The work presented in this thesis focuses on improving the thermoelectric performance of cost-efficient, widely available, non-toxic transition-metal sulphides. Three families of materials, titanium-disulphide derivatives, tetrahedrites and bornites, were investigated owing to their promising electrical and/or thermal properties. Alongside thermoelectric property measurements, structural characterisation and measurements of various physical properties were carried out in order to explain the relationship between structural and physical features. A series of composition CoxTiS2 (0 ≤ x ≤ 0.75) was synthesised and structural characterisation was carried out using a combination of powder X-ray diffraction and neutron diffraction analysis. Electrical and thermal transport property measurements, thermal analysis, magnetic susceptibility measurements and muon spectroscopy analysis were carried out to complete the investigation. The intercalation of cobalt within the van der Waals’ gap of TiS2 leads to the formation of three superstructure types, monoclinic M5S8 (x = 0.25), trigonal M2S3 (x = 1/3) and monoclinic M3S4 (x = 0.5). At intercalation levels x ≤ 0.20 and x ≥ 2/3, Co atoms randomly occupy octahedral sites. In the case of x ≥ 2/3, Co atoms also occupy tetrahedral sites. The thermoelectric properties of CoxTiS2 (0 ≤ x ≤ 2/3) have been systematically investigated. A figure of merit of 0.3 is obtained for CoxTiS2 (0.04 ≤ x ≤ 0.08) at 573 K, a 25 % improvement over pristine TiS2 and one of the highest figures of merit reported for a n-type sulphide at such a low temperature. A series with composition MoxTi1-xS2 (0 ≤ x ≤ 0.09) was synthesised and consolidated by hot-pressing and spark plasma sintering. The ball milling conditions used and the SPS processing lead to a power factor of S2ρ-1 ≈ 2.1 mW m-1 K-2 at 323 K, which is the highest reported for TiS2. The reproducibility of the electrical properties of TiS2 for application at temperatures that do not exceed 473 K was confirmed by a heat-soak experiment. Nanoparticles of TiS2 with controlled morphologies were synthesised using a scalable solution-based route. Measurements of the Seebeck coefficient were carried out on TiS2 nanoflakes and a stoichiometry of Ti1+xS2 where x lies within the range 0.02 ≤ x ≤ 0.03 was determined. The thermoelectric properties of TiS2 nanocomposites with TiO2 nanoparticles and carbon nanotubes (CNTs) were investigated. A 12 % increase in the figure of merit over that of pristine TiS2 at 573 K, ZT = 0.28, is observed for the nanocomposite with 0.5 vol% of nano-TiO2. Two families of copper sulphides were investigated for their promising p-type thermoelectric properties, tetrahedrites, including Cu12-xMnxSb4S13 (x = 0; 1) and Cu12+ySb4S13 (x = 0; 0.3; 1; 1.5 and 2), and manganese-doped bornite, Cu5Fe1-xMnxS4 (0 ≤ x ≤ 0.10). A combination of powder X-ray and neutron diffraction analysis of Cu12-xMnxSb4S13 (x = 0; 1) located the Mn atoms on the Cu(1) site in tetrahedral position. Manganese substitution leads to a substantial reduction in thermal conductivity and the figure of merit is improved with ZT = 0.56 at 573 K. Structural analysis using powder X-ray diffraction and measurements of the thermoelectric properties of a series of copper-enriched tetrahedrite Cu12+ySb4S13 (y = 0.3; 1; 1.5; 2) were carried out. At room temperature, Cu12+ySb4S13 exists as a mixture of a Cu-poor and a Cu-rich phase and a change of behaviour in the electrical and thermal properties are observed over the temperature range 380 ≤ T / K ≤ 410. An increase in the Seebeck coefficient and a simultaneous decrease in the thermal conductivity are responsible for a large improvement in the figure of merit with ZT = 0.62 at 573 K for Cu12+ySb4S13 (y = 1.5 and 2). A manganese-doped bornite series of composition Cu5Fe1-xMnxS4 (0 ≤ x ≤ 0.10) was successfully synthesised using mechanical alloying. This synthetic method offers a simplified route to large scale production as well as better thermoelectric performance (ZT = 0.55 at 543 K) compared with samples prepared by solid-state reaction (ZT ≈ 0.4 at 573 K). Structural investigation was carried out using a combination of powder X-ray diffraction and SEM/EDX analysis and the solubility limit of Cu5Fe1-xMnxS4 was determined to be x = 0.1. Repeated electrical measurements and thermogravimetric analysis were carried out in order to investigate the stability of the samples and the reproducibility of the electrical properties as a function of time and temperature.

