqref − iq = (R + Ls)iq approaching the imaginary axis. Consequently, the regulation and
transition times will extend as the negative real root approaches the
where Gi(s) is expressed by Eq. (15) imaginary axis.
In Fig. 4(b), the root trajectory is shown as the damping coefficient
KiI
Gi (s) = KiP + (15) DJ increases from 10 to 1000, where J is 1. As DJ increases, the root
s
trajectory of system gradually shifts towards the real axis, reflecting
where the state variables id and iq from Eq. (15) can be expressed as the enhanced damping and a smoother response. Specifically, DJ governs
sum of steady-state components and small perturbations. As shown in the system’s damping characteristics: for small values of DJ , the system
Eq. (16). exhibits underdamped behavior, marked by higher overshoot and a
Substitute Eq. (17) into Eq. (15), neglect the second-order pertur- quicker response. In contrast, larger values of DJ lead to overdamped
bations, and perform Laplace transform. The small-signal model of the behavior, which reduces overshoot but results in a slower response.
current loop can be denoted in Eq. (17). As the damping increases, with the increase of DJ, one negative real
root approaches the imaginary axis, as shown by the red line in Fig. 4(c),
{
prolonging regulation time. Similarly, when DL is increased from 10 to
∧
id = Id + Id
(16)
iq = Iq + Iq
∧
1000, and with the virtual inductor value set to 2mH, the resulting root
trajectory reveals the formation of an unequal pair of negative real roots,
{ ∧ ( ∧ ∧ )/ and the system enters an overdamped state. Fig. 4(d) illustrates the root
id = Gi (s) idref − id (R + Ls) trajectory as the virtual inductance Ldc is varied from 0.01mH to 2mH.
( ∧ ∧ )/ (17)
∧
iq = Gi (s) iqref − iq (R + Ls) Thus, during system parameters design, it is crucial to ensure the system
maintains adequate anti-disturbance capability, achieves the optimal
To further refine the model, small perturbations are systematically balance between stability margins and dynamic response speed, and
applied to both the inputs and outputs of Eq. (12). where, if eq is set to be provides fast and accurate regulation for stable operation, with the
0, then outer loop can be derived as follows: damping coefficient set to 50 in this process.

∧ ωc 3 ( ∧ )
ed id + + ed id (18)
∧
P= 4.3. Dynamic response of AC frequency and DC current
s + ωc 2
According to the superposition theorem and by applying the Laplace For analytical simplicity, it is assumed that there is only one DG in
transform to the system equations, it can be obtained that: both the AC and DC subnets, and the integral delay constant is neglected.
⎧ Based on the P-f and P-U droop control characteristics, the dynamics
⎪
⎪
⎪
⎪
⎨∧ 3 ωe ∧ transmission power of the BIC in the AC-DC subnets is governed by the
P c d P 3 ωc id relationship shown in Eq. (21).
= = G a (s) = = Gb (s) (19)
⎪ ∧ 2 s + ωc ed 2 s + ωc
∧
⎪
⎪ id {
⎪
⎩ Δω = − kω (ΔPac + ΔP)
(21)
ΔIdc = ΔP/( − ku (ΔPdc − ΔP))
The small-signal model of the bi-directional virtual inertia control
strategy is established based on Eqs. (10), (17), (18), (19), as depicted in where Δω and ΔIdc are the deviations in AC frequency and DC current,
Fig. 3. where the closed-loop transfer function of the BIC is: respectively, Kω and Ku are the corresponding droop coefficients, ΔP is
the transmission power of the BIC, and ΔPAC and ΔPDC are respectively

⎧
⎪
⎪
⎪
⎪∧ ∧
⎨ GPl (s)Gi (s)Gl (s)Gp (s)(Jωs + DJ ) P GPl (s)Gi (s)Gl (s)Gp (s)(Ldc Idc s + DL )
P
= =− (20)
⎪ ∧ Ls + R + G (s) + G (s)G (s)G (s) ∧ Ls + R + Gi (s) + GPI (s)Gi (s)Gl (s)
⎪ω
⎪
⎪
i PI i l
Idc
⎩

