Simultaneous network reconfiguration and capacitor allocations using a novel dingo optimization algorithm

International Journal of Electrical and Computer Engineering (IJECE)
Vol. 13, No. 3, June 2023, pp. 2384~2395
ISSN: 2088-8708, DOI: 10.11591/ijece.v13i3.pp2384-2395  2384
Journal homepage: http://ijece.iaescore.com
Simultaneous network reconfiguration and capacitor allocations
using a novel dingo optimization algorithm
Samson Oladayo Ayanlade1
, Abdulrasaq Jimoh2
, Emmanuel Idowu Ogunwole3
, Abdullahi Aremu4
,
Abdulsamad Bolakale Jimoh5
, Dolapo Eniola Owolabi6
1
Department of Electrical and Electronic Engineering, Faculty of Engineering and Technology, Lead City University, Ibadan, Nigeria
2
Department of Electronic and Electrical Engineering, Faculty of Technology, Obafemi Awolowo University, Ile-Ife, Nigeria
3
Department of Electrical, Electronic and Coputer Engineering, Faculty of Engineering, Cape Peninsula University of Technology,
Cape Town, South Africa
4
Department of Technical Operation, Ilesha Business Hub, Ibadan Electricity Distribution Company, Ilesha, Nigeria
5
Department of Electrical and Electronic Engineering, Faculty of Engineering and Technology, University of Ilorin, Ilorin, Nigeria
6
Department of Electronic and Electrical Engineering, Faculty of Engineering and Technology, Ladoke Akintola
University of Technology, Ogbomoso, Nigeria
Article Info ABSTRACT
Article history:
Received Aug 30, 2022
Revised Sep 8, 2022
Accepted Oct 1, 2022
Power loss and voltage magnitude fluctuations are two major issues in
distribution networks that have drawn a lot of attention. Combining two of
the numerous strategies for solving these problems and dealing with them
simultaneously to get more effective outcomes is essential. Therefore, this
study hybridizes the network reconfiguration and capacitor allocation
strategies, proposing a novel dingo optimization algorithm (DOA) to solve
the optimization problems. The optimization problems for simultaneous
network reconfiguration and capacitor allocations were formulated and
solved using a novel DOA. To demonstrate its effectiveness, DOA’s results
were contrasted with those of the other optimization techniques. The
methodology was validated on the IEEE 33-bus network and implemented in
the MATLAB program. The results demonstrated that the best network
reconfiguration was accomplished with switches 7, 11, 17, 27, and 34 open,
and buses 8, 29, and 30 were the best places for capacitors with ideal sizes of
512, 714, and 495 kVAr, respectively. The network voltage profile was
significantly improved as the least voltage at bus 18 was increased to
0.9530 p.u. Furthermore, the overall real power loss was significantly
mitigated by 48.87%, which, when compared to the results of other methods,
was superior.
Keywords:
Capacitor allocation
Dingo optimization
Network reconfiguration
Power loss
Voltage magnitude
This is an open access article under the CC BY-SA license.
Corresponding Author:
Samson Oladayo Ayanlade
Department of Electrical and Electronic Engineering, Faculty of Engineering and Technology, Lead City
University
Toll Gate Area, Lagos/Ibadan Express Way, Ibadan, Nigeria
Email: samson.ayanlade@lcu.edu.ng; ayanladeoladayo@gmail.com
1. INTRODUCTION
The distribution network is a subsystem of the power system that is responsible for power delivery
to the end users of electricity [1]–[3]. Therefore, the power quality delivered to customers strongly depends
on it [4]. The two basic topologies available for the distribution network are radial and ring structures [5]. In
a radial configuration, feeders radiating from the injection substation supply the loads. The key benefit of this
network design is how inexpensively it can be built and maintained [6]. However, its biggest drawback is the
possibility of a fault occurring in the center of the network, which might result in the loss of power to the

