Energy Reports 6 (2020) 1581–1589

Contents lists available at ScienceDirect

Energy Reports
journal homepage: www.elsevier.com/locate/egyr

Research paper

Multi-objective for optimal placement and sizing DG units in reducing

loss of power and enhancing voltage profile using BPSO-SLFA
∗
Abdurrahman Shuaibu Hassan a , Yanxia Sun a , , Zenghui Wang b
a
Department of Electrical and Electronics Engineering Science, University of Johannesburg, Auckland Park 2006, South Africa
b
Department of Electrical and Mining Engineering, University of South Africa, Florida 1710, South Africa

article info a b s t r a c t

Article history: Algorithms are used to optimize both single and multi-objective system limits. This research aimed
Received 24 March 2020 to detect the optimal location and size of the DGs, which can significantly minimize power loss and
Received in revised form 21 April 2020 improve the stability of the voltage. The research uses binary particle swarm optimization and shuffled
Accepted 15 June 2020
frog leap (BPSO-SLFA) algorithms for simulation and testing of an optimal power flow (OPF) on 33 and
Available online 20 June 2020
69 bus radial distribution system. The result shows that the algorithms give better DG allocation and
Keywords: minimizes the power losses but at the nascent stage of advancement. The power losses associated with
Algorithms the system have significantly reduced up to 31.8244kW using multi-DGs reconfiguration placement.
Minimize power loss The outcomes are established to verify the potency of the recommend algorithm to minimize losses,
Optimal power flow general improvement in voltage profiles and cost saving for various distribution system. However,
Distributed generation the proposed methodology can be used as a reliable method in DG settings and sizing in distribution
Cost saving & shuffled leap algorithm network system which produce better outputs rather than hybrid grey wolf optimization (GWO) and
hybrid big bang big crunch.
© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction power loss and improving the strength of the voltage. Yahiaoui
et al. (2017) proposed GWO for the optimal sizing of hybrid
The Institute of Electrical Electronics Engineering (IEEE) gives a renewable systems to lessen the cost of hybrid power generation.
concise definition of distributed generation as a power generation Other researchers have proposed a hybridization algorithm in
that is adequate compact than central generating plants and are order to replace the traditional algorithm. A particle artificial
connected closer to the distribution system (Chiradeja and Ra- bee colony (PABC)-hybrid harmony search algorithm (HSA) ap-
makumar, 2004; Pepermans et al., 2005; Singh and Parida, 2015). proach was developed to optimal size and location of the DG
The rapid development in technology, global concern about the in distribution system (Muthukumar and Jayalalitha, 2016). Their
environment and the demand of the customers for cheap and results showed that the efficiency of the proposed hybrid algo-
consistent electric power have led to an increasing interest in DG rithm in obtaining optimal solution for simultaneous placement
(Kansal et al., 2016). However, optimal allocation and sizing of of DGs and shunt capacitors in distribution networks. Kefayat
DG may bring various advantages such as voltage control through
et al. (2015) applied artificial bee colony and hybrid ant colony
cost (El-Khattam et al., 2004), boosting reliability (Rohatgi et al.,
optimization to the distribution system in order to control emis-
2006), harmonic reduction (Tagore and Gupta, 2017), minimiza-
sions expelled by the substation and overall improvement of the
tion of real and reactive losses (Mallipeddi et al., 2017), voltage
voltage stability. Doagou-mojarrad et al. (2013) reported a fuzzy
stability and network loss reduction (Esmaili et al., 2014), and
based intelligent system based on shuffled frog leap algorithm to
infeasible network configuration (Nguyen et al., 2016).
Many optimization algorithm and techniques have been devel- resolve optimal allocation problem with the aim of minimizing
oped for the application of energy technologies problem. Among power loss, energy cost and control of emission produced. Re-
the various techniques, algorithms and optimization present the sults showed that the simulation illustrate the good performance
highest potential for use in the placement of DG. Nguyen et al. and applicability of the proposed method. Another research con-
(2016) adopted the novel cuckoo based (CSA) algorithm for place- ducted by Moradi and Abedini (2012), which obtained a better
ment of DG with the technical objective of minimizing the active result in terms of reduction of loss of power, boosting voltage
regulation, and maintaining voltage stability by application of
∗ Corresponding author. particle swarm algorithm and genetic algorithm. Selim et al.
E-mail addresses: [email protected] (A.S. Hassan), (2019), proposed chaotic sine cosine algorithm for optimal lo-
[email protected] (Y. Sun), [email protected] (Z. Wang). cation and sizing DG in distribution network with minimum

https://doi.org/10.1016/j.egyr.2020.06.013
2352-4847/© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
1582 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589

power loss and high convergence rate. Their results significantly The rest of this paper is structured as follows. Section 2 ex-
shows how the new algorithm can be used to optimal allocate plains the methodology and problem formulation of BPSO-SLFA.
multiple DG units into radial distribution networks with technical Section 3 illustrates the conventional BPSO and SLFA algorithms
objective of minimizing power loss, voltage deviation and max- for DG placement to access optimal allocation for connecting
imize voltage stability index. Menesy et al. (2019), developed a DG and their significance for solving reconfiguration transient
Chaotic Harris Hawks optimization techniques for estimating the manner. Section 4 describes the results and flowchart algorithm
exact operating parameters for proton exchange of fuel cell for on four (4) different case scenarios. Lastly, the main conclusion
high electrical performance. The author’s findings are estimated are given in Section 5.
data obtained by the proposed CHHO shown a good agreement
with the experimental data of different commercial PEM fuel cell
2. Methodology
stacks.
As previously reported that the electric DG system for reduc-
tion of power loss and improvement of bus voltage stability, most Minimizing the losses of power, bus voltage stability, and max-
widely used methods are reconfiguration and DG allocation (Sam- imizing voltage stability are considered as the fitness functions
baiah and Jayabarathi, 2019). Optimal network reconfiguration for the placement of the optimal sizing of this current research
and DG allocation offer maximum essential benefits. Since combi- work.
nation between reconfiguration and DG allocation is an important
and complex non-linear constrained optimization problem. The 2.1. Minimization of power loss
optimal configuration of feeder for minimizing forms of cost and
maximizing reliability index can be obtained using hybrid PSO- Electrical energy is produced from a long distance away from
SLA (Gitizadeh et al., 2012). A hybrid particle swarm optimization the user and is distributed between the transmission lines to
(HPSO) was presented by Alrashidi and Member (2007), the re- a variety of distribution system on which the electrical utility
search aimed to lessen the real power, fuel cost, and reduction of works. Commonly, the distribution network system takes power
gaseous emission produced by the generating plant units. Simi- and sent it to the consumer of load in order to aid their demands.
larly, Khalesi et al. (2011) examined the multi-objective function Nevertheless, not all the power will be delivered 100% efficient
of locating DGs in radial distribution networks. Time varying load due to losses that occur at the distribution network line. The
is employed in this optimization to attain the practical outcomes loss of power in the network distribution system is absolutely
of the study and their needs are based on cost or benefit forms. dependent on the precise position and size of the renewable DG
They concluded that to eliminate the multi-objective problem,
system. The real loss of power in a distribution system with given
novel approach based on dynamic programming is required.
operational conditions are computed using Eq. (1), and is referred
There are various researches published on optimal allocation
to as exact loss.
and placement of DG, but very few have evaluated BPSO-SLFA, be-
nbr
cause it can be difficult to define the initial parameters. But BPSO- ∑
SLFA are proposed to decrease high dimensionality of the feature Minimize Ploss (F1) = Ii2 Ri (1)
set and to select optimized feature subsets. A new methodology i=1
is proposed to minimize loss of power and boost the voltage where:
profile using a combination of BPSO-SLFA. The BPSO-SLFA is a Ii and Ri are the current magnitude and resistance correspond-
new proposed algorithm that works based on reconfiguring the ing to the circuit branch I, number of branches is represented by
DG in a momentary path, and is used to solve multi-objective br. The branch circuit current is divided into two components,
function. This paper developed a BPSO-SLFA-based algorithm to active component (Iac) and reactive component (Irc).
optimal locate and size DG position with the suggestion of multi-
n
objective function for reconfiguration and DG installation, loss of ∑
power system, and stability of voltage. The developed methodol- PLa (F2) = I 2 aciRi (2)
ogy is analyzed on IEEE 33 & 69-bus radial distribution systems. i=1
n
The obtained results shows effectiveness and robustness of the ∑
proposed methodology in solving optimal location and sizing the PLr (F3) = I 2 rciRi (3)
DG in the DN and reduces total power loss and boost voltage i=1
profile within a defined network system. However, optimal placement of DGs can compensate the active
Furthermore, the research aimed to explain the problem for- loss components in the branch.
mulation methodology, the types of DG and fundamental back-
ground of the proposed hybrid and BPSO-SLFA algorithm, and PG = PD + Losses (4)
optimization problem for the optimal DG method and simula-
tion. However the main contribution of this research paper are The real generator power and the demand power are given as PG
summarized as follows: and PD respectively.
– The proposed algorithm is applied for resolving optimal dis- Voltage limits:
tribution network reconfiguration and optimal DG unit with the Eq. (5), is useful for improving the voltage profile as one of the
technical objective of decreasing power loss and enhancing volt- key objectives.
age profile;
NN
– First, the algorithm is exercised for resolving optimal DG place- ∑
ment that assess optimal allocation for connecting DG and their F4 = (VNi − Vrated ) (5)
significance for solving reconfiguration transient problem. Ni=1

