Submarine high voltage (HV) cables are becoming increasingly important to modern power transmission strategies. There has been a large amount of recent investment in projects such as offshore wind farms and international "megagrid" initiatives, of which submarine HV cables are essential components. A lot of research has been carried out into the thermal behaviour of HV cables buried on land. However, the performance of submarine HV cables has not been investigated extensively, despite several key differences between the two respective environments. The amount of power that can be transmitted along an HV cable is often limited thermally by the maximum operational temperature of the cable components. It is therefore crucial to understand how heat is dissipated from HV cables as comprehensively as possible to ensure reliable, economical, and efficient deployment of these assets. 2D finite element method (FEM) simulations have been developed to examine the impact that certain environmental parameters have on the dissipation of heat generated within submarine HV cables into the surrounding burial sediment. Both conductive and convective heat transfer mechanisms are considered by solving coupled heat and fluid flow equations in a representative geometrical framework. The implications of some realistic inhomogeneous burial scenarios are considered, as is the impact of environmental conditions on cable temperature response times. The FEM model suggests that the most influential environmental factor in determining the nature of the heat flow around submarine HV cables is a quantity called the intrinsic permeability. For sediments with a high permeability, convection can make a significant contribution to the overall transfer of heat from submarine HV cables into the surrounding environment, despite being neglected by traditional techniques for assessing heat flow around cables buried on land. Under these circumstances, cable temperatures are typically lower than for low permeability sediments. Consideration of the additional cooling effect provided by convective heat transfer in these situations may result an increased cable current carrying capacity, or the potential to reduce the amount of conductor material required for manufacture.

Large-scale investment in transmission and distribution power networks is planned over the next decades to meet the future demand and changes in power generation. Nevertheless, it is still of a great importance that the existing assets continue to operate reliably and their health is maintained. Moreover, the failures of distribution cables are extremely disruptive, costly to repair and have a serious impact on customer confidence. As a result, developing a reliable on-line prognostic tool is of a great importance. This research investigates a method of developing a prognostic capability for evaluation of the health and long term performance of aging distribution cable circuits. Developing such prognostic models will significantly improve the prognosis accuracy, allowing the targeting of maintenance and reduction of in-service failures. Real-time measurements taken close to underground cables can update the models giving a more accurate prognostic tool. The aging of the cable begins the moment it is installed and put in service due to a combination of mechanical, thermal, electrical and environmental factors. A thermal prognostic model is suggested. It enables prediction of the likely temperature impact on underground cable joints at the burial level and terminations according to weather conditions and known loading. Anomalies of temperature measurements along the cable compared to predicted temperatures will indicate the possible degradation activity in the cable. An experimental surface trough has been set up where operation of power cables was simulated with control system which is able to model any cable loading. The surface temperature of the cable is continuously monitored as well as the weather conditions such as solar radiation, wind speed, humidity, rainfall and air temperature. The research involved cooperation with University of Cyprus and the Electricity Authority of Cyprus which has given an opportunity to implement, install and study the performance of the condition monitoring thermal prognostic model in a distribution network with different environmental and loading conditions than found in the UK.

DC fault protection is one challenge impeding the development of multi-terminal dc grids. The absence of manufacturing and operational standards has led to many point-to-point HVDC links built at different voltage levels, which creates another challenge. Therefore, the issues of voltage matching and dc fault isolation in high voltage dc systems are undergoing extensive research and are the focus of this thesis. The modular multilevel design of dual active bridge (DAB) converters is analysed in light of state-of-the-art research in the field. The multilevel DAB structure is meant to serve medium and high voltage applications. The modular design facilitates scalability in terms of manufacturing and installation, and permits the generation of an output voltage with controllable dv/dt. The modular design is realized by connecting an auxiliary soft voltage clamping circuit across each semiconductor switch (for instance insulated gate bipolar transistor – IGBT) of the series switch arrays in the conventional two-level DAB design. With auxiliary active circuits, series connected IGBTs effectively become series connection of half-bridge submodules (cells) in each arm, resembling the modular multilevel converter (MMC) structure. For each half-bridge cell, capacitance for quasi-square wave (quasi two- level) operation is significantly smaller than typical capacitance used in MMCs. Also, no bulky arm inductors are needed. Consequently, the footprint, volume, weight and cost of cells are lower. Four switching sequences are proposed and analysed in terms of switching losses and operation aspects. A design method to size converter components is proposed and validated. Soft-switching characteristics of the analysed DAB are found comparable to the case of a two-level DAB at the same ratings and conditions. A family of designs derived from the proposed DAB design are studied in depth. Depending on the individual structure, they may offer further advantages in term of installed semiconductor power, energy storage, conduction losses, or footprint. A non-isolated dc-dc converter topology which offers more compact and efficient station design with respect to isolated DAB – yet without galvanic isolation – is studied for quasi two-level (trapezoidal) operation and compared to the isolated versions. In all the proposed isolated designs, active control of the dc-dc converter facilitates dc voltage regulation and near instant isolation of pole-to-pole and pole-to-ground dc faults within its protection zone. The same can be achieved for the considered non-isolated dc-dc converter topology with additional installed semiconductors. Simulation and experimental results are presented to substantiate the proposed concepts.

