International Journal of Power Electronics and Drive Systems (IJPEDS)

Vol. 14, No. 3, September 2023, pp. 1781~1790

ISSN: 2088-8694, DOI: 10.11591/ijpeds.v14.i3.pp1781-1790  1781

Model predictive current control for maximum power point

tracking of voltage source inverter based grid connected
photovoltaic system

Satyanarayana Burada1, Kottala Padma2

1
Department of Electrical and Electronics Engineering, Anil Neerukonda Institute of Technology and Sciences (ANITS),
Visakhapatnam, India
2
Department of Electrical Engineering, Andhra University, Visakhapatnam, India

Article Info ABSTRACT

Article history: Due to the high demand of grid connected photovoltaic systems, there is a
need to track the maximum power point of the PV system. As the output of
Received Jan 30, 2023 PV system is dc, there should be a converter, acting as medium between PV
Revised Mar 28, 2023 system and dc bus capacitor to track maximum power at all the loads.
Accepted Apr 6, 2023 Usually boost converter is acting as medium between PV system and dc link
capacitor as the duty cycle of the insulated-gate bipolar transistor in boost
converter is in between 0 to 1 for maximum loads during maximum power
Keywords: point tracking (MPPT). To make PV system stable, the balance point is dc
bus. If the dc bus voltage is constant, the system will be stable. Then the
3 phase inverters transfer power will just depend on current. For this purpose, the active
Controller current reference signal is to be generated by setting up the reference voltage
Converter across dc bus. Here to generate active reference current, PI controller is used
Model predictive current and the reference voltage is taken according to the peak voltage of the
control inverter output voltage. The proposed control strategy was evaluated on a
MPPT three-phase inverter linked to the grid and supplied by the PV system, which
Photovoltaic system is working under varying irradiation conditions.
Reference active current This is an open access article under the CC BY-SA license.

Corresponding Author:
Satyanarayana Burada
Department of Electrical and Electronics Engineering
Anil Neerukonda Institute of Technology and Sciences (ANITS)
Visakhapatnam, Andhra Pradesh, 531162, India
Email: [email protected]

1. INTRODUCTION
The decreasing of conventional power sources and reduction of PV cell cost have increased the
using of photovoltaic systems. In photovoltaic (PV) systems maximum power point tracking (MPPT)
methods becomes the part of control the power electronic converters [1]. Secondary control
algorithms [1]–[5] are used to get better sensitivity and dynamic capability of control algorithms. Usually,
renewable energy sources are used in microgrid where the group of loads are interconnected and clearly
defined their electrical boundaries and acts as single controllable entity [6], [7] with respect to grid and can
be connected in grid mode and islanded mode. for islanded networks there are some sensitivity-based
methods to minimize the current during switching action [8], [9]. Usually sliding mode control (SMC) [10]
and model predictive control (MPC) [2] algorithm is used to improve dynamic response. Even these
techniques demand powerful computing, it has eliminated by technical improvements on micro controllers.
Different algorithms are proposed to track maximum power [11]–[13] from the PV systems at
different loads. Adaptive neuro fuzzy inference and P&O MPPT algorithms are combined for better tracking

