The shear failure of reinforced concrete beams has been widely investigated over many years. Despite this, there is no consensus on the relative importance of the underlying mechanisms of shear resistance. The main objective of this thesis is to develop improved design guidelines for shear enhancement in beams with multiple concentrated loads applied on their upper side within a distance of 2d from the edge of supports (where d is the beam effective depth). The research involves a combination of laboratory testing, nonlinear finite element analysis and analytical work. Many tests have been carried out on beams with single point loads within 2d of supports but only a handful on beams with multiple point loads within 2d of supports. This is a significant omission since such loading commonly arises in practice. The author carried out a series of tests on beams loaded with up to two point loads within 2d of supports. The tests were designed to investigate the influences on shear strength of loading arrangement, cover and bearing plate dimensions. The latter two were varied to investigate the underlying realism of key assumptions implicit in the Strut and Tie Modelling (STM) technique. Detailed measurements were made of the kinematics of the critical shear crack. These measurements were used to assess the relative contributions of aggregate interlock, dowel action and the flexural compressive zone to shear resistance. Novel STMs are proposed for modelling shear enhancement in simply supported and continuous beams. NLFEA is used to assist in the development of the STM. The STM are validated with test data and are shown to give reasonable strength predictions that are of comparable accuracy to the author's NLFEA. STM gives particularly good predictions of shear resistance if the strut strengths are calculated in accordance with the recommendations of the modified compression field theory rather than the recommendations of Eurocode 2, which can result in strength being overestimated. However, the STM are shown to overestimate the influences of bearing plate dimensions and cover on shear resistance.

Nonlinear buckling behaviour of a thin-walled simply-supported stiffened panel that has uniformly spaced longitudinal stiffeners is studied. The structure is made from a linear elastic, isotropic and homogeneous material. The panel is subjected to pure compression applied at the centroid of whole cross-section. In such structures, the nonlinear interaction can occur between a global (Euler) buckling and local plate ( i.e. the stiffener or the main plate) buckling modes. The interactive buckling behaviour is usually more unstable than when the modes are triggered individually. This can lead to a significant reduction of the load-carrying capacity. The current work focuses on the case where the stiffening is only on one side of the main plate. An analytical model of a perfect thin-walled stiffened plate is formulated based on variational principles by minimizing the total potential energy. The equations of equilibrium are then solved numerically using the continuation and bifurcation software Auto to determine the post-buckling behaviour. Cellular buckling (or snaking) is revealed analytically in such a component arising from nonlinear local global interactive buckling, perhaps for the first time. In addition, the effect of varying the rigidity at the main plate -stiffener junction is studied; a rapid erosion of the cellular buckling response is revealed by increasing the joint rigidity. The initial model is then developed by including more degrees of freedom within the stiffened panel and the introduction of global and local imperfections. The results from the analytical model are validated by the finite element (FE) method using the commercial software Abaqus as well as by comparing against some experimental results taken from the literature. To obtain a greater understanding of the drivers of the structural behaviour, parametric studies are conducted for a variety of different plate and stiffener geometries as well as an investigation into the heightened sensitivity to geometric imperfections. The worst forms of local imperfection are identified in terms of the initial amplitude, number of waves and the degree of localization. The imperfection sensitivity and the parametric studies are conducted for two limiting cases, where the main plate- stiffener joint is assumed to be fully pinned or fully rigid. A framework for establishing the zone where structural designers need to consider mode interaction carefully is presented.

