Decision Support for Energy Efficiency Operations of Double Ended Ferry
Publicerad
Författare
Typ
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
A year long navigation data was available for the double ended ROPAX ship "Uraniborg"
owned by the company Rensiery Ventrafiken AB. Exploratory data analysis
shows that allocating most of the engine’s load on the stern-most engine has potential
for energy savings. Furthermore, a full black box machine learning XG Boost
model-based simulator was built in order to forecast the total fuel consumption for
a trip, given the meteocean conditions and some engine-related initial conditions
(IC), on the assumption that the speed overground and the route remains the same.
The simulator acts as a Decision Support System that allows an expert operator to
make decisions on how to allocate the Power on the ship for a given trip, as well as
to improve the Engine Speed operation of either engine.
Beskrivning
Ämne/nyckelord
Decision Support, XGBoost, Machine Learning, Ropax, Power Forecasting