A Machine-Learning Approach to Predict Main Energy Consumption under Realistic Operational Conditions

Abstract The paper presents a novel and publicly available set of high-quality sensory data collected from a ferry over a period of two months and overviews existing machine-learning methods for the prediction of main propulsion efficiency. Neural networks are applied in both real-time and predictive settings. Performance results for the real-time models are shown. The presented models were successfully deployed in a trim optimisation application onboard a product tanker.