Scholarlink

<p>Surrogate-assisted evolutionary algorithms (SAEAs) are effective for expensive optimization problems but often encounter inefficiencies in high-dimensional spaces. This paper introduces a surrogate-assisted particle swarm optimization with two-space interaction (TSPSO), designed to enhance performance in such environments. The approach divides individuals into potential and non-potential populations, segmenting the search space based on fitness values and constructing surrogate models tailored to each subspace. Throughout the iterative process, optimal solutions identified in the potential space are transferred to the non-potential space, guiding further exploration. This method helps narrow the search space, enabling quicker identification of promising directions and improving convergence rates. To mitigate premature convergence, a L<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="12530_2025_9678_Article_IEq1.gif" Format="GIF" Height="14" Rendition="HTML" Resolution="72" Type="Linedraw" Width="11" /> </InlineMediaObject> <EquationSource Format="TEX">\(\acute{e}\)</EquationSource> <EquationSource Format="MATHML"><math> <mover accent="true"> <mi>e</mi> <mo>´</mo> </mover> </math></EquationSource> </InlineEquation>vy flight strategy is applied under specific conditions, ensuring diversity in the search process. Experimental results demonstrate that TSPSO performs competitively in high-dimensional optimization tasks.</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

Two-space interaction enhanced particle swarm optimization for tackling high-dimensional problems

  • 期刊论文

  • 发表日期：

    2025年3月24日

查看全文
  • Le Wang,
  • Qinqin Fan,
  • Xuefeng Yan

摘要

Surrogate-assisted evolutionary algorithms (SAEAs) are effective for expensive optimization problems but often encounter inefficiencies in high-dimensional spaces. This paper introduces a surrogate-assisted particle swarm optimization with two-space interaction (TSPSO), designed to enhance performance in such environments. The approach divides individuals into potential and non-potential populations, segmenting the search space based on fitness values and constructing surrogate models tailored to each subspace. Throughout the iterative process, optimal solutions identified in the potential space are transferred to the non-potential space, guiding further exploration. This method helps narrow the search space, enabling quicker identification of promising directions and improving convergence rates. To mitigate premature convergence, a L \(\acute{e}\) e ´ vy flight strategy is applied under specific conditions, ensuring diversity in the search process. Experimental results demonstrate that TSPSO performs competitively in high-dimensional optimization tasks.