Journal homepage: http://ijpeds.iaescore.com

1782  ISSN: 2088-8694

of the maximum power [14]–[17]. It will give better results than the classical MPPT methods. Mechanical
trackers are also used in tracking the maximum power from the PV system [18]–[20], but this is not a
practical approach for high power rating PV systems. Advanced control algorithms are used to achieve a
better dynamic reaction and sensitive control due to the challenges on the mechanical observer. The main
advantage of sliding mode control is its fast dynamic response, robustness and insensitive to system
parameters. But this work makes the argument that the model predictive control technique is used to produce
the quick dynamic response. Model predictive control method is another method that has similar advantages
as SMC. The MPPT algorithm is combined with MPC algorithm [21], [22] to obtain better maximum power
tracking performance. Especially this combination gives good performance under variable irradiation
conditions. This algorithm manages a Z-source inverter connected to the grid.
In this paper one stage power converter has been used. As two-stage power converters [21], [22] and
[23], usage is increasing total cost of the system. But while using single stage power converter [24] series
connected PV panel group is needed to get more dc bus voltage. Even there are different solutions to increase dc
bus voltage [25]–[28] but one stage power layer with series PV panels is the most preferred
model [24]–[29]. In this grid connected single stage power converter, the important parameter is the voltage
across dc bus capacitor, which is to be maintained constant irrespective of the load for smooth and proper
operation of the Z-source inverter. In this paper boost converter is used as the medium between PV system and
dc bus capacitor as the firing angle of the power electronic switch is in between 0 to 1 for maximum loads
during MPPT, which is not the case in buck type converter or buck boost converter. In this paper the
comparison between buck, boost and buck-boost converter has been given for different load resistance values.
Here boost converter is used as the medium between PV system and dc bus capacitor and to boost up
the input voltage of the inverter. To maintain the constant dc bus voltage the active current reference signal is to
be generated by setting up the reference voltage across dc bus capacitor and as the reactive current is to be
supplied by the grid, the reference reactive current is to be maintained at zero. Here to generate active reference
current PI controller is used. And the reference voltage is taken according to the peak voltage of the inverter
output voltage. The proposed control strategy was evaluated on a three-phase inverter linked to the grid and
supplied by the PV system, which is working under varying irradiation and cloudy conditions.
Here one should think about some operational behavior of PI controller. That is according to
irradiation curve, the PV irradiation will be zero for some time and after that it starts increasing. But during the
zero-irradiation time, the generated power will be zero, so the converter cannot transfer any power to the dc bus,
but the PI controller will work as usual and it will try to make dc bus voltage equal to reference voltage. It
cannot be possible because generated power is zero and the PI controller will get saturated, when irradiation
increases, the voltage starts increasing, but as already PI controller gets saturated power cannot transfer to the
grid through inverter. So, the dc bus voltage reaches to very high value, so there is a need to reset PI controller
by designing a reset condition. Here in this work, if the dc bus voltage is greater than 1.05 times that of
reference voltage the PI controller will get reset. If there is no saturation point for the direct reference current,
the reference current may increase to very higher value which may activate the over current relay and causes to
nuisance tripping. So, a saturation point should be provided at the output of the PI controller.

2. SYSTEM MODEL AND OPERATION OF THE PROPOSED SYSTEM

2.1. Operation of the proposed system
Generally, in photo voltaic modules there is one single operating point at any given point in time
where maximum power can be drawn, called as maximum power point. In this work MPPT algorithm is used
to track the maximum power point of the photo voltaic system by controlling the duty cycle of the boost
converter for different load conditions. The generated power from PV system is transferred to dc bus by
boost converter and MPPT algorithm. The inverter transfers the power from dc bus to grid. The prediction
algorithm is used to generate reference active component of current for inverter. The phase locked loop
(PLL) is used to transform the three phase currents into dq0 transformation as the reference current signal is
dc. By reducing the control parameter error between the standard and the subsequent sampling interval
(Iabc(m+1)), the model predictive control algorithm operates. Model predictive control algorithm can be
summarized in two steps as prediction and minimizing. To forecast the control value, discretization
techniques are employed. The second stage of the model predictive control algorithm is error minimization.
In the cost function and minimization block the PI controller is used to generate switching pulses to inverter
through pulse width modulation technique, as shown in the simulation model.

2.2. Modelling of the proposed system

General circuit schema and controlling techniques are shown in Figure 1. PV system is directly
connected to the grid through boost converter and 3 phase inverters. The values of filter inductance and
capacitance can be calculated by (1) and (2), respectively [30].

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1781-1790
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1783
2
0.1𝑢𝑟𝑚𝑠
𝐿𝑓𝑖𝑙𝑡𝑒𝑟 = (1)
2𝜋𝑓𝑃𝑝ℎ𝑎𝑠𝑒

0.05𝑃𝑝ℎ𝑎𝑠𝑒
𝐶𝑓𝑖𝑙𝑡𝑒𝑟 = 2 (2)
2𝜋𝑓𝑢𝑟𝑚𝑠

Where 𝑢𝑟𝑚𝑠 is the line-to-line rms voltage from the grid side and 𝑃𝑝ℎ𝑎𝑠𝑒 is the single-phase power.
Here PV power is taken as a total system power for theoretical calculations. PV system consist 17
parallel strings and 14 series connected modules per each string and total input power is 50KW. The
inductance and capacitance of the boost converter is calculated by (3) and (5), respectively [2].

(1−𝐷)2 .𝐷.𝑉02
𝐿𝑏𝑜𝑜𝑠𝑡 = (3)
2𝑓𝑠𝑤 𝑃

Where D is the duty cycle, is the voltage across dc link capacitor, and 𝑓𝑠𝑤 is the switching frequency, P is the
total power. The duty cycle can be calculated by using the (4).
𝑉𝑀𝑃𝑃
𝐷 =1− (4)
𝑉0

Where 𝑉𝑀𝑃𝑃 is the voltage of the PV system at maximum power point.

