Optimal Power Quality Enhancement using a Radial Distribution
System with an Improved Unified Power Quality Conditioner
OLUWAFUNSO OLUWOLE OSALONI1, AYODEJI STEPHEN AKINYEMI2,
ABAYOMI ADURAGBA ADEBIYI2, KATLEHO MOLOI2, AYODEJI OLALEKAN SALAU1,3
1Department Electrical Electronic and Computer Engineering,
Afe Babalola University, Ado Ekiti, Ekiti-State,
NIGERIA
2Department Electrical Power Engineering,
Durban University of Technology, Durban,
SOUTH AFRICA
3Saveetha School of Engineering,
Saveetha Institute of Medical and Technical Sciences,
INDIA
Abstract: - Massive electric power distribution over long distances with consequential Power Quality (PQ)
challenges such as voltage sags and power losses are some of the significant attributes of a Radial Distribution
Network (RDN). Deployment of Power Angle Regulated (PAR) based Unified Power Quality Conditioner
(UPQC) in a distribution network is also securing attraction because of the latest recorded achievements and
improvements in Voltage Source Inverter (VSI) built power electronic systems. However, optimal allocation of
this kind of device to mitigate PQ problems remains a challenge for achieving set objectives. Consequently, this
study considers the best possible allocation of PAR and Improved-UPQC know as I-UPQC in the distribution
network to enhance power network performance. The identification of optimal location is achieved through the
application of hybridization of the Genetic Algorithm and Improved Particle Swarm Optimization (GA &
IPSO). The deterministic approach is based on the weight factor of various objective functions. The allocation
is attained with a selection of reactive power control between inverter connected in parallel and series and
control angle variables of the device through its dynamic involvement of total system loss derivatives.
Performances of the I-UPQC based distribution system during diverse power transfers are observed.
Convergence characteristic of deterministic approach at different disturbance percentages is analyzed and
presented. Imaginary power circulation enhanced the voltage-associated challenges at the range of 0.949 to
0.9977. Hence, power dissipation minimized to 1.15 percent compared to the initial 3.35 percent, according to
results of I-UPQC allocation in RDN utilizing mathematical and optimization technique. Additionally, the
network losses, voltage dip, and minimum bus voltage profile all fall within the regulatory standards of less
than 2%, 5%, and 5%, correspondingly. Also, the performance of the compensated network for both ordinary
and optimized scenarios indicated the fitness of the projected method in accomplishing an operational
optimization of RDN, specifically for voltage profile (VP) improvement and I-UPQC's series and shunt inverter
share imaginary power.
Key-Words: - Power Quality, Unified Power Quality Conditioner, Power Angle Control, Particle Swarm
Optimization, Genetic Algorithm.
Received: August 19, 2022. Revised: August 16, 2023. Accepted: September 13, 2023. Published: October 9, 2023.
1 Introduction
The deficiency of imaginary power in grids results
in variability, which triggers voltage drop and
oscillation. Ameliorating imaginary power at an
optimal point would eliminate voltage instability
problems and substantially enhance the grid
reaction. Man-aging imaginary power is considered
compensation, while the compensator installation is
usually placed where there are imaginary loads.
Many authors have come up with re-active power
compensation and sag/swell mitigation in recent
times. Presentation of tap changing transformer
which is used to control imaginary power in, [1], but
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
158
Volume 18, 2023
because of transformer tap has a low range limits its
application (can result in voltage swings). Optimal
capacitor placement is determined in, [2], [3], and
this is achieved using a heuristic search-based
approach. Due to factors like controlling existing
ameliorators and the financial costs of removing
some absorbing loads, this method has gained less
support than the hypothetical power injection
option, which would make its shortcomings far
more obvious. The task was completed in, [2],
where a cuckoo search-based approach had been
used to distribute static shunt capacitors in the RDN.
The process of putting in a constant and changeable
regulating capacitive load has been used in, [4]. The
hypothetical energy provided by the shunt
capacitors described in, [4], has an extremely poor
response time for unexpected varying loads.
In, [5], a fuzzy concept-based optimization
method is used to allocate multi-objective capacitors
in distribution systems. Furthermore, another power
quality device is a flexible AC transmission system
(FACT), but they find their application in optimal
imaginary power amelioration in the LV distribution
network. A distribution static compensator
(DSTATCOM), for example, is an extremely costly
FACTs device, [6]. The com-bination of VSI
connected in series and parallel connected with DC-
link back-to-back is considered as a general solution
for imaginary power compensation, which is named
UPQC, are utilized. Additionally, UPQC became a
multi-dimensional FACT technology due to its
capacity to correct for a variety of PQ-related
factors, including voltage dips, overvoltage,
imbalance voltage, imaginary current, flashes, and
distortions, [7]. By injecting a monitored and
adjustable quadrature voltage simultaneously to
address under-voltage difficulties, the VSI linked in
series can produce real power and imaginary power,
[8]. The work done in [9] present a comparative
analysis of models of UPQC.
However, the series VSI connected in series can
concurrently produce imaginary and real power, as
shown in, [10]. In most of the articles on UPQC, it
can be observed that it is either protecting a single
load or connected to two bus systems that contain
mostly non-linear or sensitive loads. Considering
the possibility of the presence of this type of load in
RDN and the level of sensitivity required for its
protection, there is a need for the proper allocation
of I-UPQC in the larger network. I-UPQC can
produce imaginary power amelioration in a
distribution network, [11]. The use of such a method
enables the de-termination of the assignment
optimizing ameliorator, in which all grid variables
seem to be in their ideal state. In, [12], imaginary
power amelioration is performed with PAC in this
work, and amelioration without including
microgrids is performed. This allows for
enhancement on, [13], by taking into account DG
supply, which is a crucial tendency to use due to
various consumer preferences. Electricity issues like
those seen in DGs studies will unquestionably be
impacted by the introduction of loads that utilize
fictitious energy in operational systems. As a result,
[14], proposes an ideal strategy that implements the
simultaneous deployment of DGs with dormant
advanced metering systems. Owing to the ability of
DGs to turn passive network to active network it
was treated in but on the contrary using DG with
UPQC is not mentioned. However, due to the clean
nature of renewable energy, no cost of fuel, its
integration into RDN through PAC of UPQC re-
quired more attention.
