Answers are grounded in this paper and cite their sources.
This algorithm combines differential evolution and gazelle optimization, achieving better results in complex optimization tasks.
In complex engineering systems, solving high-dimensional, nonlinear, and multimodal optimization problems remains a formidable challenge. Traditional optimization techniques often converge prematurely or fail to scale effectively with problem complexity. Nature-inspired metaheuristics, such as Differential Evolution (DE) and Gazelle Optimization Algorithm (GOA), have shown promise in addressing such issues due to their adaptive exploration and exploitation capabilities. While DE excels in global exploration through mutation and crossover strategies, it suffers from limited convergence precision in rugged landscapes. Conversely, the Gazelle Optimization Algorithm, inspired by the evasive and coordinated movement of gazelles under predation, provides better adaptability in exploitation but lacks the stochastic diversity for broad search spaces. Thus, combining the strengths of both may overcome their individual limitations. This paper proposes a novel hybrid approach termed Gazelle-Differential Evolution (GoDE). The algorithm synergistically integrates the exploitation ability of GOA with the exploration strength of DE. Specifically, GoDE leverages gazelle dynamics for local refinement and DE’s differential mutation for global search. A dynamic control parameter regulates the hybridization intensity, ensuring a balanced optimization process. GoDE was evaluated on 25 benchmark functions (CEC 2023) and three real-world engineering design problems (pressure vessel, welded beam, and hydro-turbine blade design). Compared to five baseline methods—Standard DE, PSO, GOA, GWO, and CMA-ES—GoDE achieved superior convergence accuracy, stability, and computation time. Results confirm its robustness in navigating complex, multimodal spaces without being trapped in local optima.
No takes yet. Share an insight, caveat, or question.
Sabaresan et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: