Ship pipeline route intelligent design via energy-field constraint quantification and improved hybrid evolutionary computation

Ship pipeline route design (SPRD) is labor-intensive and complex in constrained 3D ship environments, which often involve diverse and ambiguous engineering rules. Traditional drawing-based approaches rely heavily on designers’ experience, while most existing optimization methods are insufficient for collaborative, multi-category routing tasks on large vessels. To overcome these challenges, this study presents an intelligent SPRD framework that unifies heterogeneous layout constraints and enhances route-search performance. The framework assumes that routing criteria are independent of pipe material and does not explicitly model material-dependent properties or constraints. The ship space is discretized into grids that are classified as free, passable, restricted, attractive, or repulsive regions. Ambiguous requirements, such as maintaining proximity to bulkheads and keeping distance from hazardous equipment, are quantified via a continuous energy field and incorporated into objective functions for various pipeline routing types, including single, multiple, branch, parallel, and multi-category collaborative routing. The framework uses an enhanced Sine Cosine–Differential Evolution (SCA-DE) algorithm, which integrates adaptive parameter variation, position updating, random parameter perturbation, and dynamic population-size adjustment. Experiments on ballast-water and compartment scenarios show that the proposed SCA-DE outperforms particle swarm, genetic, and dung beetle optimizers, achieving the best fitness in most test pipelines and validating its effectiveness and versatility.