The work deals with the modelling and optimisation of reverse osmosis (RO) spiral wound elements. It is aimed at improving areas of uncertainty and possible limitations which remain with current published predictive schemes. These were compromised mainly by the lack of adequate experimental data representative of actual operating conditions. Two different mathematical models, termed the `Slit' and the `Spiral' model, were developed. These models differ on the geometrical idealisation of a spiral wound element as indicated by their names. The Solution Diffusion model is used to describe water and salt transport across the membrane. The differential equations governing the process were solved numerically using a finite difference method. The resulting computer programs enable concentrations, pressures and flow rates in the brine and permeate channels to be obtained at any point in the module. The investigation covered a wide range of feed conditions by using experimental data provided from two different types of commercial spiral wound modules. These were the ROGA-4160HR [29] and the Filmtec FT30SW2540 [28] modules. The former type dealt with data typical of brackish water desalination whereas the second type provided data typically encountered in sea-water desalination. The required intrinsic membrane characteristics were determined experimentally using small samples of membrane in a test cell in a closed loop system. For both models, the predictions agree very well with the experimental data over the entire range of operating conditions:- with the exception of some few cases, typical deviations were of the order of 6% for the module productivity and of about 10% for the permeate quality. In addition, parametric studies were performed to establish the programs consistency and the results were in accordance with the theory.

This thesis reports on Semiconductor Optical Amplifiers (SOAs) and their use in optical communication systems; in particular improving the reach of Passive Optical Networks (PON). Following a comprehensive overview of the components of optical communication systems a PON is introduced and the standard of Gigabit-PON (GPON) explained. The concept of extending the reach of GPON through the introduction of amplification is presented and the business drivers of the telecommunication operators detailed. The physics of SOAs are described followed by the parameters used to characterise them. Carrier dynamics of SOAs are explained and the methods of measurement of the carrier dynamics are detailed including the spectrogram technique. This method simultaneously measures the gain and phase recovery which is desirable for applications in long range telecommunications which require unchirped signals with a fast response for both gain and phase. Parameters of commercially available SOAs are compared with the requirements to extend the reach of PONs. Following this the fabrication tolerances for SOAs insensitive to polarisation dependent gain (PDG) are modelled. Results from SOA modelling showed that the greatest contributing factor to PDG variation was the active region thickness error. In the context of bulk production this requires a realistic tolerance of ~10nm to maintain PDG of ~1dB. A polarisation insensitive high gain SOA is designed and experimentally measured. This SOA is measured in the context of GPON and shown to extend the reach of the current standard by a record margin of 28dB. The limitation of the improvement is attributed to gain modulation sourced intersymbol interference (the patterning effect). The patterning effect has been reported in literature to be reduced through the introduction of SOAs with an active region made from quantum dot (Qdot) material. A comparative study of the gain and phase recovery time and alpha factor of various dimensional SOAs is presented. Using the spectrogram method it is shown that reducing the power and increasing the bias of the SOA can reduce the carrier recovery time. A Qdot active region SOAs is shown to considerably reduce the gain recovery time compared to a bulk SOA of similar length. The active region of the Qdot SOA alludes to a faster carrier recovery time which could be beneficial to extend the reach of PONs without patterning. However as these are more difficult to fabricate in mass production it is unknown if they are a viable solution on a commercial scale. In the context of GPON a low alpha factor is desired for minimizing chirp and phase nonlinearities during amplification of short pulses. An alpha factor study is presented and the Qdot SOA was measured to have the lowest alpha factor which could be beneficial for reducing chirp in 10G-PON.

