Back to search

National Technical University of Athens (NTUA)

Performance parameter recognition and damage prevention in marine shafting systems utilizing machine learning & artificial intelligence algorithms

Abstract

dc:description

The propulsion powertrain of ships relies crucially on the shafting system, a component subject to rigorous regulatory standards ensuring reliability and operability under adverse conditions throughout the vessel's life cycle. Unlike many shipboard systems, the shafting system's lack of redundancy options makes its infrequent failures highly perilous. Such failures pose serious threats to both crew safety and vessel integrity, resulting in unforeseen repairs and costs. This research establishes a comprehensive framework for assessing the performance of the shafting system, facilitating the identification of key operational parameters, recognizing system states, and evaluating these states through comparative assessment tools and data obtained from sensors merged with artificial intelligence techniques. This study aims to prevent failures by pinpointing and avoiding critical or potentially hazardous operating conditions through the development of multiple ML and AI algorithms. To comprehend the intricate nature of the shafting system, various tools are integrated to conduct numerical simulations accounting for the particulars of both shafts and bearings. Performance-operation maps of the entire system are calculated under different geometrical characteristics, rotational speeds, and external loads, determining the safest margins for operation. Conversely, optimal geometric characteristics can be determined by simulating various conditions and ensuring optimal system design. Thus, this framework offers a dual advantage, enhancing the operation of existing installations and optimizing the design of new applications.Additionally, an AI-based system is developed, utilizing training data from surrogate digital models. This system is designed to provide early prognosis of undesired operating conditions and, in emergencies, offer crucial instructions to the ship operators in order to mitigate potential or further damage. Complementing the computational aspect, an experimental process is established, providing the necessary theoretical background, to create an equivalent small-scale model of the shafting system with relative features similar to the large-scale installations. This serves as a test-rig for the study of novel sensor configurations, prior to a full-scale, costly implementation onboard, aiming at improving the telemetry of the shafting system. In conclusion, this holistic study of a ship's shafting system provides a thorough assessment of its current operating state, evaluates key metrics for safe and reliable operation, and broadens the understanding of the complexities involved. Ultimately, this research aims to enhance the resilience of the shafting system, ensuring the safety and efficiency of marine propulsion powertrains.

Degree

thesis:*
Grantor dc:publisher
National Technical University of Athens (NTUA)
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Rossopoulos, Georgios-Nikitas
  • Ρωσσόπουλος, Γεώργιος-Νικήτας

Subjects

dc:subject × 16

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/62434

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Rossopoulos, Georgios-Nikitas; Ρωσσόπουλος, Γεώργιος-Νικήτας. Performance parameter recognition and damage prevention in marine shafting systems utilizing machine learning & artificial intelligence algorithms. National Technical University of Athens (NTUA), 2025. http://hdl.handle.net/10442/hedi/62434