This work summarises a computational framework for dealing with dynamic multi-body frictional contact problems. It is in fact a detailed account of an instance of the Contact Dynamics method by Moreau and Jean. Hence the title. Multi-body systems with contact constraints are common. Some of them, such as machines or arrangements of particulate media, need to be predictable. Predictions correspond to approximate solutions of mathematical models describing interactions within such systems. The models are implemented as computational algorithms. The main contributions of the author are in an improved time integration method for rigid rotations, and in a robust Newton scheme for solving the frictional contact problem. A simple and efficient way of integrating rigid rotations is presented. The algorithm is stable, second order accurate, and in its explicit version involves evaluation of only two exponential maps per time step. The semi-explicit version of the proposed scheme improves upon the long term stability, while it retains the explicitness in the force evaluation. The algebraic structure of both schemes makes them suitable for the analysis of constrained multi-body systems. The explicit algorithm is specifically aimed at the analysis involving small incremental rotations, where its modest computational cost becomes the major advantage. The semi-explicit scheme naturally broadens the scope of possible applications. The semismooth Newton approach is adopted in the context of the frictional contact between three-dimensional pseudo-rigid bodies, proposed by Cohen and Muncaster. The Signorini-Coulomb problem is formulated according to the formalism of Contact Dynamics. Hybrid linearisation, parameter scaling and line search techniques are combined as the global convergence enhancements of the Newton algorithm. Quasi-static simulations of dry masonry assemblies exemplify performance of the presented framework.

Previous research has shown coupling of mechanical behaviour and water retention behaviour in unsaturated soils at a constitutive level, with degree of saturation (in addition to suction) influencing mechanical behaviour and volumetric strains influencing water retention behaviour. An innovative elasto-plastic modelling framework incorporating coupling of mechanical behaviour and water retention behaviour has been proposed by Wheeler, Sharma and Buisson (2003) for isotropic stress states. These authors presented a single constitutive model for both mechanical behaviour and water retention behaviour. They did not, however, fully validate the model against experimental results. The objectives of the current research included undertaking an experimental programme specifically designed to investigate the coupling between mechanical behaviour and water retention behaviour, and using experimental results to investigate the validity of the Wheeler et al. (2003) model. Developments and refinement of the model were also to be explored An experimental programme of suction-controlled testing was carried out on one-dimensionally compacted samples of speswhite kaolin in a single steel-walled triaxial cell and an isotropic cell. In addition to standard stress paths, such as isotropic loading, unloading, wetting, drying and shearing, many non-standard stress path tests were also performed. These produced a unique data set, providing evidence of aspects of behaviour never previously studied. In terms of constitutive model development, the Wheeler et al. (2003) model was extended to triaxial stress states, to include the role of deviator stress. In addition, bounding surface plasticity concepts were used to develop simple realistic water retention models for rigid or deformable unsaturated soils, and a new bounding surface plasticity version of the Wheeler et al. (2003) model was developed. However, a problem of theoretical inconsistency in the Wheeler et al. (2003) model was identified, which occurs if plastic volumetric strains are predicted while the soil is fully saturated. A simple, but rather unsatisfactory, solution to this inconsistency was identified, and further research is required to identify a more satisfactory solution. Codes were developed for stress point simulations with the original version of the Wheeler et al. (2003) model, the new bounding surface plasticity version of the model and the conventional Barcelona Basic Model. Simulations were performed of all the experimental tests performed in the current research, to explore the performance of the different models. Comparison of model simulations with experimental results showed that the Wheeler et al. (2003) model was able to represent basic concepts of the mechanical behaviour of unsaturated soils, but sometimes not with the same level of accuracy or flexibility as the Barcelona Basic Model. The Wheeler et al. (2003) model was however able to capture features of mechanical and water retention behaviour that could not be represented by the Barcelona Basic Model or by other conventional models for mechanical or water retention behaviour. The bounding surface plasticity version of the Wheeler et al. (2003) model sometimes produced improved predictions. There remained, however, specific aspects of behaviour that were not well matched by either versions of the Wheeler et al. (2003) model. Some of these may be solved in the future by refinement of specific constitutive equations within the Wheeler et al. (2003) model, but others appear more likely to be insoluble without a major change to the proposed modelling framework.

