Basal reinforcement along with individually capped foundation piles is used in cases where both embankment stability and surface settlement control are required. The technique has been utilised to prevent differential settlement between new embankment construction over soft soil and an existing embankment where settlement has ceased. The piled embankment solution is also adopted to prevent differential settlement between an approach embankment constructed over soft soil and the piled foundations of a bridge abutment. The study was conducted to investigate the behaviour of an idealised piled embankment incorporating basal geosynthetic reinforcement. Three-dimensional model tests at self-weight conditions were carried out to evaluate the effect of some of the factors affecting the arching mechanism and the development of surface settlement in piled embankments. The physical model was designed to represent a square grid of individually capped piles centrally located within an embankment. Three different pile cap sizes and four different geosynthetic materials were employed in the experimental study. A movable base supported on hydraulically operated jacks was used to model the soft ground. The use of a movable base permitted the simulation of a worst case scenario in which the soft ground was not involved in the load sharing mechanism. The experimental results indicated the existence of two modes of behaviour pertaining to a shallow and deep mechanism. The piled embankment geometry represented by a combination of height of fill, pile cap size and spacing was found to govern the mode of behaviour. The arching mechanism in the fill was found to be mobilised at a relatively small reinforcement deflection which supported the adoption of the two step approach utilised in designing the basal reinforcement. The circular and parabolic arc geometries were found to be adequate in describing the deflected shape of the reinforcement. The use a modified flexible cable formulation to describe the loaddeflection response of the reinforcement was found to be in good agreement with the experimental results. In addition, the validity of a number of current methods and recommendations relating to the design of piled embankments was addressed. A numerical study was undertaken using FLAC, a plane strain finite difference based programme. The calculated and measured results were compared to assess the suitability of modelling piled embankment behaviour using the finite difference programme. A parametric study was conducted to investigate the role of the basal reinforcement in the load transfer mechanism and in the prevention of surface settlement. The embankment geometry was identified as the significant factor influencing the reduction in differential settlement. A surface settlement mechanism was established based on results of the parametric study.

Structural engineers and architects have a responsibility for incorporating fire safety into their building designs in order to minimize loss of life and property. To meet this requirement, extensive research has been carried out, aimed at obtaining better understanding of the performance of steel and composite structures under fire conditions. Recent research indicates that the robustness of steel connections is vitally important to the fire resistance of steel-framed composite buildings. The development of effective connection models is a key issue in this research field. This PhD research is focused on the development of robust connection elements, for modelling steel connections at elevated temperatures. In this work, a robust simplified two-node connection element has been developed, for modelling the behaviour of the bolted end-plate connections between steel beam and column at elevated temperatures. The proposed numerical procedure is based on the model proposed by Huang (2011), incorporating additional developments to more precisely determine the tension, compression, and bending moment capacities of end-plate connections in fire. The proper failure criteria are proposed to calculate the tension capacity for each individual bolt row. In this new model, the connection failure due to bending, axial tension, compression and shear are considered. The influence of the axial force of the connected beam on the connection is also taken into account. This new model has the advantages of both the simple and component-based models. A total of 22 tests are used to validate the model. From these validations, it is evident that this new connection model has ability to accurately predict the behaviour of the end-plate connection at elevated temperatures, and can be used to represent the end-plate connections in supporting performance-based fire resistance design of steel-framed composite buildings. For modelling the behaviour of partial end-plate connections between steel beams and columns under fire conditions, a simplified robust 2-node connection element has also been developed. The rotational response of a partial end-plate connection at elevated temperatures comprises of two stages. These stages are due to the shift of the compression centre of the connection from the end of end-plate, to the centre of the beam bottom flange at large rotation. The model proposed in this research accounts for these two stage behaviours, representing the partial end-plate iv connection as a 2-node non-linear spring element. Characteristics of the spring, such as stiffness, tension, compression, shear strengths and bending moment resistance, are determined based on a component-based approach. This model therefore retains the advantages of both the simple and component-based models. Compared to normal component-based models, this simplified model has very good numerical stability under static solver condition, and is computationally efficient. Fourteen tests are used to validate the model, showing that the model is capable of accurately predicting the behaviour of partial end-plate connections under fire conditions. A series of numerical studies has been conducted on a 2D steel frame, subjected to ISO834 Fire and Natural Fire, in order to investigate the influences of the connections on the behaviour of steel structures. It is clear that the model can be used to represent the partial end-plate connections in performance-based fire resistance design of steel-framed composite buildings. According to full-scale fire tests, tensile membrane action within the concrete floor slabs plays an important role in affecting the fire resistance of composite buildings. It is well known that the development of tensile membrane actions relies on the vertical support along the edges of the slab panel. However, there is at present a lack of research into how vertical supports influence the tensile membrane actions of the slab. In this thesis, the performance of a generic three dimensional 45m x 45m composite floor subjected to ISO834 Fire and Natural Fire are investigated. Different vertical support conditions and three steel meshes are applied, in order to assess the impact of vertical supports on tensile membrane action of floor slabs. Unlike other existing large scale modelling which assumed that the connections behave as pinned or rigid for simplicity, the two robust 2-node connection element models described above are used to model the semi-rigid behaviour of end-plate and partial end-plate connections within the fire compartment. The impact of connections on the 3D behaviour of composite floors is taken into consideration. The load-transfer mechanisms of a composite floor, when connections fail due to axial tension, vertical shear and bending are investigated. Based on the results obtained, some design recommendations are proposed for enhancing the fire resistance of composite buildings.

This thesis recounts the experimental study of the dynamic response of a blast resistant glazing system to explosive loading. A combination of triaxial force sensors, pressure gauges, and laser displacement gauges capture the response in detail over a wide range of scenarios. The scenarios include low level blast loading to characterize the reaction at points around the perimeter of the window, moderate level blast loading to examine the repeatability of the blast scenario, and high level blast loading to capture the response during failure as the tensile membrane forms. The scenarios are modeled via an analytical Single-Degree-of-Freedom model as well as finite element modeling in ANSYS Explicit Dynamics. In addition, this study investigates some of the differences between experimental data and the predictions made by modeling.