Metallic utility poles, light poles, and mast arms are intermittently damaged by vehicle collision. In many cases the vehicular impact does not cause immediate failure of the structure, but induces localized damage that may result in failure under extreme service loadings or can promote degradation and corrosion within the damaged region. Replacement of these poles is costly and often involves prolonged lane closures, service interruption, and temporary loss of functionality. Therefore, an in situ repair of these structures is required. This thesis examines the failure modes of damaged metallic poles reinforced with externally-bonded fiber reinforced polymer (FRP) composites. Several FRP repair systems were selected for comparison, and a set of medium and full-scale tests were conducted to identify the critical failure modes. The material properties of each component of the repair were experimentally determined, and then combined into a numerical model capable of predicting global response. Four possible failure modes are discussed: yielding of the unreinforced substrate, tensile rupture of the FRP, compressive buckling of the FRP, and debonding of the FRP from the substrate. It was found that simple linear, bilinear, and trilinear stress-strain relationships accurately describe the response of the composite and substrate components, whereas a more complex bond-slip relationship is required to characterize debonding. These constitutive properties were then incorporated into MSC.Marc, a versatile nonlinear finite element program. The output of the FEM analysis showed good agreement with the results of the experimental bond-slip tests.

Fiber reinforced polymer (FRP) composites have become an attractive alternative to conventional methods for external-strengthening of civil infrastructure, particularly as applied to flexural strengthening of reinforced concrete (RC) members. However, durability of the bond between FRP composite and concrete has shown degradation under some aggressive environments. Although numerous studies have been conducted on concrete members strengthened with FRP composites, most of those studies have focused on the degradation of FRP material itself, relatively few on bond behavior under repeated mechanical and environmental loading. This thesis investigates bond durability under accelerated environmental conditioning of two FRP systems commonly employed in civil infrastructure strengthening: epoxy and polyurethane systems. Five environments were considered under three different conditioning durations (3 months, 6 months, and 1 year). For each conditioning environment and duration (including controls), the following were laboratory tested: concrete cylinders, FRP tensile coupons, and FRP-strengthened concrete flexural members. Numerical investigations were performed using MSC MARC finite element software package to support the outcomes of durability experimental tests. Precise numerical studies need an accurate model for the bond between FRP and concrete, a linear brittle model is proposed in this work that is calibrated based on nonlinear regression of existing experimental lap shear data. Results of tensile tests on FRP coupons indicate that both epoxy and polyurethane FRP systems do not degrade significantly under environmental exposure. However, flexural tests on the FRP strengthened concrete beams indicate that bond between FRP and concrete shows significant degradation, especially for aqueous exposure. Moreover, a protective coating suppresses the measured degradation. Also, experimental load-displacement curves for control beams show excellent agreement with numerical load-displacement curves obtained using the proposed bond iii model. Finally, a bond-slip model is predicted for concrete leachate conditioned beams by matching load-displacement curves for those beams with numerical load-displacement curves.