𝐷.𝑃
𝐶𝑏𝑜𝑜𝑠𝑡 = (5)
𝑉02 .%𝑟𝑖𝑝𝑝𝑙𝑒.𝑓𝑠𝑤

Here while designing capacitor value, percentage of ripple content to be taken as low as possible,
but keep in the mind the cost of the capacitor banks. The maximum power point tracking (MPPT) is used to
track maximum energy from PV panels at different irradiance levels. In any MPPT algorithm the duty cycle
of the converter is changed to make the load resistance is equal to internal resistance of PV module to
transfer maximum power from source to load. The comparison between buck, boost and buck-boost
converter is shown in Table 1, in which it was observed that for 90% of the loads the duty cycle value of the
boost converter is in between 0 and 1 and the continuous current can be drawn from the PV system due to the
input inductance of the boost converter.
The reference current for the inverter is determined using the conventional P&O MPPT algorithm
described in [2]. As a result, the model predictive current control approach is the main emphasis of this study,
which is utilized to control inverter current (iabc). The general voltage equation for the inverter [31] is given
in (6). Here 𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟 is the inverter voltage, 𝑉𝐿𝑓 is the voltage drop due to filter inductance, 𝑉𝑅𝑓 is the voltage
drop due to filter resistance and 𝑉𝐺𝑅𝐼𝐷 is the measured grid voltage.

𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟 = 𝑉𝐿𝑓 + 𝑉𝑅𝑓 + 𝑉𝐺𝑅𝐼𝐷 (6)

Output current equation can be written by replacing of inductor voltages [31] in (7). Here 𝑖(𝑚) is the
current sample interval’s inverter current, 𝐿𝑓 is the filter inductance, 𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟(𝑙) is the inverter voltage vector,
𝑉𝐺𝑅𝐼𝐷(𝑚) is the current sample interval’s measured grid voltage and 𝑅𝐿𝑓 is the filter resistance.
𝑑𝑖(𝑚) 1
= (𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟(𝑙) − 𝑉𝐺𝑅𝐼𝐷(𝑚) − 𝑅𝐿𝑓 𝑖(𝑚) ) (7)
𝑑𝑡 𝐿𝑓

Where 𝑉𝐺𝑅𝐼𝐷(𝑚) is the current sample interval's measured grid voltage. All measured parameters are
converted using the Clarke transformation into α-β coordinate. 𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟 stands for inverter voltage vector,
which is used to make predictions. The voltage space vectors are given in Figure 2. There are two “0” states
and six active states in three-phase inverters, as shown in the diagram. It demonstrates that there are eight
states in which the inverter can function. An array of phase states represents all voltage states (S A, SB, SC). In
(8) – α-β frame [28] can be used to express inverter voltage states. As a result, all of the control parameters in
(7) are in the α-β frame.
2
𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟(𝑙) = [𝑆𝐴(𝑙) + (−0.5 + 0.866𝑗)𝑆𝐵(𝑙) + (−0.5 − 0.866𝑗)𝑆𝐶(𝑙) ] (8)
3

In the (8), l is state number that varies from {1,2……………8}.

Model predictive current control for maximum power point tracking … (Satyanarayana Burada)
1784  ISSN: 2088-8694

Ipv(m)
PV

Dc
C Boost link 3 phase
Converter filter GRID
capacitor inverter

Iabc(m)

Vgrid(m)

δ
Vdc bus S1,S2
Vpv(m) Cost function &
Prediction algorithm Id(ref) S3,S4
MPPT Vref minimization
Ipv(m) S5,S6

PLL ꞷt Id
Vgrid(m) abc/dq0

Iabc(m+1)