The external envelope of steel framed industrial buildings normally involves the use of purlins and rails spanning between the main hot-rolled frames to support the roofing/cladding. These purlins are typically light-gauge cold-formed steel members of complex shape for which the thinness of the material means that local instabilities will significantly influence their structural behaviour. In this thesis, the finite element (FE) method (ABAQUS) has been used to develop numerical analyses to study the buckling behaviour and degree of moment redistribution in continuous and sleeved cold-formed steel 2-span purlin systems. Five types of nonlinear FE analyses have been validated against reported physical tests: (i) continuous 2-span beams subjected to uniformly distributed load (UDL), (ii) single span beams subjected to a moment gradient, (iii) single span beams subjected to pure bending (iv) sleeved 2-span beams subjected to a UDL and (v) single span sleeved sections subjected to a moment gradient. The FE analyses were used to generate a large portfolio of new results for gravity and uplift loading for continuous and sleeved 2-span arrangements covering a wide range of cross-sections by varying the flange and web dimensions and material thickness. The effects of local and distortional buckling and limited rotational capacities for single span FE models were investigated. The 2-span FE results formed the basis for a simple modification to conventional plastic design that recognises the possibility of a reduction in moment with increasing rotation in the interior support region. The assumption of full moment redistribution for gravity loading was found to be only valid for stocky sections but not for slender sections. For uplift loading in addition to the potential reductions in moment at the interior support, limitations in the span moment due to lateral torsional buckling (LTB) for slender members were also accounted for. Based on the FE results, an α-reduction framework was established to predict the collapse load for continuous and sleeved 2-span systems. It was assumed that the cross-section or LTB resistance was achieved in the span while a reduced cross-section resistance allowing for the post-peak fall in capacity was achieved at the interior support. The accuracy of the proposed design method was compared against elastic and full plastic design cases by considering their ultimate load carrying capacities. Whereas the elastic design method provides overly-conservative results and plastic design overestimates the capacity of slender sections, the proposed design method gave accurate predictions of the failure load with minimal scatter for all cases. The developed α-reduction framework provides a foundation for allowing the use of other purlin sections and interior support connections by inserting alternative cross-sectional moment capacity inputs obtained from several sources such as physical testing, hand calculations from design codes and FE analyses.

The assessment of water resources and the prediction of future changes in water management are necessary due to global climate changes. The land surface model 'Joint UK Land Environment Simulator' (JULES), which lies in the borderline between hydrological and atmospheric modelling, was implemented in this study. The study focused on the Kennet catchment in the UK; a Chalk dominated catchment that includes a variety of different geological formations. The first spatial scale, where JULES was assessed, was the point scale; the scale at which equations central to JULES are regarded as applicable. The data used originated from the Warren Farm recharge site, located in the Lambourn catchment, a tributary of the Kennet. Due to the large number of model parameters that JULES includes, a Monte Carlo sensitivity analysis was implemented on a variety of different parameters at the point scale illustrating that that some soil parameters exhibit a bigger effect on the examined hydrological fluxes than others. For larger scale modelling, i.e. scales larger than 1 km2, soil data from the National Soil Resources Institute (NSRI) were used to parameterise JULES. NSRI database has a spatial resolution of 1 km x 1 km grid squares, where each square is attributed with the percentage of the various soil series included in it. Due to the lack of chalk soil hydraulic parameters in the NSRI dataset, Warren Farm observations were used to parameterise the Chalk sites. Moving to the Kennet catchment modelling, it was shown that a weather generator, which was representing the diurnal patterns of the climatic forcing data, led to more realistic hydrological outputs compared to the one available in the standard version of JULES (2.2). The issue of whether the effect of spatial variation in surface properties could be adequately represented in JULES was explored at 1 km and catchment scales. Thus, the extent of the heterogeneity present in the Kennet catchment was investigated, with and without a surface runoff production mechanism component, with the focus being on the various soil types included in the Kennet. The amount of heterogeneity in each of the 1 km grid cells of the NSRI database was also studied, showing that the dominant soil type in the majority of the grid squares of the catchment could sufficiently represent the annual average hydrological behaviour. Lastly, it was exhibited that the inclusion of the surface runoff component allowed the model to overcome the limitation of the unrealistic surface runoff values that were produced without the implementation of that component.

