No description available.

Deterioration of reinforced concrete structures due to reinforcement corrosion is a serious problem that faces concrete infrastructure worldwide. Effect of the rebar corrosion in the shear span on the structural behaviour is not fully addressed in the published literature. This study examined the effects of corrosion of the longitudinal reinforcement in the shear span on the structural behaviour of RC beams and the effectiveness of three rehabilitation schemes on the structural performance of such beams. The experimental program consisted of testing fifteen medium-scale reinforced concrete beams (150mm wide x 350mm deep x 2400mm long) under static load. Test variables included: span to depth ratio, the degree of corrosion and the anchorage end condition and repair schemes. Two span to depth (a/d) ratios were considered: a/d=3.4 with one-point loading and a/d=2.4 with two-point loading. Two anchorage end-conditions were used: bonded or un-bonded reinforcement in the an-chorage zone. Four degrees of corrosion were chosen to simulate minor (2.5% to 5% mass loss), medium (7.5% mass loss), and severe (15% mass loss) degrees of corrosion. Corrosion was induced in the longitudinal reinforcement in the shear-span using accelerated corrosion techniques based on Faradays’ law. Three different repair scenarios were applied. The first scenario included removing the deteriorated concrete, cleaning the corroded steel and patching with a new self-compacting concrete. The second scenario included U-wrapping the beams cross-section using Glass fiber reinforced cement-based composite (GFRCM), and Carbon fiber reinforced cement-based composite (CFRCM) without removing the deteriorated concrete. The third scenario included patch repair and confinement by wrapping with GFRCM or CFRCM. Following corrosion and repair, all specimens were loaded statically to failure. Test results showed no major effect of shear-span corrosion on the flexural behaviour for the beams with end anchorage whereas a noticeable effect on the flexural behaviour was observed for beams with no end anchorage regions. The corrosion degree and the shear span to depth ratio affected the mode of failure for the specimens with no end anchorages. The type of repair significantly affected the overall behaviour of the corroded specimens. An analytical model was proposed and used to predict the load-deflection response of the tested specimens. The program calculated the mid-span deflection for a given load as an integration of the deflection of a series of elements, with the deflection being based on the elongation of the steel reinforcement in each element. A modified bond stress-slip model was incorporated into the calculations to account for the change in bond strength caused by the corrosion and/or confinement that are provided by repairs. The predicted results were in reasonable agreement with the experimental results.

Deterioration of reinforced concrete structures due to reinforcement corrosion is a serious problem that faces concrete infrastructure worldwide. Effect of the rebar corrosion in the shear span on the structural behaviour is not fully addressed in the published literature. This study examined the effects of corrosion of the longitudinal reinforcement in the shear span on the structural behaviour of RC beams and the effectiveness of three rehabilitation schemes on the structural performance of such beams. The experimental program consisted of testing fifteen medium-scale reinforced concrete beams (150mm wide x 350mm deep x 2400mm long) under static load. Test variables included: span to depth ratio, the degree of corrosion and the anchorage end condition and repair schemes. Two span to depth (a/d) ratios were considered: a/d=3.4 with one-point loading and a/d=2.4 with two-point loading. Two anchorage end-conditions were used: bonded or un-bonded reinforcement in the an-chorage zone. Four degrees of corrosion were chosen to simulate minor (2.5% to 5% mass loss), medium (7.5% mass loss), and severe (15% mass loss) degrees of corrosion. Corrosion was induced in the longitudinal reinforcement in the shear-span using accelerated corrosion techniques based on Faradays’ law. Three different repair scenarios were applied. The first scenario included removing the deteriorated concrete, cleaning the corroded steel and patching with a new self-compacting concrete. The second scenario included U-wrapping the beams cross-section using Glass fiber reinforced cement-based composite (GFRCM), and Carbon fiber reinforced cement-based composite (CFRCM) without removing the deteriorated concrete. The third scenario included patch repair and confinement by wrapping with GFRCM or CFRCM. Following corrosion and repair, all specimens were loaded statically to failure. Test results showed no major effect of shear-span corrosion on the flexural behaviour for the beams with end anchorage whereas a noticeable effect on the flexural behaviour was observed for beams with no end anchorage regions. The corrosion degree and the shear span to depth ratio affected the mode of failure for the specimens with no end anchorages. The type of repair significantly affected the overall behaviour of the corroded specimens. An analytical model was proposed and used to predict the load-deflection response of the tested specimens. The program calculated the mid-span deflection for a given load as an integration of the deflection of a series of elements, with the deflection being based on the elongation of the steel reinforcement in each element. A modified bond stress-slip model was incorporated into the calculations to account for the change in bond strength caused by the corrosion and/or confinement that are provided by repairs. The predicted results were in reasonable agreement with the experimental results.