The goal of this research is to investigate the
best apportionment to modify the I-UPQC
architecture compared to ordinary UPQC in the
RDN in such a way as to strengthen the PQ in the
power system. The contribution of the article
includes the following:
Development of proper application of GE-IPSO
in RDN
Reactive power-sharing with the utilization of
both inverters of I-UPQC simultaneously in two
bus networks analytically
Optimal PQ enhancement through the impact of
DG interconnection through I-UPQC on network
power dissipation and voltage dip/surge using
photovoltaic DG in an RDN network.
2 Basic I-UPQC Modelling
Figure 1 shows the practical block diagram of
UPQC, an integrated optimizer of PQ centered on
PAC which is employed in this study and
incorporates two series & shunt inverters. The series
inverter is utilized to reduce supply voltage dip.
Additionally, shunt inverters are employed for
correction when harmonics and the imaginary
element of the load current are present. As indicated
in Figure 2a and Figure 2b, series voltage 󰇛󰇜 and
shunt compensatory current (󰇜 would be injected
by respective series inverter and shunt inverter
under normal and dip voltage conditions. The series
inverter injects electricity under convectional
operational requirements in the UPQC-PAC design.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
159
Volume 18, 2023
Fig. 1: Basic block diagram of UPQC, [11]
Fig. 2: Evaluation of shunt and series amelioration
(a) under a voltage dip (b) at optimum conditions
2.1 Multi-Objective Planning Concept
This study presents detailed development of a multi-
objective framework that identifies the best site for
I-UPQC, the best value for ameliorating imaginary
output at the preferred destination, and the best
amount of . To concurrently identify the
optimal variable, three objectives are minimized.
They comprise, [15]:
First Goal 1: Rating of I-UPQC (VA)
Goal 1:   
 denotes the objective function. such that
,  represent the apparent powers of series and
shunt VSI, respectively.
Second Goal 2: RDN power loss
Goal 2: 
 󰇝󰇛󰇜󰇞󰇛󰇜

such that, 󰇛󰇜, 󰇛󰇜 represents the branch
line current and resistance and set comprise of all
branches in a network.
Third Goal 3: Degree of nodes with under
voltage problem (DNUVP)
Goal 3: DNUVP = 



The number of the bus with under-voltage with I-
UPQC is denoted with 
 , and 
 indicate
the number of bus haven under-voltage without I-
UPQC.
2.2 Genetically Modifies Particle Swarm
Optimization
GA-IPSO is a variety of PS that genetically
modifies to optimally solve multi-objective
problems in which equivalent Fitness Function (FF)
are attached to the particles. The state-of-art
literature review in, [16], [17], [18], show many
PSO variations. The Pareto-dominance method
populates most of the optimization strategies. The
actual goals of each of these methods are to find a
set of Non-Dominated Results (NDR) that is closer
to the set of pareto-optimal solutions (superior
convergence) and to guarantee minimal global
convergence as opposed to regular PSO. Hence,
with selection properly guided by each particle,
better convergence can be achieved. For this GA-
IPSO, two variants were used, which include PSO
and GA. The behavior of the non-dominated sorted
genetic algorithm stimulated the PSO, [19]. PSO
uses the population of individuals from the most
recent iteration and the one before it to identify non-
dominated solutions. So, the Pareto optimal
solutions are sorted using a niche technique. The
PSO, on the other hand, is built on a collection of
elite archived NDS developed by optimization
algorithms and is applied to distribute fitness to each
member of the archive while the present population
is evolving. According to the PSO's global optimal
(gbest) architecture, the IPSO particle guide
selection process is carried out, [20], [21].
Begin
// size of IPSO population
// =Highest amount of iterations
Produce preliminary population for IPSO at
random by means of the encoding system.
Evaluation of the objective functions and
explanation of the particle;
Find the first, undominated resolution.;
Learn the basic guideline;
Repetition=1;
While repetition<=
For =1,……,
Choose a guide from the list of rules to attach to
particle i,
Modify this same particle's speed and location;
Get the location and parameters for I-UPQC by
decoding the particle;
Incorporate the I-UPQC concept through power
flow;
End for
Determine the non-dominant solutions;
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
160
Volume 18, 2023
Discover the new set of instructions;
Repetition =repetition+1;
End while
Optimal solutions comprise the ultimate group of
non-dominated options.
location, size, and the parameters for I-UPQC,
End.
Pseudocodes for the scheduling of fictitious power
amelioration utilizing I-UPQC allocation and GA-
IPSO are shown in Figure 3.
2.3 Planning Algorithm
The IPSO comprises two major provision
subprogrammes, i.e., particle encrypting/decrypting
structure and power flow with the I-UPQC model.
Three significant segments consist of 1) I-UPQC
allocation in the system, 2) the quantity of
imaginary power amelioration demand, and 3) 
are parts of GA-IPSO particle coding. The first set
of segments of the particle is always transformed to
its closet digit during decoding. The solution that
violates the last constraint is set aside. The
pseudocodes for the comprehensive arrangement
algorithm are displayed in Figure 3.