Figure 1. General schematic diagram of the system

Table 1. Duty cycle of buck, boost and buck-boost converter for different loads
R0 (Ω) δ (boost) δ (buck) δ (buck boost) R0 (Ω) δ (boost) δ (buck) δ (buck boost)
2 -0.95157 0.51241 0.3388 52 0.61727 2.6128 0.7232
4 -0.37997 0.72465 0.42017 54 0.62442 2.6626 0.72697
6 -0.12674 0.88752 0.4702 56 0.63119 2.7114 0.73056
8 0.024216 1.0248 0.50613 58 0.6376 2.7594 0.734
10 0.12723 1.1458 0.53397 60 0.64369 2.8066 0.7373
12 0.20328 1.2551 0.55657 62 0.64949 2.853 0.74046
14 0.26238 1.3557 0.5755 64 0.65501 2.8986 0.7435
16 0.31002 1.4493 0.59172 66 0.66028 2.9436 0.74642
18 0.34948 1.5372 0.60587 68 0.66531 2.9878 0.74924
20 0.38286 1.6204 0.61838 70 0.67012 3.0314 0.75195
22 0.41158 1.6995 0.62956 72 0.67474 3.0744 0.75457
24 0.43663 1.775 0.63964 74 0.67916 3.1169 0.7571
26 0.45873 1.8475 0.64882 76 0.68341 3.1587 0.75954
28 0.47842 1.9173 0.65721 78 0.6875 3.2000 0.7619
30 0.49511 1.9845 0.66494 80 0.69143 3.2408 0.76419
32 0.51211 2.0496 0.67209 82 0.69522 3.281 0.76641
34 0.52668 2.1127 0.67874 84 0.69887 3.3208 0.76856
36 0.54001 2.174 0.68494 86 0.70239 3.3601 0.77065
38 0.55228 2.2335 0.69074 88 0.70579 3.3989 0.77267
40 0.56362 2.2916 0.69619 90 0.70908 3.4373 0.77464
42 0.57413 2.3481 0.70133 92 0.71226 3.4753 0.77655
44 0.58392 2.4034 0.70618 94 0.71533 3.5129 0.77841
46 0.59307 2.4574 0.71077 96 0.71832 3.5501 0.78022
48 0.60164 2.5103 0.71512 100 0.72401 3.6233 0.7837
50 0.60969 2.562 0.71926

Figure 2. Space vectors of three-phase two level inverter

3. METHOD FOR MODELING PREDICTIVE CURRENT CONTROL

Between the reference and the next sample period, the MPC algorithm minimizes the control
parameter error. MPC consists of two phases. prediction and minimization. Discrete method discretization is
employed to anticipate the control parameter. Usually, exact discretization method has high accuracy for low

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1781-1790
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1785

order and high order systems but it will take a greater number of iterations and more computational time. The
forward Euler approximation method also has high accuracy for low order systems with a smaller number of
iterations and less computational time. The local truncation error or local discretization error in the Euler
method is the error made in approximating the derivative by the difference quotient, whereas the global
discretization error at a position is the magnitude of the actual error at the point. so, (9) is a first order
differential equation. So, the forward Euler approximation [31] method is preferred.

𝑑𝑖 𝑖(𝑚+1) −𝑖(𝑚)
≈ (9)
𝑑𝑡 𝑇𝑠

In (10) can be produced by replacing the derivative in (7) with the Euler technique.

𝑖(𝑚+1) −𝑖(𝑚) 1
= (𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟 − 𝑉𝐺𝑅𝐼𝐷(𝑚) − 𝑅𝐿𝑓 𝑖(𝑚) ) (10)
𝑇𝑠 𝐿𝑓

You may get the prediction in (11) by arranging the (10).

𝑇
𝑖(𝑚+1) = [ 𝑠 (𝑉𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟(𝑙) − 𝑉𝐺𝑅𝐼𝐷(𝑚) − 𝑅𝐿𝑓 𝑖(𝑚) )] + 𝑖(𝑚) (11)
𝐿𝑓

In (11) predicts current values for all possible voltage states (1....n). Error minimization is the
second stage of the MPC method, as was already mentioned. Most control methods use the difference
between the reference and measured values. Unlike previous approaches, Utilizing the discrepancy between
the reference and predicted values, the MPC algorithm. This feature allows the algorithm to create an action
for the subsequent phase. As a result, MPC has a greater dynamic range. In MPC algorithms, in order to
calculate the error term, cost functions are used. The control algorithm's cost function is represented by (12).
Real and imaginary current values are denoted by 𝑖𝛼 and 𝑖𝛽 respectively.

𝑔 = |𝑖𝛼∗ − 𝑖𝛼(𝑚+1) | + |𝑖𝛽∗ − 𝑖𝛽(𝑚+1) | (12)

For each sample interval, the cost function is evaluated along with the prediction equation for all
potential voltage vectors. As a result, error values are generated for every possible switching position. The
best cost function with the least amount of error is selected during the minimization stage. In order to create
switching signals, the optimum cost function is used.

4. SIMULATION RESULTS
Simulation studies are used to verify the system model and proposed control structures in Figure 1.
Table 2 lists the simulation parameters. Firstly, the proposed control algorithm's power flow control
performance has been tested under various irradiation conditions. The simulation model is shown in
Figures 3(a)-(c).

