Received 15 December 2022; revised 22 January 2023; accepted 23 January 2023.

Date of publication 26 January 2023;

date of current version 17 February 2023. The review of this paper was arranged by Associate Editor Jorge Solsona.
Digital Object Identifier 10.1109/OJIES.2023.3240070

Hierarchical Model-Predictive Droop Control

for Voltage and Frequency Restoration
in AC Microgrids
IRESHA POONAHELA 1,5 (Student Member, IEEE), ABDELBASSET KRAMA 2 (Member, IEEE),
SERTAC BAYHAN 2 (Senior Member, IEEE), UGUR FESLI 3 ,
MOHAMMAD B. SHADMAND 4 (Senior Member, IEEE), HAITHAM ABU-RUB 1 (Fellow, IEEE),
AND MIROSLAV M. BEGOVIC5 (Life Fellow, IEEE)
1
Department of Electrical and Computer Engineering, Texas A&M University at Qatar, Doha 23874, Qatar
2
Qatar Environment and Energy Research Institute, Hamad Bin Khalifa University, Doha 34110, Qatar
3
Department of Electronics and Automation, Technical Sciences Vocational School, Gazi University, 06560 Ankara, Türkiye
4
Department of Electrical and Computer Engineering, University of Illinois Chicago, Chicago, IL 60607 USA
5
Department of Electrical and Computer Engineering, Texas A&M University, College Station, TX 77840 USA
CORRESPONDING AUTHOR: IRESHA POONAHELA (e-mail: [email protected]).
This work was supported by the Qatar National Research Fund (a member of Qatar Foundation) under Grants NPRP12C-33905-SP-213 and
NPRP12C-33905-SP-220.

ABSTRACT The hierarchical control structure was introduced to allow the integration of power-electronics-
based distributed generation into the microgrid in a smart and flexible manner. The main aim of the primary
controller in such a structure is to achieve accurate active and reactive power sharing, whereas the secondary
control aims to ensure voltage and frequency (V/f) stability. Generally, converter-level secondary controllers
utilize classical nested loop control that suffer from a slow dynamic response and cumbersome parameter
tuning. The existing-model-based and estimation-based secondary controllers are fast, but require complex
design methodology, high communication bandwidth, and, consequently, higher data analysis and computa-
tional burden. This article presents a simple predictive-based secondary control for the ac microgrid that is
fast and robust and has a low design complexity, low communication bandwidth, and no parameter tuning
requirement in the secondary control layer. The proposed predictive control optimally restores voltage and
frequency in the microgrid by predicting their trajectory deviations and leveraging the droop characteristic
curves. Experimental tests performed with three parallel-connected grid-forming inverters in an islanded
operation validate that the controller can accurately maintain V/f stability, while ensuring active and reactive
power sharing.

INDEX TERMS Finite-control-set model-predictive control (FCS-MPC), grid-forming (GFM) inverters,

inverse droop control, islanded ac microgrid, primary control, secondary control (SC).

I. INTRODUCTION interactive, intelligent, and efficient power system

With the global community shifting toward the usage of more paradigm [1]. In the case of any disruption to the utility
renewable power, the inclusion of a more diverse supply of grid, the MG should disconnect itself and continue to
energy to the utility grid is inevitable. With that regard, the regulate its own voltage and manage its loads autonomously
microgrid (MG) is a geographically confined smart grid that in the islanded mode of operation. It is also vital that
aids such diverse power integration with smaller controllable the transition between the two modes occurs seamlessly
elements known as distributed generators (DGs). The MG without disruption to power consumers [2], [3]. During the
allows users to actively manage not only their energy con- absence of the bulky rotating generators of the main utility
sumption but also their generation, leading to a more reliable, grid, the MG faces challenges due to the lack of inertia in

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 4, 2023 85
POONAHELA ET AL.: HIERARCHICAL MODEL-PREDICTIVE DROOP CONTROL FOR VOLTAGE AND FREQUENCY RESTORATION IN AC MGS

inverter-based resources. Virtual inertia emulation based on the event detection scheme made the system less reliable,
on virtual synchronous generators and the use of power similar to the control proposed in [19]. Therefore, in [14], a
reserve are some of the techniques that have been used small ac signal (SACS) SC approach was presented, which
to compensate for the lack of inertia in MGs [4], [5]. In does not rely on event detection. Furthermore, it enabled
addition, the islanded renewable-energy-based MG faces PI-SC to be implemented without affecting the power sharing
stability, reliability, and resilience issues due to the sporadic accuracy between GFM units. PI control was implemented
characteristic of renewable energy sources and the lack of to achieve frequency control, while an ac current signal was
stable power generation units such as gas-powered plants [6]. extracted from the voltage to produce the reference voltage
The inverter-based distributed energy resources act as with a fixed amplitude and variable frequency. However,
current-controlled inverters in grid-following mode. This hap- the use of PI controllers posed the tedious task of tuning
pens when the DGs are connected to the grid, where the output numerous variables and resulted in a slower restorative
current of the inverter is regulated. Inverters are operated performance compared to nonlinear controllers.
as grid forming (GFM) when they behave as voltage source The existing literature on SC based on nonlinear control
while regulating the output voltage of the inverter. The parallel techniques shows that the estimation-based and model-based
connection of multiple GFM inverters under the decentralized techniques have been explored. A consensus-based approach
control scheme is established in this article, where each of was used in [20] where the reactive power of the system
the GFM inverters is regulating its own voltage. The absence was communicated to all the neighboring DGs using a sparse
of a master inverter reduces the risk of single point of failure, low-bandwidth communication network. The deviation of
which increases the reliability of the MG. The primary control the angular frequency was continually updated based on the
objectives here are observed for the islanded MG operating in estimated reactive power of adjacent DGs. This approach
a decentralized manner, where the low or no use of commu- requires a much more complicated communication network
nication between the DGs reduces the risk of communication as compared to the proposed method as the reactive power
latency, data losses, and cyberattacks. information of all the neighboring DGs need to be conveyed,
The general hierarchical control structure of MGs usually whereas in the proposed method, the V/ f deviation from a
consists of zero, primary, secondary, and tertiary layers of single DG is communicated to the other DGs in a single
power management to ensure the accurate control of several control cycle. An unscented Kalman filter was implemented
factors including voltage, current, and frequency while main- in [21] using a dynamic state estimation technique where the
taining high power quality and stability in the grid [1], [7]. internal states of the available DGs in the MG were estimated.
Droop control can be used to achieve accurate power sharing Gu et al. [23] discussed a nonlinear state estimation technique
between DGs in the MG without the use of communication of the system’s voltage for decentralized SC. Dehghanpour
systems in the primary control layer. Although droop-control- et al. [22], Haughton and Heydt [24], and Hashmi et al. [25]
based inverters are able to achieve accurate power sharing, have introduced more state-estimation-based techniques.
they suffer from V/ f deviation at the primary level [8]. The Variations of predictive control have also been implemented
droop characteristic curves determine the amount of deviation for V/ f restoration. In [15], a linear generalized predictive
in V/ f in response to active and reactive power disturbances control was implemented from the z-domain of the trans-
in the MG [9]. Therefore, many papers discuss the restoration fer function for the DG system. The state-space equations
of the deviated V/ f of the MG by estimating the deviation of the system model have been realized by utilizing the
terms and adding them to the droop voltage reference, in the generalized predictive control model along with the con-
secondary control (SC) layer [10], [11], [12], [13], [14], [15], trolled autoregressive moving average model, allowing the
[16], [17]. prediction of the future state variables of the system. The
The most widely used linear control method for N-step-ahead predicted values were used in a cost function
such SC includes the proportional SC (P-SC)- and the to obtain the system’s frequency and voltage deviation. A
proportional–integral SC (PI-SC)-based methods. The model-predictive-based secondary V/ f control method was
aforementioned methods scale the error between the measured shown in [16], where the deviation of the restoration vari-
and nominal values of V/ f to obtain the deviation terms [10], ables were obtained for a decentralized control system. An
[27], [28]. Even though the P-SC is fast, it results in autoregressive model was used along with an exogenous vari-
a high steady-state error [10]. The PI-SC resolves the able model of the phase-locked loop (PLL) controller and the
steady-state error issue; however, the integral controller model-predictive controller (MPC) to obtain the state-space
reduces the speed of the transient operation of the controller equations. Finally, the adopted cost function included the min-
and results in incorrect levels of power sharing between imization of the difference between the predicted and nominal
the inverters [12]. Heydari et al. [17] and Zhang et al. frequencies. An estimation-based predictive control technique
[18] proposed a distributed PI-based secondary controller; was presented in [26], where a Kalman filter estimated the
however, it requires high-bandwidth communication and present state vector of the linear-time-invariant state-space
parameter tuning. In [13], a switched control scheme allowed model. The controller was demonstrated for secondary volt-
the controller to switch between P-SC and PI-SC to balance age restoration by using the deviation of the reactive power.
the disadvantages of each approach. However, the dependence Table 1 summarizes and distinguishes the advantages and

