Scholarlink

<p>The Artificial Hummingbird Algorithm (AHA) is a nature-inspired metaheuristic that mimics the exceptional flight behaviors of hummingbirds. Despite its promise, AHA often suffers from premature convergence and weak exploration. To overcome these issues, we propose the Balanced Artificial Hummingbird Algorithm (BAHA), which incorporates ten well-known chaotic maps into the behavioral switching mechanism of AHA. This integration introduces a dynamic trade-off between exploration and exploitation, enhancing both search efficiency and solution quality. BAHA is rigorously tested on 21 benchmark functions, three engineering design problems, and seven complex space trajectory optimization tasks from the European Space Agency (ESA). Comparative analysis shows that BAHA consistently outperforms or matches ten state-of-the-art algorithms, with the BAHA-v6 variant achieving the highest overall rank in Friedman statistical tests. These results underline the method’s robustness, adaptability, and effectiveness across diverse problem domains. Moreover, BAHA’s population-based and modular design makes it naturally compatible with parallel implementation on high-performance computing (HPC) platforms, especially for large-scale aerospace applications.</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

BAHA: balanced artificial hummingbird algorithm and its application to space industry

  • 期刊论文

  • 发表日期：

    2025年7月8日

查看全文
  • Zafer Atlı,
  • Bahaeddin Turkoglu

摘要

The Artificial Hummingbird Algorithm (AHA) is a nature-inspired metaheuristic that mimics the exceptional flight behaviors of hummingbirds. Despite its promise, AHA often suffers from premature convergence and weak exploration. To overcome these issues, we propose the Balanced Artificial Hummingbird Algorithm (BAHA), which incorporates ten well-known chaotic maps into the behavioral switching mechanism of AHA. This integration introduces a dynamic trade-off between exploration and exploitation, enhancing both search efficiency and solution quality. BAHA is rigorously tested on 21 benchmark functions, three engineering design problems, and seven complex space trajectory optimization tasks from the European Space Agency (ESA). Comparative analysis shows that BAHA consistently outperforms or matches ten state-of-the-art algorithms, with the BAHA-v6 variant achieving the highest overall rank in Friedman statistical tests. These results underline the method’s robustness, adaptability, and effectiveness across diverse problem domains. Moreover, BAHA’s population-based and modular design makes it naturally compatible with parallel implementation on high-performance computing (HPC) platforms, especially for large-scale aerospace applications.