The process dissolves Pt and Pd selectively from catalysts in a leach reactor, using an oxidant (I3-) and complement (I-), the reduced form of the oxidant. The precious metals are then electrodeposited on the cathode of an electrochemical reactor, incorporating a cation-permeable membrane, to enable the oxidant to be regenerated simultaneously at the anode without its subsequent reduction at the cathode. The iodide system was identified as being able to dissolve Pt and Pd, using tri-iodide as oxidant, at neutral pH to minimise alumina dissolution rates. The reaction kinetics were determined using techniques, such as voltammetry with rotating disc electrodes. Micro-kinetic models were developed for the oxidant reduction and metal oxidation processes, so that the behaviour of leaching metal could be predicted quantitatively under the different conditions used. As only low concentration leach solutions were produced, a cathode with high specific surface area and high mass transport rates was required to achieve acceptable capital and running costs of the electrochemical reactor. This was achieved with a cathode plate contacting a circulating bed of sub-mm graphite particles, on which adherent/coherent electrodeposits were grown. The electrical operation of the reactor was modelled successfully to give good agreement with the experimental data. Low specific electrical energy consumptions of < 1 kWh kg⁻¹ Pt were reported for the reactor. The feasibility of the overall process has been demonstrated and the running cost, of ca. £0.63 kg-1 of catalyst, calculated for the laboratory scale plant built, was significantly lower than the target set by Johnson Matthey Plc. However, for the spent catalyst sample investigated, the absolute recovery of Pt has been found to be limited to only ca. 90%, probably because of physical encapsulation.

No description available.

The potential of particulate mechanics was explored with the purpose of developing a single software modelling framework in which to model multi-physics geomechanical problems. Individual particulate models were developed for solid, fluid and granular material phases, building onto the existing Distinct Element Modelling (DEM) environment Particle Flow Code in Three Dimensions (PFC3D), with the intention that they could be combined to represent a geomechanical problem of any configuration. Advantages of utilising PFC3D were firstly, its inbuilt feature allowing inter-particle bonding so that, in the limiting case, solid material could be represented and secondly, its embedded coding language, FISH, which allows the creation of user defined variables and functions which may be used to manipulate and modify the basic DEM code. The Particle Solid Model (PSM) employed a bonded particle assembly with the concept of constructing a determinate lattice to replace a continuum material. Basic particle interactions were handled by the DEM behaviour inherent to PFC3D and complex behaviour, such as fracturing, was realised through additional code written in FISH. The Particle Fluid Model (PFM) was created with the concept of developing a macroscopic particle representation of a fluid where the particles moved freely with the flow. Fluid behaviour was implemented by preventing physical contact of the DEM particles through force-separation laws representative of pressure and viscosity fields. These were again realised via FISH code. The Particle Proppant Model (PPM) would be implemented simply through the inherent capabilities of the PFC3D DEM code and would be used to represent granular material on a grain for grain basis. After initial development of the individual models, benchmark tests were carried out to evaluate their basic capabilities.

The work contained in this portfolio thesis was carried out as part of an Engineering Doctorate (Eng.D) programme from the Institute for System Level Integration. The work was sponsored by Thales Optronics, and focuses on issues surrounding the implementation of video processing algorithms on field programmable gate arrays (FPGA). A description is given of FPGA technology and the currently dominant methods of designing and verifying firmware. The problems of translating a description of behaviour into one of structure are discussed, and some of the latest methodologies for tackling this problem are introduced. A number of algorithms are then looked at, including methods of contrast enhancement, deconvolution, and image fusion. Algorithms are characterised according to the nature of their execution flow, and this is used as justification for some of the design choices that are made. An efficient method of performing large two-dimensional convolutions is also described. The portfolio also contains a discussion of an FPGA implementation of a PID control algorithm, an overview of FPGA dynamic reconfigurability, and the development of a demonstration platform for rapid deployment of video processing algorithms in FPGA hardware.