Table 2. System specifications

Parameter Value
𝑢𝑟𝑚𝑠 (Grid voltage) 380 V
𝑓 (Grid frequency) 50 Hz
𝑃 (Nominal power) 50.72 kW
𝐿𝐹𝑖𝑙𝑡𝑒𝑟 (Filter inductance per phase) 0.0028 H
𝐿𝑏𝑜𝑜𝑠𝑡 (Inductance of boost converter) 1.6236 mH
𝐶 (DC link capacitor) 1000 µF
PV Panel (17 parallel and 14 series) 50.72 kWMPP,406 VMPP
𝑇𝑠 (Sampling time) 10 µs

Following the dynamic performance analysis, the suggested control algorithm's power flow control
performance was tested under various irradiation situations. Figure 4 shows the irradiation curve during the
course of 30 seconds [11]. This paper presents a method that overcomes the problem of the confusion during
fast irradiation change in the classical MPPT as well as in model predictive control (MPC) based MPPTs
available in the literature.

Model predictive current control for maximum power point tracking … (Satyanarayana Burada)
1786  ISSN: 2088-8694

(a)

(b)

(c)

Figure 3. Simulation models for grid connected PV system: (a) DC side of the system, (b) AC side of the
system, and (c) reference active current generation and control unit

Figure 4. Irradiation curve

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1781-1790
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1787

Figure 5 shows the PV power results. The MPPT algorithm tracks the MPP with 98 percent efficiency,
as shown in Figures 5(a) and 5(b). Even though efficiency decreased in cloudy conditions, after these times,
efficiency increased to 98 percent. Figures 5(c) and 5(d) show the voltage and current results for the PV panel
group, respectively. While the current rises to 125 A, The PV voltage is fairly stable. It results from the nature
of MPPT control in PVs and offers information on control effectiveness. The reference voltage for the PI
controller to generate direct component of reference current, should be selected based on the voltage across dc
bus capacitor, when it is disconnected from the PV system and charged by the inverter. As the inverter is mostly
buck type inverter, The dc bus voltage needs to be higher than the inverter output voltage's maximum value.
Here one should concentrate on some operational behavior of PI controller. That is according to
irradiation curve, the PV irradiation will be zero for some time and after that it starts increasing. But during
the zero-irradiation time, the generated power will be zero, so the converter cannot transfer any power to the
dc bus, but the PI controller will work as usual and it will try to make dc bus voltage equal to reference
voltage. It cannot be possible because generated power is zero and the PI controller will get saturated, when
irradiation increases, the voltage starts increasing, but as already PI controller gets saturated power cannot
transfer to the grid through inverter. So, the dc bus voltage reaches to very high value, so there is a need to
reset PI controller by designing a reset condition.

(a) (b)

(c) (d)

Figure 5. PV results: (a) power, (b) MPPT efficiency, (c) PV voltage, and (d) PV current

Here in this work, if the dc bus voltage is greater than 1.05 times that of reference voltage the PI
controller will get reset. If there is no saturation point for the direct reference current, the reference current
may increase to very higher value which may activate the over current relay and causes to nuisance tripping.
So, a saturation point should be provided at the output of the PI controller to limit the current with in
specified range. Figure 6 shows the reference voltage that must be selected for proper operation of PI controller to
generate active component of reference current. The active reference current will be generated based on the
difference between dc bus voltage and reference voltage. The active component of the reference current is shown
in Figure 7. And the voltage across the capacitor, which is also called as dc bus voltage is maintained constant,
irrespective of the load and for the given system it is maintained at 800 V as shown in Figure 8.

Figure 6. Reference voltage for dc bus voltage Figure 7. Reference current for the PI controller

Aside from the MPPT algorithm's performance, one must examine the power, effectiveness, and
harmonic distortions of the inverter side. The MPC method correctly fixes the inverter currents to the

Model predictive current control for maximum power point tracking … (Satyanarayana Burada)
1788  ISSN: 2088-8694

reference value (Iref*) as illustrated in Figure 9(a). The inverter's output currents are also altered in
accordance with the irradiation curve in Figure 4. Because The reference current value is adjusted by the
MPPT algorithm and MPC. Harmonics are less than 5% for active power values greater than 20 kW, as
shown in Figure 9(b). The THD value is less than the limit set forth in the standards [32].
Here the PI controller will trigger the inverter to make the peak current of the inverter output equal
to the active component of the reference current as shown in Figure 10. Detailed results of inverter currents
shown in Figure 11(a). The MPC algorithm successfully regulates the inverter currents to the reference.
According to recent tracking data, the combined control algorithms reliably follow the MPP under hazy and
variable irradiance conditions. Detailed results of the inverter voltages are shown in Figure 11(b). The
inverter voltages are successfully synchronized with the grid voltage of line-to-line RMS value of 380 V.
Irrespective of the load the MPC algorithm always keep the inverter voltage equals to the grid voltage.

