The ‘Zero-G’ is designated as a new class of underwater robot that is capable of unrestricted attitude control. A novel control scheme based on internal actuation using Control Moment Gyros (CMGs) is proposed to provide Zero-G Class Autonomous Underwater Vehicles (AUVs) with this unique freedom in control. The equations of motion for a CMG actuated underwater robot are derived and a nonlinear feedback control law formulated based on energy considerations of the system’s coupled dynamics. Singularities, redundancy and null motion are discussed in the context of CMGs and a mathematical escapability condition is developed based on the differential geometry of null motion. A comprehensive geometric study of the singularities of a CMG pyramid is performed and together with considerations of the inverse kinematics of attitude control form the basis of a global steering law that exactly achieves the desired torques, whilst guaranteeing real-time singularity avoidance within a constrained workspace. The development of the CMG actuated Zero-G Class underwater robot IKURA is described. This is the first Zero-G Class prototype and is the first application of CMGs to underwater robots. A series of experiments to demonstrate the practical application of CMGs and verify the associated theoretical developments is described. The open-loop dynamics of the system and the exactness and real-time applicability of the CMG steering law are verified. Experiments are carried out to assess the performance of the proposed control law by comparing the response of the robot to that using alternative control laws that neglect the hydrodynamic interactions of the body and the coupled motion of the CMGs and body respectively. The control law demonstrates a faster response with a smaller overshoot for less overall control activity than the alternative methods. The ability to actively stabilise the passively unstable translational dynamics of the robot are verified. Next, the unrestricted attitude control capability is confirmed with the robot demonstrating the necessary range of attitude control to adopt and maintain any attitude on the surface of a sphere. Finally, the ability to stabilise any attitude while translating in surge is confirmed with the robot performing vertically pitched diving and surfacing in surge. This is the first time an underwater robot has performed such a manoeuvre. This research demonstrates that CMGs are capable of actively stabilising the passively unstable dynamics of an underwater robot with essentially zero-righting moment and are capable of providing it with unrestricted attitude control. The three-dimensional manoeuvring capabilities allow Zero-G Class underwater robots to plan and optimise their missions in a fully threedimensional manner, in a way that has not been possible previously. This study concludes that the application of CMGs for attitude control opens up a path to develop sophisticated Zero-G Class underwater robots and their application to new fields of underwater research.

The ability to predict the powering performance of twin skeg LNG ship is a complex endeavour requiring appraisal of operating conditions and hydrodynamic analysis to arrive at a suitable stern design solution. Inherently coupled with the stern design process is the design optimization, namely the selection of most suitable geometrical parameters of the propulsor, control surface and their arrangements with respect to the hull. An approach to the stern design may commence with the prediction of general ship stern flow, hence its resistance and self-propulsion capabilities. Almost a century of experience exists regarding how to predict the resistance and powering capabilities of the twin skeg LNG ship. Despite this, improvement in numerical methods is still in high demand. A RANS based numerical approach is presented in this thesis to predict the resistance and powering performance of future twin skeg ships. This is supported by a meshing approach which easily blends the hull-skeg boundary layer to the free surface. Predicting the non-uniform wake in the propeller plane due to the hull-skeg and control surface interaction was identified as one of the main challenges in the stern design and powering assessment. To predict this within acceptable cost a sectorial approach was developed as part of the numerical method which discretizes the propeller plane into a series of radial and circumferential subdivisions. The local axial and tangential inflow conditions at each location can then be considered. This was coupled to a blade element momentum theory propeller code. The two-way coupling was found to be a computationally efficient tool for studying the powering performance of ships. To demonstrate the pertinence of the RANS based numerical approaches developed in this work a series of case studies has been analysed. These include: skeg-rudder-propeller interaction studies, propulsive characteristic of the KCS ship, and the resistance and self-propulsion characteristics of a future twin skeg LNG ship. These highlight the roles of the numerical approaches in the stern design process for future twin skeg ships. The techniques developed in this work enable the designer to predict the powering performance of future twin skeg LNG ships at a cost effective manner in the initial design stage.

