In the present paper, the analysis of the turning capability of the naval supply vessel presented in Part I (Broglia et al., 2015) is continued with different stern appendages, namely twin rudder and centreline skeg. The main purpose of the analysis is to assess the capability of an in-house CFD tool in capturing the different manoeuvring characteristics of the ship hulls; the test case is challenging, as the difference be- tween the two configurations lies in the complex flow structure related to rudder-propeller interactions. Moreover, although the twin rudder solution slightly improves the poor course keeping ability of the original vessel, the course stability remains poor and, consequently, large lateral motions and drift angle have to be expected during the manoeuvre. The manoeuvring capabilities of the new configuration are discussed and compared with the single rudder configuration, focusing on the nature of the hydrodynamic forces and moments acting on the main hull and appendages during the transient and stabilized phases of the manoeuvre. Emphasis will be also given to the different contributions of the propulsion system in the twin rudder configuration, that results from the different rudder-propeller interaction

Abstract The hydrodynamic characterization of control appendages for ship hulls is of paramount importance for the assessment of maneuverability characteristics. However, the accurate numerical simulation of turbulent flow around a fully appended maneuvering vessel is a challenging task, because of the geometrical complexity of the appendages and of the complications connected to their movement during the computation. In addition, the accurate description of the flow within the boundary layer is important in order to estimate correctly the forces acting on each portion of the hull. To this aim, the use of overlapping multi-block body fitted grids can be very useful to obtain both a proper description of each particular region in the computational domain and an accurate prediction of the boundary layer, retaining, at the same time, a good mesh quality. Moreover, block-structured grids with partial overlapping can be fruitfully exploited to control grid spacing close to solid walls, without propagation of undesired clustering of grid cells in the interior of the domain. This approach proved to be also very useful in reducing grid generation time. In the present paper, some details of the flow simulation around a fully appended submarine is reported, with emphasis on the issues related to the complexities of the geometry to be used in the simulations and to the need to move the appendages in order to change the configuration of the various appendages.

The turning circle manoeuvre of a naval supply vessel (characterized by a block coefficient ^CB~0.60) is simulated by the integration of the unsteady Reynolds-Averaged Navier Stokes equations coupled with the equations of rigid body motion with six degrees of freedom. The model is equipped with all the appendages, and it is characterised by an unusual single rudder/twin screws configuration. This arrangement causes poor directional stability qualities, which makes the prediction of the trajectory a challenging problem. As already shown in previous works, the treatment of the in-plane loads exerted by the propellers is of paramount importance; to this aim each propeller is simulated by an actuator disk model, properly modified to account for oblique flow effects. The main goal of the present paper is to assess the capability of the CFD tool to accurately predict the trajectory of the ship and to analyse the complex flow field around a vessel performing a turning manoeuvre. Distribution of forces and moments on the main hull, stern appendages and rudder are analysed in order to gain a deeper insight into the dynamic behaviour of the vessel. Validation is provided by the comparison with experimental data from free running tests.

The study of a planing flat plate may be considered as a topic of wide interest for academic and industrial applications. From experimental and numerical studies, flow separation occurs near the stagnation point and a thin jet sprays forward along the plate, while a clear wave pattern develops downstream. In the present study, the effect on the jet-root position caused by a cushion pressure applied on the downstream free surface is considered and the consequent variation in lift and drag coefficients is studied. This canonical problem is important in the design of Surface Effects Ships (SES), the bow seal of which may be assumed as a planing deformable surface with a cushion pressure behind it. This study focuses on the hydrodynamic interaction between the plate and the cushion pressure; as such, the plate geometry is prescribed. A two-dimensional numerical study of this problem has been performed in the present work using a finite-volume Chimera-overlapping-grids approach to numerically solve the Navier-Stokes equations; the free surface is handled by means of two-phase level-set method. Validity of the results is assessed by the comparison with theoretical and numerical results available in the literature.

A model problem of the flow under an air-cushion vessel is studied. Two different numerical techniques are used to determine the solution of the free-surface elevation and the wave resistance for a range of Froude number, Reynolds number, value of the pressure applied in the cushion, and depth of the water. The first numerical technique uses a velocity potential that satisfies linearized free-surface boundary conditions, whereas the second employs a finite-volume method to find a solution that satisfies the fully nonlinear free-surface boundary conditions. The results clearly show that for high Froude number and practical values of the cushion pressure, the linear-theory solution is in excellent agreement with the more exact nonlinear prediction. For lower Froude number the solution becomes unsteady, and the disagreement between the two methods is larger.

