Objectives With the growing development and utilization of Arctic resources, the opening and operation of Arctic waterways have made ice-strengthened ships and icebreakers essential for Arctic operations. Ice-class propellers, as the primary propulsion component of these vessels, are required to provide greater thrust to overcome substantial ice resistance (accounting for more than half of the total resistance), which necessitates higher rotational speeds. However, increasing propeller rotation speed at a constant ship velocity subjects the propeller blades to heavy loads, causing blade surface pressure to often drop to saturated vapor pressure, thus inducing cavitation. This effect is further aggravated when ice blocks slide along the hull into the inflow field in front of the propeller, causing ice blockage occurs and intensifying cavitation. Existing studies have focused on the hydrodynamic and cavitation performance of ice-class propellers under ice blockage. However, studies addressing cavitation induced by ice blocks in icy environments and the associated radiated noise remain limited. Cavitation noise, characterized by medium- and low-frequency components, exhibits long-range propagation and low attenuation. This noise not only interferes with the normal operation of precision instruments aboard ships and contributes to marine noise pollution but also affects acoustic detection equipment due to its frequency-domain acoustic properties. Therefore, this study aims to explore the cavitation phenomenon, propeller blade excitation forces, and noise radiation caused by uneven inflow resulting from ice blocks sliding into the propeller's inflow field when ice-class propellers operate in icy environments. It seeks to clarify the coupling mechanism of ice blockage-vortex cavitation-noise, thereby addressing a research gap in this field.

Methods To accurately capture the detailed evolution of the flow field and precisely predict the radiated noise, this study employed an integrated numerical method combining the hybrid large eddy simulation/Reynolds-averaged Navier−Stokes (LES/RANS) method with the Ffowcs Williams−Hawkings (FW−H) equation. Specifically, the improved delayed detached eddy simulation (IDDES), based on the hybrid LES/RANS concept, was used to solve the Navier-Stokes (N−S) equations, leveraging the high precision of LES and the high efficiency of RANS. The RANS method was applied in near-wall regions, while the LES mode was automatically activated in the wake core region to accurately resolve the transient vortex structures and their evolution process. The Schnerr-Sauer cavitation model was introduced to simulate the cavitation phenomenon on the ice-class propeller, and the FW−H equation was used to calculate the far-field noise radiation. The research object was an ice-class propeller with a scale ratio of 1:28 and a diameter of 0.25 m after scaling. An ice blockage was installed in the inflow direction in front of the propeller, with dimensions of 1.72D (length), D (width), and 0.5D (height) (D is the propeller diameter). The numerical simulation was conducted using the STAR-CCM+ platform, with the cavitation number σn fixed at 1.5, and the advance coefficient J set to 0.35, 0.45, 0.55 and the ice blockage-propeller distance L/D set to 0.15 and 0.5 as the variable parameters. The simulation process consisted of two steps: first, the RANS method was used to calculate the steady flow field; then, on the basis of the steady flow field, the IDDES method coupled with the FW−H equation was used to calculate the unsteady flow field and radiated noise. To ensure the reliability of the simulation results, grid independence was verified using the grid convergence index (GCI) method by comparing three sets of grids with different quantities (2.42×10⁷, 1.26×10⁷, and 6.35×10⁶). Finally, the medium grid (1.26×10⁷) was selected considering both calculation accuracy and efficiency. In addition, 108 acoustic monitoring points were positioned on three planes (xoz, yoz, xoy), each located10 m from the propeller shaft center, to monitor the radiated noise.

Results The numerical results showed that the hybrid LES/RANS method, combined with the Schnerr-Sauer cavitation model, demonstrated good numerical accuracy. The error between the hydrodynamic calculation results and the experimental fluid dynamics (EFD) data was within 3.0%, which was more accurate than the RANS method (with an error of approximately 5.0%). Under low cavitation number conditions (σ_n = 1.5), severe cavitation caused minimal changes in the hydrodynamic coefficients (thrust coefficient K_T, torque coefficient K_Q, and open water efficiency η₀) as the ice blockage distance decreased. As the advance coefficient J increased from 0.35 to 0.55, K_T and K_Q decreased, while η₀ increased. The single-blade excitation force exhibited periodic behavior: when the propeller blade rotated behind the ice blockage, the excitation force increased sharply, reached the maximum behind the ice blockage, and then gradually decreased. The excitation force at L/D = 0.15 occurred later and was more pronounced compared to that at at L/D = 0.50. With the increase of the advance coefficient, the average value of the single-blade excitation force decreased, but the peak value increased. The excitation force was highest at L/D = 0.15 and J = 0.55. The combined excitation forces of the four blades resulted in four excitations during one rotation period. Regarding radiated noise, the sound pressure level (SPL) exhibited peaks at the harmonics of the shaft frequency fn. At L/D = 0.15 and J = 0.35, the peaks appeared at 2fn and 4fn, with values of 150.89 dB and 139.84 dB respectively. With the increase of the advance coefficient, the peaks at 2fn and 4f_n grew, and additional peaks emerged at 6f_n, 8f_n, and 10f_n. The overall sound pressure level (OSPL) at L/D = 0.15 was greater than that at L/D = 0.50, and the OSPL increased with the increase of the advance coefficient. The ice blockage caused a sharp increase in the cavitation on the blade surface, with smaller blockage distances resulting in larger cavitation coverage areas. The evolution of cavitation morphology during one rotation period explained the changes in the excitation force. Vortex cavitation developed at the lower edge of the cavitation zone behind the ice blockage. With the increase of the advance coefficient, the total cavitation coverage area decreased, but the peak cavitation coverage area on the blade behind the ice blockage remained almost unchanged. The increase in cavitation variation contributed to the rise in both the excitation force and radiated noise.

Conclusion This study establishes an integrated numerical method to explore the coupling effects of ice blockage-vortex cavitation-noise of ice-class propellers. It accurately predicts the hydrodynamic performance, cavitation behavior, excitation force, and radiated noise of ice-class propellers under ice blockage and cavitation flow conditions. The findings clarify how the advance coefficient and the ice-propeller distance influence the hydrodynamic performance, excitation force, radiated noise, and cavitation performance of ice-class propellers under low cavitation numbers. The study reveals the mechanism by which ice blockage and cavitation jointly affect the excitation force and radiated noise: the ice blockage intensifies cavitation, and the subsequent sharp increase in the cavitation coverage area, coupled with the violent cavitation collapse, leads to higher excitation force and radiated noise. The increase in advance coefficient and decrease in blockage distance exacerbate this phenomenon. This study fills a significant gap in the field of cavitation-induced noise of ice-class propellers under ice blockage conditions, providing an important theoretical basis for the anti-cavitation design of ice-class propellers and the mitigation of excitation forces and radiated noise. The results have important guiding significance for improving the operational safety, reliability, and environmental sustainability of ice-class propellers in Arctic operations.