This thesis is concerned with the investigation of the efficacy of Zinc Oxide microvaristor compound for stress control on polymeric outdoor insulators. The preliminary work has involved a comprehensive literature survey, followed by extensive computational modelling and simulation studies as well as laboratory works covering experimental investigations and fabrication of insulator prototypes. The literature survey reviewed stress-induced degradations as the cause of ageing and insulation failures, the determination of electric field distributions, considerations for outdoor insulator modelling, and field-optimisation techniques for achieving stress relief. An 11 kV polymeric insulator has been modelled and simulated under dry-clean and wet-polluted surface conditions in order to obtain electric field distribution along the insulator creepage path. The critical high field regions on polymeric surfaces were identified. In addition, clean fog solid layer tests were carried out to experimentally examine dry band formation and electric discharges. Experimental investigations confirmed the results previously achieved from theoretical simulations. A non-linear pollution model has been developed for simulating polluted outdoor insulators. The field-dependent conductivity was derived from layer conductance measurements in a non-standard low voltage test. The proposed model was used to simulate insulators under fog and light rain conditions which respectively represent a uniform and non-uniform wetting action in practice. It was demonstrated that the nonlinear pollution model yields a more detailed and realistic field distribution compared with results obtained with models using constant/linear conductivity. Short-length microvaristor coating, having a cone-shaped structure, was introduced at both insulator ends for controlling high field, particularly near the high voltage and ground terminals. The performance of field grading was evaluated through a number of simulation scenarios. The introduction of microvaristor material with an appropriate switching characteristic has led to a substantial improvement in the electric field and heat distributions along the insulator profile. The prototype of an 11kV insulator with microvaristor grading material was fabricated in-house for preliminary tests. Lightning impulse (1.2/50 μs) flashover tests were carried out using the ‘up and down’ method, and the flashover voltage was estimated by the 50% probability breakdown, U50. The results of the lightning impulse test have indicated a considerable increase in the flashover voltage up to 21% when using microvaristor-graded insulator. Favourable field distributions obtained in the simulation study have indicated a strong correlation with the experimental results.

Small-scale embedded generation projects rated below 16A per phase are being integrated into low-voltage distribution networks in ever increasing numbers. Seen from the network operator's perspective as little more than negative load, the commissioning of such generators is subject to compliance with the Fit and Forget connection requirements of ENA Engineering Recommendation G83/1. This thesis has sought to quantify the electromagnetic switching transient implications of integrating very large volumes of embedded generation into the UK's low-voltage supply networks. Laboratory testing of a converter-interfaced PV source has been undertaken to characterise typical switching transient waveshapes, and equivalent representative source models have been constructed in EMTP-ATP. A detailed frequency-dependent travelling wave equivalent of the DNO-approved Generic UK LV Distribution network model has been developed and, by means of extensive statistical simulation studies, used to quantify the cumulative impact of geographically localised generators switching in response to common network conditions. It is found that the magnitude of generator-induced voltage and current transients is dependent on the number of concurrently switched generators, and on their relative locations within the network. A theoretical maximum overvoltage of 1.72pu is predicted at customer nodes remote from the LV transformer terminals, for a scenario in which all households have installed embedded generation. Latent diversity in switch pole closing and inrush inception times is found to reduce predicted peak transient voltages to around 25-40% of their theoretical maxima.

This research project has developed a high power RF Measurement system in the Centre for High Frequency Engineering, Cardiff University. There are two main contributions of this thesis. The first contribution is the implementation of step attenuators in the RF high power measurement system, incorporated with the approach of using the Sparameter model for correction measured waveform in measurement software processing. The step attenuator is situated between the broadband directional couplers (which obtains the incident and reflected signal from the DUT) and signal receivers (such as a sampling oscilloscope or MTA). The aim of this implementation is to extend the dynamic range of the measurement system and to develop a technique for reducing the recalibration process while the measurement system needs more attenuation in the signal condition part of the measurement system. These benefits allow the quick and convenient characterisation of the DUT under large signal excited environment. The second contribution is the further development of the Harmonic Bypass Structure (HBS), which is used instead of the step attenuator. The advantage of a HBS is that it can overcome the harmonic distortion drawback of the step attenuator approach. This drawback is caused by choosing improper high attenuation which is unsuitable for the measured signal condition. There is, therefore, a trade-off between the fundamental and harmonic frequencies. The advantage of an HBS is that it is capable of selecting the frequency band to attenuate while other frequencies can pass through to the receiver. In the meantime, the S-parameter model and modified measurement software are well suited similar to the step attenuator approach.

