STATCOM (STATic COMpensator) has been used in electrical power systems as a shunt-connected compensator for voltage support and to improve power quality. Compared with the conventional compensators such as the synchronous condenser and the SVC (Static Var Compensator), the STATCOM has a faster speed of response to deal with dynamic and transient impacts. Although the STATCOM is capable of reactive power support to improve power quality, the ability to support real power is limited due to the insufficient energy storage capability of the conventional DC-link capacitor. Therefore, the application of the STATCOM to improving power system stability has been limited. This thesis proposes a solution to enhance the performance of the STATCOM by adding supercapacitor energy storage to the DC-link of the conventional STATCOM. With the fast charge/discharge characteristics of the supercapacitors, the enhanced STATCOM can absorb and inject real power to the ac power grid virtually instantaneously. The control design of the STATCOM based on a vector control strategy is presented, including the design of an instantaneous reactive power controller based on a small-signal model of the ac power system. The control design of the supercapacitor energy storage system (SCESS) based on small-signal models of the de-to-de converter is documented. The STATCOM and the SCESS are controlled together using a feed-forward control technique. In addition, this thesis also proposes that the enhanced STATCOM can be applied to reduce instability and tripping due to the rate of change of frequency (ROCOF) protection devices caused by large load impacts. The amount of the energy required for the enhanced STATCOM to maintain the stability of the system is also discussed.

In this study two new technologies are investigated with a view to improving the efficiency and reducing the vibrations of large electrical machines. These machines are used for high powered industrial applications. Individually controlled conductors are part of an active stator project that Converteam Ltd are developing. This involves individual conductors located in each stator slot that can be controlled separately. These replace traditional polyphase windings in order to provide a high level of control over the operation of the machine. A linear magneto-mechanical finite element model has been used here to show that this control method can enable a 44% reduction in the copper loss from the machine compared to a sinusoidal supply whilst maintaining the same operational torque. This method introduces extra Fourier harmonics into the excitation pattern supplied to the machine that are not present in the supply current to traditional polyphase windings. These extra harmonics utilize saliency advantageously to produce torque. However, they also increase the vibration of the stator and may increase hysteresis loss in the iron. The bimorph concept is an idea that is unique to this thesis. This concept involves individually controlled conductors positioned through the root of each stator tooth. On application of a relatively small power input to these conductors a magnetic field is created in the stator which distorts the existing magnetic field. Under certain conditions, the magnetostriction phenomenon causes the teeth to act as a vibration absorber. The effect of this method on the Maxwell forces in the air gap is small. This enables the cancellation of components of vibration of the stator whilst the machine torque is maintained. This is a vibration control method suited to high frequency vibrations where the deformed shape of the stator includes a rocking motion of the teeth and where the resultant stator vibrations lead to tonal noise emission from the machine. This thesis investigates the two technologies mentioned above with a magnetomechanical finite element model and two experimental investigations. The thesis also contains background information relevant to this study including an introduction to electrical machines and power electronics, noise radiation and sources of noise in electrical machines, finite element modelling, vibrations of electrical machines and vibration absorbers.

In Indirect Field Orientation (IFO) of induction motors, the interest for parameters identification has increased rapidly due to the great demand for high performance drives and more sophisticated control systems that have been made possible by the development of very powerful processors, such as floating point DSPs. Accurate knowledge of the machine electrical parameters is also required to ensure correct alignment of the stator current vector relative to the rotor flux vector, to decouple the flux - and torque - producing currents and to tune the current control loops. The accuracy and general robustness of the machine is dependant on this model. Artificial intelligent technologies have been tested in the field of electro mechanics like neural networks, fuzzy logic, simulated annealing and genetic algorithms. These methods are increasingly being utilised in solving electric machine problems. This thesis addresses a novel non - intrusive approach for identifying induction motor equivalent circuit parameters based on experimental transient measurements from a vector controlled Induction Motor (I.M.) drive and using an off line Genetic Algorithm (GA) routine with a linear machine model. The evaluation of the electrical motor parameters at rated flux operation is achieved by minimising the error between experimental responses (speed or current) measured on a motor drive and the respective ones obtained by a simulation model based on the same control structure as the experimental rig. An accurate and fast estimation of the electrical motor parameters is so achieved. Results are verified through a comparison of speed, torque and line current responses between the experimental IM drive and a Matlab - Simulink model. The second part of the research work introduces a new approach based on heuristic optimisation for identifying induction motor electrical parameters under different operating conditions such as different load and flux levels. Results show via interpolation test the effect of the most important electrical parameters, the magnetising inductance Lm and rotor resistance Rr, at each different operating condition.

