Cultural history optimization algorithm: a new human-inspired metaheuristic algorithm for engineering optimization problems

Cultural history optimization algorithm: a new human-inspired metaheuristic algorithm for engineering optimization problems

연구 분야: Analysis

학회: Neural Computing and Applications

Meta-heuristic optimization methods are popular today, but they still face many problems, such as early convergence, weak scalability, and high computing cost. As engineering problems grow larger and more complex, the need for an optimizer that can search broadly, converge quickly, and keep the computation affordable becomes even more urgent. To address these issues, this paper introduces a novel human inspired metaheuristic algorithm, the cultural history optimization algorithm (CHOA), based on cultural history principles. CHOA’s performance is evaluated against 47 benchmark functions and the CEC06−2019 test suite, encompassing large-scale unimodal, multimodal, and fixed-dimension functions. Results demonstrate CHOA’s strong exploration and exploitation capabilities, achieving global optima with rapid convergence and manageable computational cost. Performance metrics, including mean cost, standard deviation, convergence acceleration, and computational burden, are compared with established metaheuristics, highlighting effectiveness. Moreover, Wilcoxon rank-sum tests confirm CHOA’s statistical superiority. As a large-scale design optimization problem, CHOA and state-of-the-art algorithms are applied to optimize a permanent magnet synchronous motor, showcasing CHOA’s local optima avoidance and scalability. Finally, the paper describes a graphical user interface (GUI) developed for CHOA to facilitate its practical application.