School of Aerospace, Transport and Manufacturing (SATM)
Permanent URI for this community
Browse
Browsing School of Aerospace, Transport and Manufacturing (SATM) by Type "Thesis"
Now showing 1 - 20 of 197
Results Per Page
Sort Options
Item Open Access Adoption of product service systems in health care.(2017-10) Mittermeyer, Stephan Alexander; Tomiyama, Tetsuo; Tiwari, AshutoshHealth care systems are constantly challenged to deliver better quality of care at lower cost. Product Services Systems (PSS) aim to output a higher value to a customer, while reducing resource input required to achieve such value and sustainability. In the health care market this could help companies increase their focus on value for the patient, but also for the health care system as such. This focus on value can ultimately help drive down health care cost, which is one of the most pressing issues in health care systems today. The potential of PSS to address some of the major challenges in the health care market was recognised early in PSS research, however adoption in this field is still below expectation. Motivated by the potential of PSS in health care this work aims to explore the current status of adoption as well as drivers and barriers to future adoption in this market and evaluates if and how PSS can be designed and implemented by companies active in this market. This work showed that PSS can be feasible and useful in this sector as they address relevant current challenges. Future changes in the health care market will likely make PSS even more relevant. Certain concepts of PSS are already applied in the market without leveraging the benefits of a fully developed PSS. Limitations in how the value for patients and other market actors is determined and made transparent is a major challenge in the adoption of PSS. An assessment method is proposed to enable companies to evaluate the value generation of their PSS offerings. In addition, a guideline for PSS design is proposed based on results of this work and field observations. This thesis contributes to a better understanding of PSS adoption in health care by investigating mechanisms in the health care market to understand if PSS can be implemented in a useful manner and how PSS can be adopted in health care in the future. As PSS consists of a number of separate concepts that may be used by themselves and also outside a PSS concept, a detailed analysis was performed to evaluate how PSS concepts are already utilized by industry, as such partial implementations may be a good starting point for full PSS adoption. Adoption of a PSS in any industry requires a measure to evaluate the success of a system implementation or the quality of PSS offerings. Given the complex market network in health care, metrics for evaluations have been identified, linking different dimensions of clinical utility to PSS. Those metrics enable companies to assess PSS systems or scenarios, but also enable development teams to focus their PSS design efforts, as those assessment metrics provide a framework for PSS requirements engineering in this market. Based on the results of the work outlined above, design guidelines were defined to support the development process of PSS in health care.Item Open Access Aerodynamic analysis of large wind farms using two-scale coupled modelling approaches.(2021-08) Ma, Lun; Tsoutsanis, Panagiotis; Antoniadis, AntoniosThe effects of turbine aerodynamics and response characteristics of the atmospheric boundary layer on the overall wind farm efficiency are investigated in this research. Various wind farm modelling strategies, which include a theoretical and several CFD models, are presented. This study consists of three main parts: (i) improve and validate an existing theoretical wind farm model, (ii) infinitely large wind farm modelling with actuator-disc and fully-resolved turbine models, and (iii) finite-size wind farm modelling with a numerical weather prediction model. In the first part, an extended theoretical model based on a two-scale coupled momentum balance method is proposed to estimate aerodynamic effects of wind turbine towers on the performance of large wind farms. The modified theoretical model predicts that the optimal turbine spacing should increase with the value of normalised support-structure drag, as well as additional parameters describing the response characteristics of the atmospheric boundary layer to the total farm drag. The Detached-Eddy simulations of a periodic array of fully staggered actuator discs (AD) show a reasonably good agreement (within 10% in the prediction of power) with the modified theoretical model. In the second part, a fully resolved (FR) NREL 5MW turbine model is employed in two URANS simulations (with and without the turbine tower) of a fully developed wind farm boundary layer. The FR-URANS results show stronger tower effects than both AD-RANS and theoretical model predictions, which is a strong indication of the necessity of considering turbine support structure within large wind farm models. The possibility of performing DDES is also investigated with the same FR turbine model and periodic domain setup. The results show complex turbulent flow characteristics within a large wind farm, where typical hairpin and hub vortices have been clearly captured. In addition, the computational cost of DDES has been found to be similar to URANS (for a given number of rotations), which is a positive sign for conducting DDES in future studies. In the third part, a numerical weather prediction model is used as a realistic farm-scale flow model to investigate how the streamwise pressure gradient, Coriolis force and acceleration/deceleration terms in the farm-scale momentum balance equation tend to change in time. The results suggest that the streamwise pressure gradient may be enhanced substantially by the resistance caused by the wind farm, whereas its influence on the other two terms appears to be relatively minor. These results suggest the importance of modelling the farm-induced pressure gradient accurately for various weather conditions in future studies of large wind farmsItem Open Access The aerodynamics of aero-engine Nacalles.(2018-02) Robinson, Matthew H.; MacManus, David G.This thesis deals with the aerodynamics of aero-engine nacelles with a focus on the influence of a short and slim nacelle design on the drag performance. As turbofan engines are designed with increasingly reduced specific thrusts in order to improve propulsive efficiency, the fan diameter tends to grow. With a larger fan, the engine weight and nacelle drag grow which may offset the benefit from the reduced specific thrust. It is imperative to determine if a reduced length and thickness nacelle, compared to a conventional design, will enable the viable use of these reduced specific thrust aero-engine designs. The research aims to answer this question with a focus on cruise drag, spillage drag, drag rise and windmill performance of isolated and installed short, and slim nacelles. An innovative optimisation process was developed with a computational fluid dynamics process included as a means to evaluate nacelle drag. This was applied to different nacelle designs in a novel design space to optimise for cruise and off-design performance with a multi-objective genetic algorithm. The optimisation routine was extensively tested and verified against a number of analytical functions to ensure it could adequately approximate optimal Pareto sets. The optimisation of both axisymmetric and non-axisymmetric nacelles was carried out on drag, spillage and drag rise Mach number as well as on two metrics which control the pressure distribution of the nacelle. Optimal nacelles were