Biomimetics, Год журнала: 2024, Номер 9(10), С. 596 - 596
Опубликована: Окт. 1, 2024
The nutcracker optimizer algorithm (NOA) is a metaheuristic method proposed in recent years. This simulates the behavior of nutcrackers searching and storing food nature to solve optimization problem. However, traditional NOA struggles balance global exploration local exploitation effectively, making it prone getting trapped optima when solving complex problems. To address these shortcomings, this study proposes reinforcement learning-based bi-population called RLNOA. In RLNOA, mechanism introduced better capabilities. At beginning each iteration, raw population divided into an sub-population based on fitness value individual. composed individuals with poor values. An improved foraging strategy random opposition-based learning designed as update for enhance diversity. Meanwhile, Q-learning serves adaptive selector strategies, enabling optimal adjustment sub-population’s across various performance RLNOA evaluated using CEC-2014, CEC-2017, CEC-2020 benchmark function sets, compared against nine state-of-the-art algorithms. Experimental results demonstrate superior algorithm.
Язык: Английский