In this thesis, a fully nonlinear finite element study of the flexural behavior of doubly reinforced concrete beams strengthened using different Carbon Fiber Reinforced Polymer (CFRP) composite strengthening configurations has been carried out. Prior to the study, a total of six beams were constructed, pre-cracked, strengthened and tested to failure under a four-point loading condition (Zhao and O'Riordan-Adjah, 2004). Then, for the purpose of this thesis work, detailed three dimensional finite element models were created not only to correlate the results obtained from the experiments, but also to predict the load capacity, failure modes and crack pattern of reinforced concrete (RC) beams strengthened using Fiber Reinforced Polymer (FRP) composites. Knowing the behavior for each of the materials that compose the beam (concrete, steel, bonding material or interface, and FRP laminates) and how to get their properties, an accurate and representative finite element model can be created. Tests and analytical (FE) results showed that the strengthened configuration plays an important role in the overall strength, failure mechanisms, and, more significantly, the ductile behavior of the beams. Considerable increases in the load-carrying capacity of the RC beams were observed. Increases that range from 12% (using FRP only on the bottom of the beam) to 35% (FRP on the bottom + 45 degrees sides' configuration as explained later) compared to the control beam before ultimate failure were obtained. Failure modes were also affected since the beam with only FRP on the bottom failed completely by debonding of the laminate while the beams with side FRP anchorage strips failed by a combination of composite debonding on the sides and concrete crushing. Finally, ductile behavior of the beams was greatly improved due to the application of the strengthening material on the side of the concrete beams, serving as an anchorage to the bottom fabric. The accuracy of the model has been validated comparing the results obtained from the six beam tests to the ones determined using the FE approach. Good agreement between the two has been found.

The new innovative composite material textile reinforced concrete (TRC) has been intensively investigated in Germany since the end 1990s. It has become increasingly important in the construction industry. Compared with conventional steel reinforcement, TRC has advantages such as higher load-bearing capacity, higher strength-to-weight ratio, better ductility, and non-corrosive behavior. This made them a subject of extensive research and diverse applications both nationally and internationally. In 2004, Xu et al. started research on bond properties of TRC in China in cooperation with Hans-Wolf Reinhardt et al. from the University of Stuttgart in Germany. Since then, there have been numerous researches on TRC in China. This article introduces a calculation method for the flexural capacity of reinforced concrete (RC) beams strengthened with TRC in China. For comparison, the dimensioning procedure in Germany is also presented. Subsequently, the two models are compared with each other in a case study. Both models in China and Germany have the same mathematical background and also provide similar results. However, they have some differences in definitions of material characteristics (e.g., design concrete compressive strength, strain, and stress distribution) and consideration of the damage resulting from the preloading stage.

L’objectif de cette thèse est la prise en compte du comportement raidissant en traction sur des poutres en béton armé du à l’adhérence entre les armatures tendues et le béton. Un vaste programme expérimental sur des poutres minces et des poutres épaisses permet de quantifier expérimental le profil de contraintes dans le béton tendu entre deux fissures de flexion. La création d’un nouvel élément fini est ensuite mise en œuvre à partir des observations expérimentales. Les tests conduits sur des tirants ont permis de mettre en évidence à la fois l’existence d’une longueur de transfert et une longueur de totale décohésion qui caractérisent le phénomène d’adhérence entre l’acier le béton entre deux fissures consécutives de flexion. La longueur de transfert correspond à la mise en tension du béton et donc l’établissement de la contribution du béton tendu à la rigidité flexionnelle, ce qui limite la diminution de rigidité de la poutre après fissuration. Les mesures de la flèche en flexion des différentes poutres testées, minces ou épaisses, permettent de caractériser la longueur de transfert et de réaliser l’objectif de la thèse qui est de prendre en compte le béton tendu dans le calcul de la rigidité flexionnelle des poutres en béton armé / The aim of this thesis is to study the stiffening effect of tensile concrete in flexural reinforced concrete beams by controlling the bond between steel and concrete. The process consists of experimentally determining the strain profile in both deep and thin flexural beams. We created a new finite element model for both deep and thin beams in accordance with experimental observations. The experimental tests we conducted on tension members highlight the existence of both transfer length and length with total bond loss characterizing the bond between tensile steel and concrete. The transfer length contributes to the production of a tension stiffening effect, thus limiting the stiffness decrease of the flexural beam. Measuring the beam deflection makes it possible to determine the transfer length, and consequently to achieve the main objective, namely the contribution of tensile concrete to beam stiffness

