An investigation has been made into the possibility of using composite materials for intermediate range structural designs. That is, designs between the small and relatively low stressed requirements such as furniture where the quantity of G.R.P. does not dominate the cost and higher requirements such as in aircraft and ships where the cost is acceptable. As an example of this level, a joint investigation with industry was made into the possibilities of using composites for the quantity manufacture of international shipping containers. The only acceptable material for such a large structure (weighing 2 tons) is Glass Reinforced Polyester and while the whole container was considered this thesis concentrates on the design and approval of the floor structure. The properties of a variety of G.R.P. glass forms and polyesters have been measured from material tests and statistical safe values derived. As stiffness is of great importance both static and dynamic elastic modulus have been measured. Floor panels have been designed and tested and these show that to meet both structural and manufacturing requirements the glass form will be a mixture including, woven rovings, uniaxial fibres, chopped strand mat and possibly a low density filler. The use of adhesive bonding as a main method of jointing has led to a programme of tests including the effect of joint geometry. These methods are to be preferred to mechanical fasteners even when joints are made to metals. It was found to be possible to design a container floor to meet the International Standard at Q structure weight which is better than existing containers. The cost of the glass fibre and polyester made the container cost appreciably higher than the price of conventional containers.

End plate connections are extensively used as moment resistant connections between members in steel frame. Surveys of the English and Scottish Steelwork Industry clearly indicate that the flush end plate connection is the most popular type of beam-to-column connection in steel-framed structures. The popularity of this connection can be attributed to the simplicity of the connection detail and economy associated with their fabrication and erection. Flush end plate connection is less rigid and has a lower moment capacity than that of an extended end plate connection. If a rigid joint is aimed extended end plate connection should be used, whereas if a semi-rigid joint is needed flush end plate can be employed. The main objectives of this project were to carry out in-depth investigation of the behaviour of this type of connection by applying finite element technique and experimental means. A three dimensional finite element prediction model of the unstiffened flush end plate beam-to-column connection was developed. Six full scale tests were conducted and the results were analysed. Comparison between analytical and experimental results was made. The analytical investigation into the contribution of the various connection components toward the moment rotation characteristics was carried out. The investigation of bolt force and prying force were also carried out. Comparison between analytical, experimental results and the results obtained by applying the design rules of Eurocode 3 was made. By comparing the experimental results with the analytical results using finite element method, it was found that the finite element method was quite capable of tackling the complex problem of flush end plate connections. Finite element computer models can be used to simulate structural behaviour of the connections, which can be useful to the design of the connections. By comparing the results of the tests, finite element analyses and the design rules of Eurocode 3, it was found that the Annex J of Eurocode 3 significantly underestimated the moment resistance capacity of many joints and appear to predict the failure type incorrectly. Recommendations on future work on column web buckling, the effect of bolt heads and nuts, the sectional fillets and the effect of welding are also made which should be carried out before a comprehensive design procedure could be developed.

In this thesis we develop the theory for two separate computer programs capable of modelling the plane strain consolidation of a soil layer under a variety of types of surface loading. We consider loading from flexible footings, rigid footings, built-up embankments and a plane frame structure on individual footings. The first program uses a simplified form of finite element analysis for the frame structure, and a method of discretizing the surface loading into a series of line-loads. A stress distribution theory applicable to a soil layer on a smooth rigid base is then combined with classical elasticity theory and a finite difference solution of the two-dimensional Terzaghi consolidation equation. The second program uses a unified finite element analysis modelling both structure and soil, and incoporating the Biot theory of consolidation. A process of smoothing the immediate nodal excess pore-pressures is developed, which allows standard types of finite element to be used in the soil model. In addition, it is shown how a set of data of void ratio against pressure from a compression test may be analysed (using smoothing splines) with a desk top computer to yield an estimate of the pre-consolidation pressure of a soil sample. Numerical results from both programs are presented and compared for a number of loading problems, and it is concluded that the finite difference program is considerably more efficient in the solution of problems involving homogeneous soils and loading from flexible and rigid footings; in contrast, the unified finite element analysis has advantages in the solution of more complex problems.

