International Journal of Applied Power Engineering (IJAPE)

Vol. 13, No. 3, September 2024, pp. 603~615

ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i3.pp603-615  603

Lion swarm optimization for grid connected PV system with

improved SEPIC

P. Annapandi1, D. Lakshmi2, B. Kavya Santhoshi3, P. Annapoorani4

1
Department of Electrical and Electronics Engineering, Francis Xavier Engineering College, Tirunelveli, India
2
Department of Electrical and Electronics Engineering, AMET Deemed to be University, Chennai, India
3
Department of Electrical and Electronics Engineering, Godavari Institute of Engineering and Technology (Autonomous),
Rajamahendravaram, India
4
Department of Electrical and Electronics Engineering, Vel Tech Multi Tech Dr. Rangarajan Dr. Sakunthala Engineering College,
Avadi, India

Article Info ABSTRACT

Article history: The wide deployment of grid-connected renewable energy system has
piqued immense attention recently, in response to rising electricity
Received Jul 11, 2023 consumption, diminishing fossil fuel reserves in addition to the need for
Revised Feb 22, 2024 reducing carbon emissions. Among the available sources of renewable
Accepted Mar 13, 2024 energy, photovoltaic (PV) power generation is the most promising
technology with enormous potential and easy access. This paper presents an
optimum control technique for grid connected PV systems. The improved
Keywords: single ended primary inductor converter (SEPIC) controls and regulates PV
output power to the optimum voltage level. The working of the improved
1𝜙 VSI SEPIC is controlled by a proportional-integral (PI) controller optimized by
Improved SEPIC meta-heuristic technique of lion swarm optimization (LSO). The constant
LSO output from the converter is then supplied to the power grid through a
PI controller single-phase voltage source inverter (1𝜙 VSI). The effectiveness of the
PV system proposed control strategy is ascertained using hardware validation with
DSPIC3050FPGA controller and MATLAB simulation generating a reduced
total harmonic distortion (THD) of 3.9% and 2.9%, respectively.
Furthermore, the proposed system generates an enhanced voltage gain of
1:10 and an efficiency of 96%.
This is an open access article under the CC BY-SA license.

Corresponding Author:
P. Annapandi
Department of Electrical and Electronics Engineering, Francis Xavier Engineering College
Tirunelveli, Tamil Nadu, India
Email: annapandian2001@gmail.com

1. INTRODUCTION
In recent days, the energy crisis is considered as a rising problem due to its influence on economic
growth of all sectors [1]. Moreover, the fossil fuel burning results in greenhouse gas emission and global
warming leading to environmental pollution. All these factors paved for the need of renewable energy
resources (RESs) [2]. The photovoltaic (PV) systems are regarded as significant clean and popular renewable
energy resource and is applied for both grid integration and power generation. The factors related to PV
system including grid integration, efficiency, grid stability, and quality of uninterrupted power are a serious
concern. Hence, the improvement of PV system efficiency and reduction of installation costs are focused by
Rodriguez et al. [3].
Generally, dozens of PV panels are used for improving the voltage at the terminals [4]. The number
of PV panels adopted for a particular application is reduced with the use of converter [5]. It is an important
component used for a removal of the power from the photovoltaic array and it maintains low current

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

604  ISSN: 2252-8792

ripples [6], [7]. Added to improved efficiency and reduced cost, the DC-DC converters provide fast current
and voltage control along with wide range of input/output voltage change ratios [8]. The conventional
converters are boost and buck-boost converters that perform enlarging of input voltage range obtained from
the PV array [9]. However, achieving conversion gain beyond six practically is not feasible in these
converters. The increased duty cycle operation compromises an efficiency of boost converter but generate
current ripples and electromagnetic interference [10]. Conventional Cuk converters have reduced switching
losses, better efficiency, and superior voltage moderation, yet they showed restrictions in delivering sharp
speed up/down voltage [11]. The single ended primary inductor converter (SEPIC) is used in various
applications related to power electronics due to its unique properties and many works concentrate on
improving its step-up voltage gain. The possibility of soft-switching performance without the addition of
auxiliary elements is a unique feature of SEPIC [12]–[14]. Generally, an increased voltage gain is attained in
SEPIC by coupled inductor as well as isolated transformer approaches. These approaches in turn exhibit the
drawbacks of isolated and coupled inductors whereas converters with non-coupled inductor have reduced
voltage gain [15]. Considering the aforementioned shortcomings, the SEPIC requires further improvement to
enhance its performance and generate improved voltage gain ratio.
In order to sustain a constant DC link voltage, a closed loop control is preferred which employs
proportional-integral (PI) controller. It is the most feasible and simplest controller used for wide operating
conditions [16]. The settling time requirement is satisfied by the proportional gain and steady state error is
decreased by an integral gain. The PI controllers track the reference values and hence a satisfactory steady
and dynamic response is determined by the fine-tuning of PI controller [17]. Considering uncertain, non-
linear and complex systems, tuning of PI controller parameters is challenging with the adopting of traditional
approaches like linear programming [18]. Initially, trial and error, conventional Ziegler-Nichols approaches
are utilized for the tuning of polarization index controller parameters but they are not appropriate for random
load variations [19]. The optimization approaches are alternate methods for tuning PI controllers in which the
PI gains are estimated by these approaches [20]. The selection of the PI controller parameters is crucial and
recently, several nature inspired algorithms are employed for the optimization process [21], [22]. The power
obtained from DC link is distributed to the grid by means of a voltage source inverter which injects an AC
output to grid with reduced total harmonic distortion (THD) [23]. Due to grid disturbances and voltage
distortion at point of common coupling (PCC) by nonlinear loads, grid synchronization is considered as a
challenging task [24]. The significant focus of grid synchronization is to enhance the control performance
thereby injecting a high-quality power in the grid [25].
Therefore, the contributions of this research are concluded as below: An efficient grid-connected PV
system is designed with improved SEPIC for boosting the input power from PV array. A novel lion swarm
optimization is proposed for the tuning of PI controller parameters which in turn effectively controls the
converter operation for maintaining a constant DC link voltage. The obtained power with reduced harmonics
is fed to a grid through a 1𝜙 VSI for providing effective grid synchronization by delivering appropriate
reactive power.

