Scholarlink

<p>The Dung Beetle Optimization Algorithm (DBO) demonstrates strong optimization capabilities and fast convergence. However, similar to many swarm intelligence algorithms, DBO suffers from limitations such as reduced convergence accuracy and a tendency to become trapped in local optima during later optimization stages. To address these issues, this paper proposes an enhanced DBO variant named OMDBO, which integrates principles from the Osprey algorithm and a multi-strategy framework. OMDBO employs piecewise linear chaotic mapping for population initialization to achieve a more dispersed distribution of individuals. During the rolling phase, OMDBO incorporates the Osprey algorithm’s exploration strategy to enhance global search capacity, thereby reducing parametric dependency and accelerating convergence. To optimize its search dynamics, OMDBO implements a dynamic proportion adjustment mechanism for beetles and refines the convergence factor balance, prioritizing early exploration before shifting to intensified local exploitation. Furthermore, to escape local optima, OMDBO universally applies an adaptive dual-phase Cauchy-Gaussian mutation operator. Rigorous benchmarking against six established swarm intelligence algorithms was conducted using 54 test functions (CEC2005, CEC2017, CEC2022) and five engineering problems. OMDBO achieved superior results in 49 benchmark functions (90.74%) and obtained optimal solutions in all five real-world applications. The experimental results demonstrate significant improvements in OMDBO’s convergence speed and optimization accuracy, confirming its effectiveness and versatility as an optimization tool.</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

An enhanced dung beetle optimization algorithm with adaptive strategies for solving complex problems

  • 期刊论文

  • 发表日期：

    2026年4月21日

查看全文
  • Zheng Li,
  • Xuechao Chen,
  • Jinlei Qin,
  • Mingyuan Du

摘要

The Dung Beetle Optimization Algorithm (DBO) demonstrates strong optimization capabilities and fast convergence. However, similar to many swarm intelligence algorithms, DBO suffers from limitations such as reduced convergence accuracy and a tendency to become trapped in local optima during later optimization stages. To address these issues, this paper proposes an enhanced DBO variant named OMDBO, which integrates principles from the Osprey algorithm and a multi-strategy framework. OMDBO employs piecewise linear chaotic mapping for population initialization to achieve a more dispersed distribution of individuals. During the rolling phase, OMDBO incorporates the Osprey algorithm’s exploration strategy to enhance global search capacity, thereby reducing parametric dependency and accelerating convergence. To optimize its search dynamics, OMDBO implements a dynamic proportion adjustment mechanism for beetles and refines the convergence factor balance, prioritizing early exploration before shifting to intensified local exploitation. Furthermore, to escape local optima, OMDBO universally applies an adaptive dual-phase Cauchy-Gaussian mutation operator. Rigorous benchmarking against six established swarm intelligence algorithms was conducted using 54 test functions (CEC2005, CEC2017, CEC2022) and five engineering problems. OMDBO achieved superior results in 49 benchmark functions (90.74%) and obtained optimal solutions in all five real-world applications. The experimental results demonstrate significant improvements in OMDBO’s convergence speed and optimization accuracy, confirming its effectiveness and versatility as an optimization tool.