Deleterious effects such as surface heating and turning moments can arise as a consequence of the impingement of thruster plumes with spacecraft surfaces. Such thrusters are normally fired for attitude control. The prediction of such effects must be undertaken at the design stage of the vehicle. In this study, the modelling of such plumes was undertaken. The following prediction techniques were implemented into computer programmes: (a) the Simons model, (b) the Method of Characteristics (MOC), and (c) the Direct Simulation Monte Carlo method (DSMC). The first two methods are derived from continuum equations whilst the third adopts a discrete particle approach. Several DSMC schemes exist for treating the collisional behaviour of the gas, and it was unclear which would be best suited for the intended application. A thorough assessment of the implementation and performance of several such schemes was therefore completed. Having determined the most suitable DSMC scheme, the three modelling techniques were then applied to the isentropic core expansion of a small hydrazine thruster plume. It was found that significant errors occur in the determination of impingement quantities through application of the continuum methods in the flow regime lying between the continuum and free molecular limits. The DSMC technique was also used to calculate the nozzle lip and backflow expansion regions of the same hydrazine thruster. A significant degree of backflow was found with flow angles of up to 140º. The sensitivity of the calculations to the conditions initially assumed were assessed and found to be important.

Certain civilian uses of the Global Positioning System (GPS) now demand levels of precision exceeding the original military specifications for the system. For example, at present, GPS is used in surveying, navigation and monitoring applications. Although GPS has been used previously to monitor the structural integrity of civilian structures such as bridges and towers, the experimental conditions have one significant difference: These structures are all immobile allowing for fixed references to be used. This research’s ultimate aim was to conduct a feasibility study into the use of GPS, and of the soon to be deployed Galileo constellation, in structural monitoring of maritime vessels. Under these conditions, a fixed base station as was used in previous structural monitoring is not a valid approach. New techniques, or variations on the current ones, were examined to deal with this lack of a fixed reference. This thesis considers the simulator that was developed in the early phases of the project and the results that were generated. An expanded version of the simulator is then explained, alongside the processes used to model structural deformation experienced by a ship at sea. Frequency analysis of the simulator results is also performed and the results detailed. It is found that frequency-domain analysis allows for the identification of different movements seen on the structure. In addition, an analysis of the impact the deployment of the Galileo constellation is conducted using simulation. This simulation found a drop of about 20% in Dilution of Precision (DOP) over several areas. This thesis then examines some of the field work, that aimed to measure rigidbody motion, that was conducted onboard Red Jet 4, a catamaran-type vessel. The experimental setup is described, and the results are briefly considered. A method for determining the optimal configuration of the receiver network is also proposed alongside an algorithm to detect plastic deformation of a vessel structure. Finally, this report considers the original project objectives and how these were or were not met. Should sufficient accuracy be achieved by a GPS receiver network, it is possible to use the system to monitor ship structures.

Noise generated by aircraft landing gears is a major contributor to the overall airframe noise of a commercial aircraft during landing approach. Because of the complex geometry of landing gears, the prediction of landing gear noise has been very difficult and currently relies on empirical tools, which have limited reliability and flexibility on the applications of unconventional gear architectures. The aim of this research is to develop an efficient and accurate numerical method to investigate the generation and far field radiation of the landing gear noise. In this thesis a hybrid approach is developed that combines near field flow computations with an integral radiation model to enable the far field signal to be evaluated without the need to directly resolve the propagation of the acoustic waves. The recent advances in the CAA methods are implemented with high-order finite difference compact schemes and a characteristics-based multi-block interface treatment. Aerodynamic noise from a generic two-wheel landing gear model, provided by Airbus LAGOON (landing gear noise database for CAA validation) program, is predicted by using the hybrid approach and compared with the LAGOON database. The unsteady flow field is computed by using a compressible Navier-Stokes solver based on high-order finite difference schemes. The calculated time history of surface data is used in a FW-H solver to predict the far field noise levels. Both aerodynamic and aeroacoustic results are compared with wind tunnel measurements in good agreement. Individual contributions from three components, i.e. wheels, axle and strut of the landing gear model are also investigated to identify the major noise source component. It is found that strong flow-body interaction noise is generated by the flow separated from tire rim impinging on the axle. Based on the same landing gear model, the comparison study using conventional CFD solver FLUENT is performed with a second-order Navier-Stokes finite volume solver to compute the unsteady near field flow and the built-in FW-H solver to calculate the far field sound propagation. The comparison suggests that although conventional CFD method can obtain good timeaveraged aerodynamic results, its ability of predicting sound radiation is limited by the inherent low-order numerical discretizations. The aerodynamic noise from the isolated undercarriage wheel with detailed hub configuration is also investigated using FLUENT. The asymmetric phenomenon in the mean flow is discovered in the wake region of the wheel, which contributes to a positive lift force for the wheel. It is predicted that the isolated wheel radiates relatively strong noise to the sides with several strong tonal noise.

