Heat Mass Transfer

DOI 10.1007/s00231-012-1038-2

ORIGINAL

Optimization of solar air collector using genetic algorithm

and artificial bee colony algorithm
Arzu S encan S ahin

Received: 25 January 2012 / Accepted: 12 June 2012

 Springer-Verlag 2012

Abstract Thermal performance of solar air collector

depends on many parameters as inlet air temperature, air
velocity, collector slope and properties related to collector.
In this study, the effect of the different parameters which
affect the performance of the solar air collector are
investigated. In order to maximize the thermal performance
of a solar air collector genetic algorithm (GA) and artificial
bee colony algorithm (ABC) have been used. The results
obtained indicate that GA and ABC algorithms can be
applied successfully for the optimization of the thermal
performance of solar air collector.
List
Ac
Cp
D
F0
FR
h
hw
I
K
ki
m_
N
Pr
Q
Re
t
ta

of symbols
Total collector aperture area (m2)
Specific heat of air (J/kg K)
Hydraulic diameter of duct (m)
Collector efficiency factor
Heat removal factor
Heat transfer coefficient (W/m2 K)
Wind convection coefficient (W/m2 K)
Solar irradiation (W/m2)
Thermal conductivity of air (W/m K)
Thermal conductivity of insulating material (W/m K)
Mass flow rate of air (kg/s)
Number of glass covers
Prandtl number
Rate of useful heat gain (W)
Reynolds number
Thickness of insulating material (m)
Temperature of air (C)

A. S encan S ahin (&)

Technology Faculty, Suleyman Demirel University,
32260 Isparta, Turkey
e-mail: arzusencan@sdu.edu.tr

ti
t0
tp
UL
UT
V
W

Inlet temperature of air (C)

Outlet temperature of air (C)
Temperature of absorber plate (C)
Overall heat loss coefficient (W/m2 K)
Top heat loss coefficient (W/m2 K)
Air velocity (m/s)
Wind velocity (m/s)

Greek symbols
s Transmittance of glass cover
a Absorptance of absorber plate
g Thermal performance
ep Absorber plate emissivity
eg Emissivity of glass cover
b Collector slope ()
1 Introduction
Solar air collectors may be used for space heating and
cooling, drying applications. Solar air collectors are relatively simple, easy to operate and have low capital costs.
The efficiency of these collectors in air heating is low due
to low convective heat transfer coefficients between
absorber and the flowing air that increases the absorber
plate temperature, resulting in higher heat loses to the
ambient [1, 2]. Therefore, studies on improving the efficiencies of the solar air collectors are very important. In
the literature, different methods for optimization of solar
air collectors have been used. Gupta and Kaushik established the optimal performance parameters for the maximum exergy delivery during the collection of solar energy
in a flat-plate solar air heater. The procedure to determine
optimum aspect ratio and optimum duct depth for maximum exergy delivery has been developed. The energy and
exergy output rates of the solar air heater were evaluated

123

Heat Mass Transfer

for various values of collector aspect ratio of the collector,

mass flow rate per unit area of the collector plate and solar
air heater duct depth [3]. Kalogirou used artificial neural
networks and genetic algorithms to find the optimum
combination of the collector area and storage-tank size to
maximize the life-cycle savings of a solar installation. The
optimization of an industrial process heat system for the
food industry is presented. The optimum solutions
obtained from the present methodology give increased
life-cycle savings of 4.9 and 3.1 % when subsidized and
non-subsidized fuel prices are used respectively, as compared to solutions obtained by the traditional trial-anderror method [4]. Chang [5] estimated the optimal tilt
angle for solar collectors. The observed data are resolved
into diffusion and beam components, and transformed into
instantaneous time frames using mathematical models.
The energy gain on installing a single-axis tracked panel
as compared to a traditional fixed panel is originally
analyzed theoretically. In addition to the observation data,
both types of radiation will be taken into account for
comparison, i.e., both extraterrestrial radiation and global
radiation predicted using empirical models. The results
show that the yearly optimal angles for six selected stations are about 0.95 and 0.88 times their latitudes for
extraterrestrial and predicted radiation, respectively. All of
the observed irradiations are less than the predicted values
for all times and stations, consequently resulting in a
flatter tilt angle, with a few exceptions in summer. Hegazy
[6] developed the analytical criterion for estimating optimum channel geometry of conventional solar air heaters to
other flat plate types. Assessment is made for two other
basic types with flow over the absorber and on both sides
of it. Heater performance characteristics are analytically
determined over a wide range of channel depth-to-length
ratios (D/L) and are compared with those of the conventional type, having a flow channel under the absorber. The
effects of air mass flow rate, emissivity of channel plates
and wind heat transfer coefficient on the accuracy of the
criterion are also investigated. For these two solar air
heaters, the criterion is found to be equally valid as it
predicts the optimal (D/L) ratio that yields best performance. Varun et al. [7] carried out experimental study to
investigate the effective efficiency of a solar air heater
duct provided with transverse and inclined ribs as artificial
roughness elements on the absorber plate. The range of
parameters considered for the present investigation; Reynolds number 2,00014,000, relative roughness pitch 38
and a fixed value of relative roughness height of 0.03. The
effective efficiency has been computed based on the
experimentally determined values for the range of
parameters considered. Further an attempt has also been
made to optimize the thermal efficiency for the same

