An Improved Control of Grid Integrated Solar PV

System Using Fractional Order PI Controller

Combined with Active Current Detection Scheme
Manoja Kumar Behera, Lalit Chandra Saikia, Satish Kumar Ramoji, Biswanath Dekaraja, Sanjeev Kumar Bhagat
Department of Electrical Engineering, National Institute of Technology Silchar, Silchar, Assam-788010, India
[email protected], [email protected], [email protected], [email protected], [email protected]
2022 4th International Conference on Energy, Power and Environment (ICEPE) | 978-1-6654-7124-4/22/$31.00 ©2022 IEEE | DOI: 10.1109/ICEPE55035.2022.9797990

Abstract—An active current detection technique (ACDT) effect, which reduces overall control stability. System
for a three-phase grid integrated photovoltaic system is characteristics and operational circumstances have an impact
presented in this research. As a result of using the triangle on the controllers ability to perform their functions [6]. It can
character function for component extraction, the suggested be less efficient to use traditional controllers at the base of a
control reduces the overall computing load. This demonstrates control structure. The conventional controller, with its
the efficacy of the proposed ACDT under a range of operating sluggish dynamics, is incapable of handling these systemic
situations. Control algorithm design also provides desired changes. It is seen in literatures [6], [7], that the control
voltage regulation, decreasing peak overshoot during transient application of fractional order (FO) shows good potential in
conditions. This will also aid in the rapid and precise
control with greater stability, which are facilitated by their
regulation of grid current concerning the reference current.
Further, the fractional-order controller is adopted for DC-link
increased degrees of freedom. FO control has been
voltage stabilization during transient dynamics of the system. outperformed by traditional controllers in a limited number
The suggested control algorithm successfully maintains a of studies, including area frequency control [8], distributed
constant power flow to meet the load requirement. Finally, energy resource management [9], and PV maximum power
simulation experiments were conducted in the Matlab point tracking (MPPT) [10]. This motivates us to employ a
Simulink in order to validate the proposed controller ability. fractional order PI controller to stabilize the VSI DC-link
The simulation result shows proposed control performs well voltage.
subject to system perturbation. Control methods for detecting phase sequence for grid
voltage were developed by the researchers. Some of them
Keywords—active current detection technique, FOPI used phase-locked loop (PLL) synchronization [11]. PLL
controller, grid-integrated PV system, maximum power point algorithms based on synchronous reference frames (SRF)
tracking, renewable energy source. [12] show distortion when subjected to unbalanced or
nonlinear loads or grid faults conditions. PLL
I. INTRODUCTION synchronization techniques are not so effective while sensed
The bloom expansion in infrastructure development and input voltage consists of both higher and lower-order
the construction of new industries around the globe increases harmonics. Researchers have also reported on-grid
the need for AC electrical power; yet, maximum loads have synchronization using the second-order generalized
nonlinear switching characteristics, resulting in a variety of integrator [13] and phase angle estimation technique [14].
power quality (PQ) issues in the power system. The whole Time-domain techniques have also been proved to be
power system was affected by the decline in PQ at the effective for real-time control in a number of investigations.
distribution level. This will cause power outages, increased The zero-crossing detection (ZCD) method [11] has a flaw
load on the grid, and more demand for reactive power in the because it only gathers phase information once every half
long run. Consequently, renewable energy sources (RESs) cycle of the supply, resulting in inferior dynamic
are rapidly being employed to meet the worlds energy needs. performance. However, the phase angle is determined by the
Since there is a lack of fossil fuels in the environment, RES filtering signal using α-β based control [15]. The time delay
are preferred for producing clean and green electricity. for acquiring the signal is then determined by these filters.
Among all RESs, solar photovoltaic (PV) has drawn The control approach based on state variable filter [16] is
attention due to its potential, simple and easy installation, also prone to unbalanced loads and frequency variations.
pollution-free, less maintenance, etc. Additionally, the control approach on weighted least square
The RES fed voltage source inverter (VSI) integration to estimation [14] is ineffective in detecting frequency
the utility grid must meet the following requirements: popper deviation and has a high computational cost.
grid synchronization, autonomous management of grid- Frequency domain control imposes a significant
injected active and reactive power, good transient response, computational overhead and introduces a one-cycle time
peak power extraction, grid currents with less overall delay into the signal. Researchers discussed many strategies
harmonic distortion [2], [3]. DC-DC boost converters were for frequency-domain control in the literature [17-20]. The
used to monitor peak power and enhance the PV voltage in a sliding discrete Fourier transform-based control [17-19] is
low-power PV array to maintain DC-link voltage [4]. potentially unstable, but the Goertzel control method [20]
Simultaneously, an appropriate control method for VSI is performs well with predetermined harmonic contents,
imposed in a grid-connected PV system to identify reference making it inefficient for random noise. Several controls
signals. Additionally, the control method is composed of two approach on active and reactive current extraction have
cascading control loops. The DC-link voltage is stabilized by proven that three-phase VSI performs better in a variety of
the first loop, while the three-phase grid currents are scenarios; however, their performance has deteriorated while
controlled by the second loop. the power source contains oscillations and noise.
In the literature, proportional-integral (PI) controllers are Some control approach on active and reactive current
often employed to maintain VSIs DC-link voltage [5]. In components extraction reported in literature are orthogonal
certain cases, PI controller integrators produce a saturation current decomposition control (OCDC) [24], Kalman filter