2. PROPOSED SYSTEM DESCRIPTION

The PV systems show a remarkable progress from a basic standalone site to large grid connected
systems. The interconnection of PV with the grid aids in ensuring ceaseless supply of power. The need for
control and supervision algorithms increases due to the wide adopting of PV systems worldwide. Here an
efficient optimization algorithm for grid connected PV using improved SEPIC is adopted as
presented in Figure 1.

Figure 1. Block diagram of the proposed work

Int J Appl Power Eng, Vol. 13, No. 2, September 2024: 603-615
Int J Appl Power Eng ISSN: 2252-8792  605

The obtained solar power from photovoltaic system relies on the intensity and illumination of the
light which necessitates the usage of a converter. Hence, improved SEPIC is employed to improve the lower
photovoltaic output voltage and also buck the output voltage at times of huge intensity. A constant output is
maintained at the DC link by a PI controller which adopts lion swarm optimization for tuning its gain
parameters. The algorithm provides superior results with improved convergence speed and robustness
thereby generating enhanced proportional and integral gains. The DC voltage is further supplied to grid
through a 1𝜙 VSI with LC filter which in turn provides effective synchronization and reduced THD with the
help of conventional polarization index controller. A whole setup provides a stable power supply to the grid
in an effective manner.

2.1. PV system
In case of a PV connected to grid, the generated power is uploaded to grid for the process of direct
transmission, distribution and consumption. Generally, operating temperature and irradiance of the PV cell
influence an output characteristic of a photovoltaic. Figure 2 depicts the PV system's equivalent circuit.
The equations for current and voltage are given by (1) and (2).
𝑉𝐷
𝐼𝑝ℎ = 𝐼𝐷 + 𝐼 + ( ) (1)
𝑅𝑠ℎ

𝑉 = 𝑉𝐷 − (𝐼 ∗ 𝑅𝑠 ) (2)

Here, the diode current 𝐼𝐷 is given by (3).

𝐼𝐷 = 𝐼 + (𝑒 (𝑉𝐷/𝑉𝑇 ) − 1) (3)

The characteristics of voltage and current which aids in the analysis of irradiance variation and temperature
effect is predict in (1) and (2). When the irradiance shows variations, the fluctuations of open circuit voltage
is small whereas short circuit current exhibits sharp fluctuations. These factors affect the output voltage of
photovoltaic system and this demands the adopting of an efficient DC-DC converter for enhancing the
photovoltaic output.

2.2. Improved SEPIC operation

The output obtained from PV system is not sufficient for the grid operation and this obtains the need
of a DC-DC converter. Among them, SEPIC is known for its improved voltage gain but can be further
improved to meet the standard of high static gain. Hence an improved SEPIC is proposed in this work which
accomplishes the addition of a diode 𝐷1 , a capacitor 𝐶1 as represented in Figure 3. The output voltage
reversing the polarity of the capacitor 𝐶𝑠 charges the capacitor 𝐶1 . The improved SEPIC performs its
operation in continuous conduction mode (CCM) and the different modes of operation are explained below.

Figure 2. Photovoltaic system Figure 3. Improved SEPIC

2.2.1. Mode 1
The mode 1 operation of improved SEPIC in which switch 𝑆 is turned off at time 𝑡0 as represent in
Figure 4. An input inductor 𝐿1 transfers stored energy to the output side through diode 𝐷0 and capacitor 𝐶𝑠 .
Moreover, the stored energy of 𝐿1 is moved to a capacitor 𝐶1 through the diode 𝐷1 . Similarly, the inductor 𝐿2
transfers the stored energy to the output through a diode 𝐷0 .