123

system under similar conditions by Taguchi method. Varun [8] optimized the thermal performance of flat plate
solar air heater by considering the different system and
operating parameters to obtain maximum thermal performance. Thermal performance is obtained for different
Reynolds number, emissivity of the plate, tilt angle and
number of glass plates by using genetic algorithm.
In literature, there are some studies based on optimization of energy systems with GA algorithm. But, studies on
optimization of energy systems with ABC algorithm are
very limited. Rao and Patel [9] used ABC algorithm for
design optimization of mechanical draft counter flow wetcooling tower. Minimizing the total annual cost for specific
heat duty requirement is considered as objective function.
Three design variables such as water to air mass ratio, mass
velocity of water and mass velocity of air are considered
for optimization. Evaluations of the cooling tower geometry and performances are based on an adaptive version of
Merkels method. Temperature and enthalpy constraints
are included in the optimization procedure. Six examples
are presented to demonstrate the effectiveness and accuracy of the proposed algorithm. The results of optimization
using ABC are validated by comparing with those obtained
by using general algebraic modeling system (GAMS)
optimization package. S encan et al. [10] used ABC algorithm to minimize the total cost of the equipment including
capital investment and the sum of discounted annual
energy expenditures related to pumping of shell and tube
heat exchanger by varying various design variables such as
tube length, tube outer diameter, pitch size, baffle spacing,
etc. Finally, the results are compared to those obtained by
literature approaches. The obtained results indicate that
ABC algorithm can be successfully applied for optimal
design of shell and tube heat exchangers. Basu [11] presented a novel optimization approach to the combined heat
and power economic dispatch problem by using bee colony
optimization algorithm. The algorithm is a swarm-based
algorithm inspired by the food foraging behavior of honey
bees. The performance of the proposed algorithm is validated by illustration with a test system. The results of the
proposed approach are compared with those of particle
swarm optimization, real-coded genetic algorithm and
evolutionary programming techniques.
In this study, the application of the ABC algorithm in
order to maximize thermal performance of the solar air
collector as different from artificial intelligence methods
available in the literature is investigated. The results of the
ABC algorithm are compared with results obtained by
using GA algorithm. ABC approach was successfully
applied for optimum thermal performance of solar air
collector. This study has provided new methodology in the
optimization of solar air collector.

Heat Mass Transfer

2 GA and ABC algorithms

The GA can be used to solve of optimization problems that
are not well suited for standard optimization algorithms.
The objective function of optimization problems can be
discontinuous, non differentiable, stochastic, or highly
nonlinear [12].
Algorithm is started with a set of solutions (represented
by chromosomes) called population. Solutions from one
population are taken and used to form a new population.
This is motivated by a hope, that the new population will
be better than the old one. Solutions which are selected to
form new solutions (offspring) are selected according to
their fitnesss, the more suitable they are the more chances
they have to reproduce. This is repeated until some condition (for example number of generation or improvement
of the best solution) is satisfied [13].
The procedure of GA for the present work is described
as follows:
Step 1
Step 2
Step 3
Step 4
Step 5
Step 6
Step
Step
Step
Step

7
8
9
10

Step 11
Step 12

Read system data and GA parameters

Generate initial population of n individuals
Calculate the fitness Fi of each chromosome in
the population
Create a new population by repeating following
steps until the new population is complete
Select two parents by tournament selection
With stochastic uniform, cross over the parent
chromosomes to form a new offspring (children)
With a mutation probability mutate new offspring
Calculate the fitness Fi of each new offspring
Place new offspring in a new population
Sort the all individuals from minimum to
maximum
Use new generated population (population size
is n) for a further run of algorithm
If the end condition is satisfied, stop, and return the
best solution in current population else go to Step 4

Flow chart of GA is shown in Fig. 1.