Fig. 3: Algorithms flow diagram for GA-IPSO, [21]
3 Network Parameters for the RDN
33-bus and 69-bus
The network of 33-bus standard IEEE test system
implementation framework is presented in this
segment, with 3.69 MW and 2.3 MVAR, 5 ties, and
68 sectionalizing lines that separate the bus. The
power flow in the test system results in power losses
of 210.5 kW, and the primary system voltage of
12.56 KV. This section displays the conventional
IEEE 33-bus test network framework with a 3.72
MW active load and the reactive load of 2.3 MVAR,
5 ties, with 68 sectionalizing lines that divide buses.
The numbers 1-68, and 69-73, correspondingly, are
assigned to the tie lines. The RDN data are
accessible in, [21], [24], [25]. Power losses occur as
a result of the test system's 225 kW power flow, and
the primary network voltage of 12.7 kV.
First scenario: Considering the system
loading of 80% with UPQC position whenever the
parallel inverter supplies imaginary power needed
for the load in an RDN without reconfiguring. In
this scenario, the UPQC inverter operates in-
dependently, with the shunt inverter bearing the
whole load of the imaginary power need of the load
for the duration of disturbances. In contrast, the
series inverter delivers the real power required by
the load during disturbances. This is accomplished
using hybrid GA-IPSO for decision-making at the
best location.
Second scenario: Considering the system
loading of 80% with UPQC position when the series
and parallel inverters combined production of
imaginary powers in RDN while reconfigured.
Hence, the UPQC-recommended PAC is enabled.
The I-UPQC series inverter is made to work with a
shunt inverter to share load-imaginary power while
simultaneously reducing voltage dip/swell. The
imagi-nary load power-sharing between the two
inverters was structured using the PAC control
technique, and the voltage dip/swell was reduced
using the real power control approach. This is
accomplished by the application of hybrid GA-IPSO
for decision-making at the best location.
Third scenario: a continuous condition of 80
% loading considering the series and parallel
inverter with the influence of PAC interlinked and
DG in an RDN is accomplished for PQ without
reconfiguration. In order to make decisions in the
best place on the network, hybrid GA-IPSO is used.
Through a shunt inverter, the PV is incorporated and
plugged by real power injection into the load in the
event of a dip. In a similar vein, the I-UPQC kept
the initial procedure going while increasing the
quantity of imaginary power delivered by the VSI
connected in series and lowering t UPQC's
magnitude for a specific level of dip/surge. By
utilizing a hybrid of GA-IPSO for decision-making,
this is accomplished through the optimal location.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
161
Volume 18, 2023
4 Simulation and Results
The investigation simulations and findings from two
RDN are presented in this part. In this study, the 33-
bus and 67-bus were employed separately along
with the I-UPQC model. All simulation iterations
were run using the certified test RDN data from,
[22], [23], [26], with balance loading as a baseline.
The only swing bus shared by the two networks was
at the substation, and the other bus were load (Q)
buses. The slack bus voltage was given as 1,0 p.u.
per unit. Three examine UPQC modes were tested
under methodologies scenario on three distinct
ratios of the highest voltage produced from the
VSI’s output to the projected load voltage , 20
%, 39.9 %, and 59.9 % in connection to dip/surge.
4.1 Utilizing Convergence Evaluation to
Assess PQ in 33-bus Systems without/with
PAC
This section presents 60 % of the study of GA-IPSO
convergence centered on the best UPQC allocation
in the RDN compared to the other simulations in
three test scenarios. Figure 4 depicts the
convergence investigation with GA-IPSO and RDN,
which considers most cases at 60% disruption. The
convergence provides the best spot for the I-UPQC's
inverters to effectively share imaginary power. The
test system's ability to improve reactive power is
also shown in Figure 4 as a result of a 60 %
injection. It is noted that the chosen approach gives
convergence of 0.151 in second scenario and 0.156
in first scenario at 59.9 % provided, during the 40th
iteration, case 3 converges at 0.148. This section
illustrates the distribution system for three sites
compared to other models.
Fig. 4: Analysis of convergence on 33-bus at a
single I-UPQC location with a 60 % injection
Figure 5 shows the convergent study of the
network for the best placements at 40% injection.
Also, Figure 5 displays a more accurate
convergence assessment provided by GA-IPSO on
the network for the whole assessment cases at 39.9
% penetration for the period of disruption. The FF is
lowered in this case to allow for an increase in
iterations, and it converges at its global minimum.
As demonstrated in Figure 4, because of the 40 %
injection, the network under test’s FF ability also
met the imaginary power amelioration distinct
requirement. In other to fulfill the need for the loads
during sag and swell, the series inverter injects extra
imaginary at bus 6 of the network. The selected
strategy is seen to produce convergence when case 3
converges at 0.148 at iteration 48, case 2 at 0.150 at
iteration 44, and case 1 at 0.155 for 40 % injection.
Fig. 5: Analysis of convergence on 33-bus at a
single I-UPQC location with a 40 % injection
The assessment in Figure 6 shows the results of
the system convergence study for the best
placements at a 20 % injection. The more accurate
convergence analysis provided by GA-IPSO is
displayed in Figure 6 on the system for all test cases
at 20 % of injection during the disruption period. In
this case, the FF is reduced to boost the repetition
amount and comes together at a global minimum.
Due to a 20 % injection, the test system FF also met
the requirements for imaginary power amelioration,
as illustrated in Figure 6. The series inverter is
located at network bus 6.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
162
Volume 18, 2023
Fig. 6: Analysis of convergence on 33 buses at a
single I-UPQC location with a 20 % injection
4.2 Evaluation of PQ Convergence on 69-bus
RDN with/without PAC
This section compares the GA-IPSO and UPQC best
allocation in the RDN to the other models for the
three test scenarios. Figure 7 shows the convergence
study of the GA-IPSO in the RDN for the whole
assessment cases of 59.9 % of injection for the
duration of the disruption. The convergence
provides the best site for the series and shunt
inverters to efficiently share imaginary power.
Similar to Figure 7, the test system FF for imaginary
power amelioration is evident due to 60% injection.
