No description available.

The research summarised in this thesis addresses the problem of determining the hydrodynamic properties of damaged ships subjected to forced oscillations in calm water. Traditionally forces of hydrodynamic reaction acting on a rigid body moving through a fluid are derived either analytically or numerically. The former approach is usually restricted to small amplitude motions of the body moving through an unbounded domain of ideal fluid. The methodology is relatively simple and computationally effective but, as experimental results suggest, accuracy of the prediction, particularly for roll motion is unsatisfactory even for intact ships. The advanced CFD-based techniques are more suitable in addressing this problem, particularly the case of a damaged ship, but they are computationally demanding. Therefore, in order to tackle the issue efficiently, there is a need for high-quality experimental data for validation of the numerical results. However, the experiments, particularly in roll , are very difficult and there is very little data available for the simpler case of intact ships and virtually none for damaged ships. As the problem involves complex nonlinear phenomena, the physical tests should be performed in a controllable environment and therefore, the 'classical' sea-keeping tests have very limited applicability in this respect. Furthermore, the calm-water experiments are usually performed with oscillations about a fixed axis and the adequacy of such an approach for investigating hydrodynamic properties of damaged ships can be questioned. That is, the physical tests on partially restricted models are of great value, particularly for validating analytical / numerical approaches, but the presence of constraints may introduce artificial conditions affecting the dynamical characteristics of the system. Accounting for this, the approach adopted in this thesis involves a freely-floating body subjected to harmonic excitations generated by an internal forcing mechanism. It is postulated that by removing all kinematic constraints the system can be analysed in the most realistic (achievable in calm water) and controllable configuration. Although use of gyroscopic moment generators for forced roll experiments is not a novelty, this methodology has never been fully exploited for measurements of hydrodynamic reaction forces acting on an unconstrained model of a damaged ship. As the experiments were unprecedented, they resulted in a modest amount of collected data but provided great opportunity for examining the nature and scale of the underlying phenomena. Furthermore, in the course of the research the methodology has been refined and has eventually reached the point at which it can be utilised to produce large amount of experimental data in an accurate and efficient way. From this perspective, the research is prenormative.

The introduction of the probabilistic framework for damage stability, namely SOLAS 2009, introduced in 1st of January 2009 was a step change in regulation history and subsequently in Naval Architecture as a whole industry. The industry, as well as the academia though, was caught unprepared for such a ground breaking change, leading to confusion and retrospectively wrong decisions and misconceptions. It has been proven that survivability is not predicted accurately by the SOLAS s-factor and that harmonisation could not actually work. Furthermore it is believed, and thus still dealt with by deterministic means, that the water on deck problem has not been resolved. To this end, the focus in this study is on passenger vessels. An attempt has been made to highlight not only the shortcomings of the probabilistic framework but also the sheer benefits from utilising such a method. SOLAS 2009 has been benchmarked against conventional methods so as to prove the later. More specifically it has been shown how the survivability of a given damage scenario can be accurately predicted and how time to capsize can actually be related to survivability in waves. In order to achieve this, established concepts like the critical significant wave height have been revised. Moreover, the proposed modifications of the s-factor can link it directly to performance-based standards such as Safe-Return-to-Port. Finally, it has be demonstrated how, when freed of its "childhood" illnesses, the probabilistic framework can provide with an irreplaceable, handy tool with which to make goal-based design and really exploit all the hidden potential leading to safer and more functional ships.

In modern marine structural design, the fatigue life and fracture prediction of local connection details of the vessel is necessary. The traditional empirical rules or numerical work has considerably advanced the qualitative and quantitative understanding of fatigue and fracture analysis. Compared with the existing methods, this thesis explores a novel geometric methodology to evaluate the stress intensity and stress concentration factors (SIF and SCF). The background and special theory was developed to give: 1. A better understanding of the singularities that commonly occur in sharp corners in ship connection details; 2. A quicker method for fatigue life estimation than present methods based on finite element analysis and/or detail classes. 3. A prediction of the stress fields so that more appropriate and reliable finite element meshes can be selected When dealing with the influence of each connection detail, a "Length Scale" estimated from the dimensions of the connection detail. This Length Scale can be converted into a Hot Spot Stress Concentration Factor for SN based fatigue calculations or used with (often simply added to) the real crack length to determine, in conjunction with a constant Y value (commonly 1.1) a Stress Intensity Factor for linear elastic fracture mechanics crack growth calculations. The method is useful both for assessment of existing structures and for design application. The thesis includes a comparison of the results from the application of this new methodology and existing fatigue analysis guidance. Within the thesis the methodology is described together with relevant conclusions.