Int J Elec & Comp Eng ISSN: 2088-8708 
Simultaneous network reconfiguration and capacitor allocations using … (Samson Oladayo Ayanlade)
2385
loads located far from the feeder. Furthermore, the bus voltage magnitude decreases progressively as the
loads are located away from the feeders. For ring distribution networks, the feeders continuously supply
power to the loads in a ring fashion such that if there is a fault in any part of the network, the power supplied
to the loads connected to the healthy sections of the network would not be interrupted. The primary downside
of this design is its high cost [7]. The radial layout is preferred by power engineers because it is inexpensive
to build and maintain. Furthermore, reconfiguring radial distribution networks is a simple process.
Switches used in network reconfiguration fall into two categories: tie switches and sectionalizing
switches [8]–[10]. Sectionalizing switches are normally closed, whereas tie switches are normally open.
Changing the status of these switches causes the network to be reconfigured. Therefore, network
reconfiguration is the act of altering the state of sectionalizing and tie switches to adjust power flow to the
loads to enhance the operational performance of the distribution networks. Various network reconfiguration
setups are feasible [11]. However, network reconfiguration should be done in such a way that the network
stays radial and no loads are cut-off from the supply after reconfiguration. One of these possible alternative
configurations provides significantly lower losses and an improved voltage profile. The network
reconfiguration method is inexpensive to deploy. Many optimization strategies were used to tackle the
network reconfiguration problem.
For instance, to tackle a network reconfiguration problem to lower distribution system power loss,
Khetrapal [12] introduced an improved harmony search algorithm (IHSA) inspired by a musician's
performance. Salkuti [13] employed a crow search method to optimize network reconfiguration while
keeping overall operational cost and network power loss as objectives. The problem of distribution network
reconfiguration was addressed in [14] by utilizing an enhanced selective binary particle swarm optimization
(IS-BPSO) method. Salau et al. [11] provided an excellent technique for resolving network reconfiguration
issues in a distribution network to mitigate losses and boost voltage profile. A new approach was also offered
to tackle this problem and offer an effective network. With diverse loading circumstances taken into account,
a modified selective particle swarm optimization approach was employed for network reconfiguration of the
network.
Capacitor deployment is another economical method of decreasing power losses [15]. Aside from
their low cost, capacitor placement helps to mitigate power loss and boost the voltage magnitude of the
networks. To obtain the greatest advantage from capacitor placements in distribution networks, however,
they must be appropriately deployed (i.e., optimally sized and placed). The literature has offered several
strategies for optimizing capacitor allocations in distribution networks. A flower pollination algorithm (FPA)
was suggested in [16] for the allocation of capacitors in various systems. A shunt capacitor was sized and
placed using a whale optimization approach in [17], [18]. A mine blast algorithm (MBA) was used in [19] to
obtain the optimal locations and capacities of capacitors in various distribution networks. The utilization of
the grey wolf, dragonfly, and moth-flame optimization approaches for the ideal capacitor placements in a
variety of distribution networks was given in [20]. Addisu et al. [21] proposed a fuzzy logic optimization
strategy for efficient voltage regulator positioning and capacitors in distribution systems. The method was
tested on the practical Ethiopian Gondar power distribution system, which has 60 nodes. These two strategies
(i.e., network reconfiguration and capacitor placements) are potent in improving the performance of radial
distribution networks while being relatively affordable to implement. Hence, hybridizing the two methods
and solving them simultaneously would be the most effective and beneficial.
There are very few works in the literature that employ simultaneous network reconfiguration and
capacitor placements to mitigate losses and boost the distribution network voltage magnitudes. The expenses
of real power losses and shunt capacitor installation, as well as enhancing the harmonic state of the network,
were modeled as multi-objective problems in [22]. A fuzzy harmony search technique was devised to find the
best solution point for multi-objective problems. The presented model was tested on two common
distribution systems: the IEEE 33-bus standard system and Taiwan Power Company's 83-bus distribution
network. While operational and power quality restrictions were present, Esmaeilian and Fadaeinedjad [23]
employed simultaneous reconfiguration and capacitor installation to decrease power loss and increase system
dependability. Because the optimization problem was discrete and non-linear, a binary gravitational search
algorithm was used to efficiently tackle the fuzzy multi-objective problems. The quick harmonic analysis
approach was adopted to execute harmonic power flow in the presence of capacitors and non-linear loads. To
evaluate the dependability of various system configurations, the state enumeration approach which depends
on the Weibull-Markov stochastic model was utilized. Furthermore, a novel encoding approach was
presented to improve the network reconfiguration procedure's performance. To test and validate the
suggested technique, the IEEE 33 and 83-bus system of Taiwan Power Company with a variety of harmonic
producing loads were used.
Namachivayam et al. [24] suggested a combined technique for network reconfiguration and
appropriate placement of capacitor banks in radial distribution networks to decrease real power loss and
increase bus voltages. Prior to the optimization procedure, suitable tie-switch combinations were constructed

 ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 13, No. 3, June 2023: 2384-2395
2386
using a graph theory-based technique to maintain radial organization and prevent node islanding. The
optimization issue was addressed with the help of a modified flower pollination algorithm and a dynamic
switching probability technique. The approach's performance was evaluated utilizing conventional 33-bus,
69-bus, and 118-bus distribution networks.
Babu et al. [25] presented simultaneous network reconfiguration and capacitor deployment in the
distribution system to reduce losses, operational costs, and enhance voltages. During network
reconfiguration, the Johnson's technique was employed to determine the smallest spanning tree, and an
adaptive whale optimization algorithm was utilized to address the problem. On IEEE 33-bus and 69-bus
networks, the suggested technique was validated. Sedighizadeh et al. [26] suggested a strategy and a
optimization technique for minimizing loss in distribution systems by simultaneous network reconfiguration
and capacitor allocation using improved binary PSO. The status of the network switches and capacitors was
represented by binary strings in the model. The technique was deployed and evaluated on IEEE 16-bus and
33-bus networks to determine the best network design in terms of losses. Using Johnson's and modified
Whale's algorithms, Anitha [27] performed simultaneous reconfiguration and capacitor allocation in
distribution networks to mitigate loss and operational cost. The proposed technique was tested on the IEEE
33-bus and 69-bus systems.
The optimization methods utilized to tackle simultaneous network reconfiguration and capacitor
allocations in the literature reviewed so far are usually trapped in local minimum points, which affects their
efficacy. Besides, better results could be obtained by using a powerful novel optimization technique.
Therefore, this paper proposes simultaneous network reconfiguration and capacitor allocation using a novel
dingo optimization algorithm (DOA) proposed by [28] to tackle the optimization problems, which is the
novelty of this research. The proposed technique was coded in MATLAB and tested on the IEEE 33-bus
network. This study's primary contribution is the application of a powerful novel optimization technique to
optimally address the simultaneous network reconfiguration and capacitor allocation problems efficiently.
The rest of the paper is structured as follows: The second section describes the methodology utilized in this
study. The simulation results are presented in the third section, and the study is concluded in the fourth
section.
2. METHOD
2.1. Objective function
The objective of network reconfiguration and capacitor allocations is to minimize overall real power
loss. As a result, this is regarded as the research's objective function. The overall network real power loss was
calculated as the summation of the losses in the line segments.
𝑂𝐹𝑚𝑖𝑛 = ∑ |𝐼𝑖|2
𝑅𝑖
𝑛𝑏
𝑖 (1)
where, nb: total number of branches, Ri=ith
branch resistance and |Ii|=ith
branch current magnitude.
2.2. Constraints
The objective function is restricted by some constraints. These constraints are divided into two
categories: equality constraints and inequality constraints. The mathematical expressions of these constraints
are given in the following sub-sections.
2.2.1. Power flow equations
During the optimization process, the power flow problems were solved utilizing the Newton-
Raphson approach. These equations are as (2) and (3):
𝑃𝑔𝑖 = 𝑃𝐷𝑖 + ∑ |𝑉𝑖|
𝑛𝑏
𝑗=1 |𝑉
𝑗|[𝐺𝑖𝑗 𝑐𝑜𝑠 𝜃𝑖𝑗 + 𝐵𝑖𝑗 𝑠𝑖𝑛 𝜃𝑖𝑗] (2)
𝑄𝑔𝑖 = 𝑄𝐷𝑖 + ∑ |𝑉𝑖|
𝑛𝑏
𝑗=1 |𝑉
𝑗|[𝐺𝑖𝑗 𝑠𝑖𝑛 𝜃𝑖𝑗 − 𝐵𝑖𝑗 𝑐𝑜𝑠 𝜃𝑖𝑗] (3)
where Vi, Vj: bus voltages at buses i and j; Pgi and PDi: active power generated and power demanded at bus i;
Qgi and QDi: reactive power generated and demanded at bus i; and θij: voltage angle between buses i and j.
2.2.2. Voltage constraints
The voltage magnitudes must be within permissible limits for the distribution network.