In addition, characteristic of GA used in this study was
given in Table 1.
The GA uses three main types of rules at each step to
create the next generation from the current population [12]:

Selection rules select the individuals, called parents that

contribute to the population at the next generation.
Crossover rules combine two parents to form children
for the next generation.
Mutation rules apply random changes to individual
parents to form children. The genetic algorithm differs
from a classical, derivative-based, optimization algorithm in two main ways, as summarized in the following
Table 2.

Fig. 1 Basic flow chart of GA [14]

Table 1 Characteristic of genetic algorithm

Initial population

Child

Mutation

Number of
generation

50

20

10

100

Table 2 Differences between genetic algorithm with classical

algorithm
Classical algorithm

Genetic algorithm

Generates a single point at each

iteration. The sequence of
points approaches an optimal
solution

Generates a population of points

at each iteration. The best point
in the population approaches an
optimal solution

Selects the next point in the

sequence by a deterministic
computation

Selects the next population by

computation which uses
random number generators

GA and applications have been discussed extensively by

various authors [1317].
The ABC algorithm, proposed by Karaboga in 2005 [18]
for real-parameter optimization, has been a recently introduced as optimization algorithm do simulate the foraging
behaviour of a bee colony. The minimal model of swarmintelligent forage selection in a honey bee colony consists
of three kinds of bees: employed bees, onlooker bees and
scout bees. Half of the colony consists of employed bees,

123

Heat Mass Transfer

and the other half includes onlooker bees. Employed bees

are responsible for exploiting the nectar sources explored
before and giving information to the waiting bees (onlooker bees) in the hive about the quality of the food source
sites which they are exploiting. Onlooker bees wait in the
hive and decide on a food source to exploit based on the
information shared by the employed bees. Scouts either
randomly search the environment in order to find a new
food source depending on an internal motivation or based
on possible external clues [1921].
Main steps of the ABC algorithm are given below [22]:
1.
2.

3.

4.

5.
6.

The food source positions are initialized.

Each employed bee produces a new food source in her
food source site (neighborhood) and exploits the better
source.
A food source is selected depending on the quality of its
solution and then an onlooker bee produces a new food
source in the selected food source site and exploits the
better source.
The source to be abandoned is determined and its
employed bee is allocated as scout for searching new
food sources.
The best food source found so far is memorized.
Steps (25) are repeated until the stop criterion is
verified.

More details on ABC algorithm can be found in Refs.

[2328].

3 Thermal performance formulations of solar air

collector
The solar air collector consists of absorber plate, transparent
cover, insulation material, frame, air passage and fins if
necessary. Figure 2 shows the schematic diagram of solar
air collector. The heat balance on the solar air collector
gives the distribution of incident solar irradiation, I, into
useful heat gain, Q, and various heat losses, refer Fig. 3.
In this section formulations that are required in order to
determine the useful energy collected and the interaction of

Fig. 3 Heat balance of solar air collector

the various constructional parameters on the performance

of solar air collector are presented. The heat gain by air
may be calculated by the following Eqs. (13):



Q Ac saI  UL tp  ta
1
Q Ac FR saI  UL ti  ta 

In this equation FR is the heat removal factor and can be

expressed as Eqs. (13):
i
_ p h
0
mc
_ p
1  eUL Ac F =mc
FR
3
UL A c
The collector efficiency factor F0 is:
F0

h
h UL

The heat transfer coefficient h can be determined from the

correlation developed for smooth flat plate collector (8).
0:4

h 0:024Re0:8 Pr

k
D

The overall heat loss coefficient UL is given by:

UL UT

ki
t

The top heat loss coefficient UT is evaluated by Eqs. (13):

2
31
N
17
6
UT 4 
0:33 5
hw
C tp ta
tp

Nf




r tp ta tp2 ta2
i


1 h2Nf
ep 0:05N 1  ep
eg  N

where
C 2501  0:0044b  90


f 1  0:04hw 0:005h2w 1 0:091N
hw 5:7 3:8W

Fig. 2 Schematic diagram of solar air collector

123

8
9
10

The efficiency of the solar air collector is given by the

following Eqs. (13):