Propeller modelling in CFD simulations is a key issue for the correct prediction of hull-propeller interactions, manoeuvring characteristics and the flow field in the stern region of a marine vehicle. From this point of view, actuator disk approaches have proved their reliability and computational efficiency; for these reasons, they are commonly used for the analysis of propulsive performance of a ship. Nevertheless, these models often neglect peculiar physical phenomena which characterise the operating propeller in off-design condition, namely the in-plane loads that are of paramount importance when considering non-standard or unusual propeller/rudder arrangements. In order to emphasize the importance of these components (in particular the propeller lateral force) and the need of a detailed propeller model for the correct prediction of the manoeuvring qualities of a ship, the turning circle manoeuvre of a self-propelled fully appended twin screw tanker-like ship model with a single rudder is simulated by the unsteady RANS solver ?navis developed at CNR-INSEAN; several propeller models able to include the effect of the strong oblique flow component encountered during a manoeuvre have been considered and compared. It is emphasized that, despite these models account for very complex and fundamental physical effects, which would be lost by a traditional actuator disk approach, the increase in computational resources is almost negligible. The accuracy of these models is assessed by comparison with experimental data from free running tests. The main features of the flow field, with particular attention to the vortical structures detached from the hull are presented as well.

The turning circle maneuver of a self-propelled tanker like ship model is numerically simulated through the integration of the unsteady Reynolds averaged Navier-Stokes (uRaNS) equations coupled with the equations of the motion of a rigid body. The solution is achieved by means of the unsteady RANS solver ?navis developed at CNR-INSEAN. The focus here is on the analysis of the maneuvering behavior of the ship with two different stern appendages configurations; namely, a twin screw with a single rudder and a twin screw, twin rudder with a central skeg. Each propeller is taken into account by a model based on the actuator disk concept; anyhow, in order to correctly capture the turning maneuvering behavior of the model, a suitable model which takes into account for oblique flow effects has to be considered. Results from a preliminary verification assessment are discussed; validation of the predicted trajectory and the kinematical parameters is provided by comparison with experimental data from free running tests. Maneuvering abilities of the two configurations are discussed; in order to better understand the different behavior of the two configurations, an in depth analysis of the force and moments on the hull and on the individual appendages is provided.

In applied hydrodynamics it is presently a general common task to simulate flow around complex shaped ships with moving appendages. As an example the simulation of a turning circle manoeuvre of a full-appended combatant ship is common in manoeuvrability studies. Nevertheless the accurate numerical simulation of turbulent, unsteady flow around a full appended maneuvering complex-shaped hull is a challenging task, because of the geometrical complexity of the appendages present and their relative movement, generating a very complex hydrodynamic flow.

The turning circle maneuver of a self-propelled tanker like ship model is numerically simulated through the integration of the unsteady Reynolds averaged Navier-Stokes (uRaNS) equations coupled with the equations of the motion of a rigid body. The solution is achieved by means of the unsteady RANS solver Xnavis developed at CNR-INSEAN. The focus here is on the analysis of the maneuvering behavior of the ship with two different stern appendages configurations; namely, a twin screw with a single rudder and a twin screw, twin rudder with a central skeg. Each propeller is taken into account by a model based on the actuator disk concept; anyhow, in order to correctly capture the turning maneuvering behavior of the model, a suitable model which takes into account for oblique flow effects has to be considered. Results from a preliminary verification assessment are discussed; validation of the predicted trajectory and the kinematical parameters is provided by comparison with experimental data from free running tests. Maneuvering abilities of the two configurations are discussed; in order to better understand the different behavior of the two configurations, an in depth analysis of the force and moments on the hull and on the individual appendages is provided.

A complementary experimental and numerical study of the interference eect for a fast catamaran is presented. Resistance, sinkage and trim are collected by towing tank experiments for Froude number in the range from 0:2 to 0:8 for several separation distances and for the monohull. Resistance coefficient curves reveal the presence of two humps, the second one strongly depending on the separation length; high interference is observed in correspondence of the second hump. To gain a deeper insight into these behaviors, a complementary analysis is carried out by a numerical campaign; simulations are performed by means of an in-house unsteady RANS solver. Verication of numerical results is provided, together with validation, which is made by the comparison with both present and other experimental data. Agreement in terms of resistance coefficient is rather good, comparison error being always smaller than 2.2%.