This work is concerned with the corrosion fatigue characteristics of full-scale reinforced concrete beams partially submerged in 3.5% NaCl solution or in tapwater of low dissolved salt content. The test beams were subjected to constant amplitude fatigue loading in uni-directional or reverse bending at slow cycle rate of 0.17 Hz and various load levels. The test programme had two stages: Stage I, carried out at relatively high load levels, represented a study of the fatigue-failure phenomenon but also provided guides for the more detailed study undertaken in Stage II which was devoted to low load conditions under which the main steel deterioration process was corrosion. In the latter stage, attention was focused on the monitoring of a number of electrochemical parameters including the corrosion rate. Late in the programme, electrochemical noise technique was also examined. The phenomenon of concrete crack blocking, previously reported in seawater environment, was observed in both test environments in this work. This phenomenon was closely examined and the mechanisms of the formation of deposits and its effects were described. A hypothesis was proposed for the structural behaviour of reinforced concrete beams during cyclic loading in aqueous environment. Failure normally occurred by the fracture of one of the main tensile bars due to fatigue, often followed immediately by yield of the remaining bar and beams collapse. Fracture surfaces were examined under SEM.Corrosion rate measurements involved formidable difficulties which had to be overcome to obtain accurate measurements. For instance, the current interruption technique for the estimation of the IR-drop was developed and established as the most appropriate method for concrete beams with complex reinforcement configuration. Extensive polarisation measurements indicated clear effect of the test condition on the technical variables involved in various measuring techniques (viz potentiodynamic and potentiostatic techniques). Based on these observations a criterion has been proposed to determine the appropriate variables necessary for the accurate determination of the polarisation resistance Rp. The work has demonstrated that the corrosion behaviour of reinforced concrete sustaining dynamic loading is extremely complex, and short term indications could not be used safely for long term predictions. Based on corrosion rate measurements and the actual corrosion pattern observed upon completion of the tests, a concept of a change in corrosion mechanism from a microcell process of relatively low corrosion rates to a macrocell process at much accelerated high rates is introduced. The prevailing mechanism depends on time of exposure, load level and reinforcement details. Results from long running fatigue tests in seawater from concurrent research were incorporated which also support this concept.

Barrages built in estuaries fundamentally alter the dynamics of the river with regard to both flow and sedimentation patterns. Therefore it is essential to ensure that these structures do not affect the sustainability of the systems in which they are built. In recent years there has been increased emphasis on assessing the effect of climate change on river flows and the impact that this has on watercourses. Therefore, to investigate morphological sustainability of barrage impoundments, the effect of climate change must be included. An assessment of the morphological sustainability of the River Tees impoundment is presented. The predictions were completed using the l-dimensional software package ISIS, which modelled flow and sediment movement within the impoundment. Fifty-year simulations were completed to predict the sediment distribution through the system under differing future scenarios. A method is proposed for extending the flow boundary for the numerical model, which uses a generic statistical modelling technique. It uses the historical flow data recorded on the Tees and forward predicts the series based on its statistical properties. Firstly, the Markov Chain method was used to predict a 50 year flow series which assumes a stable climate. The predicted series showed good correlation with the measured series in terms of both statistical properties and structure. Secondly, the method was further developed to enable climate change predictions to be incorporated. This means that the generated series can be modified to directly account for the possible influence of climate change on discharge. This technique uses a Markov model fitted in the framework of a multinomiallogit model, enabling catchment precipitation and temperature values to be linked to the discharge. Climate change predictions available for the period 2070 to 2100 were then used to create 50-year modified flow series for the River Tees under a medium\high and medium\low emissions scenario. During the period of sediment monitoring on the Tees a change to the sediment supply was noticed as a result of the high flows experienced in October/November 2000. Unfortunately, it is unclear whether the sediment supply will return to its original levels or if, as a consequence of higher flows resulting from climate change, the supply will remain at present levels. Hence three different sediment rating curves were created from the field data to deal with this uncertainty; representing high, medium and low sediment supply conditions. Using the data generated for the flow and sediment boundaries, simulations were undertaken to assess the morphological sustainability of the Tees impoundment. Simulations using a flow boundary, which assumed both a stable climate and a changed climate, as well as three different sediment supply options for each, were considered. The results show that the impoundment reaches a dynamic equilibrium during the modelled period, irrespective of the sediment supply. From this it is possible to state that the Tees Impoundment is morphologically sustainable over the next 50-80 years. Climate change, while increasing the sediment supply, actually appears to improve the sustainability of the impoundment with regards to sediment. The increased number of high flows cause more steep water surface slopes which reentrain sediments and partially flush the system. In conclusion this thesis presents an assessment of the morphological sustainability of the Tees impoundment under differing future climate scenarios for both the fluvial and sediment inputs. Within the course of the work a different technique for extending flow series assuming both a stable and changed climate has been proposed. It is hoped that these methods will be of use in future sustainability assessments; however further investigations into these methods would be beneficial.