978-1-6654-7124-4/22/$31.00 ©2022 IEEE

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on January 19,2023 at 07:35:00 UTC from IEEE Xplore. Restrictions apply.
and notch filter [21], leaky least mean fourth [23], PLL less  Compared to the usual CTF-based control, the
fast character triangular function (CTF) [25], artificial neural suggested method reduces the number of LPF from
network (ANN) [22], and HRCCD (harmonics and reactive six to three. Additionally, the proposed control
current component detection) [26] based CTF. Control based system identifies the fundamental active current
on Kalman and notch filters are challenging to build and component that is accompanied by harmonics.
have amplitude inaccuracy. However, when used in real-  The DC-link voltage is regulated using a fractional-
time, the ANN-based control method produces good results order proportional-integral (FOPI) controller for
but is imprecise when detecting a signal disturbance. An smooth power injection to the grid in the grid
orthogonal function with six ZCD and sample control and integrated PV based VSI.
hold (SCH) blocks is used to identify the reference current in  Several case scenarios are used to assess the
the OCDC method. This increases computational weight. In controllability of the proposed ACDT in the Matlab
contrast, the PLL less fast CTF and HRCCD feature six LPF Simulink environment.
(low pass filter) that introduce oscillations to the system. The reaming of this article is structured as: Section II
The following are the primary aims of the suggested conferred the double-stage grid integrated solar PV system
work based on the above fiction and facts. configuration and proposed controller model. Section III
 To track the appropriate reference grid current, a presents the Matlab simulation results of the proposed
novel current controller for grid-integrated PV controlled system. Finally, a summarized conclusion is
systems using an active current detection technique provided in Section IV.
(ACDT) is proposed.
Ua Unit Vector Template
VSI Ub Utility Grid
ipv idc Uc Estimation
Lf Rf Ttransformer PCC
L a
D
VGabc Vab
vpv Cin Sw Cdc Vdc b
Vbc c
260 V/25 kV
25 kV
IGabc
MPPT unit

400 V/25 kV
PV Array
INC based
MPPT
LPF Ua
CF
S4

S6
S1

ILar
S3

S5
S2

Vdc
Vdc ref RL Load IM
2 ILa
FOPI

LPF Ub

Ir ILr ILbr CF
1/3
IGar IGbr IGcr 2 ILb
Ua Ub Uc
LPF Uc

ILcr CF

2 ILc

Fig. 1. Two-stage grid-connected solar PV system schematic diagram.

