This thesis describes new models and investigations into switching transient phenomena related to the shunt reactors and the Mechanically Switched Capacitor with Damping Network (MSCDN) operations used for reactive power control in the transmission system. Shunt reactors and MSCDN are similar in that they have reactors. A shunt reactor is connected parallel to the compensated lines to absorb the leading current, whereas the MSCDN is a version of a capacitor bank that has been designed as a C-type filter for use in the harmonic-rich environment. In this work, models have been developed and transient overvoltages due to shunt reactor deenergisation were estimated analytically using MathCad, a mathematical program. Computer simulations were then undertaken using the ATP/EMTP program to reproduce both single-phase and three-phase shunt reactor switching at 275 kV operational substations. The effect of the reactor switching on the circuit breaker grading capacitor was also examined by considering various switching conditions. The main original achievement of this thesis is the clarification of failure mechanisms occurring in the air-core filter reactor due to MSCDN switching operations. The simulation of the MSCDN energisation was conducted using the ATP/EMTP program in the presence of surge arresters. The outcome of this simulation shows that extremely fast transients were established across the air-core filter reactor. This identified transient event has led to the development of a detailed air-core reactor model, which accounts for the inter-turn RLC parameters as well as the stray capacitances-to-ground. These parameters are incorporated into the transient simulation circuit, from which the current and voltage distribution across the winding were derived using electric field and equivalent circuit modelling. Further analysis of the results has revealed that there are substantial dielectric stresses imposed on the winding insulation that can be attributed to a combination of three factors. (i) First, the surge arrester operation during the MSCDN energisation, which causes steep voltage change at the reactor terminal. (ii) Second, the nonuniform voltage distribution, resulting in high stresses across the top inter-turn windings. (iii) Third, the rapid rate-of-change of voltage in the assumed worst-case reactor winding location. This is accompanied by a high dielectric (displacement) current through the inter-turn winding insulation. Work within this investigation proposes that a synergistic effect of high electric field and high dielectric current occurring at worst energisation, followed by the thermal effects of steady state operation contributes to the failure of air-core reactors used on the 400 kV MSCDN.

District heating systems have the potential to contribute to the UK renewable energy targets. However, there are a number of economic barriers which would have to be addressed in order to make district heating competitive in comparison with alternative heating technologies. The objective of this research was to model and analyse a district heating network and develop an optimisation method to calculate the minimum capital investment, heat losses and pump energy consumption. Firstly, modelling and analysis of a district heating network was conducted to obtain district heating design cases with minimum annual total energy consumption, annual total exergy consumption and annualised cost. Then through the analysis, a two-stage programming model was developed which synthesised design and optimal operation of a district heating network. The optimisation was used to minimise annual total energy consumption, annual total exergy consumption or annualised cost of the heat network by selecting suitable pump and pipe sizes, taking into account different parameters such as target pressure loss, temperature regime and operating strategy. The optimisation technique was used to investigate two different case studies, with high and low heat density. In all cases, a variable flow and variable supply temperature operating method was found to be beneficial. Design cases with minimum annual total energy consumption and annualised cost used rather small pipe diameters and large pressure drops. To achieve the minimum annual total exergy consumption a design case with larger pipe diameters and smaller pressure loss was found to be desirable. It was observed that by reducing the water temperature and increasing temperature difference between supply and return pipes, the annual total energy consumption, annual total exergy consumption and the annualised cost were reduced. It was also shown that district heating in an area with high heat density is more energy efficient and cost effective.

Stop signal task of response inhibition, I find that response inhibition (stopping) is slowed in the presence of angry facial expressions, and such slowing is greater in individuals high in trait neuroticism. Further, as predicted, the influence of neuroticism is moderated by individual differences in emotion regulation, such that good emotion regulation ‘buffers’ the impact of neuroticism. The implications of these findings for current cognitive models of threat-processing are discussed.