4.2. Root locus analysis of system stability the load perturbations of AC and DC subnets, respectively. Combined
with Eqs. (20), (21), the dynamic responses of the AC frequency and DC
Compared to droop control, the response speed of the inner-loop current are sequentially expanded when the active power of the AC-DC
power current control of the BIC in the AC-DC subnetwork exhibits a hybrid microgrid fluctuates, that is:
significantly faster, the power current transfer function is approximated ⎧
⎪ ΔPac
as 1. To ensure system stability, all characteristic roots are located in the ⎪
⎪ Δω =
⎨ (1/kω ) + G1 (s)(Jωs + DJ )/(Ls + R + 1 + G1 (s))
left half-plane. Progressively increasing the key parameters, the corre- (22)
sponding root trajectories can be derived from Eq. (20), based on the ⎪
⎪
⎪ ΔI =
⎩
ΔPdc
model shown in Fig. 3, as shown in Fig. 4. (1/kn ) + (Gl (s)(Ldc Idc s + DL ))/(Ls + R + 1 + Gl (s))
Fig. 4(a) presents the stability variations as the rotational inertia J is Unilateral active power transmission provides inertia for both the AC
incremented from 0.1 kg⋅m² to 10 kg⋅m², with the value of DJ set to 50. It frequency and DC voltage to support the slow recovery to steady-state
can be seen that as J increases, the poles gradually transition from values, thus the dynamic response has some degree of inertia. The pa-
conjugate complex roots to two distinct negative real roots, and the rameters J, DJ, DL, and Ldc determine the steady-state deviations of AC
system progressively transitions from an underdamped to an over- frequency and DC current, reflecting the droop characteristics of the
damped state, tending towards stability with longer response times and microgrid.
reduction in overshoot and oscillations. As J increases, the red line
represents that the negative real root is far away from the imaginary

4
K. Shi et al. Electric Power Systems Research 241 (2025) 111396

Fig. 3. Small-signal model of BIC dual-side virtual inertia control.

Fig. 4. Root trajectory diagram of closed-loop system.

5. Simulation verification the BIC operates in rectifier mode. From 4 s to 6 s, a 0.1 MW load is
applied to the AC side, the power demand is greater than that on the DC
To assess the impact of optimization on system stability and dynamic side, and the BIC operate in inverter mode. As illustrated in Fig. 5. when
response, an islanded AC-DC hybrid microgrid model is developed on both the AC and DC sides experience a 0.1 MW load increase, 0.05MW of
the Simulink platform. The performance differences in load fluctuation power is supplied to each side, with the BIC simultaneously flows
suppression are compared and analyzed between dual-side virtual 0.05MW of power to the demand side.
inertia control, conventional droop control, and Virtual Synchronous
Machine (VSM) control. The AC-DC hybrid microgrid consists of a DC Table 1
microgrid, a bidirectional interface converter, and an AC microgrid. The Simulation parameters design.
power generation unit of the DC microgrid, with a generation capacity of Subsystems Parameters Value
1.167 MW, is simulated using a P-U droop controller applied to the DC-
DC microgrid DC voltage rating Udcn (V) 700
side converter, and its voltage output is stabilized at 700 V, as shown on Active power - DC voltage droop factor Ku (MW/V) 0.0333
the left side of Fig. 1. AC microgrid generation is simulated using a P-f DC current rating Idcn (A) 0
droop-controlled AC converter, with a generation capacity of 2 MW, as BIC Filter Inductors L (H) 3e-3
depicted on the right side of Fig. 1. The simulation parameters are Resistive R (Ω) 1e-4
AC Filter Capacitors C(mF) 0.5
provided in Table 1. Damping Factor DJ 100
Virtual Damping Factor Dl 50
Virtual inductor Ldc (mH) 0.5
5.1. Bidirectional power verification Virtual inertia J (kg⋅m²) 1
AC microgrid AC phase voltage rating Uacn (V) 380
Between 1 s and 3 s, a 0.1 MW load is added to the DC side, and the AC angular frequency rating ωacn (rad/s) 314
power demand on the DC side is greater than that on the AC side. Active power - frequency droop coefficient Kω (MW/ 0.318
(rad/s))
Active power is transmitted from the AC subnet to the DC subnet, and