II. PROPOSED CONTROL MODEL AND SYSTEM DESCRIPTION

A. Proposed Control Algorithm
Fig. 1 depicts the schematic diagram of a two-stage grid-
As seen in Fig. 1, the suggested control approach
integrated PV system. An interface inductor is used to link
comprises two-stage control: control for three-phase VSI in
the PV-based VSI to the three-phase AC utility grid (25 kV,
order to generate the PWM gate pulses and another for DC-
50Hz). In order to provide enough isolation between the VSI
DC boost converter. Incremental conductance (INC) MPPT
and the utility grid, a transformer is used. The resistive and
[27] is employed with a boost converter to get peak power
inductive (RL) loads, as well as the induction motor (IM),
from solar PV systems in various climatic circumstances,
are linked at the grid integrated PV system point of common
while the suggested ACDT control uses three-legged VSI to
connection (PCC). The solar PV array and the VSI DC-link
address significant PQ difficulties. The triangle function is
are linked via a DC-DC boost converter. The VSI DC link
used in this paper to construct a rigorous control technique
capacitor is employed to reduce the DC link voltage ripple.
for detecting the active current component and estimating
PV power is properly synchronized with a three-phase
the phase reference currents. The six PWM gate pulses for
utility grid by the exact and appropriate proposed control switches for VSI depend on the reference grid current
algorithm, which converts DC to AC concurrently. correlation with real grid currents. The proposed reference
Additionally, Fig 1 illustrates the suggested control current detection technique measuredly consists of unit
mechanism for the pulse width modulation (PWM) switched vector templates estimation, ACDT, computation of current
VSI. gain, and generation of three-phase grid current reference.

