Guo Delong, Huang Huajuan
School of Mathematics and Statistics, Qiannan Normal University for Nationalities, Duyun 558000, China.
Guangxi Key Laboratory of Hybrid Computation and IC Design Analysis, Guangxi University for Nationalities, Nanning 530006, China.
Biomimetics (Basel). 2025 Sep 2;10(9):581. doi: 10.3390/biomimetics10090581.
The Honey Badger Algorithm (HBA) is a recently proposed metaheuristic optimization algorithm inspired by the foraging behavior of honey badgers. The search mechanism of this algorithm is divided into two phases: a mining phase and a honey-seeking phase, effectively emulating the processes of exploration and exploitation within the search space. Despite its innovative approach, the Honey Badger Algorithm (HBA) faces challenges such as slow convergence rates, an imbalanced trade-off between exploration and exploitation, and a tendency to become trapped in local optima. To address these issues, we propose an enhanced version of the Honey Badger Algorithm (HBA), namely the Multi-Strategy Honey Badger Algorithm (MSHBA), which incorporates a Cubic Chaotic Mapping mechanism for population initialization. This integration aims to enhance the uniformity and diversity of the initial population distribution. In the mining and honey-seeking stages, the position of the honey badger is updated based on the best fitness value within the population. This strategy may lead to premature convergence due to population aggregation around the fittest individual. To counteract this tendency and enhance the algorithm's global optimization capability, we introduce a random search strategy. Furthermore, an elite tangential search and a differential mutation strategy are employed after three iterations without detecting a new best value in the population, thereby enhancing the algorithm's efficacy. A comprehensive performance evaluation, conducted across a suite of established benchmark functions, reveals that the MSHBA excels in 26 out of 29 IEEE CEC 2017 benchmarks. Subsequent statistical analysis corroborates the superior performance of the MSHBA. Moreover, the MSHBA has been successfully applied to four engineering design problems, highlighting its capability for addressing constrained engineering design challenges and outperforming other optimization algorithms in this domain.
蜜獾算法（HBA）是一种最近提出的元启发式优化算法，其灵感来源于蜜獾的觅食行为。该算法的搜索机制分为两个阶段：挖掘阶段和寻蜜阶段，有效地模拟了在搜索空间内的探索和利用过程。尽管蜜獾算法（HBA）采用了创新方法，但仍面临收敛速度慢、探索与利用之间的权衡不均衡以及容易陷入局部最优等挑战。为了解决这些问题，我们提出了蜜獾算法（HBA）的增强版本，即多策略蜜獾算法（MSHBA），该算法在种群初始化中引入了三次混沌映射机制。这种整合旨在提高初始种群分布的均匀性和多样性。在挖掘和寻蜜阶段，蜜獾的位置根据种群内的最佳适应度值进行更新。由于种群聚集在最适应的个体周围，这种策略可能导致早熟收敛。为了抵消这种趋势并增强算法的全局优化能力，我们引入了随机搜索策略。此外，在种群中未检测到新的最佳值的三次迭代后，采用精英切向搜索和差分变异策略，从而提高算法的效率。通过对一系列既定基准函数进行的全面性能评估表明，多策略蜜獾算法（MSHBA）在29个IEEE CEC 2017基准测试中的26个中表现出色。随后的统计分析证实了多策略蜜獾算法（MSHBA）的卓越性能。此外，多策略蜜獾算法（MSHBA）已成功应用于四个工程设计问题，突出了其解决约束工程设计挑战的能力，并在该领域优于其他优化算法。