It should be noted that the selected approach yields
convergence of 0.1352 for second scenario and
0.147 for the first scenario of 60 % provided in
RDN over other models for three positions shown in
this segment. Case 3 converges at 0.139 at the 49th
iteration.
Fig. 7: Analysis convergence on 69 buses at a single
I-UPQC location with a 60 % injection
The illustration in Figure 8 depicts the results of
the convergence analysis of the network for the
optimal places at a 40 % injection. Also, Figure 8
displays a superior convergence evaluation provided
by GA-IPSO on the network for the whole
assessment scenario of 39.9 % of injection for the
period of disruption. The FF is reduced in this case
to raise the number of repetitions that converge at
the global minimum. Imaginary power
compensation, which is evident as illustrated in
Figure 6 because of a 40% injection, was also met
by the test system FF. In order to fulfill the need for
the loads during sag and swell, the series inverter
injects extra reactive at bus 6 of the network. The
adopted strategy is seen to produce convergence at
iteration 49th when case 3 converges at 0.148. At
iteration 44, case 2 converges at 0.146, and at
iteration 46, when case 1 at 40% injection converges
at 0.146.
Fig. 8: Analysis convergence on 69 buses at a single
I-UPQC location with a 40 % injection
In Figure 9, the convergence study of the system
is shown for the best spots at a 20 % injection.
Likewise, in Figure 9 presents a more accurate
convergence analysis that GA-IPSO provides on the
network for all test cases at a 25% injection rate for
the period of disruption. Here, the FF is constrained
to converge at the global minimum after an increase
in the number of iterations. Similar to how the 25%
injection met the network under consideration FF,
the reactive power mitigation evidence is presented
in Figure 9. The series inverter injects extra reactive
at bus 6 of the network to fulfill the demand from
the loads at the occurrence of sag and swell. The
accepted strategy is seen to give convergence when
case 3 converges at 0.138 at iteration 60, case 2 at
0.139 at iteration 44, and case 1 at 0.143 at 25
percent addition. Hence, a global minimum
convergence was indicated from the above
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
163
Volume 18, 2023
repetition evaluation for GA-IPSO at 25% superior
to 40%, and 60% .
Fig. 9: Analysis convergence on 69 buses at a single
I-UPQC location with a 20% injection
5 The VA Size of I-UPQC Installed on
Every Bus
As shown in Figure 10 (a) and Figure (b), in that
order, the VA capability of I-UPQC when it is
situated correctly in 33 and 69 bus RDN. The VA
rating of the device installed in the ideal position
consistently demonstrates an improvement above
the results obtained using an analytical technique at
each network bus. Employing GA-IPSO, it was
determined that the busses 17 and 61 of the
networks were the optimum locations for PQ
enhancement in terms of sharing imaginary power
amid voltage dip, surge, and the decrease of power
loss. This is a substantial improvement above the
requirement for the high I-UPQC's capability that is
positioned closer to the substation. The GA-IPSO
integrated the requirement for reactive power
sharing with the predicted greater load current
coupled with a substantial load. Therefore, in these
circumstances, more reactive power amelioration is
required. However, Figure 10 (a) and Figure (b)
show the quantity of VA distributed by VSI linked
in series at different points in the two RDN. Given a
spike in the size of series injected voltage, the
Figures show that VSI linked in series gives better
amelioration.
As seen in Figure 10(a) shows that case 1 and
case 2 have minimum capacities of 0.1 kVA and
0.070 kVA, respectively, due to imaginary capacity
sharing and ideal positioning at bus 17, whereas
case 3 has the lowest capacity on the network, with
I-UPQCpv showing a VA capacity of 0.02 kVA. It
is also observed that Case 3 still has a lower VA
rating despite the DG interconnection, and even
with a 25% dip/surge, the series inverter still
provides imaginary power amelioration. The
situation is the same for the 69-bus network as well.
For cases 3, 2, and 1, 0.50 kVA, 0.70 kVA, and 0.90
kVA of capacity have been attained. In line with the
findings, I-UPQC installed in RDN with a PAC
monitored inverter reduced its capability in bus 61,
and in the third scenario, while the VSI linked in
parallel was connected to PV, it further reduced the
rating to a very low level.
Fig. 10: I-UPQC assignment on buses 33 and 69,
respectively
5.1 Effect of Optimal I-UPQC Allocation on
RDN Loss Minimization
The examples in Figure 11 (a) and Figure (b) show
the appropriate I-UPQC distribution at every node
in the two systems and the energy losses in the bus.
Prior to I-UPQC assignment, the losses generated at
buses 33 and 69 are 202.67 and 224.98 kW,
correspondingly. The outcomes show that nodes 61
and 17 in the 69 and 33 networks system of RDN,
which are the locations where the largest amount of
Loss Reduction (LR) was accomplished, were
equipped with I-UPQC. To acquire the best
imaginary power amelioration, these are the
possible positions for the test systems. The I-UPQC
assignment causes a drop in the network's peak line
current. The highest line current measured in both
systems without using I-UPQC amount to 0.0382
and 0.0391 per unit. For examples 2, and 3, an
actual drop in the highest line current also reduced
power loss. At various levels of , however,
given that the overall imaginary power improvement
is proportionate to the imaginary load power
requirements, which is consistent, there were no
noticeable changes. A similar occurrence happened
in case 3, where the least power dissipated was
observed at 33-bus and 69-bus, individually,
accounting for 128 kW and 124 kW, which together
accounted for 60 % of the total power loss. In test
RDN, bus 5 and bus 61 for 33-bus and 69-bus were
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
164
Volume 18, 2023
the candidate buses where the least power loss was
observed owing to the I-UPQC connection.