The main goal of this research is to support concept designers’ search for inspirational and meaningful images in developing mood boards. Finding the right images has become a well-known challenge as the amount of images stored and shared on the Internet and elsewhere keeps increasing steadily and rapidly. The development of image retrieval technologies, which collect, store and pre-process image information to return relevant images instantly in response to users’ needs, have achieved great progress in the last decade. However, the keyword-based content description and query processing techniques for Image Retrieval (IR) currently used have their limitations. Most of these techniques are adapted from the Information Retrieval research, and therefore provide limited capabilities to grasp and exploit conceptualisations due to their inability to handle ambiguity, synonymy, and semantic constraints. Conceptual search (i.e. searching by meaning rather than literal strings) aims to solve the limitations of the keyword-based models. Starting from this point, this thesis investigates the existing IR models, which are oriented to the exploitation of domain knowledge in support of semantic search capabilities, with a focus on the use of lexical ontologies to improve the semantic perspective. It introduces a technique for extracting semantic DNA (SDNA) from textual image annotations and constructing semantic image signatures. The semantic signatures are called semantic chromosomes; they contain semantic information related to the images. Central to the method of constructing semantic signatures is the concept disambiguation technique developed, which identifies the most relevant SDNA by measuring the semantic importance of each word/phrase in the image annotation. In addition, a conceptual model of an ontology-based system for generating visual mood boards is proposed. The proposed model, which is adapted from the Vector Space Model, exploits the use of semantic chromosomes in semantic indexing and assessing the semantic similarity of images within a collection.

Pollution flashover constitutes the predominant parameter for the design,specification and dimensioning of high voltage outdoor insulation [1-2]. For over a quarter of a century, polymeric insulators have been used for overhead lines and substations due to their enhanced performance in highly polluted environments. This is mainly attributed to the hydrophobic properties of polymeric surfaces thatlead to a reduction of surface wetting and, thus, to a reduction of leakage current and discharge activity that could result to a flashover. However, under severe ambient conditions, discharge activity is not totally eliminated. The present investigations have demonstrated the development of dry bands and partial arcs on polymeric insulators (Chapter 3). While ‘cap and pin’ ceramic insulators - e.g. porcelain and glass - employ special anti-fog designs of increased leakage path introduced by the presence of deeper ribs, such profiles cannot be used for polymeric insulators due to moulding restrictions [1-1]. Yet, these moulding properties allow the fine texturing of the polymeric surface [1-7]. Textured insulators are a novel approach for the improvement of polymeric insulators using a surface design consisting of an array of hemispherical protuberances of various configurations and range of diameters. The objective is to reduce power dissipation (P = EJ) on the insulator surface by reducing the electric field gradient E and current density J (Chapter 2). Moreover, an increase of the longitudinal creepage path can be achieved without the need of increasing the overall length of the insulator. Where discharges do occur, the formation of parallel current paths could lead to less harmful discharges hence mitigating damage due to thermal stresses. An initial theoretical classification of the textured designs resulted to a shortlist of textured patterns that could be employed for the development of a full insulator prototype. At the next stage, these patterns were experimentally evaluated in a series of material tests in accordance to IEC-60587 inclined-plane test procedure [Chapter 5]. Rectangular silicone rubber samples with a plane surface were tested along with samples with a textured finish to assess the performance against erosion and tracking. The improvement introduced by these textured patterns in comparison with conventional non-textured samples was outstanding [2-48, 49]. The intersecting square pattern showed the best performance, hence it was chosen for the development of a full textured insulator. Conventional 11 kV silicone rubber insulators and textured insulators with the same shed profile were cast in-house using vacuum injection casting techniques. There are no international standards for the artificial testing of polymeric insulators. In this work, a proposed test procedure for non-ceramic insulators is described (Chapter 3). Clean-fog tests of conventional (Chapter 4) and textured insulators (Chapter 6), for a range of wetting and pollution severity conditions,showed that textured insulators showed an improved flashover performance that in some cases reached 26 %. Surface texturing seems to be more effective under severe conditions of wetting and pollution and when hydrophobicity is reduced, like the temporary loss of hydrophobicity observed for silicone rubber insulators in service.