2387
𝑉𝑚𝑖𝑛 ≤ 𝑉𝑖 ≤ 𝑉
𝑚𝑎𝑥 (4)
where Vmin and Vmax: least and maximum voltages (0.95 and 1.05 p.u.), and Vi: voltage magnitudes.
2.2.3. Reactive power constraint on the capacitors
The size of each of the installed shunt capacitors is constrained within the limits given by (5).
𝑄𝑆𝐶(𝑚𝑖𝑛) ≤ 𝑄𝑆𝐶 ≤ 𝑄𝑆𝐶(𝑚𝑎𝑥) (5)
where QSC(min)=100 kVAr and QSC(max) is 75% of the overall network reactive power demand [29].
2.2.4. Radial topology constraint
Without meshes, the distribution scheme should be radial. All loads are typically serviced without
interruption. The overall number of main loops created once all ties are closed is given by (6).
𝑁main loops = (𝑁𝑏 − 𝑁𝑅) + 1 (6)
The total number of sectionalizing switches
𝑁𝑅 = 𝑁𝑏 − 1 (7)
where Nb: total bus number, NR: total branch number.
A radial configuration check was carried out before starting the load flow and at different points
during the optimization process of the proposed method to ensure that the developed solutions satisfy the
radial configuration criteria. Each configuration involves the calculation of the incidence matrix C. After that,
the first column related to the slack bus is deleted, leaving a square matrix C. If the configuration is radial,
the square matrix C's determinant is equal to 1 or -1; if not, the configuration is non radial [30].
2.3. Dingo optimizer
In 2021, Bairwa [28] suggested the DOA, taking cues from the social organization and expert
collective hunting behavior of dingoes. They live in groups of 12-15 individuals, are intelligent, and have
effective communication abilities. Their social structure is well organized. The strongest member of the
group, or the alpha, is in charge of making decisions that will have an impact on every other member of the
group. Beta dingoes serve as the group's second in command, enforce group rules, and serve as a liaison
between the alpha and the other dingoes. The alphas and betas are assisted in their quest for prey and food for
the pack by all the other dingoes.
2.3.1. Encircling
Dingoes are naturally good at finding their prey. The pack of dingoes surrounds the prey after
spotting its location. The mathematical model for this behavior is (8)–(12).
𝐷𝑑
⃗⃗⃗⃗ = |𝐴 ⋅ 𝑃
𝑝
⃗⃗⃗ (𝑥) − 𝑃
⃗ (𝑖)| (8)
𝑃
⃗ (𝑖 + 1) = 𝑃
⃗𝑝(𝑖) − 𝐵
⃗ ⋅ 𝐷
⃗
⃗ (𝑑) (9)
𝐴 = 2 ⋅ 𝑎1 (10)
𝐵
⃗ = 2𝑏
⃗ ⋅ 𝑎2 − 𝑏
⃗ (11)
𝑏
⃗ = 3 − (𝐼 ∗ (
3
𝐼𝑚𝑎𝑥
)) (12)
where 𝐷
⃗
⃗ 𝑎: the distance between the dingo and prey; 𝑃
⃗𝑝: prey’s position vector; 𝑃
⃗ : dingo’s position vector; 𝐴
and 𝐵
⃗ : coefficient vectors; 𝑎1 and 𝑎2: random vector in [0, 1], 𝑏
⃗ linearly decreases from 3 to 0 at each
iteration. Dingoes change their location within the search space around the prey at random.

