Investigation on RANS prediction of propeller induced pressure pulses and sheet-tip cavitation interactions in behind hull conditions

Abstract

Propeller blades’ cavitation is not only a significant cause of Underwater radiated Noise (URN), but it is also detrimental to the propeller blades and ship engines due to resulting erosion and vibrations, respectively. Over the years, different approaches have been engaged to predict cavitation using viscous flow methodologies accurately. However, the practicalities of these methodologies have not been stamped. Data from previous experiments in open-water, full and model scale are used for verification. Details of the experimental approach used are not included in this report. This thesis investigates propeller-induced pressure pulses and resulting cavitation providing visuals of the cavities formed corresponding to the pressure broadband spectra. In decades past, sheet cavitation has been predicted more often in comparison to tip-vortex cavitation. However, this thesis aims also to shed light on blade-tip cavitation, showing a detachment from sheet to vortex cavitation from the propeller blade trailing edge tip. The upstream wake on which the propeller works is simulated from the provided ship hull. This is done by performing steady-state simulations. After which, transient simulations are performed. The transient simulations comprise the propeller’s rotations in cavitating and non-cavitating conditions. The simulations aim to depict the real-life scenario by operating the propeller in behind-hull conditions. However, the free surface and resulting kelvin waves are neglected in this thesis. The investigation analyses the accuracy of the approach used. Initially, the wake field is predicted after which spectral analyses of the pressure pulses are performed and compared to the experimental. The cavitating pattern is also reviewed. The studies engaged in this report show the RANS methodology provide a accurate prediction of the propeller rotation phenomenom in non-cavitating conditions. However, the RANS approach requires a highly independent and fine mesh to make correct predictions in cavitating conditions. The power spectral densities predicted show disparities with the experimental result. However, an analogous visualization of cavitation is achieved. The predictions is highly dependent on the accurate ship-propeller wake prediction.