Flexible rotor-bearing system stability is a very important subject impacting the design, control, maintenance and operating safety. As the rotor bearing-system dynamic nonlinearities are significantly more prominent at higher rotating speeds, the demand for better performance through higher speeds has rendered the use of linear approaches for analysis both inadequate and ineffective. To address this need, it becomes important that nonlinear rotor-dynamic responses indicative of the causes of nonlinearity, along with the bifurcated dynamic states of instabilities, be fully studied. The objectives of this research are to study rotor-dynamic instabilities induced by mass unbalance and to use smart materials to stabilise the performance of the flexible rotor-system. A comprehensive mathematical model incorporating translational and rotational inertia, bending stiffness and gyroscopic moment is developed. The dynamic end conditions of the rotor comprising of the active bearing-induced axial force is modelled, the equations of motion are derived using Lagrange equations and the Rayleigh-Ritz method is used to study the basic phenomena on simple systems. In this thesis the axial force terms included in the equations of motion provide a means for axially directed harmonic force to be introduced into the system. The Method of Multiple Scales is applied to study the nonlinear equations obtained and their stabilities. The Dynamics 2 software is used to numerically explore the inception and progression of bifurcations suggestive of the changing rotor-dynamic state and impending instability. In the context of active control of flexible rotors, smart materials particularly SMAs and piezoelectric stack actuators are introduced. The application of shape memory alloy (SMA) elements integrated within glass epoxy composite plates and shells has resulted in the design of a novel smart bearing based on the principle of antagonistic action in this thesis. Previous work has shown that a single SMA/composite active bearing can be very effective in both altering the natural frequency of the fundamental whirl mode as well as the modal amplitude. The drawback with that design has been the disparity in the time constant between the relatively fast heating phase and the much slower cooling phase which is reliant on forced air, or some other form of cooling. This thesis presents a modified design which removes the aforementioned existing shortcomings. This form of design means that the cooling phase of one half, still using forced air, is significantly assisted by switching the other half into its heating phase, and vice versa, thereby equalising the time constants, and giving a faster push-pull load on the centrally located bearing; a loading which is termed ‘antagonistic’ in this present dissertation. The piezoelectric stack actuator provides an account of an investigation into possible dynamic interactions between two nonlinear systems, each possessing nonlinear characteristics in the frequency domain. Parametric excitations are deliberately introduced into a second flexible rotor system by means of a piezoelectric exciter to moderate the response of the pre-existing mass-unbalance vibration inherent to the rotor. The intended application area for this SMA/composite and piezoelectric technologies are in industrial rotor systems, in particular very high-speed plant, such as small light pumps, motor generators, and engines for aerospace and automotive application.

The current generation of wind turbines that are being deployed around the world features, almost exclusively, a three-bladed rotor with a horizontal-axis configuration. In recent years, however, a resurgence of interest in the vertical-axis wind turbine configuration has been prompted by some of its inherent advantages over horizontal-axis rotors, particularly in flow conditions that are typical of the urban environment. The accurate modelling of the aerodynamics of vertical-axis wind turbines poses a significant challenge. The cyclic motion of the turbine induces large variations in the angle of attack on the blades during each rotor revolution that result in significant unsteadiness in their aerodynamic loading. In addition, aerodynamic interactions occur between the blades of the turbine and the wake that is generated by the rotor. Interactions between the blades of the turbine and, in particular, tip vortices that were trailed in previous revolutions produce impulsive variations in the blade aerodynamic loading, but these interactions are notoriously difficult to simulate accurately. This dissertation describes the application of a simulation tool, the Vorticity Transport Model (VTM), to the prediction of the aerodynamic performance of three different vertical-axis wind turbines - one with straight blades, another with curved blades and a third with a helically twisted blade configuration - when their rotors are operated in three different conditions. These operating conditions were chosen to be representative of the flow conditions that a vertical-axis wind turbine is likely to encounter in the urban environment. Results of simulations are shown for each of the three different turbine configurations when the rotor is operated in oblique flow, in other words when the wind vector is non-perpendicular to the axis of rotation of the rotor, and also when subjected to unsteady wind. The performance of the straight-bladed turbine when it is influenced by the wake of another rotor is also discussed. The capability of the VTM to simulate the flow surrounding vertical-axis wind turbines has been enhanced by a dynamic stall model that was implemented in the course of this research in order to account for the effects of large, transient variations of the angle of attack on the aerodynamic loading on the turbine blades. It is demonstrated that helical blade twist reduces the oscillation of the power coefficient that is an inherent feature of turbines with non-twisted blades. It is also found that the variation in the blade aerodynamic loading that is caused by the continuous variation of the angle of attack on the blades during each revolution is much larger, and thus far more significant, than that which is induced by an unsteady wind or by an interaction with the wake that is produced by another rotor. Furthermore, it is shown that a vertical-axis turbine that is operated in oblique flow can, potentially, produce a higher power coefficient compared to the operation in conditions in which the wind vector is perpendicular to the axis of rotation, when the ratio between the height of the turbine and the radius of the rotor is sufficiently low.

This research project is primarily concerned with the development of a new generation of power transmission lines for both non-real-time and real-time electromagnetic transient studies. The method proposed is entirely formulated in phase co-ordinates, avoiding the use of modal transformation matrices at every stage in the analysis. In comparison, the phase domain models presented thus far in the open literature have all incorporated the concept of modal decomposition in the initial frequency domain formulation of the problem. Only the time domain analysis is conducted in the phase domain. These models can therefore be regarded as a hybrid between the phase and modal methodologies. Algorithms are presented which allow accurate and efficient determination of the characteristic admittance matrix, Yc(), and wave propagation matrix, H(), directly in phase co-ordinates. A Padé iteration scheme is used for evaluating the characteristic admittance matrix, derived by exploiting a relationship between the matrix sign function and the matrix square root. Padé techniques have also been used to approximate the matrix exponential in order to evaluate the wave propagation function. By evaluating Yc() and H() directly in phase co-ordinates, any imbalances naturally present in the line will intrinsically be taken into account in these functions. Both methods have been extensively tested using line configurations of different size and complexity and both algorithms are shown to be very robust, accurate and efficient in all cases. One of the main difficulties in formulating the analysis entirely in phase co-ordinates for multiconductor systems concerns the unwinding of the wave propagation matrix. This is addressed in this research by evaluating a matrix phase shift function in phase co-ordinates. Since the method inherently takes into account the coupled time delays of the line, the elements of H() can be successfully unwound, irrespective of the configuration of the line, e.g. single-circuit, multi-circuit or asymmetrical.

