The focus of this research is on small, well-established, homogeneous earthfill embankment dams that are currently in use and whose performance was previously outside the Reservoirs Act 1975, but are now governed by the new Flood and Water Management Act 2010. Many uncertainties are associated with such structures, a situation that can lead to the threat of dam failure when extreme climate conditions develop. Therefore, merely carrying out a deterministic assessment for such structures is insufficient and more sophisticated models, which reflect uncertain conditions of the dam site are required. This research presents the new advanced probabilistic slope stability model with precipitation effects (APSMP) developed by integrating the First Order Second Moment method (FOSM) with the deterministic slope stability model with precipitation (ASMP) using sliding block formulation. For the purpose of this study, the selected precipitation scenarios (rainfall intensity and duration) are obtained from past Met Office rainfall records and by applying the latest probabilistic model for predicting future precipitation projections for the UK (UKCP09). It is demonstrated that by implementing APSMP the notional reliability and probability of upstream and downstream slope failure for small homogeneous earthfill embankment dams can be quantified. To reflect the critical conditions conducive to slope failure a benchmark has been developed, as a reference for comparison of the effect of precipitation on the notional reliability and performance classification of the embankment’s slopes. By considering the probabilities of failure collated from APSMP and their associated performance, the impact critical precipitation effects could have on the notional level of engineering risk associated with slope failure is also identified. Hence, the dam’s risk, as categorized by the Flood and Water Management Act 2010, can be reassessed in terms of engineering risk. From the results obtained using APSMP a more informed assessment of small homogeneous earthfill embankment dams using limited information, including the level of uncertainty associated with the available site data, can therefore be carried out. Such an approach is therefore well placed to support and enhance the decision making process when evaluating the likelihood of dam failure, its impact on infrastructure performance and public safety, especially in relation to future climate effects.

The safety and economics of nuclear utilities and High Risk Operations (HROs) is very dependent on the quality of both the management and operations personnel. The decisionmaking capability of management is important in ensuring that the operators are adequately prepared to deal effectively with accidents. This means that management has to understand the risk of power production and adequately deal with it, so that the viability of the utility is not compromised, while still operating in an economical manner. The vehicle for enabling management to function effectively is a dynamic designed organizational structure in which all personnel communicate well and is designed to use the best features of human performance according to their roles within the organization. The purpose of this thesis is to investigate the application of Beer’s Viable Systems Model of an organization to see if it fits the requirements of nuclear power plant (NPP) organizations to be both economical and safe and to propose modifications to VSM to more closely match the needs and requirements for NPP organizations. It is realized that organizations can operate effectively when they are not operating under stress, however the situations changes when they are subjected to accidents. A well designed organization is one that is prepared and can respond quickly to an accident. Because of the possible impact of accidents on organizations, the thesis studied a number of accidents that have occurred in the nuclear industry and to cover the more general case of HROs, accidents affecting other industries were also considered. Based upon the accident investigations and from the study of human behavior, insights were developed related to characteristics of both managers and operations personnel. These insights led to the development of an understanding of how VSM needs to be considered when dealing with HROs, including NPP utilities. The top down structure of VSM mirrors the basic needs of an organization, but the demands of responding to the safety requirements of an organization requires an understanding of the effects of the time response limits placed upon an organization. These requirements dictate changes to the VSM organization designed for normal commercial organizations, where time for decisions and actions are not so important and these changes are addressed. Although in normal commercial organizations risk assessment is considered, in the NPP utility and HROs business it is extremely important, since poor decisions with respect to risk can affect the viability of the organization. The thesis covers the use of risk assessment technology to improve management decision-making. Currently, the industry uses risk assessment techniques for total plant risk (more for licensing purposed) and for plant state risk assessment. The contribution of the thesis is seen as contributing to improvements in the understanding of VSM and making some modifications to it. The importance of time response of organizations in combating accidents and its human performance background is demonstrated and the mechanisms whereby performance is improved by the use of procedures and training is explained.

