See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/362027641

Multi-objective optimal power flow based gray wolf optimization method

Article in Electrical Engineering & Electromechanics · July 2022

DOI: 10.20998/2074-272X.2022.4.08

CITATIONS READS

2 68

3 authors, including:

Nabil Mezhoud Bilel Ayachi

Université 20 août 1955-Skikda Université 20 août 1955-Skikda
19 PUBLICATIONS 33 CITATIONS 11 PUBLICATIONS 11 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Nabil Mezhoud on 26 November 2022.

The user has requested enhancement of the downloaded file.

UDC 621.3 https://doi.org/10.20998/2074-272X.2022.4.08

N. Mezhoud, B. Ayachi, M. Amarouayache

Multi-objective optimal power flow based gray wolf optimization method

Introduction. One of predominant problems in energy systems is the economic operation of electric energy generating systems. In
this paper, one a new evolutionary optimization approach, based on the behavior of meta-heuristic called grey wolf optimization is
applied to solve the single and multi-objective optimal power flow and emission index problems. Problem. The optimal power flow
are non-linear and non-convex very constrained optimization problems. Goal is to minimize an objective function necessary for a
best balance between the energy production and its consumption, which is presented as a nonlinear function, taking into account of
the equality and inequality constraints. Methodology. The grey wolf optimization algorithm is a nature inspired comprehensive
optimization method, used to determine the optimal values of the continuous and discrete control variables. Practical value. The
effectiveness and robustness of the proposed method have been examined and tested on the standard IEEE 30-bus test system with
multi-objective optimization problem. The results of proposed method have been compared and validated with hose known
references published recently. Originality. The results are promising and show the effectiveness and robustness of proposed
approach. References 35, tables 3, figures 6.
Keywords: optimization, power networks, optimal power flow, emission index, grey wolf optimization.

Вступ. Однією з головних проблем енергетичних системах є економічна експлуатація систем виробництва електроенергії.
У цій статті один новий підхід до еволюційної оптимізації, заснований на поведінці метаевристики, яка називається
оптимізацією сірого вовка, застосовується для вирішення одно- та багатокритеріальних завдань оптимального потоку
потужності та індексу викидів. Проблема. Оптимальний потік потужності - це нелінійні та неопуклі задачі оптимізації з
дуже обмеженнями. Метою є мінімізація цільової функції, необхідної для найкращого балансу між виробництвом та
споживанням енергії, яка представлена у вигляді нелінійної функції з урахуванням обмежень рівності та нерівності.
Методологія. Алгоритм оптимізації сірого вовка - це натхненний природою комплексний метод оптимізації, що
використовується для визначення оптимальних значень безперервних і дискретних змінних, що управляють. Практична
цінність. Ефективність та надійність запропонованого методу були перевірені та протестовані на стандартній 30-
шинній тестовій системі IEEE із завданням багатокритеріальної оптимізації. Результати запропонованого методу були
зіставлені та підтверджені нещодавно опублікованими відомими посиланнями. Оригінальність. Результати є
багатообіцяючими та показують ефективність та надійність запропонованого підходу. Бібл. 35, табл. 3, рис. 6.
Ключові слова: оптимізація, енергетичні мережі, оптимальний потік потужності, індекс викидів, оптимізація методом
сірого вовка.

Introduction. The optimal power flow (OPF) wolf optimizer (GWO) [30, 31]. Variants of these
problem has a long history of development of more than algorithms were proposed to handle multi-objective
60 years. Since the OPF problem was first discussed by functions in electric power systems.
Carpenter in 1962, then formulated by Dommel and The proposed GWO approach is tested and
Tinney in 1968 [1]. illustrated by numerical examples based on IEEE 30-bus
Power plants coal-fired contribute a large quantity of test system.
polluting gases to the atmosphere, as they produce large Problem formulation. The OPF and EI are
amounts of carbon oxides CO2 and some toxic and nonlinear optimization problems, represented by a
dangerous gases such as emissions of sulfur oxides SOx, predefined objective function f, subject to a set of equality
and nitrogen oxides NOx [1, 2]. and inequality constraints [27, 32]. Generally, these
Over the past few years, various methods have been problems can be expressed as follows:
implemented to solve the OPF and emission index (EI) min f ( x, u ) , (1)
problems such as: quadratic programming method (QP)
subject to
[3], Newton and quasi-Newton methods [4, 5], linear and
h ( x, u )  0 ; (2)
non-linear programming methods [6, 7], and nonlinear
internal point methods (IPM) [8]. g ( x, u )  0 ; (3)
Several methods of optimization are formulated in the xmin  x  xmax and umin  u  umax , (4)
last two decades such as: artificial bee colony (ABC) [9],
where f(x, u) is a scalar objective function to be optimized;
bacterial foraging algorithms (BFA) [10], artificial neutral
and g(x, u) are, respectively, the set of nonlinear equality
networks (ANN) [11], harmony search (HS) [12], Cuckoo
constraints represented by the load flow equations and
search algorithm (CSA) [13], evolution programming (EP)
inequality constraints consists of state variable limits and
[14], differential evaluation (DE) [15], tabu search (TS)
[16], simulated annealing (SA) [17], gravitational search functional operating constraints; x and u are the state and
algorithms (GSA) [18], genetic algorithms (GA) [19], control variables vectors respectively; xmin, xmax, umin, umax
particle swarm optimization (PSO) [20], ant colony are the acceptable limits of the variables.
Hence, x and u can be expressed as given
 