2.2.2. Mode 2
In Figure 5 the equivalent circuit of mode 2 operation in which switch 𝑆 is turned on at time 𝑡1 is
represented. The diodes 𝐷0 and 𝐷1 are blocked in this mode and the inductors 𝐿1 , 𝐿2 continues to store
energy. The inductor 𝐿1 gets supplied with the input voltage and the inductor 𝐿2 gets supplied with 𝑉𝐶𝑠 and
𝑉𝐶1 in which 𝑉𝐶1 is greater than 𝑉𝐶𝑠 . A voltage across 𝐶1 is similar to the maximum voltage across all diodes
Lion swarm optimization for grid connected PV system with improved SEPIC (P. Annapandi)
606  ISSN: 2252-8792

and the switch. The total voltage across the capacitors 𝐶1 and 𝐶𝑠 indicate the output of a converter. The
average current of an inductor 𝐿1 and average current of an inductor 𝐿2 are equivalent to the input current
and output current respectively. During steady state condition, an average voltage over inductor is zero.
𝑇𝑜𝑛 (𝑉𝑐1 −𝑉𝐼 )
= (4)
𝑇𝑜𝑓𝑓 𝑉𝐼

Consider,
𝑇𝑜𝑛
𝛼= (5)
𝑇𝑜𝑓𝑓

where, 𝛼 indicates the duty cycle, 𝑇𝑜𝑛 represents the on time, 𝑇𝑜𝑓𝑓 represents the off time. The relation
between voltage across the capacitor 𝑉𝑐𝑠 and input voltage 𝑉𝐼 is given by (6).
𝑉𝑐1 1
= (1−𝛼) (6)
𝑉𝐼

Figure 4. Mode 1 operation Figure 5. Mode 2 operation

At turned-off condition of the switch 𝑆, the diodes 𝐷0 and 𝐷1 are in on state, the output voltage 𝑉0 is
mentioned as (7).

𝑉0 = 𝑉𝑐1 + 𝑉𝑐𝑠 (7)

The voltage across the capacitor 𝑉𝐶𝑠 is given by (8).

𝑉𝐶𝑠 = 𝑉0 − 𝑉𝑐1 (8)

Substitute (6) and (8) in (7) we get (9).

𝑉0 (1−𝛼)
= (1+𝛼) (9)
𝑉𝐼

The gain of the improved SEPIC which clearly indicates that the output gain highly relies on duty
cycle value is represents in (9). The obtained output DC voltage from converter is further fed to DC link at
which a constant voltage has to be maintained. In order to accomplish this, a closed loop control based on PI
controller is needed, as explained below.

2.3. Lion swarm optimized PI controller

Generally, DC link voltage has to be maintained at a constant level for achieving satisfactory results
without fluctuations. This is achieved by a PI controller which processes the error value obtained by the
comparison of the reference voltage and an actual DC link voltage. An error value applied to the PI controller
has to be zero which is attained by appropriate values of 𝐾𝑝 and 𝐾𝑖 parameters. This tuning of PI parameters
is done by optimization approaches and this work proposes lion swarm optimization for finding values of 𝐾𝑝
and 𝐾𝑖 parameters.

2.3.1. Lion swarm optimization

The lion is regarded as most social exhibiting improved levels of antagonism and cooperation. They
are classified as lion cubs, lioness and lion king where the dominance is high in lion king, since it is required to
remain stronger to retain its dominant position. The hunting lions are the lionesses, which breeds the lion cubs,
recognizes the prey location and encircles them while approaching the prey. The term following lions is

Int J Appl Power Eng, Vol. 13, No. 2, September 2024: 603-615
Int J Appl Power Eng ISSN: 2252-8792  607

associated to the lion cubs and are protected by the lion king. Lion cubs are taught about hunting by the lioness
and they seek lion king for food at times of hunger. Considering the swarm, the lions coordinate among
themselves as a team for searching food. The optimization algorithm shown in Figure 6 relies on the hunting
behavior of the swarm and the lion king is the one with best fitness value. On identifying a prey, the lion king
moves towards the prey's position whereas the lion cubs follow the lioness for learning the hunting process.
The adult lion’s proportional factor is given by 𝛽 which is a random positive number within [0,1]
range and in this work 𝛽 is set as 0.5. The term 𝛼𝑓 is used to represent the disturbance factor in moving range
of the lioness, which enhances the convergence speed by balancing the global exploration as well as local
exploitation. It is given by (10).

𝑡 10
𝛼𝑓 = 𝑠𝑡𝑒𝑝1 . 𝑒𝑥𝑝 [−30. ] (10)
𝑇

Here, 𝑇- maximum iteration, 𝑡 - current iteration, value of step within lionesses’ range of activity 𝑠𝑡𝑒𝑝1 is
provided as (11).

𝑠𝑡𝑒𝑝1 = 𝛼1 . (𝑥̅𝑚𝑎𝑥 − 𝑥̅𝑚𝑖𝑛 ) (11)

Here, for every dimension, the maximal and minimal mean value is represented by 𝑥̅𝑚𝑎𝑥 and 𝑥̅𝑚𝑖𝑛
respectively, 𝛼1 denotes the control factor. The lion cubs' moving range disturbance factor is denoted by the
symbol 𝛼𝑐 and is determined by (12).
𝑇−𝑡
𝛼𝑐 = 𝑠𝑡𝑒𝑝2 . (12)
𝑇

Where,

𝑠𝑡𝑒𝑝2 = 𝛼2 . (𝑥̅𝑚𝑎𝑥 − 𝑥̅𝑚𝑖𝑛 ) (13)

Here, 𝛼2 denotes the control factor and is any number within the scale [0,1]. The position of all lions is
denoted by (14).
x1,1 x1,2 ⋯ x1,D
𝑥=[ ⋮ ⋮ ⋮ ] (14)
xn,1 xn,2 ⋯ xn,D

Here 𝑥𝑖,𝑗 denotes 𝑗𝑡ℎ dimension related to 𝑖 𝑡ℎ lion, each 𝐷 dimensional vector 𝑥𝑖 = (𝑥𝑖,1 , 𝑥𝑖,2 , … . . 𝑥𝑖,𝐷 )
denote the state of 𝑖 𝑡ℎ lion and is provided by (15).