Retentate obtained from ultrafiltration was used as a substitute for skim milk powder, in the manufacture of ice cream and cajeta (Mexican dairy spread). The products were assessed by chemical, physical, sensory and structural analysis. Ice creams made using ultrafiltered retentate had increased ash, protein, calcium, phosphorus and magnesium, but reduced lactose, potassium and sodium contents. Physical evaluation showed that UF-products were harder, more viscous and had better melting resistance, but had lower overrun and extrusion temperature than control ice cream. In Sensory analysis UF-products scored better for iciness, sandiness and fluffiness, and resisted heat shock treatment better. No consumer preference for UF-based ice cream or control ice cream was found. The UF-ice cream took longer to soften to eating consistency. Structural examination of ice cream products by various microscopy techniques revealed air cell, ice crystal and fat droplet structures within a sugar and protein matrix. Freeze substitution was applied to ice cream for Transmission Electron Microscopy to produce unique thin sectioned samples. This showed a more agglomerated casein structure than UF-based ice cream. Heat shock changed ice cream structure. Ice crystal size increased and crystals fused into a network. Air cells could be distorted into a modified channel shape. Chemical, physical, microbiological and sensorial analysis of cajeta were carried out. UF-cajeta had slightly higher protein calcium and phosphorous contents and lower lactose, potassium and sodium contents. UF-cajeta showed better sensory attributes after storage than the control, however as shelf life was extended yeast and mould growth was possible. Structural examination of cajeta showed ultrafiltered retentate in cajeta manufacture prevented the formation of larger crystals and prevented sandiness that developed in the control product.

This thesis addresses the improvement in quality of decision making in design through the use of decomposed design evaluation. The research reported in this thesis is supported by the Design Research Methodology. To perform decomposed decision making, it is necessary to identify criteria that are deemed important for this activity. Questionnaire surveys, literature review and interviews with industry helped to identify these criteria. Reliability and realizability are two criteria that are selected for research in this thesis. The questionnaire surveys are discussed in chapter 2. A review of literature on decision making, reliability and realizability is reported in chapters 3 and 4. Methodologies for evaluating reliability and physical realizability are discussed in chapter 5. Relative reliability risk assessment methodology is applied to various examples consisting of university and industry projects in chapter 6. The application helps to reveal the strengths of the methodology and is termed ‘Verification of the methodology’. Validation issues of both the methodologies are dealt with in chapter 7 using the controlled experimental design. It is found that both the methodologies help to improve the quality of decision making during design evaluation. Relative reliability risk evaluation methodology helps to improve the quality of decision making to a substantial extent but physical realizability evaluation methodology shows only a little improvement in quality of decision making. Finally, it is suggested that the decomposed design evaluation methodology helps to improve the quality of decision making and is therefore proposed to be used by both novice and experienced designers.