– The obtained outcomes clearly indicates that scenario 4 (Multi

DG installation with network reconfiguration) is found to be more 3. Optimal sitting and sizing problem formulation
effective in reducing power losses.
– The simulated results obtained from the proposed BPSO-SLFA The radial distribution system reduces actual losses of power
are compared with the results of other techniques to assess the by optimal placement of DGs in order to boost voltage profile and
performance and effectiveness of the new proposed techniques. decreases network operating costs that are subject to different
 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589 1583

operational constraints. Objective functions of the problem are 3.2.2. The SLFO algorithm
formulated mathematically. This research work uses novel ap- Eusuff et al. (2017), developed the SLFA algorithm, a memetic
proach based on BPSO-SLFA for resolving the problems related meta-heuristic for generating hybrid optimization. This algorithm
to optimal sizing and sitting of DG in the distribution system. is a metaheuristic technique of optimization that simulates the
memetic growth of a group of frogs while trying to find the
3.1. Major contribution of this research work optimum location which has maximum amount of food avail-
able. Memetic algorithms are adapted based on population ap-
The algorithm is initially exercised to solve the placement of proaches to optimization problems in heuristic searches. The
the optimal DG, which determines optimal position for connect- word memetic has its origin in ‘‘meme’’ Dawkins (1976). ‘‘ The
ing DG and its values for reconfiguring in a momentary path. Selfish Gene’’., Oxford University Press, n.d.). Meme is assumed
The simulated outcomes are compared with the ones of other as the cultural evolution unit. The concept evolves in a way that
proposed techniques to assess the impact and performance of is similar to biological evolution, the SLFA contains a population-
the proposed technique. The flow chart describing the proposed based solution identified as memplexes which are divided into a
BPSO-SLFA is shown in Fig. 1. subset. Inside each memeplex the individual frogs do hold idea
of the other frogs and causes them to propagate the pattern.
3.2. Proposed hybrid algorithm The SLFA algorithm progresses in the form of memestic evolution
through the time loops (Eusuff and Lansey, 2004). In accomplish-
3.2.1. Basic Particle Swarm Optimization (BPSO) ing this target some network parameters must be fulfilled. One
James Kennedy and Eberhart first hosted classical optimization can describe the problem statement as in Eq. (14)
called BPSO, practical swarm optimization in the year (1995). This
Particle Swarm Optimization comprises of a collection (swarm) of Objective Function = Min (TLP) (14)
people (particles) moving in the search space, fitness values are ∑n2
where, TLP = i=1 Ii Ri is the absolute loss of real power for
determined by their trajectory movement. The particle is repre-
the radial distribution syst3‘em. The voltage is subject to the
sented by a location-indicating n-length vector and has a vector
constraint |vimin | ≤ |vi | ≤ |vimax |. Here, Ii is the total flowing
v for present position update. The velocity vector is calculated
current over the ith branch that is the position and size features
according to the following equation in (6). In PSO, the location
and each particle velocity at k iteration in the space search is of the DG. Ri is the branch resistance, and the number of branches
defined by X i k and V i k . The iterated particle velocity I is described in the system is denoted as n. The lower and upper limits of the
as k + 1 P i lbest and attained from Eq. (6). ith bus voltages are Vimin and Vimax.
The SFL estimate joins the advantages of BPSO calculations
V i k+1 = ω.V i k + C 1.R1(P i lbest − X i k ) + C 2.R2(P i global − X i k ) (6) based on inherited and social behavior. For S-dimensional vari-
where ables problems, a frog i is defined below.
The random functions are R1 and R2, and where the training
Xi = (xi1 , xi2 , . . . , xis ) (15)
coefficients are C1 and C2. This is the dimension of inertia weight.
The following result can be defined by Eq. (7). Afterward, the frogs are sorted in relation to their fitness in
a downward order. The entire population is broken into meme-
ω = ωmax − {(ωmax − ωmin ) − kmax } × k (7)
plexes, each with n frogs (p = m × n). The best and worst-fit
where, kmax donates the possible number of iterations. At the end frogs are identified as xg . Then, a method similar to BPSO is
of each iteration, the sum of the old position and the new velocity implemented in each step to boost only the worst-fit frog in every
position obtains a new location for each particle complete cycle. Consequently, the frog’s location having the worst
X i k+1 = X i k + V i k+1 (8) suitability is modified as follows:
The frog position deviations are defined as follows
The PSO formula remained unaffected. A logistic conversion
S(V i k ) is used to achieve this amendment that is written in (Di ) = rand() × (xb − xw ) (16)
Eqs. (9) and (10). New position Xw = current position (Xw + Di ) (17)
1
S(V i k+1 ) = sig mod e V i k+1 =
( )
(9) The inequality that satisfies the fitness of the position is defined
1 + exp(V i k+1 )
below