86 VOLUME 4, 2023
TABLE 1. Comparison of Proposed and Existing SC Techniques

presents extended experimental results and an enhancement

on the controller presented in [29]. The main improvement
includes the complete elimination of PI controllers.
This article presents the following unique contributions:
1) a multitime scale-predictive optimal control for voltage
and frequency restoration in the ac MG;
2) computationally efficient predictive control leveraging
droop characteristic curves for straightforward imple-
mentation with low design complexity where SC cal-
culations are performed in only a single DG of the MG;
3) flexible power sharing (i.e., equal and unequal power
sharing) to meet the desired dispatch, with virtual
impedance control and no circulating reactive currents
between the parallel GFM units;
4) robust predictive SC for GFM units with a fast-dynamic
response for V/ f restoration with no parameter tuning
and a low bandwidth for communication.

II. SYSTEM DESIGN FOR ZERO- AND PRIMARY-LEVEL

FIGURE 1. Islanded ac MG with multiple parallel-connected DGs. CONTROL
A. MG SYSTEM
The general structure of the proposed MG system is shown in
disadvantages of the proposed SC method from the existing Fig. 1. The MG consists of multiple DGs connected in parallel
techniques. at the point of common coupling (PCC) and feeding common
Although the above estimation- and model-based nonlin- resistive–inductive (RL) loads. Each DG’s framework com-
ear control systems achieve secondary V/ f control, they prises a dc voltage source, a two-level three-phase inverter, an
involve substantial mathematical calculations that carry high LCL filter, circuit breakers to individually connect the DGs to
computational burden, which could be problematic for real- the load, and line impedances to account for the lengthy cables
time implementation. Furthermore, their complex structure that connect DGs across long distances in a given area. The
makes the initial design stage more challenging. Therefore, equivalent dc resistance of the inductors is considered in the
the motivation behind this article is to present a simple real- filter models of the MG. Table 2 shows the rated values of the
time implementation of a predictive-based controller with DGs, which have identical components. The 200-V dc voltage
low communication and computational burden. In this arti- is selected due to the limitation of the available equipment for
cle, finite-control-set model-predictive control (FCS-MPC) is laboratory-based experimental testing.
implemented in the primary layer, and consequently, the pre-
dicted powers resulting from the optimally selected switching B. PRIMARY- AND ZERO-LEVEL INVERSE DROOP AND
states are used to obtain the V/ f deviations (δV, δ f ), which FCS-MPC
are the SC variables that are computed in a single DG and con- The hierarchical control structure adopted in [1] is imple-
veyed to other DGs of the MG, as seen in Fig. 1. This article mented in the proposed MG system. The lower levels of
VOLUME 4, 2023 87
POONAHELA ET AL.: HIERARCHICAL MODEL-PREDICTIVE DROOP CONTROL FOR VOLTAGE AND FREQUENCY RESTORATION IN AC MGS

TABLE 2. System Parameters (αβ). The discretized state-space model is developed using the
Duhamel integral approach as given in [30]. The discretized
equations are used to predict the control variables, as follows:
⎡ ⎤ ⎡ ⎤
Ii,αβ (k + 1) Ii,αβ (k) 
⎢ ⎥ ⎢ ⎥
⎣Vc,αβ (k + 1)⎦ = Aa ⎣Vc,αβ (k)⎦ + Bb Vi,αβ (k)
Io,αβ (k + 1) Io,αβ (k)

+ Cc Vg,αβ (k) (3)

where the matrices Aa , Bb , and Cc are defined as follows:

Aa = eaTs , Bb = a−1 (Aa − I3x3 )b, Cc = a−1 (Aa − I3x3 )c
(4)
where I3x3 is an identity matrix and the inverse of matrix a is
represented by a−1 .
control include the zero, primary, and secondary controls, c) Optimization of cost function: The presence of the LCL
which are the focus of this article, while the tertiary level is filter poses some difficulties in modeling FCS-MPC in the
out of the scope of this article. The main purpose of the lower islanded MG, where the output voltage is the main control ob-
levels of control in the proposed MG includes the following jective. The coupling of the input inductor current derivative
goals: 1) maintaining the MG voltage and frequency at the (dIi /dt) with the capacitor voltage (dVc /dt) affects the output
nominal ranges; and 2) accurate sharing of active and reactive inductor current of the DG. Therefore, the dynamics of the
power between the DGs. output inductor current needs to be taken into consideration.
The following sections will provide the detailed design and Typical linear controllers based on cascaded dual loops are
description of the zero-level FCS-MPC and the primary-level usually employed to secure the stability of the system under
droop control. this scenario. To ensure accurate power delivery from the
DG, not only the voltage but also current requires to be reg-
1) MODELING OF FCS-MPC FOR VOLTAGE CONTROL MODE ulated precisely. Therefore, the multiobjective optimization
FCS-MPC is utilized to ensure that the inverter is operated in capability of the MPC algorithm is designed to control both
the voltage control mode. The technique primarily involves the output voltage (Vo) and the input current (Ii ) from the
developing a mathematical model of the DG, the prediction converter side to their respective references as presented in the
of control variables, and the optimization of a cost function. cost function g shown in (5). The prediction can be obtained
The measured signals include the input current (Ii ), capacitor for an N-step horizon. A two-step horizon is implemented in
voltage (Vc ), output current (Io), and output voltage (Vo). this article to compensate for the time delay that arises when
a) Mathematical modeling of the LCL filter: First, the mathe- operating at high sampling and control frequencies, especially
matical model of the LCL filter is used to develop the dynamic for finite-time predictive controllers that have a large amount
equations for the filter input inductor (Li ), filter capacitor (C), of calculations to complete within one control cycle [31].
and filter output inductor (Lo). This is done by applying Kirch- Large delays between the measurement and the actuation can
hoff’s voltage and current laws. The state-space representation cause significant problems if such delay compensations are
of these equations can be developed as follows to allow easy not considered in the design of the controller. It is assumed
and efficient calculation in matrix form that the reference variables remain constant during two steps
⎡ dI ⎤ ⎡ ⎤ of prediction
i
dt Ii  
⎢ dVC ⎥
⎥ = a⎢ ⎥ 2
⎢ ⎣V ⎦ + b V + c Vo (1) g = λi Ii,ref (k + N ) − Ii (k + N )
⎣ dt ⎦ C i
dIo I
dt o + λv (Vo,ref (k + N ) − Vo (k + N ))2 (5)
where Vi is the inverter output voltage, and a, b, and c are the where λi and λv are the weighting factors for the control on the
matrices defined as follows, which are derived from (1): input current and output voltage, respectively. The discretized
⎡ −R −1 ⎤ ⎡1⎤ ⎡ ⎤ state-space equation generates the predicted capacitor voltage
i
0 0
⎢ 1Li Li
⎥ Li and not the predicted output voltage as it is needed in (5).
⎢ ⎥ ⎢ ⎥
a=⎢ ⎣ C 0 −1 ⎥
C ⎦ , b = ⎣0⎦ , c = ⎣ 0 ⎦. (2) The dynamic equation of Lo is, therefore, used as presented
1 −Ro −1 in (6) to obtain the predicted output voltage. Consequently,
0 L L
0 Lo
o o
the optimization is conducted as traditionally done for the
b) Prediction of control variables: To reduce the calculations discretized FCS-MPC controller as presented in [32]
from a 3-D space to a 2-D space, the Clarke transformation
is used, where the measured three-phase currents and volt- Voα (k + N ) = Vcα (k + N ) + ωLoIoβ (k + N )
. (6)
ages (abc) are converted into the stationary reference frame Voβ (k + N ) = Vcβ (k + N ) − ωLoIoα (k + N )

88 VOLUME 4, 2023
 FIGURE 3. Restoration of V/f by shifting droop lines.

2) DROOP CONTROL FOR POWER SHARING

Droop control is a technique extensively used for achieving
decentralized active and reactive power sharing between mul-
tiple DGs. The conventional droop method is typically used
for systems with highly inductive line impedances at the out-
put of the inverter, while the inverse droop is used for systems
with highly resistive line impedances [34]. Numerous litera-
ture works for low-voltage MGs use the inverse droop control
because the low-voltage line parameters are predominantly
resistive [35]. The low-voltage grid has a highly resistive net-
FIGURE 2. Detailed block diagram for the hierarchical control levels. work, requiring the use of inverse droop control in the primary
controllers. The inverse droop control has a Q– f and P–V
relationship, as illustrated in (9). The variables P and Q are
d) Calculation of reference values: As illustrated in Fig. 2, the actual powers, which are calculated using the measured
the output voltage reference, Vo,ref , is calculated based on the voltage (Vo,abc ) and current (Io,abc ) at the output of the filter,
inverse droop technique, which is explained in the next sec- as shown in Fig. 2. The invariant quantities are the nominal
tion. Another important aspect to consider at this stage is the frequency ( fn ), the nominal RMS voltage (Vn ), and the droop
significant voltage drop caused by line impedance across the gains k p and kq
cables that connect the ends of the DGs to the loads. This can V ∗ = Vn − k p P
lead to inaccurate power sharing between the DGs. Therefore,
and as explained in [33], in order to achieve accurate power f ∗ = fn + kq Q. (9)
sharing without the need of communication and extra sensors,
virtual impedance in the control is added to adjust the output III. SECONDARY-LEVEL PREDICTIVE-BASED CONTROL
impedance that the inverter sees at the PCC. The cable line Power sharing in the primary control level is achieved as
impedance Zline consists of a resistive impedance Rline and a consequence of drooping the voltage magnitude and fre-
inductive impedance Xline . The virtual impedance voltage VZv quency, as depicted in the previous section. The droop of
is calculated based on the virtual impedance, measured output active and reactive power levels results in a variation in the
current Io, and MG frequency ω as follows: V/ f of the system, creating steady-state deviations from the
nominal values and consequently affecting the stability and
VZvα = Ioα Rv − Ioβ Lv ω power quality of the entire system. Therefore, the secondary
(7)
VZvβ = Ioβ Rv + Ioα Lv ω controller is required to restore such deviations and bring
the system’s V/ f back to their nominal values. However,
where Rv and Lv are the virtual resistance and virtual induc- this should be realized without affecting the power sharing
tance, respectively. established in the primary layer. To achieve this task, a widely
The estimation of the input reference current, Ii,ref , requires used technique includes detecting the V/ f deviation values,
the calculation of the output current reference. Io,ref can be δV and δ f , and adding them to the droop equations [10],
calculated in the αβ frame by using the measured powers [11], [12], [13], [14], [15], [16]. This is generally achieved by
P(k), Q(k), and the reference output voltage Vo,ref . applying a vertical shift to the Q– f and P–V droop equations
Similar to the case with Lo, the dynamic equation of the at a magnitude equal to the calculated deviation terms. As
filter capacitor C is used, as shown in the following equation, seen in Fig. 3, when active or reactive loads are added, the
to estimate the predicted input current reference: primary controller shifts the operation of the DG from point 1
to point 2. This creates a deviation of V/ f from the nominal
Iiα,ref (k) = Ioα,ref (k) − ωCVcβ,ref (k) values. Therefore, this deviation is estimated and added as a
. (8)
Iiβ,ref (k) = Ioβ,ref (k) + ωCVcα,ref (k) vertical shift to the droop lines, changing the operation from

VOLUME 4, 2023 89
POONAHELA ET AL.: HIERARCHICAL MODEL-PREDICTIVE DROOP CONTROL FOR VOLTAGE AND FREQUENCY RESTORATION IN AC MGS

FIGURE 5. SC implementation using predicted powers.

