Hardiansyah HARDIANSYAH ,Junaidi JUNAIDI

Faculty of Engineering, University of Tanjungpura, Pontianak 78124, Indonesia

doi:10.15199/48.2021.02.32

Multi-Objective Ant Lion Optimizer for Solving

Environmental/Economic Dispatch
Abstract. In this paper, a new meta-heuristic algorithm, called multi-objective ant lion optimizer (MOALO) is presented to solve environmental
economic dispatch (EED) problem considering transmission losses. MOALO is inspired by the hunting mechanism of ant lions in nature. It has fast
convergence speed due to the use of roulette wheel selection technique. The effectiveness of the proposed algorithm has been tested on the
standard IEEE 30-bus test system and the results were compared with other methods reported in recent literature. The simulation results show that
the proposed algorithm outperforms previous optimization methods.

Streszczenie. Przedstawiono nowy meta-heurystyczny algorytm MOALO do rozwiązywania problemu ekonomicznego rozsyłu energii z
uwzględnieniem warunków środowiskowych. Algorytm jest inspirowany mechanizmem polowania. Daje on szybkie rozwiązanie z wykorzystaniem
zasady koła w ruletce. Algorytm sprawdzono wykorzystując system testowy IEEE-30-bus. (Wieloobiektowy algorytm mrowkolwowaty do
rozwiązywania problemu ekonomicznego rozsyłu energii z uwzględnieniem środowiska)

Keywords: multi-objective ant lion optimizer, economic dispatch, emission dispatch, combined economic emission dispatch.
Słowa kluczowe: ekonomiczny rozsył energii, owady mrówkolwaowate, emisja zanieczyszczeń.

Introduction In this paper, MOALO algorithm has been used to solve

Optimization of the modern power system plays a major the CEED problem considering transmission loss.
role in thermal power plants energy production. The Combined economic emission dispatch (CEED) solution
challenges of the engineers are to optimize the real power which was performed using MOALO algorithm was tested
of the generating units and to minimize the fuel cost of the on the standard IEEE 30-bus 6-generator test system. The
power plant. Economic dispatch (ED) is one of the most results were compared to those reported in the literature.
fundamental issues in operation and control of power
systems to allocate generations among the committed units. Problem Formulations
The main goal of the ED problem is to determine the The CEED problem targets to find the optimal
amount of real power contributed by online thermal combination of load dispatch of generating units and
generators satisfying load demand at any time subject to minimizes both fuel cost and emission while satisfying the
unit and system constraints so as the total generation cost total power demand. Therefore, CEED consists of two
is minimized. Therefore, it is very important to solve the objective functions, which are economic and emission
problem as quickly and precisely as possible [1, 2]. dispatches. Then these two functions are combined to solve
Therefore, recently most of the researchers made studies the problem. The CEED problem can be formulated as
for finding the most suitable power values produced by the follows [11]:
generators depending on fuel costs. In these studies, they (1) FT  Min f FC , EC  
produced successful results by using various optimization
where FT is the total generation cost of the system, FC is
algorithms [3-5]. Despite the fact that the traditional ED can
the total fuel cost of generators and EC is the total emission
optimize generator fuel costs, it still cannot produce a
of generators.
solution for environmental pollution due to the excessive
emission of fossil fuels.
Minimization of Fuel Cost
Currently, a large part of energy production is done with
The ED problem can be formulated in a quadratic form
thermal sources. Thermal power plant is one of the most
as follows [11]:
important sources of carbon dioxide (CO2), sulfur dioxide
 