then chosen to study the influence of nacelle incidence, the windmill condition and installation onto an aircraft on the drag performance and to provide a new quantification of these impacts. The optimisation demonstrated that under cruise conditions it is possible to have compact nacelle designs that offer reductions in drag. For example, a nacelle with a 23% reduction in length resulted in a 22% reduction in nacelle drag. However, these compact designs are more sensitive to off design condition. Specifically the spillage drag at a required drag rise Mach number of 0.87 could be 9 times higher for the reduced length nacelle. Nonetheless, it is possible to create a nacelle at the shortest length tested which had spillage of less than 6% of the cruise drag and met all requirements on drag rise to cruise at a Mach number of 0.85. This was enabled by an increase in the trailing edge radius such that it was equal to the highlight radius which improved the wave drag characteristics. Whilst the shortened nacelle was viable at low incidence, the increased wave drag resulted in the drag benefit relative to the conventional design being negated by an incidence of 6 degrees. In addition, this reduced length nacelle experienced separation at the end of runway windmill condition at 22 degrees, which is below the requirement of 30 degrees. Once installed on an aircraft the impact of reducing the nacelle length was a decrease in overall cruise aircraft drag of 3%. These studies demonstrate that there is a significant cruise benefit available from a short nacelle but that the off design conditions, most notably windmill requirements, will need to be addressed.Item Open Access Aeroelastic investigation of conventional fixed wings and bio-inspired flapping wings by analysis and experiment.(2018-09) Li, Hao; Guo, Shijun J.In this thesis, the structure and aeroelastic design, analysis and optimization of conventional fixed wing is firstly addressed. Based on the study results of conventional fixed wing, the study then focuses on the more complicated aerodynamics and aeroelasticity of flapping wing Micro Air Vehicles (MAV). A Finite Element (FE) model of a composite aircraft wing is firstly used as case study for the aeroelasticity of conventional fixed wing. A MATLAB-NASTRAN interfaced optimization platform is created to explore the optimal design of the wing. Optimizations using the developed platform show that 13% of weight reduction can be achieved when the optimization objective is set to minimize wing weight; and 18.5% of flutter speed increase can be achieved when aeroelastic tailoring of composite laminate layups is carried out. The study results further showed that the most sensitive part of the wing for aeroelastic tailoring is near the engine location, which contributes to the majority of flutter speed increment for optimization. In order to facilitate the structural design of non-circular cross section fuselage of Blended-Wing-Body (BWB) aircraft, an analytical model of 2D non-circular cross section is developed, which provides efficient design and optimization of the fuselage structure without referring to FE models. A case study based on a typical BWB fuselage using the developed model shows that by optimizing the fuselage structure, significant weight saving (17%) can be achieved. In comparison with the conventional fixed wing, insect flapping wings demonstrate more complicated aerodynamic and aeroelastic phenomena. A semi-empirical quasi-steady aerodynamic model is firstly developed to model the unsteady aerodynamic force of flapping wing. Based on this model, the aerodynamic efficiency of a Flapping Wing Rotor (FWR) MAV is investigated. The results show that the optimal wing kinematics of the FWR falls into a narrow range of design parameters governed by the dimensionless Strouhal number (St). Furthermore, the results show that the passive rotational of the FWR converges to an equilibrium state of high aerodynamic efficiency, which is a desirable feature for MAV applications. Next, the aerodynamic lift coefficient and efficiency of the FWR are calculated and compared with typical insect-like flapping wings and rotary wing. The results show that the aerodynamic efficiency of FWR in typical wing kinematics is higher than insect-like flapping wings, but slightly lower than the conventional rotary wing; the FWR aerodynamic lift coefficient (CL) surpassed the other wings significantly. Based on the numerical results, the study then continued to experimental investigations of the FWR. A prototype FWR model of weight 2.6g is mounted on a load cell to measure the instantaneous lift production. The kinematics of the wing is captured using high speed camera. Aeroelastic twist of the wing is measured using the resulting wing motion. Analyses by CFD and the quasi-steady aerodynamic model is then carried out and compared with experimental results. The study revealed that passive twist of the FWR wing due to aeroelastic effects forms desirable variations of wing Angle of Attack (AoA), which improves the aerodynamic performance of FWR. The results of the thesis provide guidance for structural, aerodynamic and aeroelastic design, analysis and optimization of conventional fixed wing, as well as bio-inspired flapping wing MAVs.Item Open Access Aeroelastic simulation of rotorcraft propulsion systems(2017) Castillo Pardo, Alejandro; Pachidis, VassiliosA close relationship between the aerospace technology level and the capability to model and simulate the physics involved during the flight has been identified throughout the aviation history. The continuous improvement in physical and mathematical models has provided a further understanding of the behaviour of the different components along with the complete vehicle. As a result, the performance modelling has experienced a large improvement. The aviation industry, which is characterised by the use of cutting edge technology, requires large investments when new concepts are introduced. The application of high fi delity simulation tools reduces considerably the investment carried out prototyping and testing. This fact is also applicable to the rotorcraft industry, where a continuous increase in the employment of helicopters has been observed throughout the last decades, expecting a sharp growth within the next 20 years. The forecasted growth in the number of helicopter operations along with the increasing concern about the environmental impact of aviation, lead the governmental bodies to set up a number of goals to reduce the carbon dioxide, nitrogen oxides, and noise emissions. Three paths were identified to reduce the environmental impact and meet the proposed goals. The fi rst one is the reduction in the number of operations. However, a sharp growth in the number of helicopter operations is expected. The second one is the optimisation of the flight procedures. Nevertheless, the potential improvement is limited. The third one is the introduction of a quieter and more,efficient type of rotorcraft. There exist two new rotorcraft con figurations which show enough potential to be studied. These are the tilt-rotor and compound helicopter. Both designs improve the cruise performance using auxiliary lift and propulsive systems, while they still exploit the vertical flight capability of helicopters. Nevertheless, the lack of reliable high fi delity models has made their development long and highly expensive. Within this context, the necessity of a simulation framework able to simulate and predict the detailed performance of novel rotorcraft con figurations is highlighted. The present work aims to lay the foundations of this comprehensive rotorcraft code by developing a computational