A Modified Droop based Decentralized Control

Strategy for an Islanded AC Microgrid

Bibhudatta Mishra Monalisa Pattnaik
Dept. of Electrical Engineering Dept. of Electrical Engineering
National Institute of Technology, Rourkela National Institute of Technology, Rourkela
Rourkela, India Rourkela, India
bibhumishra.123@[Link] pattnaikm@[Link]

Abstract—Microgrid is gaining more attention as a new The control strategies in microgrid are based on hierarchical
solution for meeting progressive power demand because of control. It is categorized into four levels such as power sharing
its smooth and reliable integration into main grid. Various control, converter output control, microgrid supervisory (sec-
system characteristics are greatly affected in a microgrid due
to a sudden load disruption. These features include current ondary) control, and the grid supervisory (tertiary) control.
sharing, voltage and frequency variation, and real and reactive The secondary control is categorized in two ways namely
power sharing among many inverters. To accomplish autonomous centralized control and decentralized control. The decentral-
voltage and frequency regulation, a droop-based decentralized ized techniques are widely implemented than centralized one
control technique for parallel functioning inverters is studied due to various advantages like simplicity, more reliability,
in this article. There is also adequate power sharing among the
inverters. Conventional P-f/Q-V and P-V/Q-f droop laws are used ignorance in interaction of different units and efficient way
in the suggested control technique. In stand-alone mode, both of solving energy management problem [2]. There are several
simple and modified droop laws are simulated for control of two techniques which come under decentralized approach namely:
parallel inverters. The new operating values for frequency and droop, distributed model predictive, consensus-based, agent-
voltage in simplified droop control are dependent on real and based, and decomposition-based techniques. The conventional
reactive power respectively, whereas in case of modified droop
control, these set points are function of both real and reactive droop control which is otherwise called PQ droop control
power. The time domain simulation is performed in both the is widely accepted for smooth sharing of active powers
MATLAB/SIMULINK and OPAL-RT digital environments. The among participating DG units keeping the system voltage and
performance assessment of the system along with the aforesaid frequency in prescribed limit. More importantly this control
control strategy effectively maintain the voltage synchronization, aspect does not use any communication link. However this
current sharing accuracy, frequency stability, and active power
sharing of both the inverters. Also under varying load condition, technique fails to yield better reactive power sharing [3]. This
this control scheme distribute the active power and restores the method is best suited for inductive lines whereas for resistive
frequency more efficiently. lines it has poor performance. In [4], a robust droop controller
Index Terms—Decentralized control, droop control, frequency is proposed for high voltage microgrid and also the effect of
regulation, microgrid, synchronization line impedance on load distribution is studied. A traditional
droop control framework is illustrated in [5] for reliable power
I. INTRODUCTION sharing among parallel coupled inverters while maintaining
the voltage and frequency stabilization. However, it suffers
The microgrid (MG) being a prominent concept of a future Slow transient response, line impedance mismatch and weak
grid, comprises of distributed energy resources (DERs), energy harmonic load sharing. To achieve effective power sharing,
storages, power converters with controllers and different loads. many droop based control methods like virtual impedance
There are three main categories of microgrids namely DC, droop [6], power factor angle droop [7], and adaptive droop
AC and hybrid AC/DC microgrid. Each MG has its own [8] give better adjustment of the voltage and frequency profile
advantages. However, now a days an AC microgrid is widely compensating the real and reactive power unbalances in an AC
accepted due to its compatibility with existing AC power net- microgrid. A fully decentralized controller is designed in [9]
work and feasibility with both renewable and non-renewable for grid connected cascaded inverters. Here, the conventional
energy resources. It needs minimum moderations to fit into the droop curves are modified by changing operating points of
existing utility grid which results in less initial investment as distributed generations. For injected real power perturbations,
compared to DC microgrid. Here, the power flow capability is the reactive power sharing error is calculated and is improved
more and the line loss is less. In AC microgrid various issues effectively.
are noticed such as system stability, protection, power quality, In this paper, an efficient droop based decentralized con-
reliability and coordinated control of DERs. Different control troller is presented for an AC microgrid. Basically, the control
strategies are implemented to resolve such issues [1]. technique includes conventional P-f/Q-V and P-V/Q-f droop
978-1-6654-1064-9/22/$31.00 © 2022 IEEE laws which does not use any communication among inverters.
 voltage controlled source which is coupled to the grid via a
line impedance as shown in Fig. 2. The complex power flow
in between VSI and grid is given by the equation (1) [3].
 ∗
V1 − V2
S = P + jQ = V I ∗ = V1
Z
(1)
V12 jθ V1 V2 j(θ+δ)
= e − e
Z Z
The active and reactive power that transfers from the converter
to the load side can be expressed as follows
V12 V1 V2
P = cos(θ) − cos(θ + δ)
Z Z (2)
V2 V1 V2
Q = 1 sin(θ) − sin(θ + δ)
Z Z
The line impedance is Zejθ = R + jX
Fig. 1: Three phase droop-controlled inverter system V1
connected in parallel P = [R(V1 − V2 cosδ) + XV2 sinδ]
R2 + X2 (3)
V1
Q= 2 [X(V 1 − V 2 cosδ) + RV2 sinδ]
When a number of inverters are operated in parallel, to achieve R + X2
better power sharing capability, modified droop controller The inverter and grid voltages are denoted by V1 and V2
having the new set points of voltage and frequency is used. For respectively. The active and reactive power are given by P and
a sudden load change condition, this control scheme manages Q respectively. The impedance angle is θ. The phase difference
the active power sharing, frequency restoration, current sharing between two voltages is symbolized by δ. ϕ represents the
accuracy more precisely. phase difference between inverter voltage and current.
Case I : Neglecting R
II. STRUCTURE OF MICROGRID AND ITS CONTROL
V1 XP
A. System configuration P ≈ V2 sinδ ⇒ δ ≈ (F or small value of δ)
X V1 V2
The microgrid consists of two identical inverters connected (4)
V1 XQ
in parallel and share a common load as shown in Fig. 1. Each Q≈ (V1 − V2 cosδ) ⇒ (V1 − V2 ) ≈
X V1
inverter is fed from separate DC source and is connected to
the load via an LC filter. The inverter is controlled using two For the high voltage transmission line, The droop laws can be
different control loops. The inner loop mainly comprises of given by
the current and voltage controllers which maintains the three f = f0 − m(P − P0 )
phase inverter voltage and frequency. The outer droop based (5)
V = V0 − n(Q − Q0 )
power control loop is responsible for effective sharing of both
active and reactive power among parallel operating inverters. Case II : Neglecting X
The output of inner loop generates the PWM signal for the V1 RP
inverters. P ≈ (V1 − V2 cosδ) ⇒ (V1 − V2 ) ≈
R V1
(6)
B. Mathematical modelling of droop control V1 −RQ
Q≈ (V2 sinδ) ⇒ δ
Droop control is simple and reliable as it assumes a highly R V1 V2
inductive impedance in between inverter and grid. The droop is For low voltage transmission line, the new operating points
characterized by maintaining the output parameters i.e. voltage for voltage and frequency are defined as
and frequency of inverter. The VSI is considered to be a
V = V0 − m(P − P0 )
(7)
f = f0 + n(Q − Q0 )
Where, V and f are the new operating point of voltage and
frequency respectively. V0 and f0 suggest the reference values
of voltage and frequency. The reference values for active and
reactive power is indicated by P0 and Q0 respectively. m and
n are the droop coefficients.
Case III : In this case for finding the new operating point of
both voltage and current, both the resistance and inductance
effects are taken into consideration.
Fig. 2: Equivalent circuit of a VSI connected to grid
 Fig. 3: Modified droop control of an inverter in islanding mode