This PhD thesis explores the utility of a land surface model (Joint UK Land-Environment Simulator, JULES) for large-scale hydrological modelling of the Peruvian Amazon - a humid tropical mountain basin where process understanding is poor and data are scarce. A sparse rain gauge network necessitates the use of large-scale data from satellite and global climate model reanalysis to complement ground observations, commanding a closer look at (1) the uncertainties (2) merging techniques to utilise multiple observations in the model forcing. A main outcome of the research is establishing the model's sensitivity to precipitation error, and at the same time, demonstrating an increasing reliability of global remote sensing products as model forcing, specifically, with data from the Tropical Rainfall Measuring Mission Multisatellite Precipitation Analysis version 7 algorithm. Furthermore, satellite-rain gauge data assimilation techniques such as mean-bias correction, double smoothing residual blending, and Bayesian combination, are shown to reduce the mean errors in the satellite-based product. Secondly, with regional calibration and an offline runoff routing scheme, JULES is shown to be reasonably skillful at reproducing the observed streamflow dynamic and extremes. Representing the subgrid heterogeneity of soil moisture using the probability distributed model (PDM) was key to improving surface runoff generation. However, evapotranspirative fluxes in the lower basin remain poorly reproduced without an adequate floodplain system representation. Finally, under the Intergovernmental Panel for Climate Change's RCP4.5 future climate scenario, which projects a warming and wetting up to the year 2035, the Peruvian Amazon basin is shown to respond nonlinearly to the increase in wet season precipitation with more than 40% increase in the peak flows compared to the baseline scenario. There is limited confidence in the projections due to climate projections uncertainty and the assumptions of model stationarity.

This thesis explores the role of consumers' choices into the integration of mobility and power systems. It will contribute to the wider literature of electric vehicles-power systems integration by explicitly accounting for consumers' preferences in shaping charging demand. This objective is achieved by developing a methodology to investigate electric vehicles (EV) charging choices in technological scenarios that enable smart charging operations. A modelling framework for the joint analysis of EV charging and activity-travel behaviour is introduced. This is based on an extension of traditional activity scheduling models that embeds the charging choice dimensions: namely the available energy after charging (that is related to the driving range) and the charging duration (defined here as the time elapsed from arrival at a charging facility until the desired battery level is achieved). This framework accommodates the interaction between charging behaviour and travel/activity behaviour, and allows us to capture the potential effects of charging service pricing and charging demand management policies on charging choices as well as along the timing dimension of travel/activity choices. A stated response survey instrument for estimating a tour-based operational version of the model is developed. Results from this empirical study provide insights into the value placed by individuals on the main attributes of the charging choice. The trade-offs between target battery levels and schedule delays potentially induced by long durations of the charging operation are also analysed. The model is then implemented into a micro-simulation framework to demonstrate the model applicability for modelling electric vehicle charging demand. The specific application shows the compatibility of charging choices under various electricity pricing scenarios with electric vehicle load flexibility - an essential requirement to enable smart charging operations.