Dmax≥ Di ≥ −Dmax
If rand ∝ S(V i k+1 ) then: X i k+1 = 1; (10)

Else: X i k+1 = 0; where, rand() is defined as a random number between 0 and 1

The function S(V i k ) is a restrictive sigmoid for achieving a new and Dmax is the maximum permissible change in a frog’s position.
change and rand is a quasi-quantity selected from a constant If this method developed improved the results, it takes the
distribution in the space of [0, 1]. However, Eqs. (11)–(13) defined place of the worst frog, or else the calculations in (16) and
the particle’s dimension limits. (17) will be repeated but with respect to the best global frog
(replaces). In this case, if no change is feasible, a new solution
1 ∝ Bi ∝ Bmax (11)
will be randomly generated to replace the frog. SFL algorithm
0 ∝ Pi ∝ Pmax (12) to optimize the location and capacity of DG to minimize losses
and boost voltage profile. The Stages are given in sequential order
as follows: The flow chart describing the projected BPSO-SLFA is
Ti = 1, 2, . . . , Tf
{ }
(13) depicted in Fig. 1.
1584 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589

Step 7: The spot and size of DGs are randomly selected in each
iteration while updating the frequency, velocity, and position.
Step 8: The main objective function is computed as in step 6.
Step 9: If the objective function attained is less than the current
best solution, then the objective function attained is considered a
new optimal solution, and the loudness and pulse emission rate
are modified.
Step 10: Repeat step 6 and step 9, before evaluating the full
iteration.
Step 11: Fix the best fitness function between all solutions and
the corresponding location, until the optimal size of the DG is
determined.
Step 12: Eventually, the base case is undergoing a comparative
review. The network output is determined by the loss of power
and stability of the voltage.

