An underwater robot usually consists of a main body and several control ducted propellers. In numerical simulation on the hydrodynamic behavior of underwater robot at a high speed motion in large computational domain with CFD approaches, some computational failures may occur due to slow computational speed and demand of super number of computational meshes. In this paper, a three-dimensional hydrodynamic model of underwater robot is proposed to investigate the hydrodynamic performance of the robot attached with ducted propellers during high-speed motion. Computational domains and geometrical model of the robot for numerical simulations are first constructed. Motion mechanism of the robot in different kinds of motion conditions are then studied. In this research, different motion styles of the robot under driven thrusts from the ducted propellers are investigated based on the proposed hydrodynamic mode. The relationship between thrusts and rotating speeds of the propellers are analyzed. The hydrodynamic performances when the propellers rotate at different positions and different rotating speeds are observed. Overlapping mesh method is utilized in CFD computations on the hydrodynamic forces of the main body of the robot and the propellers. Since great number of computational meshes and high speed computation are needed, in order to facilitate the computational process, super-computing means dealing with the excessively amount of computing power is applied in the research so that small time step sizes with high computational speeds can be achieved. The hydrodynamic performances of an underwater robot under controlled forces from the ducted propellers with a super-computing means and the numerical techniques to solve the problems arising from the simulation of the robot with high speed motion are also presented. INTRODUCTION The ocean covers 71% of the surface of the earth. The vast ocean not only contains extremely rich mineral and biological resources, but also many potentially exploitable wave and tidal energy and traces of geological movements. Therefore, exploring and developing the ocean can not only generate great economic value, but also have high scientific research value. Exploring and developing the ocean requires professional scientific equipment and devices. Underwater robot is an indispensable device for humans to explore and develop the ocean.

ABSTRACT In this paper, a systematic method to design marine ducted propellers is presented for a given duct shape and design conditions to achieve the highest efficiency while avoiding cavitation. The duct is solved by the 2D-axisymmetric Reynolds-Averaged Navier-Stokes (RANS) method and the propeller is designed and analyzed by potential flow methods (lifting line method and vortex lattice method). The initial propeller is designed based on the optimal loading from the RANS/lifting line model interaction method. A nonlinear numerical optimization method is used to refine the propeller in the effective wake (defined as the difference between the total velocity and the propeller induced velocity), with constraints to avoid cavitation for a given cavitation number. Each design is analyzed in a previously-developed and validated RANS/vortex lattice method interaction model to check if the optimization objective is achieved. INTRODUCTION In the past, different methods have been developed to analyze the complex flow interaction between the propeller and the duct. Kerwin et al. (1987) applied the vortex lattice method (VLM) on the propeller and the boundary element method (BEM, more commonly known as the panel method) on the duct. Lee and Kinnas (2006) applied the panel method on both propeller and the duct, which was improved by Kinnas et al. (2015) and Kim et al. (2018) by including the duct-induced velocities in a pseudounsteady wake alignment scheme (full wake alignment scheme, or FWA). Du and Kinnas (2019) applied a flow separation model with the panel method on a ducted propeller with a blunt trailing edge duct. Hoekstra (2006) and Majdfar et al. (2017) conducted the full-blown Reynolds-Averaged Navier-Stokes (RANS) simulations on ducted propellers. Tian et al. (2014) and Bosschers et al. (2015) used a hybrid RANS/potential flow method to calculate the open water characteristics of ducted propellers, in which the propellers are represented as distributed body forces in the RANS simulation and the strength of the body force is either from the vortex lattice method or from the panel method. In Tian et al. (2014), the effect of the duct on the propeller is considered through an effective inflow, namely the effective wake, which is defined as the difference between the total velocity and the propeller-induced velocity. Open water tests of a four-blade propeller with a square tip operating with a duct with sharp trailing edge (propeller Ka4–70 and Duct 19Am, as shown in Fig. 1) were carried out by Bosschers and van der Veeken (2008) to provide valuable experimental data to validate several numerical models mentioned above.