Flow time-series data are crucial for water resources and floods management. In catchments where flow observations are not available or data are of poor quality, a method for modelling flow time-series is needed. The overall objective of this study is to assess the applicability of recent regionalisation methods to predict flows in tropical monsoon-dominated catchments where hydrological response is particularly variable over the seasons. This PhD thesis addresses six primary challenges in the context of such catchments: 1) data quality, 2) rainfall estimation in mountainous catchments, 3) using regression for regionalising rainfall-flow response indices, 4) catchment non-stationarity, 5) conditioning rainfall-runoff models and 6) uncertainty analysis. The main novel contributions are: developing a data quality scoring system and exploring its effects on modelling; comparing a customised technique for rain gauge interpolation and using satellite products for spatial rainfall estimation; demonstrating practical difficulties in predicting land use change impacts; assessing the performance of recent conditioning methods and estimating prediction uncertainty in monsoonal areas. The main practical outcomes are: 1) the most in-depth study yet published of methods for predicting flows in northern Thailand for water resources planning; 2) recommendations towards improving data support for water resources estimation in Thailand. Using data from 44 gauged sub-catchments of the upper Ping catchment in northern Thailand from the period 1995-2006, three relevant flow response indices (runoff coefficient, base flow index and seasonal flow elasticity) were regionalised by regression against 14 available catchment properties. The runoff coefficient was the most successfully regionalised, followed by base flow index and lastly the seasonal elasticity of flow. The non-stationarity (represented by the differences between two six-year sub-periods) was significant both in the flow response indices and in land use indices; however relationships between the two sets of indices were weak. The regression equations were not helpful in predicting the non-stationarity in the flow indices except somewhat for the runoff coefficient. Rainfall estimation errors from two different estimation methods were large and believed to significantly contribute to uncertainty in regionalised flow response indices and modelled flow time-series. The three regionalised flow response indices were used individually and in combination to condition the IHACRES rainfall-runoff model using a Bayesian approach. The runoff coefficient was the most informative index. This is followed by the base flow index and lastly the seasonal flow elasticity. Using the variance of the regression coefficients and of the regression residuals had limited success in estimating the flow uncertainty intervals because uncertainty from the IHACRES model structure is not sufficiently represented by the variance of the regression. The regionalised model was considered to be too imprecise at the daily time scale but offers good support to water resources planning at the monthly and seasonal time scales. A partly subjective data quality scoring system showed the clear influence of rainfall and flow data quality on regionalisation uncertainty. Recommendations include developing more relevant soils databases, improved records of abstractions and investment in the gauge network.

Compression members, made from slender metallic plate elements, are prone to a wide range of different elastic instability phenomena. A thin-walled I-section strut, made from a linear elastic material, can suffer from the nonlinear interaction between a global (Euler) buckling mode, and a local flange plate buckling mode. The interactive buckling behaviour is usually much more unstable than when the modes are triggered individually and hence significantly reduces the load-carrying capacity of real struts. The current work focuses on such a problem using an analytical approach, the methodology of which has been well established in previous works on sandwich struts and I-section beams. An analytical model that describes the interactive buckling of a thin-walled I-section strut under pure compression based on variational principles is presented. Analytical formulations combining the Rayleigh-Ritz method and continuous displacement functions are presented to derive a series of systems that comprise differential and integral equilibrium equations for the structural component. Solving the systems of equations with numerical continuation reveals progressive cellular buckling (or snaking) arising from the nonlinear interaction between the weakly stable global buckling mode and the strongly stable local buckling mode. The resulting behaviour is highly unstable and when the model is extended to include geometric imperfections it compares excellently with some recently published experiments. Imperfection sensitivity studies reveal high sensitivity to both global and local imperfection types. The worst forms of local imperfection are identified in terms of the initial wavelength, amplitude and degree of localization. The effect of the varying rigidity of the joint of the section web and flanges is also studied and a rapid erosion of the cellular buckling response is revealed with increasing rigidity of the flange-web joint. A shell-based nonlinear finite element model is presented, primarily for validation purposes. The results from the analytical and finite element models show a good comparison, particularly for higher rigidities of the flange-web joint. A parametric study is conducted for two limiting cases, where the flange-web joint is assumed to be fully pinned or fully rigid. For a chosen set of geometries, the most undesirable interactive region is identified for both global and local slendernesses, in terms of the strut length and the flange width respectively. Practical implications are discussed in terms of the idealized buckling design curve. An analytical framework for the structural analysis of the thin-walled I-section struts that exhibit the nonlinear interaction of a global and a local buckling mode, including cellular buckling, has therefore been established.

