Machine learning techniques for modeling ships performance in waves

Abstract

This paper presents a design of a system for monitoring and recording the influence of a running sea on a vessel
in motion. Our approach is based on machine learning techniques that relate measured wave parameters (encounter
angle, wave height and wave amplitude) with measured motion characteristics of the vessel. High quality GRIB
data for wave measurements are available for some regions (e.g. North Sea and Adriatic) and we use those for
generating training sets. We store this correlation in a neural net and use this information in conjunction with the
targeted performance indicator (RMS of linear acceleration, RMS of roll or pitch angle, fuel consumption) to create
historical directed performance charts for the vessel in consideration. We use this information for rational route
planning and optimization. We report on the conclusions of experiments.

Original document

The different versions of the original document can be found in:

• https://zenodo.org/record/1485160 under the license http://creativecommons.org/licenses/by-nc-nd/4.0/legalcode

• http://dx.doi.org/10.5281/zenodo.1485159,

https://zenodo.org/record/1485160 under the license http://creativecommons.org/licenses/by-nc-nd/4.0/legalcode

• https://zenodo.org/record/1485160,

http://dx.doi.org/10.5281/zenodo.1485160 under the license http://creativecommons.org/licenses/by-nc-nd/4.0/legalcode

DOIS: 10.5281/zenodo.1485160 10.5281/zenodo.1485159