Search

Using the Atiyah-Ward construction, we examine the solutions of the self-dual Yang-Mills equations for an SU(2) gauge theory, dimensionally reduced from |R⁴ to |R². There are two main reasons for doing this: (i) To provide a large class of relatively simple examples which elucidate how non-singularity and physical field configurations are related to the parameterization of the Atiyah-Ward construction. (ii) To construct analogues, for pure non-abelian gauge theories, of the superconducting vortex solutions of the abelian Higgs model, in the hope that these will provide the dominant field configurations describing the QCD vacuum. First, Bäcklund transformations are used to construct axially symmetric solutions, and the analogues of the 't Hooft instantons. These results are then generalised, within the twister theoretic framework of the Atiyah-Ward construction, to produce an infinite dimensional parameter space of complex non-singular solutions in each of the Atiyah-Ward ansätze. The field configurations are expressible as unitary group integrals occurring in lattice gauge theories - this leads to a simple proof of non-singularity, and a convenient means of calculating properties of the field configurations using strong and weak coupling expansions. The structure of the field configurations is further elucidated using symmetry arguments and numerical computations. Finally, suggestions are made as to how these solutions may play a role in the QCD confinement mechanism.

This work presents a computational and experimental investigation on the effects of liquid properties on multiphase flow in horizontal and upward-inclined pipes. The overall aim of the research is to gain more insight into the effects of viscosity, density, and surface tension on multiphase flow behaviour. The computational part of the study simulated the drift velocity of an elongated gas bubble, commonly referred to as Taylor bubbles in a 2D domain. The computational efficiency of 2D simulation makes it a preferred option for parametric studies, where different operational parameters are systematically varied to understand their impact on the flow behaviour. By simplifying the geometry to 2D, it becomes easier to explore a wide range of parameters and assess their effects. In this work, 78 simulations were run at six different pipe inclinations to study the effects of 13 different liquids. The simulation results show that liquid density alone has little or no influence on the Taylor bubble's velocity at all pipe inclinations, while the bubble's velocity is heavily influenced by liquid viscosity at all pipe inclinations. However, surface tension appears to show unique effects on the Taylor bubble, when the pipe inclination is less than 45 degrees, surface tension seems to have no effect on Taylor bubble's drift velocity, the effects of surface tension only become notable when the flow inclination is at 45 degrees or above. This behaviour has not been reported by previous researchers based on the review done, as no previous work has singled out one liquid physical property to study its effect while keeping the other properties constant. Experimental study on the effects of liquid properties was also carried out to generate two-phase flow data using three different liquids, water, surfactant solution, and glycerol solution in a 19 mm ID pipe and 4m length. The experimental campaign was carried out in horizontal and 15 degrees upward inclinations. Two-phase flow regime and slug frequency data were acquired using a high-speed camera, quick-action solenoid valves were used to collect liquid holdup data, and a differential pressure transducer was used to measure the pressure drop across the flow section. The effects of liquid properties and variation of inclination angles on different flow parameters including flow regime, pressure drop, liquid holdup, and slug frequency were investigated and reported. Flow regime maps were also developed for all flow orientations investigated. The data generated would be a useful contribution to the gas/liquid flow database and could potentially be used to develop or improve multiphase flow correlations.

Shark skin has fascinated biologists, engineers, and physicists for decades due to its highly intricate drag-reducing structure, which has motivated a plethora of research into bio-inspired hydrodynamically efficient surfaces. Throughout this thesis the effect of shark scales on the boundary layer is investigated, with a particular focus on the role of riblets in combination with denticles. In addition to examining flows over shark scales studies are also presented investigating the behaviour of Reynolds Averaged Navier-Stokes (RANS) models close to solid boundaries, and the scaling and driving mechanisms of secondary flows over ribletted surfaces. Extensive numerical and analytical studies are carried out to determine the sensitivity of eleven turbulence closures to the near-wall grid resolution, and their consistency with asymptotic solutions. Results inform the choice of turbulence models adopted for simulations of wall bounded flows, particularly where numerical errors must be minimised. Secondary flows over longitudinal riblets are found to be driven by Reynolds stress anisotropy, consistent with Prandtl’s second type of secondary flow. The strength of the vorticity field is heavily dependent on the inner-scaled riblet spacing s+ where two distinct regimes arise; a viscous regime where vorticity production is balanced by molecular viscous diffusion, and an inertial regime where an effective turbulent viscosity balances anisotropic production. The transition between these regimes occurs when riblet tips protrude into the buffer layer and cause increased turbulent mixing (s+ ≈ 30), such that vorticity reaches its maximum before reducing as s+ increases further. Riblets in combination with shark scales do not operate as they do when applied to smooth walls. Experimental and numerical studies reveal that riblets act to reduce pressure drag acting on roughness elements, rather than the viscous forces typically associated with longitudinal riblets. The mechanisms leading to this behaviour are driven by the ability of riblets to restrict spanwise motion and maintain streamwise-aligned near-wall flow. By doing so riblets protect downstream denticles from high momentum impinging fluid, and reduce high magnitude swirl generated at the exposed denticle edges, which can otherwise lead to increased turbulent production and enhanced momentum transfer through the roughness sub-layer. These mechanisms lead to a significantly more efficient rough surface than smooth denticles, although do not necessarily lead to an overall reduced drag compared to a flat plate. These studies conclude that riblets have evolved as a mechanism to reduce or eliminate the skin friction increase due to the presence of scales. The combination of scales and riblets appears to be relatively hydrodynamically efficient in terms of skin-friction drag, whilst also acting to maintain boundary layer attachment and providing the other advantages associated with scales such as anti-fouling, abrasion resistance, and defence against parasites.

