Application of Deep Reinforcement Learning to Predict Shaft Deformation Considering Hull Deformation of Medium-Sized Oil/Chemical Tanker

The enlargement of ships has increased the relative hull deformation owing to draft changes. Moreover, design changes such as an propeller diameter and pitch have occurred compensate for reduction in engine revolution consequent ship speed. In terms propulsion shaft alignment, load stern tube support bearing increases, uneven distribution occurs between bearings, leading accidents. To prevent accidents ensure system stability, a technique is required which resulting from considered. Based on measurement data medium-sized oil/chemical tanker, this study presents novel approach predicting following through inverse analysis using deep reinforcement learning, opposed traditional prediction techniques. main reaction force, was difficult reflect previous studies, predicted with high accuracy by comparing it measured value, reasonable could be derived according deformation. learning expected expandable dynamic behavior operating vessel.