Original Research Article

Wind Engineering
1–19
Operation and control of a stand-alone Ó The Author(s) 2021
Article reuse guidelines:
power system with integrated multiple sagepub.com/journals-permissions
DOI: 10.1177/0309524X211024126

renewable energy sources journals.sagepub.com/home/wie

Nindra Sekhar and Natarajan Kumaresan

Abstract
To overcome the difficulties of extending the main power grid to isolated locations, this paper proposes the local installation of a com-
bination of three renewable energy sources, namely, a wind driven DFIG, a solar PV unit, a biogas driven squirrel-cage induction gen-
erator (SCIG), and an energy storage battery system. In this configuration one bi-directional SPWM inverter at the rotor side of the
DFIG controls the voltage and frequency, to maintain them constant on its stator side, which feeds the load. The PV-battery also sup-
plies the load, through another inverter and a hysteresis controller. Appropriately adding a capacitor bank and a DSTATCOM has also
been considered, to share the reactive power requirement of the system. Performance of various modes of operation of this coordi-
nated scheme has been studied through simulation. All the results and relevant waveforms are presented and discussed to validate the
successful working of the proposed system.

Keywords
Integration of renewable energy sources, induction generators, solar PV system, power electronic controllers, real and reactive power
control, maximum power point tracking

Introduction
Rural electrification and supply of power to isolated communities with small population has become a global con-
cern. Due to geographical constraints and grid augmentation costs, transmitting power to such locations, by
extending the main grid, is a very difficult task. Presently, most of the developed island communities in remote
places, rely on traditional diesel generators for power supply (Merabet et al., 2017; Prakash et al., 2016; Singh and
Bansal, 2018). Continuous increase in fuel cost, disturbance to ecological system through air pollution, noise,
water, and soil, make the further employment of these systems expensive and hence unfavorable. However, many
of these locations have substantial availability of renewable energy sources (RESs) like wind, solar, hydro, bio-
mass, etc. The clean energy from such sources cause no problems to the public health system, helps to reduce
energy cost and gain energy independence to households and these sources require very little maintenance (Diab
et al., 2019; regenpower.com, 2020; Wang et al., 2019). While planning RESs, certain issues like, upfront expense
of building and installing solar and wind farms, controlling the output voltage and frequency on the load side,
and ensuring cost effective and reliable storage methods must be considered.
The development in power electronic controllers and electrical machines is very helpful in overcoming
these concerns and RESs are being increasingly used to supply electricity economically for wide range of
applications (Arthishri et al., 2019; Ramabhotla and Bayne, 2019; Rezkallah et al., 2019; Wang et al., 2021).
In this context, risk of cyclones, extended wet weather, dust storms, and other such natural occurrences and
more particularly, the high cost of battery storage systems have necessitated the deployment of an additional

Department of Electrical and Electronics Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu, India

Corresponding author:
N Kumaresan, Department of Electrical and Electronics Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu 620015, India.
Email: nkumar@nitt.edu
2 Wind Engineering 00(0)

source of constant power, such as a gas generator of reasonable capacity, for certain off-grid applications
(Al-Ammar et al., 2020; Kant et al., 2017). So, the installations should be configured such that energy could
be shared primarily from different types of renewable energy sources along with adequate energy storage and
reactive power support systems. Improvements in the performance of these installations, including the appro-
priate measuring and communication systems require special support or incentives to justify hybrid applica-
tions (cat.com, 2016; Gupta et al., 2021; Yang et al., 2019). This approach not only reduces the operating
expenses when compared to purely conventional generation but also optimizes the system reliability, effi-
ciency, and flexibility (Das et al., 2019).
In the study of wind generator, Singh et al. (2020) have addressed the problems associated with high operating
temperatures of windings and bearings of the generators and have proposed design solutions for the same. Cooling
configurations based on stipulated IEC/IEEE standards have also been discussed (Singh et al., 2020). Certain bene-
ficial recommendations have been offered to enable the functioning of the generators within the safe operating tem-
perature. A thorough mathematical analysis has been illustrated to improve the overall efficiency of the system,
even if certain components could not be aligned perfectly (Singh and Sundaram, 2020). The selection procedure
has been described for the wind generator winding insulation, to withstand harmonic voltage peaks caused by the
converters employed with the generators (Singh et al., 2019).
The standalone power system network proposed in this paper is a self-sustaining and off-the-grid network with
energy generated from three different locally accessible sources, namely, wind, solar, and biogas. Such hybrid net-
work is normally supported by an energy storage system because the standalone power system is not connected to
any conventional grid (Adefarati et al., 2017; Kusakana, 2015). In the proposed system, wind and solar energy are
extracted through Doubly-fed induction generator (DFIG) and PV-MPPT (maximum power point tracking) sys-
tem, respectively. Among all the existing technologies in wind energy electric conversion system (WEECS), the
DFIG is now widely accepted, because it offers advantages viz. low rating converter for power conversion, capabil-
ity to capture the maximum energy from a wide range of wind velocities, variable speed operation, and enhanced
system efficiencies (Navin Sam et al., 2017a; Vijayakumar et al., 2012, 2013a, 2013b). DFIGs are used extensively
in high power wind turbine generators as distributed generation units (DGs) in micro grids because they supply
power at constant voltage and frequency both in sub synchronous and super synchronous speeds (Fu et al., 2018;
Navin Sam et al., 2017; Puchalapalli and Singh, 2020; Sekhar et al., 2019; Tiwari et al., 2018; Vijayakumar et al.,
2012, 2013a).
With a view to minimize the machine loss in wind-driven DFIG systems, Bagarty and Kastha (2013),
Marhaba et al. (2018), Hore and Sarma (2018), and Navin Sam et al. (2015, 2017b) have developed an algo-
rithm for determining the optimal share of reactive power for the stator side and rotor side converters of the
DFIGs employed in such units. So, in this paper, it is proposed to connect a fixed capacitor bank of appropri-
ate value, which will supply a fixed reactive power for the entire operating region and the deficit reactive power
is met by battery inverter system (BIS). It is to be noted that Navin Sam et al. (2015, 2017b), Venkatraman
et al. (2016), and Guchhait et al. (2020) have shown that the capacitor at the stator side should be varied dyna-
mically so that the net reactive power at the stator terminals will vary accordingly, as per the reactive power
requirement of the load, for operating the system with minimum power loss. Instead, for achieving this, an
attempt is made in this present paper by employing a distribution static compensator (DSTATCOM) at the
stator terminals. Its DC side voltage can be maintained with the power available from the solar panel-battery
system, while it caters to the load demand and the dynamic reactive power requirement of biogas driven
SCIG.
The proposed system consists of a wind-driven DFIG, solar PV excited Battery inverter system, and a bio-
gas engine driven SCIG. DSTATCOM is connected to the point of common coupling (PCC) of the hybrid
power system network with Solar PV-battery system on DC side and step up transformer on AC side. In this
system only one bi-directional Sinusoidal pulse width modulation (SPWM) inverter is employed at the rotor
side of DFIG. The stator voltage and frequency are maintained constant at PCC by appropriately controlling
the rotor voltage and slip frequency. Perturb and observe (P&O) method is applied to extract maximum
power from PV panel, because of its simplicity and easy implementation. Section 2 gives the block diagram
of the system components and lists the features of the various operating modes employed in the hybrid sys-
tem. DFIG and PV-MPPT control strategies are elaborated in section 3. The performance characteristics of
the system obtained by simulation, under various modes adopted, are presented, and discussed in detail in
section 4.
Sekhar and Kumaresan 3

Figure 1. Block diagram of proposed hybrid power system configuration integrated with three renewable energy sources.