Performance evaluation of miscible and near-miscible gas injection processes is available through conventional finite difference (FD) compositional simulation, which is widely used for solving large-scale multiphase displacement problems that always require large computation time. A step can be taken to reduce the time needed by considering low-resolution compositional simulation. The model can be adversely affected by numerical dispersion and may fail to represent geological heterogeneities adequately. The number of fluid components can possibly be reduced at the price of less accurate representation of phase behaviour. Streamline methods have been developed in which fluid is transported along the streamlines instead of the finite difference grid. In streamline-based simulation, a 3D flow problem is decoupled into a set of 1D problems solved along streamlines, reducing simulation time and suppressing any numerical dispersion. Larger time steps and higher spatial resolution can be achieved in these simulations, particularly when sensitivity runs are needed to reduce study uncertainties. Streamline-based reservoir simulation, being orders of magnitude faster than the conventional finite difference methods, may mitigate many of the challenges noted above. For gas injection, the streamline approach could not provide a high resolution or adequate representation for the multiphase displacement. In this work, the streamline simulations for both compositional and miscible gas injection were tested. In addition, the conventional gas injection scheme and detailed comparison between the FD simulation and the streamline approach are illustrated. A detailed comparison is given between the results of conventional FD simulation and the streamline approach for gas displacement processes. Finally, some guidelines are given on how the streamline method can potentially be used to enhance for gas displacement processes.

Deposits of turbidity currents - turbidites - commonly exhibit upward-fining grainsize profiles, reflecting deposition from flows with simple rapidly-waxing then progressively-waning velocity structures. However, turbidites with patterns of multiple cycles of inverse-to-normal grading are not uncommon. Such deposits are interpreted as being deposited under the influence of repeated waxing-waning velocity cycles within multi-pulsed turbidity currents and are termed "multi-pulsed turbidites". Multi-pulsed flow can be initiated by sequences of retrogressive submarine failures in which each slumping episode can form a pulse in the velocity structure, or may arise due to the combination of multiple flows at downstream confluences; separate flows may even run into each other over long distances. In the first case, it has been inferred that multi-pulsed deposits might carry signals of flow initiation, with each slump linked to a seismic impulse, and further, that such signals can be recognised in the vertical grading structures of distal turbidites. The focus of this research has been to establish i) how multi-pulsed flow dynamics and associated deposits vary along flow pathways and ii) the degree to which grading structures in turbidites deposited by multi-pulsed flows permit inference of flow initiation mechanisms. Initial experiment modelling of single- and multi-pulsed solute-driven gravity flows shows that internal pulses are necessarily advected forward, eventually merging with the flow head such that multi-pulsed flows transition from being cyclically waxing-waning to waxing on arrival then monotonically waning. This finding implies that initiation signals should be distorted then lost in any deposits along the flow pathway. Accordingly, an interpretational template for the spatial variation in turbidite character along flow pathways was developed, accounting for both pulse merging and flow combination at confluences. Further experiments were conducted to support a scaling analysis to estimate merging lengths; these lengths are shorter than those documented from prototype settings, and may reflect a limitation in the scope of application, arising from experimental constraints. Experiment modelling of single- and multi-pulsed sediment-driven gravity flows confirms the occurrence of the pulse merging phenomenon in turbidity currents. Analysis of associated deposits confirms the downstream spatial transition from multi- to uni-pulsed turbidites, albeit with the point of transition being more proximal in the laboratory deposit than the point of pulse merging. However, the spatial persistence of the complex velocity structure up to the point of merging need not be reflected in the associated deposit. Beyond the merging point, single-pulsed turbidites must always be deposited. Such deposits cannot be used to infer flow initiation mechanisms.

