Scholarlink

<p>This paper proposes novel variants of the shuffled particle swarm optimization (SPSO) algorithm for solving the frequency modulation sound parameter identification (FMSPI) problem. In the SPSO framework, the population is divided into several parallel groups, and each group undergoes an independent evolutionary process using particle swarm optimization (PSO). This study employs two distinct strategies to achieve an improved balance between the exploration and exploitation capabilities of the SPSO algorithm. First, it introduces a group-specific strategy for the inertia weight factor parameter, which is dynamically adjusted for each group during every iteration of the SPSO algorithm. Second, to facilitate an intensified search in promising regions of the solution space, a quasi-opposition-based strategy is incorporated into the SPSO. Experimental results on FMSPI problems demonstrate that the proposed strategies significantly enhance the performance of the SPSO algorithm. Also, in order to demonstrate the effectiveness of proposed algorithm, IEEE CEC2014 test set is used and the obtained results are compared against three other shuffled algorithms include: SFL, SCE and SDE. As evidenced the Wilcoxon signed-ranked test of reported results, the proposed algorithm demonstrates statistically superior performance versus all competitors.</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

Quasi-opposition-based learning in a shuffled particle swarm optimization for solving frequency modulation sounds parameter identification problem

  • 期刊论文

  • 发表日期：

    2025年9月30日

查看全文
  • Morteza Alinia Ahandani

摘要

This paper proposes novel variants of the shuffled particle swarm optimization (SPSO) algorithm for solving the frequency modulation sound parameter identification (FMSPI) problem. In the SPSO framework, the population is divided into several parallel groups, and each group undergoes an independent evolutionary process using particle swarm optimization (PSO). This study employs two distinct strategies to achieve an improved balance between the exploration and exploitation capabilities of the SPSO algorithm. First, it introduces a group-specific strategy for the inertia weight factor parameter, which is dynamically adjusted for each group during every iteration of the SPSO algorithm. Second, to facilitate an intensified search in promising regions of the solution space, a quasi-opposition-based strategy is incorporated into the SPSO. Experimental results on FMSPI problems demonstrate that the proposed strategies significantly enhance the performance of the SPSO algorithm. Also, in order to demonstrate the effectiveness of proposed algorithm, IEEE CEC2014 test set is used and the obtained results are compared against three other shuffled algorithms include: SFL, SCE and SDE. As evidenced the Wilcoxon signed-ranked test of reported results, the proposed algorithm demonstrates statistically superior performance versus all competitors.