Ships diesel engine performance modelling with combined physical and machine learning approach

  • Coraddu, Andrea (Principal Investigator)
  • Oneto, Luca (Principal Investigator)
  • Geertsma, Rinze (Principal Investigator)

Project: Non-funded project

Project Details

Description

The proposed research will investigate a novel approach of predicting various diesel engine performance parameters using the physical models from Delft University and the State of the Netherlands in combination with Machine Learning (ML) algorithms from Strathclyde University - NAOME, Genoa University and Damen Schelde Naval Shipbuilding. A dataset of the Holland class Oceangoing Patrol Vessels (OPV’s) from the State of the Netherlands will be used to train the machine learning algorithms and establish its performance. Moreover, the research will analyse which performance parameters can be predicted well with a physical modelling approach and which ones with a combined physical and ML approach. The research is expected to predict parameters such as engine efficiency, engine thermal loading, temperature before the turbine and exhaust valve temperature. Finally, the research will discuss how the proposed models can be used to reduce the maintenance effort on diesel engines in future using these techniques and how to integrate these into ship control systems
StatusFinished
Effective start/end date1/05/181/05/21

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.