Stellar oscillation optimizer: a nature-inspired metaheuristic optimization algorithm

This paper proposes a nature-inspired metaheuristic optimization algorithm called Stellar Oscillation Optimizer (SOO), SOO is inspired from the field of asteroseismology, which examines the oscillatory behavior of stars to understand their internal structures, physical properties, and evolutionary stages. SOO simulates the dynamic expansion and contraction phases observed in stellar pulsations for balancing exploration and exploitation. SOO was evaluated using the IEEE CEC2020 and CEC2022 benchmark datasets, which consist of 10 and 12 functions, respectively. In addition, SOO’s performance was tested on three real-world engineering design problems to further validate its effectiveness. SOO was benchmarked against 10 Physics and mathematics-inspired Optimizers, as well as 15 recent optimizers (2021-2024). SOO achieved first place in 6 out of 10 functions on the CEC2020 dataset when compared to 25 recent, physics and mathematics-inspired optimizers and ranked first in 9 out of 12 functions on the CEC2022 dataset when compared to the 15 recent optimizers. To assess the statistical significance of SOO’s performance, Wilcoxon’s signed-rank and Friedman’s tests were employed. The experimental results demonstrate that SOO consistently outperforms its competitors across a wide variety of optimization tasks. Moreover, SOO was successfully applied to multi-level image segmentation, highlighting its diversity in solving real-world practical applications. The source code of SOO is publicly available for both MATLAB at: ( https://www.mathworks.com/matlabcentral/fileexchange/161921-stellar-oscillation-optimizer-meta-heuristic-optimimization ) and PYTHON at: ( https://github.com/AliRodan/Stellar-Oscillation-Optimizer ) (For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any Authors Accepted Manuscript version of this paper arising from this submission.).