UL ti  ta
g FR sa 
:
11
I

Heat Mass Transfer

4 The objective function and optimization design

of solar air collector
In order to obtain maximum thermal performance of solar
air collector, optimum operating parameters were determined. The objective functions used in GA and ABC is:


UL ti  ta
Maximize g FR sa 
11
I
Numerical calculations started with the assumed value
of plate temperature. Using this assumed value, and known
collector geometry and system properties, equations
described in the previous section were solved for thermal
performance. The plate temperature was calculated as:


1  FR
ti  ta
tp ta FR I sa

12
FR UL
I sa
From the calculated value, new plate temperature was
computed. This new plate temperature is compared with
the previous value and the difference decides the further
course of calculations. If the difference is within acceptable
limits, the process is terminated. If the difference is outside
the tolerance limits the calculated value of tp is used as
revised value. This process continued until the solution
converged.
Thermal performance of solar air collector at different
surrounding conditions and operating parameters was
found out. Variables in the thermal performance optimization of solar air collector with GA and ABC were given
in Table 3.

5 Result and discussion

The thermal performance of a solar air collector depends
on different design components as collector slope, absorber
Table 3 Variables in optimization and assumptions
Variable

Value
2

Solar irradiation (W/m )

600

Reynolds number

2,0006,000

Air temperature (K)

290310

Air velocity (m/s)

05

Collector slope (o)

090

Emissivity of glass covers

0.850.95

Absorber plate emissivity

Number of glass covers

0.80.9
13

Total collector aperture area (m2)

1.5

Hydraulic diameter of duct (m)

0.03

Thickness of insulating material (m)

0.04

Wind velocity (m/s)

plate emissivity, emissivity of glass cover, air temperature,

air velocity and number of glass cover.
In this study for fixed solar irradiation (I = 600) and
Reynolds number varying from 2,000 to 6,000 and different number of glass covers (N = 1, 2 and 3), the optimized
values of air velocity (V), collector slope (b), absorber
plate emissivity (ep ), and emissivity of glass cover (eg ) and
air temperature (ta) were obtained. So, the maximum
thermal performance of the solar air collector was
obtained. The GA and ABC methods were used for optimization process.
By using more number of glass covers over a solar air
collector will increase the cost of the system and the system
become more complicated but it increases the thermal
performance of the system. In this study, total three sets of
glass cover (1, 2 and 3) has been considered and it has been
found out that for same solar irradiation the thermal efficiency is maximum for the three glass cover solar air
collector.
The thermal performance of a solar air collector
increases with the increase in Reynolds number. As the
Reynolds number increases the turbulence in the flow
increases which leads to higher heat transfer and which
yield higher thermal performance. In this study, Reynolds
number varying from 2,000 to 6,000 has been considered
and it has been found out that for same solar irradiation the
thermal efficiency is maximum at Reynolds number 6,000.
Figure 4 shows the thermal performance against the
number of generations during the running of the GA and
ABC. The optimum solution in both methods was reached
in less than 10 s, which is very fast. As can be seen after
about 25 generations, the optimum thermal performance
for GA method has been obtained as 0.7983. After about 27
generations, the optimum thermal performance for ABC
method has been obtained as 0.7998. The thermal performance of solar air collector obtained with ABC method for
the same working conditions is slightly high than thermal
performance obtained with GA method. Therefore, the
maximum value of thermal performance of the solar air
collector is obtained as 0.7998.

