The thesis describes research into the general motion of the components of a solid liquid mixture flowing in a horizontal circular pipe. The velocity gradient and concentration distribution of the solid particles have been investigated using high speed photography of coloured marker particles, made visible by matching the refractive indices of the bulk solid and the liquid. These results have been compared with concentration distributions obtained by gamma adsorption techniques. The three dimensional measurements of particle velocity distribution and concentration have been discussed in terms of a dynamic equilibrium between the various influences acting upon the particles. This model is supported by measurements of lateral particle movements. Investigations into methods of measuring the liquid velocity distribution using laser anemometry and photochromic dyes are reported and the feasibility of applying these techniques to a hydraulic conveying system is discussed.

Proper and sensitive monitoring capability to determine the condition of pipelines is desirable to predict leakage and other failure modes, such as flaws and cracks. Currently methods used for detecting pipeline damage rely on visual inspection or localized measurements and thus, can only be used for the detection of damage that is on or near the surface of the structure. This thesis offers reliable, inexpensive and non-destructive technique, based on ultrasonic measurements, to detect faults within Carbon steel pipes and to evaluate the severity of these faults. The proposed technique allows inspections in areas where conventionally used inspection techniques are costly and/or difficult to apply. This work started by developing 3D Finite Elements Modelling (FEM) to describe the dynamic behaviour of ultrasonic wave propagations into the pipe’s structure and to identify the resonance modes. Consequently, the effects of quantified seeded faults, a 1-mm diameter hole of different depths in the pipe wall, on these resonance modes were examined using the developed model. An experimental test rig was designed and implemented for verifying the outcomes of the finite element model. Conventional analysis techniques were applied to detect and evaluate the severity of those quantified faults. However, those signal processing methods were found ineffective for such analysis. Therefore, a more capable signal processing technique, using continuous wavelet techniques (CWT), was developed. The energy contents of certain frequency bands of the CWT were found to be in good agreement with the model predicted responses and show important information on pipe’s defects. The developed technique is found to be sensitive for minor pipe structural related deficiencies and offers a reliable and inexpensive tool for pipeline integrity management programs.

Maintenance strategies based on condition monitoring of the different machines and devices in an industrial process can minimize downtime, increase the safety of plant operations and help in the process of decision-taking for control and maintenance actions in order to reduce maintenance and operating costs. Multivariate statistical methods are widely used for process condition monitoring in modern industrial sites due to the quantity of data available and the difficulties of building analytical models in complex facilities. Nevertheless, the performance of these methodologies is still far away from being ideal, due to different issues such as process nonlinearities or varying operational conditions. In addition application of the latest approaches developed for process monitoring is not widely extended in real industry. The aim of this investigation is to develop new and improve existing methodologies for predictive condition monitoring through the use of multivariate statistical methods. The research focuses on demonstrating the applicability of multivariate algorithms in real complex cases, the improvement of these methods in terms of fault detection and diagnosis by means of data fusion and the estimation of process performance degradation caused by faults.

The principal aim of the work within this thesis was to investigate the fundamental problem of how the frictional heat generated by lubricant shearing at a rolling/sliding elastohydrodynamic contact is divided between the contacting bodies. In elastohydrodynamic lubrication (EHL), a knowledge of the temperatures of the contacting bodies is important due to the effects of temperature on lubricant viscosity in the inlet zone and hence on film thickness. A two-disc test rig previously used to study scuffing was subject to extensive modifications to allow the measurement of disk temperatures at six sub-surface locations in each disk using carefully calibrated embedded thermocouples. A series of experiments using a synthetic gas turbine engine oil was conducted at a range of sliding speeds from 10 ms"1 to 20 ms1 and loads equivalent to maximum Hertzian contact pressures between 1.0 GPa and 1.6 GPa. In each experiment, the speed was fixed and the load applied. Once the temperatures had reached steady-state conditions, the load was removed and the disks separated. The disks were run whilst they cooled until they returned to ambient temperatures. The data recorded dining these experiments were analysed using a transient two-dimensional conduction model of the outer region of the disks, which attempted to obtain an optimal agreement between calculated and measured temperatures. This was achieved by adjusting the partition of heat between the disks and the level of forced convection from the disk surface until the temperatures during the loaded phase of the test (governed by heat partition and cooling) and the cooling phase of the test (governed by the disk surface convection only) were in closest agreement with experimental measurements. Whilst the data recorded from the slow disk was found to be repeatable, there were some errors and lack of repeatability noted with the fast disk data. However, using the slow disk data, it was found that approximately 40% of the frictional heat flows into the slow disk, with the remaining 60% flowing into the fast disk. A series of thermal EHL analyses was performed, using a range of viscosity and rheological models. It was found that the heat partition predicted by the thermal EHL analysis only approached that measured during experiment when the majority of the heat was dissipated by slip at or near to the fast surface. These conditions only occurred when using a limiting shear stress rheological model in conjunction with the Bams viscosity model. The thesis also contains details of finite-element modelling carried out to study elastic-plastic deformation of asperities during the running-in process. It was found that the residual deformation following loading beyond the elastic regime always followed the same shape, with characteristic "piling up" of material around the boundary of the contact. A series of non-dimensional relations for the shape and magnitude of the residual deformation were developed, and their potential for use in an EHL rough surface solver in order to take into account plastic deformation was noted.

The one goal of this research is to present the adaptive mesh refinement (AMR) technique for one dimensional two-phase slug flows. Uniform fine meshes for these long devices are costly and, in general situations, the optimum space discretisation could not be determined a priori. The adaptive mesh refinement (AMR) procedure permits this problem to be remedied by refining the mesh locally, within regions where sharp discontinuities and steep gradients are present. With the appropriate algorithm and data organisation, it helps to reduce CPU time and speed up simulations of flows in long pipes, while preserving accuracy and acceptable execution times. The main objective of this research is to investigate the behaviour of the gas and non-Newtonian shear-thinning fluids in horizontal pipes. Predictions of drag reduction ratio and holdup are presented for the stratified flow of gas and non-Newtonian Ostwald-deWaele liquid. For slug flow regimes, the mechanistic slug unit model is adopted in order to estimate the pressure gradients along the slug unit. The slug unit model is rearranged and reinterpreted as inviscid Burgers’s equation for incompressible phases. For both stratified and slug flow regimes, three dimensional CFD (computational fluid dynamics) simulations were performed in order to compare the drag reduction ratio and pressure gradients. In stratified flows, CFD is also used in an attempt to evaluate the liquid wall friction factor and to compare the obtained values with those given by empirical standard correlations. The estimated pressure gradient and drag reductions are compared with experimental data. Calculations showed an excellent agreement between the simulation and experimental data. Shear thinning effects are also correctly modelled in this work.

Structural integrity assessments of pressurised pipes include plastic collapse as a potential failure mode. This dissertation uses analytical and numerical models as well as experiments to explore the effect of displacement controlled loading conditions and fixed load conditions on the collapse pressure of pipes with and without flaws. Open- and closed-ended pipes are considered. The displacement controlled uniform stress and bending stress represent long-range or fit-up residual stress. Such long-range residual stress, resulting from misfit between components of an engineering structure, is associated with significant elastic follow-up and is commonly treated as primary stress which contributes to plastic collapse. However, this approach may be incorrect as residual stress can be relaxed. There is little evidence that explores the relationship between longrange residual stress and plastic collapse in a pressurised pipe. In this thesis the finite element method and experiments are developed to consider the effect of long-range residual stress on plastic collapse of pressurised pipes. Pipes in a number of states are considered: short and long pipes with no flaws, part and full circumferential flaws. The flaws consist of either a crack or a slot on the internal or external surface of the pipe. Both local and global collapse pressures are considered. It is found that global collapse of pressurised pipes containing longrange residual stress is not dependent solely on the level of initial stress but also on the length of the pipe, the flaw geometries and the collapse dominated stress. It is shown that the displacement controlled stress can act either as secondary stress or as primary stress. Finally, before considering the role of long-range residual stress in failure assessment diagrams, the level of elastic follow-up for full circumferentially flawed pipes is quantified and closed-form approximations of plastic collapse pressures are proposed to consider when a long-range residual stress is treated as either a secondary stress or a primary stress. The results from this work show that conservative assessments will be made of the collapse pressures of pipes containing flaws if a long-range residual stress is simply taken to be an equivalent form of mechanical loading.

A series of experiments were designed with the dual goals of, firstly, accessing new pnictogen ring and cage architectures suitable for use as antiwear additives in industrial lubricants and, secondly, to synthesise novel coinage metal complexes utilising both organic and inorganic ligands. Several asymmetrically and symmetrically substituted cyclodiphosphazanes were synthesised and in the majority of cases the cisoid or transoid isomerism was determined in solution by 31p NMR spectroscopy. Moreover, in three cases the molecular structures of cyclodiphosphazanes were examined using X-ray crystallography. Exo,exo-diamino- and amido-phosphorus sesquisulfide compounds were also characterised by 31p NMR spectroscopy which revealed phosphorus sesquisulfide itself as a decomposition product too. 31p NMR spectroscopy was also used to identify the presence of two new metalphosphorus sesquisulfide coordination species in which the inorganic cage acts as a ditopic ligand via coordination from phosphorus and a rare example of sulfur coordination. Selected cyclodiphosphazanes and phosphorus sulfides were also blended with industrial lubricant (group 1 base oil) and subject to antiwear and friction modifying tests. Furthermore, the structures of fifteen organometallic group 11 metal complexes using alkenes, dienes, trienes, arenes, alkynes and isonitriles as ligands were elucidated using X-ray crystallography, the parameters of which, were compared to analogous and similar examples reported in the literature. These compounds were further characterised by multinuclear NMR spectroscopy, IR spectroscopy and elemental analysis, the former of which provided evidence for the behaviour of complexes in solution.

Ice pigging is an innovative pipe cleaning and product recovery method, patented at the University of Bristol. An ice pig is formed from an ice slurry, which is a two phase mixture of ice crystals suspended in an aqueous solution, most commonly brine. The process involves the flow of ice slurry through a pipe network, driven by a positive pressure differential. The process is capable of effectively removing fouling from internal pipe walls, and recovering product that would otherwise be wasted, in situations where conventional pigging is impossible or impractical. Ice pigging is becoming the worldwide industry standard cleaning technique in the water industry, and is also used for a range of further applications in many industry sectors. Additionally, it is being developed for nuclear and hydrocarbon applications. This thesis investigates the determination of ice fraction , which has been shown to be the critical ice slurry characteristic upon which the performance of an ice pig depends. However, existing ice fraction diagnostic processes are limited, either by inaccuracy or process practicality. Research set out addresses these flaws , and proposes two new ice fraction determination techniques. The first is an improvement to the existing cafetiere test, which is used primarily as a rapid, low cost approximation method. The second is ice fraction determination through electromagnetic interrogation. Development of this process included in-line investigations and the development of a salinity calibration technique. The complete process provides an in-line and near instantaneous ice fraction prediction, shown to be as accurate and reliable as the current gold standard, calorimetry method, which was used to monitor the performance of this novel technique during experimental work. This thesis concludes with a discussion regarding the simplification of the process, shown to not significantly increase process error, and the design of a testing component fit for industrial use.

No description available.

The design and development of large scale seals, under a variety of loads and how they can be monitored for fault diagnosis and life prediction, have been investigated in this study. The work builds upon ideas of developing a physically based mathematical model of an elastomeric lip seal. The approach utilised finite element analysis to simulate its in-service operating condition, in order to obtain parameters needed to be considered for credible life performance of such seal, and therefore allow an extension for new large scale problems, in which the application can be used under increased loads. Accurate modelling requires the knowledge of the boundary condition, material properties and some real life data, which are key parameters to developing a reliable optimised model and a system to enable condition monitoring of seal. The material properties of an elastomer, subjected to tension and compression loads are obtained in this work, whilst putting into consideration the large strain nonlinear elasticity exhibited by the materials. Hyperelastic material models were utilised, since they put into consideration material nonlinearity of elastomers. This eliminates the shortcomings of the conventional Hookean material law currently used in the industries for seal design study. Considering the potential for the demand of a seal monitoring and diagnostic system in the future, this thesis introduces a new, in-process cognitive expert system structure. The structure consists of a knowledge base and an inference model, that is able to evaluate the fault severity of a seal system based on the knowledge provided to it. A case study herein reveals that the structure is able to retain control rules taught to it for the assessment of a sealing process. A computer simulation has been used to analyse, illustrate and evaluate the reliability of the structure, and it shows that it can play a key role in successful application of artificial intelligence technique to monitor the seal and similar assets.