𝑥𝑖,𝑗 = 𝑥𝑚𝑖𝑛,𝑗 + 𝑟𝑎𝑛𝑑(0,1). (𝑥𝑚𝑎𝑥,𝑗 − 𝑥𝑚𝑖𝑛,𝑗 ) (15)

Here, 𝑖 = 1,2, … 𝑛 and 𝑗 = 1,2, … 𝐷, 𝑟𝑎𝑛𝑑 (0,1) specifies a random number which is uniformly distributed
in range [0,1], 𝑥𝑚𝑎𝑥,𝑗 denotes upper bound and 𝑥𝑚𝑖𝑛,𝑗 denotes lower bound of 𝑗𝑡ℎ dimension. The number of
the adult lions is given as below.

𝑛𝐿𝑒𝑎𝑑𝑒𝑟 = [𝑛. 𝛽] (16)

And the number of lion cubs is given by 𝑛 − 𝑛𝐿𝑒𝑎𝑑𝑒𝑟 the appropriate fitness values are below.

f1 ([x1,1 , x1,2 , L, x1,D ])

f2 ([x2,1, , x2,2 , L, x2,D ])
𝑓= ⋮ (17)
⋮
[fn ([xn,1 , xn,2 , L, xn,D ])]

Quality of prey sought by each lion is represented by the fitness value of its position. As a result,
their chances of survival are likewise increased. Each lion’s place in lion swarm optimization (LSO) is
modified based on its own experience as well as that of its neighbors. As previously stated, the hunting
mechanisms of each lion is different during the hunting phase. Around the prey, a circle shaped
neighborhood is created in order to support lions coming from various directions, thus the problem of
trapping in local optima is successfully avoided. The increase in population diversity is also possible using
this scheme.

Lion swarm optimization for grid connected PV system with improved SEPIC (P. Annapandi)
608  ISSN: 2252-8792

Figure 6. Flow chart of lion swarm optimization

The position with the least fitness value is the destination of the lion king, to increase that he has
priority for catch over the other lions. New status of lion king, as in (18).

𝑥𝑖 (𝑡 + 1) = 𝑔𝑏𝑒𝑠𝑡(𝑡). (1 + 𝛾. ‖𝑝𝑏𝑒𝑠𝑡𝑖 (𝑡) − 𝑔𝑏𝑒𝑠𝑡(𝑡)‖) (18)

Where, 𝑡 =present iteration value, 𝑔𝑏𝑒𝑠𝑡(𝑡)= at 𝑡, the globally best position of catch, γ = random number lies
between [0, 1] 𝑝𝑏𝑒𝑠𝑡𝑖 (𝑡)=at 𝑡, best position of 𝑖 𝑡ℎ lion.
Lionesses often hunt by recognizing their prey's position, encircling them, and then attacking them.
When a lioness engages in hunting behavior, she usually does so with the help of another lioness. The
collaboration lioness is the lioness who was chosen from the lioness group by someone other than herself.
How the new lioness position might be obtained in this case is shown in (19).

𝑝𝑏𝑒𝑠𝑡𝑖 (𝑡)+𝑝𝑏𝑒𝑠𝑡𝑐 (𝑡)

𝑥𝑖 (𝑡 + 1) = . (1 +∝𝑓 . 𝛾) (19)
2

Where, ∝𝑓 -, moving range disturbance factor of lioness. 𝑝𝑏𝑒𝑠𝑡𝑐 (𝑡)-, at t, cooperation lioness best position.
The new position of lion cub is given by (20).

Int J Appl Power Eng, Vol. 13, No. 2, September 2024: 603-615
Int J Appl Power Eng ISSN: 2252-8792  609

𝑔𝑏𝑒𝑠𝑡(𝑡)+𝑝𝑏𝑒𝑠𝑡𝑖 (𝑡) 1
. (1 + 𝛼𝑐 . 𝛾), 𝑞 ≤
2 3
𝑝𝑏𝑒𝑠𝑡𝑚 (𝑡)+𝑝𝑏𝑒𝑠𝑡𝑖 (𝑡) 1 2
𝑥𝑖 (𝑡 + 1) = . (1 +∝𝑐 . 𝛾), < 𝑞 < (20)
2 3 3
̅̅̅̅̅̅̅̅ (𝑡)+𝑝𝑏𝑒𝑠𝑡𝑖 (𝑡)
𝑔𝑏𝑒𝑠𝑡 2
{ . (1 +∝𝑐 . 𝛾), ≤ 𝑞 ≤ 1
2 3

Where, 𝑝𝑏𝑒𝑠𝑡𝑚 (𝑡) –best position followed by lion cub, as in (21).

