Mathematics

Nowadays, one of the most critical challenges of naval industries is the reduction of underwater noise due to undesired behavior of ship propellers. The goal of this talk is to provide a shape optimization pipeline for marine propellers in order to improve efficiency and reduce vibrations. A preliminary step in the pipeline is the geometrical parametrization of a propeller blade and its subsequent deformation, resulting in a large number of deformed shapes. Then, a data-driven Reduced Order Model (ROM) is built following a standard 'offline-online' procedure. In an expensive offline phase, the full order model is set up in the open-source software OpenFOAM and used to run the CFD simulations for all the deformed propellers. The collected high-fidelity snapshots and the deformed parameters are used in the online stage to build the non-intrusive ROM. The optimization is then performed using a genetic algorithm that exploits the computational speed-up of ROMs to maximize the efficiency. Finally, a proof of concept of the proposed pipeline is provided, where the optimized propeller improves the efficiency of the original propeller.