This thesis presents an investigation of techniques used to mitigate the undesirable effects of harmonics in power systems. The first part of this research develops an effective and useful comparison of alternative AC-DC converter topologies. In particular, a full evaluation of the circuit first proposed by Enjeti (known here as the Texas circuit) with a capacitively smoothed output voltage is made, specifically for operation as a 'clean power' supply interface for a variable speed drive (VSD). This mode of operation has not previously been reported in research literature. Simulation and experimental results verify the performance of the circuit and demonstrate that it draws a current with low harmonic content, but the circuit has a number of problems. This part of the research concludes that the six-switch rectifier is the most viable circuit for operation as a supply interface for a VSD due to its bidirectional power flow capability and its excellent versatility of performance. The second part of this research exploits the versatility of the six-switch rectifier and develops the current control strategy for operation of the circuit as a sinusoidal frontend and as a shunt active filter. It is found that the 'traditional' current control method suffers a significant drop in performance when the switching frequency is constrained to 2kHz due to high power levels. The major development in this thesis was an advanced current control strategy, where additional rotating frames of reference are introduced, thereby converting previously oscillatory current values to d.c. values. This is demonstrated to result in vastly improved immunity to disturbances such as supply distortion and a greatly improved steady state performance. In addition, the new controller requires no additional circuitry (apart from current transducers on the load current) and can be applied to an existing sinusoidal front end. Simulation results confirm the operation of the controller with the circuit operating as both a shunt active filter and as a sinusoidal front end. The new controller has been implemented on an experimental rig exhibiting the features of a high power inverter, i.e. low switching frequency and significant device turn-on and turn-off times, and the results confirm the superior performance of the new current controller.

GaMnN layers grown by plasma assisted molecular beam epitaxy (PAMBE) as a function of Ga/N ratio, Mn flux and growth temperature are assessed using a variety of structural characterisation techniques. At 680 C, the Ga/N ratio is found to have a dominant impact on the zinc-blende GaMnN epilayer growth rate and the resultant composition, morphology and microstructure. A maximum growth rate and an optimised microstructure are associated with growth under slightly Ga-rich conditions. A reduced growth rate and enhanced Mn incorporation are associated with growth under slightly N-rich conditions. Increasing Mn flux under N-rich conditions is considered to lead to a build up of a Mn surfactant layer during the early stages of growth and to a transition from zinc-blende single phase growth to zinc-blende/wurtzite mixed phase growth. Further, under Ga-rich conditions at low temperature, GaMnN films adopt a tilted growth mode, with close packed planes for both hexagonal and cubic phases being tilted roughly parallel to the growth surface, and this way of modified growth is also accompanied by improved Mn incorporation which is not commonly found for samples grown under Ga-rich conditions at elevated temperature. In addition, alpha-MnAs inclusions and voids extending into the GaAs buffer layer were identified in all samples, but are considered not to have a detrimental effect on layer electrical and magnetic properties.

Nowadays, numerous power electronics application requires operation at high temperatures. In order to address increasing change of reliability problems in power die attachments for high temperature and high reliability applications, sintering Ag nanoparticles has been used as bonding material for this work. Firstly, quantitative microstructure characterization of as-sintered Ag joints has been carried out. The resulting normalized thickness, pore size and porosity decreased, and grain size increased with increasing the sintering time. A time dependence of the form t1/n with n close to 2 or 3 can be further derived for the kinetics of the thinning, densification and grain growth within the sintered Ag joints. From the results can be seen, sintering kinetics is still in the intermediate stage, the densification had not been completed, and Ag grain would continue growing afterwards, which could further explain degradation behaviours of sintered joints during isothermal ageing tests and thermal cycling tests. Secondly, sintered Ag joints with four kinds of substrate metallization have been subjected to isothermal ageing tests at temperatures of 150°C, 200°C and 250°C for up to 32 days. The different microstructure patterns of sintered joints with four substrate finishes during isothermal ageing tests have been presented and compared, which could use the results to explain part of the degradation behaviours of the sintered Ag joints during thermal cycling tests and guide selection of suitable substrate finish for the die attachments in high temperature power electronic system. Furthermore, thermal cycling tests have been carried out to investigate the reliability of two sizes of sintered Ag joints and solder joints during temperature cycling between -55°C to 125°C and -55°C to 150°C. Microstructure evolution of sintered Ag joints was investigated by non-destructive and destructive characterization methods, which revealed the factors which could effect on the degradation during thermal cycling tests. With microstructure features of sintered joints observed from X-ray tomography and SAM, because a specific specimen can be evaluated over its lifetime, a true image of microstructure evolution of damage during operation can be obtained, and crack and degradation can be observed three-dimensionally.