̅̅̅̅̅̅̅̅
𝑔𝑏𝑒𝑠𝑡 (𝑡) - Position of 𝑖 𝑡ℎ lion cub
̅̅̅̅̅̅̅̅
𝑔𝑏𝑒𝑠𝑡 (𝑡) = ̅̅̅̅̅̅
𝑥𝑚𝑎𝑥 + ̅̅̅̅̅̅
𝑥𝑚𝑖𝑛 − 𝑔𝑏𝑒𝑠𝑡(𝑡) (21)

∝𝑐 – moving range disturbance factor, 𝑞-random number range [0, 1]. Based on the values of 𝑔𝑏𝑒𝑠𝑡
and 𝑝𝑏𝑒𝑠𝑡, optimal outputs are obtained which are used for the efficient tuning of the 𝐾𝑝 and 𝐾𝑖 parameters.
The obtained values of PI controller parameters outperform the existing methods which in turn efficiently
controls the operation of improved SEPIC for maintaining a constant DC link voltage.

2.4. Grid connected 𝟏𝝓 VSI

Generally, the low powered PV systems are connected to AC grid through 1𝜙 inverters. Voltage
source inverter executes active mitigation of harmonics with the implementation of an LC filter as well as the
associated circuit diagram is represented in Figure 7. A conventional PI controller achieves grid
synchronization by reducing the error between the real and reference power values. The outputs obtained are
converted to appropriate pulses to be fed to VSI. Voltage source inverter changes the DC voltage from the
DC link into AC form for grid injection. The Voltage source inverter operates as a current source rather than
a voltage source while in grid linked mode. An inverter injects higher quality power into the grid at
controllable changes in voltage, frequency, and phase angle as a consequence of the synchronization. On the
single-phase grid, this outcomes in the output being a high-quality integrated voltage. Thus, a proposed
methodology effectively presents a grid connected PV system with improved SEPIC adopting lion swarm
optimized PI controller for the production of improved outputs.

Figure 7. 1𝜙 grid connected VSI

3. RESULTS AND DISCUSSION

The grid connected PV systems contribute to majority of the installed capacity worldwide when
compared to battery utilized standalone systems. These systems adopt an efficient converter along with
control approaches for maintaining a constant DC link voltage. This work proposes an improved SEPIC with
lion swarm optimization and the corresponding simulation is performed in MATLAB/Simulink. Table 1
gives the specifications of solar panel as well as improved SEPIC converter.

Table 1. Design parameter specifications

Parameter Values Parameter Values
Power 2000 𝑊 Input capacitor 𝐶1 , 𝐶𝑠 870 𝜇𝐹
No of solar panels connected in series 36 Output capacitor 𝐶𝑂 2200 𝜇𝐹
No of panels 100 𝑊, 20 𝑝𝑎𝑛𝑒𝑙𝑠 Switching frequency 𝑓𝑠 10 𝑘𝐻𝑧
𝐿1 , 𝐿2 5 𝑚𝐻 𝑅𝑂 100 Ω

Initially, the input solar irradiance is maintained at 980 W/m 2 for 0.1s and then it is increased to
1000 W/m2 as indicated in Figure 8(a). The solar irradiance is varied to evaluate the dynamic nature of the
Lion swarm optimization for grid connected PV system with improved SEPIC (P. Annapandi)
610  ISSN: 2252-8792

suggested control technique in non-linear operating conditions. In accordance to the increase in irradiance,
the PV voltage also increases from 79 V to 80 V at 0.1 sec as depicted in Figure 8(b).
Figures 9(a) and 9(b) specifies an input current and output power waveform of a PV panel. As
demonstrated by Figure 9(a), the photo generated current from the PV is 17.2 A till 0.1s and it increases to
18.8 A corresponding to the change in irradiance. Similar to this, Figure 9(b) shows that the PV power which
is initially 1360 W, becomes 1500 W at 0.1s in response to the increase in irradiance.

(a) (b)

Figure 8. PV parameter waveform: (a) irradiation and (b) voltage

(a) (b)

Figure 9. Waveforms of PV: (a) input current and (b) output power

The output of the improved SEPIC for different control techniques is illustrated in Figure 10. The
converter output with the employment of PI controller seen in Figure 10(a) is affected by peak overshoot
condition at first and then it delivers a voltage of 270 V from 0.12 s. However, the output that is obtained is
unstable and is affected by fluctuations. Figure 10(b) illustrates GWO based PI controller output, where
peak overshoot condition arises as like PI controller, but it is able to provide a stable voltage at a quicker time
of 0.1 s. Finally, the proposed lion swarm optimized PI is successful in providing an output without peak
overshoot condition in quickest settling time of 0.08s, which is evident from Figure 10(c). Moreover, the
change in operating conditions does not have an impact on the output of the lion swarm optimized PI based
improved SEPIC. A stable voltage and current of 230 V, 5 A is obtained from the single-phase grid as shown
in Figure 11. Additionally, as Figure 12 indicates, the suggested control approach is successful in minimizing
THD at 2.9%.

Int J Appl Power Eng, Vol. 13, No. 2, September 2024: 603-615
Int J Appl Power Eng ISSN: 2252-8792  611

(a) (b)

(c)

Figure 10. Output voltage of improved SEPIC: (a) with PI controller, (b) GWO-PI,
and (c) lion swarm-PI

Figure 11. Grid output voltage and current Figure 12. THD output

3.1. Hardware results

The proposed concept is experimentally validated using field programmable gate arrays (FPGA),
and the properties of the introduced power conversion system are examined. FPGAs are a great alternative
because of qualities like great configurability and performance. To implement the hybrid system in FPGA,
the entire module has to be converted to an FPGA synthesizable module and the hardware implementation is
carried out in DSPIC3050FPGA controller as shown in Figure 13.

