Engineering and Applied Sciences Journal

GGWO: A Golden Grey Wolf Optimizer Hybridized with the Golden Sine and Cosine Algorithm for Solving Constrained Optimization Problems

Abstract

Kaiyuan Zheng

The Grey Wolf Optimization (GWO) algorithm is a well-established metaheuristic approach recognized for its exceptional ability to explore global optima and its inherent stability. Nonetheless, the original GWO algorithm faces certain limitations, including protracted convergence rates and a propensity to become ensnared in local optima. In light of these challenges, this paper introduces the Golden Grey Wolf Optimization (GGWO) algorithm, which seeks to enhance the performance and efficiency of the GWO by integrating innovative strategies that mitigate its shortcomings. By leveraging advanced techniques, GGWO aims to achieve improved convergence speed and a greater likelihood of finding global solutions, thereby offering a more robust framework for optimization tasks. Initially, the Golden Grey Wolf Optimization (GGWO) algorithm utilizes an optimized point
set for population initialization, which ensures a uniform distribution of gray wolves across the search space. This strategic initialization lays a robust groundwork for subsequent iterative processes. At the commencement of each iteration, a non- linear adjustment of the convergence factor is implemented, facilitating a broader exploration of the search space during the initial phases and gradually transitioning to a focused exploitation of local optima in later stages. This approach significantly enhances the overall optimization performance of the algorithm. Moreover, the incorporation of the golden sine strategy further enriches population diversity and boosts search capabilities. This innovative strategy not only mitigates the risk of premature convergence to local optima but also augments the algorithm’s global search proficiency. To assess the efficacy of GGWO, a series of experiments were conducted utilizing 23 classical benchmark test functions. The findings reveal that the GGWO algorithm demonstrates exceptional solving performance and robust resilience, accompanied by a notable enhancement in convergence speed. Additionally, the capability of GGWO in addressing constrained optimization challenges was validated through its application in pressure vessel design, showcasing its superior effectiveness in tackling complex practical engineering problems.