Data-driven prediction of added-wave resistance on ships in oblique waves—A comparison between tree-based ensemble methods and artificial neural networks

Malte Mittendorf,Ulrik D Nielsen,Harry B Bingham

doi:10.1016/j.apor.2021.102964

Malte Mittendorf, Ulrik D Nielsen + Show 1 more

Open Access

https://doi.org/10.1016/j.apor.2021.102964

Copy DOI

Abstract

The present paper implements machine learning methods for the prediction of the added-wave resistance of ships in head to beam wave conditions. The study is focused on non-linear regression algorithms namely Random Forests, Extreme Gradient Boosting Machines and Multilayer Perceptrons. The employed dataset is derived from results of three different potential flow methods covering a wide range of operational conditions and 18 hull forms in total. The rational data preprocessing makes up the core part of the paper having its focal point on practical application. Moreover, a rigorous hyperparameter study based on Bayesian optimization is conducted, and the validation of the final models for three case studies against numerical and experimental data as well as two established prediction techniques shows satisfactory generalization in case of the neural network. The tree-based ensemble methods, on the other hand, are not able to generalize sufficiently from the given parameter discretization of the underlying dataset.

Highlights

The maritime industry is changing: The increased environmental awareness of society and policymakers has lead to the introduction of several rules and regulations incentivizing the enhancement of energy efficiency of ships, e.g. the mandatory compliance of new and existing ship designs with the Energy Efficiency Design Index (EEDI) baseline, International Maritime Organization (2011)
To pinpoint the non-linear driving factors of RAW, according to the chosen models, the features are presented according to their relevance obtained from the already described mean decrease impurity (MDI) and mean decrease accuracy
The multilayer perceptron (MLP) as well as the Random Forest (RF) yield a performance increase of 35.51% and 36.87%, respectively

Summary

Introduction

The maritime industry is changing: The increased environmental awareness of society and policymakers has lead to the introduction of several rules and regulations incentivizing the enhancement of energy efficiency of ships, e.g. the mandatory compliance of new and existing ship designs with the Energy Efficiency Design Index (EEDI) baseline, International Maritime Organization (2011). The reliable determination of the added-wave resistance and of the factor fw is problematic both when using model test data and results of numerical calculations as has been shown by Shigunov et al (2018). For this reason, the present work provides a machine learning (ML) approach for the prediction of the added resistance in waves and compares the model estimates to traditional prediction methods. It is stressed that the ML task itself is considered as a supervised regression approach for an efficient estimate of the added-wave resistance without any detailed hull shape information

Methods

Results

Conclusion