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on January 19,2023 at 07:35:00 UTC from IEEE Xplore. Restrictions apply.
1) Unit Vector Templates Estimation procedure. Reference current (Ir) is also produced, as shown
The three-phase PCC line voltage (Vab, Vbc) is sensed to in Fig 1, after the addition of active current and DC current
compute the grid voltage of each phase (VGa, VGb, VGc) gains as (12). To keep the grid current free of harmonics, it
expressed as (1). In addition, the three-phase grid voltage is critical that the reference current be generated precisely
magnitude (VG) is determined as (2). and accurately.
2Vab  Vbc (Vab  Vbc ) (Vab  2Vbc ) I r  I Lr  I dc (12)
VGa  ;VGb  ;VGc  (1)
3 3 3 4) Three Phase Reference Grid Currents Generation
2 2 As a final step, the three-phase reference grid currents
VG  (VGa  VGb2  VGc2 ) (2) (I , I , I ) for PWM switching pulse generation are
Gar Gbr Gcr
3 expressed as (13). For the purpose of calculating three-phase
The unit vector templates weights (Ua, Ub, Uc) is estimated reference grid current, the active weight component of unit
as the ratio of the respective phase grid voltage to the vector templates for each phases is multiplied by the
magnitude of the three-phase grid voltage, expressed as (3). reference current. The hysteresis current controller and
In order to operate the suggested control, only active current reference grid current are also employed to generate a high-
components are needed. As a result, the reactive component frequency PWM signal for VSI, as shown in Fig 1. Here, a
of unit vector templates has not been approximated. As a hysteresis band of ±0.02 is used as the reference.
consequence, the control algorithms workload is minimized, I Gar  I r U a ; I Gbr  I r U b ; I Gcr  I r U c (13)
resulting in a low rate of delay for the basic load current
estimate. III. SIMULATION RESULTS
V V V
U a  Ga ;U b  Gb ;U c  Gc (3) The Matlab Simulink toolbox is used to conduct
VG VG VG simulation studies for the suggested control approach. Load
2) Active Current Detection disturbance, IM loading and solar irradiation, temperature
For identify the active load current component, LPF, change have been simulated. Additionally, Table I contains
correction factor (CF), weight of the unit vector template, the circuit parameters that were used to simulate the test
load current (ILa, ILb, ILc), and simple adder, subtractor, a scenario.
multiplier is used. In the red, yellow, and green color boxes TABLE I. SYSTEM PARAMETERS
of Fig 1, the basic active load current detection mechanism
is depicted for each phase of the three-phase system. For Sl No. Parameters Values
each phase, the active current component detection in three 1. Utility grid voltage (L-L) 25 kV
phases (ILar, ILbr, ILcr) is depicted in (4–9). The LPFs cut-off 2. Grid frequency 50 Hz
frequency is set at 10-15 Hz. Always the CF value is RGa = 0.8929 Ω
positive and less than 1, and it is used in this system to make 3. Grid Impedance
LGa = 16.58 mH
sure there are not any harmonics. The CF value used in this Lf = 0.25 mH
study is 0.21[26], [28]. The initial iterations of the active 4. Interfacing inductor
Rf = 0.0019 Ω
current component are represented by ILar1, ILbr2, and ILcr3. 5. VSI switching frequency 5 kHz
I Lar  (2  LPF  U a )[(U a  I La )  (CF U a  I Lar1 )] (4) 6. DC-link voltage (Vdc) 500 V
I Lar1  ( I La  I Lar ) (5) 7. DC-link filter capacitor (Cdc) 6000 µF
I Lbr  (2  LPF  U b )[(U b  I Lb )  (CF  U b  I Lbr1 )] (6) 8. Boost converter switching frequency 10 kHz
9. Boost inductor 5 mH
I Lbr1  ( I Lb  I Lbr ) (7)
10. PV array side capacitor (Cin) 100 µF
I Lcr  (2  LPF  U c )[(U c  I Lc )  (CF  U c  I Lcr1 )] (8) 11. PV array voltage (vpv) 321 V
I Lcr1  ( I Lc  I Lcr ) (9) 12.
Nominal PV array power at standard
100 kW
test conditions (STC)
3) Computation of Active Current Gain Load1: 40 kW, 10 kVAr
To determine the active current gain (ILa), the average of 13. Load
Load2: 20kW
the fundamental active load current component is used, as 5.4 HP, 400 V, 50 Hz,
expressed as (10). 14. Induction motor (IM)
1430 rpm
I I I
I Lr  Lar Lbr Lcr (10) A. Effect of Load Perturbation at PCC
3
Additionally, the DC current gain (Idc) is determined in To understand the influence of the load shift on the
order to maintain a fixed reference value for the DC-link system, load perturbation is done in the PCC of the grid-tied
voltage is expressed as (11) [28]. solar PV system in this case. During the control process, the
proposed algorithm is tested to whether it is resistant to such
 K  changes. In this approach, initially, a total load of (P = 60
I dc  (Vdc ref -Vdc )  K p  i  (11)
 s  kW and Q = 10 kVAr) is connected at the PCC of the
where Idc(n + 1) denotes the DC current gain; Vdc ref and Vdc system. At t = 0.5 s, a total load of 20 kW load is switched
are denoted the reference and actual DC-link voltage, off from the system and again loaded at 0.8 s. The
respectively; K p , K i and  are the proportional, integral, and simulation responses of voltage, current, power, and duty
cycle during the load change are depicted in Fig 2. The load
fractional integral order of the FOPI controller. For the variation load current drawn from the system is shown in
optimal performance of the FOPI controller, it’s three Fig 2(b). Fig 2(a) shows that grid voltage (VGabc) remains
parameters Kp, Ki, and  are needed to be tuned. As a result, constant, and while the load is decreased at PCC, the
the FOPI controller is selected by a trial and error additional current is injected into the grid. Irrespective of

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on January 19,2023 at 07:35:00 UTC from IEEE Xplore. Restrictions apply.
load variation, the DC-link voltage (Vdc) is maintained at the
reference DC voltage (Vdc ref) shown in Fig 2(c). A smooth
transition is seen in VSI injected power (Pinj) to the grid in
Fig 2(d). An accurate MPP tracking of PV voltage, power,
and respective duty cycle of the boost converter during the
process is observed in Fig 2(e) and (f). The achievement of
instant stability demonstrates the suggested control
structures dynamic performance.

(f)
Fig. 2. Simulation responses during load perturbation. (a) Grid voltage
(VGabc) at PCC, VSI injected current (IGabc) to the grid. (b) Load current
(ILabc). (c) DC-link voltage (Vdc) response. (d) VSI injected power (Pinj) to
the grid. (e) PV MPP voltage and power generated. (f) Duty cycle of the
boost converter.

B. Effect of IM Loading at PCC

In this case, the proposed control behavior of the grid-
(a) tied solar PV system during IM loading at PCC is studied.
The system is loaded with a total load of (P = 60 kW and Q
= 10 kVAr) at the PCC. At time t = 0 s, an IM (400 V, 50
Hz, 5.4 HP, 1430 rpm) is switched on to the PCC of the
system with a mechanical load of 30 Nm is applied to the
IM. At t = 0.5 s, the mechanical load is applied to IM is
increased to 45 Nm. IM generates sufficient electromagnetic
torque in response to the mechanical load shown in Fig 3(b).
With the increase in additional mechanical load, the speed
of IM slightly slows down at t = 0.5 s. As shown in Fig 3(a),
a stable grid voltage (VGabc) and smoot VSI injected current
(b) to the grid are seen during the IM loading. DC-link voltage
is kept intact with the reference voltage, showing a good
synchronization of the VSI to the grid. After supplying the
loads connected at PCC, the rest power is injected into the
grid with less oscillation in the system during the IM
loading. Though irradiance and temperature were not
changed and kept at STC, so only the duty cycle of the boost
converter during the process is provided in Fig 3(c). Hence,
seeing smooth conduct of the system behavior subject to IM
loading validates the proposed controller control ability
during IM loading on the system.
(c)

(d) (a)

(e) (b)

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on January 19,2023 at 07:35:00 UTC from IEEE Xplore. Restrictions apply.
 (c) (a)

(d) (b)

(e) (c)
Fig. 3. Simulation responses during IM loading. (a) Grid voltage (VGabc) at
PCC, VSI injected current (IGabc) to the grid. (b) Torque (Te) and Speed (N)
characteristics of IM load operation. (c) DC-link voltage (Vdc) response. (d)
VSI injected power (Pinj) to the grid. (e) Duty cycle of the boost converter.

C. Effect of Change in Irradiance and Temperature Pattern

The proposed controller during a continuous change in
solar irradiance and temperature is shown in Fig 4. For a
more robust analysis of the proposed technique, a
continuous decrease, increase of irradiance pattern, and at t
= 2 s continuous increase pattern of temperature is taken
shown in Fig 4(a). Here throughout the operation, a total (d)
load of (P = 60 kW and Q = 10 kVAr) is connected at the
PCC of the system. Fig 4(b) shows that grid voltage (VGabc)
remains stable. While irradiance is started decreased at t =
0.6 s, VSI injected power (Pinj) decreased gradually seen in
Fig 4(d) at t = 0.9 s; it goes negative means the grid now
supplies the load. A reverse power flow occurs with a 180-
degree phase shift in current injected, shown in Fig 4(b).
And again, irradiance increased gradually, and power
generation followed the irradiance and temperature pattern.
DC-link voltage (Vdc) is kept constant with insignificant
(e)
dynamics seen in Fig 4(c) during the whole operation. The
boost converter entire operating region duty cycle is shown
in Fig 4(e). It is observed that even in the severe change in
environmental conditions, proposed control is well
performed with less dynamic and undershoot overshoot and
able to maintain smooth operation of the system.

