Designing ship hull forms using generative adversarial networks

TL;DR

This paper introduces a GAN-based approach to generate ship hull forms based on performance parameters like drag and tonnage, offering a new way to design hulls aligned with owner requirements.

Contribution

It presents a novel GAN model that generates ship hull forms from performance metrics rather than geometric parameters, bridging a gap in ship design methodology.

Findings

Successfully generated ship hull data with small errors

Demonstrated the model's ability to produce hull forms based on performance parameters

Validated the approach through numerical experiments

Abstract

We proposed a GAN-based method to generate a ship hull form. Unlike mathematical hull forms that require geometrical parameters to generate ship hull forms, the proposed method requires desirable ship performance parameters, i.e., the drag coefficient and tonnage. The requirements of ship owners are generally focused on the ship performance and not the geometry itself. Hence, the proposed model is useful for obtaining the ship hull form based on an owner's requirements. The GAN model was trained using a ship hull form dataset generated using the generalized Wigley hull form. The proposed method was evaluated through numerical experiments and successfully generated ship data with small errors.