δV and δ f , with the application of the predicted powers ac-

cording to (11). It is clear that the V/ f deviations are based on
the predicted powers and the inverse droop gains; hence, they
are obtained in a straightforward manner.
Following that, a delay (D) allows the predicted active and
reactive powers to be utilized within the same control cycle.
It is assumed in this controller that δ f (k + N ) can be equal
to δ f (k) for small values of N, as the sampling time is small
enough for the approximation to be valid. Finally, as shown
in Fig. 5, the restorative terms are incorporated with the droop
control terms to form the reference voltage to the inner control
layer. This method of SC reduces the deviation of V/ f by
a significant amount, as shown in Section IV. It is possible
to achieve precise secondary restoration with this approach
as the predicted output voltage and current are a compelling
indication of the power required from the DG in the upcoming
cycle of control and can consequently represent the antici-
FIGURE 4. Flowchart of the implemented FCS-MPC algorithm.
pated V/ f deviations
⎧
⎪P(k + N ) = 1.5[Voα,opt (k + N )Ioα,opt (k + N )
⎪
⎪
point 2 to point 3. The shift is performed in order to change the ⎪
⎨
set points of V/ f in the vertical axis without changing the set +Voβ,opt (k + N )Ioβ,opt (k + N )]
(10)
points of the active or reactive powers in the horizontal axis of ⎪
⎪ Q(k + N ) = 1.5[Voβ,opt (k + N )Ioα,opt (k + N )
⎪
⎪
the droop graphs. This way, V/ f can be restored to nominal ⎩
−Voα,opt (k + N )Ioβ,opt (k + N )]
values without affecting the power sharing of the MG. The
calculation of δV and δ f is usually obtained by measuring the δ f (k + N ) = f ∗ − fn = kq Q(k + N )
error between the measured and reference V/ f , followed by a . (11)
δV (k + N ) = V ∗ − Vn = −k p P(k + N )
PI controller to drive the error to zero.
In this article, the deviation restoration is obtained using Furthermore, the droop equations show that the deviation
MPC without the use of PI controllers. The proposed method terms δ f and δV are not dependent on the droop constants
is also devoid of the requirement of measured RMS voltage (or the rating of the DG) even during unequal load sharing
or frequency in the SC level. Therefore, this layer no longer conditions. To prove this, variables kq , k p , Q, and P can
requires PLL control, which is used to obtain the frequency be defined according to (12) and (13) based on the highest
of the MG. The flowchart in Fig. 4 describes the proposed allowed deviation of V/ f from the nominal values (5% and
FCS-MPC algorithm, where the output includes the predicted 2%, respectively), required by the international IEEE 1547
powers in addition to the optimal switching sequence. Com- standard [36]. The values of Pload and Qload are assumed to
pared to the traditional FCS-MPC algorithm, the predicted be the amount of load power at a given instance, while Ptotal
values of the optimally selected output voltage and current and Qtotal are the total active and reactive power generation
from the DG, Vg,opt (k + N ) and Ig,opt (k + N ), are utilized to capacities of the MG based on the individual rated capacities
calculate the predicted powers of the upcoming control cycle. of the DGs. The general equations for the deviation terms,
Correspondingly, the optimal predicted values are used to presented in (14), can be calculated by substituting (12) and
obtain P(k + N ) and Q(k + N ) with the help of (10). The (13) into the droop equations (9). This proof shows that at
optimal values are selected after the optimization of the cost any given instance, δ f and δV are independent of the droop
function g, as depicted in Fig. 4. Thereupon, the inverse droop constants and are mainly dependent on the load requirement of
equations (9) are rearranged to obtain the general deviations the MG, allowing them to be used globally within the system.

90 VOLUME 4, 2023
 The resilience of the secondary controller in the case of a
failure to any of the DGs is an important aspect to consider.
The deviation variables for the SC layer are determined in
the largest DG unit that dominates the power contribution.
Therefore, in the case that a failure is detected in this unit, the
SC calculations need to be initiated in the controller with the
next largest unit and communicated to the smaller DGs. This
would require a prior knowledge of the maximum capacity
of each DG in order to decide which of the available units
would take over the calculation of the deviation variables. In
the case that a failure occurs in any of the other smaller DGs,
there need not be any particular action as this DG will be
disconnected from the MG
 fmax 0.02 fn
kq = Qmax = Qrated
Vmax 0.05Vn
(12)
kp = Pmax = Prated
⎧  
⎨Q = Qrated Qload
 Qtotal (13)
⎩P = Prated Pload
Ptotal
⎧  
⎨δ f = 0.02 fn Qload
 Qtotal
 . (14)
⎩δV = 0.05Vn Pload
P total

IV. EXPERIMENTAL AND SIMULATION STUDIES

The validation of the implemented control technique has been
FIGURE 6. Photograph of the MG test bed.
carried out experimentally to observe the steady-state and
transient responses of the controller with varying levels of RL
loads for similar rated and differently rated DGs. This allows 1500 W and 30 VAR. The high load switch enables the parallel
the observation of the MG controllers for equal power sharing connection and disconnection of the high power load to the
and unequal power sharing between the DGs. In addition, the MG.
robustness of the controller to model parameter variations has The following sequence of events is followed to power up
been investigated. The weighting factors have been heuristi- the entire MG.
cally tuned to be at unity to balance the control of voltage at 1) Initially, DG1 is connected to the low power load (CB1
the nominal level and the current to prevent reactive power of Fig. 1 is turned ON) and operating in droop control
circulation and maintain accurate power levels. mode. DG2 and DG3 are in the voltage tracking mode
The controller has been validated using the test bed shown and synchronized to the MG voltage but not physically
in Fig. 6, and Fig. 7 shows the detailed schematic diagram. connected to it.
The power stage for each DG includes an EA PSI 9750-12 dc 2) Next, the circuit breaker switch of DG2 (CB2 in Fig. 1)
power supply with constant 200-V dc voltage. The inverters is turned ON to physically connect DG2, while the
are built using insulated gate bipolar transistor switches and controller is simultaneously switched from the synchro-
rated to handle up to a maximum of 6.1 kW. The parameters nization mode to the droop control mode. This is done
used in the LCL filter and output line impedances are the by changing the voltage reference from the PCC voltage
same parameters as mentioned in Table 2. Three identical to the droop voltage.
DGs have been constructed and connected in parallel at the 3) Consequently, DG3 is similarly connected via CB3 .
PCC. The controllers are implemented on OPAL-RT simu- 4) Finally, the high power load is connected to test the MG
lators, where DG1 and DG2 are controlled by the OP5600 for step changes in load power.
and DG3 is controlled by the OP5700. The requirement for To observe the transient and steady-state responses of
two OPAL-RT controllers arises due to the limitations on the the secondary controller, two tests are captured during two
number of analog inputs available in a single controller. The consecutive transients, first while enabling the secondary
low power load consists of a series-connected resistor with restoration controller and next while changing the load power,
45  and inductor with 30 mH per phase, with a power rating as depicted in Figs. 8 and 10, for equal and unequal load power
of 300 W and 18 VAR and is connected at the PCC. The high sharing scenarios. The transients are designed to monitor not
power load similarly consists of 9  and 1-mH power rating of only the restoration of V/ f , but also the ability to maintain