This thesis is concerned with solar energy storage systems in terms of storage materials and storage systems for high temperature applications. The main focus has been given to either improve the thermophysical properties of the storage medium or to improve the design of the storage medium by optimizing the solar energy storage system. Nitrate salts have been chosen as the phase change material with nanoparticles as additive materials. Different types or concentrations of nanoparticles and different types of nitrate salt have been selected and studied. There are different objectives, starting from a few grams (up to 5 g) to kilograms (up to 3 kg), that have been considered through this research. In the first objective, nanosalts, which are nanoparticles seeded in the nitrate salts, have been prepared by two different methods, either the 2-step method or the 1-step method. Nanosalt prepared by the 2-step method showed higher specific heat capacity (cp) than the base salt by 10.5% with higher thermal conductivity (k) values up to 60%. Up to 6% increments in total thermal energy storage have been observed for nanosalt (iron oxide (Fe2O3) nanoparticles and binary nitrate salt). Additionally, the 1-step method was used to prepare copper oxide (CuO) nanoparticles directly inside the nitrate salt, which showed improvements in cp in comparison to base nitrate salt. In the second objective, alongside with the thermophysical properties measurements, material characterizations have been considered using different devices to show the morphology of the surface area of salts and nanosalts samples. In the third objective, an experimental rig has been designed and built to study heat transfer of salts and nanosalts. Temperature measurements have been made at different axial, radial and azimuthal locations. Both charging and cooling of the salt (or nanosalt) were studied, showing that improvements in the charging process are related to the type or concentration of the nanoparticles material. Overall, heat transfer is improved in the case of nanosalt compared to salt alone. Two different types of salt were tested in this experiment rig which was single salt (sodium nitrate) and binary solar salt (sodium nitrate: potassium nitrate by 60:40 molar ratio) with different additives materials such as CuO (by 0.5 wt. %) and Fe2O3 (by 0.1 wt. %, 0.5 wt. % and 1 wt. %). In the fourth objective, Computational Fluid Dynamics software has been used to solve the charging process of salt and nanosalt. A validation for the ANSYS CFX code (version 17.0) is conducted by comparing the experimental data with the simulation data. A good agreement is obtained for both cases of salt or nanosalt. In the fifth objective, an optimization for the solar energy storage system has been conducted. Different designs for the storage system employing finned structures were studied. The new combination effect of both nanosalt and fins system has been studied using the validated CFX code, showing a promising improvement in the charging process in comparison to salt alone, nanosalt alone or salt-fins system alone. As a result, our overall aim is to improve the thermal properties of nitrate salts and optimize the thermal energy storage system.

Vapour absorption refrigeration systems (VARSs), which utilise eco-friendly refrigerants (water), can work based on low-grade thermal energy, such as the solar energy. Using solar energy can relief the high electrical load on many national grids around the world as the peak load almost coincides with the high solar intensities time during summer. However, high initial cost, big specific size and low coefficient of performance are the main challenges that face the VARSs. Therefore, improving the efficiencies of the components of a solar refrigeration system, such as the solar collector, generator and absorber, is crucial to improve the overall efficiency of that system and to reduce its size and the cost. To improve the efficiency of a solar VARS, nanofluids are proposed in this work through direct and indirect ways. The direct way is seeding functional nanoparticles in the aqueous solution of a VARS. The expectations are that refrigerant (steam) can be generated efficiently via direct volumetric absorption of the solar energy at the generator and can be absorbed effectively via the Brownian motion of the nanoparticles at the absorber. While, the indirect way is; using aqueous nanofluids in direct solar collectors can harvest the solar energy in an efficient way, saving it in a storage tank. This stored energy is used later to generate the steam in the generator of the VARS. This work aims to investigate fundamentally the applicability of utilising nanofluids for solar absorption refrigeration systems through performing three main studies: Firstly, a comparative study of gold, copper oxide, gold and copper oxide hybrids, and carbon black nanofluids has been conducted to investigate the photo-thermal conversion efficiency. The results have shown that gold nanofluids are not feasible for solar application due to the high cost and low performance comparing to the carbon black nanofluids. Moreover, this study has demonstrated that blending different nanofluids of different narrow spectral absorption peaks can really broaden the effective spectral absorption peak but reduces its value due to the accompanied dilution of the blended nanofluids as the overall volume unavoidably increases. Secondly, a comparative study of solar steam generation among gold nanofluids, carbon black nanofluids and a thin carbon-based porous medium has been conducted. The results have also shown the infeasibility of using gold nanofluids comparing with the carbon black nanofluids due to the high cost and low absorptivity. Furthermore, this study has shown the superiority of using a thin, carbon-based porous medium in producing steam due to its capability to absorb most the solar energy in micro-sized thickness. While, very high nanoparticles concentration is required to trap and absorb the solar energy in such a thin layer, which consequently leads to an instability, high viscosity and high-cost issues. Finally, a study of the steam absorption by and generation from aqueous lithium bromide solutions seeded with carbon black and carbon nanotubes has been conducted. The results have demonstrated that a very low concentration of carbon black nanoparticles can reduce the transparency of the solutions to zero. However, seeding nanoparticles in the solutions has shown the negligible effect on the steam absorption rate, which demonstrated that the Brownian motion of the nanoparticles has a negligible effect on the steam absorption. Although the experiments conducted in this project showed negligible enhancement in the steam absorption, obvious enhancements in the photo-thermal conversion efficiency and steam generation were achieved by using nanofluids. Recommendations are suggested for future work to study other affecting aspects of seeding nanoparticles in aqueous solutions.

Turbidity currents are the largest agent of global sediment transport and their deposits, submarine fans, are the largest sedimentary structures on Earth. Submarine fans consist of networks of seafloor channels, which are vital pathways for sediment and nutrient transport to the deep ocean. This work focusses on flow dynamics within these channels, with the aim of understanding the role of the channel form on flow development and identifying implications for the development of channels and, ultimately, for submarine fans. Laboratory experiments have been conducted of continuous saline gravity currents traversing fixed-form channel models with a range of planform geometries. Both velocity and density data were gathered to investigate the effect of a channel on the flow field. Numerical simulations have also been conducted, using a Reynolds-averaged Navier Stokes model and a shear stress transport turbulence closure. These allow an extension of the laboratory analysis, both in terms of physical domain size, data resolution and measured variables. Velocity data reveal how partial confinement exerts a first order control on the vertical variation in flow structure. The channel half-depth acts to limit the height of the velocity maximum, resulting in the development of a confined, high-velocity flow core. The channel form also constrains the lateral and three-dimensional flow structure. Secondary flow rotation, characterised by a local reversal in the radial pressure gradient, is shown here to be inhibited by low channel sinuosity and large levels of overbank fluid losses. A change in cross-sectional channel profile is capable of switching the dominant cross stream basal flow direction of these structures. Furthermore, channels are shown to cause flow tuning, whereby flows of differing magnitudes entering a channel reach are rapidly modified to show a much restricted magnitude range, that remains quasi-stable thereafter. For the cases studied, this quasi-equilibrium state is characterised by a symmetrical cross channel basal stress profile. The existence of such a state could explain how seafloor channels can achieve a degree of planform stability.

With increasing demand for producing clean and pollution free energy, special attention has been paid to wind turbines and improving their performance. Reducing the effect of wingtip vortices on the wind turbine performance can be achieved by using winglets which work to weaken the impact of wingtip vortices by diffusing them away from the blade tips. The general trend of the literature has considered winglets as diffusers of the wingtip vortices. However, extending the span of the turbine rotor by attaching winglet could improve the potential of a rotor to capture more kinetic energy from moving air. Accordingly, the winglet planform and airfoil play vital roles in wind turbines performance. The present work reports on the study of the effect of winglet planform and winglet airfoil on the wind turbine performance using Computational Fluid Dynamics (CFD) tools. The National Renewable Energy Laboratory (NREL) phase VI rotor is used as a baseline rotor and the CFD results are validated with the experimental data in terms of torque, pressure and normal force coefficients for different wind speeds. In this study, two turbulence models are used, which are the SST k-ω and the Spalart-Allmaras models, which can be used to predict the properties of the fluid flow in the computational domain. Both of the models show a good match of the numerical results when compared to the experimental data, at a range of low wind speeds from 5m/s to 8m/s, due to the absence of stalled flow. At higher wind speeds of 10m/s, the SST k-ω model shows a better match between the calculated torque and the experimental measurements. Consequentially, the SST k-ω model is implemented to predict the behaviour of fluid flow in all the CFD calculations in the present study. The aerodynamic behaviour of two winglet planforms is investigated. These are rectangular and elliptical winglets to increase the NREL phase VI rotor performance. The performances of four winglet configurations are assessed when compared to the baseline power, at the range of wind speeds from 5m/s to 25m/s. The configurations are obtained by changing the winglet planforms and airfoils using the S809 and PSU 94-097 airfoils. In this regard, the elliptical planform causes a minimizing of the wingtip vortices, more than the rectangular planform, due to the reduction of the elliptical tip by 75% when compared to the rectangular tip. A rectangular planform shows a better performance than the elliptical planform in percentages of power increase. The highest percentage in the power increase is achieved by attaching the rectangular planform that tilted by a cant angle of 45o and extended by 15cm. This improvement is slightly more than 9%, at the range of low wind speeds from 5m/s to 10m/s, since the flow is almost attached. Considering the effect of winglet airfoil, the study reports that, choosing a suitable winglet airfoil is mainly dependent on the aerodynamic coefficients of the selected airfoil, such as lift coefficient (Cl), drag coefficient (Cd) and moment coefficient (Cm). For this purpose, a preliminary analysis is conducted using the Xfoil code to predict the aerodynamic coefficients of selected airfoils (S801, S803, S805A and S806A airfoils). The S806A and S805A airfoils are chosen to create two different configurations. The 3D calculations show more increase in the NREL phase VI power is achieved by attaching the configuration that created using the S806A airfoil since this airfoil has less drag coefficient.