optimization (ACO) [21], firefly algorithm (FFA) [22],
sine-cosine algorithm (SCA) [23], modified imperialist x t  PG1 , VL1 ,...VLnL , QG1 ,...QGng , S1 ,..., S nbr , (5)
competitive algorithm (MICA) [24], moth swarm algorithm
where PG, QG, VL, and Sk are the generating active power
(MSA) [25], electromagnetism-like mechanism method
at slack bus, reactive power generated by all generators,
(ELM) [26], wind driven optimization (WDO) method
magnitude voltage of all load buses and apparent power
[27], machine learning [28], teaching-learning-studying-
based optimization algorithm [29], and more recently grey © N. Mezhoud, B. Ayachi, M. Amarouayache

Electrical Engineering & Electromechanics, 2022, no. 4 57

flow in all branches, respectively; ng, nL, and nbr are, nf nf
respectively, the total number of generators, the total
number of load buses and the total number of branches.
min F   i f i with i  0 and  i  1 , (11)
i 1 i 1
The set control parameters are represented in terms nf
of the decision vector as follows:
 i  1 and i  1 : n f , i is the weighting
 
where
u t  PG2 ,..., PGng , VG1 ,...VGng , Q1com ,...Qncom , T1 ,..., TnT ,(6) i 1
factor; nf is the number of objective function considered.
where PG is the active power generation excluding the
slack generator; VG is the generators magnitude voltage; Equality constraints. These equality constraints are
T is tap settings transformers; Qcom is the reactive power the sets of nonlinear load flow equations that govern the
compensation by shunt compensator; nT and ncom are the power system, i.e.:
total number of transformers and the total number of  Pgk  Pk  Pdk ;
 (12)
compensators units, respectively. Qgk  QComk  Qk  Qdk ,
Cost without valve-point optimization. The
objective function of cost optimization f1 of quadratic cost where Pgk and Qgk are, respectively, the scheduled active
equation for all generators as given below: and reactive power generations at bus k; Pk, Qk are the
ng ng active and reactive power injections at bus k; Pdk, Qdk,
QComk are the active and reactive power loads at bus k and
f1  min  C ( Pgk )  min  ak  bk Pgk ck Pgk2 , (7)
the reactive power compensation at bus k.
k 1 k 1
Inequalitie constraints. The inequality constraints
where f1 is the total generation cost in ($/h); Pgk and ng are
g(x, u) are represented by the system operational and
the active power output generated by the ith generator and
security limits, listed below:
the total number of generators; ak, bk, ck are the cost
 Active and reactive power generations limits:
coefficients of the generator k.
min max
Cost with valve-point optimization. Generally, Pgk  Pgk  Pgk where k  1,..., ng ; (13)
when every steam valves begins to open, the valve-point min max
shows rippling. However, the characteristics of input- Qgk  Qgk  Qgk where k  1,..., n g ; (14)
output of generation units make nonlinear and non-  Voltage magnitudes and angles limits:
smooth of the fuel costs function. To consider the valve-
Vkmin  Vk  Vkmax where k  1,..., nb ; (15)
point effect, the sinusoidal function is incorporated into
the quadratic function. Typically, this function is
 kmin   k   kmax where k  1,..., nb ; (16)
represented as follows
 Tap settings transformers limits:
 
