Magnetic Resonance Imaging (MRI) is a non-invasive, non-ionising imaging modality which allows three-dimensional images to be acquired. It is particularly suited to repeat studies of the same subject. To aid comparison of the resulting images and allow any anatomical differences to be more easily visualised, it may be necessary to register the images to a common frame of reference. This ensures the same anatomy can be found at the same location in each image. For repeat scans of subjects with little or no expected anatomical change on well-maintained scanners, a rigid-body transformation model can give acceptable registration results. However, there are often geometrical differences between the serial images because of distortion and anatomical change. Various spatial transformation models exist which can be used during the registration process and the particular model used depends on the expected deformation necessary to register one image to another. This assumption is investigated in this thesis and it is shown that registration accuracy can be improved through the use of a new registration algorithm which allows spatially separate parts of an image to undergo their own independent rigid-body registration. This is done by subdividing one image into many overlapping cubic sections before using a separate process to register each sub-cube. As each process is independent of the others, this task is suited to running on a parallel computer. Following introductory chapters explaining the basics of MRI, image registration and parallel computing, this new registration algorithm is developed to be run in parallel on a supercomputer. The algorithm is validated through the use of computer-generated phantoms and a highly-structured physical phantom. The effect of sub-cube size and noise on the ensuing registration results is also investigated. For a cubic region in the centre of the physical phantom, the error in correspondence for an array of 343 regularly spaced spheres reduces from 0.562 mm using a rigid-body transformation model to 0.058 mm using the new locally adaptive registration algorithm. The registration errors associated with the new algorithm are significantly smaller (<i>p</i> < 0.0001, <i>n </i>= 343) from those using a rigid-body transformation model. Comparisons are also made with other public-domain rigid and non-rigid registration algorithms, with the non-rigid algorithms having less registration error.

This thesis aims to prove the concept that the design of a sustainable electricity system requires generator deployment within the transmission grid to be co-optimized and integrated with a national electricity policy, which adheres to the constraints of global sustainability. To combat the main human activities which jeopardize total species well-being and global sustainability at large, human appropriated net primary productivity must be reduced by a factor 4,45 and carbon emissions by a factor 1,15 through to 2030. Incorporating these constraints into a high-level electricity policy model, the results for the 27-member European Union and United States suggest that the fuel mixes in each area will show improvement in the flagship of: share of energy from renewable; emissions of greenhouse gases; and security of supply. However, to ensure best-case mix diversity, consumption must be reduced by up to 2,26% and 1,01% below current levels of the European Union and United States, respectively. Integrating the fuel mix policy with generator deployment is accomplished by co-optimizing the former with an optimized power flow, utilizing a matrix balancing algorithm to specify the space and location constraints for the generator types. A case study using mainland Portugal yields transmission loss reductions of 0,43% with 11,88% of total installed capacity deployed as distributed generation using photovoltaic. Innovative distributed wind and photovoltaic schemes in Barbados demonstrate the inclusion of sustainability principles, including attention to issues of waste, energy independence, repeatability throughout the Caribbean and social acceptance. The overall result is a unique, full-chain design tool for sustainable electricity systems.

This work address all of the stages required to take a sequence of images of an object and recover a 3D model in order to produce as system that maximises automation and minimises the demands placed on the user. To that end we present a practical implementation of an automatic method for recovering the positions and properties of the cameras used to take a series of images using a textured ground-plane. We then offer two contributions to simplify the task of segmentation an object observed in multiple images. The first, applicable to more simple scenes, automatically segments the object fixated upon by the camera. We achieve this by iteratively exploiting the rigid structure of the scene, to perform the segmentation in 3D across all the images simultaneously, and the consistent appearance of the object. For more complex scenes we move to our second algorithm that allows the user to select the required object in an interaction manner whilst minimising demands on their time. We combine the different appearance and spatial constraints to produce a clustering problem to group regions across images that allows the user to label many images at the same time. Finally we present an automatic reconstruction algorithm that improve the performance of existing state-of-the-art methods to allow accurate models to be obtained from smaller image sequences. This takes the form of a filtering process that rejects erroneous depth estimates by considering multiple depth hypotheses and identifying the true depth or an unknown state using a 2D Markov Random Field framework.

