Applied Sciences: Differential Evolution: A Survey and Analysis
Applied Sciences: Differential Evolution: A Survey and Analysis
Applied Sciences: Differential Evolution: A Survey and Analysis
sciences
Article
Differential Evolution: A Survey and Analysis
Tarik Eltaeib * and Ausif Mahmood
Computer Science and Engineering Department, University of Bridgeport, Bridgeport, CT 06614, USA;
[email protected]
* Correspondence: [email protected]; Tel.: +1-678-237-6229
Received: 26 July 2018; Accepted: 4 October 2018; Published: 16 October 2018
Abstract: Differential evolution (DE) has been extensively used in optimization studies since
its development in 1995 because of its reputation as an effective global optimizer. DE is a
population-based metaheuristic technique that develops numerical vectors to solve optimization
problems. DE strategies have a significant impact on DE performance and play a vital role in
achieving stochastic global optimization. However, DE is highly dependent on the control parameters
involved. In practice, the fine-tuning of these parameters is not always easy. Here, we discuss the
improvements and developments that have been made to DE algorithms. In particular, we present a
state-of-the-art survey of the literature on DE and its recent advances, such as the development of
adaptive, self-adaptive and hybrid techniques.
1. Introduction
Optimization algorithms are important approaches for resolving difficult optimization
problems [1]. Optimization is defined as the procedure of discovery that provides the minimum
or maximum value of a function f (x) [2,3]. There are many reasons that make these problems difficult
to solve. First, we cannot perform a comprehensive search if the problem domain space is too large.
Second, the evaluation function is noisy or varies with time, generating a series of solutions instead of
a single solution. Third, sometimes the constraints prevent arriving at a possible solution such that the
optimization approach is the only solution [4]. The mathematical representation for the optimization
involves finding an X ∗ = [x1∗ , x2∗ , x ∗N ] ∈ D N = D1 ∩ D2 ∩ . . . ∩ D N , where
where X* is the optimal solution in the N-dimensional search space DN , N is the dimension of the
optimization (the number of optimization parameters) and f imin and f imax are the objective functions.
Differential evolution (DE) is a stochastic algorithm for solving numerical continuous optimization
problems. Since its inception, the DE algorithm has become a powerful global optimizer. Developed by
Kenneth Price in 1994, DE is a promising optimization algorithm that converges to the real optimum
without using significant amounts of resources. Furthermore, its performance was validated in the
evolutionary domain by the IEEE Conference on Evolutionary Computation (CEC) in 1996 [5].
More recently, different versions of DE have secured the top ranks in many competitions between
evolutionary algorithms (EAs) by the IEEE CEC conference series (http://www.ntu.edu.sg/home/
epnsugan/index_files/cec-benchmarking.htm). However, an impressive number of different DE
algorithms have been introduced by the research community over the past decades. To understand all
the improvements that have been made to DE, a theoretical study and relative analysis of different
enhancements are presented in this paper. Because various DE algorithms involving different
techniques are comprehensively discussed, the main motivation behind this survey is to deepen
understanding of the characteristics of different DE strategies, with the goal of benefiting from the
various approaches. In fact, understanding how to combine these DEs harmoniously and their
underlying concepts could be crucial to attaining effective designs or improving the performance of
DE algorithms in particular or any optimization algorithms in general. Moreover, the literature shows
that no single algorithm has been demonstrated to be effective for various applications. Thus, this
study may provide a roadmap through which developers may gain a full understanding of this field.
DE algorithms are different from EA algorithms that shape offspring by mixing solutions with
a difference factor rate of selected individual vectors and they are an alternative to recombining
individuals through a probabilistic scheme. In fact, the differential mutation strategy is the main
component that distinguishes DE from other population algorithms. Applying the mutation to all
candidates defines an exploration rule based on other candidate solutions. Therefore, the mutation
strategy enhances a population’s capability for discovering new promising offspring based on the
current distribution of solutions within the domain space [6]. Ideally, the performance of DE is
based on two major components: the chosen strategy and the control parameters. However, the
strategy underlying DE consists of mutation, crossover and selection operators, which are utilized at
each generation to determine the global optimum. The control parameter components consist of the
population size NP, scaling factor F and the crossover rate Cr [7,8].
Despite the potential of DE, it is obvious to the research community that some adjustments
to classic DE are essential to significantly enhance its performance, especially in addressing
high-dimension problems. Stagnation, premature convergence and sensitivity to control parameters
are the main issues that influence the performance of DE [9]. Stagnation occurs when the population
cannot converge to a suboptimal solution although the diversity of the population remains high [10].
In other words, the population does not improve over a period of iterations, and the algorithm is not
capable of finding a new search domain [11]. There are many causes of stagnation, including control
parameters, which become inefficient for a specific problem in the decision space [9,10]. Many studies
have proposed a variety of ways to improve the current DE algorithm through modifications [12,13],
including the use of differential mutations with perturbations, mutations with selection pressure, and
operator adaption techniques [14–17]. Qing conducted an extensive study on differential evolution [18]
and observed that the performance of differential evolution and the quality of the results were based
on the type of technique used, either binomial or exponential [16,19].
This article provides a comprehensive survey of the different types of state-of-the-art differential
evolution algorithms available as global numerical optimizations over continuous search spaces.
Thus, this study provides a foundation for researchers who are interested in optimization in
general or who care about recent developments in DE. The growing research area is divided into
adaptive, self-adaptive, and hybridization strategies. This comprehensive study sheds light on most
improvements and developments pertaining to different types of DE families, including primary
concepts and a variety of DE formats.
This article is organized as follows. Section 2 briefly describes classic differential evolution.
Section 3 analyzes the parameter strategies and their effects on DE performance and surveys different
DE approaches. Section 4 presents the conclusions drawn from our study.
solution. Therefore, the population is initialized within only boundary constraints, the upper bound
(H) and lower bound (L), so the population can by initialized by the following:
Appl. Sci. 2018, 7, x FOR PEER REVIEW 3 of 25
Pi,j = L + (H − L)·randi,j [0,1) i = 1, 2, ..., D; j = 1, 2, ..., NP
𝑃𝑖,𝑗 = 𝐿 + (𝐻 − 𝐿). 𝑟𝑎𝑛𝑑𝑖,𝑗 [0,1), 𝑖 = 1,2, … . , 𝐷 ; 𝑗 = 1,2, … . , 𝑁𝑃
After the initialization phase, the evolution involves the three processes of mutation, crossover,
After the initialization
and selection. phase, the evolution
The classic differential evolution strategy
involvesconsists
the threeofprocesses
three randomof mutation,
vectors crossover,
v1 , v2 and
and selection.
v3 that The classic
are selected from differential evolution
the population strategy
(Equation (1)).consists
Randomly of three random
select vectors v ,from
three individuals v , and
the
vpopulation
that are vselected from the population (Equation (1)). Randomly select three individuals from the
1,2,3 ∈ [1, ..., NP], where v1 6 = v2 6 = v3
population v , , ∈ [1, … . 𝑁𝑃], where v ≠ v ≠ v
v1 v= (int) (rand()
= (𝑖𝑛𝑡) ∗ NP);
(𝑟𝑎𝑛𝑑( ) ∗ 𝑁𝑃);
Figure 1.
Figure Random vectors
1. Random vectors selected
selected in
in the
the mutation
mutation strategy
strategy (classic
(classic differential
differential evolution
evolution (DE)).
(DE)).
The mutation process is the main distinctive component of DE and is considered the strategy by
The mutation process is the main distinctive component of DE and is considered the strategy by
which DE is carried out. There are different types of mutation strategies, each one distinguished with
which DE is carried out. There are different types of mutation strategies, each one distinguished with
an abbreviation based on the classic mutation strategy described by Equation (1), i.e., DE/rand/1/bin,
an abbreviation based on the classic mutation strategy described by Equation (1), i.e., DE/rand/1/bin,
where DE represents differential evolution and “rand” represents random, which indicates that the
where DE represents differential evolution and “rand” represents random, which indicates that the
vectors are selected randomly. The number one indicates the number of difference pairs; in this strategy,
vectors are selected randomly. The number one indicates the number of difference pairs; in this
it is one pair (v1 − v2 ). The last term represents the type of crossover used. This term could be “exp”,
strategy, it is one pair (v − v ). The last term represents the type of crossover used. This term could
for exponential, or “bin”, for binomial [20]. Then, to complement the previous step (mutation strategy),
be “exp”, for exponential, or “bin”, for binomial [20]. Then, to complement the previous step
DE also applies uniform crossover to construct trial vectors vTrail , which are out of parameter values
(mutation strategy), DE also applies uniform crossover to construct trial vectors v , which are out
that have been copied from two different vectors. In particular, DE selected a random vector from
of parameter values that have been copied from two different vectors. In particular, DE selected a
the population, indicated as vx , which must be different to v1 , v2 , and v3 , and then it crosses with a
random vector from the population, indicated as v , which must be different to v , v , and v , and
mutant vector vy ; the binomial crossover is generated as follows:
then it crosses with a mutant vector 𝑣 ; the binomial crossover is generated as follows:
v [i, j] 𝑖𝑓 𝑟𝑎𝑛𝑑(0,1) < 𝐶𝑟
v [i, j] = (2)
v [i, j] 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
The crossover probability, Cr ∈ (0, 1), is a pre-defined rate that specifies the fraction of the
parameter that is transferred from the mutant. Thus, it is used to control which sources participate in
Appl. Sci. 2018, 7, x FOR PEER REVIEW 4 of 25
a given parameter. The uniform crossover rate is compared with uniform random values formed
from rand (0, 1); if the random value is smaller than or equal to Cr then the trial parameter is copied
from the
Appl.mutant 8, 1945 v , otherwise the parameter is inherited from v .