This paper presents the results of a large experimental and numerical campaign aimed to the analysis of the interference effect for a fast catamaran. Several separation distances are considered; data for resistance, sinkage and trim are collected by towing tank experiments for Froude number ranging from 0.2 to 0.8. Monohull tests are also carried out, the analysis of the interference and its dependency on the separation length being the main objective of the paper. Resistance coefficient curves reveal the presence of two humps, the second one strongly depending on the separation length; high interference is observed in correspondence of the second hump. It is found that the narrower is the configuration, the higher is the interference and the speed at which this maximum occurs. To gain a deeper insight into these behaviors, a complementary analysis, in terms of wave field, surface pressure and velocity field is carried out by an in-house unsteady RANS solver. Verification of numerical results is provided, together with validation, which is made by the comparison with both present and other experimental data. Agreement in terms of resistance coefficient is rather good, comparison error being always smaller than 2.2%.

The simulations of the flow around a high-speed vessel in both catamaran and monohull configurations are carried out by the numerical solution of the Reynold averaged Navier–Stokes (RANS) equations. The goal of the analysis is the investigation of the interference phenomena between the two hulls, with focus on its dependence on the Reynolds number (Re). To this aim, numerical simulations are carried out for values of Re ranging from 106 to 108 for two different values of the Froude number (Fr = 0.30, 0.45). Wave patterns, wave profiles, limiting treamlines, surface pressure and velocity fields are analyzed; comparison is made between the catamaran and the monohull configurations. Dependence of the pressure and viscous resistance coefficients, as well as of the interference factor, on the Reynolds number is investigated. Verification and validation for both resistance coefficients and wave cuts is also performed.

The turning circle manoeuvre of a self-propelled tanker like ship model is numerically simulated through the integration of the unsteady Reynolds Averaged Navier-Stokes (URANS) equations coupled with the equations of the motion of a rigid body. The solution is achieved by means of the unsteady RANS solver developed at CNR-INSEAN. The model is considered with two different stern appendages configurations (each one providing a different dynamic behaviour): twin screw with a single rudder and twin screw, twin rudder with a central skeg. Each propeller is taken into account by a model based on the actuator disk concept; anyhow, in order to correctly capture the turning manoeuvring behaviour of the model, a suitable description of the propeller performance in oblique flow operation has be considered. Comparison with experimental data from free running tests will demonstrate the feasibility of the CFD computations. The main features of the flow field, with particular attention to the vortical structures detached from the hull is presented as well.

The results of accurate compressible Navier-Stokes simulations of aerodynamic heating of the Vega launcher are presented. Three selected steady conditions of the Vega mission profile are considered: the first corresponding to the altitude of 18 km, the second to 25 km and the last to 33 km. The numerical code is based on the mathematical model described by the Favre-Average-Navier-Stokes equations; the turbulent model chosen for closure is the one-equation model by Spalart-Allmaras. The equations are discretized by a finite volume approach, that can handle block-structured meshes with partial overlap ("Chimera" grid-overlapping technique). The isothermal boundary condition has been applied to the lancher wall. Particular care was devoted to the construction of the discrete model; as a matter of facts, the launcher is equipped with many protrusions and geometrical peculiarities (as antennas, raceways, inter-stage connection flanges and retrorockets) that are expected to affect considerably the local thermal flow-field and the level of heat fluxes, because the flow have to undergo strong variation in space; consequently, special attention was devoted to the definition of a tailored mesh, capable of catching local details of the aerothermal flow field (shocks, expansion fans, boundary layer, etc..). The computed results are reported together with uncertainty and actual convergence order, that were estimated by the standard procedures suggested by AIAA.

In this work the numerical simulations of a submarine in straight ahead motion with the appendages at several prescribed deflection angles are performed. Due to the complex geometry involved (the presence of moving appendages), these simulations are rather demanding form the point of view of both grid generation and accuracy of the numerical method. In order to analyze these aspects, the numerical solutions are computed by means of an unsteady Reynolds averaged Navier-Stokes equations solver, which is particularly effective because of the high order discretization schemes adopted. From the point of view of mesh construction, a dynamic overset grid technique is used, where each geometrical element of the whole geometry is discretized with a set of block-structured body-conformal mesh with partial overlapping (Chimera approach). In the present paper, the details of the method and numerical results for several deflection angles of the bow and stern planes are presented.