The large scale integration of wind generation capacity into an electricity network poses technical as well as economic challenges. In this research, three major challenges introduced by wind including non-correlated power output from geographically dispersed wind farms, wind variability and wind uncertainty were studied. In order to address each of the aforementioned challenges an appropriate modelling approach and case studies were used. The impacts of power output from dispersed wind farms on the Great Britain transmission reinforcement were studied using an optimal DC load flow combined with a power generation model. It was shown that Western and Eastern HVDC links play a crucial role to bypass the Scotland to England transmission bottleneck. The impacts of wind variability on the GB gas and electricity network were investigated through application of the Combined gas and Electricity Network (CGEN) Model. Additional gas storage capacity was shown to be an efficient option to compensate for wind variability. Two-stage and multi-stage stochastic programming models were developed to examine the impact of wind forecast uncertainty on the GB electricity and gas networks. Stochastic modelling approaches were shown to be efficient methods for scheduling and operating the system under wind uncertainty. The key contributions of this thesis are the investigation of the impacts of wind generation variability on the gas network, and development of twostage and multi-stage stochastic programming models of integrated gas and electricity network.

Microwave resonant sensors use the spectral characterisation of a resonator to make high sensitivity measurements of material electromagnetic properties at GHz frequencies. They have been applied to a wide range of industrial and scientific measurements, and used to study a diversity of physical phenomena. Recently, a number of challenging dynamic applications have been developed that require very high speed and high performance, such as kinetic inductance detectors and scanning microwave microscopes. Others, such as sensors for miniaturised fluidic systems and non-invasive blood glucose sensors, also require low system cost and small footprint. This thesis investigates new and improved techniques for implementing microwave resonant sensor systems, aiming to enhance their suitability for such demanding tasks. This was achieved through several original contributions: new insights into coupling, dynamics, and statistical properties of sensors; a hardware implementation of a realtime multitone readout system; and the development of efficient signal processing algorithms for the extraction of sensor measurements from resonator response data. The performance of this improved sensor system was verified through a number of novel measurements, achieving a higher sampling rate than the best available technology yet with equivalent accuracy and precision. At the same time, these experiments revealed unforeseen applications in liquid metrology and precision microwave heating of miniature flow systems.

The microwave devices and circuits need to be characterized prior to being employed in the design of systems and components. Unfortunately the measurement systems required to characterize the microwave devices and circuits have not kept pace with the emerging telecommunication technologies demands. This has resulted into a situation where either the circuits being employed in the components are unoptimized or the yield and turn-around of optimized circuits are slow. One of the contributing factors of such situations is the limitations of the existing measurement systems to scale up in performance to fulfil the necessary requirements. This thesis presents an enhanced multi-tone, time domain waveform measurement and engineering system. The presented system allows for a more considered, and scientific process to be adopted in the characterisation and measurement of microwave power devices for modern day communications systems. The main contributions to the field of research come in two areas; firstly developments that allow for accurate time domain measurement of complex modulated signals using commercially available equipment; and secondly in the area of active impedance control, where significant developments were made allowing active control of impedance across a modulated bandwidth. The first research area addressed is the fundamental difficulty in sampling multi-tone waveforms, where the main achievements have been the realisation of a high quality trigger clock for the sampling oscilloscope and a “Time Domain Partitioning” approach to measure and average multi-tone waveforms on-board. This approach allows the efficient collection of high quality vectoral information for all significant distortion terms, for all bands of interest. The second area of research investigated suitable impedance control architectures to comprehensively investigate out-of-band impedance effects on the linearity performance of a device. The ultimate aim was to simultaneously present independent, baseband impedances to all the significant baseband (IF) frequency components and to 2nd harmonic that result from a multi-tone excitation. The main achievement in this area was the ability of the enhanced measurement system to present the broadband impedance. At baseband this has been achieved in the time domain using a single arbitrary waveform generator (AWG) to synthesise the necessary waveforms to allow a specific IF impedance environment to be maintained across a wide IF bandwidth. To engineer the RF out-of-band load terminations at RF frequencies and to emulate specific power amplifier modes, a Tektronix AWG7000 Arbitrary Waveform Generator was used to deliver the desired impedances, practically fulfilling the wideband application requirements for reliable device characterisation under complex modulated excitations.

