EGWO: An Enhanced Grey Wolf Optimizer with Adaptive Momentum for High-dimensional Optimization Problems

This study proposes an enhanced grey wolf optimizer (EGWO) to address premature convergence and diversity loss in high-dimensional optimization. By integrating adaptive dual-mode momentum, multileader guidance, and Lévy-based differential perturbation, EGWO achieves the following: (1) superior high-dimensional stability, as evidenced by Friedman ranks of 1.1 (100D) and 1.4 (50D) on CEC2020, with significantly reduced fitness values; (2) broad generalization capability, ranking first on 22/28 CEC2017 functions (100D), particularly in hybrid/composition categories. Ablation studies validate the necessity of all the components. Applied to pressure vessel design and other engineering applications, EGWO attains the lowest cost with full constraint satisfaction, demonstrating robustness in both theoretical and real-world problems.