Στην παρούσα διατριβή διερευνήθηκε πειραματικά, η συμπεριφορά διατάξεων ενίσχυσης πλακοδοκών Οπλισμένου Σκυροδέματος, σε τέμνουσα, που αποτελούνται από το συνδυασμό τρίπλευρων μανδυών από Ινοπλισμένα Πολυμερή και αγκυρίων ινών. Πρόκειται για διατάξεις ενίσχυσης που λόγω της παρουσίας των αγκυρίων, καλούνται να υπερκεράσουν τις αδυναμίες της τεχνικής των «ανοικτών» μανδυών, δηλαδή της συνήθους τεχνικής ενίσχυσης πλακοδοκών σε τέμνουσα. Οι αδυναμίες της εν λόγω τεχνικής, οφείλονται στην ανεπαρκή αγκύρωση των άκρων του μανδύα. Η πειραματική διερεύνηση της συμπεριφοράς τέτοιων διατάξεων ενίσχυσης, έγινε με εφαρμογή τους σε τέσσερα δοκίμια πλακοδοκών Οπλισμένου Σκυροδέματος, και συγκρίθηκε με τη συμπεριφορά ενός δοκιμίου αναφοράς που δεν έφερε καμία διάταξη ενίσχυσης, καθώς και με τη συμπεριφορά ενός δοκιμίου ενισχυμένου μόνο με τρίπλευρο «ανοικτό» μανδύα ΙΟΠ, απουσία αγκυρίων. Όλα τα δοκίμια, υποβλήθηκαν σε μονοτονική φόρτιση μέσω συγκεντρωμένου φορτίου με φορά ώστε να προκαλείται θλίψη στο άνω πέλμα της δοκού, δηλαδή στην πλάκα, και οι συνθήκες στήριξης ήταν τέτοιες που να προσομοιώνουν αμφιέρειστη δοκό. Τα πρώτα Κεφάλαια της διατριβής αποτελούνται από την εισαγωγή, τη βιβλιογραφική ανασκόπηση και την περιγραφή της πειραματικής διαδικασίας. Στα επόμενα Κεφάλαια, πέραν της παρουσίασης των πειραματικών αποτελεσμάτων, επιχειρείται ο προσδιορισμός της αποδοτικότητας με παράλληλη προσέγγιση της συμπεριφοράς των διατάξεων ενίσχυσης με έμφαση στον τρόπο με τον οποίον τα αγκύρια την επηρεάζουν. Στο τελευταίο Κεφάλαιο παρουσιάζεται ο κεντρικός άξονας της διατριβής περιλαμβάνοντας και τη σύνοψη των συμπερασμάτων. / In the present thesis, the effectiveness of shear strengthening schemes for T-Shaped RC beams, consisting of FRP U-Jackets and FRP anchors, was experimentally investigated. For this purpose, six full-scale of T-Shaped RC beams were produced. One specimen served as reference (unstrengthened) beam, whereas the remaining five received FRP U-jackets; out of the latter FRP anchors were used in four beams in order to enhance the effectiveness of the strengthening schemes, whereas no anchoring system was applied to the fifth beam. All specimens were tested under monotonic loading causing compression to the wide part of the section. The first chapter of this dissertation discusses the necessity of strengthening beams in shear and introduce the objective of the study. A relatively extended literature overview about shear strengthening with externally bonded reinforcement is included in the second chapter. The purpose of the third chapter is to describe the way the specimens were designed and constructed. The strengthening procedure, the experimental setup and the materials’ properties are also included in this chapter. The test results are presented and discussed in the fourth chapter. In the fifth chapter calculations regarding the effectiveness of the strengthening schemes are presented, along with an attempt to understand their general behavior while emphasizing on the way the FRP anchors affect it. The final chapter includes the general conclusions of the present study.