Sci. 2018,vector 4 of 25
The next operation is selection, in which the trail vector v competes with the target vector
v . If this trail vector v is equal or less than ( v it changes the target vector v in the next
generation otherwise v is not changed in the population, v y [i, j] i f rand (0, 1) < Cr
vTrail [i, j]= (2)
vx [i, j] otherwise (3)
v 𝑖𝑓 𝑓( v )≤ 𝑓(v )
v =
v 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
The crossover probability, Cr ∈ (0, 1), is a pre-defined rate that specifies the fraction of the
𝑓( x ) is the
whereparameter thatobjective function.
is transferred fromIfthe the new trail
mutant. Thus, vector
it is used ) is less
𝑓( v to control thansources
which or equal to the in a
participate
target given 𝑓( v ), it The
vectorparameter. replaces the target vector. Otherwise, the population
uniform crossover rate is compared with uniform random values formed maintains the target from
vectorrand
value.(0,Therefore, the different
1); if the random value DE phasesthan
is smaller prevent the population
or equal to Cr then from ever
the trial deteriorating;
parameter the from
is copied
population either vector
the mutant remains vy the same orthe
, otherwise improves.
parameter Furthermore,
is inheritedcontinued
from vx . refining of the population
is achievedTheby thenexttrial vector, is
operation although
selection,theinfitness
whichofthe thetrail
trialvector
vectorvTrail
is the same as with
competes that of
thethe current
target vector vx .
vector.IfThis
this factor is crucial
trail vector vTrailinisDE because
equal or lessit than
provides
vx itthe algorithm
changes the ability
the target vectortovmove through
x in the the
next generation
landscape usingvax isvariety
otherwise of generations
not changed [21]. The termination condition can be either a preset
in the population,
maximum number of generations or a pre-specified ( target of the objective function value [22].
vTrail i f f (vTrail ) ≤ f (vx )
2.1. Differential Evolution Strategies vx = (3)
vx otherwise
The various equations underpinning DE possess certain aspects in common when applied to
where f (x) is the objective function. If the new trail vector f (vTrail ) is less than or equal to the target
continuous optimization. All consist of an original point, sometimes referred to as the base point. The
vector f (vx ), it replaces the target vector. Otherwise, the population maintains the target vector value.
original algorithm carries out the search operation such that it finds the optimum as soon as possible.
Therefore, the different DE phases prevent the population from ever deteriorating; the population
We can generalize the DE formula to the form α = β + F × δ, where β represents the base vectors and
either remains the same or improves. Furthermore, continued refining of the population is achieved
δ the difference between vectors. Thus, the main goal of all DE equations is to provide the optimal
by the trial vector, although the fitness of the trial vector is the same as that of the current vector. This
direction based on the differential β and base vector δ (Figure 2).
factor is crucial in DE because it provides the algorithm the ability to move through the landscape
Establishing β and δ is crucial to creating an efficient strategy that can be applied to the chosen
using a variety of generations [21]. The termination condition can be either a preset maximum number
individuals from the population. However, all possible combinations of β and δ can be classified into
of generations or a pre-specified target of the objective function value [22].
the following strategies: local, random, directed, and hybrid. In random strategies, abbreviated as
“Rand”,
2.1.all individuals
Differential are formed
Evolution randomly, and there is no prior information about the objective
Strategies
function. In directed strategies, abbreviated as “DIR”, a suitable value for the base vector is chosen
according The
to thevarious equations
objective underpinning
function to ensure aDE possess
suitable certain aspects
direction. Hybrid instrategies
commoninclude
when applied
the to
continuous optimization. All consist of an original point, sometimes referred
combination of “Rand” and “DIR”, labeled RAND/DIR. In another approach, the best overall vector to as the base point. The
original
is used, not onlyalgorithm
the bestcarries
amongoutthethe searchindividuals;
selected operation such
this that it findsisthe
approach optimum
referred to asasthe
soon as possible.
“BEST”.
generalize the DE formula to the form α = β + F ×
Combining the “Rand” and “BEST” yields the hybrid RAND/BEST strategy. In addition, the and
We can δ, where β represents the base vectors
δ the difference
combination of more between
than twovectors. Thus,e.g.,
approaches, the main goal of all DEcan
RAND/BEST/DIR, equations is to provide
yield favorable theby
results optimal
direction based on the differential
exploiting the advantages of each approach. β and base vector δ (Figure 2).
Y-Axis
X-Axis
Establishing β and δ is crucial to creating an efficient strategy that can be applied to the chosen
individuals from the population. However, all possible combinations of β and δ can be classified into
Appl. Sci. 2018, 8, 1945 5 of 25
the following strategies: local, random, directed, and hybrid. In random strategies, abbreviated as
“Rand”, all individuals are formed randomly, and there is no prior information about the objective
function. In directed strategies, abbreviated as “DIR”, a suitable value for the base vector is chosen
according to the objective function to ensure a suitable direction. Hybrid strategies include the
combination of “Rand” and “DIR”, labeled RAND/DIR. In another approach, the best overall
vector is used, not only the best among the selected individuals; this approach is referred to as
the “BEST”. Combining the “Rand” and “BEST” yields the hybrid RAND/BEST strategy. In addition,
the combination of more than two approaches, e.g., RAND/BEST/DIR, can yield favorable results by
exploiting the advantages of each approach.
Table 1 shows that all DE strategies employed are formed based on the DE/rand/x variation,
which applies pairs of difference vectors:
whereas the scaling factors are frequently presumed to be the same F1 = F2 = . . . = Fk = F. Substituting
an arbitrary base vector x1 as vbest , “the best vector” from the population, provides a different DE
approach, indicated as DE/best/1:
vy = vbest + F ∗ (v1 − v2 )
Most mutation strategies can be formed by a general formula based on the sum of k scaled
difference vectors and a weighted average among the best vector and arbitrary ones:
Table 1. The differentiation operation can be carried out using many mutation strategies.
Strategy Formulation
h i
1. DE/best/1/exp V(G+1) = V(best,G) + F V(1,G) − V(2,G)
h i h i
2. DE/rand-to-best/1/exp V(G+1) = V(i,G) + λ· V(1,G) − V(2,G) + F · V(1,G) − V(2,G)
h i
3. DE/best/2/exp V(G+1) = V(best,G) + F V(1,G) + V(2,G) − V(3,G) − V(4,G)
h i
4. DE/rand/2/exp V(G+1) = V(5,G) + F V(1,G) + V(2,G) − V(3,G) − V(4,G)
h i
5. DE/best/1/bin V(G+1) = V(best,G) + F · V(2,G) − V(3,G)
h i
6. DE/rand/1/bin V(G+1) = V(1,G) + F · V(2,G) − V(3,G)
h i h i
7. DE/rand-to-best/1/bin V(G+1) = V(i,G) + λ· V(best,G) − V(i,G) + F · V(1,G) − V(2,G)
h i
8. DE/best/2/bin V(G+1) = V(best,G) + F V(1,G) + V(2,G) − V(3,G) − V(4,G)
h i
9. DE/rand/2/bin V(G+1) = V(5,G) + F V(1,G) + V(2,G) − V(3,G) − V(4,G)
One aspect common to all the mutation strategy methods is the base vector, which controls the
search direction. The difference vector provides a mutation rate term, such as a self-adaptive term, that
is added to an arbitrary or guided base vector to construct a trial individual. Over generations, the
individuals of a population reside in increasingly better positions and reform themselves. The various
combinations of these vectors can be categorized into four groups based on information pertaining to
the values gathered from the objective function: random, directed, local and hybrid.