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on January 19,2023 at 07:35:00 UTC from IEEE Xplore. Restrictions apply.
 (f) shading conditions,” Sustainable Energy Technologies and
Assessments, vol. 42, p. 100859, Dec. 2020.
Fig. 4. Simulation responses during solar irradiance and temperature
change profile. (a) Irradiance and temperature profile. (b) Grid voltage [11] F.M. Gardner, “Phaselock techniques,” John Wiley & Sons. 2005.
(VGabc) at PCC, VSI injected current (IGabc) to the grid. (c) DC-link voltage [12] M. Karimi-Ghartema, “Enhanced phase-locked loop structures for
(Vdc) response. (d) VSI injected power (Pinj) to the grid. (e) PV MPP power and energy applications,” John Wiley & Sons. 2014.
voltage and power generated. (f) Duty cycle of the boost converter. [13] S. Golestan, M. Monfared, and J. M. Guerrero, “Second order
generalized integrator based reference current generation method for
IV. CONCLUSION single-phase shunt active power filters under adverse grid
conditions,” 4th Annual International Power Electronics, Drive
An improved control technique based on CTF-based Systems and Technologies Conference, Feb. 2013.
active current detection is proposed for a two-stage grid-tied [14] H.S. Song, and K. Nam, “Instantaneous phase-angle estimation
solar photovoltaic system. This technique is combined with algorithm under unbalanced voltage-sag conditions,” IEE
an FO-controlled DC-link stabilizer. The suggested Proceedings-Generation, Transmission and Distribution, vol. 147,
technique was verified using several operational situations no. 6, pp. 409–415, Nov. 2000.
in the Matlab Simulink environment. The suggested control [15] J. Svensson, “Synchronisation methods for grid-connected voltage
method simultaneously maintains the proper voltage and source converters,” IEE Proceedings-Generation, Transmission and
Distribution, vol. 148, no. 3, pp. 229–235, May. 2001.
frequency at the PCC to ensure appropriate grid
synchronization and improve active power injection under [16] G. Saccomando, and J. Svensson, “Transient operation of grid-
connected voltage source converter under unbalanced voltage
unfavorable operating circumstances. Furthermore, a conditions,” In Conference Record of the 2001 IEEE Industry
quicker DC-link control with extremely low peak overshoot Applications Conference. 36th IAS Annual Meeting, vol. 4, pp. 2419-
and undershoot during rapid load switching has been found 2424, Sep. 2001.
in transient events. Again, the proposed approach is simple [17] B. Chitti Babu, K. Sridharan, E. Rosolowski, and Z. Leonowicz,
and can be implemented in a low-cost microcontroller. As a “Analysis of SDFT based phase detection system for grid
consequence, the designed controller has a higher detection synchronization of distributed generation systems,” Engineering
Science and Technology, an International Journal, vol. 17, no. 4, pp.
rate and higher accuracy in both steady-state and transient 270–278, Dec. 2014.
conditions. [18] E. Jacobsen, and R. Lyons, “The sliding DFT,” IEEE Signal
Processing Magazine, vol. 20, no. 2, pp. 74-80. 2003.
REFERENCES
[19] Jacobsen, E. and Lyons, R., 2004. An update to the sliding
[1] A. K. Singh, I. Hussain, and B. Singh, “Double-stage three-phase DFT. IEEE Signal Processing Magazine, 21(1), pp.110-111.
grid-integrated solar PV system with fast zero attracting normalized [20] J.-H. Kim, J.-G. Kim, Y.-H. Ji, Y.-C. Jung, and C.-Y. Won, “An
least mean fourth based adaptive control,” IEEE Transactions on islanding detection method for a grid-connected system based on the
Industrial Electronics, vol. 65, no. 5, pp. 3921–3931, May 2018. Goertzel algorithm,” IEEE Transactions on Power Electronics, vol.
[2] V. N. Lal and S. N. Singh, “Control and performance analysis of a 26, no. 4, pp. 1049–1055, Apr. 2011.
single-stage utility-scale grid-connected PV system,” IEEE Systems [21] A. Kumar, N. Gupta, V. Gupta, and B. C. Babu, “A novel orthogonal