In Sri Lanka, urban hospitals offer specialized healthcare services, while rural hospitals have limited services and normally offer only basic hospital facilities. Therefore people in rural areas have to travel relatively long distances to urban hospitals which offer specialized healthcare. This is expensive, while the poor transportation systems in rural areas make travel to urban hospitals time-consuming often involve an overnight stay. The aim of this research is to improve access to healthcare by adopting telemedicine in rural areas where the patient and the clinician in the rural hospital can contact the consultant in the urban hospital using audio, video and data communication methods for specialized healthcare services. The methods used were questionnaires and interviews in the exploratory study, which had small number of participants (83) covering three regions, followed by the main survey which had large number of participants (225) and ten regions. A model for adoption of a telemedicine system in Sri Lanka was developed from the analysis of the literature and the exploratory study. The exploratory study was conducted in three districts of Sri Lanka involving clinicians, hospital staff and the general public from both rural and urban areas, and Health Ministry officials, medical directors and consultants. The major study was conducted in ten districts and on two islands of Sri Lanka to identify the factors in the model which influenced the adoption of telemedicine. The findings of the main survey clearly indicate how important the telemedicine solution is to the rural population. The results also show that hospital staff and the general public were very much in favour of adopting telemedicine. The results also indicate that both clinicians and hospital staff believe that staff involvement, and introducing new policies and standards, will influence the adoption of telemedicine. Clinicians believe that Internet connectivity and the equipment used will not have an effect on introducing telemedicine. This model will be beneficial when implementing a telemedicine system into rural areas of Sri Lanka, which will provide patients with access to specialized healthcare services.

The Earth’s surface bears the scars of 4.5 billion years of bombardment by asteroids, despite most having been erased by tectonic activity and erosion. Asteroids predominantly orbit the Sun in the asteroid belt between Mars and Jupiter, but a large number occupy orbits close to the Earth’s. These bodies are termed Near Earth Objects (NEOs) and they present a very real impact threat to the Earth. In 1998 NASA inaugurated the ‘Spaceguard Survey’ to catalogue 90% of NEOs greater than 1 km in diameter. The smaller bodies, meanwhile, remain undetected and far more numerous. In order to understand the NEO hazard, the consequences resulting from an asteroid impact require modelling. While the atmospheric entry of asteroids is a critical part of the impact process, it is the surface impact which is most important, both onto land and into the oceans. It is the impact generated effects (IGEs) that are hazardous to human populations on the Earth and the infrastructure they occupy. By modelling these IGEs and the consequences they present for humans and infrastructure, an understanding of the global vulnerability to the hazard is developed. ‘NEOimpactor’ is the software solution built to investigate the global vulnerability to NEO impacts. By combining existing mathematical models which describe the impact and effects, a unified impact simulator tool has been developed with the capacity to model the real consequences of any terrestrial impact. By comparing the consequences of multiple impact events, a complete vulnerability assessment of the global NEO hazard is derived. The result maps are designed for ease of dissemination to explain the impact risk to a non-specialist audience. The system has identified China, US, India, Japan and Brazil as facing the greatest overall risk, as well as indicating the various factors influencing vulnerability. The results can be used for informing the international decision making processes regarding the NEO hazard and potential mitigation strategies.