In a generalized case, if both X and R are taken into TABLE I: Microgrid system and controller parameters
consideration. Then, the use of an orthogonal linear rotational Parameters Value
transformation matrix (T ) from real and reactive power P and DC voltage source 700 V
Q to the “modified” real and reactive power P ′ and Q′ is given VA rating of two inverters 5 kVA
Line inductance 2.35 mH
in equation (8) [6]. Line resistance 0.10 Ω
 ′      Filter capacitance 225 µF
P P sinθ −cosθ P
= T = Filter inductance 3 mH
Q′ Q cosθ sinθ Q m1 = m2 = 9.4e−4 Hz/kW
X R
  (8) Droop coefficients
n1 = n2 =1.3e−4 V/kVAr
Z −Z P
= R X Rated Voltage 220 V
Z Z
Q Rated frequency 50 Hz
PCC Voltage (Line to Line) 311 V
Applying the orthogonal linear rotational transformation, the Switching Frequency (fsw ) 5 kHz
frequency and voltage droop regulation becomes PI control parameters
Voltage control loop
  Kpv = 0.4, Kiv = 240
X R current control loop
f = f0 − m(P ′ − P0′ ) = f0 − m [P − P0 ] − [Q − Q0 ] Kpc = 035, Kic = 12
Z Z
  Loading 2 kW to 4 kW
R X
V = V0 − n(Q′ − Q′0 ) = V0 − n [P − P0 ] + [Q − Q0 ]
Z Z
X R
f = f0 − m (P − P0 ) + m (Q − Q0 ) parameters. The droop coefficients are chosen for ensuring
Z Z
R X stability, proper transient response and good phase matching
V = V0 − n (P − P0 ) − n (Q − Q0 )
Z Z [5]. With sudden load variation, the system is simulated in
(9) both MATLAB/SIMULINK and OPAL-RT digital platform.
The structure of a modified droop controlled inverter system The performance characteristics of the microgrid system using
is shown Fig.3. The inverter current and voltage are used for modified droop law in the real time environment is analysed.
calculation of active and reactive power. Droop control loop
sends the reference values of voltage and frequency to the The total inverter current along with the individual current
inner loop. in inverter 1 and 2 for sudden load disturbance using the
said droop control law is depicted in Fig. 4. As the droop
III. S IMULATION RESULTS coefficients for both the inverters are taken same, the current
An AC microgrid having two three-phase inverters is inves- shared by each inverter becomes equal. Fig. 5 shows the
tigated to validate the performance of both the droop control real time voltage magnitude of each inverter with sudden
schemes. The details of AC microgrid system and controller perturbation. It is observed that both the inverters maintain
parameters, along with inverter specification is given in Table synchronism throughout its operation. The total three phase
I. Here both simplified droop and modified droop laws are load current is shown In Fig. 6, which is the sum of the
realized for the said system taking the mentioned controller currents shared by individual inverter. With sudden decrease in
Fig. 4: Total and individual inverter current with load change Fig. 7: Three phase load voltage with load disturbance