The behaviour of earth electrodes at power frequency conditions is well known. Several studies are going on at present, to understand the behaviour of earthing systems at transient impulse and high frequency conditions. The study of impulse measurements in earthing systems was carried out during this project, to understand the soil electromagnetic behaviour towards high frequency and variable AC/DC/impulse current magnitudes. Several measurement techniques and instrumentation used for the measurement of the earthing systems were surveyed. The limitations and the advantages of each approach have been identified, and the range of application determined. Extensive experiments were performed on the practical earth electrodes at the Cardiff University test site at Llanrumney, and at the Dinorwig power station earthing facilities. These experiments have revealed that there is reduction of impedance of earth electrodes over the frequency range 20Hz to 120kHz. Moreover, a pronounced effect of DC current magnitude was observed on the earth resistance of the electrodes over the range of 1mA to 500mA. The numerical modelling of the test configurations did not show the reduction in earth impedance over the frequency range 20Hz to 120kHz. To understand the different trends shown by the experiments and simulation, and the effect of frequency and current magnitude, a geological literature survey was carried out. This survey revealed that when the soil water electrolyte solution is subjected to high frequency electrical currents, it exhibits conductivity dispersion phenomenon. Conductivity dispersion is a phenomenon where conductivity of the clay water electrolyte solution increases by about 30% over a frequency range 20Hz to 100kHz. The geological literature survey also revealed that the polarisation effect in the soil water electrolyte is responsible for the non-linear current effect. Moreover, during this project, a new technique of FFT from impulse, was proposed to measure the earth electrode impedance, over a frequency range which is an inherent component of the impulse signals. The FFT from impulse signals, showed a good agreement of the measured earth impedance of the earth electrodes, with the measured earth impedance using the variable frequency scan method. FFT from impulse technique has an advantage over the variable frequency scan method, from the point of view of the time required for the measurement and the simplicity of the test source, for the measurement of the earth electrode impedance. Such a technique, could have impact on the testing at high current magnitudes, where impulse generation is much easier. Finally, the future scope of work is presented to explore the measurement of earth electrode impedance above the frequency of 120kHz and current magnitudes above 5A.

In recent years aquatic vegetation has become more accepted and important in the river restoration schemes and preserving river ecology. The purpose of this thesis is to investigate the influence of emergent vegetation on velocity and turbulence fields in order to have a better understanding of the effect of vegetation on the transverse mixing processes. To achieve this objective, a series of experiments was conducted in an open channel flume with emergent rigid rods in both staggered and aligned arrangements. Detailed velocity, turbulence and dye tracer measurement were carried out for six vegetation densities relating to solid volume fractions (SVF) in the range 0.51 % to 7.79 %. In sparse vegetation (SVF < 10 %) as expected the normalised spatially-averaged longitudinal velocity reduces as the vegetation density increases with approximately 30 % to 50 % reduction when the solid volume fraction is doubled. Results indicated that in sparse vegetation, the normalised turbulence intensities increased with increasing solid volume fraction. The bulk drag coefficient increased with increasing vegetation density whilst decreased with increasing stem Reynolds number. The transverse mixing coefficient increases with both increasing vegetation density and stem Reynolds number. The current study showed that for sparse vegetation (SVF < 10%), the transverse mixing coefficient has a stronger correlation with turbulence intensity compared to transverse shear. Therefore indicating that within sparse vegetated flows, turbulence dominates over transverse shear in transverse mixing. In addition to that, transverse mixing also correlate better with double-averaged turbulence intensity compared to turbulent kinetic energy. This reflects that the turbulence in the longitudinal direction plays a greater contribution to the overall transverse dispersion than the contribution of the total turbulence in all three directions. Finally two vegetation transverse dispersions models proposed by other researcher for randomly distributed vegetation were tested against data from the current study. Both models were found to predict reasonably well.

The investigation reported in this thesis was prompted by the desire to develop an easy and reliable nondestructive method of testing to monitor changes in elastic properties of metals when they are subjected to creep, fatigue, and case hardening. The possibilities of the useful application of such a test will be invaluable to the aircraft industry. It was 'therefore suggested that the specimens used should be made from the type of materials Used in the aircraft industry. These were made of high strength alloys. The method of nondestructive testing was to excite ultrasonic surface waves on the surface of the specimens, and measuring the changes in surface wave velocity, when the material (specimens) are subjected to creep, fatigue, and case hardening. An ultrasonic goniometer was designed and constructed and, using the goniometer; critical angle reflectometry was used to excite the surface waves on the surface of the specimens. For the measurements during fatigue ,specimens were fatigued at high stress levels and low cycles. Constant amplitude alternating stresses with the specimen remaining in tension during the whole program type of loading was used. Most of the specimens were made of titanium (Ti230) but some work was also done on mild steel. The specimens used for creep were made from C263 Nickel alloy and Titanium 230. Each specimen was subjected to creep to a certain degree. The case hardened specimens were made of steel with carburized case. It was found that using ultrasonic surface waves it was possible to determine the depth of case hardening. Using a calibration curve drawn from the results obtained of the change in surface wave velocity for specimens with known case depths, it was possible to predict the case depths of specimens of unknown case depths; by measuring their surface wave velocities. Also, a frequency effect was observed whereby lower frequencies were found to better for detecting larger. case depths When specimens which were subjected to creep were considered, it was found that ultrasonic surface waves can detect early stages of creep. This was the case for both materials studied. A difference was observed in' the response to the surface waves for the two materials studied. In both cases a frequency effect was observed in that higher frequencies were better for detecting early creep. For both materials studied during fatigue, most of the change in surface wave velocity occurred during the very early stages of fatigue life. During the latter stages of fatigue life the rate of change was very much lower. No pattern was observed for the change in surface wave velocity with the number of fatigue cycles for either material. Lastly, measurements were done on brass, copper, aluminium, stainless steel and mild steel bar specimens, at the critical angle of incidence. For all the specimens it was found that the surface wave velocity changed with frequency.

