An investigation into the design of a stand-alone photovoltaic water pumping system for supplying rural areas is presented. It includes a study of system components and their modelling. The PV water pumping system comprises a solar-cell-array, DC-DC buck chopper and permanent-magnet DC motor driving a centrifugal pump. The thesis focuses on increasing energy extraction by improving maximum power point tracking (MPPT). From different MPPT techniques previously proposed, the perturb and observe (P&O) technique is developed because of its ease of implementation and low implementation cost. A modified variable step-size P&O MPPT algorithm is investigated which uses fuzzy logic to automatically adjust step-size to better track maximum power point. Two other MPPT methods are investigated: a new artificial neural network (ANN) method and fuzzy logic (FL) based method. These use PV source output power and the speed of the DC pump motor as input variables. Both generate pulse width modulation (PWM) control signals to continually adjust the buck converter to maximize power from the PV array, and thus motor speed and the water discharge rate of a centrifugal pump. System elements are individually modelled in MATLAB/SIMULINK and then connected to assess performance under different PV irradiation levels. First, the MP&O MPPT technique is compared with the conventional P&O MPPT algorithm. The results show that the MP&O MPPT has faster dynamic response and eliminates oscillations around the MPP under steady-state conditions. The three proposed MPPT methods are implemented in the simulated PV water pumping system and compared. The results confirm that the new methods have improved energy extraction and dynamic tracking compared with simpler methods.

Due to the on-going worldwide trend towards investment in de-regulated electricity markets driven by political, economic and environmental issues, increasing interconnection between modern power systems has made power system dynamic studies much more complex. The continuous load growth without a corresponding increase in transmission network capacities has stressed power systems further and forced them to operate closer to their stability limits. Large power transfers between utilities across the interconnections stress these interconnections. As a result, stability of such power systems becomes a serious issue as operational security and reliability standards can be violated. On the other hand, the evolving technology of Wide-Area Measurement Systems (WAMS) has led to advanced applications in Wide-Area Monitoring, Protection and Control (WAMPAC) systems, which offer a cost-effective solutions to tackle these challenging issues. The main focus of this research project was to develop a wide-area based stability enhancement control scheme for large interconnected power systems. A new method to identify coherent clusters of synchronous generators involved in wide area system oscillations was the initial part of the work. The coherent clusters identification method was developed to utilise measurements of generators speed deviation signals combined with measurements of generators active power outputs to extract coherency property between system's generators. The obtained coherency property was then used by an agglomerative clustering algorithm to group system's generators into coherent clusters. The identification of coherent clusters was then taken as a base to propose a new structure of a WAMS based stability control scheme. The concept of WAMS and a nonlinear control design approach (fuzzy logic theory) was used to provide a comprehensive new control algorithm. The objectives of the developed control scheme were to enhance and improve the control performance of modern power systems. Thus, allowing improved dynamic performance under severe operation conditions. These objectives were achieved by means of enhanced damping of power system oscillations, enhanced system stability and improved transfer capabilities of the power system allowing the stability limit to be approached without threatening the system security and reliability.