5
K. Shi et al. Electric Power Systems Research 241 (2025) 111396

Fig. 5. Schematic diagram of BIC power waveform in multi-mode.

gradually increases to the steady-state value at around 0.15s. Without

compromising the steady-state performance of the system, the larger the
J, the slower the response speed. Given the lower power conversion rate
of the BIC, the system primarily relies on the self-regulation of the AC
bus frequency for effective buffering and response to AC load distur-
bances. Simultaneously, the introduction of Ldc results in a similar slow
recovery trend in the dynamic response of the DC voltage. According to
the analysis, J and Ldc have no impact on the steady-state value of the
microgrid system.
Similarly, the microgrid operates with a 0.1 MW DC load at 1s. The
simulation results of the comparison of DC voltage and AC frequency for
double droop control, VSM control, and dual-side virtual inertia control
Fig. 6. AC frequency fluctuation with different control strategies for 0.1 MW with different virtual inductance are shown in Figs. 8 and 9. It is evident
AC load connection. that as Ldc increases, the DC voltage variation becomes smoother when
load fluctuations occur in the DC subnet. When Ldc is set to zero, the DC
voltage stabilizes more rapidly, enabling the DC subnet to provide
increased transient power but with no inertia support for AC frequency
and DC voltage. Thus, appropriate virtual inductance parameters
significantly enhance the inertia of the hybrid microgrid under BIC
control. Nevertheless, the final steady-state values remain unchanged,
indicating that Ldc only affects the system response speed without
changing the steady-state value.
Under dual-side virtual inertia control, a sudden increase in DC load
power primarily disturbs the DC bus voltage, causing a temporary de-
viation from its nominal value. The provided inertia support helps to
dampen these fluctuations, allowing the voltage to gradually return to
its steady-state value. This process ensures a smoother transition and
enhances the stability of the DC transmission grids during transient
disturbances.
Fig. 7. DC voltage fluctuations of different control strategies under 0.1 MW AC
load connection.

5.2. Unilateral load power surge

In islanding mode, the microgrid operates with a 0.1 MW AC load at

1s. The comparison results of AC frequency and DC voltage for double
droop control, VSM control, and dual-side virtual inertia control with
varying J is shown in Figs. 6 and 7. During the dynamic process, the
maximum fluctuations in AC frequency and DC voltage are approxi-
mately 0.05 Hz and 3 V, respectively. While the double droop control
exhibits rapid response to both AC frequency and DC voltage during a
sudden increase in AC load, it fails to provide sufficient inertia to miti-
gate the AC frequency drop. VSM control can supply inertia support for
the AC side of the AC-DC hybrid microgrid but lacks this capability for
the DC side. In contrast, dual-side virtual inertia control offers inertia
support for both AC and DC sides. However, due to the adoption of Ldc
and J, the priority of AC frequency response decreases, and then Fig. 8. DC voltage fluctuation of different control strategies under 0.1 MW DC
load connection.

6
K. Shi et al. Electric Power Systems Research 241 (2025) 111396

Fig. 9. AC frequency of different control strategies under 0.1 MW DC Fig. 13. AC frequency waveforms with different virtual inertia.
load connection.
Similarly, at t = 1s, a sudden increase of 0.1 MW load is introduced
on the AC side, leading to an immediate need for frequency support.
As shown in Figs. 12 and 13, the DC side acts as a crucial inertia
source, providing the necessary power to stabilize the frequency devi-
ation caused by the load change. This power transfer helps mitigate the
transient response on the AC side, maintaining system stability.
As the inertia on the AC side increases, the required compensation
from the DC side becomes more substantial, indicating a direct corre-
lation between the inertia levels on the AC side and the energy demand
for stabilization on the DC side.