Figure 8. DC bus voltage

(a) (b)

Figure 9. Inverter results: (a) currents and (b) THD

Figure 10. Line currents of the inverter and reference active current

(a) (b)

Figure 11. Detailed inverter results: (a) line currents and (b) line voltages

To evaluate the total efficiency of the proposed system, the transferred power to grid at the output of the
inverter was compared to ideal MPP curves. Figure 12 depicts the power losses between the inverter output power
and the desired maximum power curve. Around the nominal power, power losses are quite high. Filter inductance

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1781-1790
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1789

is to blame. The losses in a practical inductor come mainly from the parasitic equivalent resistance (which is
practically the ohmic resistance of the windings which depends on the geometry of the wire used in winding the
inductor as well as the frequency) and the core losses. To reduce the major losses in the filter inductor the toroidal
shape inductor is used. Due to its symmetry, the toroidal shape has the benefit of having a low leakage flux, or the
quantity of magnetic flux that escapes the core is low, therefore the core losses will be less and by keeping the
ohmic resistance of the inductor low the copper losses can be reduced. Even if losses increased, figure
demonstrates that for power values more than 5 kW, the minimum system efficiency is around 90%.

Figure 12. Inverter output power and ideal MPP curve

5. CONCLUSION
A combined control algorithm for PV systems is suggested in the paper. The MPPT and MPC
algorithms are used in the suggested control strategy. The integrated algorithm provided efficient tracking under
variable irradiance and foggy situations because of the quick and sensitive control capabilities of MPC. In
addition to the combined algorithm performance testing, the step-change tests demonstrate that the MPC
algorithm has a superior capability for current control. It has been done to analyze the efficiency of both control
and power structures. The MPPT method's control efficiency is 98 percent, while the inverter efficiency is 92
percent. Both the inverter and the control algorithms overall system efficiency are around 90%.

REFERENCES
[1] M. Metry, M. B. Shadmand, R. S. Balog, and H. Abu-Rub, “MPPT of photovoltaic systems using sensorless current-based model
predictive control,” IEEE Transactions on Industry Applications, vol. 53, no. 2, pp. 1157–1167, 2017, doi: 10.1109/TIA.2016.2623283.
[2] E. Irmak and N. Güler, “A model predictive control-based hybrid MPPT method for boost converters,” International Journal of
Electronics, vol. 107, no. 1, pp. 1–16, 2020, doi: 10.1080/00207217.2019.1582715.
[3] E. Kabalci, G. Gokkus, and A. Gorgun, “Design and implementation of a PI-MPPT based buck-boost converter,” Proceedings of
the 2015 7th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2015, pp. SG23–SG28, 2015,
doi: 10.1109/ECAI.2015.7301194.
[4] I. Chtouki, P. Wira, M. Zazi, B. Collicchio, and S. Meddour, “Design, implementation and comparison of several neural Perturb
and Observe MPPT methods for photovoltaic systems,” International Journal of Renewable Energy Research, vol. 9, no. 2, pp.
757–770, 2019, doi: 10.20508/ijrer.v9i2.9293.g7645.
[5] N. Genc, H. Uzmus, and I. Iskender, “Dynamic behavior of DC-DC boost converter controlled with cascade PI-ASC,” Proceedings of the
8th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2016, 2017, doi: 10.1109/ECAI.2016.7861148.
[6] E. Loukakis and E. Karapidakis, “Feasibility study of microgrid village with renewable energy sources,” 2017 52nd International
Universities Power Engineering Conference, UPEC 2017, vol. 2017-January, pp. 1–6, 2017, doi: 10.1109/UPEC.2017.8231889.
[7] A. Paspatis, K. Fiorentzis, Y. Katsigiannis, and E. Karapidakis, “Smart campus microgrids towards a sustainable energy transition—the
case study of the Hellenic Mediterranean University in Crete,” Mathematics, vol. 10, no. 7, 2022, doi: 10.3390/math10071065.
[8] E. E. Pompodakis, G. C. Kryonidis, M. C. Alexiadis, and E. S. Karapidakis, “A Three-phase sensitivity-based approach for
smooth line-switching in islanded networks,” International Journal of Electrical Power and Energy Systems, vol. 144, 2023, doi:
10.1016/j.ijepes.2022.108515.
[9] E. E. Pompodakis, G. C. Kryonidis, and E. S. Karapidakis, “Volt/Var control and energy management in Non-Interconnected
insular networks with multiple hybrid power plants,” Applied Energy, vol. 331, 2023, doi: 10.1016/j.apenergy.2022.120427.
[10] K. Amara et al., “Improved performance of a PV solar panel with adaptive neuro fuzzy inference system ANFIS based MPPT,”
7th International IEEE Conference on Renewable Energy Research and Applications, ICRERA 2018, pp. 1098–1101, 2018, doi:
10.1109/ICRERA.2018.8566818.
[11] A. A. Abdulrazzaq and A. H. Ali, “Efficiency performances of two MPPT algorithms for PV system with different solar panels
irradiances,” International Journal of Power Electronics and Drive Systems, vol. 9, no. 4, pp. 1755–1764, 2018, doi:
10.11591/ijpeds.v9.i4.pp1755-1764.
[12] N. Hashim, Z. Salam, D. Johari, and N. F. N. Ismail, “DC-DC boost converter design for fast and accurate MPPT algorithms in
stand-alone photovoltaic system,” International Journal of Power Electronics and Drive Systems, vol. 9, no. 3, pp. 1038–1050,
2018, doi: 10.11591/ijpeds.v9.i3.pp1038-1050.
[13] T. Laagoubi, M. Bouzi, and M. Benchagra, “MPPT and power factor control for grid connected PV systems with fuzzy logic
controllers,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 9, no. 1, p. 105, Mar. 2018, doi:
10.11591/ijpeds.v9.i1.pp105-113.
[14] S. Biricik, H. Komurcugil, N. D. Tuyen, and M. Basu, “Protection of sensitive loads using sliding mode controlled three-phase DVR with
adaptive notch filter,” IEEE Transactions on Industrial Electronics, vol. 66, no. 7, pp. 5465–5475, 2019, doi: 10.1109/TIE.2018.2868303.
[15] D. Haji and N. Genc, “Fuzzy and P&O Based MPPT controllers under different conditions,” in 2018 7th International Conference on
Renewable Energy Research and Applications (ICRERA), Oct. 2018, pp. 649–655. doi: 10.1109/ICRERA.2018.8566943.

