The machining characteristics of a glassy thermoplastic (Polyvinyl Chloride) and a semi-crystalline thermoplastic (High Density Polyethylene) have been studied. Chip formation mechanisms, cutting forces and surface integrity were found to be dependent, on the cutting conditions and tool geometry. Results were explained by considering the different nature of the microstructure. Segmented and discontinuous chips were produced with PVC, and continuous and segmented chips were produced with HDPE. It was observed that surface damage was closely related to the nature of chip formation in these plastics. Chip formation, surface damage and tool wear mechanisms when machining Glass-Fibre-Reinforced-Plastic (GFRP) were also studied. Cutting tools used were High-Speed-Steel (HSS), cemented carbide (P type and K type) and coated carbide (titanium carbide - and triple-coated). Discontinuous chips were always produced when machining GFRP. Sliding contact is present at the tool/chip and tool/work interface. The principal aspects of surface damage include fibre breakage, resin cracking, resin decomposition and fibre/resin interface debonding. Cutting temperature is not high, but excessive heat generates when the flank wear land develops. Coated carbide tools showed the best performance and HSS tools the poorest. The main wear mechanisms are abrasive wear with HSS tools, attrition wear with cemented carbides, and discrete plastic deformation followed by attrition wear with coated carbides.

Because of the long service expected from steam power plant it is not practicable to obtain creep data from prototype installations to assist design against excessive creep. Model techniques, however, allow accelerated creep testing in a laboratory environment, which will produce the required creep information in a period of weeks rather than years. Models are made of a lead alloy and subjected to the scaled mechanical service loads at room temperature. Similarity conditions, based on the usual stress-strain-time relationships, have been developed which allow the measured strain distribution to be used to predict the strains in the engineering component at any time during its useful service life. This prediction requires only the uniaxial creep characteristics of the model and component materials. At present the technique is limited to constant temperature conditions. A lead-antimony-arsenic alloy has been selected which can be cast in the laboratory, giving good homogeneity, isotropy and fine grain structure; this material shows sufficient creep strain due to conveniently small stresses at room temperature. The steady load stress-strain-time characteristics have been determined from uniaxial tests. The model technique has been used to study simplified steam turbine casings subjected to internal pressure. The shapes tested consisted of axially split, flanged cylinders with domed end closures containing large bossed central bores to simulate the turbine bearings and glands. The loading of the models was due to the bolting forces and due to internal pressure. Strains on the inner and outer surfaces were measured with electric resistance strain gauges.

The grinding process must achieve the desired component surface finish without producing detrimental workpiece surface layer alterations. These objectives have traditionally been achieved intuitively by skilled operators. Adaptive control, however, offers an opportunity to monitor and control surface integrity during the grinding process. The cylindrical plunge grinding machine adapted for the purposes of this research monitors normal and tangential grinding forces, wheelpower and the amplitude of chatter vibration. The control system is capable of achieving constant normal force machining by adjusting the plunge infeed rate of the wheel. This research is concerned with the feasibility of controlling surface integrity on this and simi lar grinding machines. The research has been carried out by a combination of long duration tests to examine the behaviour of grinding forces, wheelpower and vibration levels with respect to time and shorter duration tests to examine component surface finishes. The results of the tests have enabled models to be developed which relate component surface finishes to metal removal rates. normal grinding forces and chatter vibration levels. Work piece burn and its prevention have also been studied. A method of eliminating workpiece burn has been proposed which utilises in-process normal and tangential force monitoring.

The object of this work was to investigate and expand on previously carried out research into elastic-plastic crack tip fields using the first two terms of the Williams expansion to characterise the degree of crack tip constraint. As a precursor to this research a history of fracture mechanics is also presented. In the present work crack tip fields in small scale yielding have been detennined using modified boundary layer formulations in an attempt to model the influence of the second order term of the Williams expansion, the T -stress. The prime object of this thesis was to investigate and expand on previously carried out research into a two parameter characterisation of elastic-plastic crack tip fields using the second parameter of the Williams expansion(T), which attempts to characterise the degree of crack tip constraint. Modified Boundary Layer formulations in conditions of plane strain were implemented to derive a suitable reference solutions, against which the effects of out of plane strains can be compared and the validity of presently established reference fields can be gauged. The effect of out of plane non-singular stress, S, on the crack tip stress field were also considered, where constraint was largely determined by T. A wide range of analyses have been carried out, from the microstructural scale to complete engineering components in an attempt to characterise crack tip stress fields. The ability to apply two parameter fracture concepts to real engineering structures requires methods for calculating T for complex components with realistic semi-elliptical defects. A simple engineering method for achieving this was developed making use of linespring elements in the finite element package ABAQUS. This approach was validated by the calculation of T for semi-elliptical cracks at the chord-brace intersection of a tubular welded joint, modelled using the mesh generation program PATRAN. The micromechanics of cleavage, using the Ritchie-Knott-Rice model have also been constructed. This work relates the ratio of J for unconstrained and constrained geometries to critical microstructural distance, critical cleavage stress and the toughness ratio on the strainhardening effect. The elastic-plastic behaviour of short and deeply cracked bend bars has previously been described by Betegon and Hancock based on the first two terms of the Williams expansion. A local cleavage criterion has been applied to these fields to indicate the effect of loss of constraint on lower shelf toughness of shallow cracked bend bars. The work models the maximum temperature at which cleavage can occur in these geometries to show the effect of constraint and aJW ratio of cracked bend bars on the ductile-brittle transition temperature. This has also been backed by a significant experimental research program. Finally constraint dependent toughness has been considered in relation to failure assessment methodologies. A simple engineering method for modifying these Failure Assesssment Diagrams has been presented, this consists of considering the constraint matched toughness of the strucutre. This procedure recovers the original Failure Assessment Line and unifies the constraint dependent fracture toughness within defect assessment schemes which utilise Failure Assessment Diagrams.