A theoretical method has been developed to predict the forces developed due to the interaction between a ship rudder and propeller. A parallel lifting suface panel program (PALISUPAN) ha sbeen written in Occam2 which is designed to run across variable sized square arrays of transputers. thsi program forms teh basis of the theoretical method. The rudder and propeller are modelled separately. Their interaction is accounted for through an iterative process whereby their respective inflow velocity fields are modified using a circumferential average of the disturbance velcoity due to the other body. Prior to writing PALISUPAN, software techniques for the implementation of computational fluid dynamics algorithms across arrays of transputers were developed. The approach used is based on a geometric parallelism. At the outermost level on each transputer the particular CFD algorithm runs in parallel with a harness process. The harness controls teh communication across teh transputer array. to prove thsi concept an explicit finite volume solver for the two-dimensional Euler equations has been implemented. PALISUPAN itself uses a perturbation potential formulation and an explicit zero pressure loading condition is enforced at the trailing edge. Use of the communications harness greatky reduces code development time and although an implicit solver PALISUPAN gives good parallel performance. Wind tunnel tests were undertaken to derive experimental data for validation of the prediction method. These used a 3.5m x 2.5m low speed widn tunnel and a range of flow an dgeometrical parameters were tested. Total rudder forces and moments, propeller thrust and torque and quasi-steady rudder surface pressures were measured. Empirical relationships for teh prediction of rudder lift, drag and stall for use in ship manoeuvring studeis were also derived. The validated theoretical prediction for rudder-propeller interaction using PALISUPAN allows the detailed design of sjip rudder-propeller systems to be enhanced. The parallel performance of the pALISUPAN demonstrates the practicality of using transputer arrays to solve CFD problems.

A concept for a non-intrusive sloshing guidance system based on a phenomenological Rapid Sloshing Model is proposed to reduce the operational risk of sloshing damage to LNG carriers. A numerical sloshing model is implemented in a commercial Navier-Stokes Computational Fluid Dynamics (CFD) code which uses a volume-of- fluid approach for the simulation of multi- fuid problems. The effect of spatial and temporal discretisation and turbulence is investigated using systematic variation. Dimensional analysis of the multiphase flow regime and examination of the relative velocity at the fluid interface show that an inhomogeneous multiphase model is appropriate for the simulation of a violent sloshing flow. This is conrmed by the good agreement with the experimental data of Hinatsu. The effect of fluid compressibility is investigated for sloshing impacts and a criterion based on wave propagation is developed to assess the importance of compressibility. When modelling sloshing with large air bubble entrainment, the choice of fluid compressibility model is shown to have a significant influence on pressure magnitude and frequency of oscillation required for structural assessment and a thermal energy model is required. The Rapid Sloshing Model (RSM) is based on the observation that the centre of mass of a sloshing fluid tends to follow a particular trajectory. Using a phenomenological modelling approach, the forces affecting the sloshing response are approximated with mathematical functions for restoring force, damping and sloshing impacts. Calculation times for the resulting equations are typically 0.1% of real time on a desktop PC. A case study of sloshing induced by periodic rotation and translation of two-dimensional longitudinal and transverse sections of membrane LNG tanks is carried out using RSM. RSM is set up using one CFD simulation not considered in the case study and the RSM solutions are then compared to the independent CFD solutions. The fluid momentum from RSM is usually within 5%{15% of the CFD solution for excitation at and near the first resonant period at a filling level near the critical depth. An irregular surge motion profile from an ITTC two-parameter spectrum is applied to the tank and the mean error from the RSM solution remains below 15% when using momentum and transverse force. When applied to sloshing with a 10% filling level excited by an irregular seaway a mean error of 9.6% is obtained. Compared to existing phenomenological modelling approaches the RSM methodology reduces the error by an order of magnitude in sloshing scenarios of practical interest. A non-intrusive sloshing guidance system based on the Rapid Sloshing Model which is suitable for installation on existing and newbuild LNG carriers can be implemented by applying motion data measured onboard to the RSM to provide operator guidance on the sloshing severity in partially filled LNG tanks. The RSM is set up for a particular LNG carrier with existing sloshing data from the design and class approval stages.