The RAND approach consists of strategies in which the trial individual is produced without
knowledge of the value of the objective function. Similarly, the RAND/DIR approach includes
strategies that use the values of the objective function to determine a promising direction. Likewise,
the RAND/BEST approach applies the best individual approach to proceed with a trial. Additionally,
the RAND/BEST/DIR approach combines the last two groups into one that includes all of their
Appl. Sci. 2018, 8, 1945 6 of 25
collective benefits. However, a suitable direction is obtained by using the best individual to decrease
the search space and exploration time [23,24]. Thus, the “dir” and “dir-best” strategies, which use
objective function values to generate trial individuals, can produce an exploitation function. In fact,
the random selection of parents for a trial enhances exploration capabilities [25–27]. Thus, the locations
of individuals carry information about the fitness landscape. Therefore, an effective mutation strategy
that leads to uniform random vectors represents the entire search space well.
2.2. Initialization
DE is a population-based optimization technique that begins with the problem solution by
selecting the objective function at a random initial population. Predefined parameter bounds describe
the area from which the number of population (Np ) vectors in this initial population is chosen within
both the upper bound “bU ” and the lower bound “bL ”, where the subscripts L and U indicate lower
and upper, respectively. The following equation is used to develop a random number generator for
all vectors from within the predefined upper and lower bounds. The random function Random (0, 1)
produces a uniform random number within the range (0, 1).
2.3. Crossover
To balance the differential mutation search strategy, DE also applies uniform crossover to construct
trial vectors. A trial vector is constructed from values that have been copied from two diverse vectors.
In particular, DE crosses each vector as follows:
(
v j,i,g If(rand j (0, 1) ≤ Cr or j = jrand )
ui,g = u j,i,g =
x j,i,g
The crossover probability, Cr ∈ [0, 1], is predefined in the classic version of DE, and the fraction
value of the Cr control is cloned from the mutant vector. Cr is compared with a random number randj
(0, 1). If the random number is less than or equal to Cr , the trial parameter is inherited from the mutant
v j,i,g , otherwise, the parameter is cloned from the vector x j,i,g .
2.4. Selection
In this stage, we determine when the trial vector ui,g has an objective function value that is less
than or equal to that of its target vector xi,g . DE swaps the target vector in the next iteration, otherwise
the target retains its place in the population. This process is carried out by comparing each trial vector
with the target vector from which the parameters are cloned. After the population is updated, mutation,
recombination and selection are repeated until the optimum value is found or after a predefined stop
criterion is reached, such as a certain number of iterations.
(
ui,g If( f (ui,g ) ≤ f ( xi,g )
xi,g+1 =
xi,g
2.6. DE Applications
Due to the rapid rise of DE as a modest and strong optimizer, developers have applied the
technique in a wide range of domains and fields of technology
(http://www1.icsi.berkeley.edu/~storn/code.html). Yalcin proposed a new method for the 3D
Appl. Sci. 2018, 8, 1945 8 of 25
2.6. DE Applications
Due to the rapid rise of DE as a modest and strong optimizer, developers have applied the
technique in a wide range of domains and fields of technology (http://www1.icsi.berkeley.edu/
~storn/code.html). Yalcin proposed a new method for the 3D tracking of license plates from video
using a DE algorithm, which could be fine-tuned according to the license plate boundaries [28].
A color image quantization application using DE was proposed by Qinghua and Hu. The main
objective of image processing techniques during the color image quantization phase, is to decrease
the number of colors in an image with a low amount of deformation. DE can be used to adjust
colormaps and find the optimal candidate colormap [29]. With respect to the bidding market,
Alvaro et al. applied DE in developing a competitive electricity market application that finds the
optimal bids based on daily bidding activity [30]. Sickel et al. used DE in developing a power plant
control application for a reference governor to produce an optimal group of points for controlling
a power plant that was produced by [31]. Wang et al. proposed a flexible QoS multicast routing
algorithm for the next-generation Internet that improves the quality of service (QoS) of multicasts
to manage the increasing demand of network resources [32]. With respect to the electric power
systems industry, Ela et al. applied DE to determine the optimal power flow [33]. Goswami et al.
proposed a DE application for model-based well log-data inversion to discover features of earth
formations based on the dimensions of physical phenomena [34]. Another application applies network
system reconfiguration for distributing systems. The network reconfiguration application proposed by
Tzong and Lee involves the application of Improved Mixed-Integer Hybrid Differential Evolution [35].
Another DE application developed by Boughari et al. sets suitable controllers for aircraft stability and
control augmentation systems [36].
3. Parameter Control
As suggested in the previous section, the DE algorithm is a simple and effective optimization
algorithm for problems from the real world when its control parameters are properly set [8,37,38]. In
this section, we review the most current approaches for improving DE. First, the DE algorithm applies
certain control parameters to the system implementation. The accomplishment of DE is influenced
by the value of parameters, such as the crossover and mutation rate. Although some studies have
recommended certain values for these parameters, their effect on performance is complex and their
exact values are unclear. In particular, there is a wide variety of different recommended values that are
appropriate for different problems [39–41].
The mutation rate “F”, crossover rate “Cr ” and population dimension “N p” maintain balance
between exploration and exploitation [6]. Exploration is associated with finding new solutions, and
exploitation is associated with searching for new, suitable solutions; the two processes are linked in the
evolutionary search [42,43]. Therefore, the mutation and crossover rates influence the convergence
rate and the effectiveness of the search space [44].
However, specifying suitable values for these rates is not easy [45]. Three types of strategies are
used to set these parameter controls: deterministic parameter control (sometimes called random),
self-adaptive parameter control and adaptive parameter control [21,46–48]. Adaptive and self-adaptive
parameter control [16–25] have recently been proposed to dynamically alter the control parameters
without requiring the user’s prior knowledge or information about the problem behaviors throughout
the search process [49–53]. In the following sections, the self-adaptive parameter, the adaptive
parameter, and hybrid control strategies are discussed.
of a subtracted vector. Jitter multiplies each vector element of the subtracted vector by a different
scale factor, Fj . The rotation creates jitter using an essentially different procedure than the classic
DE’s constant mutation with F. However, this approach shows robustness for non-deceiving objective
functions [3]. Nonetheless, applied fixed values for each iteration, and F was created for each individual
within the range (0.4, 1), whereas the interval (0.5, 0.7) was selected for Cr [54,55].
Another approach is the composite DE (CoDE) algorithm proposed by Wang et al. In CoDE, a
trial vector is selected from a set of groups produced by utilizing diverse DE strategies [56]. The main
objective is to arbitrarily merge many trial vector strategies with different parameters at each iteration
to construct new trial vectors. These combinations help to successfully solve many problems. Wang
et al. used a group of trial vector strategies and a group of control parameter (almost three) to create
strategy and parameter candidate pools. The selected strategies are DE/rand/1/bin, DE/rand/2/bin
and DE/current-to- rand/1, and the three pair common choices for the control parameter settings were
(F = 1.0, Cr = 0.1), (F = 1.0; Cr = 0.9), and (F = 0.8; Cr = 0.2). In each generation, three different strategies
are applied, which randomly pick any of the control parameter values. Then, the trial vector is
designated the candidate with the better value of fitness. The parameters are chosen based on whether
they are frequently implemented with many DEs, and their performances are evaluated. The three pairs
of parameter settings that provide diverse effects produce new improved candidates. Furthermore, the
different values of the control parameters maintain different levels of search performance.
fact, despite its simplicity, DESAP performed better than DE in one of De Jong’s five exam problems,
whereas the other solutions are almost identical. DESAP represents an opportunity to reduce the control
parameters further by updating the size of the population, as is done with the additional parameters.
ns1 (ns2 + n f 2 )
p1 =
ns2 (ns1 + n f 1 ) + ns1 (ns2 + n f 2 )
p2 = 1 − p1
where nsi and n f i are the numbers of offspring vectors constructed by the ith i = (1, 2) strategy that
was a success or failure in the selection process over the last 50 generations. It is assumed that
this adaptation process can progressively develop the most appropriate mutation strategy at diverse
learning phases for a given problem. The mutation factors Fi are autonomously created at each iteration
based on a normal distribution, NR with a mean of 0.5 and a standard deviation of 0.3,
Fi = NR(0.5, 0.3)
The crossover rates Cri are autonomously formed based on a normal distribution with a mean of
Crm and a standard deviation of 0.1. The mean Crm approaches 0.5, is changed every 25 iterations and
is set to be the mean of the effective Cr over the previous 25 generations.
1 k
k k∑
Crm = Crsuc (k)
=1
where K is the counters of effective Cr values and Crsuc (k) indicates the kth value.