Managing shared resources in multiprocessor real-time systems can often lead to considerable schedulability sacrifice, and currently there exist no optimal multiprocessor resource sharing solutions. In addition, the choice of task mapping and priority ordering algorithms also has a direct impact on the efficiency of multiprocessor resource sharing. This thesis argues that instead of adopting a single resource sharing protocol with the traditional task mapping (e.g., the task allocation schemes that are based on utilisation only) and priority ordering (e.g., the Deadline Monotonic Priority Ordering) algorithms, the schedulability loss for managing shared resources on multiprocessors can be effectively reduced by applying a combination of appropriately chosen resource sharing protocols with new resource-oriented task allocation schemes and a new search-based priority ordering algorithm (which are independent from multiprocessor resource sharing protocols and the corresponding schedulability tests). In this thesis, a Flexible Multiprocessor Resource Sharing (FMRS) framework is proposed that aims to provide feasible resource sharing, task allocation and priority assignment solutions to fully-partitioned systems with shared resources, where each resource is controlled by a designated locking protocol. To achieve this, the candidate resource sharing protocols for this framework are firstly determined with a new schedulability test developed to support the analysis of systems with multiple locking protocols in use. Then, besides the existing algorithms, three new resource-orientated task allocation schemes and a search-based priority ordering algorithm are developed for the FMRS framework as the task mapping and priority ordering solutions. The choices of which locking protocols, task allocation and priority ordering algorithm should be adopted to a given system are determined off-line via a genetic algorithm. As demonstrated by evaluations, the FMRS framework can facilitate multiprocessor resource sharing and has a better performance than the traditional resource control and task scheduling techniques for fully-partitioned systems.

In a Cyber-Physical Control System (CPCS), there is often a hybrid of hard real-time tasks which have stringent timing requirements and soft real-time tasks that are computationally intensive. The task scheduling of such systems is challenging and requires flexible schemes that can meet the timing requirements without being over-conservative. Fixed-priority scheduling (FPS) is a scheduling policy that has been widely used in industry. However, as an open-loop scheduler, FPS has low system dynamics and no feedback from historic operation. As the working conditions of a CPCS will change due to both internal and external factors, an improved scheduling scheme is required which can adapt to changes without a costly system redesign. In recent years, there is a large research interest in the co-design of control and scheduling systems that explicitly considers task scheduling during the design of a controller. Many of these works reveal the possibility of adapting control periods at run-time in order to accommodate varying resource requirements and to optimise CPU utilization. It is also shown that control quality can be traded off for resource usages. In this thesis, an adaptive real-time scheduling framework for CPCS is presented. The adaptive scheduler has a hierarchical structure and it is built on top of a traditional FPS scheduler. The idea of dynamic worst-case execution time is introduced and its cause and methods to identify the existence of a trend are discussed. An adaptation method that uses monitored statistical information to update control task periods is then introduced. Finally, this method is extended by proposing a dual-period model that can switch between multiple operational modes at run-time. The proposed framework can be potentially extended in many aspects and some of these are discussed in the future work. All proposals of this thesis are supported by extensive analysis and evaluations.