This thesis deals with the assessment of the strength and durability of steel to composite joints for composite superstructures on ships where reduced weight is a design driver. The purpose of the work is to understand the long-term performance characteristics of hybrid connections to allow for improvements to the design of hybrid structures. Two joints were investigated in the present research. The first was a full-scale connection suitable for application in superstructures of marine vehicles, specifically a helicopter hanger on a naval vessel. The second was a generic steel/composite connection for testing performance after hygrothermal ageing. The strength and durability of the full-scale connection were examined in compression, the loading scenario representative of in-service conditions. The results indicated that the static and fatigue performance were in excess of the realistic in-service loading conditions. Failure for both static and fatigue tests were comparable and therefore good confidence in the prediction of the joint’s failure was achieved. The generic hybrid connection was artificially aged through immersion in water. The results indicated that there was no significant reduction in the performance of the joint in either static tension or bending. The numerical modelling highlighted a number of issues. Due to the geometry of the joint high stress concentration factors were observed in some locations. It is in these areas that failure of the joint was predicted in the numerical modelling. Similar results were obtained experimentally and this gave confidence in the modelling of the joint. Numerical parametric and optimisation studies were conducted to assess the influence of the joint geometry on performance characteristics obtained from both the experimental and numerical studies. This highlighted that improvements to the performance of the joint could be obtained through geometric changes alone.

With the aim of understanding and improving the sea keeping performance of marine craft the research addresses methods of influencing the motions of marine craft by stabilisation mechanisms, and identifying the influences on RIB motions. The two research projects, that attracted industrial support, represent two separate industrial interests concerned with influencing and understanding the influences on the sea keeping performance of marine craft. The first project, focused on developing a motion control system, for motion reduction and/or control of marine structures operating in the free-surface. The aim was to develop a system that would replace, eliminate or complement the use of current devices, in order to improve passenger comfort, overcoming the drawbacks of external hydrodynamic based systems. The research developed a new motion control system and theoretical and physical experimental studies were conducted. The second project, focused on identifying the influence of a RIB tube and the properties of the tube, on the dynamic motions of a RIB, including the effect of the motions on the occupants. Theoretical and experimental studies at full and model scale were conducted.

In general the use of composites results in shorter production time, lightweight and lower maintenance costs to the marine industry in the leisure, fast and fishing boats sectors. The social and economic benefits of using composite materials have made users complacent about the pollution and the health and safety issues associated with these materials. As the perception of environmental problems changes with time, alternatives with lower emissions allowing for cleaner production and easier disposal must be investigated. Glass Reinforced Thermoplastics (GRTP) have been in use for many years in the automotive industry and aerospace. These materials are fast to process, solvent free, have an unlimited pot life and demonstrate better mechanical properties such as improved toughness compared to aluminium and Glass Reinforced Thermoset (GRTS). However, building boats with GRTP requires massive investment in equipment that ship builders do not currently undertake, such as curing ovens, autoclaves and plastic welding equipment. It is, thus, necessary to define a method to measure the environmental performance of this material in the context of marine structure. The present research presents a comparative study of four materials, namely steel, aluminium, GRTS and GRTP, in the above context. The outcome of the research defines a material selection framework for marine structures focusing primarily on environmental performance. The study focused on life cycle energy and material flows to represent environmental impact over the entire life of a boat and the methodology used respects Life Cycle Assessment (LCA) standards. The influence of the conventional marine structure design approach on LCA results was highlighted by the result of a grillage and a boat design study. These two studies also showed that the contribution of in-service fuel consumption to the life cycle energy has the most significant environmental impact. This impact is two to three orders of magnitude larger than the manufacturing environmental impact of the candidate materials. A boat study taking into account the results of the two above mentioned studies overcame this limitation. This boat study, referred as a boat synthesis, uses a constant fuel consumption as a design constraint for each material. It demonstrated that in some part of the studied design space, GRTP could offer the best material alternative, whereas in some other part, aluminium is the best alternative. In addition, the study also showed that steel could also be the least environmentally damaging material under some conditions, which goes against the common practice to build all small boats in GRTS