N
FC   ai Pi 2  bi Pi  ci
(SO2) and nitrogen oxides (NOx) which create atmospheric
pollution [6]. Emission control has received increasing (2)
attention owing to increased concern over environmental i 1
pollution caused by fossil based generating units and the where Pi is the power generation of the ith unit; ai, bi, and ci
enforcement of environmental regulations in recent years are fuel cost coefficients of the ith generating unit and N is
[7]. Numerous studies have emphasized the importance of the number of generating units.
controlling pollution in electrical power systems [8].
Combined economic and emission dispatch (CEED) has Minimization of Emission
been proposed in the field of power generation dispatch, The classical ED problem can be obtained by the
which simultaneously minimizes both fuel cost and pollutant amount of active power to be generated by the generating
emissions. When the emission is minimized the fuel cost units at minimum fuel cost, but it is not considered as the
may be unacceptably high or when the fuel cost is amount of emissions released from the burning of fossil
minimized the emission may be high. A number of methods fuels. Total amount of emissions such as SO2 or NOX
have been presented to solve CEED problems such as depends on the amount of power generated by until and it
multi-objective differential evolution algorithm [9], genetic can be defined as the sum of quadratic and exponential
algorithm [10-12], simulated annealing [13], biogeography- functions and can be stated as [11]:
based optimization [14], modified bacterial foraging
 
N

algorithm [15], particle swarm optimization [16-18], artificial (3) EC    i Pi 2   i Pi   i  i exp( i Pi )

bee colony algorithm [19-21], gravitational search algorithm i 1
[22], moth swarm algorithm [23], and adaptive wind driven where αi, βi, γi, ηi and δi are emission coefficients of the ith
optimization [24]. generating unit.

PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 2/2021 153

Combined Environmental Economic Dispatch (CEED) c) Ants move over the search space using random walks
CEED is a multi-objective problem, which is a around the antlions.
combination of both economic and environmental d) The population of antlions is never evaluated. In fact,
dispatches that individually make up different single antlions assumed to be on the location of ants in the first
problems. At this point, this multi-objective problem needs iteration and relocate to the new positions of ants in the
to be converted into single-objective form in order to fulfill rest of iterations if the ants become better.
optimization. The conversion process can be done by using e) There is one antlion assigned to each ant and updates
the price penalty factor [11]. However, the single-objective its position if the ant becomes fitter.
CEED can be formulated as shown in equation (4): f) There is also an elite antlion which impacts the
(4) FT  ( w  FC  (1  w)  h  EC ) movement of ants regardless of their distance.
g) If any antlion becomes better then the elite, it will be
under the following condition,
replaced with the elite.
(5) 0  w 1 h) Step b to g are repeatedly executed until the satisfaction
where w is weighting factor: w=1 (fuel cost minimization), of an end criterion.
w=0 (NOx emission minimization), and w=0.5 (CEED i) The position and fitness value of the elite antlion are
minimization) and h is the price penalty factor. returned as the best estimation for the global optimum.
The antlions maintain the best position obtained by the
Problem Constraints ants and guide the search of ants towards the promising
There are two constraints in the EED problem which are regions of the search space. In order to solve optimization
power balance constraint and maximum and minimum limits problems, the ALO algorithm mimics random walk of ants,