Mankind’s constant pursuit of knowledge founded on innovation and creativity has led to a more complex world. A world governed by structures stretching high into the skies, objects flying faster than the speed of sound and exploration of the stars. These advancements have manifested in structures with complex geometrical designs, new building and engineering materials and the need for constant structural health monitoring (SHM). Structures comprising of a handful of components to those enlisting thousands are not impervious to failure. The ‘perfect’ material, one that will never fail, does not exist and may never. The need to remove the possibility of failure from the ever growing list of materials, components and structures has never been greater. There is an abundance of well suited techniques for a myriad of engineering problems, all suited for individual problem areas but all suffering from different weaknesses. There is no complete solution, only partial, short-lived solutions which are eclipsed by the next. The objective of this PhD is to explore a new and interesting non-destructive testing and evaluation (NDT/E) technique; nonlinear ultrasound. The work looks at assessment of both metallic and composite structures. As of recently, the technique has become well studied and documented, which has highlighted the associated benefits. Theoretically it builds on the fundamental theories of ultrasound testing techniques, but provides solutions that have amplified sensitivity in early damage detection. The techniques follow the main principle that the excitation of damaged regions results in clapping/rubbing mechanisms that give rise to further harmonic production, these further harmonics can be correlated to damage. Nonlinear ultrasound techniques were used to assess various systems that are prone to failure, these systems included adhesively bonded structures, bolted structures, structures susceptible to fatigue, compression loaded structures and impact damage of a composite plate. A series of experiments were undertaken which evaluated the ability of novel nonlinear ultrasound techniques to detect damage within these systems. Initially the primary principles of nonlinear ultrasound techniques those that affect the accuracy and repeatability of these methods were explored. The effects of hysteresis, input and output voltages of piezoelectric transducers (PZTs), modal analysis, and various nonlinear parameters were evaluated. This allowed for a clear understanding of what factors affect the accurate generation of data from these techniques. After determining various influences on results, adhesive and bolted joints were evaluated. With the aim to determine accurate nonlinear techniques that would be able to assess kissing bonds in metallic and composite joints, cracks in loaded structures and the loosened state of an individual bolt. A novel frequency specific nonlinear acoustic moment method was used to evaluate the presence of kissing bonds. The acoustic moment experiments provided good kissing bond detection probability in both metallic and composite joints. Further experiments explored the effects of structurally loading on the production of nonlinear responses. The results provided valuable insight into the potential for these techniques to assess defects in loaded structures. Thus kissing bond detection as well as the loaded state of the structure was possible using the nonlinear acoustic moment method developed. Joints account for a large proportion of mechanical and engineering structures, robust evaluation techniques would provide great savings in terms of maintenance, complete failure and lifetime service costs. The main finding was the ability to measure the loosened state of an individual bolt. Individual bolt loosened state was possible by using only two PZTs while assessing a system which included four bolts. This piece of research provides significant advancement in SHM of these structures, which has implications for small and large scale industrial joints. Fatigue failure is the most prominent failure mechanism in metallic structures, an investigation into a baseline-free method using nonlinear ultrasound was undertaken. A fatigued component was examined over its useful life using a modulated nonlinear ultrasound technique, where the development of a nonlinear ultrasound theory was used to assess the residual fatigue life by comparing the nonlinear modulated response to a theoretical model. Findings showed good correlation between the theoretical and experimental results, paving the way for further studies of baseline-free methods. A baseline-free testing method would provide a great leap forward in current testing procedures; the work highlights the potential of such methods. Further studies explored the possibility of using a computational model to predict damage levels in a material from a measured experimental nonlinear parameter. The findings of this research found good correlation between the second harmonic measured experimentally and that of the computational model. A novel nonlinear ultrasound based thermosonic technique was developed using a dual frequency excitation method. The investigation looked at determining barely visible impact damage (BVID) of a composite plate. The speed and accuracy of the method provided many advantages over current testing methods, which can be slow and deliver inconsistent results. The research relied on the determination of damage-specific resonance frequencies (DSRF), which results in focused heating at the damaged zones. The methodology explored could be used in an autonomous setup which would provide rapid assessment of composites and higher probability of damage detection. Finally the nonlinear ultrasound research completed establishes the broad range of applications for these techniques. The ease of adaptability of the techniques from metallic to composite, loaded and unloaded structures and their ability to improve other NDT/E techniques shows the great potential of nonlinear ultrasound. Through methodical application of the various nonlinear ultrasound techniques to other SHM problems there is a great degree of certainty that these methods would provide further benefits.

Ti-suboxides (TiO2-x) have been widely studied and tested for many potential applications, mostly for their tuneable electrical conductivity and chemical resistance. When the material has a low x (0 <x< 0.10), the dominant point defects in the structure consist of Ti3+ and Ti4+ interstitials and oxygen vacancies. However, for x=0.10-0.34 (Magnéli phases) the lattice is characterised by extended planar defects and crystallographic shear planes that offer metallic conductivity (104 S/m). Depending on the extent of the reduction their properties can be tuned to match the requirements of various applications such as cathodic protection, batteries, catalyst support for fuel cells and optical memory devices. This research has focused on the manufacture and characterisation of Magnéli phases fine scale fibres, with the aim to use them in a wear sensor and as electrodes in redox flow batteries. The manufacturing process was optimised to tune the properties and achieve reproducibility. A main part of the work was studying the electrical properties using Impedance Spectroscopy to analyse the electrical response of the samples across a wide range of frequencies. Additionally, the structure, density/porosity, hardness, wear rate, thermal expansion coefficient and the thermal stability were investigated to match the requirements of the applications. Thermal stability is the main disadvantage of Ti-suboxides, since a high temperature and oxygen rich environment can cause re-oxidation. Thermo-gravimetric analysis was used to determine the re-oxidation temperature and the kinetic mechanism. The Magnéli phases fibres were successfully used as the sensing element in a wear sensor that was based on resistivity measurements. They also gave very promising results when tested as electrodes for redox flow batteries.