6. Conclusion

The virtual inertia enables rapid response to perturbations,

Fig. 10. DC voltage waveforms with different virtual inductors.
mimicking the behavior of conventional generators. The dual-side vir-
tual inertia control strategy for BIC based on VSM, as proposed in this
paper, has been validated through stability analysis via small-signal
modeling. It supports bidirectional power flow, ensures efficient
power distribution between the AC and DC sides, and mitigates fre-
quency fluctuations and voltage dips during sudden power increases.
Comparative studies show the impact of varying virtual inertia on the
dynamic response of AC frequency and DC voltage, ultimately
enhancing the stability of the hybrid AC-DC microgrid. Future work
could focus on optimizing the control strategy for high-power transient
disturbances, integrating renewable energy and energy storage, and
assessing the scalability of this strategy in larger or interconnected grids.

CRediT authorship contribution statement

Fig. 11. AC frequency waveforms of different virtual inductors.
Kai Shi: Writing – review & editing, Supervision, Conceptualization.
Yang Bai: Software, Resources, Methodology, Data curation. Peifeng
Xu: Writing – review & editing, Project administration, Methodology,
Conceptualization. Yuxin Sun: Writing – review & editing, Project
administration, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial

interests or personal relationships that could have appeared to influence
the work reported in this paper.

Data availability
Fig. 12. DC voltage waveforms with different virtual inertia.
The data that has been used is confidential.
5.3. 0.1 MW load impact with varying J and Ldc in AC-DC sides
References
When the virtual inertia J is set to 1, a sudden increase of 0.1 MW [1] J. Hu, X. Liu, M. Shahidehpour, S. Xia, Optimal operation of energy hubs with
load occurs on the DC side at 1s. According to the analysis of Figs. 10 and large-scale distributed energy resources for distribution network congestion
11, the AC side serves as the inertia source for the DC side, which re- management, IEEe Trans. Sustain. Energy 12 (3) (2021) 1755–1765. July.
[2] D. Zhao, H. Wang, J. Huang, X. Lin, Storage or no storage: duopoly competition
duces the frequency response by 0.1% Hz maintaining the stability of the between renewable energy suppliers in a local energy market, IEEE J. Sel. Areas
AC side, thereby enhancing the dynamic response of the DC side. Commun. 38 (1) (2020) 31–47. Jan.