of power generation output constraint. entrapment in an antlion pit, cunstructing a pit, sliding ant
Active power balance equation towards antlions, catching prey and re-constructing the pit,
For power balance, an equality constraint should be and elitism.
To model such interactions, ants are required to move
satisfied. The total generated power should be the same as
over the search space and antlions are allowed to hunt
total load demand plus the total line loss.
N
them and become fitter using traps. Since ants move
(6) PD   Pi  PLoss stochastically in nature when searching for food, a random
walk is chosen for modeling ants' movement as follows:
i 1
where PD is the total load demand and PLoss is total X (t )  [0, cums(2r (t1 )  1, cums(2r (t2 )  1,,
(9)
transmission losses. The transmission losses PLoss can be cums(2r (tn )  1]
calculated by using B matrix technique and is defined by (7)
as, where cums calculates the cumulative sum and r(t) is
N N N defined as follows:
(7) PLoss   Pi Bij Pj   B0i Pi  B00 1, if rand  0.5
i 1 j 1 i 1
(10) r (t )   
where Bij is coefficient of transmission losses and the B0i 0, if rand  0.5 
and B00 is matrix for loss in transmission which are constant is a stochastic function where t shows the step of random
under certain assumed conditions. walk and rand is a random number generated with uniform
Minimum and maximum power limits distribution in the interval of [0,1].
Generation output of each generator should lie between In order to keep the random walk in the boundaries of
minimum and maximum limits. The corresponding inequality the search space and prevent the ants from avershooting,
constraint for each generator is the random walks should be nomalized using the following
equation:
(8) Pi min  Pi  Pi max for i  1,2,  , N ( X it  ai )  (d i  cit )
min max
(11) X it   ci
where Pi and Pi are the minimum and maximum (d it  ai )
outputs of the ith generator, respectively. where a i is minimum random walk of i-th variable,
Multi-Objective Ant Lion Optimization (MOALO) d i indicates the maximum random walk of i-th variable, c it is
In order to propose the multi-objective models of Ant
Lion Optimizer (ALO) algorithm [25], the fundamentals of minimum of i-th variable at t-th iteration, and d it is the
this algorithm should be discussed first. An algorithm should maximum of i-th variable at t-th iteration.
follow the same search behaviour to be considered as an ALO simulates the entrapment of ants in antlions pit by
extended version of the same algorithm. The ALO algorithm changing the random walks around antlions. The following
mimics the hunting mechanism of antlions and the equations have been proposed in this regard:
interaction of their favourite prey, ants, with them. (12) cit  Antlion tj  c t
Similarly to other population-based algorithm, ALO
approximates the optimal solutions for optimization (13) dit  Antlion tj  d t
problems with employing a set of random solutions. This set t t
is improved based on the principles inspired from the where c is the minimum of all variables at t-th iteration, d
interaction between antlions and ants. There are two indicates the vector including the maximum of all variables
populations in the ALO algorithm; set of ants and set of at t-th iteration, c it is the minimum of all variables for i-th
antlions. The general step of ALO to change these two sets
ant, d it is the maximum of all variables for i-th ant, and
and eventually estimate the global optimum for a given
optimization problem are as follows [26]: Antlion tj shows the position of selected j-th antlion at t-th
a) The ant set is initialized with random values and are the
main search agents in the ALO. iteration.
b) The fitness value of each ant is evaluated using an In nature, bigger antlions construct bigger pit to increase
objective function in each iteration. their chance of survival. In order to simulate this, ALO