The instrumentation and the developments presented in this thesis aim to address the challenges associated with in-service non-destructive testing of aeroengine components, imposed by severe restrictions on component accessibility. In-service non-destructive testing (NDT) is essential to ensure early detection of service-induced damage and prevent catastrophic failures of such safety-critical aeroengine components as turbine/compressor blades and disks, increasing the safety of operations and dramatically reducing the cost of engine changes. The main concern during their inspection is detection of fatigue cracking, typically initiated on the surface of the part. The application of inspection techniques used during overhaul to on-wing testing is restricted both by the complex, spatially-confined access to the components and the inaccessibility of the high-stress concentration surfaces in the full engine assembly. Two routinely-used techniques providing a solution to these challenges are the use of endoscopes for access and surface acoustic waves to inspect parts of the assembly. CHOTs are optically-activated ultrasonic transducers located on the surface of a sample, that use laser illumination to remotely generate and detect ultrasound, providing a non-contact wireless alternative to conventional piezoelectric transducers (PZTs). The benefits of optical activation are paired with fibre-optic light delivery in an endoscopic pulser -- a simple portable ultrasonic inspection system with flexible instrumentation, beneficial for testing hard-to-reach components in locations with limited access or hazardous environments. This thesis presents the instrumentation of the endoscopic pulser and the development of the self-adhesive portable CHOTs (SA CHOTs) to complement its potential field application by enabling transducer delivery to components in-service and application to large and curved parts, previously limited by laboratory-based direct on-sample CHOT fabrication. The developed transducers overcome frequent barriers for industrial adaptation of fully-optical inspection systems presented by the sample surface conditions such as reflectivity and roughness. The NDT capabilities of the CHOTs endoscopic pulser are demonstrated by performing detection of machined slots on controlled aluminium samples and representative industrial parts using 4--5 MHz surface acoustic waves. System capabilities for active structural health monitoring (SHM) as well as inspection of samples in motion are demonstrated. Although in its present configuration the size of the endoscope exceeds practical access requirements, the potential of the system for further development and miniaturisation is discussed.

The reliable performance of power electronic modules has been a concern for many years due to their increased use in applications which demand high availability and longer lifetimes. Thick Al wire bonding is a key technique for providing interconnections in power electronic modules. Today, wire bond lift-off and heel cracking are often considered the most lifetime limiting factors of power electronic modules as a result of cyclic thermomechanical stresses. Therefore, it is important for power electronic packaging manufacturers to address this issue at the design stage and on the manufacturing line. Techniques for the non-destructive, real-time evaluation and control of wire bond quality have been proposed to detect defects in manufacture and predict reliability prior to in-service exposure. This approach has the potential to improve the accuracy of lifetime prediction for the manufactured product. In this thesis, a non-destructive technique for detecting bond quality by the application of a semi-supervised classification algorithm to process signals obtained from an ultrasonic generator is presented. Experimental tests verified that the classification method is capable of accurately predicting bond quality, indicated by bonded area as measured by X-ray tomography. Samples classified during bonding were subjected to both passive and active cycling and the distribution of bond life amongst the different classes analysed. It is demonstrated that the as-bonded quality classification is closely correlated with cycling life and can therefore be used as a non-destructive tool for monitoring bond quality and predicting useful service life.

