The work carried out during this engineering doctorate degree has led to significant advances in the application of photographic measurement techniques to the characterisation of coarse water in wet steam flow, with particular emphasis on the development of a compact measurement probe suitable for application in a low-pressure steam turbine. Through the application of pulsed LED illumination for the first time in a compact probe, photographs at high magnification with excellent resolution are obtainable at significantly shorter exposure times than others reported in the literature. This has the advantage of reducing motion blur enabling reliable quantitative analysis to take place. Secondary to this, the coupling of high magnification, high resolution photography with pulsed LED illumination has been shown to provide a powerful research tool for flow visualisation across a range of applications, at an order of magnitude lower cost than commercial solutions employing pulsed laser illumination for the same purpose. This builds on the work of others elsewhere in the literature, but again has been proven at shorter exposure times enabling higher magnification with reduced motion blur. Finally a flexible optical test rig has been designed and manufactured, to act as an experimental test-bed for both photographic and phase-Doppler measurements on two-phase steam and air-water sprays. Phase-Doppler anemometry measurements have been performed on an accelerated jet of two-phase steam in order to perform size and velocity measurement of liquid droplets. This acts as a proof-of-concept for the technique in wet steam given the appropriate optical access. In addition phase-Doppler measurements of an air-atomising water nozzle were used as a validation data set to assess the accuracy and reliability of quantitative data from the photographic probe. In this case quantitative data was extracted from the images through application of a custom-designed image processing algorithm, designed to extract droplet size and velocity information from double-exposure droplet images. In general agreement is good within 10-15% of the PDA measurements. Photographic and PDA measurements have also been taken of an LP spray nozzle, as used in the Alstom model steam turbine test facility. These tests have demonstrated significant differences in the spray characteristics when spraying into air and into vacuum conditions. This provides useful insight for the positioning and application of these spray nozzles in a steamturbine environment.

Elastic-plastic crack tip constraint has been estimated for the common planar crack geometries. These include single edge cracked bars in tension and bending, centre cracked panels, and double edge cracked bars. The description of the stress field has been modified from a one parameter characterisation, based on K or J, to include a second term. The second parameter is a non-singular term, described either in terms of an elastic T-stress or a Q field. The limits of one and two parameter characterisation are discussed for single edge bars in tension and bending and for centre cracked panels. For the single edge geometries the two parameter characterisation was found to extend the characterisation well beyond the one parameter approach. For the centre cracked panels the two parameter characterisation in terms of a J-Q approach was found to be more accurate than a J-T approach at high levels of deformation. For the single edge cracked bending and tension geometries the second parameter Q is divided into two components named QT and QP. QT is an elastic term which depends on the elastic T stress and is independent of the distance from the crack tip. The second component, QP, arises from global bending on the uncracked ligament. This is a distance dependent term which depends on the level of deformation and can either be expressed as a function of the load normalised by the limit load or as a function of the plastic component of the J-integral. For single edge bend bars the constraint estimation provides a basis for a method of predicting fracture toughness using local failure approach, where the failure criterion is expressed in terms of the stress level ahead of the crack tip. Finally the constraint of mixed mode problem has been analysed and compared with that of pure Mode I problems.

he thesis introduces novel techniques to study the effects of moulding materials including different fabric architecture on the strength of pultruded adhesive joints and how to improve it. The overall objectives are (i) to devise laboratory techniques which mimic the pultrusion moulding process using glass fabric and vinyl ester resin and other moulding materials, (ii) to optimise the moulding materials and maximise the level of adhesion of the pultrusion and hence to improve the structural efficiency of the joints and (iii) to determine critical stresses and failure initiation through thickness of the composite laminate using multi-scale modeling.

This thesis presents new paradigms for a particular class of non-prehensile manipulators of nanoscale objects that are limited to modelling accurately the relative motion of objects using continuous mechanics where the contact area is not presented. This restrictions results in models which have low accuracy and a lack of understanding about the real motion of the nanoscale object. The newly developed paradigms are focused on three topics: characterisation and analysis of forces present during motion at nanoscale in two dimensional space; characterisation and analysis of the quasi-static motion of nanoscale objects using the the instantaneous centre of rotation iCOR; and characterisation and analysis of the quasi-static,impulsive and dynamic motion of nanoscale objects using motion constraints and the iCOR. For characterisation and analysis of forces present on objects being manipulated at nanoscale, new models to characterise rolling and sliding motion are introduced. For the sliding case a relation between friction load (force and torque) and slip motion (displacement and rotation) for rigid nano-object sliding on a flat and a rough surface, where the distribution of the normal contact forces is assumed to be known a priori and the friction is assumed to be independent of slip rate is introduced. Every point of frictional contact is assumed to obey Coulomb’s friction law. A developed set of equations are solved, performing high accuracy integration techniques such as the Bulirsch-Stoer Method implemented on a computing programming language such as FORTRAN. The full relation between the frictional load and the slip motion for a nano-object can thus be described by its iCOR. A new methodology to model the quasi-static motion of nanoscale objects is presented from which are derived equations that can be used to approximate the trii bological parameters of the nano-objects being manipulated for known and unknown contact pressure distributions. The characterisation of the tribological parameters, such as the coefficient of friction μ, is obtained from generated maps using the applied force or the observed iCOR location of the nano-object being manipulated. The approach has several advantages, including simplicity, robustness, and an ability to simulate classes of systems that are difficult to simulate using spatial mechanics. The final part of this thesis introduces a novel constraint-based method in combination with a minimum force principle to locate the iCOR position for nano-objects at quasi-static motion. Furthermore, the iCOR location for impulsive and dynamic motion cases are introduced. The results generated by modelling these cases can describe the full motion of the manipulated nano-object and generate knowledge of their tribological parameters.