Multiterminal High Voltage Direct Current (HVDC) systems are anticipated to enable flexible transmission of renewable energy across continents, however protection strategies for such systems are in their infancy. Fast circuit breakers have been proposed, however their implementation on a network is undecided. The protection requirements are likely to depend on both the AC/DC converter and network topologies. This thesis examines several aspects of the protection of HVDC systems, considering factors influencing circuit breaker ratings, and examining primary and backup protection philosophies. A comparison is made throughout between fault feeding and fault blocking converters, aiming to investigate the impact of fault blocking converters on HVDC protection. The ratings of HVDC circuit breakers on meshed networks are evaluated, considering factors such as the current breaking magnitude, operation time, additional inductance and energy dissipation requirements for various network and converter topologies. It is shown that the circuit breaker requirements in network areas with fault blocking converters are reduced compared to when a fault feeding converter is implemented. HVDC protection is investigated, considering primary and backup methods for several protection philosophies. The influence of the converter, network and circuit breaker topologies are examined, and the impact of the HVDC protection strategy on the connected AC systems is evaluated. It is shown that slower protection strategies using fault blocking converters might be technically feasible from both the HVDC and AC system perspectives, which could result in an effective protection strategy with a reduced requirement for circuit breakers.

No description available.

Cascaded DC/DC converter systems (or 'cascaded systems') have instability problems; i.e., although the subsystems may work well individually, the whole system may be instable due to the impedance interaction among these subsystems. To solve this problem, a family of impedance-based stabilisation methods are proposed in this thesis. First, parallel-virtual-impedance (PVI) and series-virtual-impedance (SVI) control strategies are proposed to stabilise cascaded systems via shaping the load input impedance. Theoretically, the PVI or SVI control strategy connect a virtual impedance in parallel or series with the input port of the load converter so that the magnitude or phase of the load input impedance can bemodified within a very small frequency range. Therefore, with the PVI or SVI control strategy, the cascaded system can be stabilised with minimal compromise of the load performance. Then, based on the PVI and SVI control strategies, adaptive-PVI (APVI) and adaptive-SVI (ASVI) control strategies are proposed by introducing an adaptive mechanism to change the impedance. With the APVI or ASVI control strategy, the load converter can be stably connected to different source converters without changing its internal structure. It is also shown that the ASVI control strategy is better than the APVI control strategy and can make the cascaded system more stable. Moreover, a minimum-ripple-point-tracking (MRPT) controller is proposed and utilised to solve the potential problem of the ASVI control strategy. Finally, a source-side SVI (SSVI) control strategy and a VRLC damper are proposed to stabilise the cascaded system with better source performance or input filter performance, respectively. All the proposed stabilisation and control methods are validated by extensive experimental results.

No description available.

No description available.

This research study focuses on the dielectric response of XLPE model power cables that have combinations of homo- and co-polymer insulation with furnace and acetylene carbon black semicon shields. Three dielectric spectroscopy techniques, which are frequency response analyzer and transformer ratio bridge in both frequency domain, and charging/discharging current system in time domain, were jointly used to measure the low loss XLPE cables in the frequency range from 10-4Hz to 104Hz at temperatures from 20°C to 80°C. Degassing effects and thermal ageing effects have also been studied with the spectroscopy techniques. Thermal-electric behaviour and maximum voltages for thermal breakdown have been theoretically simulated for the model cables. Three loss origins of the XLPE cables have been found with different loss mechanisms. Conduction loss due to thermally activated electron/hole hopping dominates the lower frequency range from 10-4Hz to 1Hz; Semicon loss due to its in series resistance with the insulation layer in cable equivalent circuit dominates the higher frequency range from 102Hz to 104Hz; intrinsic polarization loss of the XLPE insulation has dominant flat loss spectra in the mid-frequency range from 1Hz to 102Hz. Degassing was found to decrease the conductivity of the model cables, while thermal ageing greatly increased the conductivity. Thermal-electric simulation results with FEMLAB have shown that the position of maximum field changes from inner to outer insulation boundary under higher applied voltages. A loss mechanism model with mathematical expression for dielectric loss spectrum calculation is finally proposed to explain the total dielectric loss of polymer power cables.

Switched mode DC-DC converters exhibit a variety of complex behaviours in power electronics systems, such as sudden changes in operating region, bifurcation and chaotic operation. These unexpected random-like behaviours lead the converter to function outside of the normal periodic operation, increasing the potential to generate electromagnetic interference degrading conversion efficiency and in the worst-case scenario a loss of control leading to catastrophic failure. The rapidly growing market for switched mode power DC-DC converters demands more functionality at lower cost. In order to achieve this, DC-DC converters must operate reliably at all load conditions including boundary conditions. Over the last decade researchers have focused on these boundary conditions as well as nonlinear phenomena in power switching converters, leading to different theoretical and analytical approaches. However, the most interesting results are based on abstract mathematical forms, which cannot be directly applied to the design of practical systems for industrial applications. In this thesis, an analytic methodology for DC-DC converters is used to fully determine the inherent nonlinear dynamics. System stability can be indicated by the derived Monodromy matrix which includes comprehensive information concerning converter parameters and the control loop. This methodology can be applied in further stability analysis, such as of the influence of parasitic parameters or the effect of constant power load, and can furthermore be extended to interleaved operating converters to study the interaction effect of switching operations. From this analysis, advanced control algorithms are also developed to guarantee the satisfactory performance of the converter, avoiding nonlinear behaviours such as fast- and slowscale bifurcations. The numerical and analytical results validate the theoretical analysis, and experimental results with an interleaved boost converter verify the effectiveness of the proposed approach.