Seakeeping analysis of monohull ships at preliminary design using artificial neural networks

Abstract

Nowadays seakeeping is mostly analysed by means of model testing or numerical models. Both require of a significant amount of time and of the exact hull geometry, and this is why seakeeping is not taken into account at the early stages of ship design. Then, the main objective of this work is the development of a seakeeping prediction tool to be used in the early design. Hence this tool must be fast, accurate, and must not require the exact hull shape. To this end, an artificial intelligence (AI) algorithm has been developed. This algorithm is based on artificial neural networks (ANN) and only requires the ship coefficients of forms.

The methodology developed to obtain the predictive algorithm is presented as well as the database of ships used for training the ANN. The training data were generated using a seakeeping code based on the boundary element method (BEM). Also, the AI predictions are compared to the BEM results using both, ship shapes from the training database and from outside.

As a result it has been obtained an AI tool capable of predicting seakeeping almost instantly for a wide range of monohull merchant ships. And the difference in results, with respect to the BEM code used for the training, is lower than 5%.

Full document