A finite element model that is capable of simulating the thermo-mechanical welding process was developed by using full thermal-elasto-plastic computational analysis and validated by comparison with experimental data. It shows that distortions predicted by the finite element model agree well with measured data from previous literature and that the numerically obtained residual stress distribution is compared and agreed by both ANSYS and VrWeld software. After that, a simple method for predicting butt-welding residual stresses based on force and moment equilibrium was derived in this section. The results calculated from this simple method were a good match with the FE results. Then the author performed detailed analysis for the distribution of transverse and longitudinal residual stresses of 2D butt welding process by using 3D elements, which illustrated how the butt-welding residual stresses were distributed and accumulated during the welding process and how the boundary conditions affect the final results. A detailed parametric study for butt welding residual stresses based on 2D butt-welding by using 3D element was demonstrated. The factors carried out in the parametric study involved cut-off temperature effect, welding power effect, welding velocity effect, plate length effect and plate width effect. Lastly, the author also presented a simulation and an optimization of welding sequences for residual stress and distortion of a typical, fatigue sensitive, ship's side shell connection detail under different welding sequences.

This thesis models a Steel Catenary Riser (SCR) on soft clay subjected to randon waves. SCR-seabead interaftion is modelled using a hysteretic non-linear model in the vertical seabed direction and Coulomb friction model in the lateral direction together with an improved model that includes the breakoutsoil resistance. The influence of the uncertainty in the geotechnical parameters and the development of the trench in the seabead on the dynamic response is determind. The structural behaviour and a probabilistic approach to fatigue performance in the touchdown zone is developed. The probabilistic model can be employed as a risk analysis tool for the SCR assessment, leading to a safer and more cost0effective design of the SCR system. It is noted that confidence in seabed interaction modelling and geotechnical parameter values is needed in order to have confidence in the final numerical analysis results.

In this thesis, an approach into the applicability of Large Eddy Simulation (LES) and Detached Eddy Simulation (DES) for the prediction of forces and flow around marine bodies at high Reynolds number is presented. The main aim of the thesis was to investigate and assess the use of LES and DES based computational fluid dynamics applications as a tool for hydrodynamic prediction. A methodology for addressing the problem was followed based on mathematical models and numerical solutions translated into a three-dimensional commercial Computational Fluid Dynamics (CFD) code. Meshing strategy involved the use of adapted unstructured meshes with local refinement around key areas of the bodies. The CFD models based on LES and DES are validated against experimental measurements for three different bodies: a NACA 0012 section trefoil, a 6:1 prolate spheroid at angles of incidence and an assymetric America's Cup yacht keel. Quantitative comparisons included predictions of values of global force coefficients around the geometrics, pressure and friction coefficients on the surface, and wake analysis. Qualitative observation focused on flow generation near the bodies such as separation, velocity vectors. However, it is necessary to examine the problem further to have full confidence in their capabilities when applied to yacht hydrodynamics and more challenging flows.

No description available.

No description available.

This thesis presents the development and application of a numerical tool that couples a Navier-Stokes (NS) solver and a seakeeping solver to study the behaviour of a damaged ship upon flooding. Firstly, an NS solver that combines the volume of fluid (VOF) method with dynamic mesh techniques was developed to calculate the interactive dynamics of a damaged ship and floodwater. The VOF method was used to capture the fluid interface. The dynamic mesh techniques were employed to update the mesh following transient ship motions. To validate its performance in addressing complex free surface motion, the solver was firstly applied to several simple test cases including dam break, tank sloshing and compartment flooding. The benchmarking studies confirm that the present method is effective in solving free surface flow problems even with strong non-linearity. Then the solver was used to simulate the flooding process of a damaged barge which was fixed or freely floated in calm water . In comparison with the experimental data, both the ship and floodwater motions are well predicted by the present method. Then, a hybrid method, which couples the present NS solver and another in-house seakeeping solver based on potential flow method, was developed to study damaged ship flooding in waves. The dynamics of water flooding and sloshing in the ship compartments were calculated by the NS solver, whereas the hydrodynamic forces induced by the sea on the external hull surface were calculated using the seakeeping solver. The hybrid approach was tested by simulating the roll decay of a damaged Ro-Ro ferry and free motion of the ferry in regular beam seas. Validation against experimental data shows that the proposed method ensures acceptable numerical accuracy in flooding simulations while reducing the computational cost. In addition, the interactive dynamics of ship, sea wave, water flooding and sloshing were analyzed using the hybrid method. In conclusion, the proposed hybrid method provides a reliable and efficient tool for analyzing flooding and damaged ship dynamics. Its application could assist in understanding of the intricate flooding mechanism that remains limited for ship researchers, designers and safety authorities.