Lion swarm optimization for grid connected PV system with improved SEPIC (P. Annapandi)
612  ISSN: 2252-8792

The hardware validation of the proposed work generates the following outputs which indicates
the efficacy of the proposed system. The corresponding waveforms are presented below with a detailed
analysis. The input solar irradiance is presented in Figure 14, where a slight increase in value of solar
irradiance is seen at one point. This definitely influences the operation of PV panel which in turn generates
varying outputs.
Figure 15 demonstrates a photovoltaic panel current waveform and output voltage, which highly
relies on available temperature, solar irradiation and other environmental conditions. So, a variation in
voltage output with the change in solar irradiation is observed from the waveform, seen in Figure 15(a).
Similar to the PV voltage output, Figure 15(b) depicts the output current also increases in line with the rise in
solar irradiation. As seen in Figure 16, the improved SEPIC converter produces a stable, controlled output
with less amount of ripple contents. By boosting efficiency and decreasing input current ripples, the
improved SEPIC solves the shortcomings of regular SEPIC.

Figure 13. Hardware setup Figure 14. Solar irradiation

(a) (b)

Figure 15. Photovoltaic panel: (a) output voltage and (b) output current

Figure 16. Output voltage of improved SEPIC

Int J Appl Power Eng, Vol. 13, No. 2, September 2024: 603-615
Int J Appl Power Eng ISSN: 2252-8792  613

Figures 17(a) and 17(b) show a timeline of the voltage and current variations on the grid. These
graphs provide information about the dynamic voltage and current behaviour inside the grid system. This
indicates an enhanced grid synchronization by the proposed topology.
The THD value is to be kept as low as possible for the system's stability and improved power
quality. The implementation of the proposed approach yields a lower THD value, and Figure 18 illustrates
the corresponding output. Figure 19(a) displays the efficacy comparison between improved SEPIC and
conventional converters such as buck-boost, Cuk, boost, and SEPIC. From figure it is observed that improved
SEPIC has a better efficiency of 96% respectively. The improved SEPIC converter's performance is then
displayed in Figure 19(b), where the voltage gain value is 1:10, which is relatively higher than state of art
converter topologies.

(a) (b)

Figure 17. Grid: (a) voltage and (b) current

Figure 18. THD output

(a) (b)

Figure 19. DC-DC converter: (a) efficiency comparison and (b) voltage gain comparison

The obtained THD values for current are compared with Cuk and SEPIC as shown in Figure 20(a).
The proposed converter generates an improved THD of 2.9% and 3.9% for simulation and hardware
implementation respectively. The settling time of lion swarm optimized PI controller is contrasted with
conventional PI controller and GWO based PI controller in Figure 20(b). From given comparison chart, it is
determined the proposed control technique offers the fastest settling time of 0.08s in case of simulation and
0.09s in case of hardware validation.
Lion swarm optimization for grid connected PV system with improved SEPIC (P. Annapandi)
614  ISSN: 2252-8792

(a) (b)

Figure 20. Performance measure of (a) current THD and (b) settling time of controllers

4. CONCLUSION
An optimum control strategy for grid connected photovoltaic systems is presented in this study,
since grid-connected PV systems account for the vast majority of installed capacity worldwide when
compared to battery-based standalone systems. The voltage derived from the photovoltaic panel is
transformed to the desired level using improved SEPIC of high voltage gain and an efficiency of 96%. The
improvement of dynamic performance indices of converter in terms of over shoot and settling time is
achieved using lion swarm optimized PI controller. Thus, from the obtained outcomes, it is noted that the
suggested control technique is successful in eliminating the overshoot problem in the converter and provides
a quick settling time of 0.08𝑠 for simulation and 0.09𝑠 for hardware validation.