154 PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 2/2021

utilizes a roulette wheel operator that selects antlions based equation is used in this regard that defines the probability of
on their fitness value. The roulete wheel assists fitter antlion removing a solution from the archive:
to attact more ants. Ni
For mimicking the sliding ants towards antlions, the (19) Pi 
boundaries of random walks should be decreased c
adaptively as follows: where c is a constant and should be greater than 1 and Ni is
the number of solutions in the vicinity of the i-th solution.
ct In order to require ALO to solve multi-objective problem,
(14) ct 
I (16) should be modified due to the nature of multiobjective
dt problems.
(15) dt 
I
(20) Antlion tj  Antit , if f ( Antit )  f ( Antlion tj )
where I is a ratio, c t is the minimum vector of all variables where t shows the current iteration, Antlion tj shows the
at t-th iteration, and d t indicates the vector including the position of selected j-th antlion at t-th iteration, and
maximum of all variables at t-th iteration. Ant it indicates the position of i-th ant for t-th iteration.
The second to last step in ALO is catching the ant and
Another modification is for the selection of random
re-constructing the pit. The following equation simulates
antlions and elite in (17). We utilize a roulette wheel and
this:
(17) to select a non-dominated solution from the archive.
(16) Antlion tj  Antit , if f ( Antit )  f ( Antlion tj ) The rest of the operator in MOALO are identical to those in
where t shows the current iteration, Antlion tj shows the ALO. After all, the pseudocodes of the MOALO algorithm
are shown in Table 1.
position of selected j-th antlion at t-th iteration, and
Ant it indicates the position of i-th ant for t-th iteration. Table 1. Pseudocodes of MOALO
Multi-Objective Ant Lion Optimization (MOALO)
The last operator in ALO is elitism, in which the fittest while the end condition is not met
antlion formed during optimization is stored. The fittest ant for every ant
lion should be able to affect the movements of all ants Select a random antlion from the archive
during iterations. It is assumed that every random walks of Select the elite using Roulette wheel from the archive
ants around a chosen antlion by the roulette wheel and the Update c and d using equations (14) & (15)
elite instantaneously as follows: Create a random walk and normalize it using equations (9) &
(11)
RAt  REt Update the position of ant using equation (17)
(17) Antit  end for
2
Calculate the fitness of all ants
where R At is random walk around antlion selected by Update the archive
if the archive is full
roulette wheel at t-th iteration, R Et is the random walk Delete some solutions using Roulette wheel and (19) from the
around the elite at t-th iteration, and Ant it indicates the archive
to accommodate new solutions.
position of i-th ant for t-th iteration. end
As mentioned in the literature review, there are different end while
approaches for finding and storing Pareto optimal solutions Return archive
using heuristic algorithms. In this work, we employ an Update elite if an ant lion become fitter than the elite
archive to store Pareto optimal solutions. Obviously, the end while
Return elite
convergence of the MOALO algorithm inherits from the ALO
algorithm. If we pick one solution from the archive, the ALO
algorithm will be able to improve its quality. However, Simulation Results and Discussion
finding the Pareto optimal solutions set with a high diversity The proposed MOALO algorithm is tested on the
is challenging. standard IEEE 30-bus power system with six-generating
To overcome this challenge, we have inspired from the units in order to investigate its effectiveness. The single-line
MOPSO algorithm and utilized the leader selection and diagram of the IEEE 30-bus test system is shown in Figure
archve maintenance. For measuring the distribution of the 1 and the detailed data are given in [21, 22]. The
solutions in the archive, we use niching. In this approach, parameters of all thermal units (generation limits, fuel cost
the vicinity of each solution is investigated considering a and NOx emission coefficients) are presented in Table 2,
pre-defined radius. The number of solutions in the vicinity is followed by B-loss coefficients are presented in Table 3.
then counted and considered as the measure of distribution. The load demand of the system is 283.4 MW. The values of
To improve the distribution of the solutions in the archive, MOALO algorithm for solving CEED problem in this paper
we considered two mechanism similarly to those in are designated as follow: the number of population size, NP
MOPSO. Firstly, the antlions areselected from the solutions = 30; and the number of iterations, maxIter = 200.
with the least populated neighbourhood. The following For the purpose of comparison with the reported results,
equation is used in this regard that defines the probability of the test system is considered for two cases as follows:
choosing a solution in the archive: Case A: In this case, we take IEEE 30-bus test system with
considering transmission losses.
c
(18) Pi  Case B: In this case with neglecting transmission losses.
Ni In the first case, the best solutions for power outputs,
where c is a constant and should be greater than 1 and Ni is fuel cost and NOx emission obtained by using MOALO for
the number of solutions in the vicinity of the i-th solution. w=1, w=0, and w=0.5 are given in Table 4. The results
Secondly, when the archive is full, the solutions with obtained by MOALO for the test system along with
most populated neighbourhood are removed from the corresponding data from the literature are summarized in
archive to accommodate new solutions. The following Table 5. As can be seen in Table 5, the MOALO provided
better values for the minimum fuel cost and NOx emission in

PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 2/2021 155

regard to the values obtained by the algorithms proposed in MOALO for w=1, w=0, and w=0.5 are given in Table 6. The
[9, 14, 16, 22, 23, 24]. Figure 2 shows the convergence results obtained by MOALO for the test system along with
characteristic of fuel cost optimization with MOALO. Figure corresponding data from the literature are summarized in
3 shows the Pareto optimal solution when fuel cost and Table 7. As can be seen in Table 7, the MOALO provided
emission optimized simultaneously. better values for the minimum fuel cost and NOx emission in
In the second case, the best solutions for power regard to the values obtained by the algorithms proposed in
outputs, fuel cost and NOx emission obtained by using [14, 16, 22, 23].

