Multi-Strategy Promoted-Archimedes Optimization Algorithm and Its Application in Constellation Mission Scheduling

To address the limitations of slow convergence, low accuracy, and local optima entrapment in the Archimedes Optimization Algorithm (AOA), we propose a multi-strategy enhanced Promoted-Archimedes Optimization Algorithm (P-AOA). We first employ Halton sequence for population initialization to enhance diversity and establish a foundation for global search. We then incorporate Cauchy mutation with simulated annealing to update positions of randomly selected individuals, thereby increasing solution diversity. Adaptive perturbations based on Gaussian random walk are applied to optimal individuals to improve local optima escape capability. Recognizing the density factor’s crucial role in balancing exploration and exploitation, we integrate a golden sine-cosine strategy to optimize search performance. The enhanced algorithm is evaluated through comprehensive experiments on 15 benchmark functions and selected CEC 2017 test functions, with statistical validation using multiple nonparametric tests. Experimental results demonstrate that P-AOA achieves significant improvements in optimization precision, convergence speed, and robustness compared to existing algorithms. Furthermore, we validate its practical effectiveness by applying P-AOA to satellite constellation mission scheduling under multiple constraints, confirming its capability in solving complex discrete optimization problems.