Multi-strategy boosted Aquila optimizer for function optimization and engineering design problems
Hao Cui,Yaning Xiao,Abdelazim G. Hussien,Yanling Guo
DOI: https://doi.org/10.1007/s10586-024-04319-4
2024-03-15
Cluster Computing
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.
computer science, information systems, theory & methods