The importance of human factors has to be taken into account when determining a yacht’s performance over a course. The crew’s capabilities of technical skills, athletic performance, and his/her ability of making rational decisions under time pressure and in light of uncertainty of the future wind regime are important aspects that will determine the overall performance of a yacht-crew system. This thesis highlights the performance of such a yacht-crew system with a focus on the decision-making process of sailors. Aspects of human behaviour in sport and the decision-making process are explained considering the level of expertise and possible approaches of how to model them are shown. An artificial intelligence AI -system is developed that is capable of simulating the decision-making process of different sailing behaviours/styles as well as different expertise levels of sailors within a dynamically changing yacht racing environment. The constraints of the multiple fleet racing simulator Robo-Race (Scarponi 2008) were determined using a series of tests with real sailors identified three important constrains: (1) the predictable behaviour of the AI-yachts, (2) the predictable and unrealistic weather model and (3) the simple model describing the effects of yacht interaction. These restrictions and constraints that limited the real and AI-sailors natural sailing behaviour have been successfully removed in the updated version of Robo-Race. The new developed decision-making engine based on Decision Field Theory that uses Bayesian Belief Networks as the perceptual processor showed a clear superiority over the old rule-based decision-making engine. Extensive simulations demonstrate the feasibility of modelling various decision-making processes and therefore different behaviours and expertise levels of sailors. A good comparison was found with that obtained between the Robo-Race results and the Olympic fleet racing events.

No description available.

It is well known that ageing steel marine structures are susceptible to corrosion in its all manifestations. The most critical areas are cargo and ballast tanks of merchant ships. However, due to the regulations such as the Performance Standard for Protective Coatings, which requires a 15-year target life of coating in ballast tanks, plus the cathodic protection systems, the internal structures within cargo holds have become more problematic but poorly studied. In the underdeck area and bottom plating, the structures are not normally fully protected. Also, the complex structural arrangement may place difficulties in inspection and repair. In extreme cases, it has been reported that the corrosion rate in these areas could be 5 to 7 times higher than a normal value, and has led to catastrophic structural failures. Currently, the classification societies apply both visual and gauging methods for corrosion inspection during ship surveys. However, it is time consuming especially for large vessels and is highly experience dependent. Therefore, to improve the survey efficiency, facilitate economical maintenance decisions, and even extend the structural life, it is essential to investigate the ultimate strength of such aged and corroded steel structures. Based on the identification of existing corrosion issues in cargo tanks of oil tankers and bulk carriers and the state-of-art of corroded marine structural strength assessments, a nonlinear finite element method was adopted to investigate the influences of pitting and grooving corrosion on the structural integrity. Two full-field experimental techniques were used for a complete validation of the numerical models. Based on the repair conditions provided by classification societies, the numerical results showed that 25% locally corroded area of a plate (800 mm × 800 mm × 15 mm) with 3.75 mm remaining thickness may reduce the ultimate strength by up to 20% compared to a uniformly corroded plate. The weld-induced grooving corrosion of a width of 59 mm and a remaining thickness of 3.75 mm would cause up to 26% strength capacity reduction for a stiffened plate (4750 mm × 950 mm × 15 mm). Moreover, it was shown that the corrosion depth had a greater influence on structural performance compared to corrosion area for the same volume/material loss. By combining mechano-electrochemical protocols with the stress and strain results obtained from the modelling, it enables predictions of the ‘hot spot’ locations of mechanically-induced corrosion acceleration. Results showed that the anodic current density inside grooving corrosion damage (24 mm in width and 3.75 mm in depth) was 7 times greater vis-a-vis the unstressed condition for the stiffened plate at its ultimate strength state. The results, which are closely related to the industrial corrosion inspection and repair requirement, will not only benefit the shipping industry, but are also applicable to a whole range of marine structures (offshore platforms and steel bridges).