To accelerate the convergence, a local search technique (Quasi–Newton method) is applied to
respectable individuals after 200 generations. SaDE has been further developed by applying five
mutation strategies to resolve a group of constrained problems [60].
where di = v1 − v2 is the differential deviation, N (0.5, 0.5) means a Gaussian random number with an
average of 0.5 and a standard deviation of 0.5 and δ indicates a Cauchy random variable with a rate
parameter of t = 1.
Self-adaptive DE (SaDE) [8] was developed to resolve the issues related to control parameters and
learning technique. In SaDE, two DE learning strategies are chosen according to their performance.
The most appropriate learning technique and parameter values are increasingly self-adapted according
to the learning experience gained during evolution [61].
SaNSDE is an adaptive differential evolution algorithm that produces mutation vectors in a
manner similar to SaDE [61]. However, the difference is that the mutation factors are established based
on a normal distribution or a Cauchy distribution:
(
Normal distribution(0.5, 0.3), i f rand > f p
f =
Cauchy distribution(0.0, 1.0), otherwise
where the normal distribution (µ,σ2 ) indicates a random value of mean µ and variance σ2 and a
Cauchydistribution (µ, δ) indicates a random value with scale parameters µ and δ. The probability f p
of the spread over is adapted as follows:
ns1 (ns2 + n f 2 )
p1 =
ns2 (ns1 + n f 1 ) + ns1 (ns2 + n f 2 )
The crossover rate adaptation is similar to the method used in SaDE, but the factor Crm is changed
as a biased average of the successful values Crsuc every 25 iterations.
1 k ∆
Crm = ∑ k = 1
Wk Crsuc (k) Wk = k K
k ∑ k =1 ∆ K
where the weight is calculated with a positive improvement ∆ = f (x) − f (u) in the selection related to
each successful crossover rate CRsuc (k).
where random j = 1, 2, 3, 4 is the uniform random function ∈[0, 1]. The updated parameters are
implemented in the mutation and crossover processes to produce new, consistent vectors. This
mechanism updates the prior parameter with a new one only if the new vectors pass the selection
phase. However, jDE yields improved results with the classic DE/rand/1/ bin strategy.
Appl. Sci. 2018, 8, 1945 12 of 25
where Xgrbest ,G is the finest of the q% vectors arbitrarily selected from the existing generation, whereas
Xri ,G and Xri ,G are two distinctive vectors chosen randomly from the current population and are not
1 2
equal to Xgrbest ,G or the target.
In the p-best crossover process, each different random vector is chosen from the p best-ranking
vectors in the present population [67]. Then, a standard crossover is executed as per (5) between the
vector and the arbitrarily chosen one from the p-top vector to produce the trial vector with an identical
index. The variable p is linearly made smaller with the following generations as shown:
G−1
Np
p = ceil ·(1 − )
2 Gmax
where G is the present generation value, Gmax is the most extreme number of generations and Np is
the population number. The parameter adaption mechanism Fi is independently calculated as,
Crm = wCr ·Crm + (1 − wCr )·mean Pow (Crsuccess )wcr = 0.8 + 0.2 ∗ rand (0, 1)
n.
mean Pow (Crsuccess) = ∑ x∈Cr x 1
success (| Cr success |) 2
1
Fi ( g) =
1 + tanh(2A f i ( g))
exp (2z) − 1
tanh(z) =
exp (2z) + 1
0
Vi,g = Xi,g + Fi (Vbest, g − V1,g ) + Fi (V2,g −V3,G )
where Vbest,g is the best solution that is randomly chosen as one of the best individuals from the current
population [56]. Similarly, V1,g , V2,g and V3,g are randomly selected from the current population.
However, V3,G0 is also randomly chosen from the union between X and V .
i,g 1,g
0
V3,G = randomly (V1,g ∪ Xi,g )
JADE is also applied to the archiving process. Initially, the archive is unfilled and is added to the
parent solutions that fail in the selection process [70]. The purpose of the archive is to avoid calculation
Appl. Sci. 2018, 8, 1945 14 of 25
overhead. Moreover, the archive has a limited size; thus, if the size of the archive grows beyond R,
then a shrink operation is performed to reduce its size so that it does not exceed (α, NP).
The archive technique provides information about the direction required to improve the diversity
of the population. In addition, arbitrary F values can help expand population diversity [71].
Then, SCR is calculated, which represents the set of all effective crossover rates CRi . Furthermore,
the parameter µCR is updated in each iteration; this information is saved, and random information is
deleted from the archive file to keep its size R. µCR is calculated as follows:
Similarly, the mutation rate Fi is calculated using the Cauchy distribution (µ F , 0.1), with the
constraint that Fi = 1.
If Fi ≥ 1 or Fi ≤ 0 and µ f is initialized as 0.5, then
where S F indicates the set of all effective mutation rates Fi . Then, µ f is updated as follows:
µ f = (1 − c)·µ f + c·mean L (S F )
∑ F ∈S F F2
(S F ) =
∑ F ∈S F F
where randn(dim) T is a group of random numbers taken from a normal distribution with zero mean
and a standard deviation of 1 and has an element number equal to the dimensions of the function at
hand. The parameter “m” and the evolution of “σ” determine the overall standard deviation.
By using a shared population, the new mutated vectors are produced as the target vectors
as follows:
where xri ,g and xri ,g are two vectors randomly selected from the population, m is the average of the
1 2
present population, B is an orthonormal of eigenvectors, and D is the square root of the commensurate
none negative eigenvalues. P is a control value which maintains the contribution of the average vector
of the existing population and target ones as well. Both of the scale F and P are computed as follows:
Fi = 0.5 + 0.5·rand(0, 1)
iter
P = 0.5·(1 + )
iter − max
vi (t) = vi (t − 1) + ρ1 ∗ ( pi − xi (t − 1))
gbest represent the whole population and lbest describes the subpopulation encompassing the particle.
The gbest is practical for best results. Let pg be the best results of the population; thus, social influence
is mathematically expressed as ρ2 ∗ ( p g − xi (t − 1)). Therefore, updating the particles at each loop
as follows:
vi (t) = vi (t − 1) + ρ1 ∗ ( pi − xi (t − 1)) + ρ2 ∗ ( p g − xi (t − 1))
x i ( t ) = x i ( t − 1) + v i ( t )
where CR is the crossover rate, δd is the component of the subtract vector and β is a factor rate in
the range [0, 1]. Hendtlass proposed the first combination of DE and PSO and called it SDEA as the
individuals comply swarm principles [60]. DE is used to transfer the individuals to the promised
region in a random fashion. Yu et al. proposed an adaptive hybrid algorithm based on PSO and
Appl. Sci. 2018, 8, 1945 17 of 25
DE (HPSO-DE) with a composed population among PSO and DE [89]. The strategy incorporates the
advantages of the two algorithms and maintains population diversity. Therefore, HPSO-DE has the
ability to move to local optima [90]. Zhang et al. offered DEPSO, which apples a similar standard of
updating PSO individuals via DE [91]. DEPSO performs well with numerical integer problems but is
not efficient for small feasible space problems. Mutations are maintained by a DE operator on pi , with
a trail vector Ti = pi for the dth dimension:
where k is a random value within the domain [1, D], which includes that the mutation has at least one
dimension. CR is a crossover constant, and δ2 , is the case of N = 2 for the general difference vector δN ,
which is defined as follows:
N
δN = (∑ ∆)/N
1
∆ = PA − PB
where ∆ is the difference vector and PA and PB are randomly chosen from the p-best set. Liu et al.
offered a hybridization of PSO and DE in a pair of population scheme [92]. Three mutation strategies
are borrowed from DE (DE/rand/1, DE/current_to_best/1DE/rand/2) are applied to refresh the
former best solutions [93].
Trivedi et al. proposed a hybrid of DE and GA to resolve scheduling challenges [94]. GA
operates on the binary element variables through the DE process to enhance the related power-related
variables [95]. The advantage of a GA lies in its ability to discover a decent solution to a
problem whenever the iterative approach is too time-consuming and the mathematical approach
is unobtainable [96]. GA allows for the fast discovery of the solution. Although the genetic algorithm is
not excessively complex, the parameters and implementation of the GA generally require a tremendous
amount of tuning [97].
The advantage of DE is that, in general, it frequently shows better solutions than those yielded by
GA and other evolutionary algorithms [98–100]. Furthermore, DE is easy to apply to a wide variety of
problems regardless of noisy, multi-modal, multi-dimensional spaces, which typically make problems
difficult to optimize. Although DE consists of two important parameters, Cr and F, those parameters
do not require the same amount of tuning as those associated with other evolutionary algorithms [101].
Liao has proposed a hybridization of DE and a local search algorithm modeled after the harmony
search (HS) algorithm to find the global optimum [102]. The main goal of this type of hybridization
method is to advance the use of mixed discrete and real-valued-dimensional problems.
Boussaïd et al. proposed a hybridization of DE and biogeography based optimization (BBO)
to deliver solutions through the optimal power distribution method in a wireless sensor network
(WSN) [103,104].
Dulikravich et al. proposed a hybridized multi-objective, multi variable optimizer by combining
non-dominated sorting differential evolution (NSDE) with the strength Pareto evolutionary algorithm
(SPEA) and multi-objective particle swarm optimization (MOPSO) [105].
Guo and others have proposed a form of DE enhanced among self-adaptive parameters that
depend on simulated annealing algorithms in the collection of DE; the classic selection technique
is a greedy equation [106]. The greedy rule is easily trapped in a local optimum. However, a new
selection technique based on simulated annealing is used in this algorithm. The approach is expressed
as follows:
Xi,G+1 = Ui,G i f f (Ui,G ) ≤ f ( Xi,G )
f (Ui,G ) − f ( Xi,G )
Xi,G+1 = Ui,G i f f (Ui,G ) f ( Xi,G ) and exp (− )rand (0, 1)
tG
Appl. Sci. 2018, 8, 1945 18 of 25
f (Ui,G ) − f ( Xi,G )
Xi,G+1 = Xi,G i f f (Ui,G ) f ( Xi,G ) and exp (− ) ≤ rand (0, 1)
tG
where tG represents the Gth generation temperature. Pholdee and Bureerat offered a hybrid algorithm
involving the trial vector method of DE called the real-coded population-based incremental learning
(RCPBIL) algorithm [107]. The RPBIL can be extended to multi-objective optimization similarly to
multi-objective PBIL using binary codes for which the population serves as a likelihood vector for
single-objective problems [108]. When addressing multi-objective problems, more probability vectors
are utilized to maintain population variety. Likewise, with the binary code of PBIL, the multi-objective
style of the RPBIL uses numerous possibility matrix that appear for a real code population, where each
probability matrix is called a tray [109].
A three-dimensional matrix, represents a group of probability trays with dimensions n*nl *n T , n T
is the number of trays required for each tray drive to be used to produce a real-code subpopulation,
which has approximately NP /NT form results as its members.
An initial population is formed for the multi-starting search procedure with early likelihood
trays. An initial Pareto archive is obtained, and non-dominated results are then designated to update
the probability trays. Then, the centroid of the non-dominated solution set (RG ) is used to update a
probability tray in the series, where the rG of the set that has the lowest value of the first objective
function is applied to update the first tray and so on.
The updating procedure for each tray can be improved by substituting Xbest with rG . Subsequently,
a population yielding the updated trays is shaped. The Pareto archive is changed by substituting its
members with non-dominated solutions saved from the mixture of the current population and the
elements in the preceding archive. If the number of archive elements is larger than the constant archive
size, the clustering method is initiated to eliminate non-dominated solutions from the archive. These
steps are repeated until a stopping condition is fulfilled [110].
Neri et al. [111,112] proposed a compact DE hybridized with a memetic search to yield faster
convergence [113]. The algorithm represents the population as a multi-dimensional Gaussian
distribution and is called disturbed exploitation compact differential evolution (DEcDE) [114]. The
DEcDE algorithm utilizes an evolutionary framework based on DE logic assisted by a shallow depth
for processing the local search algorithm [114].
The output of the algorithm was introduced to create an MC model to gain high efficiency on a
diverse set of problems, regardless of its limits, in terms of complexity and memory usage. At the start
of the DEcDE algorithm, an a(2.n) probability vector (PV) is produced.
PV t = µt , σt
where µ and σ are, respectively, the mean and standard deviation values for each design variable from
a Gaussian probability distribution function (PDF) truncated within the interval [−1, 1].
PDF = [µ[i ], σ [i ]]
Zhan and Zhang proposed a differential evolution (DE) algorithm with a random walk
(DE-RW) [115]. DE-RW is analogous to the classic DE algorithm, with a minor alteration in the
crossover procedure that mixes the individual vector and the mutant vector to perform a random walk,
forming the target vector as follows:
vid rand(0, 1) > CR or d = k
rand( Ld , Hd ), else if rand(0, 1) > RW
xid otherwise
where Ld and Hd are the low and high search restrictions of the dht dimension and the parameter RW is
used to control the effect of the random walk. The parameter RW is controlled as follows:
Appl. Sci. 2018, 8, 1945 19 of 25
where g and G are the current generation number and the maximum number of generations,
respectively. A few notable DE algorithms are summarized in Table 2.
4. Conclusions
This paper presents a survey of DE and analyzes the modifications recently proposed in the
literature to enhance the performance of DE. Various state-of–the-art differential evolution algorithms
incorporating different strategies have been studied. DE performance is affected by the types of
techniques applied. Enhanced DE algorithms are categorized into three groups: adaptive, self-adaptive
and hybrid. However, as demonstrated in this article, seemingly minor modifications can greatly
enhance the performance of DE. All the DE techniques and mechanisms described here offer partial
improvements to the classic DE components to yield better performance. The selection of proper
techniques in DE is associated with the mathematical characteristics of the problems addressed.
Therefore, by properly adjusting the control parameters of different crossovers, similar levels of
performance can be attained. This article offers an essential guide to differential evolution practitioners
for acquiring optimum control parameters for DE. Thus, if information about the problem to be
addressed is available, suitable optimal values should be adopted based on the basic guidelines and
findings summarized in this paper. Researchers interested in DE can find valuable contributions in the
literature about DE.
Appl. Sci. 2018, 8, 1945 20 of 25
Author Contributions: Conceptualization of this paper survey which is carried to gain extensive ideas about
DE and verities of techniques T.E created the structure and organization of the paper, reviewed and instituted
the content in all sections and commented on the quantitative aspects of the DE algorithm. A.M. co-reviewed
the paper. T.E. contributed to the final version of the manuscript. All authors approve of the final version of
the manuscript.
Acknowledgments: The authors acknowledged the University of Bridgeport for the preparation of resources
which are required to complete this research. The University of Bridgeport library contributes by arranging
required books and researched papers in this work. The University of Bridgeport, CT USA also pledge us by
funding the cost of publishing this paper.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Das, S.; Suganthan, P.N. Differential evolution: A survey of the state-of-the-art. IEEE Trans. Evol. Comput.
2011, 15, 4–31. [CrossRef]
2. Rao, S.S.; Rao, S.S. Engineering Optimization: Theory and Practice; John Wiley & Sons: Hoboken, NJ, USA, 2009.
3. Price, K.; Storn, R.M.; Lampinen, J.A. Differential Evolution: A Practical Approach to Global Optimization;
Springer Science & Business Media: Berlin, Germany, 2006.
4. Brownlee, J. Clever Algorithms: Nature-Inspired Programming Recipes; Jason Brownlee: Melbourne, Australia, 2011.
5. Zhang, J.; Sanderson, A.C. Adaptive Differential Evolution; Springer: Berlin/Heidelberg, Germany, 2009.
6. Feoktistov, V. Differential Evolution; Springer: Dordrecht, The Nerthelands, 2006.
7. Storn, R. On the usage of differential evolution for function optimization. In Proceedings of the NAFIPS,
Biennial Conference of the North American Fuzzy Information Processing Society, Berkeley, CA, USA, 19–22
June 1996; pp. 519–523.
8. Price, K.V.; Storn, R.M.; Lampinen, J.A. The differential evolution algorithm. In Differential Evolution:
A Practical Approach to Global Optimization; Price, K.V., Storn, R.M., Lampinen, J.A., Eds.; Springer:
Berlin/Heidelberg, Germany, 2005; pp. 37–134.
9. Salman, A.; Engelbrecht, A.P.; Omran, M.G. Empirical analysis of self-adaptive differential evolution. Eur. J.
Oper. Res. 2007, 183, 785–804. [CrossRef]
10. Neri, F.; Tirronen, V. Recent advances in differential evolution: A survey and experimental analysis.
Artif. Intell. Rev. 2010, 33, 61–106. [CrossRef]
11. Onwubolu, G.C.; Davendra, D. Differential Evolution: A Handbook for Global Permutation-based Combinatorial
Optimization; Springer Science & Business Media: Berlin, Germany, 2009; Volume 175.