4. Numerical simulation and results

In the previous work, the implementation of the hybrid BPSO-

SLFA algorithm for optimal DG units, sizing shunt capacitor al-
location problems in the distribution network was not explored.
This inspires us to utilize a hybrid BPSO and SLFA algorithm
approach to detect the location and size of DG units in order to
reduce actual losses of power and maintain stability of voltage.
The projected algorithm is realistic in resolving optimal recon-
figuration of the distribution grid and optimum DG engagement
with the goal of reducing power losses and boosting the voltage
Fig. 1. A proposed flowchart for the algorithm.
profile. Nevertheless, the algorithm is used to solve optimal DG
placement which evaluates optimal allocation for connecting DG
and is a momentary way of resolving reconfiguration perfor-
1. Initialize to the start;
mance achieved. The results obtained clearly indicates that in
2. Generate, p solutions population for the frogs in a random
reducing power losses, scenario 4 (Multi DG implementation with
step;
reconfiguration of the network) in terms of reduction of losses
3. For each discrete situation i ∈ p: compute the robustness
of power are found to be more significant. The simulated results
(i);
obtained are correlated with other literature reports, and are
4. The solution of the population p is selected in downward
considered to be better in terms of reductions.
order of robustness
5. Split the solution p into m memeplexes; Scenario 1: The system with reconfiguration only
6. In every memeplex; The reconfiguration process of the distribution system per-
7. The best and worst frogs’ values be control; formed by using the optimization algorithm for handling the
8. Regulate the outfit location of frogs by applying Eqs. (16) operation of switches. This optimization is used to overcome
and (17); the concern over the selection of sectionalizing switches. The
9. Repeat on a number of iterations of the defined system; sectionalizing of switches has two states, one is open and another
10. End; one is close. The open and closed states of the switches are
11. integrate the develop memeplexes; defined by two different conditions such as 0 and 1.
12. Sort the population into fitness in their downward order;
Scenario 2: The system with DG units only
13. Verify if the end result is = true;
This scenario has a connection of multiple DG units (maximum
14. End;
3 DG) in the bus system without any reconfiguration (base case
scenario).
Proposed Methodology of BPSO-SLFA
Scenario 3: System with single DG unit reconfiguration and
Fig. 1 shows the flowchart for the location and size of the DG
installation
to be detected. The techniques for identifying the optimal size and
In scenario 3, there are two different steps processed. Firstly,
position of DGs using the proposed hybrid optimization algorithm
the reconfiguration is performed in the desired bus system and
are outlined in the steps below.
then only one DG unit is connected in the reconfigured bus
Step 1: Data of the input line and data of the bus.
system.
Step 2: Pick the number of DG units in the network for optimal
placement. Scenario 4: System with Multi DG unit reconfiguration and
Step 3: Fix the voltage limits with real and reactive limits of installation
power for the DGs, the DGs power factor is set according to the The reconfiguration is performed in the bus system. Multiple
reactive power compensation required. DG units (maximum 3 DG) are connected with the reconfigured
Step 4: Run the load flow analysis to compute the total active loss with standard 33 & 69 IEEE bus system.
of power and maintain voltage stability. The planned methodology (BPSO-SLFA), is applied and tested
Step 5: Arrange the algorithm parameters which include fre- on two standard distribution network, 33 and 69 IEEE-bus sys-
quency length, pulse rate, loudness and the number of iterations. tems with base voltage as 12.66 kV.
Step 6: Iteration is achieved through an efficient and hybrid From Table 1, it is clearly observed that base case results losses
optimization algorithm with objective function. for all four cases indicate scenario 4 with multiple DG deployment
 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589 1585

Table 1
The performance of the planned BPSO-SLFA on the IEEE 33 bus system at different case study levels.
Case 1 Before DG Installation After DG Installation
Power loss: 202.68 kW 75.3241 kW
Reduction of loss of power: – 62.8359%
Minimized voltage: 0.91075 pu 0.92841 pu
Case 2
Calculated DG size = 1500.0000 kW.
Reduction of loss of power:
27.9855
Case 3
loss of power: 202.68 kW 46.6899 kW
Loss reduction in power: – 76.9637%
Minimized voltage: 0.91075 pu 0.91981 pu
Case 4
Power loss: 202.68 kW 31.2844 kW
Reduction of loss of power: – 84.5646%
Minimum voltage: 0.91075 pu 0.9547 pu

Table 2
Performance of the proposed BPSO-SLFA on the IEEE 69 bus system at various level case study.
Case 1 Before DG Installation After DG Installation
Reduction in power losses: 224.9804 kW 142.2289 kW
Power loss reduction: – 36.7816%
Minimal voltage profile: 0.90919 pu 0.94184 pu
Case 2
Before DG After DG
Tie switches: 69 70 71 72 73 69 70 71 72 73
Power loss: 224.894 kW 152.1548 kW
Power loss reduction: – 32.3438%
DG size: 0.3296
Optimal DG position: 8
Case 3
Before DG After DG
Tie switches: 69 70 71 72 73 62 9 63 44 7
Power loss: 224.894 kW 42.0942 kW
Power loss reduction: – 81.2826%
DG size: 1.8147
Optimal DG position: 19
Case 4
Before DG After DG
Tie switches: 69 70 71 72 73 15 40 46 19 18
Power loss: 224.894 kW 67.841kW
Power loss reduction: – 69.8342%
DG size: 1.9246, 1.1350, 1.7587
Optimal DG position: 13, 10, 30

internal parameters of the proposed algorithm modified for the

MATLAB simulations are shown in Table 4.

4.1. Radial distribution systems of IEEE 33

The IEEE-33 bus system is a radial distribution system (RDS)

with a total load of 3.72 MW, 2.3 MVar, 33 buses and 32 branches
as shown in Fig. 2. The line loading system and line data are ob-
tained from Baran and Wu (1989). Table 1, describes the perfor-
mance of the proposed BPSO-SLFA for all the four cases scenario.
The line losses and voltage profiles of the IEEE 33 bus systems
with network reconfiguration before and after DG for all the cases
are shown in Fig. 5. It is clearly observed from the results, that
Fig. 2. IEEE standard 33 bus radial distribution system single line diagram (Baran the voltages have better-quality with multiple installations of DGs
and Wu, 1989).
and reconfiguration as shown from Fig. 1C.