REFERENCES
[1] K. S. Kavin and P. S. Karuvelam, “PV-based grid interactive PMBLDC electric vehicle with high gain interleaved DC-DC SEPIC
converter,” IETE Journal of Research, vol. 69, no. 7, pp. 4791–4805, Sep. 2023, doi: 10.1080/03772063.2021.1958070.
[2] A. Kumar and P. Kumar, “Power quality improvement for grid-connected PV system based on distribution static compensator
with fuzzy logic controller and UVT/ADALINE-based least mean square controller,” Journal of Modern Power Systems and
Clean Energy, vol. 9, no. 6, pp. 1289–1299, 2021, doi: 10.35833/MPCE.2021.000285.
[3] P. Rodriguez, C. Citro, J. I. Candela, J. Rocabert, and A. Luna, “Flexible grid connection and islanding of SPC-based PV power
converters,” IEEE Transactions on Industry Applications, vol. 54, no. 3, pp. 2690–2702, May 2018, doi: 10.1109/TIA.2018.2800683.
[4] F. Rong, X. Gong, and S. Huang, “A novel grid-connected PV system based on MMC to get the maximum power under partial shading
conditions,” IEEE Transactions on Power Electronics, vol. 32, no. 6, pp. 4320–4333, Jun. 2017, doi: 10.1109/TPEL.2016.2594078.
[5] M. I. Marei, B. N. Alajmi, I. Abdelsalam, and N. A. Ahmed, “An integrated topology of three-port DC-DC converter for PV-battery
power systems,” IEEE Open Journal of the Industrial Electronics Society, vol. 3, pp. 409–419, 2022, doi: 0.1109/OJIES.2022.3182977.
[6] K. S. Tey, S. Mekhilef, M. Seyedmahmoudian, B. Horan, A. T. Oo, and A. Stojcevski, “Improved differential evolution-based
MPPT algorithm using SEPIC for PV systems under partial shading conditions and load variation,” IEEE Transactions on
Industrial Informatics, vol. 14, no. 10, pp. 4322–4333, Oct. 2018, doi: 10.1109/TII.2018.2793210.
[7] M. Forouzesh, Y. Shen, K. Yari, Y. P. Siwakoti, and F. Blaabjerg, “High-efficiency high step-up DC–DC converter with dual
coupled inductors for grid-connected photovoltaic systems,” IEEE Transactions on Power Electronics, vol. 33, no. 7, pp. 5967–
5982, Jul. 2018, doi: 10.1109/TPEL.2017.2746750.
[8] K. Nathan, S. Ghosh, Y. Siwakoti, and T. Long, “A new DC–DC converter for photovoltaic systems: coupled-inductors combined Cuk-
SEPIC converter,” IEEE Transactions on Energy Conversion, vol. 34, no. 1, pp. 191–201, Mar. 2019, doi: 10.1109/TEC.2018.2876454.
[9] Q. Huang, A. Q. Huang, R. Yu, P. Liu, and W. Yu, “High-efficiency and high-density single-phase dual-mode cascaded buck–
boost multilevel transformerless PV inverter with GaN AC switches,” IEEE Transactions on Power Electronics, vol. 34, no. 8,
pp. 7474–7488, Aug. 2019, doi: 10.1109/TPEL.2018.2878586.
[10] Y. P. Siwakoti et al., “High-voltage gain Quasi-SEPIC DC–DC converter,” IEEE Journal of Emerging and Selected Topics in
Power Electronics, vol. 7, no. 2, pp. 1243–1257, Jun. 2019, doi: 10.1109/JESTPE.2018.2859425.
[11] S. Padmanaban, N. Priyadarshi, M. S. Bhaskar, J. B. Holm-Nielsen, E. Hossain, and F. Azam, “A hybrid photovoltaic-fuel cell for
grid integration with Jaya-based maximum power point tracking: experimental performance evaluation,” IEEE Access, vol. 7, pp.
82978–82990, 2019, doi: 10.1109/ACCESS.2019.2924264.
[12] S. Hasanpour, M. Forouzesh, Y. Siwakoti, and F. Blaabjerg, “A new high-gain, high-efficiency SEPIC-based DC–DC converter
for renewable energy applications,” IEEE Journal of Emerging and Selected Topics in Industrial Electronics, vol. 2, no. 4, pp.
567–578, Oct. 2021, doi: 10.1109/JESTIE.2021.3074864.
[13] D. S. Sankaralingam, M. S. B. Natarajan, M. Muthusamy, and S. B. Singh, “A novel metaheuristic approach for control of SEPIC
converter in a standalone PV system,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 13, no. 2, pp.
1082–1092, Jun. 2022, doi: 10.11591/ijpeds.v13.i2.pp1082-1092.
[14] M. Vaigundamoorthi, R. Ramesh, V. V. Prabhu, and K. A. Kumar, “MPPT oscillations minimization in PV system by controlling
non-linear dynamics in SEPIC DC-DC converter,” International Journal of Electrical and Computer Engineering (IJECE), vol.
10, no. 6, pp. 6268–6275, Dec. 2020, doi: 10.11591/ijece.v10i6.pp6268-6275.
[15] S. A. Ansari and J. S. Moghani, “A novel high voltage gain noncoupled inductor SEPIC converter,” IEEE Transactions on
Industrial Electronics, vol. 66, no. 9, pp. 7099–7108, Sep. 2019, doi: 10.1109/TIE.2018.2878127.
[16] M. S. Ali, L. Wang, H. Alquhayz, O. U. Rehman, and G. Chen, “Performance improvement of three-phase boost power factor
correction rectifier through combined parameters optimization of proportional-integral and repetitive controller,” IEEE Access,
vol. 9, pp. 58893–58909, 2021, doi: 10.1109/ACCESS.2021.3073004.
[17] S. Kakkar et al., “Design and control of grid-connected PWM rectifiers by optimizing fractional order pi controller using water
cycle algorithm,” IEEE Access, vol. 9, pp. 125941–125954, 2021, doi: 10.1109/ACCESS.2021.3110431.