12. Peng, L.; Wang, Y. Differential evolution using uniform-quasi-opposition for initializing the population.
Inf. Technol. J. 2010, 9, 1629–1634. [CrossRef]
13. Adeyemo, J.; Otieno, F. Differential evolution algorithm for solving multi-objective crop planning model.
Agric. Water Manag. 2010, 97, 848–856. [CrossRef]
14. Chang, T.-T.; Chang, H.-C. Application of differential evolution to passive shunt harmonic filter planning.
In Proceedings of the 8th International Conference on Harmonics and Quality of Power Proceedings, Athens,
Greece, 14–16 October 1998; pp. 149–153.
15. Bergey, P.K.; Ragsdale, C. Modified differential evolution: A greedy random strategy for genetic
recombination. Omega 2005, 33, 255–265. [CrossRef]
16. Fan, H.-Y.; Lampinen, J. A trigonometric mutation operation to differential evolution. J. Glob. Optim. 2003,
27, 105–129. [CrossRef]
17. Das, S.; Abraham, A.; Chakraborty, U.K.; Konar, A. Differential evolution using a neighborhood-based
mutation operator. IEEE Trans. Evol. Comput. 2009, 13, 526–553. [CrossRef]
18. Qing, A. Differential Evolution: Fundamentals and Applications in Electrical Engineering; John Wiley & Sons:
Singapore, 2009.
19. Lin, C.; Qing, A.; Feng, Q. A comparative study of crossover in differential evolution. J. Heurist. 2011, 17,
675–703. [CrossRef]
20. Guo, S.-M.; Yang, C.-C.; Hsu, P.-H.; Tsai, J.S.-H. Improving differential evolution with a successful-parent-selecting
framework. IEEE Trans. Evol. Comput. 2015, 19, 717–730. [CrossRef]
21. Eiben, Á.E.; Hinterding, R.; Michalewicz, Z. Parameter control in evolutionary algorithms. IEEE Trans.
Evol. Comput. 1999, 3, 124–141. [CrossRef]
Appl. Sci. 2018, 8, 1945 21 of 25
22. Rozenberg, G.; Bäck, T.; Eiben, A.E.; Kok, J.N.; Spaink, H.P. (Eds.) Natural Computing Series; Springer: Berlin,
Germany, 2006.
23. Wolpert, D.H.; Macready, W.G. The Mathematics of Search; Technical Report; Santa Fe Institute: Santa Fe, NM,
USA, 1995.
24. Wolpert, D.H.; Macready, W.G. No free lunch theorems for optimization. IEEE Trans. Evol. Comput. 1997, 1,
67–82. [CrossRef]
25. Zaharie, D. On the explorative power of differential evolution. In Proceedings of the 3rd International
Workshop on Symbolic and Numerical Algorithms on Scientific Computing, SYNASC-2001, Timişoara,
Romania, 2–4 October 2001.
26. Liu, J. On setting the control parameter of the differential evolution method. In Proceedings of the 8th
International Conference on Soft Computing (MENDEL 2002), Brno, Czech Republic, 5–7 June 2002;
pp. 11–18.
27. Šmuc, T. Improving convergence properties of the differential evolution algorithm. In Proceedings of the
MENDEL 2002-8th International Conference on Soft Computing, Brno, Czech Republic, 5–7 June 2002.
28. Yalcin, I.K.; Gokmen, M. Integrating differential evolution and condensation algorithms for license plate
tracking. In Proceedings of the ICPR 2006, 18th International Conference on Pattern Recognition, Hong
Kong, China, 20–24 August 2006; pp. 658–661.
29. Zhang, J.; Sanderson, A.C. JADE: Adaptive differential evolution with optional external archive. IEEE Trans.
Evol. Comput. 2009, 13, 945–958. [CrossRef]
30. Baíllo, Á.; Ventosa, M.; Rivier, M.; Ramos, A. Strategic bidding in a competitive electricity market:
A decomposition approach. In Proceedings of the IEEE Porto Power Tech Proceedings, Porto, Portugal,
10–13 September 2001; Volume 1, p. 6.
31. Van Sickel, J.H.; Lee, K.Y.; Heo, J.S. Differential evolution and its applications to power plant control. In
Proceedings of the ISAP, International Conference on Intelligent Systems Applications to Power Systems,
Niigata, Japan, 5–8 November 2007; pp. 1–6.
32. Wang, X.; Cheng, H.; Huang, M. QoS multicast routing protocol oriented to cognitive network using
competitive coevolutionary algorithm. Expert Syst. Appl. 2014, 41, 4513–4528. [CrossRef]
33. El Ela, A.A.; Abido, M.; Spea, S. Optimal power flow using differential evolution algorithm. Electr. Power
Syst. Res. 2010, 80, 878–885. [CrossRef]
34. Goswami, J.C.; Mydur, R.; Wu, P. Application of differential evolution algorithm to model-based well
log-data inversion. In Proceedings of the IEEE Antennas and Propagation Society International Symposium,
San Antonio, TX, USA, 16–21 June 2002; pp. 318–321.
35. Su, C.-T.; Lee, C.-S. Network reconfiguration of distribution systems using improved mixed-integer hybrid
differential evolution. IEEE Trans. Power Deliv. 2003, 18, 1022–1027. [CrossRef]
36. Boughari, Y.; Ghazi, G.; Botez, R.M.; Theel, F. New Methodology for Optimal Flight Control Using Differential
Evolution Algorithms Applied on the Cessna Citation X Business Aircraft—Part 1. Design and Optimization.
INCAS Bull. 2017, 9, 31.
37. Price, K.V. Differential evolution vs. the functions of the 2/sup nd/ICEO. In Proceedings of the
IEEE International Conference on Evolutionary Computation, Indianapolis, IN, USA, 13–16 April 1997;
pp. 153–157.
38. Xue, F.; Sanderson, A.C.; Bonissone, P.P.; Graves, R.J. Fuzzy logic controlled multi-objective differential
evolution. In Proceedings of the 14th IEEE International Conference on Fuzzy Systems, FUZZ ’05, Reno, NV,
USA, 25 May 2005; pp. 720–725.
39. Storn, R. Differrential Evolution—A Simple and Efficient Adaptive Scheme for Global Optimization over Continuous
Spaces; Technical Report; International Computer Science Institute: Berkeley, CA, USA, 1995.
40. Mezura-Montes, E.; Velázquez-Reyes, J.; Coello Coello, C.A. A comparative study of differential evolution
variants for global optimization. In Proceedings of the 8th Annual Conference on Genetic and Evolutionary
Computation, Seattle, WA, USA, 8–12 July 2006; pp. 485–492.
41. Vesterstrom, J.; Thomsen, R. A comparative study of differential evolution, particle swarm optimization, and
evolutionary algorithms on numerical benchmark problems. In Proceedings of the Congress on Evolutionary
Computation, CEC2004, Portland, OR, USA, 19–23 June 2004; pp. 1980–1987.
42. Fister, I.; Mernik, M.; Brest, J. Hybridization of Evolutionary Algorithms. arXiv 2013, arXiv:1301.0929.
Appl. Sci. 2018, 8, 1945 22 of 25
43. Hu, C.; Yan, X. An immune self-adaptive differential evolution algorithm with application to estimate kinetic
parameters for homogeneous mercury oxidation. Chin. J. Chem. Eng. 2009, 17, 232–240. [CrossRef]
44. Ilonen, J.; Kamarainen, J.-K.; Lampinen, J. Differential evolution training algorithm for feed-forward neural
networks. Neural Process. Lett. 2003, 17, 93–105. [CrossRef]
45. Eiben, G.; Schut, M.C. New ways to calibrate evolutionary algorithms. In Advances in Metaheuristics for Hard
Optimization; Siarry, P., Michalewicz, Z., Eds.; Springer: Berlin/Heidelberg, Germany, 2007; pp. 153–177.
46. Angeline, P.J. Adaptive and self-adaptive evolutionary computations. In Computational Intelligence: A Dynamic
Systems Perspective; IEEE Press: Piscataway, NJ, USA, 1995.
47. Eiben, A.E.; Smith, J.E. Introduction to Evolutionary Computing; Springer: Berlin, Germany; London, UK, 2003;
Volume 53.
48. Liu, J.; Lampinen, J.; Matousek, R.; Osmera, P. Adaptive parameter control of differential evolution. In
Proceedings of the MENDEL, Brno, Czech Republic, 5–7 June 2002; pp. 19–26.
49. Abbass, H.A. The self-adaptive pareto differential evolution algorithm. In Proceedings of the 2002 Congress
on Evolutionary Computation, CEC ’02, Honolulu, HI, USA, 12–17 May 2002; pp. 831–836.
