Application of machine learning to design low noise propellers

P.S. Doijode

Research output: ThesisDissertation (TU Delft)

81 Downloads (Pure)

Abstract

Over 90% of international trade is carried out over seas. Shipping is currently the cheapest mode of transoceanic transport. The traffic density of shipping lanes on seas, oceans, and also rivers is likely to increase. Consequently, the GHG, NOx, SOx and noise emissions from shipping will rise making it more difficult to meet stricter emission regulations which the IMO aims to achieve. One opportunity to reduce emissions is by designing more efficient and quieter propellers.

To design quieter and more efficient propellers an optimal blade loading solution is required. For a rigid propeller, the blade loading distribution is optimized by modifying the geometry. The propeller geometry must be modified to achieve optimal loading that maximizes efficiency and minimizes acoustic emissions. In addition to efficiency and noise considerations, propeller optimization must consider thrust, ship speed, fairing constraints as well as unsteady wake of the vessel....
Original languageEnglish
Awarding Institution
  • Delft University of Technology
Supervisors/Advisors
  • van Terwisga, T.J.C., Supervisor
  • Hickel, S., Supervisor
  • Visser, K., Advisor
Award date6 Sept 2022
Print ISBNs978-94-6419-576-7
DOIs
Publication statusPublished - 2022

Keywords

  • Explainable Machine Learning
  • Propeller Design
  • ILES
  • Barotropic model

Fingerprint

Dive into the research topics of 'Application of machine learning to design low noise propellers'. Together they form a unique fingerprint.

Cite this