 ISSN: 2088-8708
2388
2.3.2. Hunting
The mathematical modeling of dingoes assumes that the pack is aware of potential locations for
prey. The beta dingo occasionally assists the alpha dingo in the hunting process. Using (13)–(18), the
dingoes' where abouts are updated [28].
𝐷
⃗
⃗ 𝛼 = |𝐴1 ⋅ 𝑃
⃗𝛼 − 𝑃
⃗ | (13)
𝐷
⃗
⃗ 𝛽 = |𝐴2 ⋅ 𝑃
⃗𝛽 − 𝑃
⃗ | (14)
𝐷
⃗
⃗ 𝑜 = |𝐴3 ⋅ 𝑃
⃗𝑜 − 𝑃
⃗ | (15)
𝑃
⃗1 = |𝑃𝛼
⃗⃗⃗⃗ − 𝐵
⃗ ⋅ 𝐷
⃗
⃗ 𝛼| (16)
𝑃
⃗2 = |𝑃𝛽
⃗⃗⃗⃗ − 𝐵
⃗ ⋅ 𝐷
⃗
⃗𝛽| (17)
𝑃
⃗3 = |𝑃𝑜
⃗⃗⃗ − 𝐵
⃗ ⋅ 𝐷
⃗
⃗ 𝑜| (18)
The (19) to (21) are used to estimate the intensity of each dingo.
𝐼𝛼 = 𝑙𝑜𝑔 (
1
𝐹𝛼−(1𝐸−100)
+ 1) (19)
𝐼𝛽 = 𝑙𝑜𝑔 (
1
𝐹𝛽−(1𝐸−100)
+ 1) (20)
𝐼𝑜 = 𝑙𝑜𝑔 (
1
𝐹𝑜−(1𝐸−100)
+ 1) (21)
2.3.3. Attacking prey
A dingo assault on the prey has occurred when there is no update. The value of 𝑏
⃗ is decreased
linearly to model this strategy. It should be noted that the change range of 𝐷
⃗
⃗ 𝛼 is likewise reduced by
𝑏
⃗ . 𝐷
⃗
⃗ 𝛼 is a random variable in the [-3b, 3b] interval where 𝑏
⃗ is lowered from 3 to 0 between iterations. When
𝐷
⃗
⃗ 𝛼 has random values between [1, 1], a search agent's next position could be anywhere between its current
and the prey's.
2.3.4. Searching
Dingoes constantly move forward to pursue and pounce on their prey [28]. The dingo is retreating
from the prey when 𝐵
⃗ is less than 1, and toward the prey when it is greater than 1. Anytime the conditions for
termination are satisfied, DOA terminates itself.
2.4. Application of dingo optimization algorithm
The DOA approach was used to solve the problems with capacitor allocation and network
reconfiguration as:
Step 1: Input the network line and load data including the tie switches, and DOA data.
Step 2: Initialization.
A dingo is a hypothetical solution consisting of radial arrangement, capacitor placements, and sizes
when using the DOA approach. A swarm of n dingoes is denoted by
𝐷 = [
𝐷1
𝐷2
⋮
𝐷𝑛
] =
[
𝑇𝑆1
1
, … , 𝑇𝑆𝑁𝑇𝐿
1
𝑏𝑢𝑠_𝐶𝑎𝑝1
1
, … , 𝑏𝑢𝑠_𝐶𝑎𝑝𝑚
1 𝑠𝑖𝑧𝑒_𝐶𝑎𝑝1
1
, … , 𝑠𝑖𝑧𝑒_𝐶𝑎𝑝𝑚
1
𝑇𝑆1
2
2
2
2 𝑠𝑖𝑧𝑒_𝐶𝑎𝑝1
2
2
⋮⋱⋱⋱⋱⋱⋮ ⋮⋱⋱⋱⋱⋱⋮ ⋮⋱⋱⋱⋱⋱⋮
𝑇𝑆1
𝑛
𝑛
𝑛
𝑛 𝑠𝑖𝑧𝑒_𝐶𝑎𝑝1
𝑛
𝑛 ]
(22)
A dingo in the population may be described by (23).
𝐷𝑖 = [𝑇𝑆1
𝑖
𝑖
𝑖
𝑖 𝑠𝑖𝑧𝑒_𝐶𝑎𝑝1
𝑖
𝑖 ] (23)
In (23) demonstrates that each dingo's solution vector is made up of three parts. The first component reflects
the quantity of network tie switches (open branches); the second component, the number of buses selected for
the placement of capacitors; and the third component, the number of capacitor capacities.

2389
In the equations, TS1, TS2, ..., TSNTL are the tie switches in the fundamental loops; bus_Cap1,
bus_Cap2, ..., bus_Capm are the buses chosen for the placement of capacitors; size_Cap1, size_Cap2, ...,
size_Capm are the sizes of the capacitors in kVAr to be installed on the buses respectively.
Each dingo in the DOA may be seen as a solution that is generated at random at initialization. As a
consequence, each dingo, Di in the population has the following random initialization:
𝑇𝑆𝑖 = 𝑟𝑜𝑢𝑛𝑑[𝑇𝑆𝑙𝑜𝑤𝑒𝑟,𝑟1
𝑖
+ 𝑟𝑎𝑛𝑑 × (𝑇𝑆𝑢𝑝𝑝𝑒𝑟,𝑟1
𝑖
− 𝑇𝑆𝑙𝑜𝑤𝑒𝑟,𝑟1
𝑖
)] (24)
𝑏𝑢𝑠_𝐶𝑎𝑝𝑖 = 𝑟𝑜𝑢𝑛𝑑[𝑏𝑢𝑠𝑙𝑜𝑤𝑒𝑟,𝑟2
𝑖
+ 𝑟𝑎𝑛𝑑 × (𝑏𝑢𝑠𝑢𝑝𝑝𝑒𝑟,𝑟2
𝑖
− 𝑏𝑢𝑠𝑙𝑜𝑤𝑒𝑟,𝑟2
𝑖
)] (25)
𝑠𝑖𝑧𝑒_𝐶𝑎𝑝𝑖 = 𝑟𝑜𝑢𝑛𝑑[𝑠𝑖𝑧𝑒𝑙𝑜𝑤𝑒𝑟,𝑟3
𝑖
+ 𝑟𝑎𝑛𝑑 × (𝑠𝑖𝑧𝑒𝑢𝑝𝑝𝑒𝑟,𝑟3
𝑖
− 𝑠𝑖𝑧𝑒𝑙𝑜𝑤𝑒𝑟,𝑟3
𝑖
)] (26)
where r1=1, 2, ... NTL, r2=1, 2, ... m and r3=1, 2 ... m. TSlower, r1 and TSupper, r1 are the least tie switch and
maximum tie switch which are encoded in the fundamental loop r1. Capacitors are placed on any bus of the
network apart from the slack bus which represent the first bus. Hence, the lower limit (sizelower, r2) and upper
limit (sizeupper, r2) of the placement of the capacitor is from bus 2 to the last bus and the sizes of each capacitor
is from 150 kVAr to maximum power of capacitor as given in the inequality constraint of (5).
Step 3: Evaluation each dingo’s fitness
A radial configuration check is performed for the dingo population. The fitness function of a non-
radial configuration is set to infinity. The fitness function, which is employed in this work as the power loss,
is obtained by conducting the load flow for each dingo using the Newton-Raphson technique. Depending on
the parameters of the fitness function (power loss), the following outcomes are attained: i) the dingo with the
best search (Da), ii) the dingo with the second-best search (Db); and the Dingo search results after words (Dc).
Step 4: Updating the dingoes status
For i=1: Dn utilize the set of (13)–(18) to update the most recent search agent status.
Step 5: Estimation of the fitness of updated dingoes
A radial configuration check is conducted for the upgraded dingoes. A non-radial configuration's
fitness function is set to infinity. The loss is computed by performing a power flow on the dingoes (objective
function and fitness function).
Step 6: Determination of the fitness value of dingoes
Keep track of the values of Sa, Sb and Sc and Keep track of the values of 𝑏
⃗ , 𝐴, and 𝐵
⃗ .
Step 7: Termination criterion
As the iteration number approaches the maximum, the dingoes' status is updated continuously. The
IEEE 33-bus network depicted in Figure 1 was used as the basis for applying the DOA method, with
continuous lines denoting sectionalizing switches and broken lines denoting tie switches. This network's line
and load data were taken from [31].
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
19 20 21
23 24 25
26 27 28 29 30 31 32 33
22
(33)
(34)
(35)
(37)
(36)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)
(18)
(19) (20) (21)
(22)
(23) (24)
(25)
(26) (27) (28) (29) (30) (31) (32)
Figure 1. IEEE 33-bus network