Fig. 11: I-UPQC Assignment in Network Test
System of buses 69 and 33 cause power loss
5.2 Impact of Optimal I-UPQC's Placement
on Preventing Low Voltage
In I-UPQC, the shunt and series VSIs shared
imaginary power, which helped to reduce under-
voltage in terms of controlling imaginary power.
The amount of Low Voltage Compensation (LVC)
on bus was used to calculate the influence of I-
UPQC on RDN voltage. Figure 12 (a) and Figure
(b) illustrate this. The placement of I-UPQC at the
RDN network', ideal bus, as shown by the ,
revealed the same proportion of LVC at the same
nodes. In the case of the 69 buses, nine buses, or
around 12.9% of them, where sags issues are noted
without I-UPQC placement, making up around 24
out of the 33 buses, or about 63.63 %. The nodes 17
of 33-bus and 61 of 69-bus systems perform better
in the ideal I-UPQC position for dip-voltage
mitigation, with the highest sag-voltage mitigation
of 72 percent and 75 percent, respectively,
respectively, in both systems at case 3. Additionally,
I-UPQC, which was best positioned at bus 17 in the
33-bus network, gave the best under-voltage
prevention in all  cases, whereas bus 61 in the
69-bus system got results for  cases. According
to the LVC analysis, case 3 showed a higher
proportion of energy generated thanks to the
coupling of the PV to the shunt inverter. In the same
way, example 2 shows a 65 and 70 % voltage dip
mitigation, compared to 60 and 65 % in case 1. As a
result, it can be determined that example 3 exhibited
a superior LVC in Figure 12.
The series VSI can be employed in I-UPQC at
its ideal position for VA capacity reduction, voltage
dip amelioration, and imaginary power adjustment.
As a result of an increase in , VSI linked in
series supplied more amelioration. If I-UPQC is
assigned at a specific location inside the RDN,
substantial enhancements in VP, the power
dissipated, and imaginary power distribution can be
made. The lowest bus voltage from the 33-bus and
69-bus is seen from the simulation model to be
0.9979 p.u., and 0.9931 p.u., correspondingly, by
the placement of I-UPQC among all unique load
buses with many iterations.
Fig. 12: UVMN effect predicated on I-UPQC
deployment at the specific bus
The candidate bus provides the best imaginary
power amelioration. It was also observed that
an increase in  had no effect on any of the
metrics considered in this study, including voltage
profile and power loss. The findings revealed that
the chosen candidate bus, which are nodes with
a heavy load, is where I-UPQC is acquired when a
site satisfies the criterion. In the meantime, The
nodes 68 and 18 of the test system 69 and 33 RDN,
respectively, appeared to be the best places for
hypothetical power distribution and sharing for
reducing voltage dip/surge in Cases 2 and 3 at a
25% disturbance in the system, respectively.
Because the quantity of active power needed is so
little and the shunt inverter readily injects the
necessary amount of active, it appears that several
buses in both the 33 and 69 bus in Case 3 did not
absorb real power through the series inverter. That
is to say, there is zero active power injection in
those that are being monitored.
Fig. 13: 33-bus test voltage profile at a 25%
injection
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
165
Volume 18, 2023
5.3 Voltage Profile at 20% in 33 and 69 RDN
with and without PAC
Figure 13 compares the voltages in cases with and
without a UPQC PAC as well as PV connections
with UPQC under PAC control. As presented in
Table 1, the level of voltage increased to 0.9901
from 0.9463 per unit at node 18 for standard UPQC,
as compared I-UPQCPV while considering node 18,
which increase to 0.9988 from 0.9979 per unit.
Additionally, Figure 14 below provides the outcome
for 69-bus for comparing the situation with and
without a PAC of UPQC, followed by the
connectivity of PV under PAC management. As
shown in Table 2, the voltage level with UPQC
increased to 0.9730 from 0.9203 at node 68, while I-
UPQCPV increased to 0.9920 from 0.9890 per unit
at the same node. When the new device was linked
to the analytical placement through a single VSI
connected in parallel integration, the I-UPQC
efficiency shows that the VP was thus enhanced
further. Thus, the success of I-UPQC instance 3 at
disruption of 25% indicates that the inverter series
provided extra imaginary power, producing a
stronger VP in comparison to a disruption of 39.9 %
and 59.9 %. The quantity of imaginary power
generated in the first and second scenarios at a 20 %
disruption also cancels out first scenario quantity,
placing an imaginary power liability on the shunt
inverter in those cases. The case 3 design with better
VP also yields the position relating to minimum
power dissipation at node 18 for a 33-bus system
and node 68 for a 69-bus network.
Fig. 14: The 69-bus RDN voltage profile at 20%
injection
Fig. 15: At 40% injection, the VP of the 33-bus
RDN
5.4 Voltage Profile at 40 % in 33 and 69 RDN
with and without PAC
The comparison of voltage for the cases with and
without a UPQC PAC is shown in Figure 15 and
Figure 16, followed by the case with PV
connections under PAC control. According to Table
1 and Table 2, the level of voltage increased to
0.9810 from 0.9363 per unit at node 18 for standard
UPQC, as compared I-UPQC while considering the
same node, which increase to 0.9981 from 0.9900
per unit for I-UPQCPV. A PAC with UPQC was
used in the case with and without the connection of
PV under PAC control, and the results for 69-bus
are shown in Figure 16. As indicated in Table 2,
with regular UPQC, the voltage level increased to
0.9710 from 0.9303 per unit at node 17, while I-
UPQC increased to 0.9810 and I-UPQCPV to
0.9730 per unit at the same node. The entire I-
UPQC results show that the VP enhanced more
when the new device was coupled through the VSI
connected in parallel at a single integration with the
analytical positioning. A better VP than 60 %
disruption but less than 20 % disruption VP
alleviation was achieved because of the execution of
I-UPQC case 3 at a 40 % disturbance. Similar to
how case 1 places an imaginary load on the shunt
inverter, the second scenario imaginary power
injection is improved as compared to the first
scenario but significantly lower in third scenario.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
166
Volume 18, 2023
Fig. 16: At 40% injection, the VP of the 69-bus test
system
Fig. 17: At 60% injection, the VP of the 33-bus test
system
5.5 Voltage Profile at 60% in 33 and 69 RDN
with/without PAC
The voltage evaluation for cases with/without a
PAC of UPQC and for the case of PV connection
with UPQC and PAC control is shown in Figure 17
and Figure 18. The voltage on bus 18 was increased
to 0.9780 from 0.9453 p.u when standard UPQC is
used and 0.9850 p.a. when I-UPQC is used on buses
18 and Figure 19, respectively, as shown in Table 1.