Int J Appl Power Eng, Vol. 13, No. 2, September 2024: 603-615
Int J Appl Power Eng ISSN: 2252-8792  615

[18] M. I. Mosaad, H. S. M. Ramadan, M. Aljohani, M. F. El-Naggar, and S. S. M. Ghoneim, “Near-optimal PI controllers of STATCOM
for efficient hybrid renewable power system,” IEEE Access, vol. 9, pp. 34119–34130, 2021, doi: 10.1109/ACCESS.2021.3058081.
[19] V. Veerasamy et al., “A hankel matrix based reduced order model for stability analysis of hybrid power system using PSO-GSA
optimized cascade PI-PD controller for automatic load frequency control,” IEEE Access, vol. 8, pp. 71422–71446, 2020, doi:
10.1109/ACCESS.2020.2987387.
[20] T. Appala Naidu, S. R. Arya, R. Maurya, and S. Padmanaban, “Performance of DVR using optimized PI controller based gradient
adaptive variable step lms control algorithm,” IEEE Journal of Emerging and Selected Topics in Industrial Electronics, vol. 2, no.
2, pp. 155–163, Apr. 2021, doi: 10.1109/JESTIE.2021.3051553.
[21] A. Ahmad et al., “Controller parameters optimization for multi-terminal DC power system using ant colony optimization,” IEEE
Access, vol. 9, pp. 59910–59919, 2021, doi: 10.1109/ACCESS.2021.3073491.
[22] A. M. Hussien, R. A. Turky, H. M. Hasanien, and A. Al-Durra, “LMSRE-based adaptive PI controller for enhancing the performance
of an autonomous operation of microgrids,” IEEE Access, vol. 9, pp. 90577–90586, 2021, doi: 10.1109/ACCESS.2021.3091496.
[23] B. Guo et al., “A robust second-order sliding mode control for single-phase photovoltaic grid-connected voltage source inverter,”
IEEE Access, vol. 7, pp. 53202–53212, 2019, doi: 10.1109/ACCESS.2019.2912033.
[24] S. Swain and B. Subudhi, “Grid synchronization of a PV system with power quality disturbances using unscented Kalman
filtering,” IEEE Transactions on Sustainable Energy, vol. 10, no. 3, pp. 1240–1247, Jul. 2019, doi: 10.1109/TSTE.2018.2864822.
[25] S. Kewat and B. Singh, “Grid synchronization of WEC-PV-BES based distributed generation system using robust control strategy,”
IEEE Transactions on Industry Applications, vol. 56, no. 6, pp. 7088–7098, Nov. 2020, doi: 10.1109/TIA.2020.3021060.

BIOGRAPHIES OF AUTHORS

P. Annapandi received the bachelor Degree in Electrical and Electronics Engineering

from Government college of Engineering, Tirunelveli Tamil Nadu in May 2000 and Master degree
in power Electronics and Drives from College of Engineering, Guindy, Anna University, Chennai,
Tamil Nadu, in May 2006, and Ph.D. in Electrical Engineering at Anna University, Chennai, Tamil
Nadu, in 2013. His research interests include smart grid and power quality. He can be contacted at
email: annapandian2001@gmail.com.

D. Lakshmi associate professor of Department of Electrical and Electronics

Engineering, Academy of Maritime Education and Training, deemed to be University, Chennai,
Tamil Nādu, India. She has more than 22 years of expertise in the field of power system. She has
guided more than 30 UG students, 15 PG students and 4 research scholars. She has published 6
book chapters and 4 books, 41 articles in SCI and Scopus indexed journals and nearly 25 national
level and international level conferences proceedings. She has received many awards from various
organizations for her contribution towards the assigned work. She is a life member in professional
bodies such as IEEE, IEI, IAENG, and InSc. She is acting as editorial manager and reviewer for
various reputed journals. She received many funds from Govt agencies and reputed organizations
for contributing the recent research and product developments. She is a star organizer in Igen
Energathon-2023 Marathon, a new world record on longest conference. Her areas of interest are
power system operation and control, soft computing renewable energy systems, microgrid, and
electrical machines. She can be contacted at email: lakshmiee@gmail.com.

B. Kavya Santhoshi is working as an assistant professor in the Department of Electrical

and Electronics Engineering at Godavari Institute of Engineering and Technology (A), Rajahmundry,
Andhra Pradesh, India. She graduated in Electrical and Electronics Engineering at Saveetha
Engineering College (Anna University), Chennai, Tamil Nadu, India. She secured Master of
Engineering in Power Electronics and Drives at Jeppiaar Engineering College (Anna University),
Chennai, Tamil Nadu, India. She secured Ph.D. in Electrical Engineering at Anna University,
Chennai, Tamil Nadu, India. She is in the field of Power Electronics at Godavari Institute of
Engineering and Technology (A), Rajahmundry, Andhra Pradesh, India. She is in teaching profession
for more than 8 years. She has presented 30+ papers in National and International Journals,
Conference, and Symposiums. Her main area of interest includes power electronics and renewable
energy systems. She can be contacted at email: kavyabe2010@gmail.com.

P. Annapoorani obtained Bachelor’s degree in Electrical and Masters in Process

Control and Automation. She has been teaching history for over 9 years. Presently working at Vel
Tech Multi Tech Dr. Rangarajan Dr. Sakunthala Engineering College. Her area of interest in
control systems, power systems. The area of interest includes control instrumentation and
measurements. She contributed many numbers of active publications in reputed journals. She can
be contacted at email: annapoorani@veltechmultitech.org.

Lion swarm optimization for grid connected PV system with improved SEPIC (P. Annapandi)