O objetivo deste trabalho é apresentar um modelo para análise probabilística de vigas de concreto armado recuperadas à flexão, através da utilização conjunta do método de simulação de Monte Carlo e do método dos Elementos Finitos. Para uma análise da confiabilidade, foram projetadas vigas de concreto armado seguindo as recomendações da NRR 6118:2003. Após, foi considerado que as armaduras tracionadas de flexão sofreram reduções de 10%, 20% e 30% na sua área da seção transversal, sendo, então, feita uma análise da segurança estrutural remanescente. Para o projeto de recuperação das vigas danificadas, estão apresentados os procedimentos do Bulletin 14 da fédération internationale du béton (fib) para o dimensionamento de sistemas de reforço com material compósito colado externamente às estruturas. Assim, a confiabilidade destas vigas recuperadas pôde ser estimada e comparada à confiabilidade das vigas originais. Dentro do processo de simulação, a resposta em termos da carga de ruptura das vigas de concreto armado recuperadas foi obtida através de uma análise numérica não-linear utilizando um modelo de elementos finitos. Devido à importância do deslizamento entre o substrato de concreto e o sistema de reforço estrutural, foram implementados dentro do modelo de elementos finitos, elementos unidimensionais de interface, quadráticos com seis pontos nodais para a captura deste comportamento. As análises probabilísticas através do método de simulação de Monte Carlo usando um modelo de elementos finitos mostraram que se atinge bom nível de segurança no projeto de vigas de concreto armado seguindo as recomendações da NBR 6118:2003. Entretanto, danos nas armaduras tracionadas de flexão podem fazer com que a confiabilidade caía para níveis inaceitáveis. Como solução, o projeto de recuperação estrutural de acordo com o Bulletin 14 da fib fez com que a confiabilidade das vigas recuperadas fosse no mínimo igual à confiabilidade das vigas originais. Assim, a probabilidade de falha das vigas recuperadas foi menor quando comparado à probabilidade de falha das vigas originais. / The objective of this work is to present a model for probabilistic analysis of RC beams rehabilitated for flexure, through the simultaneous utilization of the Monte Carlo method and the finite element method. Initially, three RC beams were designed following the NBR 6118:2003 recommendations. In the next step, it was considered that a reduction of the steel reinforcement area of 10%, 20% and 30% had occurred. Then the procedures for strengthening systems design with externally bonded carbon fiber reinforced polymers (fib, Bulletin 14) were employed for the rehabilitation of the damaged RC beams. Finally, the reliability of the RC beams rehabilitated for flexure could be evaluated and compared to that of the damaged and undamaged RC beams. In the Monte Carlo method, the RC beams failure load was obtained numerically through a nonlinear finite element model. Due to the importance of the slip between the concrete substrate and the external reinforcement, it was introduced in the finite element model a special one-dimensional interface element, with six nodal points and quadratic shape functions. The probabilistic analysis through the Monte Carlo method using a finite element model showed the high reliability attained in the RC beams design following the NBR 6118:2003 recommendations. However, the damage in the internal steel reinforcement can lead to an unacceptable reliability level. As a solution, the design of structural recovery according to the Bulletin 14 produced a reliability level for the RC beams rehabilitated for flexure at least equal to the reliability level of the undamaged RC beams. Consequently, the failure probability of the RC beams rehabilitated for flexure was lower then the failure probability of the undamaged RC beams.

O objetivo deste trabalho é apresentar um modelo para análise probabilística de vigas de concreto armado recuperadas à flexão, através da utilização conjunta do método de simulação de Monte Carlo e do método dos Elementos Finitos. Para uma análise da confiabilidade, foram projetadas vigas de concreto armado seguindo as recomendações da NRR 6118:2003. Após, foi considerado que as armaduras tracionadas de flexão sofreram reduções de 10%, 20% e 30% na sua área da seção transversal, sendo, então, feita uma análise da segurança estrutural remanescente. Para o projeto de recuperação das vigas danificadas, estão apresentados os procedimentos do Bulletin 14 da fédération internationale du béton (fib) para o dimensionamento de sistemas de reforço com material compósito colado externamente às estruturas. Assim, a confiabilidade destas vigas recuperadas pôde ser estimada e comparada à confiabilidade das vigas originais. Dentro do processo de simulação, a resposta em termos da carga de ruptura das vigas de concreto armado recuperadas foi obtida através de uma análise numérica não-linear utilizando um modelo de elementos finitos. Devido à importância do deslizamento entre o substrato de concreto e o sistema de reforço estrutural, foram implementados dentro do modelo de elementos finitos, elementos unidimensionais de interface, quadráticos com seis pontos nodais para a captura deste comportamento. As análises probabilísticas através do método de simulação de Monte Carlo usando um modelo de elementos finitos mostraram que se atinge bom nível de segurança no projeto de vigas de concreto armado seguindo as recomendações da NBR 6118:2003. Entretanto, danos nas armaduras tracionadas de flexão podem fazer com que a confiabilidade caía para níveis inaceitáveis. Como solução, o projeto de recuperação estrutural de acordo com o Bulletin 14 da fib fez com que a confiabilidade das vigas recuperadas fosse no mínimo igual à confiabilidade das vigas originais. Assim, a probabilidade de falha das vigas recuperadas foi menor quando comparado à probabilidade de falha das vigas originais. / The objective of this work is to present a model for probabilistic analysis of RC beams rehabilitated for flexure, through the simultaneous utilization of the Monte Carlo method and the finite element method. Initially, three RC beams were designed following the NBR 6118:2003 recommendations. In the next step, it was considered that a reduction of the steel reinforcement area of 10%, 20% and 30% had occurred. Then the procedures for strengthening systems design with externally bonded carbon fiber reinforced polymers (fib, Bulletin 14) were employed for the rehabilitation of the damaged RC beams. Finally, the reliability of the RC beams rehabilitated for flexure could be evaluated and compared to that of the damaged and undamaged RC beams. In the Monte Carlo method, the RC beams failure load was obtained numerically through a nonlinear finite element model. Due to the importance of the slip between the concrete substrate and the external reinforcement, it was introduced in the finite element model a special one-dimensional interface element, with six nodal points and quadratic shape functions. The probabilistic analysis through the Monte Carlo method using a finite element model showed the high reliability attained in the RC beams design following the NBR 6118:2003 recommendations. However, the damage in the internal steel reinforcement can lead to an unacceptable reliability level. As a solution, the design of structural recovery according to the Bulletin 14 produced a reliability level for the RC beams rehabilitated for flexure at least equal to the reliability level of the undamaged RC beams. Consequently, the failure probability of the RC beams rehabilitated for flexure was lower then the failure probability of the undamaged RC beams.