Proposed hybrid system configuration and various modes of operation

The proposed hybrid system is shown in Figure 1. The DFIG and SCIG operate at variable speed by wind turbine
(WT) and biogas engine, respectively. The simple integration of these power generating units is possible by feeding
a voltage of appropriate magnitude and frequency to the rotor terminals of DFIG through an SPWM inverter
and employing a suitably designed and operated closed loop control strategy. PV panels are connected to a battery
bank through a boost converter and the PV- battery system forms the DC link of the DFIG rotor-inverter as well
as DSTATCOM. While acting as an energy storage system, the battery bank also gives a stiff DC link and makes
the boost converter output voltage virtually constant. The PV-battery system based DSTATCOM is connected
through a Y-Y step up transformer to a 415 VL-L distribution system which provides a pathway for supplying a
fraction of the load real/reactive power directly from the PV array. The battery bank is rated at 440 V, 100 Ah.
Various modes of operation of the hybrid system are possible, based on energy generation by the different
sources and reactive power support provided by the capacitor bank and DSTATCOM and these are listed in
Table 1. In each mode, there could be different cases to account for the variation in the load and the speed of the
DFIG and the BGG (Biogas Generator). In all these cases, the voltage and frequency are maintained constant at
PCC, irrespective of the variation in irradiation. The real and reactive power balance of the hybrid power system
in the different modes and cases are achieved through the closed loop control of a PV fed SPWM inverter at the
DFIG rotor.

DFIG closed loop controller and PV MPPT controller

DFIG controller must act to maintain the PCC voltage at a constant magnitude of 415 V and a frequency of
50 Hz. A simple proportional-integral (PI) controller has been used for the same, with parameter values of
Kp = 0.09, and Ki = 2.82, obtained through manual tuning. This controller consists of two parallel loops—one
for regulating the stator voltage and the other for frequency, as depicted in Figure 2. In the voltage control loop,
the voltage is sensed, and its root mean square (RMS) value is calculated. The variation in Vrms from Vref is
obtained as Verr. The PI controller generates a control signal corresponding to the required modulation index
(MI). Then, fr is computed using the rotor speed, N. After estimating fr, the unit magnitude sine wave of frequency
fr is generated. This sine wave is multiplied with the MI estimated based on the Verr shown in Figure 2, to obtain
4 Wind Engineering 00(0)

Table 1. Various operating modes of the hybrid power system.

Operating modes of the hybrid power system DFIG speed SCIG speed Irradiation Load

Mode 1: without external Q support (a) Constant Varying Varying Constant

(b) Varying Constant Varying Constant
(c) Constant Constant Varying Step change (R load)
Mode 2: Q support from three phase capacitor bank (a) Constant Varying Varying Constant
(b) Step change Constant Varying Constant
(c) Constant Constant Varying Step change (R-L load)
Mode 3: P and Q support (for load) from Constant Constant Varying Step change
DSTATCOM and Q support
from three phase capacitor bank
Mode 4: wind speed below cut in speed (a) Disconnected Varying Varying Step change

Figure 2. DFIG closed loop control scheme.

the desired voltage waveform, V*Sabc. V*Sabc is continuously compared with the carrier wave to get the SPWM
pulses for the inverter, which supplies the required three-phase voltage to the rotor terminals of DFIG. Thus, the
closed-loop controller shown in Figure 2 provides the rotor with an appropriate voltage to create the virtual grid
at the stator terminals, by maintaining a three-phase constant voltage of fixed frequency under any situations.
A DC-DC boost converter is designed and modeled using Simulink block set to interface the PV array with the
battery bank and to extract maximum power from the array. As boost converter output voltage, VBOOST is fixed
(445 V), any change in PV irradiation level is reflected in the magnitude of the boost converter output current,
IBOOST alone. Hence, here it is sufficient to sense only one signal, IBOOST for tracking the maximum power point.
The PV-MPPT controller computes the duty cycle of the power switch by sensing IBOOST and ensures that this
value is maximum corresponding to the irradiation level. The switching pulse for the power metal-oxide-
semiconductor field-effect transistor (MOSFET) is generated by comparing the duty cycle value (limited to a
range of 0.20–0.75 for safe operation of boost converter) with a 10 kHz sawtooth signal of unit magnitude. The
PV power generated is utilized for charging the battery bank as well as feeding the required DFIG rotor power
and the set value of the real-reactive power of the DSTATCOM.

Results of various modes of the hybrid power system

Mode 1: Wind, PV, and biogas generator feeding isolated load without external reactive power support
In this mode, the study has been carried out without external reactive power support and DSTATCOM. The reac-
tive power required for SCIG operation is furnished by the DFIG. The speed of both DFIG and SCIG is kept
Sekhar and Kumaresan 5

Figure 3. Waveforms of PV-boost converter system and DFIG for different irradiation levels: (a) solar irradiation variation, (b) PV
array terminal current, (c) terminal current of DC-DC boost converter, (d) voltage and power output of PV array, (e) battery
terminal voltage, and (f) DFIG stator terminal voltage.

constant/varied with continuous variation of solar irradiation, for evaluating the efficacy of this multiple renew-
able sources fed hybrid power system. The rating of DFIG is 3 hp, 415 V, 50 Hz. A set of 10 PV panels, connected
in series forming a PV array, was considered for simulation. The power rating of each panel is 210.16 W with an
open-circuit voltage of 35.7 V and a short-circuit current of 7.85 A at standard test conditions. The panel voltage
at maximum power point, Vmp is 28.4 V and the corresponding current Imp is 7.4 A. Thus, the maximum power
Pmp of the PV array is 2.1 kW. To emulate the biogas generator, a 3 hp, 415 V, 50 Hz squirrel-cage induction gen-
erator has been modeled in MATLAB/Simulink.