ng
f 2  min  ak  bk Pgk  ck Pgk
2
 Tkmin  Tk  Tkmax where k  1,..., nT ; (17)
k 1 (8)  Reactive power compensation limits:
 
min
 d k sin ek Pgk  Pgk ,  min
QComk  QComk  Qcomkmax
where k  1,....., nCom ; (18)
where dk and ek are the cost coefficients of unit with  Security constraint limits:
valve-point effect. Skj  Skjmax where k  j  1,....., nb , (19)
Active power loss optimization. The active power
loss function f3 in MW to be minimized can be expressed where nT, nCom, T and QCom are the total number of
as follows: transformers, the total number of compensator, the
transformers tap settings, the reactive power
 
nb
f 3   Gkj Vk2  V j2  2VkV j cos  kj , (9) compensation; S kjmax is the maximum apparent power
k 1
between buses k and j.
where Vk and Vj are the magnitude voltage at buses k and Grey wolf optimization (GWO) is a typical swarm-
j, respectively; Gkj is the conductance of line kj; kj is the intelligence based meta-heuristic algorithm proposed by
voltage angle between buses k and j; nb is total number of Mirjalili et al. in 2014 [33] which is inspired from the
buses. leadership hierarchy and hunting mechanism of Grey
Emission optimization. The emission function is Wolves in nature. In nature, Gray Wolf (Canis lupus)
the sum of exponential and quadratic functions of real belongs to Canidae family. It is considered as a top level
power generating. Using a quadratic equation, emission of of predators and residing at the top in the food chain.
harmful gases is calculated in (ton/h) as given below The population hierarchies of grey wolves are
ng
10 2  k   k Pgk   k Pgk2 
separated by 4 layers which are named as, alpha () is the
f 4  min
(10) fittest solution. Beta () is the second optimum solution and
k 1
delta () is the third one. Omega () is the candidate

  k exp k Pgk , solutions that are left over [30]. Generally, the populations of
where f4 is the emission function in (ton/h); k, k, k, k, grey wolves have average crowd size of 5-12 and the cluster
k are the emission coefficients of the generator k. organizes compactly through the hierarchy [30].
All multi-objective functions using aggregation The position of the wolves is considered as the
weighting function. The function used in the case of variables to be optimized and the distance between prey
weighted aggregation is given as and grey wolves determine the fitness value of the

58 Electrical Engineering & Electromechanics, 2022, no. 4

objective function. The movement of each individual is The best position of grey wolf is calculated taking
influenced by 4 processes, namely [30]: average sum of positions and given as
1. Searching for prey (exploration);   
2. Encircling prey;  X1  X1  X1
3. Hunting; X (t 1)  . (27)
3
4. Attacking prey (exploitation). D. Attacking prey. The grey wolves stop the
The following sub-section explained these operators. hunting by attacking the prey when it stops moving. It
A. Social hierarchy. The grey wolves diverge from 
each other position for searching a victim. Make use of depends on the value of a. AM is a random value in the

AM with random values to compel the search agent to interval [–2a, 2a]. In GWO, search agents update their
 positions based on the location of ,  and  and attack
diverge from the victim. The component CM provides towards the prey [32, 33]. However, GWO algorithm is
random weights for searching prey in the search space. prone to stagnation in local solutions with these operators.
B. Encircling prey. As mentioned above, grey wolves It is true that the encircling mechanism proposed shows
encircle prey during the hunt. ,  and  estimate the position exploration to some extent, but GWO needs more
of the 3 best wolves and other wolves updates their positions operators to emphasize exploration [33, 34].
using the positions of these 3 best wolves. Encircling Simulation and results. The 5 generators system,
 IEEE 30-bus system is used throughout this work to test
behavior can be represented by DM . When the wolves do the proposed algorithm. This system consist 30 buses, 6
hunting, they tend to encircle their prey. The following generators units and 41 branches, 37 of them are the
equations depicted the encircling behavior [33, 34]. transmissions lines and 4 are the tap changing
    transformers. One of these buses is chosen like as a
DM  CM X P (t )  X (t ) ; (20) reference bus (slack bus), the buses containing generators
are taken the PV buses, the remaining buses are the PQ
    buses or loads buses. It is assumed that 9 capacitors
X (t 1)  X P (t )  AM . DM , (21) compensation is available at buses 10, 12, 15, 17, 20, 21,
 23, 24 and 29. The network data, the cost and emission
where t is the current iteration; X is the position vector coefficients of the five generators are referred in [35]. The
 
of gray wolf; X P is the position of the prey; AM and one-line diagram IEEE 30-bus system is shown in Fig. 1.