This thesis describes the development of a new computer program called DANGLE (Dihedral Angles from Global Likelihood Estimates), which predicts protein secondary structure and backbone φ and Ψ dihedral angels solely from amino acid sequence information, experimental NMR chemical shift measurements and a database of known protein structures and their associated shifts. The approach uses Bayesian inferential logic to analyse the likelihood of conformations throughout Ramachandran space, paying explicit attention to the population distributions expected for different amino acid residue types. The search algorithm used by DANGLE identifies the most probable backbone conformation of a query residue by analysing the distribution of dihedral angles in fragments that possess similar secondary chemical shifts and local amino acid sequences within a database of protein structures. Upon comparison with database entries, shift and sequence fragment matching for the query residue yields a scatter of (φ,Ψ) predictions in Ramachandran space. Bayesian statistics are used to compare this with the scatter patterns obtained for known backbone configurations from database proteins, producing a “global likelihood estimate” diagram which reports on the degeneracy and precision of the predicted conformation. Simple filtering procedures can identify the most ‘predictable” residues, yielding 92% of all φ and Ψ predictions accurate to within ± 30° with respect to high-resolution X-ray structures. In contrast to previous approaches, more than 80% of all φ and Ψ predictions for glycine and pre-proline are reliable. Furthermore, DANGLE provides meaningful upper and lower bounds for the predictions which are shown to represent the precision of the prediction. DANGLE is also able to assign 86% of the secondary structure correctly. A set of dihedral angle constraints predicted by DANGLE is used in combination with conventional structure calculation protocols to determine the solution structure of the SAM domain of human SLY protein with average RMSD of 0.52 Å for all backbone atoms in the final ensemble.

In this dissertation I have created and applied a parametric model for bulk carbonate materials. The new empirical model for carbonates is stable for a wide range of carbonate structures and reproduces experimental results with reasonable accuracy. To study the surface of calcite the ab initio code SIESTA has been used. New implementation has been introduced into the SIESTA code to allow the calculation of effective charges using the modern theory of polarisation. Using these charges the calculation of the long range electrostatic effects, which are removed by the zero electric field boundary conditions, have been introduced into the phonon methodology, reproducing the LO-TO splitting within the calculated phonon modes near the F-point. Furthermore the effective charges have been used in the calculation of the infrared intensity for each phonon mode. The SIESTA implementation of DFT relies upon the evaluation of electron density on a real-space grid. Such discretization of the real-space integrals introduces an oscillatory error in the energy and forces, with the periodicity of the real-space grid. A method for reducing this error has been introduced. The SIESTA code with the new methodology has been used to study bulk calcite, {211} calcite surface and the interaction of water with the {211} sur­face. The structure and phonon frequencies for the bulk match well with experimental values. The {211} surface has been calculated showing the response of the crystal in both distortion of the ion position and the electronic configuration. Surface relaxations and phonon frequencies show no symmetry breaking reconstruction of the calcite {211} surface. Calculation of the interaction of water molecules with the {211} surface predicts the optimum position for water on the surface.

