Scholarlink

This article proposes a multi-objective algorithm based on beluga whales called Multiobjective Beluga Whale Optimization (MOBWO). It is a multi-objective version of the single-objective Beluga Whale Optimization (BWO) algorithm, which is inspired by the natural behavior of beluga whales. The procedure of the MOBWO algorithm is based on the BWO algorithm, incorporating additional mechanisms. For instance, a crowding distance mechanism is used to balance exploitation and exploration phases as the search progresses. Additionally, a non-dominated sorting strategy is integrated to preserve population diversity. To verify and validate the performance of the MOBWO algorithm, we studied 22 test problems, including 10 constrained and 12 unconstrained problems. The performance of MOBWO is compared to that of MOSMA, MOWCA, and NSGA2 using various performance metrics, such as Generational Distance (GD), Inverted Generational Distance (IGD), Maximum Spread (MS), spacing (S), spread ( \(\varDelta \) ), and hypervolume (H). The quantitative and qualitative results indicate that the proposed MOBWO offers more competitive results than the other algorithms.

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

Multobjective Belugas Whale Optimizer for Pareto Optimal Solution Search

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Amidou Zoungrana,
  • Appolinaire Tougma,
  • Kounhinir Somé

摘要

This article proposes a multi-objective algorithm based on beluga whales called Multiobjective Beluga Whale Optimization (MOBWO). It is a multi-objective version of the single-objective Beluga Whale Optimization (BWO) algorithm, which is inspired by the natural behavior of beluga whales. The procedure of the MOBWO algorithm is based on the BWO algorithm, incorporating additional mechanisms. For instance, a crowding distance mechanism is used to balance exploitation and exploration phases as the search progresses. Additionally, a non-dominated sorting strategy is integrated to preserve population diversity. To verify and validate the performance of the MOBWO algorithm, we studied 22 test problems, including 10 constrained and 12 unconstrained problems. The performance of MOBWO is compared to that of MOSMA, MOWCA, and NSGA2 using various performance metrics, such as Generational Distance (GD), Inverted Generational Distance (IGD), Maximum Spread (MS), spacing (S), spread ( \(\varDelta \) ), and hypervolume (H). The quantitative and qualitative results indicate that the proposed MOBWO offers more competitive results than the other algorithms.