framework for the aeroelastic simulation of propulsion systems. The tool is characterised by a high fi delity level able to predict the highly unsteady loads at a low computational cost. The fi rst characteristic makes this tool suitable for the design stage and noise calculations; whilst the second one enables its integration into multidisciplinary optimisation procedures. The development of this framework has required a considerable contribution to the knowledge in different areas of study, these included: structural dynamics, in flow aerodynamics, blade aerodynamics, aeroelasticity, and computational acceleration techniques. The individual models have been integrated into a cost efficient aeroelastic simulation framework, which has been extensively validated with experimental data. Very good and in some cases excellent correlation with the experimental measurements has been observed. The main contribution of this work has been the successful development of a computational framework for the aeroelastic simulation of rotorcraft propulsion systems. It accurately simulates and predicts the aerodynamic flow field and the unsteady loads generated by the rotor and transferred to the fuselage. It is easily expandable to account for interactions with other rotors, auxiliary lift surfaces, and fuselage bodies. The simulation tool estimates high fidelity low and high frequency aerodynamic loading, which enables the calculation of impulsive noise emissions. The framework computes accurate predictions of rotor power required, which enables its use as a validation tool for lower order models. The developed framework approximates the third level of Padfi eld's hierarchical paradigm, providing detailed aeroelastic information necessary for design purposes. The additions of parallel computing and an acceleration scheme results in a highly computationally effcient tool suitable for optimisation methodologies. Moreover, a considerable contribution has been made in terms of modelling of: coupled modal characteristics, aeroelastic simulation; computational enhancements of in flow models and investigation of the effect of the fuselage aerodynamic interference and coupled flexible blade modelling.Item Open Access Alignment measurements uncertainties for large assemblies using probabilistic analysis techniques.(2017-12) Doytchinov, Iordan; Tonnellier, Xavier P.; Almond, HeatherBig science and ambitious industrial projects continually push forward with technical requirements beyond the grasp of conventional engineering techniques. Example of those are ultra-high precision requirements in the field of celestial telescopes, particle accelerators and aerospace industry. Such extreme requirements are limited largely by the capability of the metrology used, namely, it’s uncertainty in relation to the alignment tolerance required. The current work was initiated as part of Maria Curie European research project held at CERN, Geneva aiming to answer those challenges as related to future accelerators requiring alignment of 2 m large assemblies to tolerances in the 10 µm range. The thesis has found several gaps in current knowledge limiting such capability. Among those was the lack of application of state of the art uncertainty propagation methods in alignment measurements metrology. Another major limiting factor found was the lack of uncertainty statements in the thermal errors compensations applied to assembly’s alignment metrology. A novel methodology was developed by which mixture of probabilistic modelling and high precision traceable reference measurements were used to quantify both measurement and thermal models compensation uncertainty accurately. Results have shown that the suggested methodology can accurately predict CMM specific measurement uncertainty as well as thermal drift compensation made by empirical, FEM and FEM metamodels. The CMM task-specific measurement uncertainties made at metrology laboratory were validated to be of maximum 7.96 µm (1σ) for the largest 2 m assemblies. The analysis of the results further showed how using this method a ‘virtual twins’ of the engineering structures can be calibrated with the known uncertainty of thermal drift prediction behaviour in the micrometric range. Namely, the Empirical, FEM and FEM Metamodels uncertainties of predictions were validated to be of maximum: 8.7 µm (1σ), 11.28 µm (1σ) and 12.24 µm (1σ).Item Open Access Analysis of the effect of impact damage on the repairability of CFRP composite laminates.(2017-02) Alzeanidi, Nasser; Ghasemnejad, HessamPolymer composite materials are common in the aerospace application such as aircraft structures including primary and secondary structures. Therefore, there has been an increasing demand for composites in both the military and civilian aircraft industry. At least 50% of the next generation of military and civil aircraft structures are likely to be made from composites. The most important properties for composite materials in aircraft application was the high strength-to-weight ratios, stiffness-to-weight ratios and easy to repair. However, the composite materials have low resistance for impact damage. Impact can lead to significant strength reduction in aircraft structure about 40% to 60% of an undamaged composite laminate strength. Therefore, establish a numerical methodology to defined the optimum repair joint to restore sufficient strength of damaged aircraft composite structures during some operations and exercise activities with limited resources which will be the main contributions to knowledge in this thesis. To achieve this contribution need to understanding of the behaviour of Carbon Fibre Reinforced Plastic (CFRP) composite laminates subject to high velocity impact and the unrepaired composite laminates and repaired (stepped joint) subject to compression after impact test. Therefore, this study consists of two parts:- first, part a combined of numerical simulation and experimental investigation have been used to evaluate the woven CFRP laminate subject high velocity impact. The selected impact velocities were (140m/s, 183m/s, 200m/s, 225m/s, 226m/s, 236m/s, 270m/s, 305m/s, 354m/s and 368m/s) in order to evaluate the induced impact damage in three different thickness of CFRP composite laminates (6 mm, 4.125 mm and 2.625 mm) these velocities were selected according the gas gun limitation. The woven composite laminate made of Hexcel G0926 Carbon Fabric 5 harness 6K, Areal Weight 370 gsm. The resin used was Hexcel RTM 6, cured for 1 hour 40 minutes at 180° C at a pressure of 100 psi, with an average thickness of 0.375mm. The laminates were comprised of 16 layers, using the following stacking sequence: [(0/90); (±45); (±45); (0/90); (±45); (±45); (0/90); (0/90); (±45); (0/90); (±45); (0/90); (±45); (0/90); (±450); (0/90)], 11 layers, using the following stacking sequence: 0/90; ± 45; 0/90; ± 45; 0/90; ±45; 0/90; ± 45; 0/90; ±45; 0/90 and 7 layers, using the following stacking sequence: ± 45; 0/90; 0/90; ±45; 0/90; 0/90; ±45. The density of woven CFRP laminates was 1.512e-3 ±1e-6 grm/mmÖ³. The penetration process and also change of kinetic energy absorption characteristics have been used to validate the finite element results. The experimental and numerical method in this study show a significant damage occurs, including delamination, compression through thickness failure, out-of-plane shear failure and in-plane tensile failure of the fibres located at the rear surface when the projectile penetrates the laminate. The penetration mechanism of the projectile had a “plugging-type” (shear) failure and the hole that was formed after impact was conical in shape were shown in experimental and also verified in the numerical