CM are the coefficient vectors calculated using the
following expressions [30, 33]:
     
AM  2 a . r 1  a and CM  2 r2 , (22)
 
where r1 and r 2 are random vectors between 0 and 1

and a is set to decreased from 2 to 0 over the course of
iterations. The 3 best solutions so far are saved and then
the other search agents (omega wolves) update their
positions according to the current best position [31, 34].
C. Hunting. Conservation of regional habitat
connectivity has the potential to facilitate recovery of the
grey wolf. After encircling,  wolf guides for hunting.
Later,  and  wolves join in hunting [33]. It is tough to
predict about the optimum location of prey. These
situations are expressed in the following expressions [33]: Fig. 1. One-line diagram of IEEE 30-bus system
   
DM   C M  X  (t )  X (t ) ; (23) The total loads of active and reactive powers are
283.4 MW and 126.2 MVAr, respectively, with 24 control
    variables. The basis apparent power used in this paper is
DM  CM  X  (t )  X (t ) ; (24) 100 MVA. The simulation results of load flow problem of
test system are summarized in Table 1.
    A. Case 1: Cost optimization without valve-point
DM  CM  X  (t )  X (t ) ; (25) effect. The cost function f1 given in (7) is optimized.
Therefore, in this case, the cost has resulted in 801.65 $/h,
   
X1  X   AM 1 DM ; which is considered 8.301 % lower than the initial case
(load flow). Figure 2 shows the convergence of cost using
   
GWO algorithm. Table 1 summarizes the optimal control
X1  X   AM 2 DM ; (26)
variables of this case.
    B. Case 2: Cost optimization with valve-point
X1  X   AM 3 DM . effect. The cost function f2 is optimized. Therefore, in this

Electrical Engineering & Electromechanics, 2022, no. 4 59

case, the cost has resulted in 836.73 $/h, which is Objective function
700
considered 4.288 % lower than the initial case. The Cost w/o valve-point effect
Cost with valve-point effect
convergence characteristic of cost for this case is 690

introduced in Fig. 2. The optimal control variables of this

case are presented in Table 1. 680

C. Case 3: Active power loss optimization. The

670
optimal control variables of this case are introduced in
Table 1. Figure 3 shows the trend for convergence 660
characteristics of active power losses using GWO
algorithm. The active power loss minimization has 650
dramatically decreased to 5.072 MW.
Table 1 640

Results of case 1, 2 and 3 for test system Iterations

Optimal values 630
0 20 40 60 80 100 120 140 160 180 200
Control variables
Base Case 1 Case 2 Case 3 Fig. 2. Convergence algorithm for cases 1 and 2
PG2, MW 40 46.53 36.57 66.930 Losses optimization, MW
0.9
PG5, MW 0 21.71 17.06 50 Losses w/o valve-point effect
PG8, MW 0 18.36 18.44 13.533 Losses with valve-point effect
0.8
PG11, MW 0 15.03 12.64 22.466
PG13, MW 0 15.26 12.45 29.854 0.7
V1, pu 1.060 1.085 1.087 1.071
V2, pu 1.045 1.066 1.064 1.061 0.6
V5, pu 1.050 1.035 1.032 1.040
V8, pu 1.070 1.038 1.036 1.040 0.5
V11, pu 1.090 1.088 1.047 1.068
V13, pu 1.090 1.022 1.027 1.064 0.4
Qcom10, MVAr 0 2.372 1.185 2.083
Qcom12, MVAr 0 0.330 4.804 2.198 0.3

Qcom15, MVAr 0 3.462 3.158 0.934 Iterations

Qcom17, MVAr 0 1.139 4.612 1.319 0.2
0 20 40 60 80 100 120 140 160 180 200
Qcom20, MVAr 0 1.667 3.320 0.864
Fig. 3. Convergence algorithm for case 3
Qcom21, MVAr 0 2.321 2.095 1.756
Qcom23, MVAr 0 1.962 2.136 1.516 Emission optimization, ton/h
3
Qcom24, MVAr 0 4.765 3.672 1.586 10 Search Agents
20 Search Agents
Qcom29, MVAr 0 3.180 2.985 3.012 30 Search Agents
2.5
T6-9 0.978 1.046 1.000 0.985 40 Search Agents
T6-10 0.969 0.971 0.995 0.975 50 Search Agents

