Abstract

As the complexity of optimization problems continues to rise, the demand for high-performance algorithms becomes increasingly urgent. This paper addresses the challenges faced by the Aquila Optimizer (AO), a novel swarm-based intelligent optimizer simulating the predatory behaviors of Aquila in North America. While AO has shown good performance in prior studies, it grapples with issues such as poor convergence accuracy and a tendency to fall into local optima when tackling complex optimization tasks. To overcome these challenges, this paper proposes a multi-strategy boosted AO algorithm (PGAO) aimed at providing enhanced reliability for global optimization. The proposed algorithm incorporates several key strategies. Initially, a chaotic map is employed to initialize the positions of all search agents, enriching population diversity and laying a solid foundation for global exploration. Subsequently, the pinhole imaging learning strategy is introduced to identify superior candidate solutions in the opposite direction of the search domain during each iteration, accelerating convergence and increasing the probability of obtaining the global optimal solution. To achieve a more effective balance between the exploration and development phases in AO, a nonlinear switching factor is designed to replace the original fixed switching mechanism. Finally, the golden sine operator is utilized to enhance the algorithm’s local exploitation trends. Through these four improvement strategies, the optimization performance of AO is significantly enhanced. The proposed PGAO algorithm’s effectiveness is validated across 23 classical, 29 IEEE CEC2017, and 10 IEEE CEC2019 benchmark functions. Additionally, six real-world engineering design problems are employed to assess the practicability of PGAO. Results demonstrate that PGAO exhibits better competitiveness and application prospects compared to the basic method and various advanced algorithms. In conclusion, this study contributes to addressing the challenges of complex optimization problems, significantly improving the performance of global optimization algorithms, and holds both theoretical and practical significance.

中文翻译：

---

用于功能优化和工程设计问题的多策略增强 Aquila 优化器

摘要

随着优化问题的复杂度不断上升，对高性能算法的需求变得越来越迫切。本文解决了天鹰座优化器 (AO) 所面临的挑战，这是一种基于群体的新型智能优化器，模拟北美天鹰座的掠夺行为。虽然 AO 在之前的研究中表现出了良好的性能，但它也面临着收敛精度差以及在处理复杂优化任务时容易陷入局部最优等问题。为了克服这些挑战，本文提出了一种多策略增强的 AO 算法（PGAO），旨在为全局优化提供增强的可靠性。所提出的算法结合了几个关键策略。最初，采用混沌地图来初始化所有搜索代理的位置，丰富了种群多样性，为全局探索奠定了坚实的基础。随后，引入针孔成像学习策略，在每次迭代过程中识别搜索域相反方向的优秀候选解，加速收敛并增加获得全局最优解的概率。为了在AO的探索和发展阶段之间实现更有效的平衡，设计了非线性切换因子来代替原来的固定切换机制。最后，利用黄金正弦算子增强算法的局部开发趋势。通过这四种改进策略，AO的优化性能显着增强。所提出的 PGAO 算法的有效性在 23 个经典、29 个 IEEE CEC2017 和 10 个 IEEE CEC2019 基准函数中得到验证。此外，还采用六个现实世界的工程设计问题来评估 PGAO 的实用性。结果表明，与基本方法和各种先进算法相比，PGAO 表现出更好的竞争力和应用前景。综上所述，本研究有助于解决复杂优化问题的挑战，显着提高全局优化算法的性能，具有理论和实际意义。