 ISSN: 2088-8708
2390
3. RESULTS AND DISCUSSION
The simulations were done with a precision of 1e-10 across a total of 20 iterations. The simulation
findings for the simultaneous network reconfiguration and capacitor allocations are summarized in Table 1.
The results shown in Table 1 indicate that opening switches 7, 11, 17, 27, and 34 resulted in optimal network
reconfiguration. The network's radiality was preserved after reconfiguration, and no load was disconnected
from the supply. Similarly, as shown in Table 1, the optimal capacitor positions were buses 8, 29, and 30,
with optimal capacitor capacities of 512, 714, and 495 kVAr, respectively. Figure 2 depicts the optimal
reconfigured network for visual inspection. The following subsections go through the remaining simulation
findings and the comparison of the results of the proposed DOA technique with those of other optimization
techniques.
Table 1. Simulation results
Base case tie switches
Tie switches after reconf. + cap.
Location/Capacitor size (kVAr)
33, 34, 35, 36, 37
7, 11, 17, 27, 34
8/512, 29/714, 30/495
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
19 20 21
23 24 25
26 27 28 29 30 31 32 33
22
(33)
(34)
(35)
(37)
(36)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)
(18)
(19) (20) (21)
(22)
(23) (24)
(25)
(26) (27) (28) (29) (30) (31) (32)
Figure 2. The optimal network configuration after simultaneous network reconfiguration and capacitor
allocations
3.1. Voltage profile
Figure 3 compares the network voltage profile before and after concurrent network reconfiguration
and capacitor allocations. When Figure 3 was visualized, the voltage magnitudes of buses 6 through 18, as
well as buses 26 through 33, were not sufficient because they were below the acceptable lower voltage limit,
with bus 18 having the least voltage value of 0.9131 p.u. In other words, the network's voltage profile was
quite subpar. Additionally, it is evident from Figure 3 that the network's overall voltage profile had greatly
enhanced, as the buses whose voltage magnitudes were below the acceptable lower voltage limit had their
voltage magnitudes significantly improved, with the least voltage magnitude at bus 18 now increased to
0.9530 p.u. following simultaneous network reconfiguration and capacitor allocations. The voltage profile
was generally remarkedly elevated, proving the viability of the proposed DOA technique in solving the
optimization problems of simultaneous network reconfiguration and capacitor allocations to upgrade the
network voltage profile and boost the network operational efficiency.
3.2. Real and reactive power losses
Figure 4 gives a quick overview of the real and reactive power losses in each branch of the
distribution networks. The network was found to have significant real and reactive power losses. However,
these losses were dramatically decreased following the optimal simultaneous network reconfiguration and
capacitor allocations. Figures 5 and 6 compare the network's real and reactive power losses at a glance before

2391
and after simultaneous network reconfiguration and capacitor allocations for better understanding. As
indicated in Figure 5, the greatest real power loss at branch 2-3 was reduced from 51.8 to 20.88 kW,
representing a loss reduction of 59.69%. Similar to this, branch 5-6, which had the most reactive power loss,
experienced a large loss reduction as the loss was dramatically decreased from 33.1 to 0.9169 kVAr, equating
to a 97.23% minimization, as seen in Figure 6.
Figure 3. Voltage profile of the network before and after simultaneous network reconfiguration and capacitor
allocations
Figure 4. Base case real and reactive power losses
All of the network's branches had general reductions in real and reactive power losses, except for
branches 2-19, 19-20, 20-21, 21-22, 3-23, 23-24, 24-25, 30-31, 31-32, 21-8, 12-22, and 25-29, where there
were modest rises. This came about as a result of the real and reactive power being redistributed along the
system branches to enhance the overall network performance. Additionally, the total real and reactive power
losses, which were formerly 202.60 kW and 135 kVAr, respectively, were dramatically decreased to
103.65 kW and 77.53 kVAr, or 48.87 and 42.57%, respectively. The simulation findings show that the
proposed DOA is efficient in addressing the simultaneous network reconfiguration and capacitor allocation
optimization problems by reducing the network power losses and thus enhancing network performance.
0
5
10
15
20
25
30
35
40
45
50
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
2-19
19-20
20-21
21-22
3-23
23-24
24-25
6-26
26-27
27-28
28-29
29-30
30-31
31-32
32-33
21-8
9-15
12-22
18-33
25-29
Line
Losses
Line
Real Power Loss (kW)
Reactive Power Loss
(kVAr)

