Scholarlink

<p>Metaheuristic and, in particular, evolutionary algorithms are&#xa0;considered computational methods used to solve complex optimization problems as well as applications in machine learning and data science.&#xa0;They are characterized by their ability to explore a search space efficiently, are usually based on a population of possible individual solutions, and are often inspired by natural phenomena. The differential evolution is an evolutionary population-based algorithm and is used for solving global optimization problems by iteratively improving a candidate solution based on an evolutionary process. This paper presents an efficient improvement to the differential evolution algorithm for efficiently solving constrained optimization problems. The improvement is based on the equation involving the differential weight (mutation/scaling factor). The proposed method is called the Mutation Recombination Differential Evolution (MRDE). The proposed improvement enables the algorithm to efficiently solve complex nonlinear constrained optimization problems with a high degree of accuracy. The proposed method was compared against a number of popular optimizers, and the results confirm the superiority of the proposed ameliorated differential evolution algorithm.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Improving the performance of the differential evolution algorithm using the mutation recombination strategy for solving constrained optimization problems

  • 期刊论文

  • 发表日期：

    2026年1月9日

查看全文
  • Ardeshir Bahreininejad,
  • Arian Bahreininejad

摘要

Metaheuristic and, in particular, evolutionary algorithms are considered computational methods used to solve complex optimization problems as well as applications in machine learning and data science. They are characterized by their ability to explore a search space efficiently, are usually based on a population of possible individual solutions, and are often inspired by natural phenomena. The differential evolution is an evolutionary population-based algorithm and is used for solving global optimization problems by iteratively improving a candidate solution based on an evolutionary process. This paper presents an efficient improvement to the differential evolution algorithm for efficiently solving constrained optimization problems. The improvement is based on the equation involving the differential weight (mutation/scaling factor). The proposed method is called the Mutation Recombination Differential Evolution (MRDE). The proposed improvement enables the algorithm to efficiently solve complex nonlinear constrained optimization problems with a high degree of accuracy. The proposed method was compared against a number of popular optimizers, and the results confirm the superiority of the proposed ameliorated differential evolution algorithm.