Model predictive current control for maximum power point tracking … (Satyanarayana Burada)
1790  ISSN: 2088-8694

[16] N. Priyadarshi, A. K. Sharma, and F. Azam, “A hybrid firefly-asymmetrical fuzzy logic controller based MPPT for PV -wind-fuel
grid integration,” International Journal of Renewable Energy Research, vol. 7, no. 4, pp. 1546–1560, 2017, doi:
10.20508/ijrer.v7i4.6131.g7195.
[17] M. Heidari, “Improving efficiency of photovoltaic system by using neural network MPPT and predictive control of converter,”
International Journal of Renewable Energy Research, vol. 6, no. 4, pp. 1524–1529, 2016, doi: 10.20508/ijrer.v6i4.4794.g6942.
[18] A. Fudholi, L. C. Haw, K. Sopian, and A. M. Omer Abdulmula, “Primary study of tracking photovoltaic system for mobile station
in Malaysia,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 9, no. 1, p. 427, Mar. 2018, doi:
10.11591/ijpeds.v9.i1.pp427-432.
[19] A. A. Sneineh and W. A. Salah, “Design and implementation of an automatically aligned solar tracking system,” International
Journal of Power Electronics and Drive Systems, vol. 10, no. 4, pp. 2055–2064, 2019, doi: 10.11591/ijpeds.v10.i4.pp2055-2064.
[20] S. H. Chong, N. N. Chandren, and C. R. Allan Soon, “Output energy maximization of a single axis photovoltaic solar tracking
system: Experimental verification,” International Journal of Power Electronics and Drive Systems, vol. 10, no. 3, pp. 1655–1661,
2019, doi: 10.11591/ijpeds.v10.i3.pp1655-1661.
[21] N. Altin, S. Ozdemir, H. Komurcugil, I. Sefa, and S. Biricik, “Two-stage grid-connected inverter for PV systems,” Proceedings -
2018 IEEE 12th International Conference on Compatibility, Power Electronics and Power Engineering, CPE-POWERENG 2018,
pp. 1–6, 2018, doi: 10.1109/CPE.2018.8372540.
[22] M. B. Shadmand, X. Li, R. S. Balog, and H. A. Rub, “Model predictive control of grid-tied photovoltaic systems: Maximum
power point tracking and decoupled power control,” 2015 1st Workshop on Smart Grid and Renewable Energy, SGRE 2015,
2015, doi: 10.1109/SGRE.2015.7208726.
[23] I. Colak, E. Kabalci, and G. Bal, “Parallel DC-AC conversion system based on separate solar farms with MPPT control,” 8th
International Conference on Power Electronics - ECCE Asia: “Green World with Power Electronics”, ICPE 2011-ECCE Asia,
pp. 1469–1475, 2011, doi: 10.1109/ICPE.2011.5944473.
[24] S. Ozdemir, N. Altin, and I. Sefa, “Single stage three level grid interactive MPPT inverter for PV systems,” Energy Conversion
and Management, vol. 80, pp. 561–572, 2014, doi: 10.1016/j.enconman.2014.01.048.
[25] S. Sajadian, R. Ahmadi, and H. Zargarzadeh, “Extremum seeking-based model predictive MPPT for grid-tied z-source inverter for
photovoltaic systems,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 7, no. 1, pp. 216–227, 2019,
doi: 10.1109/JESTPE.2018.2867585.
[26] M. Kamalirad, H. Iman-Eini, B. Farhangi, and S. Bacha, “A reliable three-phase transformerless grid-connected PV inverter with
inductive DC link,” IEEE Journal of Photovoltaics, vol. 8, no. 5, pp. 1305–1312, 2018, doi: 10.1109/JPHOTOV.2018.2846702.