My thesis consists of three chapters on issues related to the macroeconomic literature of robust control. I detail the salient properties of a RBC model with robust preferences and stochastic-volatility in technological growth. Using this as a framework, I study how to numerically approximate the worst case distribution and how agents check to see whether or not the misspecifications they consider are reasonable. This literature allows for agents to distrust the underlying statistical model governing the economy and seek policies that are ’robust’ to this model misspecification. This is accomplished in the context of a fictitious two player dynamic game between an optimising agent and her “evil” alter ego who is attempting to distort the underlying probability distribution to push the agent into unfavourable regions. The evil agent is constrained to the extent that she can distort the distribution by a parameter on the relative entropy (Kullback-Leibler divergence) between her chosen distortion and the underlying probability distribution. This parameter is somewhat arbitrary and the literature proposes a simple log-likelihood ratio test to pin it down by placing restrictions on the ability of the agent to statistically distinguish between the time series generated under the approximating and distorted model. The solution to the “evil” agents distorted distribution has no known closed form in the non-linear non-Gaussian model used but can be approximated with perturbation methods. Standard Monte Carlo methods then allow the distribution to be sampled from. The first chapter constitutes an introduction to the model used as the base unit for economic analysis in the following chapters. It introduces robustness as a concept to incorporate aversion to model misspecification. Details a very simple RBC model with the addition of stochastic volatility in technology and presents its solution methodology. Finally, the economic dynamics of the model are tested in order to validate that the model is reasonable representation of the macroeconomy and does indeed have interesting non-linearities, as well as that the non-Guassianity introduced by stochastic volatility is economically significant. This is done by considering the models moments and impulse response functions to volatility shocks. The second chapter explores the trade-offs involved in approximating and drawing from the worst case distribution using perturbation methods, a novel method using a simple completing the square argument and two Monte Carlo techniques, importance sampling and MCMC. Naturally, these techniques engender a certain degree of Monte Carlo variation in the resulting approximations. I explore the accuracy of these approximations using summary statistics and ? inaccuracy, a measure of distance between two distributions. Linear and Gaussian models of robust control allow for a analytic solution of the worst case distribution that remains Gaussian. I use this as guidance to check if a simple translated and spread Gaussian distribution approximates the worst case distribution to a similar level of accuracy as that recorded for the Monte Carlo methods. Additionally, I apply the formal empirical test of the finiteness of the variance of the importance sampling weights of ? to check the validity of using IS. This chapter serves as a practical guide for those wishing to apply these techniques as to the trade-offs involved with the various methods. The third chapter explores the statistical distinguishability between the two time series using sequential Monte Carlo methods, namely the particle filter, to evaluate the likelihoods. In comparison to the majority of the literature on robust control, the likelihoods that result from the non-linear and non-Gaussian model studied are non-analytic. Thus, the evaluation of their likelihood introduces a second source of variation in the calculation of detection error. I design a simple yet novel approach that explicitly acknowledges this in the calculation of the detection error. In order to choose the parameters for this new approach I present Monte Carlo evidence on the base unit of its construction. Also, I explore the properties of DEP across three dimensions. First, I test to see if DEP varies across the state space as well. In order to do this, and give the best chance for a difference to be found, I calculate the DEP conditional upon a given initial state. Second, for differing values of the robustness parameter I calculate the DEP in the spirit of ?, where the simulated series are generated from the unconditional distribution and the particle filter for each model also takes its initial state values from their respective unconditional distributions. Finally, for a given value of the robustness parameter I test to see if the simulation method for the worst case distribution affects the evolution of DEP, testing perturbation versus completing the square.

The work reported in this thesis was carried out in the School of Engineering Science, University of Warwick, between October 1984 and October 1987. Chapter 1 contains a review of recent developments in instrumentation that require both manipulation and measurement over the range 0.1-100nm. The instruments considered are the Scanning Tunnelling Microscope (STM), Stylus techniques, X-ray interferometry and x-ray microscopy. The rest of the thesis presents the design and assessment of a novel STM incorporating an X-ray interferometer, an ultra-high precision stylus measuring instrument and an x-ray microscope two axis specimen translation stage. Chapters 2 and 3 present an assessment of different mechanisms for the production of rectilinear motions having parasitic errors of better than mm. Theoretical and experimental investigations into monolithic parallel spring systems based on a notch type hinge and long range slideways based on a polymeric bearing sliding on a polished glass prism are presented. Optimisation of a soleniod magnet force transducer is presented as a drive technique for the former device, whilst a mechanically non-influencing feedscrew drive is described for the latter system. A stylus based measuring instrument, called "Nanosurf 2", that incorporates the polymeric slideway is presented in chapter 4. The performance of this system has been assessed and the results are presented in the following chapter. The linearity and accuracy of an electromagnetically driven, single crystal silicon, monolithic spring make it a suitable for use as the translation mechanism in a Scanning Tunnelling Microscope. Consequently, a three-axis spring has been constructed to generate the translations required for imaging. An X-ray interferometer is built into the probe axis to facilitate absolute calibration. This work is preceeded by an investigation into the generation of complex shapes in this brittle material. Finally, the current status of this work is reported with the presentation of initial experimental results.

This thesis focuses on the design concepts, implementation and preliminary evaluation of a novel micro-tribometer measuring-head test-rig intended to contribute to microtribometery (the measurement of small-scale friction, wear, etc.) which is widely recognised as a resource tool and source of important data design in many fields of advanced technology. Commercially-available instruments clearly address major needs in this respect, but there remains room to question whether their frequency response and other operational ranges suffice for all current applications. It is designed specifically to address questions such as positional stability under low-amplitude oscillatory testing and increased sliding speed and contact repetition rates on small samples. It features a sensing-head that can be used with various means of driving a sample against it. Force actuators are included both to compensate for lateral misalignment in the normal direction and to hold position (while measuring friction) in the lateral direction. Consequently, a single lightweight steel cantilever sensing beam of reasonably high stiffness can be used, helping both resonance and robustness. This design allows several modes of operation from a full null-servo, through running with a “dither” signal, to passive spring sensing similar to commercial instruments. Its purposes are therefore to reveal new information about micro-friction in new materials (especially thin film polymer contacts in MEMS), to discover whether current methods approaches their practical limits and to act as a demonstrator for new generation of high-performance micro-tribometers. As a test bed for measurement and control strategies, the first prototype has been aimed at typical forces in the mN range, sub-50 nm displacement resolution and responses up to around 100 Hz. Basic calibration shows performance acceptably close to the design ideals. Its potential is here illustrated by very promising results from demonstration experiments: the system is shown to clearly track surface topography and to reveal instances of a known frictional behaviour that contravenes Amontons’ laws. All micro-tribological tests followed the same procedure and were performed in a temperature and humidity controlled metrology laboratory, normally at 20 ± 1˚C and 40 ± 5% relative humidity.

Every sub-component of a diesel injector must perform optimally to ensure maximum fuel economy and minimum particulate and carbon-dioxide emissions. Continually developing emissions standards will drive manufacturers to guarantee fuel system performance for a significant period of the unit lifetime. Key to implementing such systems will be the ability to compensate for unit changes in use. This thesis investigates possibilities for developing existing unit and sub-component characterization techniques to enable robust performance over an extended duration. An overview of existing injector characterization techniques is given, along with possibilities for enhancement of the existing state of the art through development of measurement systems widely used within the industry. Particular attention is paid to developing methods to allow testing of injectors and sub-components under representative conditions, where system pressures of 3000 bar are commonplace. Consideration is given to the benefits of moving from existing methods, based on steady-state testing, towards dynamic transfer function testing, and methods reliant on real-time sub-system feedback, to enable better characterization of the unit in use. Key to the development of enhanced characterization will be feedback from the electrohydraulic valves operating inside the injector. These valves are critical to current heavyduty injector designs. To enable full understanding of the unit’s operation, the thesis develops mathematical representations of such component’s functionality. Software models of these systems are offered, and various model simulations discussed. Possibilities for refining these are also provided. Within this thesis the available mechanisms for taking valve position feedback measurements capable of detecting valve lifts of <30 μm with suitable resolution and accuracy are discussed. The results of experiments with various sensing devices are then presented. Subsequently, novel methods for taking measurements using the unit stator as a sensing element are developed, described in terms of mathematical models, and tested in software simulations. Sensing experiments using valve hardware are then described. The thesis closes with a discussion on important future trends in fuel system development.

Up to a decade ago fixed offshore platforms in the Gulf of Mexico were inspected below the water line only after extreme loading from hurricanes or ship collisions. However. increased investment in new platforms has led to oil companies introducing regular inspection routines. and some countries have laid down regulations for instrumentation and inspection of offshore install~ tions. to ensure safe operation and a feedback of data to designers. The conventional techniques used for inspection of offshore structures include: (a) visual inspection by divers and use of underwater television with video and sound recording; (b) magnetic particle and ultrasonic crack detection. methods. These all require diver access. and are feasible in moderate water depths and in areas with long and predictable periods of calm weather. Neither of these conditions is met in the northern North Sea. There is a demand for inspection methods which will monitor the performance of platforms under any weather conditions. and. in particular under storm conditions. The methods must give unambiguous warning about failures which could endanger the installations. Furthermore. they should depend on only a small number of transducers all positioned above the waterline for easy access and maintenance. 1 The purpose of this work was to investigate a method of detecting damage from changes in the natural frequencies of a structure. The main objective was to investigate what changes in natural frequencies would occur when a primary load-carrying memner became detached from a typical fixed offshore platform. The platform design was a four-leg. K-braced platform for which approximate dimensions were available. and which operated in 70 metres of water. A 1/20th scale model was also designed and the structural analysis programme ICES STRUDL II used to compute the natural frequencies and mode shapes of both the full scale and model platforms. Before STRUDL II was applied to the quite complex offshore platform. analyses were carried out on two simple structures to gain expe:ience in using the programme. Natural frequencies and mode shapes were computed and compared with measured values and with analytical solutions where applicable. This preliminary work also gave experience in dynamic response testing. The sensitivity of the natural frequencies of the platform was investigated by removing a single diagonal in one K-panel. and again computing the natural frequencies of the structure. The brace was then replaced and the computations were repeated with other single braces removed in turn. The reductions in the frequencies with each member removed could then be calculated. There were large changes in some of the modes and the pattern of the frequency reductions gave an indication of the location of the f~ilure. 2 The computational study was repeated for the model platform and frequency reductions similar to those for the full scale platform were found. The model platform was built and sinusoidal force excitation was applied to determine its natural frequencies andlmode shapes. Comparison of computed and measured frequencies was generally good. Two K-braces in different planes and at different elevations were severed and repaired in turn; the changes in natural frequencies compared well with the computed predictions. In the third stages of the project a dedicated Time Series Analyser became available and a number of tests were carried out with random excitation. The results for frequencies and mode shapes were in good agreement with those obtained with sinusoidal excitation. This demonstrated the feasibility of obtaining natural frequencies and mode shapes from response to ~andoM excitation. It was concluded for the project that changes in natural frequencies could provide a viable method of detecting gross damage in offshore structures. 3

The overall aim of this research has been to develop new algorithms and computer software that may be used to assess the reliability of water distribution systems. Such a tool can be used by design engineers to create systems which are both economical in total cost commensurate with meeting targets for a specified level of reliability. The introduction describes how water supply and distribution systems are normally designed, what they comprise and problems associated with failure or lack of availability of an adequate supply to the end user. This is followed by a resume of current methods and algorithms for the analysis of networks and a detailed examination of the previous work on network optimisation and reliability. Three main algorithms exist for the analysis of water networks. These are the Hardy-Cross methods, the Newton-Raphson methods and the Linear method. A computer program based on the Linear method, which is known to be the most reliable, is proposed for the hydraulic analysis part of the present work. With respect to reliability, a full discussion of the topic, including all the various factors which influence it such as the stochastic nature of customer demands, the apparently random occurrence of pipe breakages and the concept of repair time, is presented. A reliability analysis model, that incorporates simultaneously the three reliability factors mentioned, for the assessment of nodal and system availabilities, is proposed, from which an efficient computer program has been developed and tested. Two models for the design of optimal water distribution systems, based on reliability criteria, have been developed, programmed and tested. The first model makes use of the entropy principle for producing 'reliable' distributions of flow and the Linear Programming technique is used for computation of the least cost design. In the second model, however, a Genetic Algorithm procedure, that incorporates the new reliability analysis model and which is superior to other models has been formulated. The thesis concludes with a comparison between the two methods formulated as a result of this research and applied to realistic practical systems, plus suggestions for further work to improve the optimisation of water distribution networks.