50. Teo, J. Exploring dynamic self-adaptive populations in differential evolution. Soft Comput. 2006, 10, 673–686.
[CrossRef]
51. Liu, J.; Lampinen, J. A fuzzy adaptive differential evolution algorithm. Soft Comput. 2005, 9, 448–462.
[CrossRef]
52. Qin, A.K.; Suganthan, P.N. Self-adaptive differential evolution algorithm for numerical optimization. In
Proceedings of the 2005 IEEE Congress on Evolutionary Computation, Edinburgh, UK, 2–5 September 2005;
pp. 1785–1791.
53. Brest, J.; Greiner, S.; Boskovic, B.; Mernik, M.; Zumer, V. Self-adapting control parameters in differential
evolution: A comparative study on numerical benchmark problems. IEEE Trans. Evol. Comput. 2006, 10,
646–657. [CrossRef]
54. Kaelo, P.; Ali, M. Differential evolution algorithms using hybrid mutation. Comput. Optim. Appl. 2007, 37,
231–246. [CrossRef]
55. Rocha, A.M.A.; Fernandes, E.M.d.G. On charge effects to the electromagnetism-like algorithm. In Proceedings
of the 20th EURO Mini Conference: Continuous Optimization and Knowledge-Based Technologies, Neringa,
Lithuania, 20–23 May 2008; pp. 198–203.
56. Wang, Y.; Cai, Z.; Zhang, Q. Differential evolution with composite trial vector generation strategies and
control parameters. IEEE Trans. Evol. Comput. 2011, 15, 55–66. [CrossRef]
57. Zaharie, D. Control of population diversity and adaptation in differential evolution algorithms In Proceedings
of the Mendel, 9th International Conference on Soft Computing, Brno, Czech Republic, 26–28 June 2003.
58. Kumar, P.; Pant, M. A self adaptive differential evolution algorithm for global optimization. In Proceedings
of the International Conference on Swarm, Evolutionary, and Memetic Computing, Chennai, India,
16–18 December 2010; pp. 103–110.
59. Brest, J.; Bošković, B.; Greiner, S.; Žumer, V.; Maučec, M.S. Performance comparison of self-adaptive and
adaptive differential evolution algorithms. Soft Comput. 2007, 11, 617–629. [CrossRef]
60. Hendtlass, T. A combined swarm differential evolution algorithm for optimization problems. In Proceedings
of the International Conference on Industrial, Engineering and Other Applications of Applied Intelligent
Systems, Budapest, Hungary, 4–7 June 2001; pp. 11–18.
61. Yang, Z.; Tang, K.; Yao, X. Differential evolution for high-dimensional function optimization. In Proceedings
of the IEEE Congress on Evolutionary Computation (CEC 2007), Singapore, 25–28 September 2007;
pp. 3523–3530.
62. Ling, M.-X.; Wang, F.-Y.; Ding, X.; Hu, Y.-H.; Zhou, J.-B.; Zartman, R.E.; Yang, X.-Y.; Sun, W. Cretaceous ridge
subduction along the lower Yangtze River belt, eastern China. Econ. Geol. 2009, 104, 303–321. [CrossRef]
63. Babu, B.; Angira, R. Modified differential evolution (MDE) for optimization of non-linear chemical processes.
Comput. Chem. Eng. 2006, 30, 989–1002. [CrossRef]
64. Sayah, S.; Zehar, K. Modified differential evolution algorithm for optimal power flow with non-smooth cost
functions. Energy Convers. Manag. 2008, 49, 3036–3042. [CrossRef]
65. Lakshminarasimman, L.; Subramanian, S. Short-term scheduling of hydrothermal power system with
cascaded reservoirs by using modified differential evolution. IEE Proc.-Gener. Transm. Distrib. 2006, 153,
693–700. [CrossRef]
Appl. Sci. 2018, 8, 1945 23 of 25
66. Islam, S.M.; Das, S.; Ghosh, S.; Roy, S.; Suganthan, P.N. An adaptive differential evolution algorithm with
novel mutation and crossover strategies for global numerical optimization. IEEE Trans. Syst. Man Cybern.
Part B Cybern. 2012, 42, 482–500. [CrossRef] [PubMed]
67. Cai, Y.; Wang, J. Differential evolution with neighborhood and direction information for numerical
optimization. IEEE Trans. Cybern. 2013, 43, 2202–2215. [CrossRef] [PubMed]
68. Alguliev, R.M.; Aliguliyev, R.M.; Isazade, N.R. DESAMC+ DocSum: Differential evolution with self-adaptive
mutation and crossover parameters for multi-document summarization. Knowl.-Based Syst. 2012, 36, 21–38.
[CrossRef]
69. Selamat, A.; Nguyen, N.T.; Haron, H. Intelligent Information and Database Systems. In Proceedings of the 5th
Asian Conference, ACIIDS 2013, Kuala Lumpur, Malaysia, 18–20 March 2013; Springer: Berlin/Heidelberg,
German, 2013; Volume 7803.
70. Zhang, J.; Sanderson, A.C. JADE: Self-adaptive differential evolution with fast and reliable convergence
performance. In Proceedings of the IEEE Congress on Evolutionary Computation (CEC 2007), Singapore,
25–28 September 2007; pp. 2251–2258.
71. Tanabe, R.; Fukunaga, A. Success-history based parameter adaptation for differential evolution.
In Proceedings of the 2013 IEEE Congress on Evolutionary Computation (CEC), Cancun, Mexico, 20–23 June
2013; pp. 71–78.
72. Ghosh, S.; Das, S.; Roy, S.; Islam, S.M.; Suganthan, P.N. A differential covariance matrix adaptation
evolutionary algorithm for real parameter optimization. Inf. Sci. 2012, 182, 199–219. [CrossRef]
73. Ghosh, S.; Roy, S.; Islam, S.M.; Das, S.; Suganthan, P.N. A differential covariance matrix adaptation
evolutionary algorithm for global optimization. In Proceedings of the 2011 IEEE Symposium on Differential
Evolution (SDE), Paris, France, 11–15 April 2011; pp. 1–8.
74. Elsayed, S.M.; Sarker, R.A.; Essam, D.L. An improved self-adaptive differential evolution algorithm for
optimization problems. IEEE Trans. Ind. Inform. 2013, 9, 89–99. [CrossRef]
75. Lichtblau, D. Relative position indexing approach. In Differential Evolution: A Handbook for Global
Permutation-Based Combinatorial Optimization; Springer: Berlin/Heidelberg, Germany, 2009; pp. 81–120.
76. Blum, C.; Puchinger, J.; Raidl, G.R.; Roli, A. Hybrid metaheuristics in combinatorial optimization: A survey.
Appl. Soft Comput. 2011, 11, 4135–4151. [CrossRef]
77. Dragoi, E.-N.; Dafinescu, V. Parameter control and hybridization techniques in differential evolution:
A survey. Artif. Intell. Rev. 2016, 45, 447–470. [CrossRef]
78. Goldberg, D.E.; Holland, J.H. Genetic algorithms and machine learning. Mach. Learn. 1988, 3, 95–99. [CrossRef]
79. Vas, P. Artificial-Intelligence-Based Electrical Machines and Drives: Application of Fuzzy, Neural, Fuzzy-Neural, and
Genetic-Algorithm-Based Techniques; Oxford University Press: Oxford, UK, 1999; Volume 45.
80. Panigrahi, B.K.; Suganthan, P.N.; Das, S.; Dash, S.S. Swarm, Evolutionary, and Memetic Computing. In
Proceedings of the Third International Conference SEMCCO, Chennai, India, 16–18 December 2010.
81. Nwankwor, E.; Nagar, A.K.; Reid, D. Hybrid differential evolution and particle swarm optimization for
optimal well placement. Comput. Geosci. 2013, 17, 249–268. [CrossRef]
82. Vitaliy, F. Differential Evolution—In Search of Solutions; Springer: New York, NY, USA, 2006.
83. Fister, I.; Fister, I., Jr. Adaptation and Hybridization in Computational Intelligence; Springer: Cham, Switzerland,
2015; Volume 18.
84. Thangaraj, R.; Pant, M.; Abraham, A.; Bouvry, P. Particle swarm optimization: Hybridization perspectives
and experimental illustrations. Appl. Math. Comput. 2011, 217, 5208–5226. [CrossRef]
85. Eberhart, R.; Kennedy, J. Particle swarm optimization. In Proceedings of the IEEE International Conference
on Neural Network, Perth, Australia, 27 November–1 December 1995; pp. 1942–1948.
86. Shi, Y.; Eberhart, R.C. Empirical study of particle swarm optimization. In Proceedings of the Congress on
Evolutionary Computation, CEC ’99, Washington, DC, USA, 6–9 July 1999; pp. 1945–1950.