model. The residual kinetic energy in numerical model is 5.0 % larger than experimental data which is significantly matched in all simulated cases. In part two a finite element model is established to optimise the repair joint to restore sufficient strength of damaged composite laminate and used compression after impact test to compare the compression failure load of the sample. In order to achieve this an optimised repair models of stepped lap joints with variable parameters such as number of steps and length of steps have been experiment the undamaged composite laminate and composite laminate subject to high velocity impact and also created a numerical model for these experimental. The experimental CAI failure load of undamaged 7 Plies CFRP composite laminate higher than the failure load of damaged specimens by approximately 23%. The undamaged 11 Plies CFRP composite laminate failed at approximately 40% higher than the damaged specimens. Moreover, the difference between the experimental and numerical results of above tests was about 10%. The numerical model of repaired composite laminate show the damage initiated at the end of overlap and the average compression failure load of the stepped lap joint increased with the increasing of the number of step and length of step. The 85% and 90% of compressive failure load has been restored.Item Open Access Application of compressor water injection for the reduction of civil aircraft NOᵪ emissions.(2018-12) Block Novelo, David Alejandro; Igie, Uyioghosa; Nalianda, DevaiahGas turbine Nitrogen Oxide (NOx) emissions are directly proportional to combustion temperature. These contaminants are associated with respiratory diseases and damage to the local water quality and wildlife. Higher demand on civil aviation, coupled to high-pressure ratio (and thus, temperature-ratio) engines, have caused aviation-borne NOᵪ Gas turbine Nitrogen Oxide (NOx) emissions are directly proportional to combustion temperature. These contaminants are associated with respiratory diseases and damage to the local water quality and wildlife. Higher demand on civil aviation, coupled to high-pressure ratio (and thus, temperature-ratio) engines, have caused aviation-borne NOᵪ emissions to double since 1990. This is of concern around airports, at operations below 3,000 ft. where the concentration of air traffic is high and the population faces direct exposure to engine contaminants. This thesis explores the use of atomized water droplets into an engine compressor as a way of intercooling the cycle and in doing so reducing NOᵪ emissions. The use of water injection is proposed to be applied only during take-off and climb up to 3,000 ft. The analysis of water injection is firstly applied to common turbofan architectures (2 and 3-spool), under varied ambient conditions. The gas turbines are simulated by means of an in-house performance simulating tool, Turbomatch. The changes in cycle temperature when water injection is applied, are accounted for by means of a stand-alone analytical compressor model. The platform calculates the thermodynamic exchange between the gas path of the engine and the water droplets in the Lagrangian frame of reference. The engine models are then integrated into an in-house aircraft performance simulating tool, Hermes. Two types of aircraft, narrow and wide-body, are considered for operations with the water injection system. The performance benefits noted in the stand-alone engine section, are evaluated considering the extra system weight for different missions ranging from 500 to 11,000 km. The observed theoretical trends are then confirmed by means of an experiment performed on a stationary gas turbine. The test includes performance monitoring (pressures, temperatures, mass flows), water droplet measurements, and exhaust emissions analysis. The most optimistic case of water injection shows a reduction of NOx emissions greater than 50%, for the period when water is used. This technology, when applied after the fan compressor, is effective at ambient temperatures as low as 5°C and is more promising in 3-spool engines. For the shortest mission considered, equivalent to a journey from London to Paris, the aircraft benefits from a small fuel saving, despite of the extra weight. For longer missions, there is a negligible fuel penalty (0.05%) derived from the extra payload. In all the cases Landing and Take-Off (LTO) emissions are estimated to be reduced by 42-43%. A reduction of NOx emissions of 25% is achieved experimentally when injecting 2% water-to-air ratio. The study concludes that compressor water injection is a feasible solution that can significantly reduce the environmental footprint of aviation emissions to double since 1990. This is of concern around airports, at operations below 3,000 ft. where the concentration of air traffic is high and the population faces direct exposure to engine contaminants. This thesis explores the use of atomized water droplets into an engine compressor as a way of intercooling the cycle and in doing so reducing NOᵪ emissions. The use of water injection is proposed to be applied only during take-off and climb up to 3,000 ft. The analysis of water injection is firstly applied to common turbofan architectures (2 and 3-spool), under varied ambient conditions. The gas turbines are simulated by means of an in-house performance simulating tool, Turbomatch. The changes in cycle temperature when water injection is applied, are accounted for by means of a stand-alone analytical compressor model. The platform calculates the thermodynamic exchange between the gas path of the engine and the water droplets in the Lagrangian frame of reference. The engine models are then integrated into an in-house aircraft performance simulating tool, Hermes. Two types of aircraft, narrow and wide-body, are considered for operations with the water injection system. The performance benefits noted in the stand-alone engine section, are evaluated considering the extra system weight for different missions ranging from 500 to 11,000 km. The observed theoretical trends are then confirmed by means of an experiment performed on a stationary gas turbine. The test includes performance monitoring (pressures, temperatures, mass flows), water droplet measurements, and exhaust emissions analysis. The most optimistic case of water injection shows a reduction of NOᵪ emissions greater than 50%, for the period when water is used. This technology, when applied after the fan compressor, is effective at ambient temperatures as low as 5°C and is more promising in 3-spool engines. For the shortest mission considered, equivalent to a journey from London to Paris, the aircraft benefits from a small fuel saving, despite of the extra weight. For longer missions, there is a negligible fuel penalty (0.05%) derived from the extra payload. In all the cases Landing and Take-Off (LTO) emissions are estimated to be reduced by 42-43%. A reduction of NOx emissions of 25% is achieved experimentally when injecting 2% water-to-air ratio. The study concludes that compressor water injection is a feasible solution that can significantly reduce the environmental footprint of aviation.Item Open Access Application of data analytics for predictive maintenance in aerospace: an approach to imbalanced learning.