Fluid power control is dominated by the throttling orifice. This is an inherently inefficient methodology that is responsible for low system efficiencies. The field of digital fluid power seeks to replace the throttling orifice with on-off valves and in the process greatly improve the efficiency of fluid power systems. One implementation of these on-off valves is the Switched Inertance Hydraulic System (SIHS) which operates in a similar way to Switched Mode Power Supplies (SMPS) in power electronics. In order to realise SIHS it is necessary to have valves that can switch large flow rates between high and low pressure supplies quickly. This report details the development of such a valve. It is demonstrated empirically that by using multiple grooves on a single spool a flow rate of 55L/min (at 10bar pressure drop) can be achieved whilst switching in <1ms. This is achieved through cascading a State Variable Feedback (SVF) controller with Iterative Learning Control (ILC) feedforward. The addition of novel stop learning conditions to the simple proportional lag compensated ILC scheme allow the valve to be tested to the limit of its abilities giving a minimum switching time of 0.5ms, where the limitation proved to be the range of the accelerometer used. Using the valve in a SIHS yielded promising initial results with efficiencies above 80\% being achieved across a range of switching ratios.

The development of marine renewable energy is attracting increasing attention due to its great potential in meeting human energy demands with limited negative environment impact. Various wave energy converter concepts have been proposed in attempt to convert wave energy into usable energy. Both experimental and numerical methods have been widely used to investigate the hydrodynamic performance of these devices in operational conditions and their survival characteristics in extreme sea states. This study focuses on developing a numerical procedure that can predict wave loads and run-up on fixed and moving offshore and coastal structures more accurately. The wave induced motions of flap-type wave energy converter (WEC) and its efficiencies are also investigated. The ultimate objectives of the study are to develop a rigorous approach for the safe and cost efficient design of general offshore structures and leading to the better design of wave energy converters with increased efficiency, and provide best practice guideline to the wave energy converter developers and researchers and engineers in the field. Non-linear hydrodynamic modelling in viscous flow has been used in the simulations. Even for moderate waves, nonlinear effects are important due to wave-structure interaction and also the expected large motions under operational conditions. It seems likely that estimates of performance will be unreliable unless the nonlinear effects associated with such large amplitude motions are properly accounted for. Extreme conditions are also be analysed to ensure device integrity. OpenFOAM, a free, open-source CFD package, has been applied in this work due to its strong capability in coastal and offshore engineering. The built-in viscous solvers interFoam and interDyMFoam have been selected and extended to model wave interactions with fixed and moving offshore and coastal structures, respectively. The solvers have been firstly extended to generate various wave conditions, including regular waves, focused wave groups and tsunami waves etc. New module has also been developed to advance the wave absorption capability in attempt to reduce computational cost of the numerical model by using smaller computational domain. In order to simulate the motion of WECs in waves, the code has been further developed to have functions on determining the wave-induced motions of WECs and on updating the computational domain automatically according to the motion of the WEC. By comparing with published experimental data, theoretical and numerical results on various physical problems, including wave interactions with varied seabed, a fixed vertical circular cylinder, a rotating half-submerged rectangular barge and a flap-type wave energy converter etc. it is confident to say that OpenFOAM is very capable of modelling nonlinear wave interactions with coastal and offshore structures accurately.

A range of electrical characterization techniques previously used for DSSC have been transferred to the study of planar perovskite devices. These include impedance spectroscopy (EIS), intensity modulated photovoltage spectroscopy (IMVS) and open-circuit voltage decay measurements (OCVD). An investigation into the observed response from these measurements has been carried out in order to gain a deeper understanding of device operation. Multiple processes with time constants on the microsecond, millisecond and second timescale were observed. The complimentary frequency and time domain techniques have been employed, showing excellent agreement between the two types of measurement. The high frequency (microsecond) process was found to be purely electronic in nature, which was linked to recombination. The geometric capacitance was shown to dominate this response, with accumulation of charge in the planar perovskite layer not observed. The lower frequency (millisecond and second timescale) processes were found to be linked to the coupling between recombination and the movement of ions. The low frequency EIS and IMVS measurements revealed that the recombination resistance was frequency dependent. The rate of change of the recombination resistance was found to be linked to the diffusion of ionic species. Activation energies for these processes were obtained (EA=0.55-0.66 eV) and shown to be in good agreement to computationally calculated values from literature for iodide vacancy migration. The same slow processes were also studied in the time domain using open-circuit photovoltage rise and decay measurements from well-defined equilibrium conditions. Comparable activation energies were also found using these techniques. The vacancy defect concentration was calculated to be 3x1019 cm-3, which is high enough for ionic double layers at the contacts to completely screen the built-in voltage across the perovskite at equilibrium in the dark. The slow dynamic processes observed under illumination or applied bias are therefore due to the rearrangement of ions in response to a changing electric field. As this rearrangement occurs, the rate of recombination is altered.