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.

Over the last decade, increasing numbers of multi-national corporations, public institutions and individual property owners have become interested in installing solar photovoltaics and small wind turbines. To best inform this broad range of actors, this research aims to assess the financial viability of such investments across broad city regions whilst maintaining accuracy at individual properties. Publicly available digital representations of urban surfaces are central to meeting this aim because they can be used to assess the area, slope and orientation of potential solar photovoltaic (PV) installation sites and to define how vertical wind profiles are altered by urban areas. A first study utilised digital surface models (DSMs) across seven UK cities to assess the roof spaces available for solar PV and also incorporated socio-economic factors to define the propensity for cities to install the technology. Despite changes to financial incentives that had recently occurred, the technologies remained viable at a very large number of locations and could theoretically meet large percentages (16% to 43%) of the cities’ electricity demands. The accuracy of slope, orientation and available area estimation in roof geometry modelling was then improved through the development of a neighbouring buildings method. In 87% of 536 validated results, the method identified the correct roof shape and roof slope was estimated to a mean absolute error of 3.76° when compared to 182 measured roofs. Work was then undertaken to improve solar insolation modelling. A radiative transfer model was created that incorporated shading based on DSM data. It estimated the power output of 17 solar PV installations across four UK cities with +2.62% mean percentage error when its 2013 insolation estimates were converted to power outputs using a 0.8 performance ratio. The validation data showed that the RTS model outperformed the market-leading esri ArcMap solar radiation software which incurred a -15.97% mean percentage error. This method was then adapted to be deployable on a city scale and predicted solar insolation with a mean percentage error of -4.39% despite the process being made far more computationally efficient. A method to estimate long-term average wind speeds for urban areas was then developed that produced results of comparable accuracy to an existing model but with considerably reduced computational demand and complexity in deployment. The mean absolute error inwind speed estimation was just 1.75% greater using the simplified methodology than the existing model. Finally, the improved modelling of roof geometries, solar insolation and long-term mean wind speed were brought together to evaluate the city-scale potential for solar PV and small to medium wind microgeneration. The research has shown that wind and solar PV microgeneration at sites that pay back within nine years could theoretically meet 88.5% of annual domestic electricity demand in the city of Leeds, or would be the equivalent of providing electricity to 300,319 homes. Current financial contexts were used to define a baseline scenario from which hypothetical changes to a variety of factors influencing microgeneration viability were investigated. When the costs and revenues were defined from a pessimistic, but still realistic, perspective the percentage of the study area’s electricity demand that could theoretically be met by wind and solar PV microgeneration fell to 0.1%. This suggests that government policy will continue to play a key role in the future growth of UK wind and solar PV deployment.

No description available.