T4-12 0.966 0.974 0.996 0.991 2

T27-28 0.932 0.993 0.999 0.973

Cost, $/h 874.22 801.65 836.73 – 1.5

Losses, MW 17.56 – – 5.072

Emission, ton/h 4.100 – – – 1
Slack, MW 260.96 175.43 196.4 105.687
CPU time, s 19.820 79.710 83.77 91.791 0.5

D. Case 4: Emission optimization. In this case, the

emission reduction yielded 0.215 ton/h. The optimal 0
Iterations
0 50 100 150 200 250
control variables settings for this case are detailed in
Fig. 4. Convergence algorithm for case 4
Table 1. The convergence characteristics of emission is
shown in Fig. 4. 840
Objective functions
E. Case 5: Cost and active loss optimization. The 820
Cost
Cost
and power loss optimization
and power loss with valve-point effect

control variables of this case are tabulated in detail in 800

Cost
Cost
and emission optimization
and emission with valve-point effect
Table 2. The cost and the power losses has resulted in 780
814.45 $/h and 7.4 MW, respectively. The convergence
760
result of this case is presented in Fig. 5.
F. Case 6: Cost and emission optimization. The 740

control variables of this case are tabulated in detail in 720

Table 2. The cost and emission has resulted, respectively, 700

in 801.88 $/h and 0.267 ton/h. Figure 6 shows the 680

convergence algorithm obtained in case 5. 660

G. Case 7: Cost, active power loss and emission. 640

Iterations
The control variables of this case are presented in detail in 0 20 40 60 80 100 120 140 160 180 200

Table 2. The cost optimization obtained in this case is Fig. 5. Convergence algorithm for cases 5 and 6
presented in Fig. 2.