VOLUME 4, 2023 91
POONAHELA ET AL.: HIERARCHICAL MODEL-PREDICTIVE DROOP CONTROL FOR VOLTAGE AND FREQUENCY RESTORATION IN AC MGS

FIGURE 7. Schematic diagram of the MG test bed.

accurate power sharing between the DGs. The first transient

invokes the secondary controller, while the second transient
includes the disconnection of the high power load. For each
case, the presented values are the three-phase sinusoidal PCC
voltages, RMS PCC voltage, sinusoidal current of phase A
from each of the DGs, the system frequency error ( fn − f (k)),
and active powers and reactive powers of each of the DGs. A
nominal frequency of 60 Hz and voltage amplitude of 100 V
are maintained in the MG during the absence of the utility
grid.
FIGURE 8. Case 1: Equal load power sharing. (a) Steady-state and
transient response. (b) Zoomed area of the first transient. (c) Zoomed area
A. CASE 1: EQUAL LOAD POWER SHARING of the second transient.
The amount of the total load handled by each DG is decided
based on the decentralized droop controller, specifically the
droop constants. The droop constants include the rating of sharing as dictated by the inverse droop laws. The total load
the DGs to obtain equal load sharing. Each DG is rated at connected is 1800 W and 48 VAR. Each DG equally supplies
1 kW and 100 VAR. As illustrated in Fig. 8(a), initially only 600 W and 16 VAR to the combined common load. During
primary control is enabled. The zoomed windows in Fig. 8(b) the first transient, as V/ f restoration is enabled, it is observed
and (c) show that the RMS voltage has deviated by 8%, while that the voltage error reduces to 2.6% and the frequency error
the frequency has risen by 0.14% to enable accurate power to 0.01%, while each DG maintains sharing the same amount

92 VOLUME 4, 2023
FIGURE 9. THD of voltage and current at the PCC. (a) Phase-A THDV o .
(b) Phase-A THDIo .

of power. The minor rise in active and reactive power by a

few units is due to the overall increase of the system volt-
age. During the second transient, the total active and reactive
load reduces by 84% and 47%, respectively. The frequency is
maintained at 60 Hz, and the voltage error is slightly reduced
as the total load is very low. As presented in Fig. 9, the
total harmonic distortion (THD) of the voltage and current are
1.3% and well within the ranges imposed by the IEEE 1547
standards [36].

B. CASE 2: UNEQUAL LOAD POWER SHARING

To test for unequal load power sharing, DG1 is rated the same
as before, but the ratings of DG2 and DG3 are reduced by half.
Therefore, they individually supply half of the power supplied FIGURE 10. Case 2: Unequal power sharing. (a) Steady-state and transient
response. (b) Zoomed area of the first transient. (c) Zoomed area of the
by DG1 at all times, as seen in Fig. 10(a). Thus, the DGs cover
second transient.
the total load in a 2:1:1 ratio. Fig. 10(b) and (c) shows that
the average deviation leads to a steady-state voltage error of
7%, while the frequency rises by an average of 0.2%. DG1 0.01% error and the average voltage error remains low at
supplies 900 W and 24 VAR, while DG2 and DG3 each supply 2.5%, while the power sharing ratios stay unaffected.
half of that. During the first transient as V/ f restoration is
enabled, it is observed that the average voltage error reduces
to 2.7 V and the average frequency error to 0.03 Hz, while C. CASE 3: ROBUSTNESS TO PARAMETER MISMATCH
each DG maintains accurate power sharing. During the second MPC in the primary control layer involves model-based con-
transient, the frequency is maintained close to 60 Hz with trol and, therefore, uses the rated values of the LCL filter

VOLUME 4, 2023 93
POONAHELA ET AL.: HIERARCHICAL MODEL-PREDICTIVE DROOP CONTROL FOR VOLTAGE AND FREQUENCY RESTORATION IN AC MGS

FIGURE 12. Comparison of PI-based and predictive-based restoration for

(a) RMS voltage and (b) frequency.

D. CASE 4: COMPARISON WITH LINEAR CONTROL

A comparison with the well-established PI-based secondary
controller is highlighted in Fig. 12, where the connection and
disconnection of an RL load has been simulated at time in-
stances 3 and 4 s, respectively. The results indicate that while
the overshoot in both the controllers is similar, the predictive-
based controller has a shorter settling time. The absence of
integral control allows the predictive controller to perform
faster and reach a steady state more swiftly. Furthermore, an
assessment of power sharing shows that the predictive-based
controller shares the power at correct ratios, while the PI-
based controller has a 20% error in its power sharing ratios.

