Modeling of Ship Propulsion Performance

Benjamin Pjedsted Pedersen, Jan Larsen

AbstractThe paper addresses prediction of propulsion power from measurement of external variables. Propulsion pwer was prediceted with an relative error of less than 2.7%, using a single hidden layer neural network while classical methods have an relative error of more than 5%. The prediction was carried out for four different states with the following input variables: ship speed, relative wind speed and direction, air temperature and sea water temperature.
KeywordsPropulsion; performance; monitoring; non-linear neural network regression; neural networks; fuel consumption
TypeConference paper [With referee]
ConferenceWorld Maritime Technology Conference WMTC2009, Jan. 21-24
Year2009    Month January
PublisherThe Institute of Marine Engineers
AddressMumbai, India
Electronic version(s)[pdf]
Publication link
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing