The impact of corrosion on the nonlinear stress-strain behaviour of reinforcing bars under monotonic and cyclic loading was explored experimentally. The corrosion procedure was simulated in a laboratory environment using an accelerated corrosion procedure. A total of 132 corroded test specimens were produced and tested under mono tonic and cyclic loading. 23 corroded bars from the tests specimens were taken for further statistical analysis of their corrosion patterns. An advanced 3D optical measurement technique was employed to scan the surface of corroded bars. A novel stochastic signal processing technique was used for corrosion pattern analysis and subsequent development of probabilistic distribution models for the geometrical properties of corroded bars (cross section area, second moment of area etc.). Finally the scanned bars were tested under monotonic buckling and cyclic loading. The impact of the corrosion patterns on the nonlinear stress-strain behaviour of corroded bars was then investigated using nonlinear finite element modelling of the tested bars which was compared with the experimental results. Using the experimental and numerical data a new phenomenological uniaxial material model is also developed for reinforcing bars. The new material model accounts for the influence of corrosion damage, inelastic buckling and low-cycle fatigue degradation. This model is then implemented into the OpenSees platform as a new uniaxial material class known as Corroded ReinforcingSteel. Finally the material model is validated against 10 buckling critical flexural RC columns from the UW-PEER experimental RC column database. In addition, a new modelling technique is developed for modelling the flexural behaviour of buckling critical RC bridge piers using a distributed plasticity fibre beamcolumn model. These new models for the assessment of corrosion damaged RC columns have significantly improved the accuracy of previous models. This will help bridge managers and owners to develop a rigorous maintenance strategy to evaluate and predict the performance of their bridge network so that repairs can be prioritised and targeted at the most critical structures.

Failure of pile-supported structures are still observed in liquefiable soils after most major earthquakes. As a result, the behaviour of pile foundations during liquefaction phenomena remains a constant source of attention to the earthquake engineering community. In this context, the present research attempts to investigate the effects of soil liquefaction on the dynamic behaviour of pile-supported structures. Firstly, the thesis reports a field investigation carried out in the region affected by liquefaction events observed after the 2012 Northern Italy earthquake sequence. The collected information are used to gain insight into the mechanism governing the onset of liquefaction in the real field. The thesis also presents a series of multi-stage cyclic triaxial tests that aim to investigate both pre and post-liquefaction behaviours of two types of silica sands. These findings are subsequently used to develop a new set of p - y curves which are capable of capturing the strain-hardening behaviour observed in the experiments. The main component of the present research consists of an experimental investigation carried out using a shaking table. A preliminary study is reported in which tests are performed to investigate the effects of artificial boundaries of the model container on the response of the enclosed soil. It is found that the use of absorbing boundaries, which are made of conventional foams, is capable of reducing reflections and generation of body waves from the artificial boundaries, which in turn minimise the so called boundary effects. Subsequently, four physical models consisting of two single piles and two 2 x 2 pile groups are tested on a shaking table. The dynamic response is conveniently identified in terms of two modal parameters, namely: fundamental period and damping ratio. The experimental results suggest that the fundamental period of the pile-supported structures may increase considerably due to the soil softening caused by liquefaction. On the other hand, the damping ratio of these structures increase to values in excess of 20%. Based on the these findings , it is noted that the seismic demand imposed by the shaking on the models may vary with the excess pore water pressure generation, which in extreme case may lead to full liquefaction conditions. In particular, it is observed that the highest acceleration demand, and consequently maximum inertia force, experienced by the models occur during the transient to liquefaction. Finally, a series of numerical analyses are performed using the Winkler approach with the proposed p - y curves. The numerical results show that the models correctly replicated the distribution of the maximum bending moments measured after the onset of liquefaction but consistently underestimated the maximum moments by a factor ranging from 2 to 3. It is also found that the capacity spectrum method can be used as a simple and convenient tool for assessing the seismic demand.

Vibration serviceability of structures has become an evermore important issue in structural engineering practice. A particularly challenging problem is posed by modelling pedestrian loading on lively structures such as bridges. This is because pedestrians have the capacity to interact with vibrating structures which can lead to amplification of the structural response. However, current design guidelines are often inaccurate and limiting as they do not sufficiently acknowledge this effect and do not allow for consideration of different loading scenarios which might govern the design criteria in different cases. Therefore, this thesis addresses the uncertainty which pertains to the bidirectional pedestrian-structure interaction and its consequences for structural stability. This thesis has three main aims. The first aim is to investigate pedestrian-structure interaction on the ground vibrating in a single lateral and vertical direction by means of a mathematical model. The second aim is to establish a versatile modelling tool for the design of structures against lateral instability due to the loading from crowds with different distributions of defining parameters. The third aim is to evaluate pedestrian actions on laterally-oscillating ground in the laboratory environment while avoiding the implications of artificiality and allowing for unconstraint gait. To achieve these aims a synthetic multidisciplinary biological approach is adopted. The advancement in the structural engineering field afforded by the biological approach relies in appreciation of operational complexities of biological systems, in general, and human locomotion, in particular. This allows fundamental understanding of pedestrian behaviour and its determinants in the presence of vibration applied by the walking surface to be gained with the ultimate goal of better characterisation of the self-excited forces exerted by the pedestrians, which are the main concern in structural engineering since they can drive structural instability. With all this in mind, a biomechanically-inspired inverted pendulum pedestrian model is explored in the context of forces generated on the supporting surface undergoing lateral or vertical motion. Consideration is given to the gait balance control mechanisms specific to these cases. It is shown that the net, destabilising or beneficial, self-excited forces can be generated without necessarily assuming synchronisation, but considering simple mechanics and control requirements of walking. Probabilistic stability criteria against lateral instability of structures under the action of crowds are then formulated and used in combination with the self-excited forces derived from the studied pedestrian model to show their applicability. The experimental setup is developed based on a laterallydriven instrumented treadmill. A motion capture system is utilised to obtain data on the kinematics of human gait and to facilitate an immersive virtual environment representative of a bridge, delivered through a head mounted display, and the treadmill belt control mechanism allowing for automatic adjustment of its speed to that of the walker. The results from the tests conducted on the setup give supporting evidence for applicability of the inverted pendulum pedestrian model and the foot placement balance control mechanism, also revealing influence of the visual environment on the selfexcited forces. Owing to the advances in experimental protocols and data processing, attractive patterns in pedestrian stepping behaviour on laterally-oscillating ground are identified and analysed in the context of structural stability. The premise underlying this thesis is that, despite the ubiquitous complexity of human behaviour, walking gait obeys a set of universal rules which are trackable and quantifiable. Uncovering these rules can help in defining more reliable models of pedestrian loading and structural response. Therefore, the research approach presented in this thesis departs from the approach found in the existing codified design guidelines, calibrated by top-down empirical models, and current modelling trends based on stochastic load description towards the bottom-up approach formulated on more deterministic treatment of the pedestrian loading.

This thesis is concerned with the design development of cold-formed steel portal frames at both ambient and elevated temperatures. A full-scale site test was carried out to determine the behaviour in fire, of a portal frame structure comprised entirely from cold-formed steel. Laboratory component tests were used to investigate joint behaviour at ambient temperature. Numerical modelling was used to extend the investigations without the need for further physical testing. Good agreement between the experimental tests and non-linear finite element (NLFEA) models was obtained. The finite element models were then used in an extended study, the purpose of which was to develop the basis for a performance based design approach. At ambient temperature, the inclusion of joint stiffness in models was demonstrated as important in order to accurately predict the behaviour of frames. Design recommendations are proposed that are based on both the experimental and finite element results. At elevated temperature, the importance of base fixity and in-plane restraint from side-rails was highlighted as being crucial in preventing undesirable outwards collapse. Design recommendations in the form of a mathematical model, protection advice and construction details are proposed, that are based on both the experimental and finite element results. In the absence of published design recommendations, the guidance and NLFEA shell model presented can be used by practicing engineers to assist in design of cold-formed steel portal frames in fire boundary conditions.