High performance variable speed induction motor drives have been commercially available for industrial applications for many years. More recently they have been proposed for applications such as hybrid automotive drives, and some pump applications on more electric aircraft. These applications will require the drive to operate in the presence of faults i.e. they must be “Fault Tolerant” and be capable of “Fault Ride Through”. The aim of this project was therefore to investigate fault ride through control strategies for induction motor drives, particularly with respect to open circuit winding or power converter faults. Three objectives were identified and addressed to meet this aim. a) A new simulation model for an induction motor was created which reflects both saturation and space harmonics effects within the drive under both symmetric (healthy) and asymmetric (faulted) conditions. The model has a relatively low computational requirement to allow it to be used in conjunction with the simulation of high performance control algorithms and power electronic equipment. For operation in both healthy and faulty conditions, comparisons show that the simulated saturation and space harmonic effects match those obtained from an experiment system. Therefore this model is a very useful tool for the development and optimisation of new control strategies for fault tolerant drive systems. b) A novel on-line fault detection and diagnosis algorithm based on the measurement of the third harmonic component in the motor line currents has been proposed. The location of the open circuit fault is detected based on detecting a magnitude reduction for the third harmonic component of the current flowing to the motor terminals, and can be implemented in real time to give a fast response with little additional computational overhead. c) A new open circuit fault tolerant control strategy has been designed for a delta connected induction machine suddenly affected by an open circuit winding fault. The fault ride through is achieved without any modification to either the power converter or the motor circuit. A novel feedforward compensation algorithm is introduced which considerably reduces the current and the torque ripple in the faulted drive motor. Two methods for controlling the neutral point voltage are also presented so that the available voltage capacity of the inverter is maximised in both normal and fault mode. For high speed operation, two different methods for field weakening control are presented, so that the available voltage capacity is maximized in both normal and fault mode. This thesis describes the theoretical derivation of the new models and algorithms, and presents experimental results from a 4kW laboratory prototype to validate the proposals. The full fault tolerant system is experimentally demonstrated on a delta connected machine which suffers an open circuit winding fault. The improved motor performance under fault conditions is clearly seen.

The performance of piezoelectric cantilever beam energy harvesters subjected to base excitation is considered in this work. Based on the linear assumption, a theoretical model is developed to predict the mechanical and electrical responses of the harvester and in comparison to other theoretical models, more accurate mode shape functions are used for the structural part of the harvester. The model is validated against experimental measurements and parameter studies are carried out to investigate the maximum power output in different situations. In some applications, like powering tyre pressure monitoring sensors (TPMS), energy harvesters are subjected to large amplitude shocks and high levels of acceleration, which can cause large bending stresses to develop in the beam, leading to mechanical failure. In this work, a bump stop is introduced in the energy harvester design to limit the amplitude of vibration and prevent large amplitude displacement. A theoretical model is developed to simulate the energy harvester impacting a stop, and the model is used to investigate how the electrical output of the harvester is affected by the stop. The work demonstrates how the model can be used as a design tool for analysing the compromise between the electrical output and structural integrity. Nonlinear behaviour of the energy harvester is observed to have a significant effect on the resonance frequencies when the harvester is subjected to large amplitude base accelerations. To correctly predict the behaviour of the harvester, piezoelectric material nonlinearity and geometric nonlinearity are incorporated in the theoretical model. It is found that the nonlinear softening effect is dominated by the material nonlinearity, while the geometric nonlinearity is less significant. The nonlinear energy harvester model is used in conjunction with the bump stop and results obtained using the linear and nonlinear models are compared to experimental measurements to investigate the importance of using a nonlinear model. The inclusion of nonlinear behaviour is shown to improve significantly the accuracy of predictions under some circumstances. The energy harvester models developed in this work are used to simulate the electrical power generated in a TPMS application, where the harvester embedded in the tyre is subjected to large radial accelerations as the tyre rolls along the road. The simulated results are compared to reported experimental work and agreement is found between the results.

