PIANO - Physics Informed Machine Learning Architecture for Optimal Auxiliary Wind Propulsion

Research Project, 2024 – 2027

The IMO decarbonization goals are stimulating R&D on wind-assisted propulsion systems (WAPS) and boosting their installations in the shipping industry. To facilitate the fast market uptake of WAPS and maximize utilization of renewable wind energy from WAPS for ship propulsions, further socio and technical R&D are necessary to, 1) increase the automation efficiency of WAPS control and operations; 2) win market confidence; 3) gain best operation practices; and 4) facilitate further optimization of future WAPS design. Especially adding WAPS onboard ships can completely change a ship's performance characteristics. Therefore, systematical crew training is urgently needed to better prepare seafarers for efficient WAPS ship operations. However, dedicated maritime simulator-based training on shipping energy efficiency with respect to WAPS, and systematical investigation of human factors related to operating WAPS in real/simulated shipping environments, are rarely available.

 
To overcome the above issues, the main objective of this PIANO project is to develop physics-informed machine learning architecture for ship hull-WAPS-rudder-engine-CPP dynamic coupling models, transparent evaluation guidelines, simulator-based training scenarios, and their demonstration cases for WAPS operations and human factors.