International Journal of Information Technology and Applications, Vol. 1, No. 4, pp. 167-177, December 2024.
Abstract: To address the issues of easy entrapment in local optima and insufficient convergence precision in the Dung Beetle Optimization (DBO) algorithm, an improved DBO algorithm with a hybrid strategy (SADBO) is proposed. Initially, the tent chaotic mapping strategy is used to initialize the population, making the initial positions of the dung beetles more evenly distributed and enhancing population diversity. Secondly, the Sine Algorithm (SA) is introduced to improve global exploration capabilities. Finally, adaptive t-distribution is applied to perturb individuals, which assists the algorithm in evading localized optimum. Comparative experiments with four other algorithms and the original DBO algorithm demonstrate that SADBO surpasses them on the basis of accuracy and convergence speed on multiple benchmark test functions, proving the proposed algorithm’s superiority.