4.2. IEEE 69 radial distribution systems results

with reconfiguration option is found to be more effective in
As shown in Fig. 3, the IEEE-69 bus system is a radial distri-
decreasing losses of power from 202.68 kW to 31.2844 kW. The bution system (RDS) with a total real and reactive power load of
1586 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589

Fig. 3. The IEEE 69 bus-radial network single line diagram (Sahoo and Prasad, 2006).

Table 3
Compares the results of the various methods for the test systems of the allocation of multiple DG units.
Case Approach Installed DG Power loss (kW) Loss reduction (%)
3 DG Hybrid (Kansal et al., 2016). 13, 0.79 81.050 61.62
3 DG Hybrid* proposed 13, 10, 30 31.2844 84.5646
3 DG Hybrid (HGWO) (Sanjay et al., 2017). 30,24,13 2946 72.884
3 DG Hybrid Big bang big crunch (Sedighizadeh et al., 2014). 7,9,14,28,32 139.53 78.73

Note: Hybrid* is the proposed new algorithm.

3.80 MW, 2.69 MVar, 69 buses and 68 branches. The system load Table 4
line and line data are taken from Sahoo and Prasad (2006). The Parameters of the proposed algorithm for the test systems under assessment.

performance of the proposed BPSO-SLFA for all the four (4) cases Test system

scenario are presented in Table 2. The line losses and Voltage Balanced 33 &66 bus
N = 20, dimension of search = 5, Maximum weight (wmax) = 0.9,
profiles of the standard IEEE-69 bus systems with reconfiguration
Minimum weight (wmin) = 0.4
before and after DG for all the cases are shown in Fig. 4. From Minimum average weight (wavg) kW = 00.31 × 103 , Maximum average
Fig. 4D it is observed that the voltage profile has improved with weight (wavg) kW = 00.015 × 103 population size (pop size) = 10 bMVA
multiple installations of DGs with reconfiguration and compari- = 100, bkV = 12.66 nbb = 33 & p.f = 1
son was given in Table 3, and shows the effectiveness of the new
algorithms against existing ones. Figs. 4 and 5 shows the voltage
profiles and line losses of IEEE-33 and 69 bus systems with and From Table 2, it is noticed that the Tie switches at 15, 40, 46, 19
without DG reconfiguration. From Figs. 4 and 5, it is clearly
and 18 are open at a suitable location and the size of the DG is
observed that the minimized voltage profile has improved after
given as 1.9246, 1.1350 and 1.7587 with precise optimal position
installation of multiple DG reconfiguration network. This shows
the effectiveness of the proposed algorithm. Table 2 illustrates of 13, 10 and 30 respectively. The quality, power loss reduction
the optimal network reconfiguration of all the cases under the and improvement in voltage profile for the cases are observed
study structure of the IEEE 69 bus system after the instantaneous in CASE 4, which demonstrate the preeminence of the proposed
reconfiguration of multiple DGs using the proposed technique. BPSO-SLFA.
 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589 1587

Fig. 4. (A–D) Voltage profile improvement before reconfiguration and after reconfiguration for all the four cases.

5. Conclusion system bestowing to good enactment and the preeminent results

rather than Hybrid GWO and Hybrid big bang big crunch.
In this study, the authors have utilized BPSO-SLFA algorithm 1. More benefits of the recommended (PBSO-SLFA) algorithms
to assign DG of optimum size DG in power network distribu-
have been achieved compared with other optimization
tion. It is clearly observed that the technical goal of hybrid DG
algorithm, to solve optimal sizing and placement problem.
installation and reconfiguration in scenario 4 was achieved. The
2. The power losses associated with the system, have sig-
proposed study was investigated on IEEE33 and 69-bus power
network system by DG installation at the 13th, 10th and 30th nificantly reduced up to 31.8244 kW, using multi-DGs re-
bus and succeeding in terms of reduction in network power configuration placement which is expected as part of the
losses and improvement in voltage profile. The obtained result objective of the research work.
shows that BPSO-SLFA algorithm has the best result in power loss 3. Significant improvement in voltage profile and Voltage sta-
reduction, voltage profile improvement and enhancing reliability
bility boosted with reconfiguration and installation of mul-
versus Hybrid GWO and Hybrid big bang big crunch for both
test results. Hence, the proposed methodology can be used as a tiple DG units has been achieved using controlled power
reliable method in DG setting and sizing in distribution network factor.
1588 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589

Fig. 5. (A–C) Voltage profile improvement before reconfiguration and after reconfiguration for all the cases.

