Skip to main navigation Skip to search Skip to main content

A hybrid modelling approach to evaluating the effects of a vessel’s biofouling state on hull and propeller performance

  • Iliya Valchev

Student thesis: Master's Thesis

Abstract

With the sophistication of shipboard systems, maintenance is becoming more time-consuming and skill-dependant. Precise and timely planning, which can be achieved through Condition Based Maintenance, is key in overcoming these challenges, as well as ensuring the longevity and efficiency of assets. The rapid growth of ship monitoring systems means that failures and performance shifts can be linked to variations in measured parameters. This knowledge can be used to limit performance losses, through the identification of undesirable trends. The current project seeks to predict the health status of the hull and the propeller, through the combination of performance modelling approaches from two distinct schools, first-principle techniques and data-driven methods, with the data collection capabilities of modern vessels. Reductions of the vessel’s performance at a certain operational and environmental setting, when compared to a reference ’clean vessel state’ can be traced to the extent of fouling of the hull & propeller, whose condition can be determined separately. A hybrid methodology, combining the accuracy, speed, and flexibility of data-driven methods with the physical knowledge of first-principle models is considered the ideal candidate for the above tasks. The main aim of the research project is to develop a set of real-time fault detection, which can be used to supplement a maintenance strategy, reducing maintenance loads and crew requirements. The developed novel hybrid prediction models utilise both data-driven and first-principle approaches, which has not been previously applied to the problem of evaluating the vessel performance shift due to biofouling. Finally, a case study is employed to demonstrate the numerical tool’s ability to identify shifts in ship capability. Due to the generous collaboration of DAMEN Shipyards and The Royal Netherlands Navy, the developed methodology is validated and tested on real-world data.
Date of Award29 May 2025
Original languageEnglish
Awarding Institution
  • University Of Strathclyde
SponsorsUniversity of Strathclyde
SupervisorMaurizio Collu (Supervisor) & Andrea Coraddu (Supervisor)

Cite this

'