The lateral stiffening effects of cladding and partition walls, which are usually unaccounted for in a building structure's design, are investigated in this research project. Direct and iterative, linear elastic finite element analyses of representative modules of these components and their supporting primary structure were performed. These were used to study their general lateral load behaviour, and to establish their modes of interaction and induced forces. As a result, new and practical analogous strut models have been devised to allow their incorporation in, and the analysis of, the total building structure. The strut models permitted the effects of the non-structural elements' interaction on the static and dynamic responses of tall building structures to be studied. The ultimate objective of this work has been to contribute towards the development of new procedures of analysis and design of building structures braced by precast concrete cladding panels and non-loadbearing concrete blockwork walls.

The mechanical behaviour of a rubber-like elastic material is characterized by its strain energy function, which depends on the type of rubber, including its molecular structure and cross-linking. This thesis presents the results of a series of uniaxial tests that were conducted on natural rubber samples, for the purpose of determining the form of its strain energy function. The simplest model that provides a good match with the stress-strain data is chosen to identify the strain energy function. The validation of the model chosen is performed through an experiment involving fluid loading of a natural rubber membrane that is fixed along a circular boundary. The experimental results of the deformation of the membrane are in good agreement with the values obtained from computational modelling. The studies indicate that the Mooney-Rivlin form of the strain energy function can accurately predict the mechanical behaviour of natural rubber at moderately-large strains. / Le comportement mécanique des matériaux élastiques comme le caoutchouc est caractérisé par sa fonction d'énergie de déformation, qui dépend du type de caoutchouc, y compris sa structure moléculaire et sa réticulation. Cette thèse présente les résultats d'une série de tests uni-axiaux qui ont été réalisés sur des échantillons de caoutchouc naturel, afin de déterminer sa fonction d'énergie de déformation. Le modèle qui correspond le plus précisément aux données contrainte-déformation est utilisé pour prédire la fonction d'énergie de déformation. La validation du modèle a été faite en réalisant des tests expérimentaux comportant le chargement liquide d'une membrane de caoutchouc naturel qui est fixée le long d'une frontière circulaire. Les résultats expérimentaux de la déformation de la membrane sont en accord avec les valeurs obtenues par la simulation numérique. Les études indiquent que pour les déformations modérées, le modèle d'hyperélasticité Mooney-Rivlin peut prédire avec exactitude le comportement mécanique du caoutchouc naturel.

The recirculating flows behind a sudden expansion in an open channel are computed using three different turbulence models: (i) a standard single-length-scale $ kappa$-$ epsilon$ model, (ii) a two-length-scale $ kappa$-$ epsilon$, and (iii) a constant-eddy-viscosity model. The performance of these models is evaluated by comparing the numerical results with the experimental data obtained from the previous investigation. / The flow simulation is characterized by two basic dimensionless parameters: a turbulent Reynolds number, $Re sb{T},$ which defines the level of eddy viscosity, and a bed-friction number, S, which represents the effect of bed friction. The study shows that in the limit of shallow water depth, that is S $>$ 0.10, the mean flow is quite successfully predicted by all employed models. However, in the limit of deep water depth, S $<$ 0.10, both $ kappa- epsilon$ models under-predict the length of the recirculating region due to the high level of computed eddy viscosity. On the other hand, the study indicates that the constant viscosity model gives quite acceptable results for most engineering applications. / Advantageously using the constant viscosity model's simple concept, an attempt is made to define a criterion for numerical stability of the computational procedure. The stability of the algorithm is assessed by varying the flow Reynolds number, the bed-friction number as well as the mesh size. The Courant number, a dimensionless parameter, is then introduced and correlated with the $Re sb{T}$ and S, thus providing the means to determine the stability of the numerical calculations. / As most of the recirculating flows observed in natural waterways are dominated by the bed-friction effect, accurate simulation of the mean flow field is possible even with an incorrect model for the lateral exchange process.

In this thesis, the effect of geometrical roughness of a non-deformable road surface on the mobility performance of a pneumatic tyre is investigated. A system function which represents the theoretical relationship between the drawbar-pull and the random traction surface geometry is developed in the frequency domain, associated with the energy conservation law. / Prior to the analysis, measurement of road profile is achieved using an ultrasonic distance detector, and characterization of the stochastic road profile is evaluated by a power spectral density function. Further to this, the resulting drawbar-pull and tractive efficiency for a given slip rate, applied torque, tyre characteristics (as a function of inflation pressure) and vehicle moving velocity are evaluated through the above-mentioned system function. / The enveloping function, describing the modification effect of the original road profile by a flexible tyre, is also experimentally investigated. Systematic laboratory tests are carried out by using four pseudo-random rough traction surfaces, for comparison and verification of the theoretical predictions. / Through combined theoretical and experimental investigations, the following conclusions are obtained: (1) As the traveling speed increases, tractive efficiency is gradually decreased. (2) Higher slip rates give less tractive efficiency, however, an optimum slip rate between 2% and 10% provides the highest efficiency. (3) Tractive efficiency increases hyperbolically with an increase in the inflation pressure of the tyre. (4) A substantial decrease in tractive efficiency is obtained when the condition of the road surface becomes sufficiently rough that the road would be classified as "poor" or rougher under the ISO recommendations, at a speed of 5 m/s (18 km/h).