Table 2. Generation limits, fuel cost and NOx emission coefficients for IEEE 30-bus test system [21]
Unit Pi min Pi max ai bi ci αi βi γi ηi δi
1 5 150 10 200 100 4.091e-2 -5.554e-2 6.940e-2 2.0e-4 2.857
2 5 150 10 150 120 2.543e-2 -6.047e-2 5.638e-2 5.0e-4 3.333
3 5 150 20 180 40 4.258e-2 -5.094e-2 4.586e-2 1.0e-6 8.0
4 5 150 10 100 60 5.326e-2 -3.550e-2 3.380e-2 2.0e-3 2.0
5 5 150 20 180 40 4.258e-2 -5.094e-2 4.586e-2 1.0e-6 8.0
6 5 150 10 150 100 6.131e-2 -5.555e-2 5.151e-2 1.0e-5 6.667

Table 3. Transmission loss coefficients [21]

 0.1382 - 0.0299 0.0044 - 0.0022 - 0.0010 - 0.0008 

 - 0.0299 0.0487 - 0.0025 0.0004 0.0016 0.0041 

 0.0044 - 0.0025 0.0182 - 0.0070 - 0.0066 - 0.0066 
Bij   
 0.0022 0.0004 - 0.0070 0.0137 0.0050 0.0033 
 0.0010 0.0016 - 0.0066 0.0050 0.0109 0.0005 
 
 0.0008 0.0041 - 0.0066 0.0033 0.0005 0.0244
B0i   0.0107 0.0060 - 0.0017 0.0009 0.0002 0.0030 
B 00  0.00098573

29

28
27

30
25
26

23 24

15 18 19

17 20

21

14 16 22

13 12 10

11 9

1 4 6 8
3

2 5

Fig. 1. Single-line diagram of IEEE 30-bus test system [20]

Table 4. The best solutions obtained by using MOALO (Case A)

Generation (MW) Fuel Cost NOx Emission PLoss
w
P1 P2 P3 P4 P5 P6 ($/h) (ton/h) (MW)
1 12.0969 28.6312 58.3557 99.2854 52.3970 35.1899 605.99837 0.20453 2.55619
0 37.3419 50.1791 51.2265 46.6136 51.2815 50.1836 639.75214 0.18672 3.42623
0.5 23.2164 37.2784 54.3552 77.2134 52.4097 41.5729 612.02569 0.19264 2.64599

156 PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 2/2021

 630
MOALO

625
Objective function ($/h)

Emission (ton/h)
620

615

610

605
0 20 40 60 80 100 120 140 160 180 200
Iteration
Fig. 3. Fuel cost and emission optimization with MOALO (Case A)
Fig. 2. Fuel cost optimization with MOALO (Case A)

Table 5. Comparison of best solution (Case A)

Fuel cost minimization (w=1) NOx emission minimization (w=0) CEED minimization (w=0.5)
Methods Fuel cost NOx emission Fuel cost NOx emission Fuel cost NOx emission
($/h) (ton/h) ($/h) (ton/h) ($/h) (ton/h)
MODE [9] 606.41060 0.2221 643.5190 0.1942 614.1700 0.2043
MBFA [14] 607.6700 0.2198 644.4300 0.1942 616.4960 0.2002
MOPSO [16] 607.7900 0.2193 644.7400 0.1942 615.0000 0.2021
GSA [22] 605.9984 0.2207 646.2070 0.1942 612.2530 0.2036
MSA [23] 605.9984 0.2207 646.2049 0.1942 612.2519 0.2038
AWDO [24] 605.9984 0.2207 646.2070 0.1942 612.2528 0.2036
MOALO 605.99837 0.20453 639.75214 0.18672 612.02569 0.19264

Table 6. The best solutions obtained by using MOALO (Case B)