An air-cooled synchronous generator has been studied for its airflow and thermal analysis using Computational Fluid Dynamics (CFD) in conjunction with experimental validation. Due to the temperature dependent resistive losses in the machine’s windings, any improvement in cooling provides a direct reduction in losses and an increase in efficiency. In addition, detailed modelling of machine windage components provide an insight for efficiency savings associated with airflow. A full 3600 CFD generator model has been constructed including all major solid components in conjunction with the air fluid regions. Fluid flow, turbulence, rotation as well as conjugate heat transfer modelling was solved throughout the whole machine. The CFD model has been validated using experimental testing on a newly commissioned rig. Airflow parameters of mass flow rate, torque and local pressures were used to validate the fluid modelling. Thermal measurements of stator temperature, heat flux and calculated heat transfer coefficients were used to validate the conjugate heat transfer modelling of the machine. Temperature distributions throughout all machine components were analysed. These have been explained by detailed analysis of local surface heat transfer coefficients and the associated local air flow structures which determine them. This versatility and detail in the thermal analysis of electrical machines is unique to CFD modelling. The methodology presented in this thesis demonstrates advancement in the scale and complexity for CFD analyses of electrical machines. This level of cooling scrutiny will enable informed design developments for step changes in machine efficiency.

A novel direct control of high performance bidirectional quasi-Z-source inverter (HPB-QZSI), with optimized controllable shoot-through insertion, to improve the voltage gain, efficiency and to reduce total harmonic distortion is investigated. The main drawback of the conventional control techniques for direct current to alternating current (DC-AC) conversion is drawn from the multistage energy conversion structure, which implies complicated control, protection algorithms and reduced reliability due to the increased number of switching devices. Theoretically, the original Z-source, Quasi-Z-source, and embedded Z-source all have unlimited voltage gain. Practically, however, a high voltage gain (>2 or 3), will result in a high voltage stress imposed on the switches. Every additional shoot-through state increases the commutation time of the semiconductor switches, thereby increasing the switching losses in the system. Hence, minimization of the commutation time by optimal placing of the shoot-through state in the switching time period is necessary to reduce the switching loss. To overcome this problem, a combination of high performance bidirectional quasi-Z-source inverter with a sawtooth carrier based sinusoidal pulse width modulation (SPWM) in simple operation condition for maximum boost control with 3rd harmonic injection is proposed. This is achieved by voltage-fed quasi-Z-source inverter with continuous input current, implemented at the converter input side which can boost the input voltage by utilizing the extra switching state with the help of shoot-through state insertion technique. This thesis presents novel control concepts for such a structure, focusing mainly on the control of a shoot-through insertion. The work considers the derivation and application of direct controllers for this application and scrutinizes the technical advantages and potential application issues of these methodologies. Based on the circuit analysis, a small signal model of the HPB-QZSI is derived, which indicates that the circuit is prone to oscillate when there is disturbance on the direct current (DC) input voltage. Therefore, a closed-loop control of shoot-through duty cycle is designed to obtain the desired DC bus voltage. The DC-link boost control and alternating current (AC) side output control are presented to reduce the impacts of disturbances on loads. The proposed strategy gives a significantly high voltage gain compared to the conventional pulse width modulation (PWM) techniques, since all the zero states are converted into shoot-through states. The simulated results verify the validity and superiority of the proposed control strategies.

No description available.

Series compensation techniques can be very effective when applied in an electrical network to increase the power transfer capacity of existing power lines. Distributed Static Series Compensation (DSSC) is a power electronics based series compensation scheme in which a DSSC device comprises of a single-phase H-bridge voltage source converter, a dc link capacitor and a low pass filter suspended from the power line via a single turn transformer. The application of DSSC in the 11kV distribution network is investigated in this thesis. This is followed by a study of existing control strategies employed in DSSC and Static Synchronies Series Compensation (SSSC) schemes. Most of these controllers are based on dq transformation methods in which balanced conditions are assumed and zero sequence currents are assumed to be negligible. While this might be a reasonable assumption at transmission level voltages, but it can be argued that in the presence of unbalanced loads and currents (a common feature of lower voltage distribution networks) these strategies can be inaccurate, leading to the wrong amount of compensation being injected. In addition some of the studied controllers are based on the 90° phase shift of line current. Practically, the injection angle must be slightly different in order to compensate the internal losses of the DSSC. The need for the diversion from the 90° can change over the time and this can threaten the stability of the system. A new single-phase control strategy based on the instantaneous power exchange between the DSSC devices and each of the three phase conductors is proposed in this thesis to address this issue. The new control method does not employ a dq transformation and is immune from the probable errors resulting from the presence of unbalanced network conditions. In the same time the injection angle is not fixed and it is adjusted by the controller. The operation of DSSC can be categorized in two modes and transfer function of system is obtained based on these two modes. The transfer function is used in the design of controller. This is followed by analyzing immunity of the designed controller against change of system parameters. The proposed scheme is simulated (using PSCAD software) to examine the operation of the new control method and the resulting impact on the 11kV distribution feeder, including the ability to divert power from one line to another and the ability to improve network voltage profiles. Performance of DSSC using the proposed controller is compared with performance of DSSC when the traditional controllers are employed.

