Column separation on both the upstream and downstream sides of a valve in an aviation kerosene pipeline was the subject of an investigation involving the method of characteristics to solve the partial differential equations governing pressure transient propagation. Particular attention was given to obtaining accurate velocity results at the instant the predicted pressure at a section reached vapour pressure. A test rig utilizing L.56 aluminium alloy fuel piping and other aircraft standard components and pumping Aviation Kerosene Specification 2494, was employed to investigate the phenomenon and test the computing procedures. For separation upstream of a valve following closure, a comparison of the computed and observed results indicated an accuracy within 3% for the first peak following valve closure and 5% for the cavity duration. Computed results within 10% of the observed were obtained for the later peaks following cavity collapse. Observation and filming of the sequence of events downstream of the valve during and following closure indicated that the air released during initial separation remained out of solution. The effect of this air was significant but could be included, in terms of its partial pressure, in the cavity boundary equations. The predicted cavity collapse pressures were consistently above those observed. Predicted values of maximum and minimum pressures, and their event times, following valve closure were, at worst, within 10% of the observed results. Analysis of the released gas indicated that it had normal air composition. Measurement of column velocity from the films and the use of a hot film probe and anemometer supported the assumptions made with reference to column motion. The hot film probe results demonstrated that this flow measurement technique was practical in this application.

No description available.

This research presents a vibration analysis for a thin isotropic plate containing an arbitrarily orientated surface crack. The work has been motivated by the well known applicability of various vibrational techniques for structural damage detection in which the detection and localisation of damage to thin plate structures at the earliest stage of development can optimise subsystem performance and assure a safer life, and is intended to be an enhancement to previous work on cracked plates for which the orientation of the crack angle was not included. The novelty of this research activity has been in the assimilation of a significantly enhanced crack model within the analytical model of the plate, in modal space, and taking the form of a specialised Duffing equation. The governing equation of motion of the plate model with enhanced crack modelling is proposed to represent the vibrational response of the plate and is based on classical plate theory into which a developed crack model has been assimilated. The formulation of the angled crack is based on a simplified line-spring model, and the cracked plate is subjected to transverse harmonic excitation with arbitrarily chosen boundary conditions. In addition, the nonlinear behaviour of the cracked plate model is investigated analytically from the amplitude-frequency equation by use of the multiple scales perturbation method. For both cracked square and rectangular plate models, the influence of the boundary conditions, the crack orientation angle, crack length, and location of the point load is demonstrated. It is found that the vibration characteristics and nonlinear characteristics of the cracked plate structure can be greatly affected by the orientation of the crack in the plate. The dynamics and stability of the cracked plate model are also examined numerically using dynamical systems tools for representing the behaviour of this system for a range of parameters. Finally the validity of the developed model is shown through comparison of the results with experimental work and finite element analysis in order to corroborate the effect of crack length and crack orientation angle on the modal parameters, as predicted by the analysis. The results show excellent predictive agreement and it can be seen that the new analytical model could constitute a useful tool for subsequent investigation into the development of damage detection methodologies for generalised plate structures.

The research presented in this thesis is concerned with the Design of a Flexible Manufacturing Cell open to further expansion into Flexible Manufacturing System. The work carried out includes practical and theoretical studies designed to meet the following objectives: Part One: Presents the traditional design method used for the development of an FMC. Part Two: Presents an introduction to the formal 'Design Function Deployment', currently discussed at the Engineering Design Centre at the city University. Part Three: Re-works the FMC design problem using the 'DFD' methodology as currently exists. Presents the use of the 'DFD' methodology with some additional features correlating modifications to costs. Part Four: Presents the general conclusions about the FMC and its flexibility, and compares the design problem with and without DFD. Enhancements to the DFD methodologies are proposed. Where practicable, correlation between the results obtained from field experience and theoretical studies are attempted. The author intends the contents of this thesis to add to the existing knowledge of global design process and enable the identification of specific areas of work which should be pursued for further studies.