87. Trelea, I.C. The particle swarm optimization algorithm: Convergence analysis and parameter selection.
Inf. Process. Lett. 2003, 85, 317–325. [CrossRef]
88. Kennedy, J.; Mendes, R. Population structure and particle swarm performance. In Proceedings of the 2002
Congress on Evolutionary Computation, CEC ’02, Honolulu, HI, USA, 12–17 May 2002; pp. 1671–1676.
Appl. Sci. 2018, 8, 1945 24 of 25
89. He, Z.; Wei, C.; Yang, L.; Gao, X.; Yao, S.; Eberhart, R.C.; Shi, Y. Extracting rules from fuzzy neural network
by particle swarm optimisation. In Proceedings of the IEEE International Conference on Evolutionary
Computation Proceedings. IEEE World Congress on Computational Intelligence, Anchorage, AK, USA,
4–9 May 1998; pp. 74–77.
90. Das, S.; Abraham, A.; Konar, A. Particle swarm optimization and differential evolution algorithms: Technical
analysis, applications and hybridization perspectives. In Advances of Computational Intelligence in Industrial
Systems; Springer: Berlin/Heidelberg, Germany, 2008; pp. 1–38.
91. Van Sickel, J.H.; Lee, K.Y.; Heo, J.S. Differential evolution and its applications to power plant control.
In Proceedings of the International Conference on Intelligent Systems Applications to Power Systems
(ISAP 2007), Niigata, Japan, 5–8 November 2007; pp. 1–6.
92. Yu, X.; Cao, J.; Shan, H.; Zhu, L.; Guo, J. An adaptive hybrid algorithm based on particle swarm optimization
and differential evolution for global optimization. Sci. World J. 2014, 2014, 215472. [CrossRef] [PubMed]
93. Zhang, W.-J.; Xie, X.-F. DEPSO: Hybrid particle swarm with differential evolution operator. In Proceedings
of the IEEE International Conference on Systems, Man and Cybernetics, Washington, DC, USA, 8 October
2003; pp. 3816–3821.
94. Liu, B. Uncertain risk analysis and uncertain reliability analysis. J. Uncertain Syst. 2010, 4, 163–170.
95. Liu, H.; Cai, Z.; Wang, Y. Hybridizing particle swarm optimization with differential evolution for constrained
numerical and engineering optimization. Appl. Soft Comput. 2010, 10, 629–640. [CrossRef]
96. Hästbacka, J.; de la Chapelle, A.; Mahtani, M.M.; Clines, G.; Reeve-Daly, M.P.; Daly, M.; Hamilton, B.A.;
Kusumi, K.; Trivedi, B.; Weaver, A. The diastrophic dysplasia gene encodes a novel sulfate transporter:
Positional cloning by fine-structure linkage disequilibrium mapping. Cell 1994, 78, 1073–1087. [CrossRef]
97. Das, S.; Abraham, A.; Konar, A. Automatic clustering using an improved differential evolution algorithm.
IEEE Trans. Syst. Man Cybern. Part A Syst. Hum. 2008, 38, 218–237. [CrossRef]
98. Goldberg, D.E. Genetic Algorithms in Search, Optimization and Machine Learning; Addison: Reading, MA, USA, 1989.
99. Mezura-Montes, E. Nature-Inspired Algorithms Evolutionary and Swarm Intelligence Approaches. In
Proceedings of the 7th Mexican Internatioal Conference of Arifical Intelligence “MICAI“, Instituto Tecnol
ogico de Monterrey, Monterrey, Mexico, 27–31 October 2008.
100. Xu, X.; Li, Y. Comparison between particle swarm optimization, differential evolution and multi-parents
crossover. In Proceedings of the International Conference on Computational Intelligence and Security,
Harbin, China, 15–19 December 2007; pp. 124–127.
101. Codreanu, I. A parallel between differential evolution and genetic algorithms with exemplification in
a microfluidics optimization problem. In Proceedings of the International Semiconductor Conference
(CAS 2005), Sinaia, Romania, 3–5 October 2005; pp. 421–424.
102. Sentinella, M.R. Comparison and integrated use of differential evolution and genetic algorithms for space
trajectory optimisation. In Proceedings of the IEEE Congress on Evolutionary Computation (CEC 2007),
Singapore, 25–28 September 2007; pp. 973–978.
103. Hegerty, B.; Hung, C.-C.; Kasprak, K. A comparative study on differential evolution and genetic algorithms
for some combinatorial problems. In Proceedings of the 8th Mexican International Conference on Artificial
Intelligence, Seattle, WA, USA, 8–12 July 2006.
104. Liao, T.W. Two hybrid differential evolution algorithms for engineering design optimization. Appl. Soft Comput.
2010, 10, 1188–1199. [CrossRef]
105. Boussaïd, I.; Chatterjee, A.; Siarry, P.; Ahmed-Nacer, M. Two-stage update biogeography-based optimization
using differential evolution algorithm (DBBO). Comput. Oper. Res. 2011, 38, 1188–1198. [CrossRef]
106. Boussaïd, I.; Chatterjee, A.; Siarry, P.; Ahmed-Nacer, M. Hybridizing biogeography-based optimization with
differential evolution for optimal power allocation in wireless sensor networks. IEEE Trans. Veh. Technol.
2011, 60, 2347–2353. [CrossRef]
107. Moral, R.; Sahoo, D.; Dulikravich, G. Multi-objective hybrid evolutionary optimization with automatic
switching. In Proceedings of the 11th AIAA/ISSMO Multidisciplinary Analysis and Optimization
Conference, Portsmouth, VA, USA, 6–8 September 2006.
108. Guo, H.; Li, Y.; Li, J.; Sun, H.; Wang, D.; Chen, X. Differential evolution improved with self-adaptive control
parameters based on simulated annealing. Swarm Evol. Comput. 2014, 19, 52–67. [CrossRef]
Appl. Sci. 2018, 8, 1945 25 of 25
109. Pholdee, N.; Bureerat, S.; Yıldız, A.R. Hybrid real-code population-based incremental learning and
differential evolution for many-objective optimisation of an automotive floor-frame. Int. J. Veh. Des.
2017, 73, 20–53. [CrossRef]
110. Pholdee, N.; Bureerat, S. Hybridisation of real-code population-based incremental learning and differential
evolution for multiobjective design of trusses. Inf. Sci. 2013, 223, 136–152. [CrossRef]
111. Pholdee, N.; Bureerat, S. Hybrid real-code population-based incremental learning and approximate gradients
for multi-objective truss design. Eng. Optim. 2014, 46, 1032–1051. [CrossRef]
112. Bureerat, S.; Pholdee, N.; Park, W.-W.; Kim, D.-K. An Improved Teaching-Learning Based Optimization for
Optimization of Flatness of a Strip During a Coiling Process. In Proceedings of the International Workshop
on Multi-disciplinary Trends in Artificial Intelligence, Chiang Mai, Thailand, 7–9 December 2016; pp. 12–23.
113. Neri, F.; Cotta, C. Memetic algorithms and memetic computing optimization: A literature review. Swarm Evol.
Comput. 2012, 2, 1–14. [CrossRef]
114. Zou, D.; Wu, J.; Gao, L.; Li, S. A modified differential evolution algorithm for unconstrained optimization
problems. Neurocomputing 2013, 120, 469–481. [CrossRef]
115. Zhan, Z.-H.; Zhang, J. Enhance differential evolution with random walk. In Proceedings of the ACM
14th Annual Conference Companion on Genetic and Evolutionary Computation, Philadelphia, PA, USA,
7–11 July 2012; pp. 1513–1514.
116. Yu, W.-J.; Zhang, J. Multi-population differential evolution with adaptive parameter control for global
optimization. In Proceedings of the 13th Annual Conference on Genetic and Evolutionary Computation,
Dublin, Ireland, 12–16 July 2011; pp. 1093–1098.
117. Bujok, P.; Tvrdik, J.; Polakova, R. Differential evolution with rotation-invariant mutation and competing-strategies
adaptation. In Proceedings of the 2014 IEEE Congress on Evolutionary Computation (CEC), Beijing, China,
6–11 July 2014; pp. 2253–2258.
118. Trivedi, A.; Srinivasan, D.; Biswas, S.; Reindl, T. A genetic algorithm–differential evolution based hybrid
framework: Case study on unit commitment scheduling problem. Inf. Sci. 2016, 354, 275–300. [CrossRef]
119. Pant, M.; Thangaraj, R.; Grosan, C.; Abraham, A. Hybrid differential evolution-particle swarm optimization
algorithm for solving global optimization problems. In Proceedings of the ICDIM 2008, Third International
Conference on Digital Information Management, London, UK, 13–16 November 2008; pp. 18–24.
© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).