Fig. 4 GA and ABC results for thermal performance

123

Heat Mass Transfer

Fig. 5 GA and ABC results for collector slope

123

Fig. 6 GA and ABC results for absorber plate emissivity

0.9

Emissivity of glass cover

Figure 5 shows the collector slope against the number of

generations during the running of the GA and ABC. The
optimum solution in both methods was reached in less than
10 s, which is very fast. As can be seen after about 38
generations, the optimum collector slope for GA method
has been obtained as 82. After about 76 generations, the
optimum collector slope for ABC method has been
obtained as 40. The optimum collector slope of the solar
air collector can be accepted as 40 because the thermal
performance obtained with ABC method is slightly high
than thermal performance obtained with GA method as can
be seen in Fig. 4.
Figure 6 shows the absorber plate emissivity against the
number of generations during the running of the GA and
ABC. The optimum solution in both methods was reached
in less than 10 s, which is very fast. As can be seen after
about 62 generations, the optimum absorber plate emissivity for GA method has been obtained as 0.8537. After
about 75 generations, the optimum absorber plate emissivity for ABC method has been obtained as 0.9415. The
optimum absorber plate emissivity of the solar air collector
can be accepted as 0.9415 because thermal performance
obtained with ABC method is slightly high than thermal
performance obtained with GA method as can be seen in
Fig. 4.
Figure 7 shows the emissivity of glass cover against the
number of generations during the running of the GA and
ABC. The optimum solution in both methods was reached
in less than 10 s, which is very fast. As can be seen after
about 23 generations, the optimum emissivity of glass
cover for GA method has been obtained as 0.802. After
about 76 generations, the optimum emissivity of glass
cover for ABC method has been obtained as 0.8043. The
optimum emissivity of glass cover of the solar air collector
can be accepted as 0.8043 because thermal performance
obtained with ABC method is slightly high than thermal
performance obtained with GA method as can be seen in
Fig. 4.
Figure 8 shows the air temperature against the number
of generations during the running of the GA and ABC. The

GA
ABC

0.88
0.86
0.84
0.82
0.8
10

20

30

40

50

60

70

80

90

100

Number of generation

Fig. 7 GA and ABC results for emissivity of glass cover

Fig. 8 GA and ABC results for air temperature

optimum solution in both methods was reached in less than

10 s, which is very fast. As can be seen after about 50
generations, the optimum air temperature for GA method
has been obtained as 291.6581 K. After about 40 generations, the optimum air temperature for ABC method has
been obtained as 290.0798 K. The optimum air temperature of the solar air collector can be accepted as
290.0798 K because thermal performance obtained with
ABC method is slightly high than thermal performance
obtained with GA method as can be seen in Fig. 4.
Figure 9 shows the air velocity against the number of
generations during the running of the GA and ABC. The
optimum solution in both methods was reached in less than
10 s, which is very fast. As can be seen after about 43
generations, the optimum air velocity for GA method has
been obtained as 3.7359 m/s. After about 77 generations,

Heat Mass Transfer

Air velocity (m/s)

4
GA
ABC

10

20

30

40

50

60

70

80

90

100

Number of generation

Fig. 9 GA and ABC results for air velocity

Table 4 Optimization results with GA and ABC algorithm
Method

ep

eg

ta

GA

82

3.7359

0.8537

0.8020

291.6581

0.7983

ABC

40

3.6859

0.9415

0.8043

290.0798

0.7998

of solar air collector is low. The thermal performance of

solar air collector depends on many parameters as air
temperature, air velocity, collector slope, emissivity of
glass cover and absorber plate emissivity. In this study in
order to maximum the thermal performance of solar air
collector, GA and ABC methods were used. The optimized
values of air velocity, collector slope, absorber plate
emissivity, emissivity of glass cover and air temperature
were obtained. The results of this work show that GA and
ABC can be successfully used for the optimization of
thermal performance of solar air collector. ABC and GA
method are found to be the most accurate and quick
according to traditional methods. The algorithms proposed
here can help the manufacturer and engineers for optimization of thermal performance of solar air collector in
engineering applications.
Acknowledgments The author thanks to Bayram Klc and Ulas
Klc for helpful comments.

the optimum air velocity for ABC method has been

obtained 3.6859 m/s. The optimum air velocity of the solar
air collector can be accepted as 3.6859 m/s because thermal performance obtained with ABC method is slightly
high than thermal performance obtained with GA method
as can be seen in Fig. 4.
The optimized values of air velocity, collector slope,
absorber plate emissivity, emissivity of glass cover and air
temperature were obtained. So, the maximum thermal performance of the solar air collector was obtained. The GA
and ABC methods were used for optimization process. A
comparison between the results obtained by GA and ABC
for this case study is given in Table 4. Thermal performance
of solar air collector obtained with ABC method for the
same working conditions is slightly high than thermal performance obtained with GA method. The maximum value of
thermal performance comes out at 0.7998 having three glass
covers and solar irradiation of 600 W/m2 at Reynolds
number 6,000. The optimized set of value for this is velocity
V = 3.6859 m/s, collector slope b = 40o, absorber plate
emissivity ep = 0.9415, emissivity of glass covers eg =
0.8043, air temperature is 290.0798 K.
The optimum solution in both methods was reached in
less than 10 s. The time required to find the optimum
solution using GA and ABC methods is greatly reduced. In
addition, the solution reached is more correct than by using
the traditional trial-and-error method which most of the time
relies on the intuition of the user to find a good solution.