E. CASE 5: TUNING OF WEIGHTING FACTORS FOR

FIGURE 11. Robustness against model parameter mismatch. (a) ARTE. ZERO-LEVEL CONTROL
(b) Average THD.
The multiobjective cost function of the MPC controller has
two weighting coefficients, as seen in (5). The selection of
components to perform calculations during the control se- optimal values for each parameter is significant in keeping the
quence. In reality, the actual inductances and capacitances operations of the controller within stable limits at different
deviate from the rated values due to heat and operating points. operating points. Owing to the absence of the utility grid,
To observe the robustness of the controller to such variations, the regulation of the MG voltage falls upon the inverters, and
the values of each component are varied in the control algo- therefore, the output voltage is considered an essential control
rithm, while the physical system is kept unchanged. Four tests variable. Therefore, its weighting factor (λv ) is kept at unity,
are performed, where the first three involve the deviation of while the current weighting coefficient (λi ) is tuned for an
one of the components, Li , C, or Lo, to values ranging from optimal value below and above unity. Initially, the value of λi
−20% to +20% of its rated value. The fourth test involves is kept at unity; then, step increases were imposed. It is seen in
the variation of all three parameters together. The effect of Fig. 13(a) that, as λi increases, the voltage gradually deviates
parameter mismatch is observed for: from the nominal value. When reaching a λi of 5, the voltage
1) the percentage absolute reference tracking error (ARTE) control loses stable operation. This occurs since current track-
of the variables in the cost function, i.e., %eVo and %eIi ; ing supersedes voltage tracking, leading to system instability.
2) the THD of the output voltage and output current of each Furthermore, the controller was tested for lower values of λi .
DG, i.e., THDVo and THDIo. Fig. 13(b) shows that the controller maintains stability until
The ARTE is calculated according to the following equa- the current control no longer exists at a null value of λi .
tion, as explained in [37]: Without current control, accurate power sharing between DGs
is not possible, which leads to voltage collapse and loss of
|xref (k)−x(k)|
%ex,k = xRMS (k) , k = a, b, c stability. Therefore, a unity weighting factor is selected for
%ex,a +%ex,b +%ex,c (15)
%ex = both control variables in order to enable the optimal operation
3
of the MG.
where x is the observed variable. The experimental tests performed above prove that the pro-
Fig. 11 shows that the error is maintained well below 2% posed predictive-based secondary controller is able to restore
for the given range of parameter mismatch, while the THD is voltage and frequency to nominal values while maintaining
maintained under 5%. This proves that the entire controller, power sharing levels between parallel-connected inverters. It
complete with primary and SC, shows a high range of robust- presents a faster dynamic response to transient changes and
ness against parameter variations. has a stable operation in the steady state despite large load

94 VOLUME 4, 2023
 frequency, while accurately maintaining active and reactive
load power sharing between the distributed generation units.

ACKNOWLEDGMENT
This publication was made possible by grants NPRP12C-
33905-SP-213 and NPRP12C-33905-SP-220 from the Qatar
National Research Fund (a member of Qatar Foundation). The
statements made herein are solely the responsibility of the
authors.

REFERENCES
[1] J. M. Guerrero, J. C. Vasquez, J. Matas, L. G. De Vicuña, and
M. Castilla, “Hierarchical control of droop-controlled AC and DC
microgrids—A general approach toward standardization,” IEEE Trans.
Ind. Electron., vol. 58, no. 1, pp. 158–172, Jan. 2011.
[2] S. D’silva, M. Shadmand, S. Bayhan, and H. Abu-Rub, “Towards
grid of microgrids: Seamless transition between grid-connected and
islanded modes of operation,” IEEE Open J. Ind. Electron. Soc., vol. 1,
pp. 66–81, 2020.
[3] S. Bayhan and H. Abu-Rub, “A simple control technique for distributed
generations in grid-connected and islanded modes,” in Proc. IEEE 27th
Int. Symp. Ind. Electron., 2018, pp. 1237–1242.
[4] K. S. Ratnam, K. Palanisamy, and G. Yang, “Future low-inertia power
systems: Requirements, issues, and solutions—A review,” Renewable
Sustain. Energy Rev., vol. 124, 2020, Art. no. 109773.
[5] K. Feng and C. Liu, “Distributed hierarchical control for fast frequency
restoration in VSG-controlled islanded microgrids,” IEEE Open J. Ind.
Electron. Soc., vol. 3, pp. 496–506, 2022.
FIGURE 13. Effect of (a) increasing and (b) decreasing the weighting factor. [6] A. Khan, M. Hosseinzadehtaher, M. B. Shadmand, S. Bayhan, and
H. Abu-Rub, “On the stability of the power electronics-dominated grid:
A new energy paradigm,” IEEE Ind. Electron. Mag., vol. 14, no. 4,
pp. 65–78, Dec. 2020.
changes. The controller has a low design complexity and low [7] A. Mohammed, S. S. Refaat, S. Bayhan, and H. Abu-Rub, “AC micro-
tuning requirement and is robust to system parameter varia- grid control and management strategies: Evaluation and review,” IEEE
tions. For future studies of the proposed predictive-based SC, Power Electron. Mag., vol. 6, no. 2, pp. 18–31, Jun. 2019.
[8] Y. Sun, X. Hou, J. Yang, H. Han, M. Su, and J. M. Guerrero, “New
some important aspects to analyze are the effects of nonlinear perspectives on droop control in AC microgrid,” IEEE Trans. Ind.
loads, unbalanced loads, and mismatched feeder impedance Electron., vol. 64, no. 7, pp. 5741–5745, Jul. 2017.
on the effectiveness of the secondary controller. Stability anal- [9] M. C. Chandorkar, D. M. Divan, and R. Adapa, “Control of parallel
connected inverters in standalone AC supply systems,” IEEE Trans. Ind.
ysis to study the practical limitations of the controller, along Appl., vol. 29, no. 1, pp. 136–143, Jan./Feb. 1993.
with the controller’s susceptibility to communication failure, [10] J. Rocabert, A. Luna, F. Blaabjerg, and P. Rodríguez, “Control of power
will be topics for future research. converters in AC microgrids,” IEEE Trans. Power Electron., vol. 27,
no. 11, pp. 4734–4749, Nov. 2012.
[11] M. Hua, H. Hu, Y. Xing, and J. M. Guerrero, “Multilayer control
V. CONCLUSION for inverters in parallel operation without intercommunications,” IEEE
Trans. Power Electron., vol. 27, no. 8, pp. 3651–3663, Aug. 2012.
This article discussed a multitime scale-predictive control [12] H. Shi, F. Zhuo, H. Yi, F. Wang, D. Zhang, and Z. Geng, “A
for voltage and frequency restoration in an ac islanded MG novel real-time voltage and frequency compensation strategy for
with multiple inverse-droop-based, parallel-connected, GFM photovoltaic-based microgrid,” IEEE Trans. Ind. Electron., vol. 62,
no. 6, pp. 3545–3556, Jun. 2015.
inverters. Droop constants along with the optimally selected [13] J. M. Rey, P. Martí, M. Velasco, J. Miret, and M. Castilla, “Secondary
predicted powers are leveraged to estimate the voltage and switched control with no communications for islanded microgrids,”
frequency trajectory of the next time step. Therefore, the IEEE Trans. Ind. Electron., vol. 64, no. 11, pp. 8534–8545, Nov. 2017.
[14] B. Liu, T. Wu, Z. Liu, and J. Liu, “A small-ac-signal injection-
key advantages of the proposed method, in addition to the based decentralized secondary frequency control for droop-controlled
absence of parameter tuning in the SC layer, are a noncom- islanded microgrids,” IEEE Trans. Power Electron., vol. 35, no. 11,
plex design concept that is straightforward to implement and pp. 11634–11651, Nov. 2020.
[15] M. Jayachandran and G. Ravi, “Decentralized model predictive hierar-
nonreliant on computationally intensive model-based calcula- chical control strategy for islanded AC microgrids,” Electr. Power Syst.
tions or complex-estimation-based techniques. The design of Res., vol. 170, pp. 92–100, 2019.
the hierarchical control layers presents a system that is com- [16] C. Ahumada, R. Cárdenas, D. Saez, and J. M. Guerrero, “Secondary
control strategies for frequency restoration in islanded microgrids with
pletely devoid of PI controllers. Simulation and experimental consideration of communication delays,” IEEE Trans. Smart Grid,
results were conducted for the controller and, compared to the vol. 7, no. 3, pp. 1430–1441, May 2016.
PI-based controller, showed significantly better response and [17] R. Heydari, T. Dragicevic, and F. Blaabjerg, “High-bandwidth sec-
ondary voltage and frequency control of VSC-based AC microgrid,”
operation. It has been proven that the controller is robust, has IEEE Trans. Power Electron., vol. 34, no. 11, pp. 11320–11331,
a fast dynamic feedback, and is able to restore voltage and Nov. 2019.