Dye-sensitized solar cells are a class of photovoltaics that have shown promise in producing electricity at a reasonable price. Although the processes limiting performance of the devices are quite well understood, their quantification has not been incorporated into a single consistent framework. In this study this framework, based on continuum charge transport equations, is presented and used to investigate the effectiveness of common characterisation methods. Approximate analytical solutions to the model are also derived and it is shown that these can be used to solve the device model inverse problem by fitting the solutions to impedance spectroscopy measurements. Experimental results indicate that the overall device model is a good description of the system and that it can be used to quantify different power loss mechanisms. Additionally some initial work was undertaken to formulate a charge transport model for a new class of photovoltaics called perovskite cells. The cell is modelled as a p-i-n heterojunction where the perovskite absorber is an intrinsic semiconductor sandwiched between two selective contacts. Simulations indicate that a significant built-in field drives free charges towards the contacts significantly improving charge collection.

Optimal and sub-optimal excitation control of a single machine system is investigated. Optimal excitation control is derived by using singular optimal control theory, while sub-optimal control is derived by using function minimisation techniques employing dynamic sensitivity functions. Both bang-bang and continuous feedback controls have been considered. A dual mode control arrangement has been used where the control operates in a bang-bang mode during large disturbances and in a continuous feedback mode during small disturbances. It was found that substantial overall improvements were achieved by the optimal and sub-optimal controls in the generator performance both under small and large disturbances. It was also found that the sub-optimal controls furnished a substantial improvement in transient performance of the system under a wide range of operating conditions and parameters. These conclusions have been confirmed on a practical micromachine system. A sub-optimal control consisting of non-linear functions of state has been investigated. This non-linear feedback control made substantial improvement in the system performance under a large range of operating conditions and parameters.

This research aims to improve the quality and availability of wave field information available to the developers and operators of wave energy converters (WECs) to aid in their design and operation. Applications relate to improving performance in varying wave climates, reducing extreme and fatigue-causing loads and reducing risk in critical marine operations through providing access to array-based and near-realtime surface elevation information. This Thesis describes a design process, leading from conceptual design, through to critical review. Design, assembly, commissioning and testing of multiple novel sensors involved technical work spanning mechanical, electrical, communications, signal processing and manufacturing disciplines. This required project management of areas including budget, procurement, IPR and programme scheduling. Three experimental procedures are outlined which were used to test the feasibility of a novel instrument conceived to meet the potential requirement for improved surface elevation data in large hydraulic test facilities and at sea. The first involves a method in the laboratory to assess the physical ("mechanical-only") surface tracking ability of long, floating, ribbon-like sensor elements that are aware of their position in two dimensions. Showing mean errors in wave height tracking of 6% and wave period tracking errors of 0.1% in irregular waves, across the widest available test range, results from surface tracking tests justify the subsequent testing of actual sensor implementations. Two approaches are taken: the first involves the modification and testing of a sensor technology comprising position-aware optical fibres with the second approach involving the design, fabrication and testing of floating sensors based on micro-electro-mechanical (MEM) sensor technology. Whilst wave period errors (individual time domain wave-by-wave comparisons) remain low for the optical fibre system at approximately 1% with standard deviations of approximately 10%, wave height errors are significant. Mean wave height error (depending on processing technique) range from -6% to 4% with standard deviations of 18% to 25% across irregular sea states. Performance is shown to be affected by wave steepness with wave trough tracking showing higher performance compared to wave crest tracking. Preliminary testing of the MEM-based sensor ribbons (in array form capable of measuring position in three dimensions) show wave height errors in regular waves to be on average 1.3% with standard deviations of relative error of 8.4%. Wave period errors and their standard deviations were below 1%. In irregular waves, mean significant wave height is under-predicted, across a range of directional seas, by 3% with standard deviation, across the tests and individual ribbons forming the array, of 7.5%. Peak wave period is under-predicted by 1.3% with standard deviation of 2.2%. Time domain statistics are not reported but it is expected that - as with the optical fibre system - performance degradation would occur when moving to irregular seas. Wave height error magnitude excludes the application of the developed sensors from small-scale tank testing where mm levels of accuracy are required. With further work, however, sensors based on this concept could potentially be used in larger scales and at sea where spatial wave field information is necessary, where wave period is critical and where other sensor techniques require baseline data.