 ISSN: 2088-8708
2392
Figure 5. Real power loss before and after simultaneous network reconfiguration and capacitor allocations
Figure 6. Reactive power loss before and after simultaneous network reconfiguration and capacitor
allocations
3.3. Comparison of DOA results with those of other optimization methods in the literature
The findings were compared to those of the modified flower pollination algorithm (MFPA), binary
gravitational search algorithm (BGSA), the adaptive whale optimization algorithm (AWOA), improved
binary particle swarm optimization (IBPSO), fuzzy harmony search (Fuzzy-HS), and modified whale
algorithm (MWA) in the literature to confirm the applicability of the proposed DOA approach. Table 2
provides documentation of the comparison of the DOA solutions with those of other optimization methods.
Column two of Table 2 lists the switches that were opened (i.e., tie switches) following simultaneous
network reconfiguration and capacitor allocations. Table 2's third column contains the capacitor locations and
sizes, whereas column four contains the percentage decrease in real power loss for each method as well as for
the proposed DOA technique.
The proposed DOA optimization approach demonstrated superiority over existing optimization
strategies in the literature, with a real power loss reduction percentage of 48.87%, as shown in Table 2. The
percentage real power reductions obtained by other optimization methods, as indicated in Table 2, were not
as high as that of the proposed DOA technique. Also, it can be seen that the IBPSO and MFPA techniques
0
5
10
15
20
25
30
35
40
45
50
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
2-19
19-20
20-21
21-22
3-23
23-24
24-25
6-26
26-27
27-28
28-29
29-30
30-31
31-32
32-33
21-8
9-15
12-22
18-33
25-29
Real
Power
Loss
(kW)
Line
Base Case
Recong. + Cap. Allocations
0
5
10
15
20
25
30
35
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
2-19
19-20
20-21
21-22
3-23
23-24
24-25
6-26
26-27
27-28
28-29
29-30
30-31
31-32
32-33
21-8
9-15
12-22
18-33
25-29
Reactive
Power
Loss
(kVAr
)
Line
Base Case
Recong. + Cap. Allocations

2393
have the least power loss reduction of 31.14%, closely followed by MWA at 32.68%, BGSA at 37.42%,
AWOA at 42.68%, and Fuzzy-HS at 44.49%. So, the simultaneous network reconfiguration and capacitor
allocation optimization challenges were better addressed by the proposed DOA approach, as Table 2 clearly
shows.
Table 2. Proposed DOA's performance summary against other existing optimization techniques
Algorithm Ties Switches Capacitor size in kVAr (Location) % Reduction
BGSA [21]
MFPA [22]
AWOA [23]
IBPSO [24]
Fuzzy-HS [20]
MWA [25]
Proposed DOA
10, 14, 28, 32, 33
7, 14, 9, 32, 37
33, 34, 35, 36, 25
7, 14, 9, 32, 37
6, 9, 32, 34, 37
33, 34, 35, 36, 25
7, 11, 17, 27, 34
450 (6), 300 (12), 900 (30), 600 (29)
750 (6), 150 (28), 850 (29)
400 (24), 250 (25), 150 (30)
900 (2), 300 (4), 300 (15), 300 (23), 300 (25), 600 (31), 600 (32)
450 (11), 900 (5), 150 (19), 150 (29), 150 (23)
400 (24), 250 (25), 150 (30)
512 (8), 714 (29), 495 (30)
37.42
31.14
42.68
31.14
44.49
32.68
48.87
4. CONCLUSION
The IEEE 33 bus network's simultaneous network reconfiguration and capacitor allocation
optimization problems were resolved by employing a novel DOA optimization approach. With the help of the
proposed technique, the optimization problem posed by simultaneous network reconfiguration and capacitor
allocations was adequately addressed, and the overall network voltage profile was greatly enhanced as the
minimum voltage magnitude was increased to 0.9530 p.u. Also, significant improvements were made in the
network's overall real and reactive power losses as the total real and reactive power losses decreased by 48.87
and 42.57%, respectively. According to the comparison findings, the proposed method's outcomes were
noticeably superior to those of other optimization approaches in the literature. The simultaneous solution of
network reconfiguration and capacitor allocation may be achieved using DOA, a powerful optimization
approach.
REFERENCES
[1] S. O. Ayanlade and O. A. Komolafe, “Distribution system voltage profile improvement based on network structural
characteristics,” in Faculty of Technology Conference, 2019, pp. 75–80.
[2] A. Jimoh, S. O. Ayanlade, F. K. Ariyo, and A. B. Jimoh, “Variations in phase conductor size and spacing on power losses on the
Nigerian distribution network,” Bulletin of Electrical Engineering and Informatics, vol. 11, no. 3, pp. 1222–1233, Jun. 2022, doi:
10.11591/eei.v11i3.3753.
[3] S. A. Salimon, G. A. Adepoju, I. G. Adebayo, and S. O. Ayanlade, “Impact of shunt capacitor penetration level in radial
distribution system considering techno-economic benefits,” Nigerian Journal of Technological Development, vol. 19, no. 2,
pp. 101–109, Aug. 2022, doi: 10.4314/njtd.v19i2.1.
[4] S. O. Ayanlade, O. A. Komolafe, I. O. Adejumobi, and A. Jimoh, “Distribution system power loss minization based on network
structural characteristics,” in 1st International Conference on Engineering and Environmental Science, 2019, pp. 849–861.
[5] R. Parasher, “Load flow analysis of radial distribution network using linear data structure,” arXiv preprint arXiv:1403.4702, Mar.
2014.
[6] M. G. Hemeida, A. A. Ibrahim, A.-A. A. Mohamed, S. Alkhalaf, and A. M. B. El-Dine, “Optimal allocation of distributed
generators DG based manta ray foraging optimization algorithm (MRFO),” Ain Shams Engineering Journal, vol. 12, no. 1,
pp. 609–619, Mar. 2021, doi: 10.1016/j.asej.2020.07.009.
[7] J. A. Ganesh and S. Michline Rupa, “Power flow analysis for radial distribution system using backward/forward sweep method,”
International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering, vol. 8, no. 10,
pp. 1627–1632, 2014.
[8] A. S. Abubakar, K. R. Ekundayo, and A. A. Olaniyan, “Optimal reconfiguration of radial distribution networks using improved
genetic algorithm,” Nigerian Journal of Technological Development, vol. 16, no. 1, Feb. 2019, doi: 10.4314/njtd.v16i1.2.
[9] M. Al Samman, H. Mokhlis, N. N. Mansor, H. Mohamad, H. Suyono, and N. M. Sapari, “Fast optimal network reconfiguration
with guided initialization based on a simplified network approach,” IEEE Access, vol. 8, pp. 11948–11963, 2020, doi:
10.1109/ACCESS.2020.2964848.
[10] R. A.V. Sudhakara and R. M. Damodar, “Application of whale optimization algorithm for distribution feeder reconfiguration,”
i-manager’s Journal on Electrical Engineering, vol. 11, no. 3, 2018, doi: 10.26634/jee.11.3.14119.
[11] A. O. Salau, Y. W. Gebru, and D. Bitew, “Optimal network reconfiguration for power loss minimization and voltage profile
enhancement in distribution systems,” Heliyon, vol. 6, no. 6, Jun. 2020, doi: 10.1016/j.heliyon.2020.e04233.
[12] P. Khetrapal, “Distribution network reconfiguration of radial distribution systems for power loss minimization using improved
harmony search algorithm,” International Journal on Electrical Engineering and Informatics, vol. 12, no. 2, pp. 341–358, Jun.
2020, doi: 10.15676/ijeei.2020.12.2.11.
[13] S. R. Salkuti, “Multi-objective based optimal network reconfiguration using crow search algorithm,” International Journal of
Advanced Computer Science and Applications, vol. 12, no. 3, 2021, doi: 10.14569/IJACSA.2021.0120310.
[14] R. Pegado, Z. Ñaupari, Y. Molina, and C. Castillo, “Radial distribution network reconfiguration for power losses reduction based
on improved selective BPSO,” Electric Power Systems Research, vol. 169, pp. 206–213, Apr. 2019, doi:
10.1016/j.epsr.2018.12.030.
[15] S. Biswal, A. Ghosh, S. Kumar, N. Chakraborty, and S. K. Goswami, “Cuckoo search algorithm based cost minimization by
optimal DG and capacitor integration in radial distribution systems,” in 2018 20th National Power Systems Conference (NPSC),
Dec. 2018, pp. 1–6, doi: 10.1109/NPSC.2018.8771773.

 ISSN: 2088-8708
2394
[16] A. Y. Abdelaziz, E. S. Ali, and S. M. Abd Elazim, “Optimal sizing and locations of capacitors in radial distribution systems via
flower pollination optimization algorithm and power loss index,” Engineering Science and Technology, an International Journal,
vol. 19, no. 1, pp. 610–618, Mar. 2016, doi: 10.1016/j.jestch.2015.09.002.
[17] M. O. Okelola, O. W. Adebiyi, S. A. Salimon, S. O. Ayanlade, and A. L. Amoo, “Optimal sizing and placement of shunt
capacitors on the distribution system using whale optimization algorithm,” Nigerian Journal of Technological Development,
vol. 19, no. 1, pp. 39–47, Jun. 2022, doi: 10.4314/njtd.v19i1.5.
[18] D. B. Prakash and C. Lakshminarayana, “Optimal siting of capacitors in radial distribution network using whale optimization
algorithm,” Alexandria Engineering Journal, vol. 56, no. 4, pp. 499–509, Dec. 2017, doi: 10.1016/j.aej.2016.10.002.
[19] S. M. Abd Elazim and E. S. Ali, “Optimal locations and sizing of capacitors in radial distribution systems using mine blast
algorithm,” Electrical Engineering, vol. 100, no. 1, pp. 1–9, Mar. 2018, doi: 10.1007/s00202-016-0475-1.
[20] A. A. Z. Diab and H. Rezk, “Optimal sizing and placement of capacitors in radial distribution systems based on grey wolf,
dragonfly and moth–flame optimization algorithms,” Iranian Journal of Science and Technology, Transactions of Electrical
Engineering, vol. 43, no. 1, pp. 77–96, Mar. 2019, doi: 10.1007/s40998-018-0071-7.
[21] M. Addisu, A. O. Salau, and H. Takele, “Fuzzy logic based optimal placement of voltage regulators and capacitors for distribution
systems efficiency improvement,” Heliyon, vol. 7, no. 8, Aug. 2021, doi: 10.1016/j.heliyon.2021.e07848.
[22] S. Esmaeili, H. D. Dehnavi, and F. Karimzadeh, “Simultaneous reconfiguration and capacitor placement with harmonic
consideration using fuzzy harmony search algorithm,” Arabian Journal for Science and Engineering, vol. 39, no. 5,
pp. 3859–3871, May 2014, doi: 10.1007/s13369-014-0971-4.
[23] H. R. Esmaeilian and R. Fadaeinedjad, “Distribution system efficiency improvement using network reconfiguration and capacitor
allocation,” International Journal of Electrical Power and Energy Systems, vol. 64, pp. 457–468, Jan. 2015, doi:
10.1016/j.ijepes.2014.06.051.
[24] G. Namachivayam, C. Sankaralingam, S. K. Perumal, and S. T. Devanathan, “Reconfiguration and capacitor placement of radial
distribution systems by modified flower pollination algorithm,” Electric Power Components and Systems, vol. 44, no. 13,
pp. 1492–1502, Aug. 2016, doi: 10.1080/15325008.2016.1172281.
[25] M. R. Babu, C. V. Kumar, and S. Anitha, “Simultaneous reconfiguration and optimal capacitor placement using adaptive whale
optimization algorithm for radial distribution system,” Journal of Electrical Engineering and Technology, vol. 16, no. 1,
pp. 181–190, Jan. 2021, doi: 10.1007/s42835-020-00593-5.
[26] M. Sedighizadeh, M. Dakhem, M. Sarvi, and H. H. Kordkheili, “Optimal reconfiguration and capacitor placement for power loss
reduction of distribution system using improved binary particle swarm optimization,” International Journal of Energy and
Environmental Engineering, vol. 5, no. 1, Apr. 2014, doi: 10.1007/s40095-014-0073-9.
[27] S. Anitha, “Simultaneous reconfiguration and optimal capacitor placement for loss and cost reduction in radial distribution
system,” International Journal of Scientific and Engineering Research, vol. 11, no. 4, pp. 351–359, 2020.
[28] A. K. Bairwa, S. Joshi, and D. Singh, “Dingo optimizer: a nature-inspired metaheuristic approach for engineering problems,”
Mathematical Problems in Engineering, vol. 2021, pp. 1–12, Jun. 2021, doi: 10.1155/2021/2571863.
[29] E. S. Ali, S. M. Abd Elazim, and A. Y. Abdelaziz, “Improved harmony algorithm and power loss index for optimal locations and
sizing of capacitors in radial distribution systems,” International Journal of Electrical Power and Energy Systems, vol. 80,
pp. 252–263, Sep. 2016, doi: 10.1016/j.ijepes.2015.11.085.
[30] M. Mohammadi, A. M. Rozbahani, and S. Bahmanyar, “Power loss reduction of distribution systems using BFO based optimal
reconfiguration along with DG and shunt capacitor placement simultaneously in fuzzy framework,” Journal of Central South
University, vol. 24, no. 1, pp. 90–103, Jan. 2017, doi: 10.1007/s11771-017-3412-1.
[31] K. Dharageshwari and C. Nayanatara, “Multiobjective optimal placement of multiple distributed generations in IEEE 33 bus
radial system using simulated annealing,” in 2015 International Conference on Circuits, Power and Computing Technologies,
Mar. 2015, pp. 1–7, doi: 10.1109/ICCPCT.2015.7159428.
BIOGRAPHIES OF AUTHORS
Samson Oladayo Ayanlade graduated with a Bachelor of Technology (B.Tech)
degree in Electronic and Electrical Engineering from Ladoke Akintola University of
Technology, Ogbomoso, Oyo State, Nigeria in 2012 and an M.Sc. in Power System
Engineering from Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria in 2019. He is
currently a Ph.D. student at Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria. His
research interests are primarily in the areas of power system optimization and power system
stability. He can be contacted by email: samson.ayanlade@lcu.edu.ng and
ayanladeoladayo@gmail.com.
Abdulrasaq Jimoh received the Bachelor of Engineering (B.Eng.) degree in
Electrical Engineering from the University of Ilorin in 2002 and the M.Sc. degree in Electrical
Power System Engineering from the University of Lagos in 2010. He is a registered engineer
with the Council for Regulation of Engineering in Nigeria, a Fellow of the Nigerian Society of
Engineers and a Fellow of the Nigerian Institution of Power Engineers. He was the Technical
Engineer of the Ibadan Electricity Distribution Company (IBEDC), Ibadan, Nigeria.
Currently, he is the Business Manager with IBEDC and is also pursuing an MPhil/Ph.D.
degree at Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria. His research interests
include distribution system voltage control and improvement, and the economic operation of
distribution systems. He can be contacted by email at jimohabdulrasaq@gmail.com.