Magnetic characteristics of grain oriented electrical steel (GOES) are usually measured at high flux densities suitable for its applications in power transformers. There are limited magnetic data at low flux densities which are relevant for the characterisation of GOES for applications in metering instrument transformers and low frequency magnetic shielding in MRI (magnetic resonance imaging) medical scanners. Magnetic properties of convention grain oriented (CGO) and high permeability grain oriented (HGO) electrical steels were measured and compared at high and low flux densities at power magnetising frequency. HGO was found to have better magnetic properties at both high and low magnetisation regimes. This is because of the higher grain size of HGO and higher grain-grain misorientation of CGO. As well as its traditional use in non-destructive evaluation, Barkhausen Noise (BN) study is a useful tool for analysing physical and microstructural properties of electrical steel which control their bulk magnetic properties. Previous works deal with measurements carried out at high flux densities (0.2 T and above) but this work demonstrates that BN has different characteristics at low flux densities. The results show that the amplitude sum and the rms BN signals are higher for HGO than CGO steels at high flux densities. Below 0.2 T, the BN signal becomes higher for CGO steel. This is because of grain size/misorientation effects. Mechanically scribing of HGO samples on one surface transverse to the rolling direction was found to reduce the BN amplitude at high flux densities due to the decrease of domain width by scribing. The trend reverses again at low flux density. Removal of the coating from the surface of CGO and HGO electrical steels was found to increase the BN due to the widening of the 180° domains as a result of the release of the tensile stress imparted to the materials during coating. The BN characteristics of decoated samples with a 3 MPa tension applied were found to be similar to those observed before decoating demonstrating the close similarity between the effects of coating stress and externally applied stress on BN due to their similar roles in domain refinement. A strong correlation between average velocity of domain wall movement and changes in BN in conventional and high permeability steels was found which demonstrates that the dominant factor responsible for BN v emission is the mean free path of domain wall movement and hence the width of the predominant 180° domains in these materials. BN of commercially produced non-oriented electrical steel was found to be influenced by silicon contents and sample thickness. BN was found to increase with decreasing strip thickness and increase with increasing silicon contents owing to eddy current shielding effects. The rms values of the BN and the total sum of amplitudes were found to increase with the rate of change of flux density at all the peak flux densities measured. The findings show that the influence of sample thickness and silicon content is significant and must be taken into consideration when measuring and interpreting BN in non-oriented electrical steel.

The work reported in this thesis is concerned with the design, development and testing of a wireless 802.11 condition monitoring sensor for an electrical substation environments. The work includes a comprehensive literature review and the design and development of a novel continuous wireless data acquisition sensor. Laboratory and field tests were performed to evaluate the data acquisition performance of the developed wireless sensor. The sensor‟s wireless immunity to interference performance was also evaluated in laboratory and field tests. The literature survey reviews current condition monitoring practices in electrical substation environments with a focus on monitoring high voltage insulators and substation earth impedance. The data acquisition performance of the wireless sensor was tested in a laboratory using two artificially polluted insulators, in a fog chamber that applied clean fog. Analysis of the test results were found to be in good agreement with those recorded directly through a data acquisition card and transmitted via coaxial cable. The wireless impedance measurement of a 275kV transmission earth tower base field test was also performed and was found to be in agreement with previous published results from standard earth measurements. The sensor‟s wireless interference performance was evaluated at a field test site when no high voltage experiments were taking place. The sensors wireless interference performance was then tested in a laboratory environment before and during high voltage tests taking place. The results of these tests were compared to each other and to published results. These tests demonstrate the suitability of the sensor‟s design and its immunity to interference. The experimental work conducted using the developed wireless sensor has led to an understanding that continuous wireless data acquisition is possible in high voltage environments. However, novel condition monitoring systems that make use of such wireless sensors, have to take into account data losses and delays adequately. Furthermore, a solar power source was designed and constructed to be used for outdoor substation applications and the solar battery charging performance of the wireless sensor was tested in a solar laboratory.