(2021-05) Dangut, Maren David; Jennions, Ian K.; King, SteveThe use of aircraft operational logs to predict potential failure that may lead to disruption poses many challenges and has yet to be fully explored. These logs are captured during each flight and contain streamed data from various aircraft subsystems relating to status and warning indicators. They may, therefore, be regarded as complex multivariate time-series data. Given that aircraft are high-integrity assets, failures are extremely rare, and hence the distribution of relevant data containing prior indicators will be highly skewed to the normal (healthy) case. This will present a significant challenge in using data-driven techniques to 'learning' relationships/patterns that depict fault scenarios since the model will be biased to the heavily weighted no-fault outcomes. This thesis aims to develop a predictive model for aircraft component failure utilising data from the aircraft central maintenance system (ACMS). The initial objective is to determine the suitability of the ACMS data for predictive maintenance modelling. An exploratory analysis of the data revealed several inherent irregularities, including an extreme data imbalance problem, irregular patterns and trends, class overlapping, and small class disjunct, all of which are significant drawbacks for traditional machine learning algorithms, resulting in low-performance models. Four novel advanced imbalanced classification techniques are developed to handle the identified data irregularities. The first algorithm focuses on pattern extraction and uses bootstrapping to oversample the minority class; the second algorithm employs the balanced calibrated hybrid ensemble technique to overcome class overlapping and small class disjunct; the third algorithm uses a derived loss function and new network architecture to handle extremely imbalanced ratios in deep neural networks; and finally, a deep reinforcement learning approach for imbalanced classification problems in log- based datasets is developed. An ACMS dataset and its accompanying maintenance records were used to validate the proposed algorithms. The research's overall finding indicates that an advanced method for handling extremely imbalanced problems using the log-based ACMS datasets is viable for developing robust data-driven predictive maintenance models for aircraft component failure. When the four implementations were compared, deep reinforcement learning (DRL) strategies, specifically the proposed double deep State-action-reward-state-action with prioritised experience reply memory (DDSARSA+PER), outperformed other methods in terms of false-positive and false-negative rates for all the components considered. The validation result further suggests that the DDSARSA+PER model is capable of predicting around 90% of aircraft component replacements with a 0.005 false-negative rate in both A330 and A320 aircraft families studied in this researchItem Open Access Application of mechanical surface treatments to improve fatigue crack growth life of aircraft fuselage materials(2018-06) Liu, Yao; Ganguly, SupriyoMechanical treatment for surface processing is a cost-effective tool and has the potential to improve the dynamic strength of a component or structure significantly through creation of a residual compressive stress state. This research is aimed to investigate mechanical surface processing treatments, e.g. deep surface rolling, machine hammer peening, in aircraft fuselage structural alloys to reduce fatigue crack growth rate and improve damage tolerance. The study also revealed that such processing could be used effectively to improve damage tolerance properties of such safety critical structures. However, optimisation of such processes is important as distortion from the processing would need to be minimised, to maximise the benefit from the residual compressive stress field. This thesis focuses on the application of deep surface rolling to understand the underpinning interaction between stress states and a long fatigue crack under a variably distributed residual stress field. Centre notch of 8 mm length were machined in Middle-tension M(T) specimens of 1.6 mm thickness 2024-T351 and 2524-T351 clad aluminium alloys. The M(T) specimens were locally rolled by a deep surface rolling process to create a spatially resolved compressive residual stress fields on both sides of the notch and under different loads. Prior to application of deep surface rolling on the M(T) specimens, the process was trialled on similar thickness specimens to ensure minimum distortion so that it can be applied on both the surfaces. The spatial position of the DSR patches with respect to the crack tip were varied to understand the interaction of the stress field on crack propagation and how the benefit of the process can be maximised. Following rolling of M(T) specimens, fatigue testing were performed at a stress ratio R = 0.1 and maximum stress of 100 MPa. A three-dimensional finite-element (FE) model of the DSR process was developed to predict the residual stress field and distortion. This model was validated with experimentally measured residual stress data and distortion. An analytical method based on experimental residual stress data, was developed to determine the residual stress intensity factor (Kres). The crack closure behaviour was taken account for the prediction of the fatigue crack growth rate (FCGR). Despite formation of a compressive residual stress (CRS) field through the thickness below the DSR patch it was found that improvement of fatigue performance depends on the location of the patch with respect to the crack tip. It was observed that the rolling load parameters and distance from the crack tip are vital in the reduction of crack propagation behaviour. The former balances the stress field and distortion while the later determines the crack driving force, when the crack enters the compressive residual stress field, and a large distance between the crack tip and stress field will cause acceleration of the crack before it enters the compressive stress field. The analytical method of computing Kres was successfully contributed to the prediction of FCGR and showed good agreements with experiments. In a further study, the analytical method was used to calculate Kres by using the predicted residual stress field from FEA (finite element analysis). Based on the predicted Kres, the predicted FCGR showed a good agreement with experiments as well. The application of DSR to the metal fatigue enhancement is significantly effective and cost-effective. By optimising DSR process to intentionally treat the high possibility of fatigue damage region, the fatigue life can be significantly enhanced, resulting in improvement in damage tolerant design of aerospace structures or components.Item Open Access Areal artefact manufacturing using SPDT(2018-08) Zhao, Junguo; Giusca, Claudiu; Goel, SauravWith the increasing importance of the surface engineering, surface topography measuring instrument has been used in wider range of applications, which requires trustworthy calibration process to deliver traceability so that the instrument is able to give comparable and reliable measurement. The calibration standard / artefact is designed to transfer traceability easily and reliably. In current market, the feature of the artefact used for evaluation the surface topography measuring process are not sufficiently accurate. This insufficiency may be solved by using certain types of calibration standard specified in ISO standard however they are not commercially produced. In this project, one of the desired types called ‘radial sinusoidal shape’ was produce by SPDT (single point diamond turning) manufacturing method. The feature parameters of the artefact are designed to meet the instrument measurement requirement and the machining path is generated with consideration of the tooling geometry. To assess the repeatability in z direction of the turning machine, a step height experiment was designed and conducted. The measurement result indicates that the repeatability of the machine