Abstract A numerical method to observe thrust behavior of a ducted propeller attached in an underwater vehicle under the influence of flow field of the vehicle's main body is proposed, the thrust characteristics of ducted propeller and the velocity distribution around the propeller when the vehicle in turning motion is studied numerically, the hydrodynamic relationship between the thrust and the velocity components around the propeller is investigated. In the research, 3D geometric models of the duct, propeller and main body of the vehicle are first constructed according to their geometrical feature. Computational fluid dynamics method based on the finite volume method and sliding mesh technique are applied to solve the Navier-Stokes equations which govern the fluid motion around the duct, propeller and underwater vehicle when they are in a turning motion. These equations are solved numerically with the CFD code FLUENT. With the proposed numerical simulation approach, the characteristics of thrust of the ducted propeller and the velocity distribution around the propeller under different working conditions are analyzed, the thrust issued from the ducted propellers and the hydrodynamic relationship between thrust and the velocity components under the influence of vehicle's flow field are observed. Results of the numerical simulation indicate that influence of fluid field caused by the underwater vehicle on the thrust of the ducted propeller is not negligible; there are strong relationships between thrust issued from the propeller and axial induced velocity on the propeller disk, and between the hydrodynamic torque on the propeller and the induced circumferential velocity. It is believed that analyzing correctly the hydrodynamic relationships between thrust and velocity components around ducted propeller in an underwater vehicle is a key to understand more precisely the thrust mechanism of the propeller. Introduction Ducted propeller is one of the major active control devices for manipulation of an underwater vehicle, the trajectory and attitude of an underwater vehicle is usually controlled by user on water surface through an umbilical cable sending operational signals to the ducted propellers attached the vehicle (Avila and Adamowski, 2011; Jaulin, 2009; Li et al, 2005). To understand correctly the hydrodynamic relationship between thrust issued from the propeller and velocity components around it under the influence of the vehicle's flow field is of great importance in designing effective trajectory and attitude control devices for the underwater vehicle. The most common method nowadays to analyze the hydrodynamic performance of a propeller installed in an underwater vehicle is to establish a fitting relationship between the propeller coefficient, torque coefficient and advance coefficient of the propeller by means of open-water propeller test with linear function method or least square method (Bachmayer et al, 2000; Fossen and Blanke, 2000). In these methods, rate of advance on the propeller disk is usually determined based on the vehicle's motion due to the difficulty of evaluating its value on the propeller disk. Obviously the influence of the wake caused by vehicle motion on the thrust evaluation of the propeller is not considered on this computation method although this influence actually exists (Kim and Chung, 2006). Therefore, how to forecast accurately the thrust behaviors of a ducted propeller under the influence of underwater vehicle flow field becomes a basic element for understanding comprehensively the hydrodynamic and control nature of the vehicle when it is driven by the control force of the propeller.

Abstract QCM has been applied to open-water thrust and torque predictions of the ducted propellers with different positions of the propeller and duct. The QCM results are in good agreement with the experimental data of the propellers positioned inside a duct. The results of propellers in the downstream of the duct show no significant dependence on the length of the straight trailing vortex of the duct. Comparison with the results calculated without any duct has indicated no improvement of the calculated open-water characteristics by positioning the present duct in front of the present propeller.

ABSTRACT: A formerly developed vortex lattice method (VLM) is coupled with Renolds Averaged Navier Stokes based Computational Fluid Dynamics (CFD) tools, in order to predict the hydro-dynamic performance of ducted propulsion systems. The analysis of the performance for ducted propellers subject to steady and unsteady cavitating inflow is via applying a hybrid numerical method which couples a vortex lattice method (MPUF-3A) with RANS solvers for analyzing the viscous flow around the propulsor and the drag force on the hub and duct surfaces. The presence of the propeller is represented through distributed pressure gradients, or in other words, body forces. The body forces can be obtained from the potential flow method, and re-associated to the fluid domain revolved by the propeller blades. The propeller force distributions are considered as source terms (body forces) in the momentum equations of RANS solver. The effects of viscosity on the effective wake and on the performance of the propeller blade, as well as on the predicted hub and duct forces, can be assessed. INTRODUCTION The demand of high-speed marine vehicles for commercial has increased drastically in recent decades. Therefore, marine propulsors are designed with more complex geometries to satisfy various requirements, such as high efficiency, less noise and vibration, better course stability, lower vessel resistance and economic operation, etc. Following this trend, ducted propellers have become more and more popular in the contemporary design. The complexity of the geometric characteristics of the propulsion systems introduces complicated flow fields around the propulsors and thus creates a lot of challenges to the design and computational analysis. Using numerical methods to solve ducted propulsion systems is much more challenging than those of open propellers because of the complicate geometric configurations and the interactions among the components. The primary problems and challenges are caused by hydrodynamic cavitation and viscosity.

ABSTRACT This paper presents a method for estimating the propulsion performance and efficiency of an AUV thruster based on CFD (Computer Fluid Dynamics) analysis. A discussion of the factors influencing the numerical simulation accuracy and convergence rate, namely the number, size, and type of mesh elements required to describe each part of the model, is presented. The axial symmetry characteristics of the propeller, nozzle and the encompassing MRF (Moving Reference Frame) region have been used to reduce complexity and hence computation time. The paper, then, describes a method for estimating the motor horsepower and RPM of an AUV thruster based on simulated resistance tests and propeller open water (POW) tests. The engine horsepower for an AUV equipped with a ducted propeller and RPM is evaluated from the speed-power curve. The effectiveness of the ducted propeller is studied by comparing the performance characteristics for test results obtained without the nozzle versus those obtained for a range of nozzle shapes. The best design of the nozzle shape is finally determined from speed-power curves of the given duct. INTRODUCTION An understanding of the fluid interaction between the hull and propeller is necessary to estimate propulsion performance and efficiency of an Autonomous Underwater Vehicle (AUV) thruster. Resistance and POW (Propeller Open Water) tests are essential preliminary steps that are required to determine hull and propeller efficiency. Self-propulsion testing can then be carried out to estimate the required power as well as obtain the relative rotative efficiency (ηR), the RPM, and the propulsion factors such as thrust deduction factor (t) and wake fraction factor (w). In order to increase efficiency the propeller thrust coefficient (CTh) must be reduced. The ducted propeller concept was first introduced by Kort (1934), and has been widely researched during the 1950s and 1960s (Manen and Oosterveld, 1966; Wessinger and Maass, 1968).