The growing capabilities and widespread proliferation of information and communication technologies (ICT) into virtually every aspect of lifestyle, combined with the continuing challenges faced by transport systems, has ensured ongoing interest in the interactions between ICT and travel behaviour. Yet, despite more than three decades of efforts to understand these relationships, few point of consensus have so far emerged, partly due to the rapidly evolving character of ICT, and partly due to the inherent complexity of such interactions. This thesis seeks to develop novel understandings of such interactions by introducing a number of extensions to the existing modelling frameworks. This is achieved through three interrelated research objectives which seek to explore the topic from macro, micro, and temporal perspectives. The macro perspective takes the form of a structural equation analysis of the relationships between ICT use and travel behaviour across four countries: Canada, the United States, the United Kingdom, and Norway, with the data for the latter three obtained by pooling separate datasets on ICT use and travel behaviour. The micro perspective seeks to develop a microeconomic model of an individual maximising utility through joint choice of activities, including in-travel activities, ICT use, as well as the choice of travel mode, timing and route, with the decisions motivated by contribution towards satisfaction, productivity, and consumption. The model is subsequently tested in the empirical contexts of rail business travel time, business travel time valuation, and conceptualisation of the ICT and travel behaviour interaction scenarios reported elsewhere in the literature. The final, temporal perspective analyses the comparatively least explored topic of evolution in the relationships between ICT use and travel behaviour over time. This is achieved by analysing repeated cross-sectional data using structural equation modelling, and interpreted with reference to the theory of diffusion of innovations. The thesis also discusses a number of potential research, policy and industrial applications of its empirical and theoretical contributions.

Aluminium alloys are nonlinear metallic materials with continuous stress-strain curves that are not well represented by the simplified elastic, perfectly plastic material model used in most existing design specifications. The aims of this study are to develop a more efficient design method for aluminium alloy structures by rationally exploiting strain hardening. The key components of this study include laboratory testing, numerical modelling and development of design guidance for aluminium alloy structures. As part of the present study, the experimental programme included tests on 11 stub columns, 40 simply supported beams, 46 continuous beams and corresponding tensile coupon tests. Numerical investigations of aluminium alloy simply supported beams and continuous beams were also conducted. The validated finite element models were used for extensive parametric studies, generating 96 results for beams under three-point bending, 96 under four-point bending and 210 for continuous beams. The experiments and numerical simulations have shown the following key features of the inelastic behaviour of aluminium alloy structural elements: (1) the significance of strain hardening, indicated by the ultimate stress over the yield stress, could be up to 50%; (2) non-slender section capacities could be generally up to 40% higher than the yield limits in compression, and 50% greater than the plastic moments in bending; (3) the experimental and numerical ultimate loads of continuous beams on non-slender sections go beyond the calculated loads corresponding to the occurrence of the first hinge by more than 10%. Previous experimental data on aluminium alloy stub columns and simply supported beams were also collected. These collected test data were used together with the newly generated experimental and numerical results obtained from this study, totalling about 900 data, to assess the design predictions of the American, Australian/New Zealand and European specifications. On average, the existing design methods under-estimated the capacity of aluminium alloy stub columns by around 15%~22%, simply supported beams by around 18%~40% and continuous beams by around 27% ~ 50%. Existing section classification limits in Eurocode 9 (2007) were also assessed, and while they were found to be safe, some improved limits were proposed. The combined experimental and numerical results were used to develop and calibrate a new design method, termed the continuous strength method (CSM). Two key components of the CSM - a base curve and a bi-linear material model for aluminium alloys have been proposed in this study. Global plastic analysis allowing for moment redistribution has also been adopted in the CSM. Unlike current practices, the CSM has the merits of adopting the continuous treatment for the cross-section deformation response, rationally exploiting the available capacity beyond the yield limit and reasonably allowing for redistributing the internal forces. The capacity predictions of aluminium alloy structural members have been improved by more than 30% using the CSM. Reliability analyses have also been performed to assess the reliability level of different design methods according to the American Institute of Steel Construction (2010) and European Standard EN1990 (2002) approaches. The CSM has been shown to be safe, efficient and consistent for aluminium alloy structural members.