Case (a): Constant DFIG speed and load, varying SCIG speed, and irradiation level. The speed of DFIG is set at 1300 rpm.
It is to be noted that the SCIG is directly connected to the PCC (without out any power electronic controller) and
hence it is to be run at more than its synchronous speed for operating as a generator. So, based on the rotational
speed, the SCIG will supply the real power at PCC and takes reactive power for its operation. The solar irradia-
tion level is varied continuously between 300 and 1000 W/m2 as given in Figure 3(a). The load connected to the
system is 2 kW, unity power factor (UPF).
A simple P&O algorithm for tracking the maximum power of PV array has been programmed with an initial
value of the duty cycle as 0.3. For a given duty cycle, the output voltage of PV array, VPV will be constant and
equal to the fixed battery terminal voltage, VBATT. IPV, corresponding IBOOST and PPV and VBATT are shown in
Figure 3(b) to (e). The power supplied by PV array is to be utilized for charging the battery bank when it is in
excess of the requirement of the DFIG rotor power. It is assumed that the initial state of charge (SOC) of the bat-
tery is 92%, which leads to a voltage of 445 V, as shown in Figure 3(e).
It is well-known that the power output of a grid connected SCIG is constant for a given running speed, NSCIG.
While this can be a potential advantage in the operation of a microgrid, the drawback is that it also draws reactive
power from the grid for its operation, that is, from the DFIG since in this case, there is no other active AC three-
phase source in the system. Hence, the constraint in this case is that the DFIG must supply the fraction of load as
well as the reactive power requirement, QSCIG even at low values of DFIG rotor speed, NDFIG. Second challenge is
that the controller should be capable of maintaining the stator voltage magnitude, VDFIG ST irrespective of the var-
iation in NSCIG and the magnitude of load power, PLOAD.
The efficient operation of the controller considering all the above factors for a PLOAD of 2 kW UPF, NDFIG of
1300 rpm, and varying NSCIG can be seen in Figure 3(f). The magnitude of VDFIG ST is maintained as 415 V.
The rotor circuit and its resulting effects on the stator side of DFIG, that is, at the PCC of hybrid power system
are studied and presented. The terminal voltage of SCIG and load will be the same as that of the DFIG. Hence, to
show the sharing of power by these two induction machines, only the current waveforms are shown along with the
load current waveform in Figure 4.
6 Wind Engineering 00(0)

Figure 4. Three-phase current waveforms: (a) DFIG stator terminal, (b) SCIG terminal, and (c) load terminal.

Figure 5. Real and reactive powers at different terminals: (a) PLOAD, (b) PDFIG, (c) PSCIG, (d) QDFIG, and (e) QSCIG.

It is observed that when NSCIG increases from 1520 to 1590 rpm at time t = 3 seconds, there is an increase in
the stator current of SCIG, ISCIG. During this interval, the stator current of DFIG, IDFIG ST decreases to balance
the load requirement. This change in real power supplied by DFIG is reflected in the rotor inverter MI as men-
tioned earlier. On the contrary, when NSCIG decreases from 1590 to 1550 rpm at time t = 7 seconds, ISCIG also
decreases, which leads to an increase in DFIG stator current, IDFIG ST to balance load requirement.
The power supplied by the DFIG, PDFIG and by SCIG, PSCIG are given in Figure 5 and power absorbed by the
load is shown in Figure 5(a). It is clearly seen from Figure 5(b) and (c) that with increased speed of operation,
the SCIG delivers more power to the load, thereby reducing the real power delivered, PDFIG by the DFIG. Also,
the reactive power needed for SCIG is supplied by the DFIG, QDFIG that is, QSCIG = QDFIG. Thus, it can be seen
that the DFIG is capable of generating real and reactive power independently.
It can be inferred that change in irradiation level does not have any effect on the three-phase PCC hybrid system
voltage, and hence on the power sharing by the DFIG and SCIG. Although the PV-Battery system forms the DC
link of the DFIG rotor-inverter, for the sake of brevity, henceforth, the DC side waveforms are not presented. It is
to be noted that the PV-battery system is present in all modes of operation that follow.

Case (b): Constant SCIG speed and load, varying DFIG speed, and irradiation level. In this mode, DFIG is operated at differ-
ent speeds, while the NSCIG is maintained constant at 1575 rpm. The PV irradiation level is varied continuously
between 300 and 1000 W/m2. The load connected in the system is 3 kW, UPF.
Figure 6(a) and (b) shows the injected voltage and current in the rotor circuit of the DFIG. It is observed that
the frequency of the rotor circuit waveforms varies with its speed and is equal to the corresponding slip frequency.
The magnitude of the rotor voltage varies with the MI generated by the DFIG controller. The corresponding stator
terminal voltage is maintained at 415 V as shown in Figure 6(c). For the different running speeds, the stator fre-
quency is maintained at 50 Hz by the injected voltage in the rotor circuit at slip frequency. Load power consump-
tion is shown in Figure 6(d). The power sharing by the SCIG and DFIG is shown in Figure 6(e) and (f) for feeding
the connected 3 kW, UPF load. The DFIG delivers the reactive power required by the SCIG for its operation as
depicted in Figure 6(g) and (h)
Sekhar and Kumaresan 7

Figure 6. Electrical parameters of the hybrid system for case (b): (a and b) DFIG rotor terminal voltage and current, (c) DFIG
stator terminal phase voltage, (d) Load real power demand, (e and f) Real power supplied by DFIG and SCIG for load, (g) Reactive
power supplied by DFIG to SCIG, and (h) Reactive power absorbed by SCIG from DFIG.

Figure 7. Response of the hybrid power system during step change in load for case (c) of Mode1: (a) DFIG stator terminal voltage
(b) Load real power demand and, (c and d) Real power supplied by SCIG and DFIG.

