An Enhanced Beluga Whale Optimization Algorithm for Engineering Optimization Problems

This paper presents an enhanced beluga whale optimization algorithm (EBWOA) for engineering optimization problems. To enhance the performance and address the challenges of poor convergence and suboptimal solution stagnation commonly faced by the original beluga whale optimization algorithm (BWOA), EBWOA employs a two-step approach. In the initial stage, a dynamic update factor is introduced to accelerate convergence during the exploitation phase of BWOA. Subsequently, the second stage incorporates the Cauchy mutation operator to inject diversity into the population, preventing it from becoming entrapped in local optima. The proposed enhancement is validated on 15 classical benchmark functions and CEC-19 functions in terms of solution quality and convergence speed. To assess the efficiency of EBWOA, the algorithm is applied to a real-world industrial problem, specifically, the spiral steel pipe manufacturing system (SSPM), serving as a case study and four classical engineering design optimization problems. The simulation results demonstrated the superiority of EBWOA in optimizing the fuzzy availability of the SSPM industrial system and successfully solving all four constrained engineering design problems when compared to other recent metaheuristics.