Scholarlink

The Particle Swarm Optimization algorithm is one of the most popular swarm-based algorithms for solving global optimization problems, as evidenced by its prevalence in the literature. Although the PSO algorithm demonstrates satisfactory performance in terms of global optimization, it still exhibits shortcomings in terms of convergence and stagnation at local optima. Consequently, researchers have been exploring alternative implementations to address these limitations. As a result, the PSO algorithm has undergone numerous improvements and implementations since its original proposal. The improvements include different ways of initializing the swarm and self-configuration of the parameters, as well as hybridizations with other well-known algorithms. However, a comparative analysis of the population diversity behavior between the different variants has not been performed. As is known, population diversity within Evolutionary Algorithms can be used as an indication of the distribution of solutions or population particles within the search space. This paper aims to analyze some of the PSO variants, focusing on the population diversity of each version. Subsequently, the behavior of the variants on engineering problems and the CEC 2017 dataset will be analyzed.

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

Diversity Measurement in Different PSO Variants Applied to Global Optimization and Classical Engineering Problems

  • 章节

  • 发表日期：

    2025年

查看全文
  • Angel Casas-Ordaz,
  • Jorge Ramos-Frutos,
  • Mario A. Navarro,
  • Eduardo H. Haro,
  • Bernardo Morales-Castañeda,
  • Marco Perez-Cisneros,
  • Noé Ortega-S,
  • Daniel Zaldivar Navarro

摘要

The Particle Swarm Optimization algorithm is one of the most popular swarm-based algorithms for solving global optimization problems, as evidenced by its prevalence in the literature. Although the PSO algorithm demonstrates satisfactory performance in terms of global optimization, it still exhibits shortcomings in terms of convergence and stagnation at local optima. Consequently, researchers have been exploring alternative implementations to address these limitations. As a result, the PSO algorithm has undergone numerous improvements and implementations since its original proposal. The improvements include different ways of initializing the swarm and self-configuration of the parameters, as well as hybridizations with other well-known algorithms. However, a comparative analysis of the population diversity behavior between the different variants has not been performed. As is known, population diversity within Evolutionary Algorithms can be used as an indication of the distribution of solutions or population particles within the search space. This paper aims to analyze some of the PSO variants, focusing on the population diversity of each version. Subsequently, the behavior of the variants on engineering problems and the CEC 2017 dataset will be analyzed.