VOLUME 4, 2023 95
POONAHELA ET AL.: HIERARCHICAL MODEL-PREDICTIVE DROOP CONTROL FOR VOLTAGE AND FREQUENCY RESTORATION IN AC MGS

[18] Y. Zhang, A. Mohammadpour Shotorbani, L. Wang, and [37] V. Yaramasu, M. Rivera, B. Wu, and J. Rodriguez, “Model predictive
B. Mohammadi-Ivatloo, “Enhanced PI control and adaptive gain current control of two-level four-leg inverters—Part I: Concept, algo-
tuning schemes for distributed secondary control of an islanded rithm, and simulation analysis,” IEEE Trans. Power Electron., vol. 28,
microgrid,” IET Renewable Power Gener., vol. 15, no. 4, pp. 854–864, no. 7, pp. 3459–3468, Jul. 2013.
2021.
[19] J. M. Rey, C. X. Rosero, M. Velasco, P. Martí, J. Miret, and M. Castilla,
IRESHA POONAHELA (Student Member, IEEE)
“Local frequency restoration for droop-controlled parallel inverters in
received the bachelor’s degree in electrical and
islanded microgrids,” IEEE Trans. Energy Convers., vol. 34, no. 3,
computer engineering from Texas A&M Univer-
pp. 1232–1241, Sep. 2019.
sity at Qatar (TAMUQ), Doha, Qatar, in 2017. She
[20] Y. Han, “Consensus-based enhanced droop control scheme for ac-
is currently working toward the Ph.D. degree in
curate power sharing and voltage restoration in islanded micro-
electrical engineering with Texas A&M University
grids,” in Modeling and Control of Power Electronic Converters
(TAMU), College Station, TX, USA.
for Microgrid Applications. Cham, Switzerland: Springer, 2022,
From 2017 to 2018, she was a Research As-
pp. 239–261.
sistant with the TEES Smart Grid Center and
[21] S. Yu, L. Zhang, H. H.-C. Lu, T. Fernando, and K. P. Wong, “A DSE-
TAMUQ. She was a Teaching Assistant with the
based power system frequency restoration strategy for PV-integrated
Department of Electrical Engineering, TAMU, in
power systems considering solar irradiance variations,” IEEE Trans.
2021 and 2023. Her research interests include the control of power in power
Ind. Informat., vol. 13, no. 5, pp. 2511–2518, Oct. 2017.
electronic converters, renewable energy integration, grid-tied and islanded
[22] K. Dehghanpour, Z. Wang, J. Wang, Y. Yuan, and F. Bu, “A survey
microgrid operations, proportional–integral control, and model-predictive
on state estimation techniques and challenges in smart distribution
control.
systems,” IEEE Trans. Smart Grid, vol. 10, no. 2, pp. 2312–2322,
Mar. 2019.
[23] W. Gu, G. Lou, W. Tan, and X. Yuan, “A nonlinear state estimator- ABDELBASSET KRAMA (Member, IEEE) re-
based decentralized secondary voltage control scheme for autonomous ceived the bachelor’s and master’s (Hons.) degrees
microgrids,” IEEE Trans. Power Syst., vol. 32, no. 6, pp. 4794–4804, in electrical engineering from the Kasdi Merbah
Nov. 2017. University of Ouargla, Ouargla, Algeria, in 2013
[24] D. A. Haughton and G. T. Heydt, “A linear state estimation formulation and 2015, respectively, and the Ph.D. degree in
for smart distribution systems,” IEEE Trans. Power Syst., vol. 28, no. 2, electrical engineering from El Oued University, El
pp. 1187–1195, May 2013. Oued, Algeria, in 2019.
[25] K. Hashmi, R. Ali, M. Hanan, W. Aslam, A. Siddique, and M. M. Khan, From 2018 to 2022, he was a Researcher
“Reactive power sharing and voltage restoration in islanded AC micro- with Texas A&M University at Qatar, Doha,
grids,” Turkish J. Elect. Eng. Comput. Sci., vol. 30, no. 3, pp. 818–838, Qatar, where he was involved in different research
2022. projects related to smart grid. He is currently with
[26] A. Morattab, O. Akhrif, and M. Saad, “Decentralised coordinated Qatar Environment and Energy Research Institute (QEERI), Doha, Qatar,
secondary voltage control of multi-area power grids using model as a Scientist. His research interests include renewable energy conversion,
predictive control,” IET Gener., Transmiss. Distrib., vol. 11, no. 18, microgrid control, and smart grid applications.
pp. 4546–4555, 2017. Dr. Krama is a Professional Active Member of the IEEE Industrial Elec-
[27] A. Karaki, M. Begovic, S. Bayhan, and H. Abu-Rub, “Frequency and tronics Society. He is currently the Secretary of the IEEE-IES Qatar Section
voltage restoration for droop controlled AC microgrids,” in Proc. 2nd Chapter. He was the recipient of many prestigious awards, such as the student
Int. Conf. Smart Grid Renewable Energy, 2019, pp. 1–6. excellent award from Ouargla University in 2015, Best Presentation Award
[28] S. Bayhan, H. Abu-Rub, J. I. Leon, S. Vazquez, and L. G. Franquelo, from American Algerian Foundation (AUF) in Summer University in 2018,
“Power electronic converters and control techniques in AC microgrids,” Outstanding Research Award from Electrical and Computer Engineering at
in Proc. IECON 43rd Annu. Conf. IEEE Ind. Electron. Soc., Oct. 2017, Texas A&M University at Qatar in 2021, and Best Paper Award from SGRE
pp. 6179–6186. 2022 Conference.
[29] I. Poonahela, A. Krama, S. Bayhan, H. Abu-Rub, M. Begovic, and M.
B. Shadmand, “Predictive voltage and frequency restoration for decen-
tralized FCS-MPC based droop controlled DGS in AC microgrids,” SERTAC BAYHAN (Senior Member, IEEE) re-
in Proc. IEEE 3rd Int. Conf. Smart Grid Renewable Energy, 2022, ceived the bachelor’s, M.S., and Ph.D. degrees
pp. 1–6. in electrical engineering from Gazi University,
[30] B. Long, Z. Zhu, W. Yang, K. T. Chong, J. Rodríguez, and J. M. Guer- Ankara, Türkiye, in 2008 and 2012, respectively.
rero, “Gradient descent optimization based parameter identification for He is currently a Senior Scientist with the
FCS-MPC control of LCL-type grid connected converter,” IEEE Trans. Qatar Environment and Energy Research Insti-
Ind. Electron., vol. 69, no. 3, pp. 2631–2643, Mar. 2022. tute, Hamad Bin Khalifa University, Doha, Qatar,
[31] P. Cortes, J. Rodriguez, C. Silva, and A. Flores, “Delay compensation in where he is a Faculty Member with the rank of
model predictive current control of a three-phase inverter,” IEEE Trans. an Associate Professor with the Sustainable Di-
Ind. Electron., vol. 59, no. 2, pp. 1323–1325, Feb. 2012. vision, College of Science and Engineering. He
[32] J. Rodriguez et al., “Predictive current control of a voltage source has acquired $13 million in research funding and
inverter,” IEEE Trans. Ind. Electron., vol. 54, no. 1, pp. 495–503, authored or coauthored more than 170 papers in mostly prestigious IEEE
Feb. 2007. journals and conferences. He is also the coauthor of two books and five book
[33] A. Vijay, N. Parth, S. Doolla, and M. C. Chandorkar, “An adaptive chapters. He was a recipient of many prestigious international awards.
virtual impedance control for improving power sharing among inverters
in islanded AC microgrids,” IEEE Trans. Smart Grid, vol. 12, no. 4, UGUR FESLI received B.Sc. degree from Gazi
pp. 2991–3003, Jul. 2021. University, Ankara, Turkey, in 2006, and the M.Sc.
[34] A. Engler, “Applicability of droops in low voltage grids,” Int. J. Distrib. degree from Zonguldak Karaelmas University,
Energy Resour., vol. 1, no. 1, pp. 1–6, 2005. Zonguldak, Turkey, in 2009, both in electrical elec-
[35] M. Aquib, A. Vijay, S. Doolla, and M. C. Chandorkar, “Power tronic engineering. He is currently working toward
sharing enhancement strategy for parallel-connected voltage-source the Ph.D. degree in energy system engineering with
inverters with common DC and AC buses,” in Proc. IEEE Int. Gazi University.
Conf. Power Electron., Smart Grid, Renewable Energy, 2022, He is currently a Lecturer with the Department
pp. 1–6. of Electronics and Automation, Gazi University
[36] IEEE Standard for Interconnection and Interoperability of Distributed Technical Sciences Vocational School. His cur-
Energy Resources With Associated Electric Power Systems Interfaces, rent research interests include power systems and
IEEE Standard 1547-2018, 2018, pp. 1–138. power electronics.