60 Electrical Engineering & Electromechanics, 2022, no. 4

 880
Objective functions Table 3
Multi-obective result f1+f3+f4 Comparison of obtained results for cases 5, 6 and 7
Multi-obective result f2+f3+f4
860
Methods Cost, Losses, Emission,
840 Methods Reference $/h MW $/ton
820
Case 5
Proposed – 814.45 7.40 0.2524
800 MSA [25] 859.191 4.540 –
780
ABC [9] 854.913 4.982 –
PSO [20] 878.873 7.810 –
760
DE [15] 820.880 5.594 –
740
Case 6
Proposed – 801.88 – 0.267
720
GA [19] 820.166 – 0.271
700
Iterations MICA [24] 865.066 – 0.222
0 20 40 60 80 100 120 140 160 180 200
Case 7
Fig. 6. Convergence algorithm for case 7 –
Proposed 823.00 6.038 0.227
Table 2 GA [19] 793.605 8.450 0.187
Results of cases 4, 5, 6 and 7 for test system IABC [9] 851.611 4.873 0.223
Optimal values ABC [9] 854.916 4.982 0.228
Control variables
Case 4 Case 5 Case 6 Case 7 DE [15] 867.980 5.563 0.266
PG2, MW 76.762 60.385 47.081 53.489 obtained results validate the advantage of the proposed
PG5, MW 50 26.084 20.674 30.009 approach over many other methods used to solve the optimal
PG8, MW 26.991 15.136 21.764 34.998 power flow in terms of solution quality. It is concluded that
PG11, MW 30 20.436 13.838 18.426
the proposed method has the ability to obtain near global
PG13, MW 40 23.063 15.590 23.746
solution with stable convergence characteristics. Thus, the
V1, pu 1.042 1.078 1.083 1.073
may be recommended the proposed approach as a promising
V2, pu 1.032 1.064 1.065 1.060
algorithm for solving some more complex engineering
V5, pu 1.003 1.034 1.033 1.032
problems. The versatility of optimization is illustrated by
V8, pu 0.999 1.038 1.040 1.039
various tests by changing the parameters of proposed
V11, pu 1.004 1.098 1.069 1.082
approach such as number of population size and control
V13, pu 1.011 1.049 1.045 1.051
Qcom10, MVAr 2.887 3.674 2.488 2.286
parameter 0 coefficient. The simulation results
Qcom12, MVAr 2.193 3.143 1.277 1.414 demonstrated the effectiveness and robustness of the
Qcom15, MVAr 1.092 2.047 2.774 1.749 proposed methodology.
Qcom17, MVAr 1.771 2.508 1.688 4.259 Conflict of interest. The authors declare that they
Qcom20, MVAr 3.213 2.539 2.294 2.561 have no conflicts of interest.
Qcom21, MVAr 2.972 1.584 1.297 3.274 REFERENCES
Qcom23, MVAr 3.749 1.330 3.604 1.828 1. Dommel H., Tinney W. Optimal Power Flow Solutions.
Qcom24, MVAr 3.506 4.274 1.192 2.970 IEEE Transactions on Power Apparatus and Systems, 1968, vol.
Qcom29, MVAr 3.247 0.313 2.277 2.971 PAS-87, no. 10, pp. 1866-1876. doi:
T6-9 1.078 1.036 1.041 1.010 https://doi.org/10.1109/TPAS.1968.292150.
T6-10 0.939 0.940 0.922 0.995 2. Talaq J.H., El-Hawary F., El-Hawary M.E. A summary of
environmental/economic dispatch algorithms. IEEE
T4-12 1.006 0.971 0.974 0.994
Transactions on Power Systems, 1994, vol. 9, no. 3, pp. 1508-
T27-28 0.924 0.980 0.973 0.982 1516. doi: https://doi.org/10.1109/59.336110.
Cost, $/h – 814.45 801.88 823.00 3. Nicholson H., Sterling M.H. Optimum Dispatch of Active and
Losses, MW – 7.40 – 6.038 Reactive Generation by Quadratic Programming. IEEE
Emission, ton/h 0.215 – 0.267 0.227 Transactions on Power Apparatus and Systems, 1973, vol. PAS-92,
Slack, MW 63.681 145.69 173.28 128.768 no. 2, pp. 644-654. doi: https://doi.org/10.1109/TPAS.1973.293768.
CPU time, s 74.987 81.601 86.01 99.374 4. Santos A.Jr., da Costa G.R.M. Optimal-power-flow solution
by Newton’s method applied to an augmented Lagrangian
For the IEEE-30 bus system, 24 control variables function. IEE Proceedings - Generation, Transmission and
(5 generators excluding slack bus, 6 generators magnitude Distribution, 1995, vol. 142, no. 1, pp. 33-36. doi:
voltages, 4 transformers taps and 9 reactive powers https://doi.org/10.1049/ip-gtd:19951586.
compensators) were optimized. Tables 3 shows a 5. Giras T.C., Talukdar S.N. Quasi-Newton method for
comparison between the obtained results. optimal power flows. International Journal of Electrical Power
& Energy Systems, 1981, vol. 3, no. 2, pp. 59-64. doi:
Conclusions. In this paper, the grey wolf optimization https://doi.org/10.1016/0142-0615(81)90010-7.
approach is implemented and applied successfully to solve 6. Stott B., Hobson E. Power System Security Control
the multi-objective optimal power flow. The obtained results Calculations Using Linear Programming, Part I. IEEE Transactions
with proposed method in all cases are much better. on Power Apparatus and Systems, 1978, vol. PAS-97, no. 5, pp.
Therefore, in the multi-objective case, taking into account 1713-1720. doi: https://doi.org/10.1109/TPAS.1978.354664.
generation cost, the active power losses optimization and 7. Sasson A. Nonlinear Programming Solutions for Load-Flow,
Minimum-Loss, and Economic Dispatching Problems. IEEE
emission optimization all results were significantly Transactions on Power Apparatus and Systems, 1969, vol. PAS-88,
decreased to 823 $/h, 6.038 MW and 0.227 ton/h, which are no. 4, pp. 399-409. doi: https://doi.org/10.1109/TPAS.1969.292460.
considered 5.85 %, 61.61 % and 44.63 %, respectively, 8. Capitanescu F., Glavic M., Ernst D., Wehenkel L. Interior-point
lower than the initial case (load flow). With comparison, the based algorithms for the solution of optimal power flow problems.

