Abstract

The growing demand for efficient and reliable propulsion systems in small and medium-sized vessels has revealed the limitations of conventional shaft-based systems, which often suffer from low transmission efficiency, high maintenance costs, and excessive noise and vibration. Even advanced pod-type propulsors frequently fail to meet the specific efficiency needs of small underwater vehicles operating in confined spaces. The hubless rim-driven thruster (RDT) presents a novel propulsion concept by integrating the ducted propeller and electric motor into a single, compact unit, eliminating the traditional bulky shaft and sealing components. This design enhances hydrodynamic efficiency and offers superior resistance to entanglement, making it an ideal solution for challenging marine environments. This study explores the parametric design and optimization of hubless RDT blades to enhance propulsive performance in shallow-water and small-scale underwater applications. Using STAR-CCM+ simulations, we systematically analyze the impact of paddle blade pitch angle and chord length on hydrodynamic performance. The resulting comprehensive dataset is utilized to train a machine learning surrogate model, facilitating rapid performance prediction and optimization. By employing an improved adaptive Sparrow Search Algorithm, we identify the optimal geometric parameters, achieving a 3–7% increase in propulsion efficiency while maintaining structural reliability under optimal operating conditions. The findings of this research provide a robust, data-driven methodology for designing the next generation of highly efficient and reliable propulsion systems for small underwater vehicles, contributing significantly to the field of marine engineering.