This thesis consists of three parts. Part I (chapters 1-4) gives a review and description of the basis for the numerical modelling of tension structures. The discussion in Part I leads to the conclusion of a need for an interactive design procedure for tension structures which is the subject under consideration in Part II (chapters 5-7). In the design of tension structures, an area which requires special attention is the dynamic response often initiated by the action of a natural wind. In Part III (chapters 8 and 9), this area is examined in detail and a strategy is proposed to give an improved modelling of dynamic response. The numerical procedure developed is assessed by comparison with previously reported test results for a pneumatic dome. Chapter 1 gives a general introduction to tension structures, and their main characteristics and behaviour are briefly described. From both the structural and architectural points of view, tension structures (classified as either prestressed cable nets or doubly curved membranes) do offer a number of benefits which arise from their characteristics and behaviour. The different types of cable nets which can be contructed are outlined, and various possible types of membrane structures and membrane materials which can be used are described. In addition, the form-finding and load analysis stages in the design process of tension structures are briefly discussed. An overview of Parts I, II and III of this thesis is also included. As a result of the flexible nature of a tension structure, large deformations often occur under loads acting normal to the surface. In addition, the coated woven fabrics exhibit material non-linearities, ie. the material properties vary under loads. In other words, a full non-linear structural analysis accounting for both material and geometric non-linearities is required in order to give a realistic modelling of the behaviour of a tension structure. Chapter 2 reviews numerical methods which have been widely reported for dealing with nonlinear structural analysis. From the review, it will be noted that the dynamic relaxation (DR) method is well suited to solving the highly non-linear problems which occur particularly in the case of tension structures. The dynamic relaxation method with a finite element idealisation of the structure is chosen as the solution scheme for all the analysis work in this thesis, and a detailed description of the method is given in chapter 3. Features of the method which are particularly useful for the design of tension structures are: (a) the effective decoupling of the equations of compatibility and equilibrium which allows complex material properties modelling and the use of slip cables, etc., and (b) the use of a 'kinetic' damping procedure which permits gross changes in support geometries to be made during interactive form-finding without the possibility of numerical instability. Although the main surface spanning elements may be purely tensile, many tension structures will employ compression and bending elements for their support. For example, as a means of providing support to a large span tension structure, a compression boundary is considered to be an efficient alternative to tension anchorages. In a sense, a compression boundary is complimentary to the tension elements in the structure as these elements also act as supports to the compression boundary. This gives the advantage of a compression boundary comprising of slender sections. The compression boundary is modelled as a series of beam elements. The moment-curvature equations of a beam element expressed in the form of natural stiffness relations, are developed in chapter 4. In addition, the non-linearities, both geometric and material, and boundary conditions which can be dealt with by the beam elements are considered. An outline is given of the implementation of the beam elements using the dynamic relaxation method. Included in chapter 4 are also the results to test problems which have been set up in order to validate the underlying theory and implementation of the beam elements. As tension structures often exhibit complex surface curvatures, a study of surfaces, their properties and behaviour is appropriate, and useful in the understanding of concepts applied in the design process. This study is the subject of chapter 5 which focusses on the relevant topics of differential geometry. A few useful ideas from differential geometry form the basis of certain procedures implemented into the form-finding and patterning stages being considered in chapter 6. The derivation of the equilibrium equations for a surface when acted on by applied loads is also given in chapter 5. The discussion in chapter 6 is about the stages of form-finding and static load analysis in the design process. A review of the available solution methods for the form-finding problem are given in this chapter. In these methods, the solution can be for either the unknown geometry or unknown stresses, or both. The adopted approach in this thesis is to solve for the unknown equilibrium geometry given the stress distributions, and initial and boundary conditions. The controls which can be used during form-finding to achieve the desired geometries of cable nets and membrane structures are discussed. The equilibrium geometry derived from the form-finding stage has to be subsequently evaluated for its performance under loads at the load analysis stage. After an equilibrium geometry which behaves satisfactorily under loads has been achieved, the corresponding cutting patterns are developed in the case of membrane structures. Recent advances in interactive computer graphics technology have made it possible to develop a fully interactive CAD system for tension structures. The development of such a CAD system is the subject of discussion in chapter 7. The CAD system integrates together the form-finding, load analysis and fabrication patterning stages, resulting in a continuous design process. It demonstrates how the various concepts discussed in Part I of this thesis fit together within an interactive environment implemented with an effective and functional user interface. It is illustrated in chapter 7 how such a user interface has been achieved. The CAD system fully exploits the capabilities offered by the available computer hardware such that the computations involved during analysis of the structure, in generation of surface shaded graphic images and so on, can be executed at very high speeds. As a result, the CAD system can respond quickly to the user and is thus consistent with the interactive nature of the design process. The discussion in chapter 7 also provides an insight into the various procedures involved throughout the design process. The CAD system has produced a number of benefits of which the main one is the saving in design time which has been achieved. As the CAD system is highly user friendly, only a short learning period is required, thus enabling it to be used more widely among designers. The CAD system also serves as a useful tool for the communication of ideas between the engineer and the architect. In the design of a tension structure unlike that of a conventional building, there is often close cooperation between the engineer and the architect right from the early stages of conceptual development. In chapter 8, the possible loads which may act on a tension structure during its service life are considered. These loads are applied to the structure at the load analysis stage in the design process. In most cases, the design loads are those due to snow and wind. An accurate assessment of the loads is essential in order to achieve a structurally sound and economic design. In addition, it should be possible to represent the loads in a form which can be easily applied in the structural analysis. The considerations which are involved in the assessments of the snow and wind loads are outlined.

