Power prediction for a vessel without recorded data using data fusion from a fleet of vessels
作者：
Highlights：
• Neural networks are used due to the complexity of the interactions within the data.
• A fusion of data from different vessels extends the range of accurate extrapolation.
• Validates that ship powering can be predicted to a mean error of less than 2%.
• Achieves 4% error for prediction for vessels where there is no available data.
• Higher errors are shown for predictions outside of the common domain of the dataset.
摘要
•Neural networks are used due to the complexity of the interactions within the data.•A fusion of data from different vessels extends the range of accurate extrapolation.•Validates that ship powering can be predicted to a mean error of less than 2%.•Achieves 4% error for prediction for vessels where there is no available data.•Higher errors are shown for predictions outside of the common domain of the dataset.
论文关键词：Machine learning,Shaft power prediction,Neural networks,Ocean engineering,Naval architecture
论文评审过程：Received 16 June 2020, Revised 10 May 2021, Accepted 22 September 2021, Available online 29 September 2021, Version of Record 13 October 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115971
