Scholarlink

<p>This study provides an effective cooperative multi-populations swarm intelligence approach, called CMPMO-SGO, which is based on the Social Group Optimization (SGO) algorithm to&#xa0;address multi-/many-objective optimization problems. In particular, this study first puts forth a unique cooperative multi-population framework- namely&#xa0;CMPMO/des, with dual elite selection. CMPMO/des achieves significantly high performance on solutions convergence and diversity with four excellent strategies: the ability to independently optimize one objective within each subpopulation, collaboration among subpopulations via a global archive, dual elite selection mechanisms, and the inclusion of logistic chaotic single-dimensional perturbation. Afterward, due to its very high performance, SGO is employed as the single objective&#xa0;optimizer in each subpopulation. Sixteen state-of-the-art multi/many-objective optimization algorithms, two CMPMO/des based many-objective genetic algorithms named CMPMO-GA and many-objective Harris Hawks optimization named CMPMO-HHO, were extensively compared with the performance of CMPMO-SGO on 34 multi-objective and 19 many-objective benchmark problems. The results demonstrate that CMPMO-SGO provides promising performance in addressing multi/many-objective optimization problems by utilizing the CMPMO/des framework.</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

Cooperative multi-population social group optimization for many-objective optimization

  • 期刊论文

  • 发表日期：

    2025年8月13日

查看全文
  • Anima Naik

摘要

This study provides an effective cooperative multi-populations swarm intelligence approach, called CMPMO-SGO, which is based on the Social Group Optimization (SGO) algorithm to address multi-/many-objective optimization problems. In particular, this study first puts forth a unique cooperative multi-population framework- namely CMPMO/des, with dual elite selection. CMPMO/des achieves significantly high performance on solutions convergence and diversity with four excellent strategies: the ability to independently optimize one objective within each subpopulation, collaboration among subpopulations via a global archive, dual elite selection mechanisms, and the inclusion of logistic chaotic single-dimensional perturbation. Afterward, due to its very high performance, SGO is employed as the single objective optimizer in each subpopulation. Sixteen state-of-the-art multi/many-objective optimization algorithms, two CMPMO/des based many-objective genetic algorithms named CMPMO-GA and many-objective Harris Hawks optimization named CMPMO-HHO, were extensively compared with the performance of CMPMO-SGO on 34 multi-objective and 19 many-objective benchmark problems. The results demonstrate that CMPMO-SGO provides promising performance in addressing multi/many-objective optimization problems by utilizing the CMPMO/des framework.