Electrical Engineering & Electromechanics, 2022, no. 4 61

 Electric Power Systems Research, 2007, vol. 77, no. 5–6, pp. 508- considering the cost, emission, voltage deviation and power
517. doi: https://doi.org/10.1016/j.epsr.2006.05.003. losses using multi-objective modified imperialist competitive
9. Mouassa S., Bouktir T. Artificial Bee Colony Algorithm for algorithm. Energy, 2014, vol. 78, pp. 276-289. doi:
Solving OPF Problem Considering the Valve Point Effect. https://doi.org/10.1016/j.energy.2014.10.007.
International Journal of Computer Applications, 2015, vol. 112, no. 25. Mohamed A.-A.A., Mohamed Y.S., El-Gaafary A.A.M.,
1, pp. 45-53. Available at: Hemeida A.M. Optimal power flow using moth swarm
https://www.ijcaonline.org/archives/volume112/number1/19634- algorithm. Electric Power Systems Research, 2017, vol. 142, pp.
1208 (accessed 23 May 2021). 190-206. doi: https://doi.org/10.1016/j.epsr.2016.09.025.
10. Zakaria Z., Rahman T.K.A., Hassan E.E. Economic load 26. El-Hana Bouchekara H.R., Abido M.A., Chaib A.E. Optimal
dispatch via an improved Bacterial Foraging Optimization. 2014 Power Flow Using an Improved Electromagnetism-like Mechanism
IEEE 8th International Power Engineering and Optimization Method. Electric Power Components and Systems, 2016, vol. 44, no.
Conference (PEOCO2014), 2014, pp. 380-385. doi: 4, pp. 434-449. doi: https://doi.org/10.1080/15325008.2015.1115919.
https://doi.org/10.1109/PEOCO.2014.6814458. 27. Mezhoud N., Ayachi B., Bahri A. Wind Driven
11. Lai L.L. Intelligent System Applications in Power Optimization Approach based Multi-objective Optimal Power
Engineering: Evolutionary Programming and Neural Networks. Flow and Emission Index Optimization. International Research
Wiley Publ., 1998. 286 p. Journal of Multidisciplinary Technovation, 2022, pp. 21-41. doi:
12. Khazali A.H., Kalantar M. Optimal reactive power dispatch https://doi.org/10.54392/irjmt2223.
based on harmony search algorithm. International Journal of 28. Hasan F., Kargarian A., Mohammadi A. A Survey on
Electrical Power & Energy Systems, 2011, vol. 33, no. 3, pp. Applications of Machine Learning for Optimal Power Flow.
684-692. doi: https://doi.org/10.1016/j.ijepes.2010.11.018. 2020 IEEE Texas Power and Energy Conference (TPEC), 2020,
13. Abd Elazim S.M., Ali E.S. Optimal Power System pp. 1-6. doi: https://doi.org/10.1109/TPEC48276.2020.9042547.
Stabilizers design via Cuckoo Search algorithm. International 29. Akbari E., Ghasemi M., Gil M., Rahimnejad A., Andrew
Journal of Electrical Power & Energy Systems, 2016, vol. 75, Gadsden S. Optimal Power Flow via Teaching-Learning-
pp. 99-107. doi: https://doi.org/10.1016/j.ijepes.2015.08.018. Studying-Based Optimization Algorithm. Electric Power
14. Yuryevich J., Kit Po Wong. Evolutionary programming Components and Systems, 2021, vol. 49, no. 6–7, pp. 584-601.
based optimal power flow algorithm. IEEE Transactions on doi: https://doi.org/10.1080/15325008.2021.1971331.
Power Systems, 1999, vol. 14, no. 4, pp. 1245-1250. doi: 30. Mohamed A.-A.A., El-Gaafary A.A.M., Mohamed Y.S.,
https://doi.org/10.1109/59.801880. Hemeida A.M. Multi-objective Modified Grey Wolf Optimizer
15. Abou El Ela A.A., Abido M.A., Spea S.R. Optimal power for Optimal Power Flow. 2016 Eighteenth International Middle
flow using differential evolution algorithm. Electric Power East Power Systems Conference (MEPCON), 2016, pp. 982-
Systems Research, 2010, vol. 80, no. 7, pp. 878-885. doi: 990. doi: https://doi.org/10.1109/MEPCON.2016.7837016.
https://doi.org/10.1016/j.epsr.2009.12.018. 31. Dilip L., Bhesdadiya R., Trivedi I., Jangir P. Optimal Power
16. Abido M.A. Optimal Power Flow Using Tabu Search Algorithm.
Flow Problem Solution Using Multi-objective Grey Wolf
Electric Power Components and Systems, 2002, vol. 30, no. 5, pp.
Optimizer Algorithm. In: Hu Y.C., Tiwari S., Mishra K., Trivedi