The thesis is concerned with the design, analysis, fabrication am evaluation of integrated optic lenses which are inhomogeneous either in physical shape or in refractive index profile. The thesis has nine chapters. Chapter one, the introduction, illustrates the importance of these lenses within the domain of integrated optiCS, where the complicated mathematical functions required to describe the lens profiles are most easily realised. Connections are made between the study of these lenses and the exciting new field of optical computing. A special class of non-uniform lenses which are conceptually perfect optical instruments forms the main area of interest in the present study. Historically, the development of these lenses has followed two distinct lines, related to two possible methods of physically obtaining the required variation in path of light rays passing through the lens. In one method the optical path is made to vary directly, whilst the other method involves controlling the fi'lysical path, and thus the optical path, through the principle of equivalence. The dual development has been continued in the field of integrated optiCS, where lenses based on direct control of the optical path are termed variable-index lenses and those based on physical path control are termed geodesic lenses. The perfect variable-index lens studied in this work was the well-known Luneburg lens. Perfect geodesic lens designs have also been published. The design formulae for both types of lens are presented in chapter two. A simpler lens, of spherical geometry, is also presented which is easily analyzed and characterised and which serves as an archetypal model against which the performance of the more sophisticated lenses can be assessed. Chapter three investigates the problems involved in modelling fabrication conditions in a thermal-evaporation-invacuum environment so that lens profiles can actually be constructed. Chapter four goes into methods of tracing rays through these lenses in some detail. Ray-tracing has long been the classical tool of optical designers, providing a useful guide to optical performance. Ray methods, which effectively provide image error evaluations, are not entirely-appropriate for those lenses which are conceptually perfect within the geometrical optics approximation. Diffraction effects prevent the lenses from attaining true perfection. In such cases the wave-field produced by the lenses in the image space is the important quantity. In chapter five, the beam-propagation method (BPM) is used to study diffraction arrl associated effects in inhomogeneous lenses. '!he method allows the propagation of complicated waveshapes in lnhomogeneous media, normally a difficult task. Furthermore, anlsotropic effects and the interaction between acoustic waves aoo optical waves can be studied with the method. Negative focal shifts are reported which are not predicted by geometrical optics or the usual approximate diffraction theories. The fabrication of lenses is considered in chapter six. Planar waveguide measurements car r ied out on the var ious materials used in the study are presented. A major problem in the fabrication of geodesic lenses, that of obtaining a uniform wavegulde layer over the complete lens area, is dealt with in some detail in chapter seven. In chapter eight, extensive tests on the experimental performance of several lenses are reported. Near diffraction-limited performance is reported for geodesic lenses. More limited performance figures are obtained for Luneburg lenses though the possibility of high performance is lndicated if profile resolution can be improved. The themes of the thesls are pulled together for discussion in chapter nine and conclusions are drawn as to the relative merits of the various lenses. Possible means of improving fabrication procedures, thus driving lenses closer to ultimate resolution limits, are presented. The greatest problem faced is that of scatter ing in the waveguide, which appears to be accentuated as the waveguide traverses the lens surface. If the scattering problem can be successfully dealt with it is concluded that integrated optical lenses could be important and viable components in addresslng the problem of fast, high-throughput data processing.

Allpass filters are commonly employed in many applications to perform group delay equalisation in the passband. They are non-minimum phase by definition and are characterised by poles and zeros in mirror-image symmetry. SI allpass filters of both cascade biquad and bilinear-LDI ladder types have been in existence. These were implemented using Euler based integrators. Cascade biquads are known to have highly sensitive amplitude responses and Euler integrators suffer from excess phase. The equalisers that are proposed here are based on bilinear integrators instead of Euler ones. Derivation of these equalisers can proceed from either the s-domain, or directly from the z-domain, where a prototype is synthesised using the respective continued-fractions expansions, and simulated using standard matrix methods. The amplitude response of the bilinear allpass filter is shown to be completely insensitive to deviations in the reactive ladder section. Simulations of sensitivities and non-ideal responses reveal the advantages and disadvantages of the various structures. Existing DI multirate filters have to date been implemented as direct-form FIR and IIR polyphase structures, or as simple cascade biquad or ladder structures with non-optimum settling times. FIR structures require a large number of impulse coefficients to realise highly selective responses. Even in the case of linear phase response with symmetric impulse coefficients, when the number of coefficients can be halved, significant overheads can be incurred by additional multiplexing circuitry. Direct-form IIR structures are simple but are known to be sensitive to coefficient deviations and structures with non-optimum settling times operate entirely at the higher clock frequency. The novel SI decimators and interpolators proposed are based on low sensitivity ladder structures coupled with FIR polyphase networks. They operate entirely at the lower clock frequency which maximises the time available for the memory cells to settle. Two different coupling architectures with different advantages and disadvantages are studied.

This thesis reports the results of a three year, full-time research project investigating the generation and communication of product descriptions within the conceptual phase of the engineering design process. The research pays particular attention to the role played by the designer's sketch in communicating new product ideas. The investigation commences with a literature review of existing design process models (Chapter 2), which helps to define the area under investigation while presenting modern views of the process in relation to classic examples from established design research. Chapter 3 presents a literature review of the methods currently used to support communication of product descriptions. These methods of Specification are assessed and particular attention is given to new computer-based recording methods such as DOORS and Cradle. Suggestions for improving the efficiency of such models are put forward and the text-only bias of such systems is identified. This comparison of the existing systems thus identifies the research questions. Having identified the possible improvement to be gained by the incorporation of visual material in addition to the universal text description, Chapter 4 presents a literature review assessing the roles of the conceptual sketch in engineering design. As well as presenting views of drawing from philosophical, psychological and scientific standpoints, this section compares attempts made to support the engineer's sketching activity by computer means. This chapter concludes that efforts made to provide effective computer support of sketching by freehand methods are preferred to attempts made to replicate the process with current computer tools. The resulting research experiment, the methodology of which is described in Chapter 5, uses students from the final year of the Product Design Engineering course at Glasgow School of Art and the University of Glasgow. The main aim of the experiment is to identify means of including sketching within the kind of text-based support methods discussed in Chapter 3. It also observes the volume and pattern of information produced by sketch activity throughout the conceptual stages of the design process and aims to find methods which would enable sketches to indicate the general progress of a design. The findings are detailed in Chapter 6.