The main objective of this thesis is to perform a comprehensive simulation study of the statistical variability in well scaled fully depleted ultra thin body silicon on insulator (FD-UTB SOI) at nanometer regime. It describes the design procedure for template FDUTB SOI transistor scaling and the impacts of statistical variability and reliability the scaled template transistor. The starting point of this study is a systematic simulation analysis based on a welldesigned 32nm thin body SOI template transistor provided by the FP7 project PULLNANO. The 32nm template transistor is consistent with the International Technology Roadmap for Semiconductor (ITRS) 2009 specifications. The wellestablished 3D ‘atomistic’ simulator GARAND has been employed in the designing of the scaled transistors and to carry out the statistical variability simulations. Following the foundation work in characterizing and optimizing the template 32 nm gate length transistor, the scaling proceeds down to 22 nm, 16 nm and 11 nm gate lengths using typically 0.7 scaling factor in respect of the horizontal and vertical transistor dimensions. The device design process is targeted for low power applications with a careful consideration of the impacts of the design parameters choice including buried oxide thickness (TBOX), source/drain doping abruptness (σ) and spacer length (Lspa). In order to determine the values of TBOX, σ, and Lspa, it is important to analyze simulation results, carefully assessing the impact on manufacturability and to consider the corresponding trade-off between short channel effects and on-current performance. Considering the above factors, TBOX = 10nm, σ = 2nm/dec and Lspa = 7nm have been adopted as optimum values respectively. iv The statistical variability of the transistor characteristics due to intrinsic parameter fluctuation (IPF) in well-scaled FD-UTB SOI devices is systematically studied for the first time. The impact of random dopant fluctuation (RDF), line edge roughness (LER) and metal gate granularity (MGG) on threshold voltage (Vth), on-current (Ion) and drain induced barrier lowering (DIBL) are analysed. Each principal sources of variability is treated individually and in combination with other variability sources in the simulation of large ensembles of microscopically different devices. The introduction of highk/ metal gate stack has improved the electrostatic integrity and enhanced the overall device performance. However, in the case of fully depleted channel transistors, MGG has become a dominant variability factor for all critical electrical parameters at gate first technology. For instance, σVth due to MGG increased to 41.9 mV at 11nm gate length compared to 26.0 mV at 22nm gate length. Similar trend has also been observed in σIon, increasing from 0.065 up to 0.174 mA/μm when the gate length is reduced from 22 nm down to 11 nm. Both RDF and LER have significant role in the intrinsic parameter fluctuations and therefore, none of these sources should be overlooked in the simulations. Finally, the impact of different variability sources in combination with positive bias temperature instability (PBTI) degradation on Vth, Ion and DIBL of the scaled nMOSFETs is investigated. Our study indicates that BTI induced charge trapping is a crucial reliability problem for the FD-UTB SOI transistors operation. Its impact not only introduces a significant degradation of transistor performance, but also accelerates the statistical variability. For example, the effect of a late degradation stage (at trap density of 1e12/cm2) in the presence of RDF, LER and MGG results in σVth increase to 36.9 mV, 45.0 mV and 58.3 mV for 22 nm, 16 nm and 11 nm respectively from the original 29.0 mV, 37.9 mV and 50.4 mV values in the fresh transistors.

The work presented is specifically concerned with improving cumulative fatigue damage life predictions in the region of long lives where most of the stress levels experienced by a specimen or structure are below the constant amplitude fatigue limit. Further, an attempt has been made to improve these predictions in such a manner as to be feasable in the design process, i.e. mathematical simplicity and minimum data reaquirements to enable predictions of improved accurancy to be made. Several existing cumulative damage prediction rules are examined with reference to their accuracy, areas of application and suitability for the design process. Results are presented for constant amplitude tests on four different specimen configurations and various forms of block loading tests with finally a triangular modulation of anplied load on a single specimen configuration in rotating bending. The cumulative damage experimental results are used to investigate the resultant effect of loads below the original constant amplitude fatigue limit. The observed behavioural patterns have been rationalized to form the basis of a cumulative damage rule. The main feature of the rule is the division of the fatigue life into two stages termed Stage A and Stage B which are defined as follows:- Stage A Initiation and Micro-crack propagation. Stage B Macro-crack propagation. A linear summation is performed on each stage simulating observed patterns of behaviour, using a modified S-N curve derived from the standard constant amplitude S-N curve. The length of each stage is determined by the length of that same stage in a constant amplitude test at the highest damaging stress level in the spectrum. The division of the life between Stage A and Stage B has heen shown to be a unique quantity dependant on specimen geometry and an empirical relationship has been derived relating the KT value of the notch to the slope of a life in Stage A versus Total life to failure plot. As a consequence of different behaviour in Stages A and B different cumulative damage behaviour is predicted for differing notch severities. Certain aspects of size effects are examined and discussed and the general areas of apnlication of the cumulative damage rule are outlined. Finally, the proposed rule is tested against other cumulative damage rules and experimental results presented in the cumulative damage literature. Suggestions are presented for further research work to further investigate and extend the areas of applicability of the proposed method.