Detecting and dealing with congestion in delay tolerant networks is an important and challenging problem. Current DTN forwarding algorithms typically direct traffic towards particular nodes in order to maximise delivery ratios and minimise delays, but as traffic demands increase these nodes may become unusable. This thesis proposes Café, an adaptive congestion aware framework that reduces traffic entering congesting network regions by using alternative paths and dynamically adjusting sending rates, and CafRep, a replication scheme that considers the level of congestion and the forwarding utility of an encounter when dynamically deciding the number of message copies to forward. Our framework is a fully distributed, localised, adaptive algorithm that evaluates a contact’s next-hop potential by means of a utility comparison of a number of congestion signals, in addition to that contact’s forwarding utility, both from a local and regional perspective. We extensively evaluate our work using two different applications and three real connectivity traces showing that, independent of the network inter-connectivity and mobility patterns, our framework outperforms a number of major DTN routing protocols. Our results show that both Café and CafRep consistently outperform the state-of-the-art algorithms, in the face of increasing traffic demands. Additionally, with fewer replicated messages, our framework increases success ratio and the number of delivered packets, and reduces the message delay and the number of dropped packets, while keeping node buffer availability high and congesting at a substantially lower rate, demonstrating our framework’s more efficient use of network resources.

Matrix converter is a direct AC-AC converter topology that is able to directly convert energy from an AC source to an AC load without the need of a bulky and limited lifetime energy storage element. Due to the significant advantages offered by matrix converter, such as adjustable power factor, capability of regeneration and high quality sinusoidal input/output waveforms, matrix converter has been one of the AC – AC topologies that receive extensive research attention for being an alternative to replace traditional AC-DC-AC converters in the variable voltage and variable frequency AC drive applications. Multilevel matrix converter is an emerging topology that integrates the multilevel concept into the matrix converter topology. Having the ability to generate multilevel output voltages, the multilevel matrix converter is able to produce better quality output waveforms than conventional matrix converter in terms of harmonic content, but at the cost of higher number of power semiconductor device requirement and more complicated modulation strategy. In this research work an indirect three-level sparse matrix converter is proposed. The proposed converter is a hybrid combination between a simplified three-level neutral-point-clamped voltage source inverter concept and an indirect matrix converter topology. This multilevel matrix converter topology has a simpler circuit configuration and is able to generate three-level output voltages, making this topology an attractive option in industrial applications. In this work a comprehensive simulation study is carried out to investigate the operation of the proposed converter. The performance of the proposed converter is compared to the conventional indirect matrix converter topology and a multilevel neutral-point- clamped matrix converter in order to identify the advantages and disadvantages offered by the proposed converter. A study of the semiconductor losses in the indirect three-level sparse matrix converter is also included. Finally, the operation of the proposed converter is experimentally validated using a laboratory prototype.

Today the majority of UK's energy needs are met by fossil fuels. An energy sector that uses 30% of this energy and generates 28% of the total emissions is domestic sector. To reduce the emissions generated by fossil fuels UK government decided to increase the energy coming from renewable sources by 2020. A renewable energy that can contribute is solar energy. Solar thermal collectors and photovoltaics are two means of transforming solar energy to thermal and electrical energy. The limited space in the roofs and the cost of the technologies will prevent families to use both systems together in their roof A hybrid energy system combine the use of two or more alternative power sources will help to increase the system's total efficiency. The photovoltaic/thermal (PV/T) system is a hybrid structure that converts part of the sun's radiation to electricity and part to thermal energy. This research work focuses on the production of new approaches on hybrid PV/T systems. PV/T systems using water and air have been introduced and a literature review conducted in order to identify positives and negatives of these systems. Experiments also conducted by using water and air as heat transfer medium, and the results helped to work as a benchmark performance to the new approaches. These technologies were heat pipes, phase change materials and micro encapsulated phase change materials. The technologies exist for years but their use in the specific application is new. A literature review was undertaken to provide an understanding of these technologies and identified findings that have contributed to the design of the systems. Experimental work was carried out incorporating these technologies in the rear of a PV and the results indicated comparable performance with PV/T-water and PV/Tair. Five performance indicators were employed to help with the comparison of the systems. These were electrical and thermal efficiency, the total energy efficiency, the primary energy saving efficiency and the exergy efficiency. From these five indicators the primary energy saving efficiency that shows how much fossil fuel is saved and the exergy efficiency that could give the optimum working conditions of each system was the most valuable ratings. For the PV/PCM model a new simulation program was developed to help validate the experimental work. Also an environmental and economic study was undertaken to compare if the new systems could help reduce the C02 emissions and if they were feasible to become commercial products. Finally the conclusions gained have been presented and recommendations fo r future work have been made.