The prediction of the hydrodynamics of immiscible liquid-liquid flow is essential for the accurate design of process intensification technologies using micro capillaries and packed bed systems. It is also very complex mainly because of the shear exerted between the phases and the inertial effects that are present. The aim of this work relies on the computational fluid dynamics (CFD) study of liquid-liquid slug flow in capillaries, offering insight on the effectiveness of monoliths and packed bed reactors in biodiesel production for process optimisation. The main parameters for modelling slug flow in capillaries were investigated. An attempt to predict the terminal droplet velocity was developed by relating the drag force and the Reynolds number over a single droplet of different sizes dispersed in a continuous flow in a capillary. The results showed a significant effect of the film thickness and droplet length on the Stokes-coefficient suggesting predominance of Stokes flow for the conditions under study. Also, the motion of a droplet in pressure driven horizontal flow was investigated. The numerical predictions revealed a notable influence of the film thickness, slug and drop length on the droplet velocity. Moreover, the study of the interfacial forces in the limits of high and low viscosity ratios was developed using an alternative method to the Volume-of-Fluid method. The velocity and shear profiles across the two-phases were efficiently achieved and visualisations of the internal hydrodynamics structures in the continuous and dispersed phases were compared to similar studies from the literature. Furthermore, an efficient predictive tool based on slug flow correlations from the literature was developed to calculate the film thickness, droplet velocity and pressure drop in a capillary when the properties of the fluids and the inlet flow rates are known. The results are in good agreement with those predicted by CFD methods and with experimental data found in the literature. This tool can be useful for design purposes of technologies involving two-phase flows in capillaries. Also, it can be helpful for predicting initial conditions and input parameters in CFD models applied for two-phase flow in porous systems. Finally, the first stage of a pressure drop model for liquid-liquid flow in porous media is proposed based on slug flow correlations. The model includes the influence of interfacial forces and inertial effects in porous media and can be further implemented to predict operating conditions of a packed bed reactor for biodiesel production.

Many Mixed Criticality scheduling algorithms have been developed with an assumption that lower criticality level tasks may be abandoned to guarantee the schedulability of higher criticality tasks when the criticality level of the system changes. But it is valuable to explore means by which all of the tasks remain schedulable throughout criticality level changes. This thesis introduces a semi-partitioned model which allows all of the tasks to remain schedulable if only a bounded number of cores increase their criticality levels. In such a model, some lower criticality tasks are allowed to migrate instead of being abandoned. Different possible semi-partitioned approaches are proposed and analysed in this thesis. It is concluded from the experiments results that the semi-partitioned algorithm provides improved schedulability and performance of multi-core mixed criticality systems while enables all tasks to keep executing in the majority of scenarios.

The Sellafield site in Cumbria, UK maintains three steam heated evaporators which operate at sub-atmospheric pressure. They are used to evaporate and thus concentrate nitric acid based highly active liquor, which is a by-product arising from reprocessing spent nuclear fuels. Corrosion occurs on the internal surfaces of the evaporators in contact with the liquors. The rates of corrosion are a function of the local surface temperatures and heat transfer rates. Therefore accurate heat transfer predictions inside the evaporators is highly desirable as it enables good predictions to the rates of corrosion. The aim of this study was to use engineering calculations and computational fluid dynamics (CFD) to provide predictions of the continuity, momentum and energy transfer occurring inside evaporators operating at sub-atmospheric pressure. Three evaporator scales were chosen for this study: (i) two small unscaled cylindrical test rigs which were 0.1 m diameter, and had liquid fill depths of 0.1 m and 2.215 m respectively; (ii) a test rig representing a quarter scale slice of an industrial evaporator, and had a liquid fill depth of 0.8m; (iii) and lastly Sellafield Evaporator C, which represented a full scale industrial evaporator design, and had a fill depth of 2.35 m. Thermal resistance investigations were performed on the unscaled cylindrical test rigs which proved conduction heat transfer through the walls removed all sensitivity to the specified boundary conditions. Single phase CFD simulations were also performed on the test rigs which showed a symmetrical geometry assumption could not be used to simplify the modelling approach. Two phase Eulerian-Eulerian CFD simulations were performed on the one quarter scaled test rig. A custom length scale for use in the interfacial area density was developed and used. The length was a function of a user prescribed rate constant. In the simulations evaporation at the free surface was modelled. Published experimental data was used to validate the simulations, and showed that the length scale required a rate constant of 1 Hz to simulate evaporating flows at sub-atmospheric pressure. The CFD models which were developed were applied to the simulations of the full scale industrial evaporator design, Sellafield Evaporator C. The results shows the evaporator behaved similar to an unconstrained thermosyphon reboiler with distinct counter rotating convection cells. Indications of nucleate boiling was not present on all heat transfer surfaces as previously thought, which meant surface corrosion rates may be lower than anticipated.