ABSTRACT The ducted propeller is widely used as a propulsor in submarines, torpedos, UUVs (unmanned underwater vehicles), underwater robots, and DPDs (diver propulsion devices) because of its high propulsive efficiency. Its merits also include collision protection, efficiency enhancement, cavitation reduction, etc. However, full 3-D flow analysis is hard because of complexity of geometry and interactions between the propeller and the duct. The present work aims to design the propursive system and analyze the turbulent flow field of a ducted marine propulsor using Navier-Stokes equations with the MRF (Multiple Reference Frames) technique at Fluent code. INTRODUCTION Underwater robot is gaining importance for the role in building marine structures, researching deep sea ecology, and developing marine energy. For stable operation of underwater robot, efficient propulsion systems are required. The ducted propeller is widely used as a propulsor in submarines, torpedos, UUVs (unmanned underwater vehicles), underwater robots, and DPDs (diver propulsion devices) because of its high propulsive efficiency. Its merits also include collision protection, efficiency enhancement, cavitation reduction, etc. However, the flow of ducted propeller is very complex because it involves strong flow interaction between the blade impeller and duct. Kerwin et al. (1987) analyzed the flow of ducted propulsors with the panel method. Jung et al. (2001, 2002), Jang et al. (2004a) and Park et al. (2005a) applied the sliding multi-block method analyze the ducted propulsor. Jang et al. (2004a) and Park et al. (2005b) also analyzed the flow of a waterjet which has roter-stater interaction flow. The present work aims to design the propursive system and analyze the turbulent flow field of a ducted marine propulsor using Navier-Stokes equations with the MRF (Multiple Reference Frames) technique of a commercial Fluent code. Three-blades Propeller The 3-blades propellers as shown in Figs. 1~2 are designed based on 2-dimensional strip theory.

ABSTRACT A lifting surface panel model of a prototype tip-driven propeller has been developed. The model includes the propeller, surrounding duct, stators and bearing casings. It was used to predict the likely performance of the prototype thruster under various advance speeds, propeller speeds and duct profiles, and the results have been compared with the experimental data. The paper presents the development of the computational model comparison of predictions with experimental results, and the future use of the model to optimise the hydrodynamic characteristics of the propeller and duct. INTRODUCTION Ducted propellers have been used for many years in the marine industry to both protect the propeller from damage, and to improve upon the propeller efficiency under certain load conditions, Consequently a wide range of tests have been carried out on the units and much performancee data has been published. One reason for losses in a dueted propeller system is the small clearance gap between the blade tips and the inner surface of the duet. In this region tip vortices occur and subsequently reduce propeller performance. In an attempt to eliminate tip vortices, the use of ring-propellers was studied (Oosterveld, 1972) whereby the propeller blades were attached to a profiled ring which rotated with the propeller. This did overcome the clearance gap problems, however, any gains were offset by the increased frictional resistance of the rotating ring. Further developments of the ring-propeller led to the ducted ringpropeller in which the propeller was attached to a thin ring, which sat flush within an external duct. This did offer an improvament in efficiency over the open ring-propeller, but not to the extent of matching that of a standard ductad propeller, which still remained the most efficient.

ABSTRACT: This paper presents the key results of a comprehensive study into the nature and magnitude of the mechanisms which contribute to the loads on both open and ducted propellers in the presence of ice. Both model test and full scale trials measurements and visual observations were employed. The nature of the principal components of ice loading are identified and their parametric dependencies are presented. Differences between the interaction of open and ducted propellers with ice are explained and the load limiting effect claimed for the ducted propeller is addressed. The model test results are correlated with equivalent full scale events with remarkably good agreement. The study represents a major advance towards the prediction of design loads for Arctic marine propulsion systems. INTRODUCTION When a vessel operates in an ice regime, especially in thick ice broken ice travels along the hull underneath the vessel and is subsequently milled by the propeller. This results in significantly higher forces and moments than are generated in an open water environment and the propeller and propulsion train must be designed for these conditions. The traditional approach to design of ice class propellers and propulsion systems has been to increase the open water design loads using an ice torque factor (Browne and Van der Pas, 1989). This is a function of the ice class of the vessel, which is synonymous with the severity of ice conditions expected in service. The ice torque factor is determined from measurements of shaft loading and propulsion system service experience from earlier vessels. However, it suffers similarly to other purely empirical approaches in that it is no better than the limited data on which it is based and cannot be extrapolated with confidence to cover new propulsion systems and significantly different operating conditions.