Forest structure is the complex 3D arrangement of all components within the forest architecture. This includes stems, foliage, branches (the components of trees) but also includes non-tree components such as understorey shrubs and herbs. Understanding the structural components of forests is critical when considering forest ecosystems. The structure of a forest can affect functional and compositional characteristics such as productivity and species richness with structure being an important factor influencing animal-habitat associations. Structural characteristics of forests include the size distribution and spatial organisation of trees, and the horizontal and vertical density of objects within the understorey. Trees are the dominant feature of any forest, but the understorey is also very important when considering forest characteristics. Examining the links between the spatial distribution of understorey material and ecological parameters, such as diversity and productivity, has an important role in ecological studies. There are multiple field survey techniques that can be applied when collecting data for a forest survey. For a technique to be an effective survey tool it should be readily quantifiable, repeatable, cost-effective, easily assessed, ecologically meaningful and where possible not contain observer bias. Traditional methods of forest survey are very common as they offer reliable, low cost estimations of forest structural parameters such as diameter, height and understorey cover. Recent developments within 3D data collection using terrestrial laser scanning (TLS) have allowed foresters and ecologists to reproduce the structural parameters collected during traditional forest surveys. These developments have shown the usefulness of 3D data collection in assessing forest structure, but have focused on replicating existing forest metrics rather than developing new ones. For TLS to reach its full potential within the field of forest ecology, new metrics and indices need to be developed specifically for laser scan analysis. This study developed and tested new methods of forest survey, concentrating on understorey vegetation, using commercially available TLS. Results showed that these new techniques can provide novel structural assessments of the understorey layers of forests for use in forest ecology surveys, not available through traditional methods. Using a new index describing the vertical component of forest understorey, it was shown how the relationship between deer browsing and forest structure can be identified through feature extraction from laser scanning. The method developed required minimal manual processing and was applied to large data sets. The structural changes between high and low deer density sites were also observed through the creation of an understorey density profile. This method, specifically targeted at the lower layers of the understorey, successfully identified structural change at the decimetre level. Using microtopography estimates from understorey point clouds it was shown how understorey complexity corresponded with vegetation surfaces extracted through TLS. This suggests that correlation between understorey structure (and therefore habitat type) and the microtopography of vegetation surfaces may be used for detailed assessment of understorey structural characteristics utilising TLS. In addition to the development of novel analysis methods, new techniques for acquiring TLS data of forest understorey were examined. The use of a standardised methodology for temporal surveying, utilising a common digital terrain model and fixed ground control, as developed here, provides a framework from which further data can be acquired. This approach offers a relatively quick, efficient, non-destructive assessment of temporal change within forests. A novel method of forest survey utilising handheld mobile laser scanning (HMLS) was also tested, showing its potential to complement static TLS surveying by providing increased survey coverage and allowing point cloud processing to be considered for areas which are otherwise difficult to access.

Micro Electro-Mechanical Systems (MEMS) rate sensors are inertial devices based on a resonating structure to measure and detect rate of angular rotation. For applications requiring high performance angular rate measurements it is important to be able to design MEMS rate sensors with high quality factors (Q-factor or Q). However, the device performance is affected by physical damping mechanisms which influence the overall quality factor of the device. High performances from a damping perspective can be achieved by identifying the dominant damping mechanism and considering different ways to reduce damping. For vacuum encapsulated devices, thermoelastic damping (TED) and support loss are the most important damping mechanism in MEMS resonators. This thesis focuses on understanding and quantifying the effects of thermoelastic damping and support loss on the damping performance of a supported ring resonator. An extensive review of different methods of modelling TED and support loss is presented. It is concluded that analytical models are not applicable to complex structures to predict damping due to their restrictive assumptions, and numerical approaches are required for accurate prediction. In this thesis a finite element based model using a fully coupled thermo-elastic approach is developed to quantify the thermoelastic dissipation in supported ring resonators and a detailed parameter study is conducted to understand the influence of ring geometry, support legs, micro-machined slots around the ring circumference, and material properties. The results show that damping in the support legs can have significant influence on TED in the resonator, and the optimum leg geometry can be identified to achieve high-Q. It is also observed that the addition of slots improves Q for resonators having higher energy loss. However, for high-Q rings slots have a detrimental effect. The results are considered for designing high-Q ring resonators with reduced levels of thermoelastic damping, and devices fabricated based on high TED Q designs are tested experimentally to compare the performance with predicted values. To analyse support loss behaviour in supported rings, a finite element model is developed using Perfectly Matched Layer (PML) technique to accurately predict the dissipation of energy through the elastic wave propagation from the resonator into its substrate. The model is validated by identifying key modelling issues, and strategies have been developed to effectively calculate PML parameters are discussed. The model is used to analyse support loss in supported ring resonators, and it is concluded that support loss is negligible compared to TED. Further investigations consider the effect of key design parameters on support loss, including a quantitative evaluation of the influence of an unbalanced vibration on the support loss by considering systematic design variation and manufacturing imperfections in the supported ring resonators. It is concluded that mass imperfection in the resonator and cycling asymmetry of the central supporting structure not only induce frequency splitting between degenerate pair of modes but also increase support loss due to the unbalance of the mode. It is expected that the results and contributions of this work will aid the development of high performance MEMS angular rate sensors by reducing the damping behaviour.