Reinforced concrete (RC) moment frame structures make up a notable proportion of buildings in earthquake-prone regions in the United States and throughout the world. The beam-column (BC) joints are the most crucial regions in a RC moment frame structure as any deterioration of strength and/or stiffness in these areas can lead to global collapse of the structure. Thus, accurate simulations of the joint behavior are important for assessment of the local and global performance of both one-way and two-way interior BC joints. Such simulations can be used to study the flexural-shear-bond interaction, the failure modes, and sensitivity of various parameters of structural elements. Most of the existing analytical approaches for interior BC joints have either failed to account for the cyclic bond-slip behavior and the triaxial compressive state of confined concrete in the joint correctly or require so many calibrations on parameters as to render them impractical. The core motivation for this study is the need to develop robust models to test current design recommendations for 3D beam-column-slab subassemblies subjected to large drifts. The present study aims to first evaluate the flexural-shear-bond interactive behavior of two-way beam-column-slab interior connections by both finite element and nonlinear truss methodologies. The local performance such as bond-slip and strain history of reinforcing steel are compared with the experimental results for the first time. The reliability of applied finite element approach is evaluated against a series of one-way interior BC joints and a two-way interior beam-column-slab joint. The accuracy and efficiency of the nonlinear truss methodology is also evaluated by the same series of joints. Results show good agreement for finite element method against both global and local response, including hysteretic curve, local bond-slip development and beam longitudinal bar stress/strain distributions. The nonlinear truss model is also capable in obtaining satisfactory global response, especially in capturing large shear cracks. A parametric study is exhibited for a prototype two-way interior beam-column-slab joint described in an example to ACI 352R-02, to quantify several non-consensus topics in the design of interior BC connections, such as the joint shear force subjected to bidirectional cyclic loading, the development of bond-slip behavior, and the failure modes of two-way interior joints with slab. Results from connections with different levels of joint shear force subjected to unidirectional loading show that meeting the requirements from ACI 352 is essential to maintain the force transfer mechanism and the integrity of the joint. The connections achieved satisfactory performance under unidirectional loading, while the bidirectional monotonic loading decreases the joint shear force calculated by ACI 352 by 10%~26% based on current results. Poorer performance is obtained for wider beams and connections fail by shear in the joint rather than bond-slip behavior when subjected to bidirectional cyclic loading. In general, the study indicates that the ACI352-02 design methodology generally results in satisfactory performance when applied to 2D joints (planar) under monotonic and cyclic loads. Less satisfactory performance was found for cases of 3D joints with slabs. / Doctor of Philosophy / Reinforced concrete (RC) moment frames are one of the most popular structure types because of their economical construction and adaptable spaces. Moment frames consist of grid-like assemblages of vertical columns and horizontal beams joined by cruciform connections commonly labelled as beam-column joints. Because of the regularity of the grid and the ability to have long column spacing, moment frames are easy to form and cast and result in wide open bays that can be adapted and readapted to many uses. In RC structures, steel bars embedded in the concrete are used to take tensile forces, as concrete is relatively weak when loaded in tension. Forces are transferred between the steel and concrete components by so-called “bond” forces at the perimeter of the bars. The proper modeling of the behavior of bond forces inside the beam-column joints of reinforced concrete moment frames is the primary objective of this dissertation. Reinforced concrete moment frames constitute a notable proportion of the existing buildings in earthquake-prone regions in the United States and throughout the world. The beam-column joints are the most crucial elements in a RC moment frame structure as any deterioration of strength and/or stiffness in these areas can lead to global collapse of the structure. Physical experimentation is the most reliable means of studying the performance of beam-column joints. However, experimental tests are expensive and time-consuming. This is why computational simulation must always be used as a supplemental tool. Accurate simulations of the behavior of beam-column joints is important for assessment of the local and global behavior of beam-column joints. However, most of the existing analytical approaches for interior beam-column joints have either failed to account for the bond-slip behavior and the triaxial compressive state of confined concrete in the joint correctly or require so many calibration parameters as to render them impractical. The present study aims to provide reliable numerical methods for evaluating the behavior of two-way beam-column-slab interior joints. Two methods are developed. The v first method is a complex finite element model in which the beam-column joint is subdivided into many small 3D parts with the geometrical and material characteristics of each part carefully defined. Since the number of parts may be in the hundreds of thousands and the geometry and material behavior highly non-linear, setting up the problem and its solution of this problem requires large effort on the part of the structural engineer and long computation times in supercomputers. Finite element models of this type are generally accurate and are used to calibrate simpler models. The second method developed herein is a nonlinear truss analogy model. In this case the structure is modelled as nonlinear truss elements, or elements carrying only axial forces. When properly calibrated, this method can produce excellent results especially in capturing large shear cracks. To evaluate the accuracy and to quantify the current seismic design procedure for beam-column joints, a prototype two-way interior beam-column-slab joint described in an example to ACI 352R-02, the current design guide used for these elements in the USA, is analytically studied by the finite element methodology. The study indicates that the ACI352-02 design methodology generally results in satisfactory performance when applied to one-way (planar) joints under monotonic and cyclic loads. Less satisfactory performance was found for cases of three-dimensional (3D) joints with slabs.

