Evolutionary Algorithms Literature Research Articles

A new proximity metric based on optimality conditions for single and multi-objective optimization: Method and validation

With the widespread application of Evolutionary Algorithms (EAs), their performance needs to be evaluated using more than the usual performance metrics. In the EA literature, various metrics assess the convergence ability of these algorithms. However, many of them require prior knowledge of the Pareto-optimal front. Recently, two Karush–Kuhn–Tucker Proximity Metrics (KKTPMs) have been introduced to measure convergence without needing prior knowledge. One relies on the Augmented Achievement Scalarization Function (AASF) method (AASF-KKTPM), and the other on Benson’s method (B-KKTPM). However, both require specific parameters and reference points, making them computationally expensive. In this paper, we introduce a novel version of KKTPM applicable to single-, multi-, and many-objective optimization problems, utilizing the Penalty-based Boundary Intersection (PBI) method (PBI-KKTPM). Additionally, we introduce an approximate approach to reduce the computational burden of solving PBI-KKTPM optimization problems. Through extensive computational experiments across 23 case studies, our proposed metric demonstrates a significant reduction in computational cost, ranging from 20.68% to 60.03% compared to the computational overhead associated with the AASF-KKTPM metric, and from 16.48% to 61.15% compared to the computational overhead associated with the B-KKTPM metric. Noteworthy features of the proposed metric include its independence from knowledge of the true Pareto-optimal front and its applicability as a termination criterion for EAs. Another feature of the proposed metric is its ability to deal with black box problems very efficiently.

A novel evolutionary optimization algorithm inspired in the intelligent behaviour of the hunter spider

During the past decade, solving complex optimization problems with metaheuristic algorithms has received considerable attention among practitioners and researchers. Hence, many metaheuristic algorithms have been developed over the last years. Many of these algorithms are inspired by various phenomena of nature. Evolutionary intelligence is a research field that models the behaviour of insects or animals. Several algorithms arising from such models have been proposed to solve a wide range of complicated optimizable systems. In this paper, a novel evolutionary technique called Artificial Spider Algorithm (ASA) for solving optimization tasks in unconstrained problems with high nonlinearity is proposed. The ASA is based on the simulation of spider behaviour. For this purpose, a new metaphysical method according to spinning web and hunting insects via spider is inspired in nature. In order to illustrate the proficiency of the proposed approach, it is compared to other well-known evolutionary methods. The comparison investigates several test functions that are commonly considered within the literature of evolutionary algorithms. The result shows a high performance and effectiveness of this method for searching a global optimum, as well as the cost reduction noticeably for various benchmark functions.

A hybrid optimization approach based on clustering and chaotic sequences

Evolutionary computation algorithms represent a class of stochastic methods that can be applied to a wide set of different complex optimization problems. Recently, the combination of approaches extracted from different computation techniques represents one of the most successful trends in evolutionary optimization. With this integration, the idea is to overcome the limitations of each single method and to reach a synergetic effect through their integration. In this paper, a hybrid optimization algorithm for solving optimization problems is introduced. The approach, called cluster–chaotic-optimization, combines the classification characteristics of a clustering method with the randomness of chaotic sequences to conduct its search strategy. Under the proposed method, at each generation, the population is divided into different clusters according to its space distribution. Then, individuals are modified considering two kinds of operators: intra-cluster and extra-cluster. In the intra-cluster operation, individuals of the same cluster are locally adjusted considering the position of the best element of the cluster in terms of its fitness value. On the other hand, in the extra-cluster operation, the best individual of each cluster is globally attracted to the best element of the complete population. In both operations, the adjustment on each individual position is produced by using deterministic rules and chaotic sequences. With such mechanisms, the proposed method efficiently examines the search space based on the spatial associations produced by the individuals during the optimization process. To exhibit the performance and robustness of the proposed method, different comparisons to other well-known evolutionary methods and hybrid approaches are conducted. The comparison considers several standard benchmark functions and real-world engineering problems which are typically found in the literature of evolutionary algorithms. The results suggest a high performance of the proposed methodology.