Figure 18 shows the results for 69 test systems in
order to compare the scenario with/without PAC
installation to UPQC, with the influence of PAC and
PV connection. As shown in Table 1, the voltage
level increased to 0.9599 to 0.9303 per unit at node
68 for regular UPQC, while with I-UPQC, it
increased to 0.9720, and with I-UPQCPV to 0.9705
per unit at the same 68 node. According to the
overall I-UPQC function, the VP enhanced more
when the latest device was linked to the VSI
coupled in parallel at a single integration with the
optimal location. The overall I-UPQC result proves
the VP enhanced more when the new device was
linked single connection with the analytical
positioning through the shunt inverter. But when the
I-UPQC was linked to the PV across the VSI
connected in parallel, the results were even better.
6 Comparison of Findings
Subsequently, the aforesaid findings, a
comprehensive and accurate examination of the
effects of I-UPQC in taking into account VP
enhancement and loss minimization by correlating
the base case, first, second and third scenarios
become necessary. Though, UPQC, I-UPQC, and I-
UPQCPV are crucial points of evaluation in Figure
19 and Figure 19, Table 1 and Table 2 provide a
summary of this influence for both 33-bus and 69-
bus. Figure 18 and Figure 19 depict the hypothetical
power distribution between the two inverters in
series and shunt configuration in a similar manner.
Fig. 18: Power dissipation of 33-bus RDN in all
cases
6.1 Comparison of Power Dissipated in the
33-bus with the best I-UPQC Allocation
As indicated in Table 1, the impact of I-UPQC
outcomes over conventional UPQC with-out PAC
are compared in 33-bus RDN overall. The I-UPQC
demonstrates that bus 6 of 33-bus has a series
inverter that provides extra imaginary power in
PAC-controlled models. When examining the
outcomes from various scenarios, it is found that a
25% disturbance in RDN with I-UPQC gives greater
power dissipation minimization. Because of the
ability of the parallel inverter to supply the real
power demanded by the load in the occurrence of a
dip and absorbed imaginary power in the event of a
surge, the network connected to PV via I-UPQC
exhibits a greater decrease in comparison. Figure 19
depicts the percentage power loss decrease in each
example, and I-UPQC was found to have the lowest
proportion of 15.21 %. According to the findings,
case 3 PV connections through the shunt inverter
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
167
Volume 18, 2023
can significantly reduce loss if it is properly
constructed and regulated at the proper position.
Table 1. Comparing the I-UPQC model to the
standard UPQC in the 33-test system
Scenarios
UPQC
Position
Name
Base Case
No UPQC
Lowest VP (p.u.)
Dissipated Power
(kW)
Lowest VSI
First
scenario:
UPQC
Bus 18
Lowest VP (p.u.)
Dissipated Power
(kW)
% LR (kW)
Lowest VSI
Shunt injected
(kVar)
Second
scenario: I-
UPQC
Bus 18
Lowest VP (p.u.)
Dissipated Power
(kW)
% Loss Decrease
(kW)
Lowest VSI
Series injected
(kVar)
Shunt injected
(kVar)
Third
scenario: I-
UPQC
Bus 18
Lowest VP (p.u.)
with PV
Dissipated Power
(kW)
% LR (kW)
Lowest VSI
Series injected
(kVar)
Shunt injected
(kVar)
6.2 Comparison of 33-bus Imaginary Power
Sharing in Cases with and without PAC
Figure 19 illustrates the injection of imaginary
power between the two inverters in the 33-bus RDN
for the period of imaginary power improvement in
the case of dip reduction at 25%. According to the
UPQC in the RDN, case 1, a complete imaginary
load was carried by the shunt inverter and was 0.202
kVar, whereas the series provided all real power.
However, whereas the shunt inverter injects 0.101
kVar, the series inverter only supplies 0.051 kVar.
Case 3 ultimately performs better due to its
involvement in imaginary power amelioration
during the disruption when the I-UPQC device is
triggered. In the condition of integrated PV through
the shunt inverter, the series VSI provides 0.1499
kVar, and the parallel VSI produces 0.1011 kVar in
node 6. Due to PAC control, it has been indicated
that I-UPQC outperforms in terms of imaginary
power improvement.
Fig. 19: Imaginary power amelioration for
parallel/series inverters in 33-test system
6.3 Evaluation of Power Dissipation on 69-
Test System RDN with/without PAC
Table 2 shows the total evaluation of the effect of I-
UPQC findings above regular UPQC without PAC
in three scenarios for the 69-test system RDN. At
node 61 of 69 test system, the I-UPQC demonstrates
that series inverters provided so much imaginary
power in PAC regulated models. Under a
disturbance of 25%, it is found that radial networks
with I-UPQC offer a greater decrease in power loss
when comparing the findings from various
scenarios. The network connected to PV through I-
UPQC exhibits a higher decrease in comparison due
to the parallel inverter's ability to provide the real
power that the consumer needs in the occurrence of
undervoltage and absorbed imaginary power in the
event of a surge. In the 69-test system, I-UPQC had
the lowest percentage, or 18.50%, where the
proportion of power dissipation minimization for
each example is shown in Figure 20. According to
the findings in case 3, PV connections through the
shunt inverter can significantly reduce loss if it is
properly constructed and regulated at the right
position.