Fig. 5: Individual inverter voltage with load change Fig. 8: Active power shared by inverters with load change

Fig. 6: Three phase load current with load disturbance Fig. 9: Frequency response of inverters with load change

load, the load current decreases proportionately to their droop sudden load change is depicted in Fig. 8. The corresponding
coefficients. It is noticed that the load voltage which is shown frequency response of individual inverter is shown in Fig.
in Fig. 7 is synchronized with the output voltage of individual 9. Even though the disturbance is made, the frequency level
inverter. of both the inverters merely get affected. It maintains the
The active power equally shared by each inverter during a prescribed frequency limit. To observe the frequency response
 IV. CONCLUSION
This article focuses on a droop based decentralized control
configuration in a system of two 3-phase inverters. Two
distinct droop laws are used in the control. The proportionate
load sharing is provided from each inverter in the AC mi-
crogrid. Active and reactive power have a direct relationship
with frequency and voltage respectively for simplified control,
however new operating frequency and voltage set points in
modified droop are dependent on both real and reactive
powers. The real-time results simulated in OPAL-RT platform
using modified droop law accurately shares current and active
power. Also, a comparative analysis of frequency restoration
(a) using both control schemes is presented in the simulation
results. The frequency drop in modified control is slightly more
than the simplified droop as the effect of both inductance and
resistance are taken into consideration.
R EFERENCES
[1] B. Sahoo, S. K. Routray, and P. K. Rout, “AC, DC, and hybrid control
strategies for smart microgrid application: A review,” Int Trans Electr
Energ Syst., vol. 31, no. 1, pp. e12683, 2021.
[2] M. Castilla, LG. Vicuña, and J. Miret. “Control of power converters in
AC microgrids.” In Microgrids design and implementation, pp. 139-170.
Springer, Cham, 2019.
[3] M. Arias, D. G. Lamar, M. RodrIguez, M. Hernando, and A. Fernandez,
“Simple droop voltage control system for parallel operation of UPS,”
2008 Twenty-Third Annual IEEE Applied Power Electronics Conference
and Exposition, pp. 1946-1951, 2008.
[4] Z. Shuai, S. Mo, J. Wang, Z. J. Shen, W. Tian, and Y. Feng, “Droop
(b) control method for load share and voltage regulation in high-voltage
microgrids,” Journal of Modern Power Systems and Clean Energy, vol.
4, no. 1, pp. 76-86, 2016.
[5] A. El Moubarek Bouzid, P. Sicard, A. Yamane, and J. Paquin, “Simulation
of droop control strategy for parallel inverters in autonomous AC micro-
grids,” 2016 8th International Conference on Modelling, Identification
and Control (ICMIC), pp. 701-706, 2016.
[6] W. Yao, M. Chen, J. Matas, J. M. Guerrero, and Z. Qian, “Design and
Analysis of the Droop Control Method for Parallel Inverters Considering
the Impact of the Complex Impedance on the Power Sharing,” IEEE
Transactions on Industrial Electronics, vol. 58, no. 2, pp. 576-588, 2011.
[7] Y. Sun, L. Li, G. Shi, X. Hou, and M. Su, “Power Factor Angle Droop
Control—A General Decentralized Control of Cascaded Inverters,” IEEE
Transactions on Power Delivery, vol. 36, no. 1, pp. 465-468, 2021.
[8] M. M. Bijaieh, O. M. Anubi, and C. Konstantinou, “Distributed Adaptive
AC Droop Control in D-Q Coordinates for Inverter-Based Microgrids,”
2020 IEEE Power & Energy Society General Meeting (PESGM), pp. 1-5,
2020.
[9] X. Hou, Y. Sun, H. Han, Z. Liu, W. Yuan, and M. Su, “A fully decen-
(c) tralized control of grid-connected cascaded inverters,” IEEE Transactions
on Sustainable Energy, vol. 10, no. 1, pp. 315-317, 2019.
Fig. 10: (a) Frequency response in simple droop (b)
Frequency response in modified droop (c) Comparision

of both the droop laws, simulation is carried out with load

change from 2 kW to 4 kW during the time interval of 0.4
sec to 0.6 sec. The simulation results of frequency response of
individual inverter using simplified and modified droop laws
are represented in Fig. 10 (a) and (b) respectively. Moreover,
the comparison of frequency response between these two
droop laws is shown in Fig. 10 (c). It is observed, as compared
to simplified droop, the frequency drop in modified droop
technique is somewhat higher since, frequency is dependent
on both the active and reactive power in this situation.