An explanation is provided for the slab behaviour in composite beams at web holes where the concrete slab carries heavy vertical shear. This is based on the truss concept, and requires consideration of shear studs in the hole region as vertical tension members. According to this, a structural action between the concrete slab and shear connectors for carrying or transferring vertical shear to the steel beam was clearly identified. / On the basis of the slab behaviour identified, truss idealizations capable of determining the slab shear carrying capacity in a rational manner were developed. Then, the ultimate strength for composite beams at web holes was formulated including the truss idealizations. Another ultimate strength analysis accounting for the slab shear carrying capacity in a simple manner, which was also developed during this research project, is given. This provided the fundamental solution procedure for the plastic analysis used. / A series of nine tests was carried out with particular attention being directed to the verification of the proposed truss analogy. The major test parameters included the configurations of the studs in the hole region, the width of the concrete slab and stud detailing near the high moment end of the hole. The ultimate strength predictions were made by the two methods developed, and compared with previous and present test results.

The general bearing mode of failure often controls the capacity of cold-formed steel bolted connections loaded in shear. Previous research has shown that the absence of washers can significantly lower the bearing resistance of these connections. Experimental studies have also revealed that the bearing capacity can be formulated with a gradated bearing coefficient that is governed by the ratio of fastener diameter to sheet thickness. / A gradated bearing coefficient method, which was based on an investigation of bolted connections composed of mild steel as well as thin high strength steel, has recently been proposed to cold-formed steel design code committees in North America, Australia and New Zealand. However, the new bearing method does not include provisions for the design of washerless bolted connections. For this reason a total of 168 bolted connection shear tests designed to fail by bearing were performed. The test program included connections with and without washers, which were divided into single and double shear, as well as high strength steel (Grade 550) and mild sheet steel (Grade 230) categories. The thickness of the sheet steel material ranged between 0.43 mm and 1.53 mm. In addition, data for 381 washerless bolted connection specimens from previous experimental investigations was included in an evaluation of current cold-formed steel design standards and specifications. Modifications to the connection design procedure contained in the 2001 North American Specification were proposed.

Alkali-aggregate reactivity (AAR) is known to cause detrimental effects in reinforced and prestressed concrete structures. To be able to design for these disruptive deformations and structural distress, it is necessary to have a clear understanding of the basic mechanics of reactivity, and its effect on cracking, engineering properties of the deteriorating concrete, and the resulting structural deformations. / This research will provide a brief overview of alkali-aggregate reactions in concrete, the causes of the problem and its symptoms in existing concrete structures. The goal of this study is to assess the role of the steel reinforcement on the response of AAR-affected concrete; the paper will also review how the reaction affects the mechanical properties of concrete. Expansion and compression tests as well as Poisson's elastic modulus are being studied. The results show that reinforcement has a significant effect in decreasing the expansive effect of the AAR. / Some preventive measures are reviewed, focusing on the minimization of AAR with the use of quality aggregate, portland cement, and mineral and chemical admixtures.

In this research program two coupling beam specimens were constructed and tested under reversed cyclic loading. One specimen was reinforced with horizontal reinforcing bars and hoops in accordance with the design and detailing requirements for ductile beams. Another specimen was reinforced with diagonal reinforcing bars in addition to the horizontal reinforcing bars and hoops. The diagonal reinforcing bars were not locally confined by closely spaced hoops to see if there would be adequate confinement and anti-buckling resistance. Both the specimens had an identical span-to-depth ratio of 3.0. The testing results showed that the diagonally reinforced coupling beam had better performance than the conventionally reinforced coupling beam when subjected to reversed cyclic loading. It is also concluded that the use of diagonal reinforcing bars is easier to construct than the diagonal reinforcement with closely spaced hoops.

A single-storey precast concrete structure constructed in 1963 with masonry-infilled walls was selected for a seismic evaluation. The structure is located in the Montreal region, and is considered a post-disaster emergency building since it houses spare parts for municipal facilities. The warehouse was first evaluated through a visual inspection and identified as being seismically vulnerable. A series of three-dimensional models of the structure were analyzed to evaluate the seismic performance of the structure with respect to the seismic design requirements of the 1995 and 1965 edition of National Building Code of Canada (NBCC). The structural models were analyzed using a linear static analysis accounting for the effects of different diaphragm configurations on the lateral load resisting columns and the effects of masonry-infilled walls. The seismic analyses showed that the warehouse does not satisfy either of the NBCC 1995 and NBCC 1965 requirements. A seismic rehabilitation scheme is proposed comprising the addition of reinforced concrete shear walls, steel braces and an upgraded roof diaphragm. The proposed retrofit satisfies the seismic design requirements of NBCC 1995. Finally, the proposed seismic design provisions of NBCC 2005 are briefly summarized and the impact of the new seismic design provisions on the retrofitted structure is discussed.

An experimental study was conducted to investigate the effect of confinement on the response of ductile shear walls subjected to reversed cyclic loading. Two shear walls, with identical dimensions, were designed in accordance with the Canadian Standard Association A23.3-04, except that a greater amount of confinement was provided in boundary elements. One specimen used normal-strength concrete, and the other used high-strength concrete. A number of confinement analytical models were selected to predict the behaviour of each specimen. The test data of two moderately confined shear walls (Liu 2004) was used to compare the responses of ductile shear walls at different levels of confinement. These predictions and comparisons with the test results are presented and discussed in terms of load-carrying capacities, ductility, moment-curvature responses and energy dissipation. / The test results indicate that the increased confinement of the boundary elements significantly enhanced the displacement ductility without appreciable change in load-carrying capacity. In addition, the beneficial effects of increased confinement were more pronounced in the normal-strength concrete shear walls than in the high-strength concrete shear walls.