Filament winding is a process for the placement of reinforcement fibres on to a rotating surface in a specified geometric pattern. A conventional straight line interpolation controller is not very appropriate for filament winding because the fibre pay-out-eye is at some distance away from the mandrel surface and its movement from one point to another may disturb the position of previously laid fibre on the mandrel. Filament winding demands a controller which can produce fast and smooth carriage movements, and have a path-anticipation capability so that while moving around curves the effect of pay-out-eye movement on previous fibre positions is minimised. This problem can be overcome by using an interpolation technique which determines the pay-out-eye path by considering more than two data points, and whose profile is continuous in nature up to its second derivative. In this project an IDM PC based filament winding controller, using B-spline interpolation technique, is developed. To test the performance of the controller a 3-axis CNC filament winding machine was designed and manufactured. For optimum performance, emphasis was given to the low inertia of the machine carriage, while at the same time maintaining the system's structural stiffness. To reduce the machine carriage weight, the DC servomotors were installed on the machine's frame, instead of the carriage, and a timing belt arrangement was used for power transmission. The controller's electronic hardware was assembled using servo amplifiers, DAC cards, and a purpose built optical encoder interface card. The controller software was developed using TURBO C++ as the main programming language, whereas the hardware interface routines were written in Assembly Language. Problems of winding path deviation as a result of B-spline approximation were tackled using knowledge based programming techniques. The results showed a considerable improvement in winding speed and less fibre slippage in the case of non-geodesic winding patterns, resulting in higher accuracy of fibre placement on the mandrel.

The principal aim of this thesis was to investigate the contact and deformation of rough surfaces such as those found on the teeth of gears. Freshly manufactured surfaces undergo a process known as “running-in”, in which the surface geometry is altered as a result of contact under load. Plastic deformation can occur which induces significant residual stresses, and it has been suggested that these may have implications for the subsequent fatigue life of the surfaces. In this thesis, finite element analysis (FEA) has been used to perform full elastic-plastic contact analysis based on profiles from gear teeth which are used in micropitting tests in order to determine the detailed nature of deformation and the magnitude and distribution of the residual stresses. FEA was performed using Abaqus, and the techniques were first developed using known contact problems of smooth elastic bodies. Plastic behaviour was subsequently introduced, guided by previous studies in the literature. Profiles from real surfaces were then used to study the behaviour of typical gear surfaces under load. Experiments were carried out in which the rough surfaces of crowned steel discs were loaded together, with relocated profiles taken before and after loading. The aim was to provide experimental verification of the residual deformations predicted by FEA. Good agreement was found between the analysis and experiments carried out at different loads. Regions of surface and subsurface residual tensile stress were predicted to occur in proximity to heavily loaded asperity contacts. Greater plastic deformation resulted in increased magnitudes of residual stress. Significant residual tensile stresses were predicted in regions where crack initiation has been shown to occur in practice at depths typical of micropitting failures in gears. It is concluded that residual effects of initial plastic deformation taking place during running-in can be a significant factor in micropitting failures in gears.

The thesis describes a numerical model for evaluating the variation of friction and wear of a self lubricating bearing liner over its useful wear life. Self-lubricating bearings have been in widespread use since the mid-1950s, particularly in the aerospace industry where they have the advantage of being low maintenance components. They are commonly used in relatively low speed, reciprocating applications such as control surface actuators, and usually consist of a spherical bearing with the inner and outer elements separated by a composite textile resinbonded liner. A finite element model has been developed to predict the local stiffness of a particular liner at different states of wear. Results obtained using the model were used to predict the overall friction coefficient as it evolves due to wear, which is a novel approach. Experimental testing was performed on a bespoke flat-on-flat wear test rig with a reciprocating motion to validate the results of the friction model. These tests were carried out on a commercially-available bearing liner, predominantly at a high contact pressure and an average sliding speed of 0.2 ms-1. Good agreement between predicted and experimentally measured wear was obtained when appropriate coefficients of friction were used in the friction model, and when the reciprocating sliding distance was above a critical value. A numerical wear model was also developed to predict the trend of backlash development in real bearing geometries using a novel approach. Results from the wear model were validated against full-scale bearing tests carried out elsewhere by the sponsoring company. Good agreement was obtained between the model predictions and the experimental results for the first 80% of the bearing wear life, and explanations for the discrepancy during the last 20% of the wear life have been proposed.