is unsatisfactory when the feed distance smaller than 100 nm. The wavelength and the amplitude of machined radial sinusoidal shape was measured by stylus profiler, followed by the measurement uncertainty analysis. The measurement result was compared with the design to evaluate quality of the manufacturing process. To estimate the systematic error of the profiler, CCI was used to measure the machined radial sinusoidal shape. The measurement result was also compared with the design.Item Open Access Artificial immune systems for case based reasoning of unmanned aircraft flight data(2017-09) Pelham, Jonathan Gerald; Fan, Ip-Shing; McFeat, Jim; Jennions, Ian K.UAS(Unmanned Aerial Systems) mishaps are high, and their pilots face many control challenges. The reliability of UAS has been seen as a dominant mishap cause but in several instances the aircraft could have been saved if the health state of the aircraft had been understood at an earlier point by the pilot. Manned and unmanned aircraft pilots both benefit from the use of their own experience in the detection and mitigation of faults during flight. However it has been suggested that pilots within a GCS(Ground Control Station) face difficulties in maintaining their situational awareness due to the nature of remote control. The use of a cognitive framework as a basis for case based reasoning is suggested as a way to integrate through life learning into the Safety Management System. The population of the case base for such a system would require a large investment of time to create. The use of machine learning is suggested and evaluated to address this issue by generating cases for CBR. This has seen some success and even the use of an AIS(Artificial Immune System) in this thesis. An AIS was used in order to try to address the problem of cost and time caused by high pre-processing required by common machine learning methods. A simulation of the Aerosonde UAS was created and multiple flights simulated to build up a set of representative set flight data. Several fault cases were included in the simulated flights of varying severities. Different machine learning schemes were evaluated using the data set and their effectiveness compared in order to evaluate the ability of the algorithm to learn from flight data without extensive pre-processing. The complex dataset made the problem difficult but in analysis the AIS performed slightly better than the neural network with which it was compared. In due time and with development it's computational cost could be reduced and its effectiveness increased. The benefit of an automated method to learn from aircraft incidents and mishaps can readily be seen in a fleet scenario where it would be uneconomical to analyse flight data of unmanned aircraft in the same way that it would be done for manned aircraft. This semi-supervised approach reduces personnel requirements and enhances the ability of operators to learn from mishaps by relating mishap cases to the current situation and being transparent in their alerting criteria.Item Open Access Assumption management in model-based systems engineering: an aircraft design perspective.(2021-12) El Fassi, Soufiane; Riaz, Atif; Guenov, Marin D.Early design of complex systems is characterised by significant uncertainty due to lack of knowledge, which can impede the design process. In order to proceed with the latter, assumptions are typically introduced to fill knowledge gaps. However, the uncertainty inherent in the assumptions constitutes a risk to be mitigated. In fact, assumptions can negatively impact the system if they turn out to be invalid, such as causing system failure, violation of requirements, or budget and schedule overruns. Within this context, the aim of this research was to develop a computational approach to support assumption management in model-based systems engineering, with an explicit consideration of the uncertainty in assumptions. To achieve the research aim, the objectives were to: (1) devise methods to enable assumption management in a model-based design environment; and (2) devise methods to manage risk of change due to invalid assumptions, with an explicit consideration of both assumptions and margins. The scope was limited to the early stages of aircraft design. To evaluate this research, a demonstration was performed based on two use cases to assess whether the methods work as intended. The developed methods were demonstrated to industry experts in order to obtain feedback on expected usefulness in practice, thus assessing the impact of this research. The experts concluded that the proposed methods are innovative, useful and relevant to industry, where these methods can lead to: (i) fewer undesired iterations, due to earlier identification and management of risks associated with assumptions; and (ii) a better margin balance, due to timely and interactive margin revision. Future work includes further industrial evaluation, extending the research scope and studying the scalability and associated costs of the proposed methods.Item Open Access Authoring digital contents for augmented reality in maintenance.(2018-04) Palmarini, Riccardo; Erkoyuncu, John Ahmet; Roy, RajkumarTechnicians’ performance is a major driver in maintenance and each process can be prone to time and quality variances as well as errors due to factors such as experience, complexity and environment. Augmented Reality (AR) is an emerging technology that has been applied in a wide variety of disciplines and has been demonstrated to have a role with improving efficiency, effectiveness and decision-making within industrial maintenance. AR has not reached its full potential yet and its implementation in Industry is slowed down by three main limitations: hardware restricted capabilities, object recognition robustness and contents-related issues. This PhD project focuses on easing the implementation of AR by overcoming the AR technology selection challenges and the AR contents-related issues. In order to reach the aim, the student has provided three main contributions to knowledge: 1) a process to select AR technology for maintenance (IPSAR), 2) a method for creating AR step-by-step procedures (FARP) and 3) a method for providing remote assistance (ARRA). FARP and ARRA methods have been developed and tested. The first allows recording procedures in an ad-hoc designed “AR-format” and is able to show “step-by-step” procedures. It aims to support deskilling the maintenance process and reducing the error rate by simplifying the delivery of maintenance with efficient and effective guidance. The second overcomes current remote video-call assistance limitations by improving spatial referencing. ARRA module allows to provide AR-assistance by overlaying virtual objects on the real environment of a remote maintainer. The methods proposed by the student could boost the implementation of AR and open the doors for a bright future in which AR supports technicians thus reducing operational costs and training and improving human performances.Item Open Access A basis for the representation, manufacturing tool path generation and scanning measurement of smooth freeform surfaces(2017-02) Morantz, Paul; Shore, PaulFreeform surfaces find wide application, particularly in optics, from unique single-surface science programmes to mobile phone lenses manufactured in billions. This thesis presents research into the mathematical and algorithmic basis for the generation and measurement of smooth freeform surfaces. Two globally significant cases are reported: 1) research in this thesis created prototype segments for the world’s largest telescope; 2) research in this thesis made surfaces underpinning the redefinition