Evaluation of Different Deep-Learning Models for the Prediction of a Ship’s Propulsion Power

Adverse conditions within specific offshore environments magnify the challenges faced by a vessel’s energy-efficiency optimization in Industry 4.0 era. As data rate and volume increase, analysis of big using analytical techniques might not be efficient, or even infeasible some cases. The purpose this study is development deep-learning models that can utilized to predict propulsion power vessel. Two are discriminated: (1) feed-forward neural network (FFNN) (2) recurrent (RNN). Predictions provided these were compared with values measured onboard. Comparisons between two types networks also performed. Emphasis was placed on different pre-processing phases, as well optimal configuration decision process for each developed models. Factors parameters played significant role outcome, such number layers network, evaluated.