A comparison of physics-informed data-driven modeling architectures for ship motion predictions

• Published in: Ocean engineering 2023-10, Vol.286, p.115608, Article 115608
• Main Authors: Schirmann, Matthew L., Gose, James W., Collette, Matthew D.
• Format: Article
• Language: English
• Subjects: Digital twin; Gaussian process; Machine learning; Neural network; Research vessel; Ship motions
• ISSN: 0029-8018; 1873-5258
• DOI: 10.1016/j.oceaneng.2023.115608

How to select the optimal formulations for building blended physics-machine learning models for ship motions is not currently clear. This work compares and contrasts two approaches to this problem: (1) A black-box deep learning approach based on a new neural network architecture that can better handle varying wave conditions, and (2) a clear-box model based on updates to linear response amplitude operators via a Gaussian process regression. Both models are trained and evaluated on a dataset consisting of more than 15,000 30-minute-long motion observation windows from two research vessels at sea in the Atlantic and Pacific oceans. Three different hindcast weather services are used, including two models from the EU’s Copernicus system and NOAA’s WAVEWATCH III. The evaluation shows that a tradeoff exists between the formulations, with the black-box formulation offering higher accuracy and the cost of less transparency. The weather hindcast used has a small impact on the results, and the ability of both models to generalize predictions between near-sister ships is also encouraging for the practical application of these techniques. •Machine learning is applied to 15,000 30-minute at-sea ship motion records.•A black-box and a clear-box machine learning approach are compared.•A custom neural network model structure allows varying wave feature lengths.•Models can generalize between near-sister ships with some accuracy loss.•Weather sources also impact accuracy but to a smaller extent than model structure.