Case (c): Step change in load. To demonstrate the effective operation of the system for a sudden change in load, the
NDFIG is set at 1300 rpm, NSCIG at 1560 rpm, and the irradiation level is changed continuously. The load is chan-
ged from 3 to 1.5 kW at time t = 3 seconds.
In this case, it can be seen from Figure 7(a) that the DFIG stator terminal voltage is maintained as 415 V. As
the SCIG output power is uncontrolled, it will deliver a fixed power corresponding to the running speed and so
the power generated from DFIG will vary to meet load demand as seen in Figure 7(b) to (d). The magnitude of its
rotor voltage and current decreases when the load decreases from 3 to 1.5 kW as seen from Figure 8(a) and (b).
Correspondingly, its rotor real and reactive power also decrease as shown in Figure 8(c) and (d). It can be seen that
the change in real power demand at stator side reflects in rotor power of DFIG, PDFIG ROTOR. However, as the
reactive power demand at stator side is the same, the change in rotor reactive power is small. For a fixed SCIG
speed, its reactive power demand is met by the DFIG, as shown in Figure 8(e) and (f).
From all the three cases of operation of the hybrid system under Mode 1, it was observed that QSCIG = QDFIG.
This poses a constraint on the maximum load that can be supplied by the system. As the DFIG and SCIG are rated
for only 3 kVA each, the actual total real power delivered should be 4.5 kW, and reactive power 3.96 kVAr,
8 Wind Engineering 00(0)

Figure 8. Variation of voltage, current, real, and reactive power in DFIG and SCIG for load change: (a and b) DFIG rotor terminal
voltage and current, (c and d) Real and reactive power at DFIG rotor terminal, (e) Reactive power supplied by DFIG to SCIG and,
(f) Reactive power absorbed by SCIG from DFIG.

Figure 9. DFIG stator terminal voltage.

considering that both machines deliver rated reactive power. Hence, it is suggested that an external reactive power
source, such as a capacitor bank could be employed, that will meet out the QSCIG demand alone. In that case, the
load connected to the proposed system can be increased. The simulation of the proposed hybrid power system with
capacitor bank connected at the PCC is carried out and results are presented in the following subsection in
Mode 2.

Mode 2: Reactive power sharing between wind and biogas generator and three-phase capacitor bank
Case (a): Constant DFIG speed and load, varying SCIG speed, and irradiation level. A delta connected capacitor bank of
2 kVAr is added at the stator terminals of DFIG. NDFIG is set at 1300 rpm and an R-L load of rating 3.75 kVA,
0.8 p.f. (lag) is connected to the system. NSCIG is varied from 1520 to 1590 rpm. Irradiation level is changed con-
tinuously. It can be seen from Figure 9 that the voltage at PCC of the system is maintained as 415 V, irrespective
of change in NSCIG.
The real power sharing by DFIG, SCIG, and reactive power sharing by three-phase capacitor bank, QCAP are
shown in Figure 10. It can be seen that SCIG and DFIG supply PLOAD = 2400 W, while the reactive power
demand of the load, QLOAD = 1800 VAr is met by the DFIG. In this mode, QSCIG = QCAP. It is to be noted that
the QSCIG varies proportionally with NSCIG. When QSCIG exceeds QCAP, the reactive power shortage is met by the
DFIG. It is to be noted that the real/reactive power balance is maintained for all operating points. The corre-
sponding rotor real and reactive power waveforms are shown in Figure 10(h) and (i).

Case (b): Constant SCIG speed and load, step change in DFIG speed, and varying irradiation level. In this mode, NDFIG is chan-
ged from 1250 to 1350 rpm at time t = 3 seconds, while the NSCIG is maintained constant at 1575 rpm. The PV
irradiation level varies continuously. The load connected in the system is 3 kVA, 0.9 p.f. (lag) R-L load. In this
mode of operation also, the stator terminal voltage of DFIG is maintained as 415 V, as in Figure 11.
As the value of NSCIG is constant, the power generated by SCIG and reactive power absorbed for the operation
will also be constant. With variation in NDFIG, the real power balance for the load is maintained by the DFIG
and SCIG and reactive power balance is maintained by DFIG and capacitor bank in tandem, while SCIG delivers
fixed real power and absorbs fixed reactive power. Figure 12(a) to (g) show the real/reactive power balance in this
hybrid power system when NDFIG is changing. The frequency of rotor variables is equal to the corresponding slip
frequency, as shown in Figure 12(h) and (i).
Sekhar and Kumaresan 9

Figure 10. Real/reactive power at different terminals: (a and b) load, (c and d) DFIG stator, (e and f) SCIG (g) capacitor bank, and
(h and i) DFIG rotor.

Figure 11. DFIG stator terminal voltage.

Figure 12. Electrical parameters at different terminals of hybrid system: (a and b) load, (c and d) DFIG, (e and f) SCIG, (g)
capacitor bank, and (h and i) DFIG rotor voltage and current waveforms.

Case (c): Step change in load. To demonstrate the effective operation of the system for sudden change in load, the
NDFIG is set at 1300 rpm, NSCIG at 1575 rpm, and the irradiation level is changed from 1000 to 900 W/m2. The
load connected in the hybrid power system is changed from 3 kVA, 0.9 p.f. (lag) to 4 kW, UPF at time t = 3 sec-
onds. In this mode of operation also, the DFIG stator terminal voltage shown in Figure 13 is maintained as 415 V.
10 Wind Engineering 00(0)

Figure 13. DFIG stator terminal voltage.

Figure 14. Real/reactive power at different terminals: (a and b) load, (c and d) DFIG stator, (e and f) SCIG, (g) capacitor bank, and
(h and i) DFIG rotor.

As the value of NSCIG is constant, the power generated, and the reactive power absorbed by the SCIG will also
be constant. With variation in load, the real power balance for the load is maintained by the DFIG and SCIG.
Reactive power balance is achieved by DFIG and capacitor bank in tandem with SCIG. Figure 14(a) to (g) show
that the real/reactive power balance in the system is maintained for step change in load. The corresponding rotor
power injected is shown in Figure 14(h) and (i).
In mode1, the maximum load that could be supplied from DFIG and SCIG is limited to each 3 kW, UPF.
Throughout mode 2, it was seen that the inclusion of the 2 kVAr delta-connected capacitor enabled operation of
the system for loads greater than 3 kW, UPF as well as lagging power factor linear loads. The response of the sys-
tem to change in NSCIG, NDFIG, Irradiation, and load was tested and presented. It is observed that it is possible to
expand the range of loads in the system, if the reactive power of the SCIG is met by an external apparatus. In the
next mode 3, the operation of the hybrid power system is tested when a DSTATCOM is interfaced in the PCC to
deliver a portion of the load real power and complete load reactive power requirements. This is carried out to
make a pathway for direct power flow from the PV-battery system to the PCC of the hybrid power system.

Mode 3: Wind, PV, and biogas generator feeding isolated load with real and reactive power support (for load
alone) from DSTATCOM and reactive power support from three-phase capacitor-bank
In this mode of operation, three-phase terminals of the DSTATCOM module is connected at the PCC of the
hybrid power system. The PV-battery system forms the DC link of the DFIG rotor-inverter as well the
DSTATCOM. The DSTATCOM is operated to supply 50% of the PLOAD and complete reactive power demand
of load, QLOAD. The NDFIG is set as 1300 rpm, and NSCIG is set as 1550 rpm. Irradiation level is initially equal to
1000 W/m2. From 5 to 11 seconds, irradiation is considered as 800 W/m2, thereafter it is changed to 950 W/m2.
Initially, the DFIG-PV-Battery-SCIG system is allowed to settle down before the DSTATCOM breaker is closed
at time t = 2.1 seconds. The load connected in the system is changed periodically to show the efficient power
sharing by the three active sources in the hybrid power system. The different load settings are shown in Table 2.
Sekhar and Kumaresan 11

Table 2. Load setting for Mode 3.