Fig. 20: Loss of power in all 69-bus network test
instances
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
168
Volume 18, 2023
Table 2. Comparing the I-UPQC model to the
standard UPQC in the 69-test system
Scenarios
UPQC
Position
Name
Peak
load
Base Case
No UPQC
Lowest VP per
unit
0.9203
Dissipated Power
(kW)
51.59
Lowest VSI
0.8369
First
scenario:
UPQC
Bus 68
Lowest VP per
unit
0.9720
Dissipated Power
(kW)
23.55
% LR (kW)
40.35
Lowest VSI
0.9041
Shunt injected
(kVar)
0.269
Second
scenario: I-
UPQC
Bus 68
Lowest VP per
unit
0.9870
Dissipated Power
(kW)
35.14
% LR (kW)
27.4
Lowest VSI
0.9053
Size of UPQC in
(kVar)
0.61
Series injected
(kVar)
0.133
Shunt injected
(kVar)
0.126
Third
scenario: I-
UPQC
Bus 68
Lowest VP per
unit
0.9930
with PV
Dissipated Power
(kW)
14.79
% LR (kW)
18.5
Lowest VSI
0.9362
Size of UPQC in
(kVar)
0.41
Series injected
(kVar)
0.149
Shunt injected
(kVar)
0.131
6.4 Comparison of 69-bus Imaginary Power
Sharing with and without PAC
Figure 21 shows how the two inverters in the 69-test
system RDN share imaginary power when in an
imagined power improvement in the case of dip
reduction at 24.9%. The shunt inverter assumed the
responsibility for all imaginary power in scenario 1
of 0.269 kVar, according to the UPQC in the RDN.
Hence, the series provided most of the real power.
However, the shunt inverter only injects 67.5
reactive power, whereas the series inverter produces
84.5 reactive power. Case 3 ultimately performs
better by taking part in imaginary power
amelioration during the disruption when I-UPQC
was in use. When connected PV via the parallel
inverter, the series inverter offers 183.6 reactive
power, whereas the parallel inverter gives 135.6
reactive power on the node 6. It has been
demonstrated that I-UPQC outperforms PAC in
aspects of compensating for imaginary power.
Fig. 21: Compensation of imaginary power in 69-
test system among both parallel/series inverter
7 Conclusion
A specific quantity of under-voltage in the RDN
was effectively reduced by using a spe-cial PAC of
the UPQC configuration known as I-UPQC. For the
study, the I-UPQC model was best included using a
load-flow method with GA-IPSO. The results of this
investigation led to the following conclusion. I-
UPQC located at an optimal node can reduce VA
capacity while reducing imaginary power
consumption. I-UPQC installed at a particular RDN
node significantly reduces power loss. Additionally,
certain nodes achieve better VP enhancement at
sending and receiving end of the RDN. However,
the results show that a candidate bus must be
situated optimally with a greater capacity I-UPQC
to perform at its best. As a result, the performance
of I-UPQC in RDN depends on the position. It also
demonstrates higher power reduction with an ideal
method than analytical placements and improves
under-voltage. The voltage sags, VP variations, and
actual power loss of the tested systems were all
adequately managed with the help of the research
methods and outcomes. It is evident that I-UPQC
showcases the proper sharing of imaginary power
while optimal placement enhances its operation in
RDN. The outcomes showed that, at a preset degree
of disturbance in a particular area, I-UPQC with a
lower capacity could mitigate sag/swell. I-UPQC
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
169
Volume 18, 2023
coupled to PV using the same control strategy
achieves a better improvement with active power
absorbed via a parallel inverter. Although the
assessment of optimal improvement strategy was
effectively conducted, the dynamic behavior was
not, and this can be a topic for more investigation.
References:
[1] D. Ćetković, G. Klobučar, and V. Komen,
“Analysis of the Impact of Reactive Power
Compensation on the Electric Power Quality
of the LV Consumers in Distribution
Networks of Various Characteristics,” in 2022
45th Jubilee International Convention on
Information, Communication and Electronic
Technology (MIPRO), 2022, pp. 796799.
[2] G. Chen, J. Zhang, and M. Taleb Ziabari,
“Optimal allocation of capacitor banks in
distribution systems using particle swarm
optimization algorithm with time-varying
acceleration coefficients in the presence of
voltage-dependent loads,” Aust. J. Electr.
Electron. Eng., vol. 19, no. 1, pp. 87100,
2022.
[3] O. O. Oluwole, “Optimal allocation of
distributed generation for power loss
reduction and voltage profile improvement,”
University of Cape Town, 2016.
[4] Y. Maataoui, H. Chekenbah, O. Boutfarjoute,
and R. Lasri, “Voltage control using fuzzy
logic for radial distribution network with high
penetration of photovoltaic generators,” in
E3S Web of Conferences, 2022, vol. 351, p.
1030.
[5] H. Lotfi and A. A. Shojaei, “A dynamic
model for multi-objective feeder
reconfiguration in distribution network
considering demand response program,”
Energy Syst., pp. 130, 2022.
[6] H. B. Tolabi, M. H. Ali, and M. Rizwan,
“Simultaneous reconfiguration, optimal
placement of DSTATCOM, and photovoltaic
array in a distribution system based on fuzzy-
ACO approach,” IEEE Trans. Sustain.
Energy, vol. 6, no. 1, pp. 210218, 2014.
[7] A. Gupta, “Power quality evaluation of
photovoltaic grid interfaced cascaded H-
bridge nine-level multilevel inverter systems
using D-STATCOM and UPQC,” Energy,
vol. 238, p. 121707, 2022.