A Novel Intelligence Algorithm Based on the Social Group Optimization Behaviors

The collective intelligent behaviors of insects or animal groups in nature have maintained the survival of the species for thousands of years. In this paper, a novel swarm intelligence algorithm called the social group entropy optimization (SGEO) algorithm is proposed for solving optimization tasks. The proposed algorithm is based on the social group model, the status optimization model, and the entropy model, which are the main contributions of this paper. First, the social group model and the feedback mechanism between Leaders and Followers are developed to reduce the probability of local optimum. Second, the status optimization model is described to reveal the changing rule about the population behavior states, to support the conversion between different social behaviors during evolution, to promote the algorithm to optimize quickly, and to avoid local optimization. Third, the entropy model is introduced to analyze the entropy of social groups, the change rule of difference entropy, and to set the information entropy as behavior’s criterion of state optimization. In addition, the mathematical model of the SGEO is deduced from the group theory, matter dynamics, and the information entropy theory. The convergence and parallelism of it have been analyzed and verified theoretically. Moreover, to test the effectiveness of the SGEO, it is used to solve benchmark functions’ problems that are commonly considered within the literature of evolutionary algorithms. Experimental results are compared with those of three other state-of-the-art algorithms. The superior performance of the SGEO validates its effectiveness and efficiency for the optimization problems, especially for the high-dimension problems.

A global optimization algorithm inspired in the behavior of selfish herds

In this paper, a novel swarm optimization algorithm called the Selfish Herd Optimizer (SHO) is proposed for solving global optimization problems. SHO is based on the simulation of the widely observed selfish herd behavior manifested by individuals within a herd of animals subjected to some form of predation risk. In SHO, individuals emulate the predatory interactions between groups of prey and predators by two types of search agents: the members of a selfish herd (the prey) and a pack of hungry predators. Depending on their classification as either a prey or a predator, each individual is conducted by a set of unique evolutionary operators inspired by such prey-predator relationship. These unique traits allow SHO to improve the balance between exploration and exploitation without altering the population size. To illustrate the proficiency and robustness of the proposed method, it is compared to other well-known evolutionary optimization approaches such as Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), Firefly Algorithm (FA), Differential Evolution (DE), Genetic Algorithms (GA), Crow Search Algorithm (CSA), Dragonfly Algorithm (DA), Moth-flame Optimization Algorithm (MOA) and Sine Cosine Algorithm (SCA). The comparison examines several standard benchmark functions, commonly considered within the literature of evolutionary algorithms. The experimental results show the remarkable performance of our proposed approach against those of the other compared methods, and as such SHO is proven to be an excellent alternative to solve global optimization problems.

Probabilistic Analysis of the (1+1)-Evolutionary Algorithm.

We give a detailed analysis of the optimization time of the [Formula: see text]-Evolutionary Algorithm under two simple fitness functions (OneMax and LeadingOnes). The problem has been approached in the evolutionary algorithm literature in various ways and with different degrees of rigor. Our asymptotic approximations for the mean and the variance represent the strongest of their kind. The approach we develop is based on an asymptotic resolution of the underlying recurrences and can also be extended to characterize the corresponding limiting distributions. While most of our approximations can be derived by simple heuristic calculations based on the idea of matched asymptotics, the rigorous justifications are challenging and require a delicate error analysis.

Micro-time variant multi-objective particle swarm optimization (micro-TVMOPSO) of a solar thermal combisystem

Abstract Multi-objective optimization (MOO) algorithms usually require a high number of objective function evaluations to approximate the Pareto-optimal solutions, which can be time-consuming in many engineering applications. To overcome this issue, MOO algorithms using a small population, often referred to as micro-MOO algorithms, aim at approximating the true Pareto front using a smaller number of objective function evaluations. Such algorithms are not common in the evolutionary algorithms literature and their usage in building engineering is seldom. This paper proposes a micro-time variant multi-objective particle swarm optimization (micro-TVMOPSO), which is a revised version of the micro-MOPSO algorithm. First, the proposed algorithm is applied along with eight other MOO algorithms to 24 benchmark problems, and their performance is compared by using two metrics. Although the proposed micro-TVMOPSO algorithm faced difficulties in solving some complicated Pareto fronts, it outperformed the eight optimization algorithms on 10 of the 24 selected benchmark problems. After the comparison of proposed micro-TVMOPSO with several MOO algorithms for different benchmark problems, the micro-TVMOPSO is applied to a case study in engineering: the design optimization of a residential solar thermal combisystem using two conflicting objective functions, the life cycle cost (LCC) and energy use (LCE). Different patterns of variation of the decision variables were observed from the non-dominated solutions found by micro-TVMOPSO, which would have not been possible to notice by performing a single-objective optimization. For instance, the increase of number of solar collectors from 1 to 10 had the impact of increasing the LCC by 84% and decreasing the LCE by 63%. The results indicated that the number of solar thermal collectors is the variable having the most effect on both LCC and LCE. When the number of collectors increases, more energy is harvested, and larger tanks and less auxiliary electric power are needed.

A swarm optimization algorithm inspired in the behavior of the social-spider

Swarm intelligence is a research field that models the collective behavior in swarms of insects or animals. Several algorithms arising from such models have been proposed to solve a wide range of complex optimization problems. In this paper, a novel swarm algorithm called the Social Spider Optimization (SSO) is proposed for solving optimization tasks. The SSO algorithm is based on the simulation of cooperative behavior of social-spiders. In the proposed algorithm, individuals emulate a group of spiders which interact to each other based on the biological laws of the cooperative colony. The algorithm considers two different search agents (spiders): males and females. Depending on gender, each individual is conducted by a set of different evolutionary operators which mimic different cooperative behaviors that are typically found in the colony. In order to illustrate the proficiency and robustness of the proposed approach, it is compared to other well-known evolutionary methods. The comparison examines several standard benchmark functions that are commonly considered within the literature of evolutionary algorithms. The outcome shows a high performance of the proposed method for searching a global optimum with several benchmark functions.

Structured population genetic algorithms: a literature survey

The Genetic Algorithm (GA) has been one of the most studied topics in evolutionary algorithm literature. Mimicking natural processes of inheritance, mutation, natural selection and genetic operators, GAs have been successful in solving various optimization problems. However, standard GA is often criticized as being too biased in candidate solutions due to genetic drift in search. As a result, GAs sometimes converge on premature solutions. In this paper, we survey the major advances in GA, particularly in relation to the class of structured population GAs, where better exploration and exploitation of the search space is accomplished by controlling interactions among individuals in the population pool. They can be classified as spatial segregation, spatial distance and heterogeneous population. Additionally, secondary factors such as aging, social behaviour, and so forth further guide and shape the reproduction process. Restricting randomness in reproduction has been seen to have positive effects on GAs. It is our hope that by reviewing the many existing algorithms, we shall see even better algorithms being developed.

A review of multiobjective test problems and a scalable test problem toolkit

When attempting to better understand the strengths and weaknesses of an algorithm, it is important to have a strong understanding of the problem at hand. This is true for the field of multiobjective evolutionary algorithms (EAs) as it is for any other field. Many of the multiobjective test problems employed in the EA literature have not been rigorously analyzed, which makes it difficult to draw accurate conclusions about the strengths and weaknesses of the algorithms tested on them. In this paper, we systematically review and analyze many problems from the EA literature, each belonging to the important class of real-valued, unconstrained, multiobjective test problems. To support this, we first introduce a set of test problem criteria, which are in turn supported by a set of definitions. Our analysis of test problems highlights a number of areas requiring attention. Not only are many test problems poorly constructed but also the important class of nonseparable problems, particularly nonseparable multimodal problems, is poorly represented. Motivated by these findings, we present a flexible toolkit for constructing well-designed test problems. We also present empirical results demonstrating how the toolkit can be used to test an optimizer in ways that existing test suites do not