Journal, vol. 11, no. 3, pp. 1601–1611, Sep. 2017. current decomposition control for grid-connected voltage source
[3] W. Longhui, Z. Fang, Z. Pengbo, L. Hongyu, and W. Zhaoan, “Study converter,” IEEE Transactions on Industry Applications, vol. 55, no.
on the influence of supply-voltage fluctuation on shunt active power 6, pp. 7628–7641, Nov. 2019.
filter,” IEEE Transactions on Power Delivery, vol. 22, no. 3, pp. [22] R. Panigrahi and B. Subudhi, “Performance enhancement of shunt
1743–1749, Jul. 2007. active power filter using a Kalman filter-based ${{{\RM H}}_\infty
[4] A. Guichi, S. Mekhilef, E. M. Berkouk, and A. Talha, “Optimal }$ control strategy,” IEEE Transactions on Power Electronics, vol.
control of grid-connected microgrid PV-based source under partially 32, no. 4, pp. 2622–2630, Apr. 2017.
shaded conditions,” Energy, vol. 230, p. 120649, Sep. 2021. [23] N. Kumar, B. Singh, and B. K. Panigrahi, “LLMLF-based control
[5] M. Mishra and V. N. Lal, “An advanced proportional multiresonant approach and LPO MPPT technique for improving performance of a
controller for enhanced harmonic compensation with power ripple multifunctional three-phase two-stage grid integrated PV system,”
mitigation of grid-integrated PV systems under distorted grid voltage IEEE Transactions on Sustainable Energy, vol. 11, no. 1, pp. 371–
conditions,” IEEE Transactions on Industry Applications, vol. 57, no. 380, Jan. 2020.
5, pp. 5318–5331, Sep. 2021. [24] Y. Singh, I. Hussain, B. Singh, and S. Mishra, “Single-phase single-
[6] M. K. Behera and L. C. Saikia, “An improved voltage and frequency stage grid tied solar PV system with active power filtering using
control for Islanded Microgrid using BPF Based Droop control and power balance theory,” Journal of The Institution of Engineers
optimal third harmonic injection PWM scheme,” IEEE Transactions (India): Series B, vol. 99, no. 3, pp. 301–311, Mar. 2018.
on Industry Applications, vol. 58, no. 2, pp. 2483–2496, Mar. 2022. [25] M. B. Ammar, M. Chaabene, and Z. Chtourou, “Artificial neural
[7] M. K. Behera and L. C. Saikia, “Combined Voltage and Frequency network based control for PV/T panel to track Optimum thermal and
Control for Diverse Standalone Microgrid Networks Using Flexible electrical power,” Energy Conversion and Management, vol. 65, pp.
IDC with Novel FOC: A Real-Time Validation,” IETE Journal of 372–380, Jan. 2013.
Research, pp. 1–26, Sep. 2021. [26] A. K. Verma and B. Singh, “Harmonics and reactive current detection
[8] S. Debbarma, L. C. Saikia, and N. Sinha, “AGC of a multi-area of a grid-interfaced PV generation in a distribution system,” IEEE
thermal system under deregulated environment using a non-integer Transactions on Industry Applications, vol. 54, no. 5, pp. 4786–4794,
controller,” Electric Power Systems Research., vol. 95, pp.175-183, Sep. 2018.
Feb. 2013. [27] H.-J. Chiu, Y.-K. Lo, T.-P. Lee, Q. S. Chen, W. L. Yu, J.-X. Lee, F.
[9] M. B. Delghavi, S. Shoja-Majidabad, and A. Yazdani, “Fractional- Shih, and S.-C. Mou, “A battery charger with maximum power point
order sliding-mode control of islanded distributed energy resource tracking function for low-power photovoltaic system applications,”
systems,” IEEE Transactions on Sustainable Energy., vol. 7, no. 4, International Journal of Circuit Theory and Applications, vol. 39, no.
pp. 1482-1491, May. 2016. 3, pp. 241–256, Mar. 2011.
[10] M. K. Behera and L. C. Saikia, “A new combined extreme learning [28] S. K. Jalan, B. C. Babu, K. Sridharan, and N. Gupta, “An improved
machine variable steepest gradient ascent MPPT for PV system based control strategy of grid‐tied solar photovoltaic system using active
on optimized PI-FOI cascade controller under uniform and partial current detection method,” International Journal of Circuit Theory
and Applications, vol. 49, no. 3, pp. 602–615, Feb. 2021.

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on January 19,2023 at 07:35:00 UTC from IEEE Xplore. Restrictions apply.