2395
Emmanuel Idowu Ogunwole obtained his Bachelor of Technology (B.Tech)
degree in 2012 from the department of Electrical Engineering, Ladoke Akintola University of
Technology, Ogbomosho, Oyo State, Nigeria. He further obtained his Master of Science
(MSc) degree in 2020 in the discipline of Electrical Engineering at the University of
KwaZulu-Natal, Durban, South Africa. He is currently pursuing his doctorate in the Electrical
Engineering Department at the Cape Peninsula University of Technology, South Africa. His
research interests span several areas in the field of electrical engineering which are; power
systems analysis and optimization, energy management, and distributed computing. He is a
member of the following professional bodies; the Nigerian Society of Engineers (NSE), the
Council for the Regulation of Engineering in Nigeria (COREN), the Society for Automation,
Instrumentation, Mechatronics, and Control (SAIMC), and the South African Institute of
Electrical Engineers (SAIEE). He can be contacted at email: emmanuelidowu18@gmail.com.
Abdullahi Aremu earned a Bachelor of Science degree in Electrical and
Computer Engineering from the Federal University of Technology, Minna in 2002 and a
Master of Science degree in Electrical Power System Engineering from Obafemi Awolowo
University, Ile-Ife in 2012. He is a registered engineer with the Council for Regulation of
Engineering in Nigeria, a member of the Nigerian Society of Engineers, and a Fellow of the
Nigeria Institution of Electrical and Electronic Engineers. He was the Manager of Network
Planning and Design at Ibadan Electricity Distribution Company (IBEDC), Ibadan, Nigeria.
Currently, he is a technical engineer with IBEDC. He can be contacted at email:
rms_ng@yahoo.com.
Abdulsamad Bolakale Jimoh was born in 1999 and is a final year student in the
Department of Electrical and Electronic Engineering at the University of Ilorin, Kwara State,
Nigeria. His research interests include energy management and power economics. He can be
contacted at email: jimohabdulsamad@gmail.com.
Dolapo Eniola Owolabi obtained her Bachelor of Technology (B.Tech) degree in
2021 from the department of Electrical and Electronic Engineering, Ladoke Akintola
University of Technology, Ogbomosho, Oyo State, Nigeria. She is currently an MSc. student
at the institution where she received her Bachelor of Technology (B. Tech) degree. She is a
graduate member of the Association of Professional Women Engineers of Nigeria
(GMAPWEN), and a graduate member of the Nigerian Society of Engineers (GMNSE). Her
research interests are in the areas of power system optimization and energy management
systems. She can be contacted by email: owolabidolapo343@gmail.com.

Simultaneous network reconfiguration and capacitor allocations using a novel dingo optimization algorithm

Recommended

More Related Content

Similar to Simultaneous network reconfiguration and capacitor allocations using a novel dingo optimization algorithm (20)

More from IJECEIAES (20)

Recently uploaded (20)

Simultaneous network reconfiguration and capacitor allocations using a novel dingo optimization algorithm