Ship performance monitoring using machine-learning

The hydrodynamic performance of a sea-going ship varies over its lifespan due to factors like marine fouling and the condition anti-fouling paint system. In order accurately estimate power demand fuel consumption for planned voyage, it is important assess ship. current work uses machine-learning (ML) methods using onboard recorded in-service data. Three ML methods, NL-PCR, NL-PLSR probabilistic ANN, are calibrated data from two sister ships. models used extract varying trend in ship’s time predict change through several propeller hull cleaning events. predicted compared with corresponding values estimated friction coefficient (ΔCF). found be performing well while modeling state ships ANN model best, but results NL-PCR not far behind, indicating that may possible use simple solve such problems help domain knowledge.