This thesis focuses on scheduling for hypervisor systems in the automotive domain. Current practices are primarily implementation-agnostic or are limited by lack of visibility during the execution of partitions. The tasks executed within the partitions are classified as event-triggered or time-triggered. A scheduling model is developed using a pair of a deferrable server and a periodic server per partition to provide low latency for event-triggered tasks and maximising utilisation. The developed approach enforces temporal isolation between partitions and ensures that time-triggered tasks do not suffer from starvation. The scheduling model was extended to support three criticality levels with two degraded modes. The first degraded mode provides the partitions with additional capacity by trading-off low latency of event-driven tasks with lower overheads and utilisation. Both models were evaluated by forming a case study using real ECU application code. A second case study was formed inspired from the Olympus Attitude and Orbital Control System (AOCS) to further evaluate the proposed mixed-criticality model. To conclude, the contributions of this thesis are addressed with respect to the research hypothesis and possible avenues for future work are identified.

One of the main obstacles to overcome regarding the uptake of renewable energy technologies, specifically wind and solar energy, is their intermittency. Current energy storage techniques are costly and in-efficient. Fuel cells are a promising candidate for future energy storage, as part of an integrated system combining renewable energy with hydrogen production as the storage vector with reconversion. The Polymer Electrolyte Fuel Cell (PEFC) has the greatest potential for use with micro-generated renewable power and is suitable for the widest range of applications. Hence it has received a great deal of attention from research institutions and industry over the last few decades. However, they suffer performance limitations due to flooding by liquid water in the porous components forming the electrodes of the cell. Two numerical investigations utilising different methods to probe multiphase transport in porous media, and one experimental investigation into the flow through partially saturated porous media, are presented. The porous media under investigation are typical materials for PEFC gas diffusion layers (GDLs), and the influence of compression of the material on the multiphase transport is investigated. In addition, a further study assessing the suitability of pore-scale capillary pressure models for predicting multiphase flow behaviour is included as a final research chapter.

Numerical computations have been performed to evaluate the influence of bedform roughness on turbulent transport of sediments in geophysical flows. Special attention is paid to turbidity currents, which are responsible for the transport of sedimentary rocks far into the deep ocean. It has been suggested that enhanced turbulence mixing in flows over rugose topography contributes to the unexpectedly large runout lengths of naturally occurring turbidity currents. One of the objectives of this study is to provide evidence for against this conjecture. We perform computations over a wide range of periodic arrays of rectangular roughness elements, We find that a strong peak in turbulent mixing occurs when the width-to-height ratio equals a critical value of seven. We also find that a strong peak in resistance to flow occurs at the same critical value. These are competing effects, with the former acting to promote, and the latter acting to diminish runout length. So we are not able to conclude definitively that the enhancement of mixing is responsible for long runout lengths. We continue by considering flows over periodic arrays of shapes which are representative of bedforms that occur in the natural environment. We again find a strong correlation between the optimisation of both turbulence mixing and resistance to the flow. We are unable to distinguish bedform shapes that promote long runout length relative to the flat bed case. However, we are able to distinguish those bedform shapes that have large resistance to flow and large turbulence mixing compared to those that have low resistance and low turbulent mixing, with the latter case occurring for widely spaced asymmetric dunes with a long low angled slope facing the flow. Finally, we develop a model for flow and sediment transport which takes into account erosion and deposition from the bottom boundary. We first apply this model to flow over fixed dune shapes, in order to assess the influence of bedform shape on flow capacity, stratification, and the energy budget. An important result of this study is that flow capacity is optimised for the class of bedform shapes that promote low flow resistance and low turbulent mixing. We conclude by applying the model to the two-way coupled flow of a mobile dune, starting from an initially symmetric inherited dune morphology. We find that, for sufficiently large grain sizes, the dune evolves into a sequence of asymmetric dunes, rather than to a flat bed, and that the long-time evolution tends to be towards those dune shapes that promote large relative flow capacity. However, the model has a discrepancy in that it is unable to prevent the dune shape exceeding the maximum angle of repose. Hence, further work is required before these results can be regarded as reliable.

