Scholarlink

<p>The Crayfish Optimization Algorithm (COA) is inspired by the foraging, summering, and competitive behavior patterns of crayfish. However, when faced with multiple local optima in the solution space, the algorithm may converge slowly or prematurely to a local optimum. To address this issue, we propose a enhanced Crayfish Optimization Algorithm (ECOA). In ECOA, a nonlinear control parameter based on a sine function is introduced. Additionally, a hierarchical update mechanism is proposed to prevent the algorithm from falling into local optima. Meanwhile, a new boundary adjustment strategy, termed dimension-by-dimension centroid boundary control, is also introduced to effectively utilize population information. Finally, dynamic centroid opposition-based learning is incorporated to enhance convergence speed and population diversity. The CEC2017 test suite is used to evaluate the optimization performance of ECOA. ECOA is compared with 13 well-known algorithms, including GSA, WOA, MFO, AGWO, HPHHO, CPSOGSA, HEOA, DMOA, WO, KOA, SMO, SFOA, and COA. The experimental results demonstrate that ECOA significantly outperforms the other algorithms in terms of stability, accuracy, and convergence speed. Finally, through four engineering problems, the KELM hyperparameters were optimized for flood prediction, and 3D UAV path planning. It further verifies the effectiveness of ECOA in solving practical problems. From the statistical results, it can be observed that for the 10, 30, 50, and 100 dimensions of CEC2017, the Friedman mean rank of ECOA are 3.73, 3.06, 2.94, and 2.62, respectively. Except for the second ranking in the 10-dimensional case, ECOA ranks first in all other cases. Additionally, for all three real-world applications, ECOA achieves the best results.</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

Enhanced crayfish optimization algorithm for global optimization and real-world applications

  • 期刊论文

  • 发表日期：

    2025年8月30日

查看全文
  • Jiangxue Xie,
  • Haisong Huang,
  • Shengwei Fu,
  • Ziten Lu,
  • Feifei Li,
  • Man Su

摘要

The Crayfish Optimization Algorithm (COA) is inspired by the foraging, summering, and competitive behavior patterns of crayfish. However, when faced with multiple local optima in the solution space, the algorithm may converge slowly or prematurely to a local optimum. To address this issue, we propose a enhanced Crayfish Optimization Algorithm (ECOA). In ECOA, a nonlinear control parameter based on a sine function is introduced. Additionally, a hierarchical update mechanism is proposed to prevent the algorithm from falling into local optima. Meanwhile, a new boundary adjustment strategy, termed dimension-by-dimension centroid boundary control, is also introduced to effectively utilize population information. Finally, dynamic centroid opposition-based learning is incorporated to enhance convergence speed and population diversity. The CEC2017 test suite is used to evaluate the optimization performance of ECOA. ECOA is compared with 13 well-known algorithms, including GSA, WOA, MFO, AGWO, HPHHO, CPSOGSA, HEOA, DMOA, WO, KOA, SMO, SFOA, and COA. The experimental results demonstrate that ECOA significantly outperforms the other algorithms in terms of stability, accuracy, and convergence speed. Finally, through four engineering problems, the KELM hyperparameters were optimized for flood prediction, and 3D UAV path planning. It further verifies the effectiveness of ECOA in solving practical problems. From the statistical results, it can be observed that for the 10, 30, 50, and 100 dimensions of CEC2017, the Friedman mean rank of ECOA are 3.73, 3.06, 2.94, and 2.62, respectively. Except for the second ranking in the 10-dimensional case, ECOA ranks first in all other cases. Additionally, for all three real-world applications, ECOA achieves the best results.