An experimental study of typical connections commonly used in precast concrete buildings subjected to reversed cyclic loads is presented, along with an analytical study of nonlinear behaviour of these specimens subjected to monotonically increasing loads until failure. / Five different connection specimens were constructed and tested under reversed cyclic loading to determine their seismic responses. The specimens represent a series of horizontal connections between double-tee flanges or between a double-tee flange and lateral load resisting element in a typical roof of a precast concrete structure. / The analytical study consisted of two parts: linear finite element analysis of the connection to determine the test specimen geometry and to define its boundary conditions, and non-linear finite element analysis, using the NONLACS program, to predict the complete responses of the connections subjected to monotonically increasing loads. / The experimental results were compared with the predictions made using theoretical calculations, the CPCI design method and non-linear finite element analysis. / It is noted that the CPCI design method gives conservative predictions for the ultimate strength of the connection reinforced with a bent deformed bar anchor and a 90$ sp circ$ hook, and the one with two deformed bars oriented at 45$ sp circ$ and headed studs, both welded to a steel plate. The predicted ultimate strength of the connection with a bent deformed bar anchor welded to a steel plate and one with two straight deformed bar, oriented at 90$ sp circ$ and welded to an embedded steel angle, are close to the experimental results. The ultimate strength of the connection wit the headed studs welded to a steel plate predicted by CPCI method is overestimated. Design guidelines to increase the strength of the connections and to improve the displacement ductility are presented.

The horizontal interaction between the walls and frames in wall-frame structures is investigated to determine its role in stiffening the structure against lateral, wind or earthquake loading. Discontinuous wall-frame structures are studied to assess the effects of the discontinuities on the horizontal interaction, and on the structures' lateral stiffness. The discontinuities include stepped walls or frames, curtailed walls, and different configurations of stiffened-storey frames. / In each case, a continuum solution is developed and used to describe the behaviour of the structure, and to evaluate the effects of the discontinuities. Corresponding discrete finite elements analyses are also performed to verify the accuracy of the algebraic solutions and to determine the nature and magnitude of the resulting effects. / For stepped and curtailed wall-frame structures, it is found that the walls can be reduced or curtailed without modifying significantly the overall horizontal interaction and lateral stiffness. It is found that for stiffened-storey frame structures, the horizontal interaction and lateral stiffness are significantly increased. This constitutes an entirely new type of high-rise wall-frame structure of greater efficiency.

Six reinforced concrete specimens, modeling the wall of a nuclear reactor containment structure, were tested under cyclic thermal effects and cyclic applied loading. To study the behaviour of the reactor wall, three of the specimens were tested under axial tension, while the other three were tested in flexure. The specimens were subjected to cyclic loads, strains, and a combination of loads and strains, equivalent to 25 and 50 years in the life of the actual structure. The stiffness of the specimen and the cracks on the specimen were monitored. / The test results confirmed that the largest increases in crack widths and the largest drop in stiffness occurred in the cycling that would be equivalent to the first 25 years of the structure's service life. Proposed procedures are given to determine the responses of the tension and bending specimens for monotonic loading, after the equivalent of 25 years, and after the equivalent of 50 years. The influence of cycling was accounted for by determining suitable tension stiffening factors, as a function of the number of cycles of loading and the level of tensile strain.

There is a need to enhance our understanding of the corrosion mechanisms and their effect on the level and loss of prestressing forces. This research program consisted of accelerated corrosion tests on eight prestressed concrete prisms, measuring 127 x 127 x 2,060 mm, with three different levels of prestress (0, 30, and 60% of the ultimate strength of prestressing strand). The aim of the investigation was to simulate and monitor the corrosion of prestressing strands in prestressed systems. The specimens were submerged in a saline solution and subjected to an impressed constant voltage, aimed at accelerating the corrosion of the strands. / During the corrosion period, various electrochemical tests were performed to monitor the corrosion activity as a function of elapsed time. Strains at selected locations were measured regularly to check the instantaneous level of prestress. At the end of the experiment, the prestressing strands were removed from the concrete, visually inspected, and weighed. They were also tested in tension to determine their residual tensile strength and their mode of failure. / An assessment of the electrochemical results did not point to any relationship between the initial levels of prestress and the corrosion activity. A similar observation was made in verifying the effect of the initial level of prestress on the corrosion activity as manifested in the residual tensile strength of the strands or their final measured weights. / This research program was therefore valuable to the practicing engineer in the design and maintenance of durable prestressed concrete structures. The findings confirmed that the risk and extent of corrosion in all prestressing strands in corrosive environments are roughly similar, regardless of the initial prestressing level.

A study of the response of eight full-scale deep beams was carried out at McGill University. Four beams were tested by Li (2003) and this thesis reports on the testing of the remaining four beams. The deep beams reported in this thesis were 2000 mm long by 400 mm thick. Two beams had an overall depth of 520 mm and the other two beams had an overall depth of 810 mm. Two beams were reinforced with main tension tie reinforcement only, while the other two contained both vertical and horizontal uniformly distributed reinforcement. / These beams were tested under concentrated load to investigate the influence of span-to-depth ratio and the influence of uniformly distributed horizontal and vertical reinforcement. The presence of uniformly distributed steel resulted in higher capacities, better crack control and also served to control bond splitting failures near the supports. Four approaches were used to predict the capacities: a plane-section model, a simplified strut-and-tie model, a model based on the 1996 FIP Recommendations and a refined strut-and-tie model. The 1996 FIP (Federation Internationale de la Precontrainte) Recommendations gave conservative predictions suitable for design. The refined strut-and-tie model gave the most accurate predictions due to the fact that this approach accounted for the contributions of both the horizontal and vertical uniformly distributed reinforcement in the strut-and-tie model.