Higher education increasingly emphasises the importance of learner self-regulation and autonomy. Self-regulated learners are active participants in their own learning and employ strategies for sustaining motivation, metacognitive thinking, and self-monitoring. This work identifies four central aspects which are investigated in an individual project based learning setting, namely motivation, time management, progress awareness, and monitoring. Monitoring is the key driver of learner self-regulation. Time management has proven to enhance perceived control over time, health, and academic achievement. Progress awareness supports learner self-observation and self-evaluation. Finally, monitoring is the process of generating feedback both internally (own feedback) and externally (from others). In this work, a 17-week quasi-experimental study involving 378 participants was conducted, preceded by a less successful trial. The study employed a web-based monitoring system, combined with a monitoring scheme in the context of Master of Science (MSc) summer projects in the School of Electronics and Computer Science at the University of Southampton, implementing a set of features for each learning aspect. In the monitoring scheme, monitors met with groups of students weekly to monitor project progress. Feedback was submitted online by students, monitors, and supervisors. It is shown that there are positive and significant relationships between feature use and weekly student progress and motivation ratings, and also with their dissertation mark. This suggests that some system features enhanced student self-motivation beliefs, self observation, and self-reflection. Also, features were ranked as to their impact on student self-regulated learning, and a narrative case study exploring processes behind the effects is provided. High impact features were graphical student progress visualisations, a ranking table, the virtual project page listing past feedback and providing project management tools, and weekly progress feedback. Evidence for between-monitor effects on student dissertation mark as well as system use and successful system support for information exchange is also presented. The contribution of this thesis is novel and noteworthy since it (1) shows the effects of web-based monitoring features on self-regulated learning, (2) shows how features can be used for implementing principles of good practice, and (3) draws on the effect of monitors in the context of this study.

The application of Riemann Methods formulated in the Lagrangian reference frame to the numerical simulation of non-linear events in solid materials is investigated. Here, solids are characterised by their ability to withstand shear distortion since they possess material strength. In particular, numerical techniques are discussed for simulating the transient response of solids subjected to extreme loading. In such circumstances, the response of solids will often be highly non-linear, displaying elastic and plastic behaviour, and even moderate compressions will produce strong shock waves. This work reviews the numerical schemes or 'hydrocodes' which have been adopted in the past in order to simulate such systems, identifying the advantages and limitations of such techniques. One of the most prominent limitations of conventional Lagrangian methods is that the computational mesh or grid has fixed-connectivity i.e. mesh nodes are connected to the same nodes for all time. This has significant disadvantages since the computational mesh can easily become tangled as the simulated material distorts. The majority of conventional hydrocodes are also constructed using outdated artificial viscosity schemes which are known to diffuse shock waves and other steep features which may be present in the solution. In the work presented here, a novel two-dimensional Lagrangian solver has been developed Vucalm-EP which overcomes many of the limitations of conventional techniques. By employing the Free-Lagrange Method, whereby the connectivity of the computational mesh is allowed to evolve as the material distorts, problems of arbitarily large deformation can be simulated. With the implementation of a spatially second-order accurate, finite-volume, Godunov-type solver, non-linear waves such as shocks are represented with higher resolution than previously possible with contemporary schemes. The Vucalm-EP solver simulates the transient elastic-perfectly plastic response of solids and displays increased accuracy over alternative Lagrangian techniques developed to simulate large material distortion such as Smoothed particle Hydrodynamics (SPH). Via a variety of challenging numerical simulations the Vucalm-EP solver is compared with contemporary Euler, fixed-connectivity Lagrangian, and meshless SPH solvers. These simulations include the solution of one- and two dimensional shock tube problems in aluminium, simulating the collapse of cylindrical shells and modelling high-velocity projectile impacts. Validation against previously published results, solutions obtained using alternative numerical techniques and analytical models illustrates the versatility and accuracy of the technique. Thus, the Vucalm-EP solver provides a numerical scheme for the Lagrangian simulation of extensive material distortion in materials with strength, which has never previously been possible with mesh-based techniques.

Over 120 cases of on-orbit breakups have now been recorded. Many more undetected events are believed to have occurred. Each time an object breaks up, whether by explosion or collision, a cloud of debris is formed. The overall objective of the PhD is to examine the interaction between the debris clouds produced by on-orbit fragmentation events and specific space systems. A breakup event will give rise to concentrations of debris which, for some time after the event, will have spatial densities considerably higher than the background flux. Thus, a detailed knowledge of the extent to which the cloud will grow over a given time period, and an accurate assessment of the risk of collision for a spacecraft passing through it, may prove to be important in mission planning and satellite shielding design. The SDS (Space Debris Simulation) software suite has been developed to carry out the analysis presented in this thesis and now represents the state-of-theart in debris cloud modelling. The integrated structure of the developed software enables a wide variety of analyses to be conducted and simulations of both historic and potential future orbital fragmentation events to be performed. Program BREAKUP uses a combination of empirical and analytical models to simulate catastrophic and non-catastrophic collisions, and also variable intensity explosive fragmentations. Included in BREAKUP is a novel parametric model for producing and controlling non-isotropic fragment spreads. TRAJECTORY acts as a test-bed for orbit propagation techniques, providing the facility for convenient and direct method comparison. EVOLUTION enables the complex dynamics of debris cloud growth to be visualised and in particular the effects of propagation method to be examined. Program TARGET employs a novel implementation of the method of probabilistic continuum dynamics to perform collision hazard assessments for spacecraft which encounter debris clouds. Among the additional new developments included in TARGET are the consideration of atmospheric drag, a direct interface with a non-isotropic cloud model, the use of a cellular target spacecraft representation and impact energy-related damage assessment algorithm, and a built-in satellite constellation analysis facility. A number of case studies are presented to illustrate the modelling capabilities of the SDS software suite, including the simulation of several historic fragmentation events and the debris cloud collision risks to ENVISAT-1 and the Iridium™ satellite constellation. The results produced by the models are validated by comparisons with other simulation software and, wherever possible, with actual breakup event, debris impact and spacecraft, orbit, data.

