Zero-Emission Hydrofoil Fast Ferries: Design, Modelling, and Performance Evaluation

Hydrofoil technology can potentially reduce the resistance and energy requirement of high-speed ferries, thus lowering fuel consumption and emissions or extending the range of zero-emission vessels. Although substantial amounts of research has been done on this technology, previous efforts have mainly focused on developing simulation methods or analysing earlier generations of vessels with surface-piercing hydrofoil systems or service speeds over 45 knots. Recently, a new generation of hydrofoil fast ferries has emerged, with fully submerged hydrofoil systems and service speeds of 25-35 knots. Battery-electric versions have been suggested, and their designers claim that their hydrofoil technology substantially increases the range and the feasibility of zero-emission operation of fast ferry routes. However, no scientific studies have been performed to support these claims. This thesis takes the first steps towards closing this research gap. In addition, it studies the importance of take-off, surface roughness, and air resistance for the energy requirements, evaluates the relative shares of resistance originating from the main vessel components, and explores an idea for improving the transport efficiency in waves by flight control systems based on artificial neural networks and reinforcement learning. To address these questions, a fully transparent and algorithmic design process has been developed for hydrofoil vessels, accompanied by a simulation tool to assess vessel performance. The design algorithm approximates the geometries of contemporary vessel concepts from a combination of public information and physical constraints set by cavitation and structural considerations, and accounts for hydrofoil interaction effects, practical span limitations, and requirements for passive pitch stability. The simulation tool builds upon well-established non-linear dynamic lifting line theory, with modifications that adapt it for hydrofoil simulations and that exploit a series of simplifications and optimisations to decrease the simulation time. The tool was validated by comparison to aerodynamic and hydrodynamic experiments and new Reynolds-averaged Navier-Stokes (RANS) simulations. The force predictions for a hydrofoil in calm water and waves originating from the final version of the simulation tool corresponded within 2.9% from those of RANS simulations with 83.9 million cells. A framework of simulation modules for kinetics, kinematics, and flight control systems was