An experimental comparison is made between two turbulent boundary layers produced in a low-speed water channel subjected to different pressure gradient distributions. Both flows involve identical favourable streamwise components, generated via a lateral contraction of the flow area; and in the second case, an additional spanwise gradient is imposed by curving the walls. The measurement system and methods are developed in full, with rigorous testing and validation allowing the uncertainty and accuracy of the results to be estimated. Hot-wire anemometry is employed to take measurements of velocity using miniature single-film probes. A Hydrogen bubble visualisation system enables an inspection of the coherent turbulent structures in the boundary layer near the wall. The mean-velocity measurements show a continued rise in the Reynolds number downstream accompanied by a fall in the coefficient of friction, in spite of a relatively high streamwise acceleration. This unorthodox behaviour was found to occur for both flows. In response to the acceleration, changes in the statistical moments of streamwise velocity show an increased dominance of high velocity fluctuations near the wall. This corresponds with the results of the structure visualisations which reveal a rise in the mean spanwise spacing of the low-speed fluid elements. The pressure gradients of the two cases are generated using a novel approach which aims to make the effects from each strain easier to evaluate. The additional spanwise component in the second case induced a cross ow in the boundary layer which reached 11% that of the local external velocity. Despite this, the measurements and low-speed streaks show the turbulence to be relatively insensitive to this level of three-dimensionality. A simple numerical method is presented to model the development of the low-speed fluid streaks in different mean-velocity distributions. Using this model, greater cross flow magnitudes in the boundary layer are tested for which a positive streak dependence is found.

Investigation of turbulence response during constant temporal acceleration and deceleration can assist in improving the understanding of turbulence evolution and flow physics. Such flows have potential importance in engineering applications for example the air flow through the main trachea during the breathing cycle experience temporal acceleration and deceleration. The previous experimental and theoretical investigations based on conventional computational fluid dynamics (CFD) modelling could not provide the detailed information about turbulence response in the near-wall region in such types of flows. In particular, the response of near-wall structures has not been studied for turbulent flow with temporal acceleration and deceleration. In the present study, turbulent flows involving temporal acceleration and deceleration has been investigated using DNS and LES. A fully implicit fractional step method is implemented in the present study. The Navier-Stokes equations are discretised using finite volume method. Second-order- implicit Crank-Nicolson method is used for temporal discretisation for the convective and viscous terms. Second-order accuracy of spatial discretisation is achieved using four neighbouring points to calculate velocity gradients. A uniform grid is used in the streamwise and spanwise directions while a non-uniform grid is employed in the wall-normal direction. The numerical implementation has been validated for three test cases. The dynamic subgrid-scale model has been implemented for LES calculations. The LES model implementation has been validated through comparison with benchmark data available in literature. As one of the first DNS of accelerating turbulent flow, this study has produced a comprehensive database of turbulent statistics which can be used for unsteady turbulence modelling and validation. The detailed investigation has substantially enhanced the understanding of turbulence response for such flows. The flow physics has been studied in detail using turbulent kinetic energy budget analysis, vorticity analysis, anisotropy invariant maps and energy spectra. The evolution of new turbulent structures during the acceleration has been investigated using low-speed streaks and λ2 plots and many interesting ow characteristics have been found. The effect of different acceleration rates has been studied using LES. The turbulence propaga- tion in the core region has been studied for different acceleration rates. Turbulent flow subjected to constant temporal deceleration has also been investigated using LES. The effect of different deceleration rates has been also studied. The turbulent flow response to temporal deceleration has been analysed using the rms velocity and vorticity, kinetic energy budget and Reynolds stress anisotropy tensor analysis.

Turbulent skin-friction control is the subject of much research and the use of transverse (spanwise) oscillating motions offers the means of obtaining a significant reduction in skin-friction. Dielectric barrier discharge (DBD) actuators can be used to generate spanwise oscillating waves but the difficulty in placing a sensor in the area of plasma gives rise to problems in recording near-wall velocities. A modified version of the Shyy et al. (2002) DBD model was integrated into a direct numerical simulation (DNS). This numerical model was used in a series of two-dimensional simulations, in initially quiescent ow, and the results were compared to results reported from experimental investigations. A close affinity was found confirming that the DBD model is satisfactory. Both a temporal and a spatial, spanwise oscillating ow were investigated. Only one plasma profile was investigated. Three actuator spacings were investigated. Only the largest actuator spacing resulted in a gap between each plasma profile that was larger than the plasma profile width itself. A spatially uniform plasma configuration produced larger DR% than spanwise wall oscillation for both spatial and temporal waves, maximum DR = 51% compared to a DR = 47% for a spanwise wall oscillation. Increased skin-friction reductions originated from the displacement of the Stokes layer. The spatial wave produced lower skin-friction values than temporal waves for all the configurations. For both spatial and temporal waves the performance of the discrete configurations in producing an overall skin-friction reduction decreased with increasing actuator spacing. Using both temporal and spatial waves, the configuration with the largest spacing, which is relatively small, did not produce a drag reduction for any case that was tested.