This work introduces new enhancements to the Bees Algorithm in order to improve its overall performance. These enhancements are early neighbourhood search process, efficiency based recruitment for neighbourhood search process, hybrid strategy involving tabu search, new escape mechanism to escape locals with similar fitness values and autonomy to minimise interaction between search process and the user. The proposed enhancements were applied alone or in pair to develop improved versions of the Bees Algorithm. Three Enhanced Bees Algorithms were introduced: the Early Neighbourhood Search and Efficiency Based recruitment Bees Algorithm (ENSEBRBA), the Hybrid Tabu Bees Algorithm (TBA) and the Autonomous Bees Algorithm (ABA). The ENSEBRBA with an empowered initialisation stage and extra recruitment for neighbourhood search is introduced to improve performance of the Bees Algorithms on high dimensional problems. The TBA is proposed as a new version of the Bees Algorithm which utilises the memory lists to memorise less productive patches. Moreover, the local escape strategy was also implemented to this algorithm. Proposed modifications increased the productivity of the Bees Algorithm by decreasing number of evaluations needed to converge to the global optimum. iii The ABA is developed to provide independency to the Bees Algorithm, thus it is able to self tune its control parameters in a sub-optimal manner. All enhanced Algorithms were tested on continuous type benchmark functions and additionally, statistical analysis was carried out. Observed experimental results proved that proposed enhancements improved the Bees Algorithm’s performance.

Deformation mechanism and dislocation dynamics in grain oriented electrical steel (GOES) is not well established during the continuous flattening anneal process. This work deals with the study of deformation mechanisms during the process and the effect of lattice defects created during the process on the final magnetic properties of GOES. A heat transfer model of the continuous flattening anneal furnace was developed to calculate the temperature profile of the strip throughout the process. The heat transfer model showed the stability of peak strip temperature at 850oC when the line speed was varied from 60-90m/min. A deformation mechanism map was constructed for two varieties of GOES. The main implication of this model is the knowledge of the effect of process parameters like stress, temperature and strain rate on the formation of dislocation structure in GOES during continuous flattening anneal process. LAFFAS (Lab Annealing Furnace for Flattening Anneal Simulation) was constructed to simulate the continuous flattening anneal process and produce samples for dislocation study and magnetic testing. The high temperature mechanical behaviour of Conventional Grain Oriented+ ® (CGO+ - new GOES product) and its effect on magnetic properties were also analysed. Polygonization in GOES and the factors affecting polygonization were studied in detail. Initial orientation of the grains was determined to be a rate controlling factor for degree of polygonization along with temperature and annealing time. The degree of polygonization was shown to be an important parameter affecting the change in domain width. A localised decrease in specific total loss was observed at 1.5T and 50Hz in grains where polygonization was complete. An increase in specific total loss of about 10-35% at 1.5T and 50Hz in bulk polycrystalline material was observed due to the sensitivity of polygonization to initial texture resulting in incomplete polygonization in a high percentage of grains.

The gear meshing is very complex process as high non-linearity behaviour is involved during the process. Normally most analysis of gear tooth contact is on the basis of a hypothesis that the contacted gear teeth are cantilever beams to obtain the approximate results quickly. However, many important factors have been ignored in simulating gear tooth contact using cantilever beam hypothesis, those factors include tooth profile variations, gear rotation speed and input torque, coefficient of friction and geometry distortion. Many of non-linear responses such as noise, vibration, abrasion and so on make the actual contact process much more complicated than those in the cantilever beam hypothesis although it is a reliable reference for integrated gear performance investigation. The main contribution of this thesis is in the investigation of the typical static and dynamic performances of meshing gears by simulating gears rolling-sliding contact processes using non-linear finite element method. A typical spur gear model has been proposed first for performance analysis and the model with integrated parameters could contribute reliable finite element conditions to simulate actual gear meshing. The bending stress and its resultant stress of contacted teeth with static model and dynamic model have been then simulated and the results can be used to provide information on the contacted teeth stresses distribution. The detailed investigation of transmission error in gear meshing has been carried out to understand harmful noise and vibration of gear set and a micro geometry modification of gear tooth profile has been proposed, i.e. tip relief. The gear contact pressure distributions and transmission errors have been then analyzed with micro geometry modification and the analysis is under static and dynamic boundary conditions. Finally, the most significant contribution of the thesis is the investigation of the frictional effect on dynamic performance, including dynamic stress and transmission errors.