Disasters can have devastating effects on our communities and can cause great suffering to the people who reside within them. Critical infrastructure underpins the stable functioning of these communities and the severity of disasters is often linked to failure of these systems. Traditionally, the resilience of infrastructure systems is assessed by subjecting physically based models to a range of hazard scenarios. The problem with this approach is that it can only inform us of inadequacies in the system for the chosen scenarios, potentially leaving us vulnerable to unforeseen events. This thesis investigates whether network graph theory can be used to give us increased confidence that the system will respond well in untested scenarios by developing a framework that can identify generic system characteristics and hence describe the underlying resilience of the network. The novelty in the work presented in this thesis is that it overcomes a shortcoming in existing network graph theory by including the effects of the spatial distribution of geographically dispersed systems. To consider spatial influence, a new network generation algorithm was developed which incorporated rules that connects system components based on both their spatial distribution and topology. This algorithm was used to generate proxy networks for the European, US and China air traffic networks and demonstrated that the inclusion of this spatial component was crucial to form the highly connected hub airports observed in these networks. The networks were then tested for hazard tolerance and in the case of the European air traffic network validated using data from the 2010 Eyjafjallajökull eruption. Hazard tolerance was assessed by subjecting the networks to a series of random, but spatially coherent, hazards and showed that the European air traffic network was the most vulnerable, having up to 25% more connections disrupted compared to a benchmark random network. This contradicts traditional network theory which states that these networks are resilient to random hazards. To overcome this shortcoming, two strategies were employed to improve the resilience of the network. One strategy ‘adaptively’ modified the topology (crises management) while the other ‘permanently’ modified it (hazard mitigation). When these modified networks were subjected to spatial hazards the ‘adaptive’ approach Page i produced the most resilient network, having up to 23% fewer cancelled air routes compared to the original network, for only a 5% change in airport capacity. Finally, as many infrastructure networks are flow based systems, an investigation into whether graph theory could identify vulnerabilities in these systems was conducted. The results demonstrated that by using a combination of both physically based and graph theory metrics produced the best predictive skill in identifying vulnerable nodes in the system. This research has many important implications for the owners and operators of infrastructure systems. It has demonstrated the European air traffic network to be vulnerable to spatial hazard and shown that, because many infrastructure networks possess similar properties, may therefore be equally vulnerable. It also provides a method to identify generic system vulnerabilities and strategies to reduce these. It is argued that as this research has considered generic networks it can not only increase infrastructure resilience to known threats but also to previously unidentified ones and therefore is a useful method to help protect these systems to large scale disasters and reduce the suffering for the people in the communities who rely upon them.

The thesis is mainly concerned with the Finite element static and dynamic models for the behaviour of reinforced concrete using non - associated flow rule. Applications of these models to the analysis of plates and some threedimensional problems are reported. The thesis can be broadly divided into four parts. The first part describes the geometric and material nonlinearities, and the non symmetry of stiffness matrix arising from the use of non associated flow rule. Some techniques are proposed to achieve symmetry by the use of scalar measures, such as incremental work done, plastic multiplier and a non associator corrector. The performance of these techniques is illustrated by a wide range of problems ranging from granular materials such as soils and concrete, to non granular materials such as metals. The second part of the thesis gives the description of the model for static behaviour of concrete. A five parameter model is adopted for the yield criterion, while the modified Drucker Prager function describes the plastic potential function. A work hardening plasticity, reflecting the multiaxial stress state is adopted and an energy recovery strategy for unloading (during crack closure) is reflected in the model. The model is examined by a number of numerical examples. The third part gives the description of a model for the dynamic behaviour of reinforced concrete. The rate sensitivity of such material properties as the elasticity modulus, yield stress and fluidity parameter are highlighted and the parameters involved are identified using some existing experimental data. The model is implemented in a three dimensional program and used in analysing some numerical examples. The fourth part of the thesis is devoted to the study of some model response to various factors, the choice of which is user dependent. Such factors include mesh discretisation, time and load increment size, numerical integration rule etc. Some conclusions that could be beneficial to the program user are arrived at. The inclusion of such information in a user - manual may be of some help to the the program user, who may be unaware of the inherent assumptions made in the model and program development.

