Ship hull resistance minimization using surrogate modelling and an improved dung beetle optimizer

Abstract

The optimization of hull forms is a crucial aspect of ship design optimization. Using surrogate models and intelligent optimization algorithms can significantly enhance the efficiency of hull form optimization. To improve the algorithm's performance, this paper proposes modifications to and validates the dung beetle algorithm. These modifications include introducing Circle chaotic mapping, a sine-cosine fusion mutation Cauchy operator, and the Levy flight strategy at different stages of the algorithm. Based on the improved algorithm and the random forest surrogate model, a 24,000 TEU container ship is used as the research target. Three semi-parametric deformation methods extract design variables to find the hull form optimization parameters for minimum resistance. Comparative analysis of the hull forms before and after the improvements demonstrates that the optimization scheme proposed in this paper decreases the optimal iteration times by about 1% compared to traditional research methods, and significantly reduces ship resistance.