A new power prediction method using ship in-service data: a case study on a general cargo ship
Artikel i vetenskaplig tidskrift, 2023

To increase energy efficiency and reduce greenhouse gas (GHG) emissions in the shipping industry, an accurate prediction of the ship performance at sea is crucial. This paper proposes a new power prediction method based on minimizing a normalized root mean square error (NRMSE) defined by comparing the results of the power prediction model with the ship in-service data for a given vessel. The result is a power prediction model tuned to fit the ship for which in-service data was applied. A general cargo ship is used as a test case. The performance of the proposed approach is evaluated in different scenarios with the artificial neural network (ANN) method and the traditional power prediction models. In all studied scenarios, the proposed method shows better performance in predicting ship power. Up to 86% percentage difference between the NRMSEs of the best and worst power prediction models is also reported.

artificial neural networks (ANN)

Power prediction

GHG emission

in-service data

ship performance

Författare

Ehsan Esmailian

Kumera Marine AS (Hjelseth)

Norges teknisk-naturvitenskapelige universitet

Youngrong Kim

Norges teknisk-naturvitenskapelige universitet

Chalmers, Mekanik och maritima vetenskaper, Maritima studier

S. Steen

Norges teknisk-naturvitenskapelige universitet

Kourosh Koushan

SINTEF Ocean

Norges teknisk-naturvitenskapelige universitet

Ship Technology Research

0937-7255 (ISSN) 20567111 (eISSN)

Vol. In Press

Ämneskategorier

Energiteknik

Farkostteknik

Marin teknik

DOI

10.1080/09377255.2023.2275378

Mer information

Senast uppdaterat

2023-11-20