Time, t (seconds) Load present in the system

0\t ł 4 3 kW, UPF

4\t ł 7 4 kW, UPF
7 \ t ł 10 4 kVA, 0.8 p.f. (lag)
10 \ t ł 13 3 kVA, 0.8 p.f. (lag)

Figure 15. Voltage at PCC of hybrid power system: (a) per-phase voltage waveform and (b) per-phase rms value.

Figure 16. Block diagram representation of reference current generation for DSTATCOM.

For successful operation of the DSTATCOM, it is essential that the DFIG stator terminal voltage behaves as a
stiff three-phase grid and as shown in Figure 15, as the DFIG controller maintains this grid at a voltage of 415 V,
for all operating conditions. This has been achieved irrespective of multiple sources injecting power into this hybrid
system.
First, the working of DSTATCOM is explained through reference current generation as depicted in Figure 16
and simulation results, followed by demonstration of power sharing by the different sources in the system. Lastly,
the waveforms of the PV-battery system, rotor-side inverter, and DSTATCOM DC-link are presented to show the
overall integrated operation of the proposed hybrid power system.
The purpose of the DSTATCOM is to provide a pathway for supplying a fraction of the load real/reactive
power directly from the PV array. As the load terminal voltage is fixed, the load kVA will be reflected in the cur-
rent drawn by the load alone. So, it is sufficient to sense the load current for determining the load power that is to
be compensated by DSTATCOM. As shown in Figure 16 the three-phase current at the load terminal is trans-
formed to the dq0 axis with the d-axis aligned with phase-a of ILOAD. By transforming the three-phase waveform
from abc-domain to dq0-domain, the oscillating time domain waveform is converted to a more convenient DC
form for the purpose of easy control action. The selection of reference waveform to compute angle theta, u for
transforming the variables in abc domain to dq0 domain is carried out. However, since the PCC phase voltage will
be in-phase with the load phase current for UPF load, the three-phase to ground voltage at DFIG terminal has
been considered for evaluating angle u, which is computed using abc-ab transformation.
12 Wind Engineering 00(0)

Figure 17. (a) Angle of DFIG terminal phase voltage and (b and c) load current in dq reference frame and reference current for
DSTATCOM.

Figure 18. Current injected by DSTATCOM in the transformer windings: (a and b) Current injected in transformer primary and
secondary winding.

Figure 19. Reference current tracking by HCC: (a) Current injected in transformer primary winding following load reference
current, and (b) Equivalence of injected currents in transformer primary and secondary winding.

The component of load current along the d-axis, IDL represents PLOAD, whereas the component along q-axis,
IQL represents QLOAD. In order that 50% of PLOAD is supplied from DSTATCOM, the magnitude of correspond-
ing reference current, IDL REF will be 0.5 3 IDL with K = 0.5. As QLOAD will be supplied as such from
DSTATCOM, the magnitude of corresponding reference current, IQL REF will be equal to IQL. These reference
currents IDL REF and IQL REF are transformed back to the abc-coordinate system as shown in Figure 16. This will
be the current that the DSTATCOM will inject into the PCC to compensate for the given load power setting. As
the DC-link voltage is 445 V, a step-up transformer with a turns ratio of 1.667 has been used to interface the
DSTATCOM with the PCC. The values of IDL and IQL are scaled appropriately before computing IDL REF and
IQL REF.
Figure 17(a) shows the angle of grid phase voltage, evaluated through abc-ab transformation. The load phase
current is transformed from abc-ab frame of reference using the computed angle u. Subsequent transformation of
load current from ab to dq0 frame of reference is carried out to get IDL and IQL. The current required to be
injected by the DSTATCOM is computed by the control algorithm and reference currents IDL REF and IQL REF
are also computed. The reference currents for different load settings are shown in Figure 17(b) and (c).
For injecting the current into PCC, the reference currents must be transformed back to the abc domain through
dq0-ab transformation, followed by ab-abc transformation. The generated reference current, IL REF is the refer-
ence current that is to be injected into the PCC by DSTATCOM to achieve load real/reactive power
compensation.
A hysteresis current controller (HCC) of bandwidth 0.02, compares the IL REF with the current injected in the
primary winding of the step-up transformer, IINJ PRI continuously. Based on the deviation between IL REF and
IINJ PRI, the pulses for the upper and lower insulated-gate bipolar transistor (IGBT) switches in the three legs of
the DSTATCOM inverter are switched ON/OFF. The corresponding waveforms of primary and secondary wind-
ing currents are shown in Figure 18.
The reference current tracking by the HCC is shown in Figure 19 along with the load current and currents in the
transformer primary and secondary windings. The injected primary current IINJ PRI is made to follow the generated
Sekhar and Kumaresan 13

Figure 20. Current measured at different terminals in the hybrid system for Mode 3: (a) Load current (b) DFIG stator terminal
current (c) SCIG current, and (d) DC-link current.

Figure 21. Real and reactive power balance before and after closing DSTATCOM: (a) Real power absorbed by load, (b, c and d)
Real power supplied by DSTATCOM, DFIG stator terminals and SCIG, (e) Reactive power absorbed by load, (f and g) Reactive
power supplied by DFIG stator terminals and DSTATCOM, and (h and i) Reactive power balance in SCIG and Capacitor bank.