Le comportement de l’interface acier-béton a une grande importance lorsque la fissuration des structures en béton armé est étudiée. Une approche par éléments finis a été proposée par (Torre-Casanova, 2013) et (Mang, 2016) pour représenter l'interface acier-béton dans les simulations de structures à grandes dimensions Le modèle proposé permet de calculer le glissement tangentiel entre l'acier et le béton. L’objectif de cette étude est d’améliorer ce modèle initial pour le rendre plus efficace et plus représentatif. Le document est découpé en trois parties : 1) Le modèle initial de liaison est évalué. Puis amélioré tant en chargement monotone qu’alterné. Le nouveau modèle est validé par plusieurs applications numériques. 2) L'effet de confinement est implémenté dans le modèle de liaison acier-béton. L'effet sur le comportement structural du confinement actif est étudié en utilisant le nouveau modèle. A partir des simulations proposées, il est montré, par l’utilisation du nouveau modèle, que l’effet de confinement actif peut jouer un rôle sur les comportements monotone que cyclique. 3) L'effet goujon est étudié avec le nouveau modèle liaison acier-béton. Deux campagnes expérimentales différentes sont simulées avec différents modelés de renforts (1D barre et poutre) et d’interface (liaison acier-béton et liaison parfaite). Les résultats montrent que le nouveau modèle de liaison acier-béton permet de mieux reproduire les résultats expérimentaux par rapport au modèle de liaison parfaite aux échelles globale et locale. / In numerical applications of reinforced concrete structures, the steel-concrete interface behavior has a vital importance when the cracking properties are investigated. A finite element approach for the steel-concrete interface to be used in large-scale simulations was proposed by (Torre-Casanova, 2013) and (Mang, 2016). It enables to calculate the slip between the steel and concrete in the tangential direction of the interface element representation. The aim is here to improve the initial bond-slip model to be more efficient and more representative. The document is divided into three parts: 1) The existing bond-slip model is evaluated. The bond-slip model is then improved by considering transversal and irreversible bond behaviors under alternative loads. The new bond-slip model is validated with several numerical applications. 2) Confinement effect is implemented in the bond-slip model to capture the effect of external lateral pressure. According to the performed numerical applications, it is demonstrated how the active confinement can play a role, through the steel-concrete bond, during monotonic and cyclic loading cases. 3) Dowel action is finally investigated with the new bond-slip model. Two different experimental campaigns (Push-off tests and four-point bending tests) are reproduced with different reinforcement (1D truss and beam) and interface (new bonds-slip and perfect bond) models. The results show that the proposed simulation strategy including the bond slip model enables to reproduce experimental results by predicting global (force-displacement relation) and local behaviors (crack properties) of the reinforced concrete structures under shear loading better than the perfect bond assumption which is commonly used in the industrial applications.

Dieser Beitrag befasst sich mit der experimentellen und analytischen Bestimmung der Verankerungslängen textiler Bewehrungsstrukturen einer Textilbetonverstärkungsschicht. Die experimentelle Untersuchung des Verbundverhaltens erfolgte anhand von Pull-Out-Versuchen. Die analytische Betrachtung des Verbundproblems geschieht aufbauend auf multilinearen Lösungen der Verbunddifferentialgleichung anhand der experimentell ermittelten Kraft- Rissöffnungs-Beziehungen. Mit Hilfe eines separaten Modells wird aus der so ermittelten Verbundspannungs-Schlupf-Beziehung (VSB) die zur Verankerung einer entsprechenden Kraft F erforderliche Verankerungslänge lE bestimmt. Die Überprüfung der Berechnung erfolgt anhand von unabhängig in experimentellen Versuchen zur Bestimmung der Verankerungslänge ermittelten Werten. Es konnte eine gute Übereinstimmung der berechneten mit den versuchstechnisch bestimmten Verankerungslängen festgestellt werden.

No description available.

Dans cette thèse, plusieurs approches de modélisation de composites renforcés par des fibres sont proposées. Le matériau étudié est le béton fibré, et dans ce modèle, on tient compte de l’influence de trois constituants : le béton, les fibres, et la liaison entre eux. Le comportement du béton est analysé avec un modèle d’endommagement, les fibres d'acier sont considérées comme élastiques linéaires, et le comportement sur l'interface est décrit avec une loi de glissement avec l’extraction complète de la fibre. Une approche multi-échelle pour coupler tous les constituants est proposée, dans laquelle le calcul à l'échelle macro est effectué en utilisant la procédure de solution operator-split. Cette approche partitionnée divise le calcul en deux phases, globale et locale, dans lesquelles différents mécanismes de rupture sont traités séparément, ce qui est conforme au comportement du composite observé expérimentalement. L'identification des paramètres est effectuée en minimisant l'erreur entre les valeurs calculées et mesurées. Les modèles proposés sont validés par des exemples numériques. / In this thesis, several approaches for modeling fiber-reinforced composites are proposed. The material under consideration is fiber-reinforced concrete, which is composed of a few constituents: concrete, short steel fibers, and the interface between them. The behavior of concrete is described by a damage model with localized failure, fibers are taken to be linear elastic, and the behavior of the interface is modeled with a bond-slip pull-out law. A multi-scale approach for coupling all the constituents is proposed, where the macro-scale computation is carried out using the operator-split solution procedure. This partitioned approach divides the computation in two phases, global and local, where different failure mechanisms are treated separately, which is in accordance with the experimentally observed composite behavior. An inverse model for fiber-reinforced concrete is presented, where the stochastic caracterization of the fibers is known from their distribution inside the domain. Parameter identification is performed by minimizing the error between the computed and measured values. The proposed models are validated through numerical examples.