The UK government has a target of achieving an 80% reduction in CO2 emissions with respect to the values from 1990 by 2050. Therefore, renewables based distributed generations (DGs) coupled with substantial electrification of the transport and heat sectors though low carbon technologies (LCTs), will be essential to achieve this target. The anticipated proliferation of these technologies will necessitate major opportunities and challenges to the operation and planning of future distribution networks. Smartgrid technologies and techniques, such as energy storage systems (ESSs), demand side response (DSR) and real time thermal ratings (RTTRs), provide flexible, economic and expandable solutions to these challenges without resorting to network reinforcement. This research investigates the use of ESS and DSR in future distribution networks to facilitate LCTs with a focus on the management and resolution of thermal constraints and steady state voltage limit violation problems. Firstly, two control schemes based on sensitivity factors and cost sensitivity factors are proposed. Next, the impacts of a range of sources of uncertainties, arising from existing and future elements of the electrical energy system, are studied. The impacts of electric vehicle charging are investigated with Monte Carlo simulation (MCS). Furthermore, to deal with uncertainties efficiently, a scheduling scheme based on robust optimization (RO) is developed. Two approaches have been introduced to estimate the trade-off between the cost and the probability of constraint violations. Finally, the performance of this scheme is evaluated. The results of this research show the importance of dealing with uncertainties appropriately. Simulation results demonstrate the capability and effectiveness of the proposed RO based scheduling scheme to facilitate DG and LCTs, in the presence of a range of source of uncertainties. The findings from this research provide valuable solution and guidance to facilitate DG and LCTs using ESS, DSR and RTTR in future distribution networks.

The aerospace industry is moving towards more-electric aircraft. These electrical systems are lighter, more efficient and smaller compared to present hydraulic system. The permanent magnet electrical machine is an obvious choice to replace these hydraulic systems because of its high torque density. However for low speed applications, there are significant drag torque issues which negate some of the machine’s advantages. This thesis introduces a new permanent magnet assisted segmental rotor switched reluctance machine. This machine was first designed by using finite element software and then compared with a conventional segmental rotor switched reluctance machine, showing an increase in torque by increasing non saturation region of the stator lamination. On this basis, a fault-tolerant permanent magnet assisted segmental rotor switched reluctance machine was designed to replace the permanent magnet machine used in the nose wheel of the aeroplane. Both two-dimensional and three-dimensional finite element analyses were conducted to analyse the dominant end winding effect. Fault-tolerant segmental rotor switched reluctance machine was then built and tested. Static analysis was conducted to determine current-flux linkage and torque values at different rotor angles. The machine was analysed both with and without magnets to assess the effect of reverse magnetization with magnets at the stator tooth tip. Mutual inductance was also found using the same test rig. Dynamic testing of the machine was done to determine open circuit voltage and short-circuit current. A novel pot core permanent magnet assisted inductor was also designed and compared with the conventional E-E core inductor. The magnet was used to hold the inductance of the inductor constant for high current values and the inductor was then tested to determine magnet’s effect. A dominant inner loop effect was found which was proved by doing several tests. Various recommendations were made to further improve overall performance of the machine and the inductor.