O objetivo deste trabalho é apresentar um modelo para análise probabilística de vigas de concreto armado recuperadas à flexão, através da utilização conjunta do método de simulação de Monte Carlo e do método dos Elementos Finitos. Para uma análise da confiabilidade, foram projetadas vigas de concreto armado seguindo as recomendações da NRR 6118:2003. Após, foi considerado que as armaduras tracionadas de flexão sofreram reduções de 10%, 20% e 30% na sua área da seção transversal, sendo, então, feita uma análise da segurança estrutural remanescente. Para o projeto de recuperação das vigas danificadas, estão apresentados os procedimentos do Bulletin 14 da fédération internationale du béton (fib) para o dimensionamento de sistemas de reforço com material compósito colado externamente às estruturas. Assim, a confiabilidade destas vigas recuperadas pôde ser estimada e comparada à confiabilidade das vigas originais. Dentro do processo de simulação, a resposta em termos da carga de ruptura das vigas de concreto armado recuperadas foi obtida através de uma análise numérica não-linear utilizando um modelo de elementos finitos. Devido à importância do deslizamento entre o substrato de concreto e o sistema de reforço estrutural, foram implementados dentro do modelo de elementos finitos, elementos unidimensionais de interface, quadráticos com seis pontos nodais para a captura deste comportamento. As análises probabilísticas através do método de simulação de Monte Carlo usando um modelo de elementos finitos mostraram que se atinge bom nível de segurança no projeto de vigas de concreto armado seguindo as recomendações da NBR 6118:2003. Entretanto, danos nas armaduras tracionadas de flexão podem fazer com que a confiabilidade caía para níveis inaceitáveis. Como solução, o projeto de recuperação estrutural de acordo com o Bulletin 14 da fib fez com que a confiabilidade das vigas recuperadas fosse no mínimo igual à confiabilidade das vigas originais. Assim, a probabilidade de falha das vigas recuperadas foi menor quando comparado à probabilidade de falha das vigas originais. / The objective of this work is to present a model for probabilistic analysis of RC beams rehabilitated for flexure, through the simultaneous utilization of the Monte Carlo method and the finite element method. Initially, three RC beams were designed following the NBR 6118:2003 recommendations. In the next step, it was considered that a reduction of the steel reinforcement area of 10%, 20% and 30% had occurred. Then the procedures for strengthening systems design with externally bonded carbon fiber reinforced polymers (fib, Bulletin 14) were employed for the rehabilitation of the damaged RC beams. Finally, the reliability of the RC beams rehabilitated for flexure could be evaluated and compared to that of the damaged and undamaged RC beams. In the Monte Carlo method, the RC beams failure load was obtained numerically through a nonlinear finite element model. Due to the importance of the slip between the concrete substrate and the external reinforcement, it was introduced in the finite element model a special one-dimensional interface element, with six nodal points and quadratic shape functions. The probabilistic analysis through the Monte Carlo method using a finite element model showed the high reliability attained in the RC beams design following the NBR 6118:2003 recommendations. However, the damage in the internal steel reinforcement can lead to an unacceptable reliability level. As a solution, the design of structural recovery according to the Bulletin 14 produced a reliability level for the RC beams rehabilitated for flexure at least equal to the reliability level of the undamaged RC beams. Consequently, the failure probability of the RC beams rehabilitated for flexure was lower then the failure probability of the undamaged RC beams.