This thesis investigates the concept and design of a portable on-line transfer function analyser (TFA). It is eminently suitable for the identification of plants and other controlled feedback systems in which normal operating records are available. A point by point representation in the frequency domain, requiring a maximum of three records, allows Nyquist plots to be carried out, either visually or by plotter facilities. The basic theory relies heavily upon statistical concepts whereby, least squares estimates of the transfer function are obtained from a combination of heterodyning, exact filtering and adaptive loops. The resultant output, on both channels (real and imaginary), is the culmination of the solution to two linear differential equations with stochastic coefficients, so mechanised when the adaptive loops reach a stable equilibrium. Throughout, emphasis is placed upon the electronics combining the best of analog and digital techniques, in order that six parallel paths may be analysed in similar mode. This is especially true of the heterodyning and filtering operations. Practical shortcomings of the instrument noted by comparing estimates with those from the best currently available commercial apparatus, operating on deterministic signals. Examples of a feedback loop, subjected to both deterministic and random stimuli, with and without the presence of extraneous noise sources, are used to illustrate the ease and simplicity by which the instrument can be used in place of complex computing schemes, which tend, in consequence, to be solely of local academic interest. The practical features of the thesis have led to the submission of four papers to the technical press. Two of them deal, exclusively, with a capacitor ratio commutated filter – not apparently described in publications to date - which is also the subject of a proposed patent application in conjunction with NRDC. It is intended, in the near future, to submit the complete instrument as the basis for a second patent proposal.

The thesis presented herein comprises of the work undertaken to research novel methods of Phased array ultrasonic inspection of complex steam turbine blade roots as found in the power generation industry. The research was conducted as part of the Engineering Doctorate scheme, administered by the Research Centre for Non-Destructive Evaluation (RCNDE), in conjunction with RWE npower and the University of Warwick. Steam turbine blades, and in particularly last stage blades of low pressure steam turbines, are amongst the most highly stressed components on a power generating plant. Two of the most common blade root fixing types include ‘curved axial entry fir tree roots’ (CAEFTR), and axial pinned roots, both of which are prone to cracking due to the high stresses to which they are subjected under operating conditions. Failure of the blade root fixings of such components, leading to the release of the blades, has historically led to the catastrophic failure and destruction of the whole turbine; the cost of collateral damage to plant components and the loss in generating income are seconded only by the risk these failures pose to life. Due to the high price of failure, NDT plays a critical part in the support and management of engineering maintenance, offering insight into the condition and integrity of turbine components through regular planned inspection regimes. It will be shown in this thesis how the invention of a novel continuous wedge, used to refract ultrasound into the critical regions of the blade roots, has significantly improved the ability to detect defects. Combined with the development of bespoke scanning frames these wedges facilitate the efficient and accurate acquisition of scanned data to assess the integrity of the component. By combining the latest reverse engineering, modelling and simulation tools with novel application of rapid prototyping, the author has been able to demonstrate significant reduction in design cycles whilst improving accuracy, sensitivity and repeatability of the applied inspections. Furthermore, application of this design philosophy has led to the development of inspection techniques which have facilitated the inspection of remote regions of the blade roots where manual access is limited or impossible. The developments and techniques invented during this research have been successfully deployed across numerous RWE npower and customer projects, leading to estimated savings in excess of £1m.

This thesis presents the results of a three year research project carried out by the author from October. 1964 to June. 1967 while a Commonwealth Scholar in this country. The contents of the thesis are divided into three parts. Part I describes the analytical and experimental studies on a second-order bang-bang servo under non-ideal operating conditions. The sensitivity and stability of the bang-bang system have been, investigated. This work was carried out during 1964-65 while a research scholar at the Queen's University of Belfast. Part II of the thesis is entitled "Self-optimisation using Pseudo-Random Binary Sequences". In this part, the performance of the original hill-climbing system which employs discontinuous parameter adjustment and of the modified system using continuous parameter adjustment is described. It is shown that the system with continuous parameter adjustment has a superior performance. A novel analogue running averager has been developed which is simple and inexpensive. Part III describes an investigation into the frequency response of a sample-and-hold element at certain sampling frequencies.