This thesis investigates space-vector modulation and one-dimensional modulation applied to the cascaded H-bridge multi-level converter as a model for one port of the UNIFLEX-PM power converter system. The UNIFLEX-PM converter is a modular system including galvanic isolation at medium frequency intended to replace transformers in future distribution and transmission systems. Power converters in this application must produce good quality voltage waveforms with low power loss. In this work, modulation methods are developed using theoretical analyses and simulation studies, before being verified experimentally using a low voltage, laboratory-based power converter operating at the low switching frequencies applicable to high-power applications. Using space-vector modulation, the relationship between the phase of the sampling process and the distortion of the line voltages is used to reduce the harmonic distortion of the output voltages. Different loads are attached to the cells of the cascaded H-bridge converter and limits are derived determining the range of loads for which it is possible to equalize the capacitor voltages. An algorithm which uses redundant states to balance the capacitor voltages without increasing the switching frequency is applied to space-vector modulation and one-dimensional modulation and its performance is compared to the derived limits. The geometrical effect of capacitor voltage ripple on the space-vector diagram is used to derive the influence on the spectrum of the line-voltages. It is identified that second and fourth harmonics of the capacitor voltages contribute to fifth and seventh harmonics of the line voltages. A feed-forward scheme to compensate for the ripple of the capacitor voltage is derived and is shown to reduce the magnitude of un-wanted harmonics. All the methods developed in this thesis can be applied to converters with any number of cells.

The research described in this thesis focuses on the mitigation of inter-turn short-circuit (SC) faults in Fault tolerant Permanent Magnet (FT-PM) machines. An analytical model is proposed to evaluate the inter-turn SC fault current accounting for the location in the slot of the short-circuited turn(s). As a mitigation strategy to SC faults at the design stage, a winding arrangement called VSW (Vertically placed Strip Winding) is proposed and analysed. The proposed analytical model is benchmarked against finite element (FE) calculation and validated experimentally. The results demonstrate that the proposed winding arrangement in the slot improves the fault tolerance (FT) capability of the machine by limiting the inter-turn SC fault current regardless the fault location in the slot. Electromagnetic and thermal studies are conducted to verify the merits and drawbacks of the proposed winding compared to the conventional winding using round conductors (RCW). The study shows that the proposed winding scheme, in addition to being fault-tolerant, has an improved bulk radial conductivity, can achieve a good fill factor, but has a significantly higher frequency-dependent AC copper loss. To predict the AC losses an analytical model based on an exact analytical 2D field solution is proposed. This model consists of first solving the two-dimensional magneto-static problem based on Laplace’s and Poisson’s equations using the separation of variables technique. Then, based on that solved solution, by defining the tangential magnetic field (Ht) at the slot opening radius, Helmholtz’ equation is solved in the slot sub-domain. Subsequently, an FE and MATLAB® coupled parametric design is undertaken to maximise the VSW wound machine’s efficiency whilst maintaining its FT capability. The proposed analytical models for prediction of the SC fault current and AC copper losses are integrated into the coupled optimisation. It is shown that the effective losses of the VSW can be minimised through the parametric design while maintaining the required level of machine performance. Using an existing FT-PM machine of which the rotor is kept unchanged two stators were designed, manufactured and wound with RCW and VSW respectively and experimental tests are carried out to validate the analytical models and the new winding concept.