This thesis is concerned with the use and development of optimisation techniques for process supervision and control. Two major areas related to optimisation are combined namely model predictive control and dynamic data reconciliation. A model predictive control scheme is implemented and used to simulate the control of a coal gasification plant. Static as well as dynamic data reconciliation techniques are developed and used in conjunction with steady-state optimisation and model predictive control schemes. The inaccuracy of process data due to measurement errors can be considerably reduced by data reconciliation techniques. This in turn improves process knowledge and control system performance. The static and dynamic data reconciliation techniques developed in this thesis are tested using dynamic models of process plants. In the steady-state case, a static data reconciliation algorithm that uses a static model of the process is implemented. This algorithm has capabilities of estimating measured variables, unmeasured variables, systematic bias and unknown physical parameters. The technique is applied to static optimisation to show the improvements in performance of the optimiser when using reconciled data. In order for static data reconciliation to be applied, it is necessary to employ a steady-state detection scheme since the underlying assumption is that the process is at steady-state. An algorithm for steady-state detection is implemented and tested in conjunction with the static data reconciliation technique. In the dynamic case, a moving horizon estimator that employs a dynamic model of the process is used to reconcile dynamic process data. An algorithm for the detection, identification and elimination of gross errors is implemented and tested. Furthermore, an algorithm for the detection and identification of systematic bias is developed and implemented. These techniques are then applied in combination to the dynamic model of a process. The effect of dynamic data reconciliation on the performance of model predictive control is observed by means of applying the above techniques to such a scheme. The various algorithms outlined above are implemented in software and tested using appropriate simulations. It is shown that it is possible to implement a steady-state detection algorithm and to successfully use it in conjunction with static data reconciliation. The application of static data reconciliation to steadystate optimisation shows a marked improvement in the performance of the optimiser. It is further shown that it is possible to combine bias and gross error detection and identification algorithms and to successfully apply them to dynamic data reconciliation procedures. The application of dynamic data reconciliation techniques to model predictive control shows improvement in the performance in cases where the objective is not purely economic.