The understanding of the performance of a propeller in realistic operative conditions is nowadays a key issue for improving design techniques, guaranteeing safety and continuity of operation at sea, and reducing maintenance costs. In this paper, a summary of the recent research carried out at CNR-INSEAN devoted to the analysis of propeller loads in realistic operative scenarios, with particular emphasis on the in-plane loads, is presented. In particular, the experimental results carried out on a free running maneuvering model equipped with a novel force transducer are discussed and supported by CFD (Computational Fluid Dynamics) analysis and the use of a simplified propeller model, based on Blade Element Momentum Theory, with the aim of achieving a deeper understanding of the mechanisms that govern the functioning of the propeller in off-design. Moreover, the analysis includes the scaling factors that can be used to obtain a prediction from model measurements, the propeller radial force being the primary cause of failures of the shaft bearings. In particular, the analysis highlighted that cavitation at full scale can cause the increment of in-plane loads by about 20% with respect to a non-cavitating case, that that in-plane loads could be more sensitive to cavitation than thrust and torque, and that Reynolds number effect is negligible. For the analysis of cavitation, an alternative version of the BEMT solver, improved with cavitation linear theory, was developed.

Marine propellers in behind-hull conditions develop, in addition to thrust and torque, in-plane loads that are strictly related to fatigue stress of the propulsive shaft bearings, hull-induced vibrations and the dynamic response of the ship while maneuvering or experiencing wave induced motions. An in-depth understanding of their nature as well as their quantification in typical design and off-design operative scenario is fundamental for improving ship design criteria. This issue is tackled in the present work by means of URANS simulations and simplified propeller theories to assess the correlation between inflow conditions and propeller loads. In particular, the analysis is carried out for the same twin screw model recently considered in free running maneuvering model tests (Ortolani et al., 2015a, 2015b) and further aims to provide a complementary and deeper insight to the outcome of these experiments. The first part of the study is focused on the straight ahead motion and the steady turning maneuvers with rudder deflections of 15°, 25° and 35° and Froude number equal to 0.26.

The present investigation focuses on the effects of the stern appendages and the propulsion system on the hydro-loads generated by the propeller during off-design conditions, with particular emphasis on the in-plane components. Recent experimental investigations carried out by free running model tests [7,8] and CFD analysis [5] for a modern twin screw model, highlighted that maneuvers at small drift angles and yaw rates might be as critical as the tighter ones due to complex propeller-wake interactions. Therefore, design criteria should take into account also these operative conditions, in order to reduce the effects of propeller-wake interaction phenomena that degrade the overall propulsive efficiency, induce shaft/hull structural vibration and increase noise emission. In the present study we analyze the effects of geometric and propulsive modifications with respect to the twin screw configuration studied in [5]. In particular, the effect of the centreline skeg, propeller direction of rotation and control strategies of the propulsion plant on the propeller bearing loads have been investigated from the analysis of the nominal wake in maneuvring conditions, computed by unsteady RANSE simulations coupled with a propeller model based on Blade Element Theory. The considered test cases were turning circle maneuvers with different rudder angles at FN = 0.265.

The numerical study presented in Part I (Dubbioso et al., 2017) on the bearing loads developed by the propellers of a twin screw model during quasi-steady conditions is extended to transient maneuvers. In the previous study, numerical simulations highlighted that the hydrodynamic loads might experience significant peak at moderate turning rates due to complex interaction of the propeller with the wake. In the present paper, the complete turning circle maneuver at ? 1/4 35 ? at Fr 1/4 0:265 is numerically simulated in order to analyze the character of the blade loads during the transient phases after the actuation of the rudder (start and pull-out). The analysis shows that the overall degradation of the propeller performance may occur also at kinematic conditions weaker than those usually considered as the most critical ones (in general, tight maneuvers); therefore, these conditions should be accounted for also in the early design phases.

To comply with the more and more restrictive international standards and regulations for noise and vibration levels on board passenger ships, a renewed interest on secondary N&V sources, with respect to propeller and machinery sources, has been observed. In particular, the increase of ship performances in terms of velocity has been directed on study the hydrodynamic noise sources and among the others turbulent boundary layer (TBL). The great difficulties encountered in simulating the wall pressure fluctuations (WPF) due to TBL at high Reynolds numbers and for complex configurations typical of a real ship have pushed the research community to develop models for WPF based on theoretical considerations and model scale tests. In particular, scaling laws for pressure spectra have been established at least for simple geometries and flow conditions and models of cross spectral density for their spatial characterization have been obtained. Unfortunately, model scale tests do not allow reaching Reynolds number values comparable with full scale conditions. Therefore, to validate current models an experimental campaign devoted to WPF measurements have been performed on the hull of a Ro-Ro Pax vessel. Numerical simulations of the flow around the ship hull were performed to evaluate mean flow parameters.