To compete in the current economic climate steel companies are striving to reduce costs and tighten process windows. It was with this in mind that a property prediction model for continuous annealed steels produced at Tata Steel’s plants in South Wales was developed. As continuous annealing is one of the final processes that strip steel undergoes before being dispatched to the customer the final properties of the strip are dependent on many factors. These include the annealing conditions, previous thermo-­‐mechanical processing and the steel chemistry. Currently these properties, proof stress, ultimate tensile strength, elongation, strain ratio and strain hardening exponent, are found using a tensile test at the tail end of the coil. This thesis describes the development of a model to predict the final properties of continuous annealed steel. Actual process data along with mechanical properties derived using tensile testing were used to create the model. A generalised regression network was used as the main predictive mechanism. The non-­‐linear generalised regression approach was shown to exceed the predictive accuracy of multiple regression techniques. The use of a genetic algorithm to reduce the number of inputs was shown to increase the accuracy of the model when compared to those trained with all available inputs and those trained using correlation derived inputs. Further work is shown where the fully trained models were used to predict the relationships that exist between the processing conditions and mechanical properties. This was extended to predict the interaction between two process conditions varying at the same time. Using this approach produced predictions that mirrored known relationships within continuous annealed steels and gives predictions specific to the plant that could be used to optimise the process.

In the past decade the Scanning Electron Microscope (SEM) has taken on a significant role in the micro-nano imaging field. A number of researchers have been developing computational techniques for determining depth from SEM images. Depth from Automatic Focusing (DFAF) is one of the most popular depth computation techniques used for SEM. However, images captured with SEM may be distorted and suffer from problems of misalignment due to internal and external factors such as interaction between electron beam and surface of sample, lens aberrations, environmental noise and artefacts on the sample. Distortion and misalignment cause computational errors in the depth determination process. Image correction is required to reduce those errors. In this study the proposed image correction procedure is based on Phase Correlation and Log-Polar Transformation (PCLPT), which has been extensively used as a preprocessing stage for many image processing operations. The computation process of PCLPT covers the pixel level interpolation process but it cannot deal with sub-pixel level interpolation errors. Hence, an image filtering stage is necessary to reduce the error. This enhanced PCLPT was also utilised as a pre-processing step for DFAF which is the first contribution of this research. Although DFAF is a simple technique, it was found that the computation involved becomes more complex with image correction. Thus, the priority to develop a less complicated and more robust depth computation technique for SEM is needed. This study proposes an optimised Blind Image Deconvolution BID) technique using the Bees Algorithm for determining depth. The Bees Algorithm (BA) is a swarm-based optimisation technique which mimics the foraging behaviour of honey bees. The algorithm combines exploitative neighbourhood search with explorative global search to enable effective location of the globally optimal solution to a problem. The BA has been applied to several optimisation problems including mechanical design, job shop scheduling and robot path planning. Due to its promise as an effective global optimisation tool,the BA has been chosen for this work. The second contribution of the research consists of two improvements which have been implemented to enhance the BA. The first improvement focuses on an adaptive approach to neighbourhood size changes. The second consists of two main steps. The first step is to define a measurement technique to determine the direction along which promising solutions can be found. This is based on the steepness angle mimicking the direction along which a scout bee performs its figure-of-eight waggle dance during the recruitment of forager bees. The second step is to develop a hybrid algorithm combining BA and a Hill Climbing Algorithm (HCA) based on the threshold value of the steepness angle. The final contribution of this study is to develop a novel technique based on the BA for optimising the blurriness parameter with BID for determining depth. The techniques proposed in this study have enabled depth information in SEM images to be determined with 68.23 % average accuracy.

The magnetostriction of the core laminations is one of the primary causes of transformer acoustic noise. The magnetostriction of grain oriented silicon steel is extremely sensitive to compressive stress applied along its rolling direction which increases the magnitude of magnetostriction drastically. A measurement system using piezoelectric accelerometers has been built and optimised for magnetostriction measurements under stress within the range of 10 MPa to -10 MPa. This system was used for characterisation of wide range of samples which were prepared and processed under different thermal and mechanical conditions. In this study the influence of factors such as strip thickness, coating stress, annealing under tension, cutting stress and rotational magnetisation on the magnetostriction of silicon steel under stress were investigated. It was observed that the increase of strip thickness leads to the decrease of the magnetostatic energy and therefore a reduction in the volume of closure domains in the stress patterns leading to magnetostriction under stress increasing in proportion with the thickness. Also a gradual increase in coating weight resulted in an effective increase of tensile stress introduced to the surface of the steel which was evaluated by analysis of the stress shift of the magnetostriction curve. An investigation of sample cutting techniques showed that the water jet cutting introduced an advantageous tensile stress along the cut edge in the RD of the steel. Measurement of the pk-pk magnetostriction under rotational magnetisation was shown to be significantly higher than due to uniaxial magnetisation under high compression. The acquired data was used to develop a new prediction model based on the Boltzmann function capable of evaluating the influence of those factors on magnetostriction in GO steel. The final model was able to accurately describe the effect of all studied aspects being present during the production of the steel and affecting the magnetostriction sensitivity of the final material.