CRediT authorship contribution statement of Johannesburg for Global Excellence and Stature (GES), South
Africa.
Abdurrahman Shuaibu Hassan: Investigation, Writting, Re-
view, Methodology and Editing. Yanxia Sun: Supervision, Visu- References
alization and Editing. Zenghui Wang: Supervision, Validation,
Checking codes and Proofreading. Alrashidi, M.R., Member, S., 2007. Hybrid particle swarm optimization approach
for solving the discrete OPF problem considering the Valve Loading effects.
IEEE Trans. Power Syst. 22 (4), 2030–2038, november 2007 22.
Declaration of competing interest Baran, M.E., Wu, F.F., 1989. Optimal sizing of capacitors placed on a radial
distribution system mesut. IEEE Trans. Power Deliv. 4 (1), 4, January 1989.
The authors declare that they have no known competing finan- Chiradeja, P., Ramakumar, R., 2004. An approach to quantify the technical
benefits of distributed generation. IEEE Trans. Energy Convers. 19, 764–773.
cial interests or personal relationships that could have appeared
http://dx.doi.org/10.1109/TEC.2004.827704.
to influence the work reported in this paper. Dawkins, R., 1976. The Selfish Gene. Oxford University Press, N.Y., n.d. No Title.
Doagou-mojarrad, H., Gharehpetian, G.B., Rastegar, H., Olamaei, J., 2013. Optimal
Acknowledgments placement and sizing of DG ( distributed generation ) units in distribution
networks by novel hybrid evolutionary algorithm. Energy http://dx.doi.org/
10.1016/j.energy.2013.01.043.
This research work is partially supported by grants from the El-Khattam, W., Bhattacharya, K., Hegazy, Y., Salama, M.M.A., 2004. Optimal
National Research Foundation South African with grant number investment planning for distributed generation in a competitive electricity
(No. 112108 and 112142), and reward grants from National Re- market. IEEE Trans. Power Syst. 19, 1674–1684. http://dx.doi.org/10.1109/
search Foundation the South African grant number (No. 95687 TPWRS.2004.831699.
Esmaili, M., Firozjaee, E.C., Shayanfar, H.A., 2014. Optimal placement of dis-
and 114911), grants from Tertiary Education Support Programme tributed generations considering voltage stability and power losses with
Eskom, South Africa, research grants from the URC University observing voltage-related constraints. Appl. Energy 113, 1252–1260. http:
of Johannesburg, South Africa, and grants from the University //dx.doi.org/10.1016/j.apenergy.2013.09.004.
 A.S. Hassan, Y. Sun and Z. Wang / Energy Reports 6 (2020) 1581–1589 1589