Adaptive hypermedia systems, such as intelligent tutoring systems, aim to reduce reliance upon a teacher. However, such systems have some drawbacks such as inconsistency when estimating a learner‟s knowledge level, and a lack of a pedagogically informed approach to teaching and learning. These drawbacks may be addressed by a competency model. Such a model has the benefits of an improved pedagogical approach to e-learning and a more consistent profile of learners‟ competences. Such a model also renders competences machine processable, sharable, and modifiable. The aim of this research is to investigate and design a competence-based system which provides appropriate study materials from the Web to the learner without any intervention from the teacher. Each step within the system for deriving the study material links from the learners‟ competences is described in detail. A competence structure is designed from a set of intended learning outcomes. An XML-schema represents the information within a competence structure to support machine processing. Experiments were carried out to evaluate the competence-based system for recommending links by considering the learner‟s reaction, by comparing the learning improvement between the competence-based approach and other approaches, and by exploring the effects of search engines used and keywords on the search results. From these experiments, some conclusions have been drawn, such as: learning paths with more nodes are more helpful, and Web links of a competence node with a lower level of Bloom‟s taxonomy showed higher ratings than those with a higher level of Bloom‟s taxonomy. In addition, a competence-based system is accepted by learners at the reaction level. A freely-browsing and a competence-based system produced equal improvements in learners‟ learning. Different types of search engines (Google and Google API) and categories of keywords (SM and CA+SM+CO) show no significant differences between the qualities of study material links in helping learners achieve their competences. Furthermore, the links from Google were found to be as good as those from an educational search engine. Some future work is suggested, for example, more exploration of a complex competence structure and learning paths, improvements on the usability and accessibility of the application, and more in-depth consideration of self-assessment.

Virtual simulations of the behaviour of mechanical systems are of widespread use in academia and industry. Mechanical structures are often analysed using the finite element method, where deterministic models with one particular set of physical parameters are employed. However, the underlying assumption that the input data is precisely known is in general not valid, because there are uncertainties about the parameters, often until the last stage of the design cycle and even when the product is in service. Furthermore, every manufacturing process naturally introduces some product variability, which is inevitable. These effects can be compensated for by the application of safety factors. However, with the increasing requirements towards product performance, the effects of non-deterministic properties are of growing concern and advanced methods are needed that properly take them into account. In this context, it is often more important to predict the variation in the response than attempt to further improve the accuracy of a deterministic model. A number of viable methods to take non-deterministic properties into account already exist, but their computational efficiency and applicability have to be increased. In this thesis, a framework for the non-deterministic analysis of built-up structures using component mode synthesis (CMS) is presented. It is shown how several coordinate systems in CMS can be used advantageously for the quantification and propagation of non-deterministic data. A specific approach, based on considering the variation in component natural frequencies only, is introduced and its efficiency and accuracy investigated. The application of perturbational relations for uncertainty propagation in CMS is discussed. The framework of CMS is also used to combine qualitatively different uncertain data and the inclusion of experimental data is addressed. Overall, CMS methods can be used to reduce the numerical costs, improve the applicability of the approaches and also gain some physical insight for a structural dynamic problem with non-deterministic properties. Furthermore, several contributions are made to simulation methods that are usually applied in connection with the CMS approach. Different concepts for non-deterministic modal superposition are presented, which can be used to estimate the variation in frequency response functions from uncertain modal data. The application of the Line-Sampling simulation method, as an advanced Monte Carlo approach, to structural dynamic problems is shown. Finally, the modelling of spatial variations in components using random fields and the implementation in existing finite element models are addressed. Numerical examples are presented throughout.