EBWO: a multi-strategy collaborative enhanced Beluga Whale Optimization algorithm

Abstract

Abstract To address the limitations of the Beluga Whale Optimization (BWO) algorithm, including insufficient population diversity, susceptibility to local optima, and room for improvement in convergence speed, this paper proposes a multi-strategy enhanced Beluga Whale Optimization (EBWO). The EBWO integrates three improvement strategies: (1) An elite set strategy to preserve high-quality individuals, thereby maintaining diversity and guiding the search; (2) Adaptive Cauchy mutation to introduce perturbations, enhancing global exploration, preventing premature convergence, and balancing exploration and exploitation; (3) Incorporation of a differential evolution mutation mechanism, leveraging global best individual information and differential vector perturbations to accelerate convergence and improve computational efficiency. Comprehensive validation using the CEC-2022 benchmark test suite, statistical analysis, and real-world engineering optimization problems demonstrates that, compared to the original BWO, the EBWO achieves significantly higher convergence accuracy, markedly faster convergence speed, and exhibits superior global search capability and stability. The EBWO provides a more efficient and stable novel approach for solving complex optimization problems.