7
K. Shi et al. Electric Power Systems Research 241 (2025) 111396

[3] N.L. Díaz, A.C. Luna, J.C. Vasquez, J.M. Guerrero, Centralized control architecture [15] B. She, F. Li, H. Cui, J. Wang, Q. Zhang, R. Bo, Virtual Inertia Scheduling (VIS) for
for coordination of distributed renewable generation and energy storage in Real-time economic dispatch of IBR-Penetrated Power Systems, IEEe Trans.
islanded AC microgrids, IEEe Trans. Power. Electron. 32 (7) (2017) 5202–5213. Sustain. Energy 15 (2) (2024) 938–951. April.
July. [16] J. Pahasa, P. Potejana, I. Ngamroo, MPC-Based Virtual Energy Storage System
[4] N.T. Mbungu, R.M. Naidoo, R.C. Bansal, V. Vahidinasab, Overview of the optimal Using PV and Air Conditioner to Emulate Virtual Inertia and Frequency Regulation
smart energy coordination for microgrid applications, IEEe Access. 7 (2019) of the Low-Inertia Microgrid, IEEe Access. 10 (2022) 133708–133719.
163063–163084. [17] P. Utkarsha, N.K.S. Naidu, B. Sivaprasad, K.A. Singh, A Flexible Virtual Inertia and
[5] Y. Han, K. Zhang, H. Li, E.A.A. Coelho, J.M. Guerrero, MAS-Based distributed Damping Control Strategy for Virtual Synchronous Generator for Effective
coordinated control and optimization in microgrid and microgrid clusters: a Utilization of Energy Storage, IEEe Access. 11 (2023) 124068–124080.
comprehensive overview, IEEe Trans. Power. Electron. 33 (8) (2018) 6488–6508. [18] D. Chen, Y. Xu, A.Q. Huang, Integration of DC Microgrids as Virtual Synchronous
Aug. Machines into the AC Grid, IEEE Trans. Ind. Electron. 64 (9) (2017) 7455–7466.
[6] V.M. Garrido-Arévalo, W. Gil-González, O.D. Montoya, L.F. Grisales-Noreña, J. Sept.
C. Hernández, Optimal dispatch of DERs and battery-based ess in distribution grids [19] K. Shi, Y. Chen, P. Xu, Y. Sun, M. Liu, X. Wang, A Generalized DC electromotive
while considering reactive power capabilities and uncertainties: a second-order force control strategy for Bi-directional interface converters in a standalone hybrid
cone programming formulation, IEEe Access. 12 (2024) 48497–48510. microgrid, IEEE Trans. Energy Convers. 38 (2) (2023) 1155–1165. June.
[7] I. Ponce, F. Milano, Modeling hybrid AC/DC power systems with the complex [20] G. Magdy, A. Bakeer, M. Nour, E Petlenkov, A New Virtual Synchronous Generator
frequency concept, IEEE Trans. Power Syst. 39 (4) (2024) 6004–6013. July. Design Based on the SMES System for Frequency Stability of Low-Inertia Power
[8] S. Peyghami, H. Mokhtari, F. Blaabjerg, Autonomous Operation of a Hybrid AC/DC Grids, Energies. (Basel) 13 (2020) 5641.
Microgrid With Multiple Interlinking Converters, IEEE Trans Smart Grid 9 (6) [21] A. Bakeer, A. Chub, A. Abid, S.A. Zaid, T.A.H. Alghamdi, H.S Salama, Enhancing
(2018) 6480–6488. Nov. Grid-Forming Converters Control in Hybrid AC/DC Microgrids Using Bidirectional
[9] M. Ahmed, L. Meegahapola, M. Datta, A. Vahidnia, A Novel Hybrid AC/DC Virtual Inertia Support, Processes 12 (1) (2024) 139.
Microgrid Architecture with a Central Energy Storage System, IEEE Transactions [22] Z. Zhang, J. Fang, C. Dong, C. Jin, Y. Tang, Enhanced grid frequency and DC-Link
on Power Delivery 37 (3) (2022) 2060–2070. voltage regulation in hybrid AC/DC microgrids through bidirectional virtual
[10] Q. Xu, J. Xiao, P. Wang, C. Wen, A decentralized control strategy for economic inertia support, IEEE Trans. Ind. Electron. 70 (7) (2023) 6931–6940. July.
operation of autonomous AC, DC, and hybrid AC/DC MGs, IEEE Trans. Energy [23] Ziwen Liu, et al., Improved power flow control strategy of the Hybrid AC/DC
Convers 32 (4) (2017) 1345–1355. Dec. Microgrid Based on VSM, IET Gener. Transm. Distrib. 13 (1) (2019) 81–91.
[11] S.M. Malik, Y. Sun, W. Huang, X. Ai, Z. Shuai, A generalized droop strategy for [24] Dan Liu, et al., Bidirectional virtual inertia control strategy for hybrid distributed
interlinking converter in a standalone hybrid microgrid, Appl. Energy 226 (2018) generations integrated distribution systems, Math. Probl. Eng., Hindawi (2022)
1056–1063. 1–10, pagesNovember.
[12] Y. Fu, Z. Zhang, Y. Mi, Z. Li, F. Li, Droop control for DC multi-microgrids based on [25] K. Yin, et al., A new grid side inertia support control method for cascaded power
local adaptive fuzzy approach and global power allocation correction, IEEE Trans converters in Bi-directional EV Chargers, in: 2023 3rd Power System and Green
Smart Grid 10 (5) (2019) 5468–5478. Sept. Energy Conference (PSGEC), Shanghai, China, 2023, pp. 473–478.
[13] B. Fan, et al., A novel droop control strategy of reactive power sharing based on [26] W. Wu, et al., A virtual inertia control strategy for DC microgrids analogized with
adaptive virtual impedance in microgrids, IEEE Trans. Ind. Electron. 69 (11) virtual synchronous machines, IEEE Trans. Ind. Electron. 64 (7) (2017)
(2022) 11335–11347. Nov. 6005–6016. July.
[14] R.J. Best, P.V. Brogan, D.J. Morrow, Power system inertia estimation using HVDC
power perturbations, IEEE Trans. Power Syst. 36 (3) (2021) 1890–1899. May.