Objective Hull fouling severely impairs the sailing efficiency of ships. In actual ship operations, fouling removal is usually conducted within the scheduled dry-docking period. However, this approach often fails to perform fouling removal in a timely manner at the optimal time, resulting in a substantial increase in ship fuel consumption costs. To address this issue, this study proposes an operational data-driven hull fouling assessment method. This method can real-time evaluate the performance loss caused by hull fouling, thereby providing a basis for fouling removal decision-making.

Method First, based on the data collected from the ship during the non-fouling period and the meteorological forecast data, a multi-layer neural network model is established. This model is designed to achieve accurate prediction of fuel consumption per nautical mile. Then, the hull fouling condition is assessed by comparing the deviation between the prediction results of the non-fouling model and the actual measured values. Additionally, a distance threshold screening method is adopted to filter the data in the assessment segment. This step aims to avoid prediction errors caused by model drift and ensure the reliability of the assessment results.

Results Three segments of non-training data were selected for validation, corresponding to the periods before hull fouling, during hull fouling, and after fouling removal, respectively. For the data of the pre-fouling and post-fouling-removal periods, the percentage deviation of the model's predicted fuel consumption per nautical mile was approximately 7%. In contrast, for the data of the in-fouling period, the percentage deviation of the model's predicted fuel consumption per nautical mile exceeded 13%, showing a significant increase in model deviation during the hull fouling period.

Conclusions The validation results demonstrate that the proposed method can effectively assess the hull fouling condition of ships. The percentage deviation between the model's predicted values and the actual measured values can be regarded as the incremental fuel consumption caused by hull fouling. This finding is conducive to further calculating the benefits of fouling removal.