of one of the seven SI base units – the kelvin - and also what will be the newly (and permanently) defined value for the Boltzmann constant. Theresearchdemonstratestwounderlyingphilosophiesofprecisionengineering, the critical roles of determinism and of precision measurement in precise manufacturing. The thesis presents methods, and reports their implementation, for the manufacture of freeform surfaces through a comprehensive strategy for tool path generation using minimum axis-count ultra-precision machine tools. In the context of freeform surface machining, the advantages of deterministic motion performance of three-axis machines are brought to bear through a novel treatment of the mathematics of variable contact point geometry. This is applied to ultra-precision diamond turning and ultra-precision large optics grinding with the Cranfield Box machine. New techniques in freeform surface representation, tool path generation, freeform tool shape representation and error compensation are presented. A comprehensive technique for very high spatial resolution CMM areal scanning of freeform surfaces is presented, with a new treatment of contact error removal, achieving interferometer-equivalent surface representation, with 1,000,000+ points and sub-200 nm rms noise without the use of any low-pass filtering.Item Open Access Behaviour monitoring: investigation of local and distributed approaches(2017) Turchi, Dario; Shin, Hyo-Sang; Tsourdos, AntoniosNowadays, the widespread availability of cheap and efficient unmanned systems (either aerial, ground or surface) has led to significant opportunities in the field of remote sensing and automated monitoring. On the one hand, the definition of efficient approaches to information collection, filtering and fusion has been the focus of extremely relevant research streams over the last decades. On the other hand, far less attention has been given to the problem of ‘interpreting’ the data, thus implementing inference processes able to, e.g., spot anomalies and possible threats in the monitored scenario. It is easy to understand how the automation of the ‘target assessment’ process could bring a great impact on monitoring applications since it would allow sensibly alleviating the analysis burden for human operators. To this end, the research project proposed in this thesis addresses the problem of behaviour assessment leading to the identification of targets that exhibit features “of interest”. Firstly, this thesis has addressed the problem of distributed target assessment based on behavioural and contextual features. The assessment problem is analysed making reference to a layered structure and a possible implementation approach for the middle-layer has been proposed. An extensive analysis of the ‘feature’ concept is provided, together with considerations about the target assessment process. A case study considering a road-traffic monitoring application is then introduced, suggesting a possible implementation for a set of features related to this particular scenario. The distributed approach has been implemented employing a consensus protocol, which allows achieving agreement about high-level, non-measurable, characteristics of the monitored vehicles. Two different techniques, ‘Belief’ and ‘Average’ consensus, for distributed target assessment based on features are finally presented, enabling the comparison of consensus effects when implemented at different level of the considered conceptual hierarchy. Then, the problem of identifying targets concerning features is tackled using a different approach: a probabilistic description is adopted for the target characteristics of interest and a hypothesis testing technique is applied to the feature probability density functions. Such approach is expected to allow discerning whether a given vehicle is a target of interest or not. The assessment process introduced is also able to account for information about the context of the vehicle, i.e. the environment where it moves or is operated. In so doing the target assessment process can be effectively adapted to the contour conditions. Results from simulations involving a road monitoring scenario are presented, considering both synthetic and real-world data. Lastly, the thesis addresses the problem of manoeuvre recognition and behaviour anomalies detection for generic targets through pattern matching techniques. This problem is analysed considering motor vehicles in a multi-lane road scenario. The proposed approach, however, can be easily extended to significantly different monitoring contexts. The overall proposed solution consists in a trajectory analysis tool, which classifies the target position over time into a sequence of ‘driving modes’, and a string-matching technique. This classification allows, as result of two different approaches, detecting both a priori defined patterns of interest and general behaviours standing out from those regularly exhibited from the monitored targets. Regarding the pattern matching process, two techniques are introduced and compared: a basic approach based on simple strings and a newly proposed method based on ‘regular expressions’. About reference patterns, a technique for the automatic definition of a dictionary of regular expressions matching the commonly observed target manoeuvres is presented. Its assessment results are then compared to those of a classic multi-layered neural network. In conclusion, this thesis proposes some novel approaches, both local and distributed, for the identification of the ‘targets of interest’ within a multi-target scenario. Such assessment is solely based on the behaviour actually exhibited by a target and does not involve any specific knowledge about the targets (analytic dynamic models, previous data, signatures of any type, etc.), being thus easily applicable to different scenarios and target types. For all the novel approaches described in the thesis, numerical results from simulations are reported: these results, in all the cases, confirm the effectiveness of the proposed techniques, even if they appear to be open to interpretation because of the inherent subjectivity of the assessment process.Item Open Access Bioinspired symmetry detection on resource limited embedded platforms(2017-07) Elliott, Alexander William; Zbikowski, RafalThis work is inspired by the vision of flying insects which enables them to detect and locate a set of relevant objects with remarkable effectiveness despite very limited brainpower. The bioinspired approach worked out here focuses on detection of symmetric objects to be performed by resource-limited embedded platforms such as micro air vehicles. Symmetry detection is posed as a pattern matching problem which is solved by an approach based on the use of composite correlation filters. Two variants of the approach are proposed, analysed and tested in which symmetry detection is cast as 1) static and 2) dynamic pattern matching problems. In the static variant, images of objects are input to two dimentional spatial composite correlation filters. In the dynamic variant, a video (resulting from platform motion) is input to a composite correlation filter of which its peak response is used to define symmetry. In both cases, a novel method is used for designing the composite filter templates for symmetry detection. This method significantly reduces the level of detail which needs to be matched to achieve good detection performance. The resulting performance is systematically quantified using the ROC analysis; it is demonstrated that the bioinspired detection approach is better and with a lower computational cost compared to the best state-of-the-art solution hitherto available.Item Open Access Biomimetic vision-based collision avoidance system for MAVs.