Permeability is widely considered as a measure of durability. The main objective of this study was to investigate some of the significant factors that influence the measured relative gas permeability of concrete. In particular the study set out to reduce the time taken to perform a permeability test. Initially the effect of conditioning at 105°C and 85°C on the strength and permeability of Normal Strength Concrete (NSC) and High Strength Concrete was examined. It was shown that although similar compressive and tensile strength results were obtained irrespective of conditioning temperature, marginally higher permeabilities were reported for the concrete conditioned at 105°C. Concrete specimens were then conditioned to obtain varying degrees of percentage weight loss to determine whether conditioning to a completely dry state could be avoided and whether reliable results could be achieved from concrete specimens similar to those found in-situ. A power relationship resulted between percentage weight loss and relative permeability; therefore, it is possible to estimate the permeability of a specimen from the test results without the need to condition it fully. Next, a study into the effect of gas path length on the relative permeability of NSC and a higher strength concrete was performed. From the study a linear relationship was observed between the two variables for both concretes signifying that smaller specimens can be tested and reliable permeability results obtained. However, coefficients of variation of between 30% and 45% have been noted repeatedly for relative gas permeability results from the experimental investigations. Finally, an analytical and a finite difference numerical model were developed to replicate the relative permeability test set up. Both of the models showed excellent agreement with the experimental data and therefore relative permeability test results can be used to obtain intrinsic permeability coefficients. The benefit of the models is that within the limits of the relative permeability test set up intrinsic permeability coefficients can be obtained for specimens of any diameter and length.

Low Energy Dynamic Compaction (LEDC) is a recently adopted approach for the rapid improvement of the mechanical behaviour of soil to a relatively shallow depth over a limited area. This method of compaction is termed `low energy' because the energy input per blow is low compared with that imparted by traditional dynamic compaction techniques. The field apparatus for this method was designed originally by BSP/DRA for the rapid repair of bomb damaged airfield runways, but later adopted as a method of ground improvement. This thesis describes the design and operation of a unique model compactor simulating the action of a low energy compactor in the centrifuge. The centrifuge study has been undertaken at 20g using a tamper mass of 0.875 kg falling through 100mm onto a stiff aluminium target, having a mass 0.268 kg and a diameter of 100mm. This simulated a field scale tamper mass of approximately seven tonnes falling through two meters onto a target having a base area of 3.14m2. This research required the development, for use in the centrifuge, of a process monitoring system proposed by Allen (1996), where the improvement in the ground characteristics may be evaluated with the execution of the improvement process. The procedure was based on the principles of the WAK test ( Briaud and Lepert, 1990 ) which was devised to provide, by means of a very simple test, an estimate of the static stiffness of a soil/footing system, thus removing the necessity for expensive pre- and post- test investigations. The monitoring system required the innovative deployment of miniature instrumentation on the pounder and the target to measure both input force and output accelerations In order to analyse the soil response to the impact it is necessary to obtain both the signature of the energy input and soil response in terms of acceleration and transient earth pressures for each impact. This was achieved by the use of a dynamic load cell mounted on the drop weight, an accelerometer on the target and accelerometers and dynamic earth pressure cells embedded within the soil mass.