High voltage DC power supplies have been used in a wide range of applications such as Radio Frequency (RF) tube drives, industrial, military, aerospace, medical, and domestic applications. Recently as a result of advances in power electronics and magnetic materials, development of compact high voltage DC power supplies has become an active area of research, and resonant power converters have been introduced as a promising solution, owing to their potential for high efficiency and high power density. Several resonant converter topologies have been investigated to be used as modulators for supplying RF tubes (i.e. Magnetrons, Klystrons, and Traveling Wave Tubes) for various applications. The main aim of this study is to develop a control methodology for maintaining soft switching of a series resonant series loaded (SRSL) power converter based modulator for driving an industrial magnetron with variable load conditions. This thesis considers the design of a high voltage, high frequency, compact power supply for applications where the load is variable and/or nonlinear. One particular application where this is the case is when driving an industrial magnetron with a variable output RF energy requirement. The magnetron appears to the resonant converter as a variable load which can negatively affect the efficiency of the power conversion unless control of the converter is carefully considered. A compact power supply based on the SRSL resonant converter with an extended combined phase and frequency control is proposed. A novel control method based on three dimensional (3D) lookup tables has been developed, in order to control the load resonant converter whilst maintaining soft switching under variable load conditions. A direct quadrature (DQ) based modelling approach is used to develop a suitable model of the converter for control design. Based on the characteristics of the magnetron, an emulator prototype is also proposed to represent the magnetron load behavior in a low voltage laboratory environment. A detailed design and implementation procedure is presented, including the hardware design and control of the resonant power converter and the magnetron emulator. Simulation and experimental results are provided to validate the approach and in order to demonstrate the feasibility of the proposed converter modelling approach and control strategy. A good correlation between simulation and experimental results is obtained.

In this thesis, a dual inverter topology is considered as an alternative to a multilevel converter for the control of high speed machines. Instead of feeding to one end of the stator with a single power converter, this topology feeds from both sides of the stator winding using two converters, thus achieving multilevel output voltage waveforms across the load. A large amount of published work in the area of open end winding power converter topologies are focused on symmetrical voltage sources. This published research recognises the advantages of the converter system in terms of increased reliability, improved power sharing capability and elimination of common mode voltages when compared to traditional single sided three phase converter solutions. However isolated DC supplies come with the price of additional components thus increase size, weight and losses of the converter system. The aim of this project is, therefore, to investigate on reducing size, weight and losses of the open end winding motor drive by eliminating the need for isolated supply as well to achieve multilevel output voltage waveform. A traditional open-end winding induction motor drive has been analysed in terms of weight and losses and it has been clearly identified that the isolation transformer not only increases the size and weight of a drive system but also includes additional losses. A modified dual inverter system has then been proposed where one of the bridge inverters is floating, thus eliminated the need for isolated supplies. An asymmetric DC voltage sources ratio of 2:1 is utilised to achieve multilevel output voltage waveform across the load. The switching sequences are also analysed to identify the charging and discharging sequences to achieve control over floating capacitor voltage. This thesis describes the theoretical derivation of the modified converter model and algorithms as well as experimental results from an 11kW laboratory prototype.

Scientists at European Organization for Nuclear Research (CERN) are currently conducting feasibility studies for the Compact linear collider (CLIC); their proposed next experimental setup for gathering information on the fundamental particles of matter. This experiment will involve the simultaneous pulsing of 1300 klystron modulators to produce a 140us, 39GW pulse with a 50 Hz repetition rate. This proposal presents many demands for the connected power system as an effort is made to "hide" this pulse from the local distribution network - instead drawing only the constant average power of approximately 300MW. This challenge is considered in this work. In order to understand the optimal approach both the power system architectures and power electronics interfaces must be considered simultaneously. An approach to the optimisation of the power system architecture is described in this thesis. It is clear from this exercise that the optimum power converter topology for the interface between the electricity distribution network and the klystron modulators is the Modular Multilevel Converter (MMC). This converter is mainly used in modern HVDC transmission circuits as a result of its high efficiency and ability to produce high quality AC waveforms. Pulsing of the klystron modulators does however create further challenges for the inner control loops of an MMC. The placement of the pulse can create imbalances in the DC capacitors of the MMC submodules which may result in tripping of the converter if not corrected. This thesis proposes three arm balancing solutions to be applied together with the decoupled AC and DC side controller designed for the specified application. These proposed solutions to the aforementioned problems are successfully validated using simulation work in PLECS and using data from a laboratory scale prototype of one of the MMC interface power converters.