IL REF by the action of this controller. Also, the equivalent transformer primary and secondary currents are in-phase
which is an indication that the phase angle of the current waveform that is injected in the PCC is preserved.
Before closing the DSTATCOM breaker, the load current is supplied by the DFIG and SCIG. After closing
the breaker, the DSTATCOM injects power into the PCC of the system. As ISCIG is fixed due to constant NSCIG,
before and after the DSTATCOM is connected, the DFIG stator current will reduce for maintaining the power
balance in the system. This is shown in Figure 20.
The real/reactive power balance in the proposed system is shown in Figure 21. The SCIG generates a constant
power of 1.4 kW before and after DSTATCOM breaker is closed at t = 2.1 seconds. In the time interval 0 \ t
ł 2.1 seconds, DFIG supplies the remaining 1.96 kW to meet load power requirement. However, QDFIG is almost
zero as QLOAD is equal to zero, and QSCIG = QCAP. When the breaker is closed at t = 2.1 seconds, the real power
injected by DSTATCOM, PSTATCOM is nearly 50% of PLOAD, that is, 1.46 kW and reactive power injected by
DSTATCOM, QSTATCOM is zero, as the load is a purely resistive one. Now, PDFIG decreases to 2490 W to main-
tain the power balance in the system. In case of lagging p.f. loads, it can be seen that QSTATCOM = QLOAD,
PSTATCOM = 0.5 PLOAD, PSCIG = 1.04 kW, and PDFIG = PLOAD 2 PSCIG 2 PSTATCOM. However,
QDFIG G 0, when DSTATCOM is connected in the system.
The corresponding power flow in the DC side of the proposed hybrid system is presented. The DC link voltage
is maintained as 445.1 V by the battery bank, which is charged when PPV is in excess of DC side power at rotor
inverter and DSTATCOM. The corresponding currents satisfy Kirchhoff’s Current Law (KCL) at all times, irre-
spective of the load connected or variation in NDFIG, NSCIG, and irradiation. This can be inferred from Figure 22.
14 Wind Engineering 00(0)

Figure 22. Power flow in DC side of the proposed hybrid system and corresponding response form rotor side inverter: (a and b)
Battery bank voltage and current, (c and d) PV array and boost converter output current, (e) DC-link current, (f, g and h) Current,
real and reactive power at DFIG rotor terminals, and (i) Modulation index.

As the DC link voltage is constant and all the currents in the DC link satisfy KCL, the power balance is also
achieved. The corresponding real/reactive power injected by the rotor side inverter is shown in Figure 22(g) and
(h) and its MI is shown in Figure 22(i).

Mode 4: Operation of DSTATCOM as 50 Hz inverter system, powering the isolated loads in the absence of
wind
The proposed hybrid power system shown in Figure 1 is fed by the wind, biogas, and solar renewable energy
sources. In this system even if the wind speed is less or power generation from wind energy is zero, then it is possi-
ble to supply the AC loads with solar PV panels, if we operate the three-phase SPWM inverter with 50 Hz fre-
quency. The simulation of the system has been carried out, and results are presented to demonstrate the operation
of the 50 Hz inverter system, with and without biogas generator (SCIG).

Case (a): PV-battery-50 Hz inverter system powering the isolated loads in the absence of wind (without biogas driven
SCIG).. This scheme powering the three-phase isolated loads is explained through simulation results. The isolated
load connected in the system is varied from 0.5 to 2 kW, UPF. The 50 Hz inverter must provide constant AC vol-
tage of 415 (VL-L) for the load. A simple PI controller has been used for the same and its parameters, Kp = 0.09,
and Ki = 12 have been obtained through manual tuning.
The operation of PV-battery-50 Hz inverter system is tested with an increment of load in steps of 0.5 kW start-
ing from 0.5 kW, with a time interval of 2 seconds. The load demand is limited to the solar PV array power rating
of 2 kW. The inverter maintains constant voltage of 415 (VL-L) at its terminals, VINV for different load values,
which is given in Figure 23. The load voltage and current are same as those of the inverter shown in Figure 23(a).
As the load increases, the current supplied by the inverter, IINV is also increased as in Figure 23(b). The power
absorbed by the load is same as that of the power supplied by the inverter, PINV as shown in Figure 23(c) since
the inverter powered by the PV which is the only source in the system.

Case (b): PV-battery-50 Hz inverter system with biogas driven SCIG, powering the isolated loads in the absence of wind. This
scheme, with biogas generator (SCIG) is explained through simulation results. The isolated load connected in the
system is 2 kW, UPF. The reactive power support for SCIG is provided through delta connected 2 kVAr capacitor
bank and inverter. The NSCIG varies from 1520 to 1560 rpm as mentioned in Figure 24(a) to share the load. The
50 Hz inverter maintains the constant voltage 415 (VL-L), which is shown in Figure 24(b) and voltage at load and
SCIG terminals are same as the inverter output voltage. As the NSCIG increases, load shared by SCIG increases
and reactive power absorption (QSCIG) also increases. The waveforms of load, inverter, and SCIG currents are
shown in Figure 24(c) to (e).
Sekhar and Kumaresan 15

Figure 23. Waveforms of 50 Hz inverter and load: (a) inverter terminal voltage, (b) inverter current, and (c) load and inverter
power.

Figure 24. Waveforms of load, 50 Hz inverter, and SCIG: (a) SCIG speed, (b) inverter terminal voltage, (c) load current, (d)
inverter current, and (e) SCIG current.

The load power and power sharing by inverter and SCIG are given in Figure 25(a) to (c). The reactive power
absorbed by the SCIG from 1520 to 1560 rpm and reactive power supplied by inverter and capacitor are given in
Figure 25(d) to (f). The reactive power absorbed by the SCIG is proportional to its rotor speed.
The performance of the standalone power system for various modes is given in the Table 3. Real and reactive
power balance in the system along with the highlight in each mode is also delineated in Table 3. Each mode in the
proposed system has its own exclusivity in supplying load with the available renewable energy sources and reactive
power support.

Conclusion
A hybrid power system consisting a wind generator, solar panels, battery bank, and a biogas driven generator has
been considered for installation in isolated locations. Four different modes of operation of the system based on
16 Wind Engineering 00(0)

Figure 25. Waveforms of load, 50 Hz inverter, capacitor, and SCIG: (a) load real power, (b) inverter real power, (c) SCIG real
power, (d) SCIG reactive power, (e) capacitor reactive power, and (f) inverter reactive power.

Table 3. Performance of proposed hybrid power system for various operating modes; VPCC = 415 V and fPCC = 50 Hz; Solar
irradiation level is varied from 300 to 1000 W/m2.

Operating mode NDFIG (rpm) NSCIG Load P balance in the Q balance in Main
(rpm) system the system feature

Mode 1 (a) 1300 1520 to 2 kW, UPF PLoad = QSCIG = QDFIG Supplying
1590 PDFIG + PSCIG loads
without
external Q
support
(b) 1150 to 1350 1575 3 kW, UPF
(c) 1300 1560 3 to 1.5 kW, UPF
Mode 2 (a) 1300 1520 to 3.75 kVA, PLoad = QLoad = Supplying
1590 0.8 p.f. (lag), R-L load PDFIG + PSCIG QDFIG and loads more
QSCIG = than 3 kW,
QCAP + QDFIG UPF and
lagging loads
(b) 1250 to 1350 1575 3 kVA, 0.9 p.f. (lag), R- QLoad = QDFIG and
L load QSCIG = QCAP
(c) 1300 1575 3 kVA, 0.9 p.f. (lag) to
4 kW, UPF
Mode 3 1300 1550 3 to 4 kW, UPF (from PLoad = PDFIG QSCIG = QCAP Supplying
0 to 7 seconds) + PSCIG loads more
and 4 to 3 kVA, than 3 kW,
0.8 p.f. (lag) UPF, and
(from 7 to 3 kVA, p.f.
13 seconds) (lag)
With
DSTATCOM
connection
(after t =
2.1 seconds)
PLoad = PDFIG + QLoad =
PSTATCOM + PSCIG QSTATCOM and
QSCIG = QCAP
Mode 4 No wind 1520 to 0.5 to 2 kW, UPF PLoad = PINV QSCIG = Suppling
1560 + PSCIG QCAP + loads
QINV without
wind energy
Sekhar and Kumaresan 17

energy generation by the different sources and reactive power support provided by capacitor bank and
DSTATCOM, have been studied. Under each mode again, different cases are considered, in respect of the nature
of speed of the generators and the magnitude and power factor of the load. In all these modes and their sub cases,
the solar irradiation is varied from one level to another linearly.
In the first mode with no reactive power support provided in the system, when the speed of the DFIG and that
of SCIG are both kept constant or one is constant and that of the other is varied, the real power of the UPF load,
is shared between the DFIG and SCIG based on their speed values. The reactive power requirement of the SCIG
is supplied by DFIG. The phenomenon is the same, even when there is a step change in the load. From these
results, it was inferred that an additional reactive power source would be required, if the load is of a lagging power
factor, so that the load connected to hybrid system can be increased. Accordingly, a capacitor bank was added
that will meet out the QSCIG demand alone. With these aspects, the second mode was taken up for investigation.
In this mode also similar patterns of the generator speeds were considered, and lagging p.f. loads were taken. It
was observed that if the speed of the DFIG is kept constant and that of SCIG is varied, DFIG and SCIG share
the load real power. The reactive power demand of load is met by the DFIG and that of the SCIG is supplied by
the capacitor bank. If the reactive power needed by the SCIG, which varies in proportion to its speed, exceeds that
of the capacitor, then the shortage is met by the DFIG. Alternatively, if the speed of the SCIG is constant, the real
power generated, and reactive power absorbed by it remain the same. With variation in the speed of DFIG, it sup-
plies a part of the total reactive power needed by the SCIG and the load and the balance is taken care of by the
capacitor. Similar observations were made even if a step change in load is brought in.
Subsequently a DSTATCOM was introduced at the load terminals of the hybrid system and operated to share
50% of the real power and complete reactive power demand of the load. The load connected in the system was
changed periodically both in kVA and power factor and the power sharing by the three active sources controlled
through the hysteresis controller employed in the system, has been demonstrated successfully. The simulation has
been carried out using abc-to-dq0 and dq0-ab transformations of the three-phase load current.
Lastly, the system is studied assuming that there is no wind power generation, the wind speed being below cut-
in value. AC load is supplied at the set voltage and 50 Hz frequency, employing solar PV panels, and operating
the three-phase SPWM inverter with 50 Hz frequency, using a simple PI controller. Both cases of without and with
the biogas engine driven SCIG have been considered. If the SCIG is not used, the load should be limited to the rat-
ing of the solar panels. If the SCIG included, then it also contributes real power, depending on its rotor speed, but
its reactive power requirement cannot be met only by the inverter and the capacitor bank should be added to the
system.
In all these cases, it is shown that, irrespective of the variation in irradiation, the inverters and controllers
designed and operated in the system, maintain constant voltage and frequency at the load terminals. Further, the
real and reactive power balance is accomplished through the closed loop control of a PV fed SPWM inverter at
DFIG rotor. All the relevant wave forms of voltage, current, and real and reactive power are presented and fully
discussed, and they validate the usefulness of the proposed hybrid system operated by suitably formulated control
strategies.
In the future work, proposed standalone power system topology can be tested using an experimental set up.
The wind energy will be extracted effectively by implementing MPPT controller for DFIG. The reactive power
control and optimization can be achieved in a smooth way by using effective controllers and soft computing tech-
niques. Utilizing the available recent research findings, detailed investigation of temperature rise of components of
the wind generator, conforming to stipulate standards and ways to improve the overall system efficiency of the
scheme, can be taken up. This will enable the installation to attain the full benefit of the integrated multiple renew-
able sources of energy.

Acknowledgements
The authors gratefully acknowledge Dr. M. Subbiah, Former Professor, Department of Electrical and Electronics Engineering,
National Institute of Technology, Tiruchirappalli, for proof reading and valuable suggestions in the preparation of this article.
The authors further acknowledge Suhanya MS, PhD scholar, Department of Electrical and Electronics Engineering, National
Institute of Technology, Tiruchirappalli, for supporting in carrying out MATLAB/Simulink work in the preparation of this
article.
18 Wind Engineering 00(0)

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this
article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article:
This work was supported in part by the National Institute of Wind Energy (Formerly Centre for Wind Energy Technology),
Chennai, India, an autonomous Research and Development Institution under the Ministry of New and Renewable Energy,
Government of India.