No description available.

No description available.

No description available.

In chapter 2, a description of compartment fire physics is given. The development of fires in enclosures is explained and the main physical phenomena are described (rate of heat release, fire plume, vent flow, heat transfer to partitions...). The analytical models and correlations most widely used to model these phenomena are given. The first part of chapter 3 is a review of the existing compartment fire models, from the well-known nominal fire curves to the most sophisticated computational fluid dynamic fire models. Analyses of some of these models are made and the needs for improvements are pointed out in the second part of chapter 3. The description of the zone model developed in this work is given in chapters 4, focused on the formulation of the model. In chapter 5, the different fire scenarios in relation with the model are presented as well as their impact on the design of structural steel elements. Comparisons of the compartment fire model included in OZone with full scale fire tests are presented in chapter 6. These comparisons enable to assess the code and to define some limits beyond which the code is not able to predict reasonably the fire course. Finally in chapter 7, the new parametric fire curves developed by the author are presented.

Thesis--Ph. D., University of Hong Kong. / Mimeographed.

The work described in this thesis involved the prediction and back-analysis of ground movements and pore water pressures around a temporary excavation, lOm deep, 105m long and 35m wide (50m long and lOm wide at its base) in the heavily overconsolidated Oxford Clay. An experimental programme was carried out which was designed to investigate the nonlinearity and the anisotropy in the soil's response. A series of computer controlled triaxial stress path tests were carried out on specimens of Oxford Clay trimmed from block samples. The initial anisotropic stress state resulted in a highly stress path dependent stiffness, and the stressstrain behaviour was closely linked to the time at which the specimen had been held at constant stress prior to a change in loading. In the triaxial apparatus, shear wave propagation tests using square wave input functions into bender element apparatus were significantly affected by near-field effects and by additional wave components. Tests carried out using sine wave inputs provided a far more consistent output allowing correlation analyses and easier visual identification of the travelling shear wave. The shear modulus of the soil at very small strains could then be determined. A stress path dependent, non-linear, cross-anisotropic elastic model was developed and implemented into the finite element program CRISP. A genuine prediction of the Elstow excavation was carried out while the instrumentation data from the site investigation were temporarily withheld. It was found that the horizontal displacements were modelled satisfactorily but that the vertical displacements were in error by as much as 2 or 3 times. This discrepancy was attributed to volumetric changes suggested by the instrumentation data. A parametric analysis was carried out in which the effects of the initial stress state, the degree of anisotropy, and the degree of non-linearity were investigated. This showed that although it was possible to improve the accuracy of the prediction locally, it was not possible to improve on the overall pattern of behaviour predicted by the first non-linear cross-anisotropic analysis.

This thesis details the results of an investigation into a technique for determination of "useful" structural parameters from forced vibration testing. The implementation of this technique to full scale civil engineering structures was achieved by several developments in the experimental and computational fronts: a vibration generator and a computer-aided-testing system for the former and two computational algorithms for the latter. The experimental developments are instrumental to exciting large structures and acquisition of large quantities of useful data in digital format. These data serve as inputs to the computational algorithms whose outputs are structural parameters. These parameters are in either modal or spatial forms which cannot be measured directly but have to be extracted from the raw data. The modal-parameter-extraction method is based on direct Least-Square fitting technique and is simple to implement. The technique can yield good accuracy if the residual effects from out-of-range modes are removed from the raw data before fitting. The spatial-parameter- extraction method distinguishes itself from other conventional methods in the way that the orthogonality property is not explicitly used. This method is applicable to situations where conventional methods are not; i.e. in cases if modal matrices are not square. Some success was achieved in cases in which computer synthesized or good quality laboratory test data were used. Full scale field tests of a tall office block and a slender tower were carried out and their modal models obtained. Attempts to obtain spatial models of these structures were not carried out, however, as this task can be a separate research topic in its own right. Further research in such application is still required.