Traceable low force metrology should be an essential tool for nanotechnology. Traceable measurement of micro- and nanonewton forces would allow independent measurement and comparison on material properties, MEMS behaviour and nanodimensional measurement uncertainties. Yet the current traceability infrastructure in the UK is incomplete. This thesis describes the incremental development of the low force facility at the National Physical Laboratory (NPL). The novel contribution of this thesis has three components. First, specific modifications to the NPL Low Force Balance were undertaken. This involved developing novel or highly modified solutions to address key issues, as well as undertaking detailed comparions with external ans internal traceability references. Second, a triskelion force sensor flexure was proposed and mathematically modelled using both analytical and finite element techniques, and compared to experimentally measured spring constant estimates. The models compared satisfactorily, though fabrication defects in developed prototype artefacts limited the experimental confirmation of the models. Third, a piezoelectric sensor approach for quasistatic force measurement was proposed, experimentally evaluated and rejected. Finally, an improved design for a low force transfer artefact system is presented, harnessing the findings of the reported investigations. The proposed design combines proven strain-sensing technology with the advantageous triskelion flexure, incorporating an external stage and packaging aspects to achieve the requirements for a traceable low force transfer artefact.

The rational operator-based approach to depth from defocus (DfD) using pill-box point spread function (PSF) enables texture-invariant 3-dimensional (3D) surface reconstructions. However, pill-box PSF produces errors when the amount of lens diffraction and aberrations varies. This thesis proposes two DfD methods, one using the Gaussian PSF that addresses the situation when diffraction and aberrations are dominant, and the second based on the generalised Gaussian PSF that deals with any levels of the problem. The accuracy of DfD can be severely reduced by elliptical lens distortion. This thesis also presents two correction methods, correction by distortion cancellation and correction by least squares fit. Each method is followed by a smoothing algorithm to address the low-texture problem of DfD. Most existing human activity recognition systems pay little attention to an effective way to obtain training silhouettes. This thesis presents an algorithm to obtain silhouettes from any view using 3D data produced by Vicon Nexus. Existing background subtraction algorithms produce moving shadow that has a significant impact on silhouette-based recognition system. Shadow removal methods based on colour and texture fail when the surrounding background has similar colour or texture. This thesis proposes an algorithm based on known position of the sun to remove shadow in outdoor environment, which is able to remove essential part of the shadow to suffice recognition purpose. Unlike most recognition systems that are either speed-variant, temporal-order-variant, inefficient or computational expensive, this thesis presents a near real-time system based on embedded silhouettes. Silhouettes are first embedded with isometric feature mapping, and the transformation is learned by radial basis function. Complex human activities are then learnt with spatial objects created from the patterns of embedded silhouettes.

Acoustic cavitation plays an important role in a broad range of biomedical, chemical and engineering applications, because of its magnificent mechanical and chemical effects. Particularly, the irradiation of the multi-frequency acoustic wave could be favouritely employed to promote these effects, such as enhancing the intensity of sonoluminescence, increasing the efficiency of sonochemical reaction, and improving the accuracy of ultrasound imaging and tissue ablation. Therefore, a thorough understanding of the bubble dynamics under the multi-frequency acoustic irradiation is essential for promoting these effects in the practical applications. The objective of this PhD programme is to investigate the bubble dynamics under dual-frequency excitation systematically with respect to bubble oscillations, the acoustical scattering cross section and the secondary Bjerknes force (a mutual interaction force between two oscillating bubbles). Spherical gas bubbles in water are considered. Both analytical analysis based on perturbation method and numerical simulations have been performed in this thesis. The analytical solutions of the acoustical scattering cross section and the secondary Bjerknes force under dual-frequency excitation have been obtained and validated. The value of the secondary Bjerknes force can be considered as the linear combination of the forces derived under the single-frequency approaches. The predictions of those analytical solutions will be impaired for the cases with large acoustic pressure amplitudes. The numerical simulations reveal some unique features of the bubble dynamics under dual-frequency excitation, e.g., the combination resonances (i.e., their corresponding frequencies corresponding to the linear combinations of the two component frequencies) and the simultaneous resonances (i.e., the simultaneous occurrence of two resonances in certain conditions). The influence of a number of paramount parameters (e.g., the pressure amplitude, the equilibrium bubble radii, the power allocation between the component waves, the phase difference and the driving frequency) on the bubble dynamics under dual-frequency excitation is also investigated with demonstrating examples. Based on that, the parameters for optimizing the dual-frequency approach are proposed. In addition, the effects of thermal effects and mass transfer on the bubble dynamics have also been discussed.