[27] N. Genc and Y. Koc, “Experimental verification of an improved soft-switching cascade boost converter,” Electric Power Systems
Research, vol. 149, pp. 1–9, 2017, doi: 10.1016/j.epsr.2017.04.015.
[28] E. Irmak and N. Guler, “Model predictive control of grid-tied three level neutral point clamped inverter integrated with a double
layer multi-input single output DC/DC converter,” Proceedings - 2018 IEEE 12th International Conference on Compatibility,
Power Electronics and Power Engineering, CPE-POWERENG 2018, pp. 1–6, 2018, doi: 10.1109/CPE.2018.8372538.
[29] A. Sangwongwanich and F. Blaabjerg, “Mitigation of interharmonics in PV systems with maximum power point tracking
modification,” IEEE Transactions on Power Electronics, vol. 34, no. 9, pp. 8279–8282, 2019, doi: 10.1109/TPEL.2019.2902880.
[30] M. W. Hussain and M. A. Qureshi, “Analysis and design of passive filters for power quality improvement in 3φ grid-tied PV
systems,” 2021 4th International Conference on Energy Conservation and Efficiency, ICECE 2021 - Proceedings, 2021, doi:
10.1109/ICECE51984.2021.9406278.
[31] A. Lashab, D. Sera, and J. M. Guerrero, “A dual-discrete model predictive control-based MPPT for PV systems,” IEEE
Transactions on Power Electronics, vol. 34, no. 10, pp. 9686–9697, 2019, doi: 10.1109/TPEL.2019.2892809.
[32] IEEE Power and Energy Society (Transmission and Distribution Committee), “IEEE Recommended Practice and Requirements
for Harmonic Control in Electric Power Systems,” IEEE Std 519-2014, vol. 2014, 2014.

BIOGRAPHIES OF AUTHORS

Satyanarayana Burada received B. Tech degree in Electrical and Electronics

Engineering from JNTU Kakinada, India in 2012, M. Tech degree in Power Systems from
National Institute of Technology (NIT), Surat, India in 2016 and currently pursuing PhD in grid
connected photo-voltaic systems area from Andhra University, Visakhapatnam, India. He
presented papers in 2 IEEE conferences held at Delhi Technological university in 2016 and
Amity University in 2017. From 2016 until now he has been a lecturer in the department of
Electrical and Electronics Engineering, Anil Neerukonda Institute of Technology and Sciences,
Visakhapatnam, India. He can be contacted at email: [email protected].

Kottala Padma has done her B. Tech from Sri Venkateswara University in 2005,
MTech in Power Systems from Andhra University in 2010 and PhD from Andhra University
in 2015. Presently working as Associate Professor of Electrical Engineering Department at
Andhra University, Visakhapatnam, India. She has published 10 papers in SCI, Scopus, Web
of Science indexed journals and 2 papers in springer book chapters. Her research interests
include Power system aspects, Renewable energy systems, Microgrids, Internet of things, soft
computing techniques, Transmission and Distribution systems. She has guided 2 Ph.D.
Students and currently guiding 8 Research Scholars and has successfully supervised 30 PG
thesis. She has been presented with Global Teacher Award and best research award. She can
be contacted at email: [email protected].

Int J Pow Elec & Dri Syst, Vol. 14, No. 3, September 2023: 1781-1790