96 VOLUME 4, 2023
 MOHAMMAD B. SHADMAND (Senior Mem- MIROSLAV M. BEGOVIC (Life Fellow, IEEE)
ber, IEEE) received the Ph.D. degree in electrical received the B.S. and M.S. degrees in electrical
engineering from Texas A&M University, College engineering from Belgrade University, Belgrade,
Station, TX, USA, in 2015. Yugoslavia, and the Ph.D. degree in electrical en-
From 2015 to 2016, he was an Instructor with the gineering from Virginia Polytechnic Institute and
Department of Electrical and Computer Engineer- State University, Blacksburg, in 1989.
ing, Texas A&M University. From 2016 to 2017, He is the Head of the Department of Electrical
he was a Research Engineer with the Renewable and Computer Engineering and the College of En-
Energy and Advanced Power Electronics Research gineering Excellence Professor with Texas A&M
Laboratory, College Station. From 2017 to 2020, University, College Station, TX, USA. Prior to that,
he was an Assistant Professor with the Department he was a Professor and the Chair of the Electric
of Electrical and Computer Engineering, Kansas State University, Manhat- Energy Research Group, School of Electrical and Computer Engineering, and
tan, KS, USA. Since 2020, he has been an Assistant Professor with the an Affiliated Faculty Member of the Brook Byers Institute for Sustainable
Department of Electrical and Computer Engineering, University of Illinois Systems and the University Center of Excellence in Photovoltaic Research,
Chicago, Chicago, IL, USA. His current research interests include advanced Georgia Institute of Technology, Atlanta, GA, USA. He delivered more than
model-predictive control, grid-tied power electronics interfaces with advance 100 keynote and invited presentations worldwide.
functionalities, grid situational awareness, and control of smart microgrid Prof. Begovic was a recipient of the IEEE Power and Energy Society (PES)
systems. 2019 Meritorious Service Award. He was a Chair of the Emerging Tech-
nologies Coordinating Committee of the IEEE PES, the IEEE PES Treasurer
(from 2010 to 2011), and the IEEE PES Distinguished Lecturer. He served
HAITHAM ABU-RUB (Fellow, IEEE) received as the President-Elect, the President, and the Immediate Past President of the
the M.Sc. degree in electrical engineering from IEEE PES from 2012 to 2018.
the Gdynia Marine Academy, Gdynia, Poland, in
1990, and the Ph.D. degree in electrical engineer-
ing from the Gdansk University of Technology,
Gdansk, Poland, in 1995.
He has research and teaching experiences at
many universities in many countries, including
Qatar, Poland, Palestine, USA, U.K., and Germany.
For five years, he was the Chair of Electrical and
Computer Engineering Program, TAMUQ, and is
the Managing Director of the Smart Grid Center at the same university. He
has authored or coauthored more than 550 journal and conference papers, two
issued patents, six books, and six book chapters. His research interests include
power electronic converters, renewable energy systems, electric drives, and
smart grids.
Dr. Abu-Rub was the recipient of Fulbright and Alexander von Humboldt
scholarships. He has secured and successfully supervised and supervising
many advanced research projects with over $20M funds. His research group
contains many Ph.D. students and postdoctoral fellows. He is a Fellow of the
IEEE and Co-Editor-in-Chief of the IEEE TRANSACTIONS ON INDUSTRIAL
ELECTRONICS.

VOLUME 4, 2023 97