Domain knowledge-driven decomposition-based large-scale optimization for ship cabin structures

This paper proposes a domain knowledge-driven large-scale optimization algorithm for ship cabin structures based on a decomposition optimization framework.

The proposed algorithm combines domain mechanical knowledge with a general black box optimization algorithm, groups the design variables into location variables and size variables, and decomposes the original problem into a series of low-dimensional subproblems. Due to the monotonicity and locality of each bending stress, shear stress, and deformation constraint, subproblems with larger constraint margins are prioritized for optimization. All of the location variables are grouped into one subproblem, and the corresponding subproblem's objective function is to maximize the minimum constraint margin. Each girder size variable is separately grouped, and the corresponding subproblem's objective function is the weight of the cabin structure. Additionally, a surrogate model is introduced to quickly predict the constraints of each subproblem, and the sample infill criterion is adopted only in the constraint surrogate model.

The experimental results show that the algorithm can reduce the overall weight of the cabin structure by 43.5% compared to the upper bound.

The proposed algorithm has higher optimization efficiency and can obtain a better optimization solution compared to both the differential evolution algorithm directly using the using finite element method and the general black box optimization algorithm.