[8] R. K. Chauhan, J. P. Pandey, and M. Hasan,
“Comparative performance of the various
control techniques to mitigate the power
quality events using UPQC,” in Advanced
Computational and Communication
Paradigms, Springer, 2018, pp. 3140.
[9] P. K. Ray, P. S. Puhan, A. K. Das, D.
Pradhan, and L. Meher, “Comparative
Analysis of Different Control Techniques
Implementation in UPQC for Power Quality
Improvement,” in Sustainable Energy and
Technological Advancements, Springer, 2022,
pp. 147161.
[10] A. Patel, S. K. Yadav, and H. D. Mathur,
“Utilizing UPQC-DG to export reactive
power to grid with power angle control
method,” Electr. Power Syst. Res., vol. 209, p.
107944, 2022.
[11] O. O. Osaloni and A. K. Saha, “Voltage
Dip/Swell Mitigation and Imaginary Power
Compensation in Low Voltage Distribution
Utilizing Improved Unified Power Quality
Conditioner (I-UPQC),” in International
Journal of Engineering Research in Africa,
2020, vol. 49, pp. 84103.
[12] O. O. Osaloni and A. K. Saha, “Impact of
Improved Unified Power Quality Conditioner
Allocation in Radial Distribution Network,”
Int. J. Eng. Res. Africa, vol. 59, pp. 135150,
2022.
[13] A. Patel, H. D. Mathur, and S. Bhanot,
“Improving Performance of UPQC-DG for
Compensation of Unbalanced Loads,” in 2018
8th IEEE India International Conference on
Power Electronics (IICPE), 2018, pp. 16.
[14] J. H. Yi, R. Cherkaoui, M. Paolone, D.
Shchetinin, and K. Knezovic, “Optimal Co-
Planning of ESSs and Line Reinforcement
Considering the Dispatchability of Active
Distribution Networks,” IEEE Trans. Power
Syst., 2022.
[15] A. Parizad, H. R. Baghaee, A. Yazdani, and
G. B. Gharehpetian, “Optimal distribution
systems reconfiguration for short circuit level
reduction using PSO algorithm,” in 2018
IEEE Power and Energy Conference at
Illinois (PECI), 2018, pp. 16.
[16] P. Ramsami and R. T. F. A. King, “Multi-
Objective Optimisation of Photovoltaic
Systems and Unified Power Quality
Conditioners with Simultaneous Distribution
Network Reconfiguration,” in 2020 3rd
International Conference on Emerging Trends
in Electrical, Electronic and Communications
Engineering (ELECOM), 2020, pp. 191197.
[17] T. Guan, F. Han, and H. Han, “A modified
multi-objective particle swarm optimization
based on levy flight and double-archive
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
170
Volume 18, 2023
mechanism,” IEEE Access, vol. 7, pp.
183444183467, 2019.
[18] O. Osaloni and K. Awodele, “Analytical
Approach for Optimal Distributed Generation
Allocation in Primary Distribution Networks,”
in Proc. 2016 South African Universities
Power Engineering Conference, pp. 115120.
[19] P. Jangir and I. N. Trivedi, “Non-dominated
sorting moth flame optimizer: A novel multi-
objective optimization algorithm for solving
engineering design problems,” Eng. Technol.
Open Access J., vol. 2, no. 1, pp. 1731, 2018.
[20] G. Carpinelli, F. Mottola, D. Proto, and A.
Russo, “A Decision Theory Approach for the
Multi-objective Optimal Allocation of Active
Filters in Smart Grids,” in 2022 20th
International Conference on Harmonics \&
Quality of Power (ICHQP), 2022, pp. 16.
[21] O. O. Osaloni, “Power quality improvement
in low voltage distribution network utilizing
improved unified power quality conditioner.,”
2020.
[22] C. U. Eya, A. O. Salau, S. L. Braide, S. B.
Goyal, V. A. Owoeye, and O. O. Osaloni,
“Assessment of Total Harmonic Distortion in
Buck-Boost DC-AC Converters using
Triangular Wave and Saw-Tooth based
Unipolar Modulation Schemes”, WSEAS
Transactions on Power Systems, vol. 17, pp.
324-338, 2022.
[23] O. O. Osaloni, A. S. Akinyemi, A. A.
Adebiyi, A. O. Salau, "An Effective Control
Technique to Implement an IUPQC Design
for Sensitive Loads in a Hybrid Solar PV-Grid
Connection", WSEAS Transactions on Power
Systems, vol. 18, pp. 26-38, 2023.
[24] O. O. Osaloni, and A. K. Saha. Distributed
Generation Interconnection with Improved
Unified Power Quality Conditioner for Power
Quality Mitigation,” 2020 International
SAUPEC/RobMech/PRASA Conference, 1-6.
[25] H. Kassahun, A. O. Salau, O. O. Osaloni, and
O. Olaluyi: “Power System Small Signal
Stability Enhancement Using Fuzzy Based
STATCOM,” PRZEGLĄD
ELEKTROTECHNICZNY 8, 27-32, 2023.
[26] O. O. Osaloni, A. S. AKINYEMI, A. A.
Adebiyi, T. O. IBITOYE Power Loss
Analysis with Dispersed Generation in
Multifunction Power Conditioner Design to
Improve Power Quality WSEAS
Transactions on Power Systems, vol. 18, pp.
94-103, 2023.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The authors equally contributed in the present
research, at all stages from the formulation of the
problem to the final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
I want to use opportunity to appreciate Durban
University of Technology and Afe Babalola
University for the success of this research work.
Conflict of Interest
The authors have no conflicts of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2023.18.17
Oluwafunso Oluwole Osaloni, Ayodeji Stephen Akinyemi,
Abayomi Aduragba Adebiyi,
Katleho Moloi, Ayodeji Olalekan Salau
E-ISSN: 2224-350X
171
Volume 18, 2023