The retrieval methods use the explicit elements and associations which appear in a structured design representation, without requiring an understanding of the designs or their application domains. They depend instead on simple similarity measures concerning the basic representation elements, and integrate this basic evidence within the structured representations. Structure is handled dynamically, so that fragments are defined by the best-matching level for each query. The propagation of evidence through the structure of fragments allows different types and levels of representation to be related. These principles are developed are evaluated within an IR framework consisting of textual information, not least since text can and does express design information, especially concerning experience and the design process. Moreover, IR provides standard test collections to quantify retrieval performance. Although usually based on fairly long passages of unstructured text, such a collection can be used to evaluate the suggested approach to retrieving structured mechanical design fragments, which is difficult to assess directly. Rather than transferring the structured representation to flat text, which would lose information, a structure is imposed on a textual test collection and an analogy drawn between structured design representations and structured text. The results show that formalised knowledge structures such as classifications are not necessary for retrieving design information. Instead, informal knowledge and 'obvious' connections between representation elements can lead to improved retrieval performance, according to the standard IR measures of precision and recall. However, connections are not always applicable in every context, and retrieval performance suffers if any ambiguity in the representation elements or their similarity measures is not resolved before making connections. The representation structure forms a convenient context for this process of resolving ambiguity. There are several applications of this work. One is in Design Reuse, initially via a retrieval system which suggests standard components to replace specially-designed parts. Allowing for imprecision means that it can be used relatively early in the design process.

This thesis explores how a vehicular sensor platform may be built and how data from a variety of sensors can be sampled and stored. Applications exploiting such platforms will infer higher-level information from the raw sensor data collected. We present the design and implementation of one such application which involves processing vehicles’ location histories into an up-to-date road map. Our experience shows that there is a problem with programming this kind of application: the number of vehicles and the nature of computational infrastructure available are not known until the application is executed. By comparison, existing approaches to programming applications in wireless sensor networks tend to assume that the nature of the network architecture is known at design-time. This is not an appropriate assumption to make in vehicular sensor networks. Instead, this dissertation proposes that the functionality of applications is designed and implemented at a higher level and the problem of deciding how and where its components are to be executed is left to a computer. We call this ‘late physical binding’. This approach brings the benefit that applications can be automatically adapted and optimised for execution in a wide range of environments. We describe a suite of transformations which can change the order in which the components of the program are executed whilst preserving its semantic integrity. These transformations may affect several of the application’s characteristics such as its execution time or energy consumption. The practical utility of this approach is demonstrated through a novel programming language based on Jawa. Two examples of diverse applications are presented which demonstrate that the language and compiler can be used to create non-trivial applications. Performance measurements show that the compiler can introduce parallelism to make more efficient use of resources and reduce an application’s execution time. One of the applications belongs to a class of distributed systems beyond merely processing ventricular sensor data, suggesting that the late physical binding paradigm has broader application to other areas of distributed computing.

Shape memory alloys (SMAs) are particularly attractive as actuators in smart structures, due to their large actuation strain/stress, high power/weight ratio, single actuation mechanism and low noise etc. However, the lack of reliable predictive models and computational and design tools severely limits their applications. This thesis is concerned with the models appropriate for the numerical implementation, and the optimal design procedure of SMA actuators. To understand the thermomechanical characteristics of NiTi SMAs, a number of tests have been carried out, including iso-thermal tension tests under different constant temperatures, iso-thermal tension tests with different constant strain rates, iso-stress thermal cycling tests under different constant stresses, repeated iso-stress thermal cycling tests and repeated iso-thermal superelastic mechanical cycling tests. The stress-temperature phase diagrams for phase transformations are constructed. Under the framework of generalized plasticity with an internal-variable formalism, a three-dimensional phenomenological model is developed. Based on the stress-temperature phase diagram, the evolution equations of phase fractions are derived. The model reproduces the basic features of SMAs, such as the shape memory effect (SME) and superelasticity (SE), and can deal with incomplete phase transformations. Using return mapping algorithms, the incremental numerical formulation of the model is implemented into ABAQUS, through a user-defined material subroutine. Numerical examples and comparisons between the simulations and experiments have shown the capacity of the model and the feasibility of the subroutine. The design method of a typical bias SMA actuation unit is developed. The relationship of the actuation stress, strain and temperature is obtained by solving a coupled problem combining a thermomechanical constitutive model, heat conduction equations and the spring response. A novel lightweight SMA actuator integrating SMA actuation units into a truss structure is proposed, optimised and manufactured to demonstrate the possibility of SMA actuators in the application of smart structures.

No description available.

No description available.