Generation (MW) Fuel Cost NOx Emission PLoss
w
P1 P2 P3 P4 P5 P6 ($/h) (ton/h) (MW)
1 10.9719 29.9766 52.4299 101.6199 52.4298 35.9720 600.11141 0.20523 -
0 36.8275 49.6415 50.7287 45.8475 50.7287 49.6261 631.98790 0.18686 -
0.5 23.7145 37.8025 51.4852 77.1837 51.4853 41.7288 606.39137 0.19250 -

Table 7. Comparison of best solution (Case B)

Fuel cost minimization (w=1) NOx emission minimization (w=0) CEED minimization (w=0.5)
Methods Fuel cost NOx emission Fuel cost NOx emission Fuel cost NOx emission
($/h) (ton/h) ($/h) (ton/h) ($/h) (ton/h)
MBFA [14] 600.17 0.2200 636.73 0.1942 610.906 0.2000
MOPSO [16] 600.12 0.2216 637.42 0.1942 608.65 0.2017
GSA [22] 600.11141 0.222145 638.27344 0.194203 606.79829 0.203289
MSA [23] 600.11141 0.22215 638.27583 0.194203 606.80105 0.20329
FFA [23] 600.11141 0.22214 638.27398 0.194203 606.79835 0.20329
PSOGSA[23] 600.11141 0.22215 638.27452 0.194203 606.79841 0.20329
MOALO 600.11141 0.20523 631.98790 0.18686 606.39137 0.19250

Conclusion Authors: The authors are with Electrical Engineering Department,

In this paper, a new approach based on multi-objective Faculty of Engineering, University of Tanjungpura, Pontianak
ant lion optimizer (MOALO) algorithm has been presented 78124, Indonesia.
E-mail: [email protected]; Assoc. Prof. Hardiansyah
and successfully applied to solve the combined economic
HARDIANSYAH;
emission dispatch problem considering transmission losses. E-mail: [email protected]; Assoc. Prof. Junaidi JUNAIDI;
The problem has been formulated as multiobjective The correspondence address is:
optimization problem with competing fuel cost and E-mail: [email protected]
environmental impact objectives. The effectiveness of
proposed algorithm is demonstrated on the standard IEEE REFERENCES
30-bus test system with six generating units. The [1] M. A. A b i d o , “Environmental/economic power dispatch using
comparison of the results obtained with other methods multiobjective evolutionary algorithms”, IEEE Transactions on
reported in the literature shows the superiority of the Power Systems, 18(2003), No. 4, 1529-1537.
proposed algorithm and its potential for solving the [2] S. Krishnamurthy, and R. Tzoneva, “Multi objective dispatch
problem with valve point effect loading of fuel cost and
combined economic emission dispatch problems in large-
emission criterion”, International Journal of Computer and
scale power systems. The results obtained from the test Electrical Engineering, 4(2012), No. 5, 775-784.
systems have indicated that the proposed technique has [3] S. Y. Lim, M. Montakhab, and H. Nouri, “Economic dispatch of
better performance in terms of minimum fuel costs and NOx power system using particle swarm optimization with
emissions than other optimization methods reported in the constriction factor”, International Journal of Innovations in
literature. Energy Systems and Power, 4(2009), No. 2, 29-34.

PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 2/2021 157

[4] Z. L. Gaing, “Particle swarm optimization to solving the algorithm”, International Journal of Electrical Power and
economic dispatch considering the generator constraints”, Energy Systems, 32 (2010), No. 7, 794-803.
IEEE Transactions on Power Systems, 18(2003), No. 3, 1187- [16] M. A. Abido, “Multi-objective particle swarm optimization for
1195. environmental/economic dispatch problem”, Electric Power
[5] Wanchai KHAMSEN and Chiraphon TAKEANG, “Hybrid of Systems Research, 79(2009), 1105-1113.
lamda and bee colony optimization for solving economic [17] Y. M. Chen, and W. S. Wang, “A particle swarm approach to
dispatch”, PRZEGLAD ELEKTROTECHNICZNY, 9(2016), solve environmental/economic dispatch problem”,
220-223. International Journal of Industrial Engineering Computations,
[6] T. Ratniyomchai, A. Oonsivilai, P. Pao-La-Or, and T. 1(2010), 157-172.
Kulworawanichpong, “Particle swarm optimization for solving [18] Anurag Gupta, K. K. Swarnkar, and K. Wadhwani, “Combined
combined economic and emission dispatch problems”, in economic emission dispatch problem using particle swarm
Proceedings of the 5th IASME/WSEAS International optimization”, International Journal of Computer Applications,
Conference of Energy and Environment: 211-216, 2010. 49(2012), No. 6, 1-6.
[7] C. Palanichamy, and N. S. Babu, “Analytical solution for [19] S. Hemamalini, and S. P. Simon, “Economic/emission load
combined economic and emissions dispatch”, Electric Power dispatch using artificial bee colony algorithm”, ACEEE
Systems Research, 78(2008), No. 7, 1129-1137. International Journal on Electrical and Power Engineering,
[8] N. Cetinkaya, “Optimization algorithm for combined economic 1(2010), No. 2, 27-33.
and emission dispatch with security constraints”, International [20] Y. Sonmez, “Multi-objective environmental/economic dispatch
Conference on Computer Science and Application (ICCSA solution with penalty factor using artificial bee colony
2009), 150-153. algorithm”, Scientific Research and Essays, 6(2011), No. 3,
[9] L. H. Wu, Y. N. Wang, X. F. Yuan, and S. W. Zhou, 2824-2831.
“Environmental/economic power dispatch problem using multi- [21] D. Aydin, S. Ozyon, C. Yasar, and T. Liao, “Artificial bee
objective differential evolution algorithm”, Electric Power colony algorithm with dynamic population size to combined
Systems Research, 80(2010), No. 9, 1171-1181. economic and emission dispatch problem”, International
[10] L. A. Koridak, M. Rahli, and M. Younes, “Hybrid optimization Journal of Electrical Power, 54(2014), 144-153.
of the emission and economic dispatch by the genetic [22] Jordan Radosavljevic, “Gravitational search algorithm for
algorithm”, Leonardo Journal of Sciences, (2008), Issue 14, solving combined economic and emission dispatch”, Infoteh-
193-203. Jahorina, 14(2015), 148-153.
[11] U. Güvenç, “Combined economic emission dispatch solution [23] M. Jevtic, N. Jovanovic, J. Radosavljevic, and D. Klimenta,
using genetic algorithm based on similarity crossover”, “Moth swarm algorithm for solving combined economic and
Scientific Research and Essays, 5(2010), No. 17, 2451-2456. emission dispatch problem”, Elektron Elektrotech, 23(2017),
[12] Simon Dinu, Ioan Odagescu, and Maria Moise, No. 5, 21-28.
“Environmental economic dispatch optimization using a [24] M. Jevtic, N. Jovanovic, and J. Radosavljevic, “Solving a
modified genetic algorithm”, International Journal of Computer combined economic emission dispatch problem using
Applications, 20(2011), No. 2, 7-14. adaptive wind driven optimization”, Turkis Journal of Electrical
[13] J. Sasikala, and M. Ramaswamy, “Optimal λ based economic Engineering & Computer Sciences, 26(2018), 1747-1758.
emission dispatch using simulated annealing”, International [25] S. Mirjalili, “The ant lion optimizer”, Advances in Engineering
Journal of Computer Applications, 1(2010), No. 10, 55-63. Software, 83(2015), 80-98.
[14] P. K. Roy, S. P. Ghoshal, and S. S. Thakur, “Combined [26] S. Mirjalili, P. Jangir and S. Saremi, “Multi-objective ant lion
economic and emission dispatch problems using optimizer: a multi-objective optimization algorithm for solving
biogeography-based optimization”, Electrical Engineering, engineering problems”, Applied Intelligence, 46(2017), No. 1,
92(2010), No. 4-5, 173-184. 79-95.
[15] P. K. Hota, A. K. Barisal, and R. Chakrabarti, “Economic
emission load dispatch through fuzzy based bacterial foraging

158 PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 2/2021