(2017-05) Isakhani, Hamid; Aouf, Nabil; Whidborne, James F.This thesis proposes a secondary collision avoidance algorithm for micro aerial vehicles based on luminance-difference processing exhibited by the Lobula Giant Movement Detector (LGMD), a wide-field visual neuron located in the lobula layer of a locust’s nervous system. In particular, we address the design, modulation, hardware implementation, and testing of a computationally simple yet robust collision avoidance algorithm based on the novel concept of quadfurcated luminance-difference processing (QLDP). Micro and Nano class of unmanned robots are the primary target applications of this algorithm, however, it could also be implemented on advanced robots as a fail-safe redundant system. The algorithm proposed in this thesis addresses some of the major detection challenges such as, obstacle proximity, collision threat potentiality, and contrast correction within the robot’s field of view, to establish and generate a precise yet simple collision-free motor control command in real-time. Additionally, it has proven effective in detecting edges independent of background or obstacle colour, size, and contour. To achieve this, the proposed QLDP essentially executes a series of image enhancement and edge detection algorithms to estimate collision threat-level (spike) which further determines if the robot’s field of view must be dissected into four quarters where each quadrant’s response is analysed and interpreted against the others to determine the most secure path. Ultimately, the computation load and the performance of the model is assessed against an eclectic set of off-line as well as real-time real-world collision scenarios in order to validate the proposed model’s asserted capability to avoid obstacles at more than 670 mm prior to collision (real-world), moving at 1.2 msˉ¹ with a successful avoidance rate of 90% processing at an extreme frequency of 120 Hz, that is much superior compared to the results reported in the contemporary related literature to the best of our knowledge.Item Open Access Boundary layer ingestion performance assessments with application to business jets.(2018-07) Sanders, Drewan S.; Laskaridis, PanagiotisAdvancements in propulsion system performance are reliant on improvements in propulsive efficiency, through increases in turbofan bypass ratio. This requires larger nacelle diameters, which incur external aerodynamic penalties. Business jets cruise at high subsonic Mach numbers, and are therefore normally propelled by high specific thrust turbofans. The business jet may benefit from a BLI propulsion system, whereby the specific thrust may be reduced without incurring such heavy penalties in external drag rise. The aim of the research is to perform a design exploration study on BLI applied to a business jet, with emphasis on external aerodynamics. Methods are developed to thoroughly analyse aerodynamic coupling between propulsor and airframe. A multi-physics, control-volume based approach led to the development of near-field momentum-based, far-field momentum-based and energy-based net-vehicle-force formulations. The former two, allowed for a set of thrust-force accounting systems to be defined. Energy-based methods, allowed for flow-field decompositions into different physical mechanisms. These include flow phenomena internal and external to the jet plume. The practical implications associated with applying these methods to RANS CFD solutions, is examined. This hinges around viscous stress tensor field continuity in the flow domain. It was found that the k — w SST turbulence model, along with a Green-Gauss Cell-Based gradient scheme, produced a continuous viscous stress tensor field. Having resolved this, the assessment methods were applied to solutions of non-propelled and propelled bodies. These methods were applied to control volumes having varying extents, which showed the far-field momentum-based method to be sensitive to spurious affects. The energy-based formulation, on the other hand, was observed to be relatively insensitive spurious affects. Good agreement (within 4%) was found between the forces predicted by all three methods over a non-propelled body. A very close agreement was observed between far-field momentum-based and energy-based results (within 1%) over the propelled body. However, much larger discrepancies were observed when compared against the near-field results. This was attributed to the increase in flow-field complexity, which now contained BL, shock and jet interaction regions. A design exploration study was performed by retrofitting a business jet with a fuselage concentric propulsor, powered by the baseline podded engines. A preliminary parametric study was first performed to gauge conditions favourable to BLI benefit. A ram drag approach to modelling BLI benefit was based on a flat plate analogy to obtain boundary layer profiles. Thrust-split, BLR, fan efficiency and intake pressure recoveries, were varied parametrically to asses potential benefits. An optimum SFC benefit between 5-7.5% was achieved at thrustsplits between 30-35%, when ingesting 65-90% of the BL thickness. This guided the the parametric CFD studies, where two tail-cone positions were examined. The first was placed at the top of the tail-cone, and the second positioned midway along the tail-cone. Benefits were only realised for the latter, where a 3-4% improvement in SFC was realised for a thrust-split around 20%, by ingesting 40% of the BL thickness. Energy breakdowns and decompositions were performed on all of the cases. One of the significant outcomes of this research was revealing that a significant proportion of the thrust force may be attributed to the isentropic expansion region within the jet plume's core.Item Open Access Broad band anti-reflection 2-16μm coatings on diamond and ZnSe.(2018-08) Hakim Khalili, Mohammad; Endrino Armenteros, JoseIn this project, Thin Metal Films an optical coating company is requiring the upgrade of a desktop research type radio frequency magnetron sputtering machine which was custom made for a European funded project and was gifted to it after the project completion. As the machine had no usage history, it had to be investigated and its capabilities and characteristics identified. The first part of the project, focused on restoring the machine using a systematic approach by utilising a technique called Plan Do Check Act, where a continuous feedback loop is used to identify problems and finding solution to them. Within the restoration process the aim was to improve the coating uniformity and machine’s repeatability. The restoration of the machine consumed almost the entire project duration leaving a little time for second part of the project. The second part was to use the restored machine, to develop single and multilayer Broad Band Anti-Reflective Coatings for Zinc Selenide and diamond optics. The optics are to be used in Attenuated Total Reflectance accessory units for Infrared spectroscopy. Based on the literature review and multiple decision criteria, yttria and ytterbia were showing promising results in adhering very well to a diamond surface and improving its transmission. Therefore, using Macleod software, a single layer quarter wavelength coating was designed and theoretically examined. It was found that a quarter wavelength ytteria at 5000 nm and a quarter wavelength ytterbia at 3370 nm can increase the transmission on both ZnSe and diamond optics to maximum of 98% and 95% accordingly. The coatings have not actually been deposited on the optics due to both targets breaking before conducting the actual experiments.