The author developed two methods for predicting the discharge capacity of uniform meandering compound channels. The first method utilised an Artificial Neural Network (ANN) functional approximator which was taught to replicate the relationship between 9 of the key parameters and the magnitude of F* which was exhibited by the flow data gathered during the Series B extension (1993-1996) programme. The ANN approximator was constructed using a 'Matlab' software environment and was supported on a PC. The second method consisted of a semi-physical / semi-empirical method which was named the Enhanced zonal method. This method comprised formulations which explicitly determined the discharge capacity of the 3 individual flow zones in meandering compound channels whilst accounting for difference induced by the characteristic 4 flow region behaviour. The author demonstrated that both of these methods produce more accurate discharge capacity predictions than the James and Wark [1] [1992] method for the majority of available flow data sets. The James and Wark [1] [1992] method was the optimal method prior to the Series B extension (1993-1996) programme. The ANN approximator gave the most accurate predictions when the parameters of the compound channels to which it was applied fell within the range of the parameters incorporated in the ANN training data set. However, the author demonstrated that the Enhanced zonal method is the most reliable discharge capacity prediction method over the full range of uniform meandering compound channel configurations. The author developed two refined one-dimensional (1D) numerical models (for application to both steady and unsteady flow conditions) which incorporated the Ackers [1991] and James and Wark [1] [1992] methods to determine the conveyance characteristics at representative uniform cross-sections in natural meandering compound channels. The author demonstrated, using a case study of the River Dane, that these refined 1D models were able to predict the water surface profiles in natural channels to a high degree of accuracy.

A computer program (called PGROUPN) has been developed for the non-linear analysis of pile groups subjected to any combination of vertical loads, horizontal loads and moments. The code is based on a complete boundary element approach and may be regarded as a generic descendant of the program PGROUP (Banerjee & Driscoll, 1976) which has been extended in order to make the analysis numerically efficient for practical problems and to include the effects of soil nonlinearity by means of a stepwise linear incremental procedure. One of the main advantages of a non-linear analysis system over a linear elastic approach is that it has the desirable effect of demonstrating a reduction of the corner loads in larger groups in both the vertical and horizontal senses. This observation is of basic importance in practice because it offers the prospect of tangible improvements in design procedures and potential saving of materials. The non-linear analysis put forward in this thesis may be applied to large pile groups embedded in cohesive soil, specifically fully saturated clay under undrained conditions. The soil is modelled as an elastic-perfectly plastic material, which is assumed to behave linearly elastic at small strain levels, but fails when the stresses at the pile-soil interface reach certain limiting values. The analysis only requires the definition of three soil parameters whose physical meanings are clear, ie the (initial tangent) Young’s modulus Es, the Poisson’s ratio vs and the undrained shear strength Cu. This represents a significant advantage over more common load-transfer approaches which are based on either empirical parameters or the results of full-scale pile load tests. The validity and accuracy of the proposed PGROUPN solution have been verified by comparison with alternative numerical analysis for single piles and pile groups subjected to axial and lateral loads. Benchmark solutions in the linear and non-linear range are presented, and the critical question of estimation of soil parameters is addressed. Finally, two published case histories are described which demonstrate the applicability of the method to practical problems.