6 Conclusions
The determination of optimum system and operating
parameters is important because the thermal performance

References
1. Kalogirou SA (2004) Solar thermal collectors and applications.
Prog Energy Combust Sci 30:231295
2. Duffie JA, Beckman WA (1991) Solar engineering of thermal
processes. Wiley, New York
3. Gupta MK, Kaushik SC (2008) Exergetic performance evaluation
and parametric studies of solar air heater. Energy 33:16911702
4. Kalogirou SA (2004) Optimization of solar systems using artificial
neural-networks and genetic algorithms. Appl Energy 77:383405
5. Chang TP (2009) Performance evaluation for solar collectors in
Taiwan. Energy 34:3240
6. Hegazy AA (2000) Performance of flat plate solar air heaters with
optimum channel geometry for constant/variable flow operation.
Energy Convers Manage 41:401417
7. Varun S, Patnaik A, Saini RP, Singal SK (2009) Performance
prediction of solar air heater having roughened duct provided
with transverse and inclined ribs as artificial roughness. Renew
Energy 34:29142922
8. Varun S (2010) Thermal performance optimization of a flat plate
solar air heater using genetic algorithm. Appl Energy 87:1793
1799
9. Rao RV, Patel VK (2011) Optimization of mechanical draft
counter flow wet-cooling tower using artificial bee colony algorithm. Energy Convers Manage 52:26112622
10. S encan A, Klc B, Klc U (2011) Design and economic optimization of shell and tube heat exchangers using artificial bee
colony (ABC) algorithm. Energy Convers Manage 52:33563362
11. Basu M (2011) Bee colony optimization for combined heat and
power economic dispatch. Expert Syst Appl 38:1352713531
12. MATLAB Global Optimization Toolbox 3 Users Guide, The
MathWorks, Inc., 2010
13. Chakraborty D, Sharma CP, Das B, Abhishek K, Malakar T
(2009) Distribution system load flow solution using genetic
algorithm. In: ICPS09 international conference on power systems, pp 16
zdemir A, Lim JY, Singh C (2005) Post-outage reactive power
14. O
flow calculations by genetic algorithms: constrained optimization
approach. IEEE Trans Power Syst 20:12661272

123

Heat Mass Transfer

15. Sen Z, Oztopal A, Sahin AD (2001) Application of genetic
algorithm for determination of AngstroE`m equation coeffcients.
Energy Convers Manage 42:217231
16. Gosselin L, Tye-Gingras M, Mathieu-Potvin F (2009) Review of
utilization of genetic algorithms in heat transfer problems. Int J
Heat Mass Transf 52:21692188
17. S encan S ahin A, Klc B, Klc U (2011) Optimization of heat
pump using fuzzy logic and genetic algorithm. Heat Mass Transf
47:15531560
18. Karaboga D (2005) An idea based on honey bee swarm for
numerical optimization. Technical ReportTR 06, Oct 2005.
Erciyes University, Engineering Faculty, Computer Engineering
Department, Kayseri
19. Karaboga D, Basturk B (2007) A powerful and efficient algorithm
for numerical function optimization: artificial bee colony (ABC)
algorithm. J Glob Optim 39:459471
20. Karaboga D, Basturk B (2008) On the performance of artificial
bee colony (ABC) algorithm. Appl Soft Comput 8:687697
21. Akay B, Karaboga D (2012) A modified artificial bee colony
algorithm for real-parameter optimization. Inf Sci 192:120142

123

22. Oliveira IMS, Schirru R (2011) Swarm intelligence of artificial

bees applied to in-core fuel management optimization. Ann Nucl
Energy 38:10391045
23. Karaboga D, Akay B (2009) A comparative study of artificial bee
colony algorithm. Appl Math Comput 214:108132
24. Karaboga N (2009) A new design method based on artificial bee
colony algorithm for digital IIR filters. J Franklin Inst 346:
328348
25. Karaboga D, Akay B (2011) A modified artificial bee colony
(ABC) algorithm for constrained optimization problems. Appl
Soft Comput 11:30213031
26. Karaboga D, Ozturk C (2011) A novel clustering approach:
artificial bee colony (ABC) algorithm. Appl Soft Comput 11:
652657
27. Sonmez M (2011) Artificial bee colony algorithm for optimization of truss structures. Appl Soft Comput 11:24062418
28. Samanta S, Chakraborty S (2011) Parametric optimization of
some non-traditional machining processes using artificial bee
colony algorithm. Eng Appl Artif Intell 24:946957