In real-time systems, a schedulable task-set guarantees that all tasks complete before their deadlines. In functional programming, atomic execution provides the correctness of the program. Priority-based functional reactive programming (P-FRP) allows the usage of functional programming in the real-time system environment. The abort-and-restart (AR) is a scheme to implement P-FRP but an appropriate scheduling approach does not exist at the moment. Hence, efficient analysis is needed for the AR model. In this thesis, the schedulability analysis for the AR model is introduced and it shows that finding the critical instant for the AR model with periodic and sporadic tasks is intractable, and a new formulation is derived. Afterwards, a new priority assignment scheme is developed that has the performance close to the exhaustive search method, which is intractable for large systems. The technique of deferred preemption is employed and a new model, deferred abort (DA), provides better schedulability and dominates the non-preemptive model. Lastly, a tighter analysis is introduced and the technique of the multi-set approach from the analysis of cache related preemption delay is employed to introduce a new approach, multi-bag. The multi-bag approach can apply to both the AR model and the DA model. In the experiments, the schedulability of the AR model is improved at each stage of the research in this thesis.

Interpretation and Symbolic Model Checking are powerful techniques in the field of testing. These techniques can verify the correctness of systems by exploring the state space that the systems occupy. As this would normally be intractable for even moderately complicated systems, both techniques employ a system of using approximations in order to reduce the size of the state space considered without compromising on the reliability of the results. When applied to Real-time Systems, and in particular Worst Case Execution Time Estimation, Abstract Interpretation and Symbolic Model Checking are primarily used to verify the temporal properties of a system. This results in a large number of applications for the techniques, from verifying the properties of components to the values given variables may take. In turn, this results in a large problem area for researchers in devising the approximations required to reduce the size of the state space whilst ensuring the analysis remains safe. This thesis examines the use of Abstract Interpretation and Symbolic Model Checking, in particular focusing on the methods used to create approximations. To this end, this thesis introduces the ideas of Information Theory and Lossy Compression. Information Theory gives a structured framework which allows quantifying or valuing information. In other domains, Lossy Compression utilises this framework to achieve reasonably accurate approximations. However, unlike Abstract Interpretation or Symbolic Model Checking, lossy compression provides ideas on how one can find information to remove with minimal consequences. Having introduced lossy compression applications, this thesis introduces a generic approach to applying lossy compression to problems encountered in Worst Case Execution Time estimation. To test that the generic approach works, two distinct problems in Worst Case Execution Time estimation are considered. The first of these is providing a Must/May analysis for the PLRU cache; whilst common in usage, the logical complexity of a PLRU cache renders it difficult to analyse. The second problem is that of loop bound analysis, with a particular focus on removing the need for information supplied by annotations, due to the inherent unverifiability of annotations.

Experimental and numerical investigations have been conducted in order to evaluate the accuracy of the Mixture Model, a depth-resolved and time-averaged multiphase numerical model, in predicting the behaviour of dilute surge-type turbidity currents. The effects of turbulent dispersion and turbulence modulation upon sediment transport within turbidity currents are directly modelled via their incorporation into the Mixture Model. Modelled predictions of flow front propagation and deposit density are compared against both experimental data and refined two-fluids model from previous studies. When modelled using the formulation of Chen & Wood (1985), turbulence modulation does not affect on the propagation of dilute turbidity currents significantly. Turbulent dispersion can be modelled by incorporating the formulation of Simonin (1991) into the slip equation of the Mixture Model. Its effect is strongest in dilute flows carrying fine particles and diminishes when either grain size or flow concentration increases. Modelled turbulent dispersion effects are too strong in simulations of flows carrying silicon carbide particles; Mixture Model simulations agree poorly with both experimental data or refined two-fluids model results of the deposit mass profile. Yet turbulent dispersion is essential to ensure that model predictions of flows carrying glass beads compare well with experimental data. The reasons for the discrepancy between modelling approaches best suited to each of these flow types remains poorly understood. A new analytical approach is developed to evaluate the effect of the lift force on particles of small, intermediate and large particle Reynolds number immersed in two-dimensional shear flows. The lift force always reduces the magnitude of the particle settling velocity and may push particles forward or backward, depending on the sign of both the lift coefficient and the flow vorticity. Given plausible velocity profiles within natural turbidity currents, the effect of lift force on the sand-like particles immersed in such turbidity currents is negligible. It may become significant when the ratio of the particle density to the flow density approaches unity. New experiments are presented for flows over the flow concentration range 0.25 – 5% and grain size range 58 - 115μm. The data are used to facilitate a more complete validation of the Mixture Model, based on flow front propagation rates, deposit mass density and deposit grain characteristics. Modelling results for first two variables are in good agreement with the experimental data, when turbulent dispersion effects are incorporated. For reasons which remain unclear, the model cannot simulate the unexpected experimental result that deposit grain size is largely unfractionated if the standard deviation of the source material is less than 11 but significantly fractionated if it exceeds 18. This discrepancy requires further work.