This thesis is concerned with a study of the design of reinforced concrete slab structures. Such a study is most opportune at the present time when the basis of the design of reinforced concrete structures is being re-examined and design recommendations being reformulated within the concepts of limit state design. The work is divided into four parts. In Part 1 the methods of analysis available for the solution of reinforced concrete slab problems are first discussed. Particular attention is paid to the idealizations involved in the use of these methods and the degree to which they may be employed by the designer. Subsequent to an appraisal of existing design philosophies for reinforced concrete structures the recommendations for slab design contained in the present code of practice for structural concrete (CP 114 1957) are critically examined and a number of unsatisfactory features of these recommendations are pointed out. The actual behaviour of reinforced concrete slabs is discussed with particular reference to membrane effects which are generally ignored in existing methods of analysis and design recommendations. Having established the importance of membrane effects methods of analysis for reinforced concrete slabs acknowledging these effects are developed in Part 2. In particular in Section 4a finite difference method of elastic analysis for beam-slab structures which allows for full composite action between the slab and the supporting beams is presented. Using this technique the behaviour of various square uniformly loaded beam-slab panels is investigated. In sections 5 and 6 methods of analysis for predicting the ultimate load behaviour of simply supported rectangular reinforced concrete slabs are developed. These analyses allow for the menbrane effects which are observed in the large deflection behaviour of such structures. Correlation between these new solutions, existing methods of analysis and experimental results are examined. In Part 3 the results of three separate experimental investigations carried out by the author are reported. The behaviour of a series of Perspex beam-slab panels is compared with the predictions of the finite difference analysis refered to above and an alternative analytical approach using the finite element technique. A series of tests on reinforced concrete beam-slab panels is described in Section 8. The behaviour of the specimens is carefully studied with a view to determining the salient features which must be allowed for in the design of such structures. The correlation between the behaviour of the test panels and the methods of analysis described in Part 2 is examined. The results of a series of tests on rectangular simply supported uniformly loaded reinforced concrete slabs are presented in Section 9. Although working load conditions are considered, particular attention is paid to the collapse modes for such structures. In particular the formation at large deflections of in-plane bending hinges is noted. (The results of these tests are compared in Section 6 with the analysis developed by the author which allows for this effect). In Part 4a new basis for the design of reinforced concrete slab structures is presented. These proposals are based on an examination of the existing recommendations and experimental evidence presented by the author and other research workers. The need for methods of analysis which deal realistically with the behaviour of reinforced concrete slabs is apparent and the possible use of the analytical developments described in Part 2 is outlined.

Occupant Restraint Systems (ORS) have been widely used in Public Service Vehicles (PSVs). A Wheelchair Tiedown and Occupant Restraint System (WTORS) has been developed to provide effective occupant protection for disabled people who are seated in wheelchairs. An international laboratory study had been conducted to produce a compliance test protocol that included specification of the sled deceleration versus time history and the crash pulse corridor. Currently effort at the international level is being focused through the International Standards Organisation (ISO) to produce standards for WTORS and transportable wheelchairs. Dynamic sled testing of WTORS was conducted in Middlesex University Road Safety Engineering Laboratory (MURSEL) to develop a test protocol in a WTORS System. This research has been concerned with the effects to which the occupant of a wheelchair secured by a WTORS is subjected in a frontal impact. Both occupant Forward Facing Frontal (FFF) and Rearward Facing Frontal (RFF) impact configurations have been considered. A Surrogate wheelchair with a tiedown restraint System, a Surrogate occupant restraint System, and an Anthropomorphic Test Dummy (ATD) were used to facilitate highly controlled tests. Production wheelchairs were also crash tested to validate the response of the Surrogate System. A 48 km/h-20g crash pulse falling within the ISO standard crash pulse corridor was specified. The Crash Victim Simulation (CVS), one of the computer modelling methods, and Finite Element Analysis (FEA) models were designed to study the dynamic response of a restrained wheelchair and its occupant in a crash environment. Two CVS computer packages: MADYMO®, DYNAMAN® and one of FEA programs: PAFEC were used in WTORS models to predict the occupant response during impacts and hence provide data to optimise future system design. A modelling protocol for WTORS was developed based on the results of ninety (90) sled tests of WTORS Surrogate system and forty (40) dynamic tests of production wheelchairs. To illustrate the potential of these models the results of simulations were validated by sled tests. A random effects Statistical method was used to quantify the results. The load-time histories were also traced to qualify the test and model results. A literature review highlighted twenty years of wheelchair crash research. The correlation between computer model and experimental results was made more accurately. The modelling technique of interconnection of FEA models into CVS program was also introduced. The velocity profile and the natural frequency of WTORS analysis were used to explain why the wheelchair and dummy experienced acceleration amplifications relative to the sled. The shoulder belt load at floor-mounted configuration was found to be higher than that at B pillar configuration. Energy principles were also applied to show why more compliant wheelchair tiedown Systems subjected restraints to a less severe crash environment. A decomposition of forces using the computer model showed why quasi-static analysis is insufficient in WTORS design. It is concluded that the B pillar anchorage of the occupant diagonal strap is superior to the floor-mounted configuration.