In the design of hot rolled steel structures, stability is one of the most important considerations. In terms of flexural members, e.g., a steel I-section beam, bent about the major principal axis, lateral buckling about the minor principal axis in combination with torsional buckling is the main stability concern. To prevent lateral-torsional buckling from occurring, lateral braces are typically installed perpendicular to the longitudinal axis of the flexural member, either on the top and/or bottom flange or near the shear-centre of the beam. Nethercot developed an equation to predict the critical elastic lateral-torsional moment resistance of cantilever beams, which involves a series of effective length factors, which are dependent on the brace configuration, type and position of loading, as well as cantilever system. This effective length factor approach has become widely used as a simplified method to calculate the elastic lateral-torsional buckling moment resistance of cantilever steel beams. However, no specific guidelines for the design of the lateral braces are available in design standards. / This thesis will include background information on elastic beam buckling theories, an introduction to the development of research on lateral brace requirements, and a discussion on the currently used lateral bracing requirements in North American and British Standards. Details of a study concerning the lateral bracing design requirements for cantilever I-beams, carried out with the use of the finite element software Buckling Analysis of Stiffened Plates (BASP), are provided. Concerning the required brace stiffness, the results from the numerical parametric study are compared with the results from the Canadian Steel Design Standard CSA-S 16 and the SSRC (Structural Stability Research Council) method of the United States. / It is shown that the current code methods are generally conservative for top flange loaded cantilever beam cases in terms of the predicted elastic lateral-torsional buckling capacity. Furthermore, the current bracing design requirements provide adequate results in most cases, except for the prop cantilever scenario without a tip brace.

Dam safety is always a high priority issue since failures result in severe property damage and loss of life. Many countries require that all existing dams need to be inspected and evaluated regularly. Monitoring instruments have long been an integral part of the dam safety process by providing information for detailed analyses as well as for monitoring the current condition of dams. In this thesis, statistical models are used to infer behavior characteristics of a multiple arch dam. First, a statistical model known as the Hydrostatic-Season-Time (H-S-T) model is fitted to each individual displacement component from pendulum data for the Daniel Johnson Dam, a multiple arch concrete dam located in Northern Quebec. Principal component analysis (PCA) was used with the data from pendulums. A modified PCA analysis was also performed after removing thermal effects. A C++ language program (DASOD) was developed for the estimation of the H-S-T model, visualization of results, and data preparation for the principal component analysis. (Abstract shortened by UMI.)

Wood poles are the most common support structures for distribution lines in North America. Wood poles typically have a service life of 40-50 years but may be replaced prematurely when they fail or are damaged by climatic loads or when they have degraded prematurely. The distribution network of Hydro-Quebec comprises more than 2 million wood poles and the selection of an optimal class of poles for new or refurbished lines can potentially represent significant savings. In this project, wood poles from class 1 to 5 and steel poles are considered for new or refurbished lines using life-cycle-cost analysis. / The evaluation of life-cycle-costs of a line for a service life of 50 years was performed in two steps. A reliability model was first developed to estimate the probability of failure of single poles with climatic loads. The results of the reliability analysis were then integrated in an economic model that computes the Net Present Value of construction costs and pole replacement costs associated with failure under climatic loads of the whole line. Maintenance costs were neglected in this study since they were determined to be approximately similar for the various classes of poles. / The results indicate that the Class 4 poles that are currently used for the distribution network are not optimal from an economic point of view. For distribution lines located in regions with moderate ice hazard expositions, a Class 2 pole has a 15% cost advantage over the Class 4 poles. For distribution lines located in regions with severe ice hazard expositions, a Class 2 pole has a 30% advantage over the Class 4 poles.

A ductile shear wall must be capable of undergoing significant plastic deformations, while absorbing significant energy without brittle failures when subjected to a major earthquake. In general, the sectional ductility is improved with increasing amounts of compressive reinforcement, increasing the concrete compressive strength and the ultimate concrete strain. This research program evaluated the reversed cyclic loading responses of two reinforced concrete ductile flexural shear walls that were designed in accordance with the Canadian Standards Association Standard A23.3-04. The two shear walls were designed with an identical amount and detailing of longitudinal and confining reinforcement but with different concrete strengths (35 MPa and 70 MPa, respectively). The purpose was to investigate the influence of concrete strength on the reversed cyclic loading behaviour, including the overall response, shear capacity, deflection, displacement ductility, and cumulative energy dissipation. / The experimental results indicated that with the higher concrete strength, the ductility, strength and energy absorption of the shear wall were slightly increased.

This research investigated how cold temperatures affect oil/water partitioning of two commonly encountered aromatic hydrocarbon pollutants, naphthalene and benzene. Although it is generally assumed that equilibrium concentrations of organic solutes decrease with temperature, there is little data on oil-water partitioning coefficients of environmentally significant solutes or methodology description for the calculation of such coefficients. Partitioning experiments showed that equilibrium aqueous concentrations were found to depend largely on a single compound property, the excess heat of solution of the compound. Equilibrium concentrations of naphthalene decreased with decreasing temperature between 25ºC and 8ºC. But for those compounds exhibiting a solubility minimum at ambient temperature, such as benzene, equilibrium concentrations at low temperatures below the solubility minimum were found to be close to those at 25ºC.