Eusuff, M.M., Lansey, K.E., 2004. The shuffled frog leaping algorithm. In: World Pepermans, G., Driesen, J., Haeseldonckx, D., Belmans, R., D’haeseleer, W., 2005.
Water Congress ASCE 2004. pp. 1–8. Distributed generation: Definition, benefits and issues. Energy Policy 33,
Eusuff, M., Lansey, K., Pasha, F., 2017. Shuffled frog-leaping algorithm : a 787–798. http://dx.doi.org/10.1016/j.enpol.2003.10.004.
memetic meta- heuristic for discrete optimization Shuffled frog-leaping Rohatgi, A., Pregelj, A., Begovic, M., 2006. Recloser allocation for improved
algorithmdoi:10. : a memetic meta-heuristic. Eng. Optim. 0273. http://dx.
reliability of DG-enhanced distribution networks. IEEE Trans. Power Syst. 21
doi.org/10.1080/03052150500384759.
(3), 1442–1449, AUGUST 2006 Recloser. 21.
Gitizadeh, M., Vahed, A.A., Aghaei, J., 2012. Multistage distribution system expan-
sion planning considering distributed generation using hybrid evolutionary Sahoo, N.C., Prasad, K., 2006. A fuzzy genetic approach for network re-
algorithms. Appl. Energy http://dx.doi.org/10.1016/j.apenergy.2012.07.010. configuration to enhance voltage stability in radial distribution systems.
Kansal, S., Kumar, V., Tyagi, B., 2016. Electrical Power and Energy Systems Energy Convers. Manage. 47 (2006), 3288–3306. http://dx.doi.org/10.1016/
Hybrid approach for optimal placement of multiple DGs of multiple types j.enconman.2006.01.004, 47, 3288–3306.
in distribution networks. Int. J. Electr. Power Energy Syst. 75, 226–235. Sambaiah, K.S., Jayabarathi, T., 2019. Optimal reconfiguration and renewable
http://dx.doi.org/10.1016/j.ijepes.2015.09.002. distributed generation allocation in electric distribution systems. Int. J.
Kefayat, M., Ara, A.L., Niaki, S.A.N., 2015. A hybrid of ant colony optimization Ambient Energy 1–29. http://dx.doi.org/10.1080/01430750.2019.1583604.
and artificial bee colony algorithm for probabilistic optimal placement and Sanjay, R., Jayabarathi, T., Raghunathan, T., Ramesh, V., Mithulananthan, N.,
sizing of distributed energy resources. Energy Convers. Manage. 92, 149–161.
2017. Optimal allocation of distributed generation using hybrid grey Wolf
http://dx.doi.org/10.1016/j.enconman.2014.12.037.
optimizer. IEEE Access 5, 14807–14818. http://dx.doi.org/10.1109/ACCESS.
Khalesi, N., Rezaei, N., Haghifam, M., 2011. Electrical Power and Energy Systems
DG allocation with application of dynamic programming for loss reduction 2017.2726586.
and reliability improvement. Int. J. Electr. Power Energy Syst. 33, 288–295. Sedighizadeh, M., Esmaili, M., Esmaeili, M., 2014. Application of the hybrid Big
http://dx.doi.org/10.1016/j.ijepes.2010.08.024. Bang-Big Crunch algorithm to optimal recon fi guration and distributed
Mallipeddi, R., Suganthan, P.N., Amaratunga, G.A.J., 2017. Generators and ca- generation power allocation in distribution systems. J. Energy Energy http:
pacitors in distribution network. http://dx.doi.org/10.1016/j.asoc.2017.07. //dx.doi.org/10.1016/j.energy.2014.09.004.
004. Selim, A., Kamel, S., Jurado, F., 2019. J. Appl. Soft Comput. J. 105938. http:
Menesy, A.S., Sultan, H.M., Selim, A., Ashmawy, M.G., 2019. Developing and //dx.doi.org/10.1016/j.asoc.2019.105938.
applying Chaotic Harris Hawks Optimization technique for extracting pa-
Singh, A.K., Parida, S.K., 2015. Electrical Power and Energy Systems Novel
rameters of several proton exchange membrane fuel cell stacks. IEEE Access
sensitivity factors for DG placement based on loss reduction and voltage
PP (1), http://dx.doi.org/10.1109/ACCESS.2019.2961811.
Moradi, M.H., Abedini, M., 2012. Electrical Power and Energy Systems A combi- improvement. Int. J. Electr. Power Energy Syst. 1–4. http://dx.doi.org/10.
nation of genetic algorithm and particle swarm optimization for optimal DG 1016/j.ijepes.2015.04.010.
location and sizing in distribution systems. Int. J. Electr. Power Energy Syst. Tagore, A.K., Gupta, A.R., 2017. Harmonic Load Flow Analysis of Radial Distri-
34, 66–74. http://dx.doi.org/10.1016/j.ijepes.2011.08.023. bution System in Presence of Distributed Generation. IEEE, pp. 147–151,
Muthukumar, K., Jayalalitha, S., 2016. Optimal placement and sizing of dis- 2017.
tributed generators and shunt capacitors for power loss minimization in Yahiaoui, A., Fodhil, F., Benmansour, K., Tadjine, M., Cheggaga, N., 2017. Grey
radial distribution networks using hybrid heuristic search optimization wolf optimizer for optimal design of hybrid renewable energy system PV-
technique. Int. J. Electr. Power Energy Syst. 78, 299–319. http://dx.doi.org/ Diesel Generator-Battery: Application to the case of Djanet city of Algeria.
10.1016/j.ijepes.2015.11.019.
Sol. Energy 158, 941–951. http://dx.doi.org/10.1016/j.solener.2017.10.040.
Nguyen, T.T., Truong, A.V., Phung, T.A., 2016. A novel method based on adaptive
Cuckoo search for optimal network reconfiguration and distributed genera-
tion allocation in distribution network. Int. J. Electr. Power Energy Syst. 78,
801–815. http://dx.doi.org/10.1016/j.ijepes.2015.12.030.