Browse by author
Lookup NU author(s): Dr Yi Zhou, Dr Kayvan PazoukiORCiD, Dr Rosemary NormanORCiD
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
Operational measures and policies by International Maritime Organization have been proposed to reduce shipfuel consumption and associated emissions. These initiatives emphasise the need for precise predictive modelsthat account for the impact of real operational conditions on Fuel Oil Consumption (FOC). FOC is an essentialfinancial concern for fishing vessels, often comprising over 50% of operating costs. Accurate prediction of FOCduring fishing trips and events is essential but challenging. The challenge lies in the distinct characteristics of fuelconsumption across the operation modes, and the various factors that influence it. This study introduces a novelgrey-box modelling approach for the main engine FOC for a tuna purse seiner, combining a Multi-head BidirectionalLong Short-Term Memory (Bi-LSTM) network with a domain-knowledge based operating modedetection methodology. This approach leverages the sequential data processing capabilities of attention enhancedBi-LSTM along with domain-specific insights. As a result, the FOC model developed using the proposedapproach demonstrated a mean accuracy of at least 97.66% across 10 unseen cruising trips and at least90.93% across eight fishing events, respectively, which suggests that this methodology has the potential to offeraccurate predictions for decision support systems aimed at optimising the operations of tuna purse seiners.
Author(s): Zhou Y, Pazouki K, Norman R
Publication type: Article
Publication status: Published
Journal: Ocean Engineering
Year: 2025
Volume: 324
Print publication date: 30/04/2025
Online publication date: 24/02/2025
Acceptance date: 18/02/2025
Date deposited: 28/02/2025
ISSN (print): 0029-8018
ISSN (electronic): 1873-5258
Publisher: Elsevier
URL: https://doi.org/10.1016/j.oceaneng.2025.120733
DOI: 10.1016/j.oceaneng.2025.120733
Altmetrics provided by Altmetric
