The penalty function method has been widely used for solving constrained optimization problems. In the method, an extended objective function, which is the sum of the objective value and the constraint violation weigh...
详细信息
ISBN:
(纸本)9781728121536
The penalty function method has been widely used for solving constrained optimization problems. In the method, an extended objective function, which is the sum of the objective value and the constraint violation weighted by the penalty coefficient, is optimized. However, it is difficult to control the coefficient properly because proper control of the coefficient varies in each problem. In this study, the equivalent penalty coefficient value (EPC) is proposed for population-based optimization algorithms (POAs). EPC can be defined in POAs where a new solution is compared with the old solution. EPC is the penalty coefficient value that makes the two extended objective values of the solutions the same. Search that gives priority to the objective value is realized by selecting a small EPC. Search that gives priority to the constraint violation is realized by selecting a large EPC. The adaptive control of the penalty coefficient can be realized by selecting an appropriate EPC. The proposed method is introduced to differential evolution and the advantage of the proposed method is shown by solving well-known constrained optimization problems.
Brain storm optimization (BSO) is a newly proposed optimizationalgorithm inspired by human being brainstorming process. After its appearance, much attention has been paid on and many attempts to improve its performan...
详细信息
Brain storm optimization (BSO) is a newly proposed optimizationalgorithm inspired by human being brainstorming process. After its appearance, much attention has been paid on and many attempts to improve its performance have been made. The search ability of BSO has been enhanced, but it still suffers from sticking into stagnation during exploitation phase. This paper proposes a novel method which incorporates BSO with chaotic local search (CLS) with the purpose of alleviating this situation. Chaos has properties of randomicity and ergodicity. These properties ensure CLS can explore every state of the search space if the search time duration is long enough. The incorporation of CLS can make BSO break the stagnation and keep the population's diversity simultaneously, thus realizing a better balance between exploration and exploitation. Twelve chaotic maps are randomly selected for increasing the diversity of the search mechanism. Experimental and statistical results based on 25 benchmark functions demonstrate the superiority of the proposed method.
To solve complicated function optimization problems, a function optimizationalgorithm is constructed based on the Susceptible-Infective-Susceptible (SIS) epidemic model, the function optimizationalgorithm is called ...
详细信息
To solve complicated function optimization problems, a function optimizationalgorithm is constructed based on the Susceptible-Infective-Susceptible (SIS) epidemic model, the function optimizationalgorithm is called SIS algorithm, or SISA in short. The algorithm supposes that some male and female organisms exist in an ecosystem;each individual is characterized by a number of features;an infectious disease exists in the ecosystem and infects among individuals, the infection rule is that female individuals infect male individuals or male individuals infect female individuals, the disease attacks a part of features of an individual. The infected individuals can be cured;the cured individuals can be infected again after a period of time. The physique strength of an individual is decided synthetically by the infection, cure and susceptibility of certain features. The S-I operator is used to transfer feature information from male to female or female to male, the I-S operator is used to transfer feature information from male to male or female to female, the I-S operator and S-S operator are used to transfer feature information among individuals without sex difference. The individuals with strong physique can continue to grow, while the individuals with weak physique stop growing. Results show that the algorithm has characteristics of global convergence and high convergence speed for the complicated functions optimization problems, especially for high dimensional function optimization problems. (C) 2013 Elsevier B.V. All rights reserved.
The design of effective optimizationalgorithms is always a hot research topic. An optimizer ensemble where any population-based optimization algorithm can be integrated is proposed in this study. First, the optimizer...
详细信息
The design of effective optimizationalgorithms is always a hot research topic. An optimizer ensemble where any population-based optimization algorithm can be integrated is proposed in this study. First, the optimizer ensemble framework based on ensemble learning is presented. The learning table consisting of the population members of all optimizers is constructed to share information. The maximum number of iterations is divided into several exchange iterations. Each optimizer exchanges individuals with the learning table in exchange iterations and runs independently in the other iterations. Exchange individuals are generated by a bootstrap sample from the learning table. To maintain a balance between exchange individuals and preserved individuals, the exchange number of each optimizer is adaptively assigned according to its fitness. The output is obtained by the voting approach that selects the highest ranked solution. Second, an optimizer ensemble algorithm (OEA) which combines multiple population-based optimization algorithms is proposed. The computational complexity, convergence, and diversity of OEA are analyzed. Finally, extensive experiments on benchmark functions demonstrate that OEA outperforms several state-of-the-art algorithms. OEA is used to search the maximum mutual information in image registration. The high performance of OEA is further verified by a large number of registration results on real remote sensing images.
Among various complexities affecting the performance of an optimizationalgorithm, the search space dimension is a major factor. Another key complexity is the discreteness of the search space. When these two complexit...
详细信息
Among various complexities affecting the performance of an optimizationalgorithm, the search space dimension is a major factor. Another key complexity is the discreteness of the search space. When these two complexities are present in a single problem, optimizationalgorithms have been demonstrated to be inefficient, even in linear programming (LP) problems. In this paper, we consider a specific resource allocation problem constituting to an integer linear programming (ILP) problem which, although comes from a specific industry, is similar to other practical resource allocation and assignment problems. based on a populationbasedoptimization approach, we present a computationally fast method to arrive at a near-optimal solution. Compared to two popular softwares (glpk, Makhorin, 2012 and CPLEX, Gay, 2015), which are not able to handle around 300 and 2000 integer variables while continuing to run for hours, respectively, our proposed method is able to find a near-optimal solution in less than second on the same computer. Moreover, the main highlight of this study is to propose a customized populationbasedoptimizationalgorithm that scales in almost a linear computational complexity in handling 50,000 to one billion (10) variables. We believe that this is the first time such a large-sized real-world constrained problem is ever handled using any optimizationalgorithm. We perform sensitivity studies of our method for different problem parameters and these studies clearly demonstrate the reasons for such a fast and scale-up application of the proposed method. (C) 2017 Elsevier B.V. All rights reserved.
The penalty function method has been widely used to solve constrained optimization problems. In the method, an extended objective function, which is the sum of the objective value and the constraint violation weighted...
详细信息
The penalty function method has been widely used to solve constrained optimization problems. In the method, an extended objective function, which is the sum of the objective value and the constraint violation weighted by the penalty coefficient, is optimized. However, it is difficult to control the coefficient properly because the proper control depends on each problem. Recently, the equivalent penalty coefficient (EPC) method, which is a new adaptive penalty method for population-based optimization algorithms (POAs), has been proposed. The EPC method can be applied to POAs where a new solution is compared with the old solution. The EPC value, which makes the two extended objective values of the solutions the same, is used to control the coefficient. In this study, we propose to apply the EPC method to particle swarm optimization (PSO) where a new solution is compared with the best solution found so far. To improve the performance of constrained optimization, a mutation operation is also proposed. The proposed method is examined using two topologies of PSO. The advantage of the proposed method is shown by solving well-known constrained optimization problems and comparing the results with those obtained by PSO with a standard constraint-handling technique.
暂无评论