Wind power is an essential part of the strategy to address challenges facing the energy sector. Operation of the electricity network in 2020 will require higher levels of response and reserve from generation. The provision of inertial response from wind turbines was investigated. A model was developed for the simulation of frequency on the mainland UK system, including a simplified model for a synchronous generator to represent Full Power Converter turbines. Two different methods of inertia response, the step method and the inertia coupling method, were modelled and introduced into the turbine torque speed control. Simulations illustrated the effects on primary frequency control for a high penetration of wind turbines. Results are shown for different demand levels with generation losses of 1320GW and 1800GW. A comparison of the inertia functions is included and the effect of wind speed and the constant speed region of the maximum power extraction curve. For the scenarios modelled only a small change in turbine output was required for inertia response (0.02p.u). Without inertia response a large increase in synchronous plant response was needed. A test rig was constructed consisting of a Full Power Converter bridge and a synchronous generator driven by a dc machine. Power converters were designed and constructed by the candidate. Vector control of both the generator converter and grid converter was implemented on a dedicated control platform. The inertia coupling function was implemented and a test frequency deviation injected to represent a load generation imbalance. Results compared closely to those from the model and demonstrated the capability to closely couple turbine speed to system frequency with adjustment of the response via a filter if desired. The experimental work confirmed the adequacy of the simplified generator model and further confirmed the possibility of using inertia response. The inertia coupling function was considered suitable for use for the UK system.

This thesis is concerned with the investigation of trifluoroiodomethane (CF3I) gas mixtures as an alternative for an insulation medium in high voltage applications. The work has involved a broad review of literature, followed by developing a test rig for carrying out experimental investigations, extensive computational modelling and simulation studies as well as extensive laboratory tests on CF3I gas and its gas mixtures. The literature survey reviewed the current trend of efforts taken by researchers to find solutions for minimizing the usage of sulphur hexafluoride (SF6) as a gas insulator, focusing on CF3I and its mixtures. The physical properties of CF3I are investigated, along with thermal and electrical properties. A new test rig has been designed and constructed specifically to be used for gas insulation research. The test rig is integrated with wireless temperature and humidity sensors, as well as an electrode gap length control system. The test rig is completed with a gas recovery system to ensure proper gas handling is carried out after each test. Extensive laboratory experimental investigations on CF3I mixtures have been completed, focusing on the mixture of CF3I-CO2 gas with a ratio of 30%-70%. Standard lightning impulse of 1.2/50 has been used, with both positive and negative polarity. The effects of electrode configuration, impulse polarity, electrode gap length, gas pressure, and CF3I content have been investigated. Insulation properties such as 50% breakdown voltage (U50) and V-t characteristics for each test condition are investigated and presented, as well as the electric field behaviour. Finite element method (FEM) has been used to determine the electric field behaviour of a given test condition. This study revealed that CF3I gas mixtures perform better under more uniform field condition. It was also found that an increase in gas pressure will increase the insulation strength and an increase in CF3I content is more likely to give benefit in conditions with a more uniform field when compared to less uniform field conditions. Also, relation between liquefaction temperatures of a CF3I-CO2 mixture with varying CF3I content has been developed for various pressures based on literature. Observations on solid by-products of CF3I have also been carried out. It has been found that iodine particles are deposited on both high voltage and ground electrodes, which can affect the insulation properties of CF3I and its mixtures.