ORCID iD
N Kumaresan https://orcid.org/0000-0003-3449-2835

References
Adefarati T, Bansal RC and John Justo J (2017) Techno-economic analysis of a PV-wind-battery-diesel standalone power sys-
tem in a remote area. The Journal of Engineering 13: 740–744.
Al-Ammar EA, Habib HUR, Kotb KKM, et al. (2020) Residential community load management based on optimal design of
standalone HRES with model predictive control. IEEE Access 8: 12542–12572.
Arthishri K, Kumaresan N and Ammasai Gounden N (2019) Analysis and application of three-phase SEIG with power conver-
ters for supplying single-phase grid from wind energy. IEEE Systems Journal 13(2):1813–1822.
Bagarty DP and Kastha D (2013) Reactive power allocation for loss minimisation in a stand-alone variable speed constant fre-
quency double output induction generator. IET Power Electronics 6(5): 851–861.
cat.com (2016) Hybrid microgrids: The time is now. Available at: https://www.cat.com/en_US/by-industry/electric-power/Arti-
cles/White-papers/white-paper-hybrid-microgrids-the-time-is-now.html (accessed 10 May 2020).
Das BK, Al-abdeli YM and Woolridge M (2019) Effects of battery technology and load scalability on stand-alone PV/ICE
hybrid micro-grid system performance. Energy 168: 57–69.
Diab AZ, Sultan HM, Mohamed IS, et al. (2019) Application of different optimization algorithms for optimal sizing of PV/
wind/diesel/battery storage stand-alone hybrid microgrid. IEEE Access 7: 119223–119245.
Fu Y, Zhang H, Mi Y, et al. (2018) Control strategy of DFIG in hybrid micro-grid using sliding mode frequency controller and
observer. IET Generation, Transmission & Distribution 12(11): 2662–2669.
Guchhait PK, Banerjee A and Mukherjee V (2020) Comparative study using soft computing techniques for the reactive power
compensation of a hybrid power system model. Ain Shams Engineering Journal 11(1):87–98.
Gupta Y, Doolla S, Chatterjee K, et al. (2021) Optimal DG allocation and Volt–Var dispatch for a droop-based microgrid.
IEEE Transactions on Smart Grid 12(1): 169–181.
Hore D and Sarma R (2018) Neural network–based improved active and reactive power control of wind-driven double fed
induction generator under varying operating conditions. Wind Engineering 42(5): 381–396.
Kant K, Jain C and Singh B (2017) A hybrid diesel-wind PV-based energy generation system with brushless generators. IEEE
Transactions on Industrial Informatics 13(4): 1714–1722.
Kusakana K (2015) Optimal scheduled power flow for distributed photovoltaic/wind/diesel generators with battery storage sys-
tem. IET Renewable Power Generation 9(8): 916–924.
Marhaba MS, Farhangi S, Iman-Eini H, et al. (2018) Reactive power sharing improvement of droop controlled DFIG wind tur-
bines in a microgrid. IET Generation, Transmission & Distribution 12(4): 842–849.
Merabet A, Tawfique Ahmed K, Ibrahim H, et al. (2017) Energy management and control system for laboratory scale micro-
grid based wind-PV-battery. IEEE Transactions on Sustainable Energy 8(1): 145–154.
Navin Sam K, Kumaresan N and Ammasai Gounden N (2017a) Wind-driven stand-alone DFIG with battery and pumped
hydro storage system. Sadhana - Academy Proceedings in Engineering Science 42(2): 173–185.
Navin Sam K, Kumaresan N, Ammasai Gounden N, et al. (2015) Analysis and control of wind-driven stand-alone doubly-fed
induction generator with reactive power support from stator and rotor side. Wind Engineering 39(1): 97–112.
Navin Sam K, Kumaresan N, Ammasai Gounden N, et al. (2017b) Optimal reactive power controller for wind driven stand-
alone doubly-fed induction generators. Wind Engineering 41(2): 124–143.
Prakash SL, Arutchelvi M and Jesudaiyan AS (2016) Autonomous PV-array excited wind-driven induction generator for off-
grid application in India. IEEE Journal of Emerging and Selected Topics in Power Electronics 4(4): 1259–1269.
Puchalapalli S and Singh B (2020) A single input variable FLC for DFIG-based WPGS in standalone mode. IEEE Transactions
on Sustainable Energy 11(2): 595–607.
Ramabhotla S and Bayne SB (2019) Cost and availability optimization of wind energy with distributed energy resources of a
microgrid. Wind Engineering 43(6): 559–572.
Sekhar and Kumaresan 19

regenpower.com (2020). Regen microgrid: Solar-wind-diesel hybrid systems. Available at: https://regenpower.com/off-grid/
hybrid-systems/ (accessed 18 May 2020).
Rezkallah M, Singh S, Chandra A, et al. (2019) Comprehensive controller implementation for wind-PV-diesel based standalone
microgrid. IEEE Transactions on Industry Applications 55(5): 5416–5428.
Sekhar N, Lakshmana Rao P, Kumaresan N, et al. (2019) Experimental investigation on a new hybrid system employing wind-
driven DFIG and solar PV panels. Journal of the Institute of Engineers (India) Series B 100: 561–574.
Singh G and Sundaram K (2020) Manufacturing deformation impact on the performance of electrical generator for the wind
turbine application. Wind Engineering. Epub ahead of print 12 November 2020. DOI: 10.1177/0309524x20968816.
Singh G, Lentijo S and Sundaram K (2019) The impact of the converter on the reliability of a wind turbine generator. In: Pro-
ceedings of the ASME 2019 power conference, Salt Lake City, Utah, USA, 15–18 July 2019, p.V001T06A016. New York:
ASME.
Singh G, Sundaram K and Matuonto M (2020) A solution to reduce overheating and increase wind turbine systems availability.
Wind Engineering. Epub ahead of print 18 March 2020. DOI: 10.1177/0309524x20910992.
Singh R and Bansal RC (2018) Review of HRESs based on storage options, system architecture and optimisation criteria and
methodologies. IET Renewable Power Generation 12(7): 747–760.
Tiwari SK, Singh B and Goel PK (2018) Design and control of autonomous wind-solar system with DFIG feeding 3-phase 4-
wire loads. IEEE Transactions on Industry Applications 54(2): 1119–1127.
Venkatraman K, Selvan MP and Moorthi S (2016) Predictive current control of distribution static compensator for load com-
pensation in distribution system. IET Generation, Transmission & Distribution 10(10): 2410–2423.
Vijayakumar K, Kumaresan N and Ammasai Gounden N (2012) Operation and closed loop control of wind driven stand-alone
DFIG using single inverter-battery system. IET Power Applications 6(3): 162–171.
Vijayakumar K, Kumaresan N and Ammasai Gounden N (2013a) Operation of inverter assisted wind-driven slip-ring induction
generator for stand-alone power supplies. IET Electric Power Applications 7(4): 256–269.
Vijayakumar K, Kumaresan N, Ammasai Gounden N, et al. (2013b) Real and reactive power control of hybrid excited wind-
driven grid-connected DFIGs. IET Power Electronics 6(6): 1197–1208.
Wang R, Sun Q, Gui Y, et al. (2019) Exponential-function-based droop control for islanded microgrids. Journal of Modern
Power Systems Clean Energy 7: 899–912.
Wang Z, Yi H, Zhuo F, et al. (2021) Analysis of parameter influence on transient active power circulation among different gen-
eration units in microgrid. IEEE Transactions on Industrial Electronics 68(1): 248–257.
Yang D, Jiang C, Cai G, et al. (2019) Optimal sizing of wind/solar/battery/diesel hybrid microgrid based on typical scenarios
considering meteorological variability. IET Renewable Power Generation 13(9): 1446–1455.

Appendix
Symbols
fr frequency at the DFIG rotor terminals,
Hz
fs frequency at the DFIG stator terminals,
Hz
Iabc L three phase load current fed as inputs to
the controller, A
IDL F filtered component of IDL, A
K proportional controller gain varies from 0
to 1
N DFIG rotor speed, rpm
P number of poles of DFIG
Verr deviation in Vrms from Vref, V
Vrms actual voltage at DFIG stator terminals,
V
Vref reference/desired voltage at DFIG stator
terminals, V
VPCC voltage at PCC, V
fPCC frequency at PCC, Hz