A Machine Learning Approach for Estimating the Power
of a Ship: Utilizing Historical Operational Data

Milovanovic, Mark

dc.contributor.author	Milovanovic, Mark	en
dc.date.accessioned	2023-04-06T07:36:07Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/57492
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.25189
dc.rights	Default License
dc.subject	Power Prediction	en
dc.subject	Neural Networks	en
dc.subject	Fouling	en
dc.subject	Ship Performance	en
dc.subject	Machine Learning	en
dc.title	A Machine Learning Approach for Estimating the Power of a Ship: Utilizing Historical Operational Data	en
heal.type	bachelorThesis
heal.classification	Marine Engineering	en
heal.dateAvailable	2024-04-05T21:00:00Z
heal.language	en
heal.access	embargo
heal.recordProvider	ntua	el
heal.publicationDate	2023-03-09
heal.abstract	The primary objective of this thesis is to address a major concern in the shipping industry, which is predicting power consumption due to fouling in ships. Fouling results from mechanical damage and the growth of living organisms such as barnacles, slime, and seaweed on the hull’s surface, leading to reduced efficiency and increased power consumption. The study examines the effectiveness of hybrid models for power prediction by combining physical models with a data-driven neural network model. The research utilizes data collected from two sister crude oil tanker ships between January 2019 and December 2022. Wind and wave resistance calculations are also incorporated to achieve a precise understanding of the ship’s performance. The study evaluates the performance of two separate models, one taking into account idle periods as parameters and the other not. Additionally, the study investigates the correlation between the level of biofouling on a ship and its chlorophyll concentration. The research results demonstrate that the proposed method achieved promising results, as indicated by the metrics used. Furthermore, the study highlights the potential for future research to explore the correlation between idle periods and the level of biofouling.	en
heal.advisorName	Παπαλάμπρου Γεώργιος	el
heal.advisorName	Papalambrou, Georgios	en
heal.committeeMemberName	Samouilidis, Manolis
heal.committeeMemberName	Papadopoulos, Christos	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ναυπηγών Μηχανολόγων Μηχανικών	el
heal.academicPublisherID	ntua
heal.fullTextAvailability	false