469-483. doi: https://doi.org/10.1080/15325000252888425.
M. (eds) Intelligent Communication and Computational
17. Jeon Y.-J., Kim J.-C. Application of simulated annealing
Technologies. Lecture Notes in Networks and Systems, 2018,
and tabu search for loss minimization in distribution systems.
International Journal of Electrical Power & Energy Systems, vol. 19. Springer, Singapore. doi: https://doi.org/10.1007/978-
2004, vol. 26, no. 1, pp. 9-18. doi: 981-10-5523-2_18.
https://doi.org/10.1016/S0142-0615(03)00066-8. 32. Ayachi B., Boukra T., Mezhoud N. Multi-objective optimal
18. Bhowmik A.R., Chakraborty A.K. Solution of optimal power flow considering the multi-terminal direct current.
power flow using nondominated sorting multi objective Electrical Engineering & Electromechanics, 2021, no. 1, pp. 60-
gravitational search algorithm. International Journal of 66. doi: https://doi.org/10.20998/2074-272X.2021.1.09.
Electrical Power & Energy Systems, 2014, vol. 62, pp. 323-334. 33. Pradhan M., Roy P.K., Pal T. Grey wolf optimization
doi: https://doi.org/10.1016/j.ijepes.2014.04.053. applied to economic load dispatch problems. International
19. Yaşar C., Özyön S. A new hybrid approach for nonconvex Journal of Electrical Power & Energy Systems, 2016, vol. 83,
economic dispatch problem with valve-point effect. Energy, pp. 325-334. doi: https://doi.org/10.1016/j.ijepes.2016.04.034.
2011, vol. 36, no. 10, pp. 5838-5845. doi: 34. El-Fergany A.A., Hasanien H.M. Single and Multi-objective
https://doi.org/10.1016/j.energy.2011.08.041. Optimal Power Flow Using Grey Wolf Optimizer and
20. Abido M.A. Optimal power flow using particle swarm Differential Evolution Algorithms. Electric Power Components
optimization. International Journal of Electrical Power & and Systems, 2015, vol. 43, no. 13, pp. 1548-1559. doi:
Energy Systems, 2002, vol. 24, no. 7, pp. 563-571. doi: https://doi.org/10.1080/15325008.2015.1041625.
https://doi.org/10.1016/S0142-0615(01)00067-9. 35. Lee K.Y., El-Sharkawi M.A. (Eds.). Modern Heuristic
21. Ketabi A., Alibabaee A., Feuillet R. Application of the ant Optimization Techniques. John Wiley & Sons, Inc., 2008. 586 p.
colony search algorithm to reactive power pricing in an open doi: https://doi.org/10.1002/9780470225868.
electricity market. International Journal of Electrical Power &
Energy Systems, 2010, vol. 32, no. 6, pp. 622-628. doi: Received 17.04.2022
https://doi.org/10.1016/j.ijepes.2009.11.019. Accepted 25.05.2022
22. Padaiyatchi S.S. Hybrid DE/FFA algorithm applied for different Published 20.07.2022
optimal reactive power dispatch problems. Australian Journal of
Electrical and Electronics Engineering, 2020, vol. 17, no. 3, pp. Nabil Mezhoud1, PhD of Power Engineering,
203-210. doi: https://doi.org/10.1080/1448837X.2020.1817233. Bilel Ayachi1, PhD of Power Engineering,
23. Attia A.-F., El Sehiemy R.A., Hasanien H.M. Optimal Mohamed Amarouayache1, PhD of Power Engineering,
power flow solution in power systems using a novel Sine-Cosine 1
Electrotechnical Laboratory Skikda (LES),
algorithm. International Journal of Electrical Power & Energy Electrical Engineering Department, Faculty of Technology,
Systems, 2018, vol. 99, pp. 331-343. doi: University 20 August 1955, Skikda, Algeria,
https://doi.org/10.1016/j.ijepes.2018.01.024. e-mail: [email protected] (Corresponding author);
24. Ghasemi M., Ghavidel S., Ghanbarian M.M., Gharibzadeh [email protected]; [email protected]
M., Azizi Vahed A. Multi-objective optimal power flow
How to cite this article:
Mezhoud N., Ayachi B., Amarouayache M. Multi-objective optimal power flow based gray wolf optimization method. Electrical
Engineering & Electromechanics, 2022, no. 4, pp. 57-62. doi: https://doi.org/10.20998/2074-272X.2022.4.08

62 Electrical Engineering & Electromechanics, 2022, no. 4

View publication stats