Este trabalho tem como objetivo investigar o comportamento pós-pico de pilares com concretos de alta resistência confinados com e sem adição de fibras metálicas, submetidos à flexo-compressão. Para tanto, realizou-se inicialmente uma extensa revisão bibliográfica sobre o tema, na qual, elencaram-se os principais fatores que influenciam o comportamento pós-pico desses elementos estruturais. Em seguida, desenvolveram-se estudos paramétricos com o intuito de estabelecer os modos de influência de cada fator. Com base nesses estudos preliminares, um programa experimental, dividido em duas fases, foi proposto. Na primeira fase, ensaiaram-se vinte e seis pilares de concreto armado e 14 pilares de concreto simples. Três fatores foram estudados: a taxa de adição de fibras metálicas, a taxa de armadura transversal e a resistência à compressão do concreto. Os pilares possuíam seções transversais quadradas com dimensões 15 cm x 15 cm e altura de 50 cm e foram ensaiados à compressão centrada com controle de deslocamento. Observou-se que os três fatores analisados influenciam diretamente a ductilidade desses elementos estruturais e que, por meio de adição de fibras metálicas, é possível garantir índices de ductilidade aceitáveis para os pilares com concreto de alta resistência. Outrossim, analisando-se os diagramas força vs. deformação dos pilares, observou-se que para o primeiro pico de força, toda a seção resiste a esforços de compressão e que a adição de fibras melhora o trabalho conjunto entre o cobrimento e o núcleo do pilar. Finalizando esta fase, propuseram-se modificações para o modelo para concreto confinado de Cusson e Paultre (1995), de modo que, permitisse ao mesmo modelar o comportamento do concreto confinado com e sem adição de fibras metálicas. Na segunda fase do programa experimental, quinze pilares de concreto armado foram ensaiados a compressão excêntrica. Essa fase teve como objetivo investigar a influência da excentricidade, da taxa de armadura transversal e da taxa de adição de fibras metálicas no comportamento pós-pico desses elementos estruturais. Os pilares, nessa fase, possuíam seção transversal quadrada com dimensões de 15 cm x 15 cm e altura de 170 cm. Para realização dos ensaios foi confeccionado um par de rótulas unidirecionais, as quais apresentaram excelente desempenho, conseguindo transferir integralmente o momento externo aplicado ao pilar. Observou-se que, tanto a flambagem das barras da armadura longitudinal, quanto a perda de massa de concreto nesses pilares ocorrem de modo bem mais crítico que nos pilares submetidos à compressão centrada. Constatou-se, ainda, que quando os valores dos três fatores analisados são elevados, a ductilidade desses elementos estruturais é melhorada. Finalmente, foi observado que o efeito da flexão faz com que as tensões de confinamento se distribuam de modo diferenciado dentro da seção transversal dos pilares; contudo, observou-se que a tensão de confinamento na região comprimida da seção transversal pouco é modificada. / This thesis aims to investigate the ductility of high strength concrete columns, with confinement and/or with steel fibre addition, under combined axial load and bending. For this, an extensive survey was performed, and the main factors, which influence the column ductility, were pointed out. Then, parametric studies were done, aiming to establish the mode of influence of each factor. Based on these preliminaries studies, an experimental investigation, divided in two steps, was proposed. In the first part, twenty-six reinforced concrete columns and fourteen concrete columns were tested and three factors were analysed: the ratio of fibre addition, the transversal reinforcement ratio and the concrete compressive strength. The columns have 15 cm x 15 cm square cross section and were 50 cm high. They were tested under concentric load with displacement control. It was observed that all factors interfere in the columns\' ductility, and that, by adding fibre in the concrete, is possible to provide acceptable ductility index to high strength concrete columns. The load vs. column strain diagrams were studied and it was verified that for the first load peak, all the column cross section resists the axial load. Besides that, it was observed that the fibre addition improves the work together between the cover and the column core. Finally, several modifications to the concrete confinement model of Cusson and Paultre (1995) were proposed to allow this model to analyse the compressive behaviour of the confined concrete with or/not fibre addition. In the second part of the experimental program, fifteen reinforced concrete columns were tested under eccentric load. The effects of the load eccentricity, the ratio of steel fibre addition and the transversal reinforcement ratio on the column ductility were investigated. The columns have 15 cm x 15 cm square cross section and were 170 cm high. A pair of unidirectional hinges was design, which presented exceptional performance, transferring integrally the external applied moment to the columns. A most critical model of longitudinal reinforcement buckling and mass lost was observed. In addition, it was verified that when the values of all factors were increased, the column ductility was improved. Finally, a different distribution mode of the confinement pressure on the column cross section was observed, when combined axial load and bending is applied. Nevertheless, the confinement pressure on the compressive region of the column cross section is not modified.

Este trabalho tem como objetivo investigar o comportamento pós-pico de pilares com concretos de alta resistência confinados com e sem adição de fibras metálicas, submetidos à flexo-compressão. Para tanto, realizou-se inicialmente uma extensa revisão bibliográfica sobre o tema, na qual, elencaram-se os principais fatores que influenciam o comportamento pós-pico desses elementos estruturais. Em seguida, desenvolveram-se estudos paramétricos com o intuito de estabelecer os modos de influência de cada fator. Com base nesses estudos preliminares, um programa experimental, dividido em duas fases, foi proposto. Na primeira fase, ensaiaram-se vinte e seis pilares de concreto armado e 14 pilares de concreto simples. Três fatores foram estudados: a taxa de adição de fibras metálicas, a taxa de armadura transversal e a resistência à compressão do concreto. Os pilares possuíam seções transversais quadradas com dimensões 15 cm x 15 cm e altura de 50 cm e foram ensaiados à compressão centrada com controle de deslocamento. Observou-se que os três fatores analisados influenciam diretamente a ductilidade desses elementos estruturais e que, por meio de adição de fibras metálicas, é possível garantir índices de ductilidade aceitáveis para os pilares com concreto de alta resistência. Outrossim, analisando-se os diagramas força vs. deformação dos pilares, observou-se que para o primeiro pico de força, toda a seção resiste a esforços de compressão e que a adição de fibras melhora o trabalho conjunto entre o cobrimento e o núcleo do pilar. Finalizando esta fase, propuseram-se modificações para o modelo para concreto confinado de Cusson e Paultre (1995), de modo que, permitisse ao mesmo modelar o comportamento do concreto confinado com e sem adição de fibras metálicas. Na segunda fase do programa experimental, quinze pilares de concreto armado foram ensaiados a compressão excêntrica. Essa fase teve como objetivo investigar a influência da excentricidade, da taxa de armadura transversal e da taxa de adição de fibras metálicas no comportamento pós-pico desses elementos estruturais. Os pilares, nessa fase, possuíam seção transversal quadrada com dimensões de 15 cm x 15 cm e altura de 170 cm. Para realização dos ensaios foi confeccionado um par de rótulas unidirecionais, as quais apresentaram excelente desempenho, conseguindo transferir integralmente o momento externo aplicado ao pilar. Observou-se que, tanto a flambagem das barras da armadura longitudinal, quanto a perda de massa de concreto nesses pilares ocorrem de modo bem mais crítico que nos pilares submetidos à compressão centrada. Constatou-se, ainda, que quando os valores dos três fatores analisados são elevados, a ductilidade desses elementos estruturais é melhorada. Finalmente, foi observado que o efeito da flexão faz com que as tensões de confinamento se distribuam de modo diferenciado dentro da seção transversal dos pilares; contudo, observou-se que a tensão de confinamento na região comprimida da seção transversal pouco é modificada. / This thesis aims to investigate the ductility of high strength concrete columns, with confinement and/or with steel fibre addition, under combined axial load and bending. For this, an extensive survey was performed, and the main factors, which influence the column ductility, were pointed out. Then, parametric studies were done, aiming to establish the mode of influence of each factor. Based on these preliminaries studies, an experimental investigation, divided in two steps, was proposed. In the first part, twenty-six reinforced concrete columns and fourteen concrete columns were tested and three factors were analysed: the ratio of fibre addition, the transversal reinforcement ratio and the concrete compressive strength. The columns have 15 cm x 15 cm square cross section and were 50 cm high. They were tested under concentric load with displacement control. It was observed that all factors interfere in the columns\' ductility, and that, by adding fibre in the concrete, is possible to provide acceptable ductility index to high strength concrete columns. The load vs. column strain diagrams were studied and it was verified that for the first load peak, all the column cross section resists the axial load. Besides that, it was observed that the fibre addition improves the work together between the cover and the column core. Finally, several modifications to the concrete confinement model of Cusson and Paultre (1995) were proposed to allow this model to analyse the compressive behaviour of the confined concrete with or/not fibre addition. In the second part of the experimental program, fifteen reinforced concrete columns were tested under eccentric load. The effects of the load eccentricity, the ratio of steel fibre addition and the transversal reinforcement ratio on the column ductility were investigated. The columns have 15 cm x 15 cm square cross section and were 170 cm high. A pair of unidirectional hinges was design, which presented exceptional performance, transferring integrally the external applied moment to the columns. A most critical model of longitudinal reinforcement buckling and mass lost was observed. In addition, it was verified that when the values of all factors were increased, the column ductility was improved. Finally, a different distribution mode of the confinement pressure on the column cross section was observed, when combined axial load and bending is applied. Nevertheless, the confinement pressure on the compressive region of the column cross section is not modified.

There are approximately 4000 railway bridges in Sweden managed by the Swedish Administration of Transport (Trafikverket), of which a common construction type is the reinforced concrete (RC) trough bridge, which is a structure that consists of a slab carried by two longitudinal main beams which transfer loads towards the supports. A substantial amount of the RC trough bridge population is approaching the end of their service lives which consequently implies that the replacement of some of these bridges can be expected in the near future. Extending their service lives can yield positive effects from a financial- as well as an environmental perspective, and therefore it is highly beneficial to evaluate their capacities as realistically as possible. One factor that may help improve accuracy during the determination of their capacities is an evaluation of how it is affected by the location of the railway track on the bridge.  In current design codes defined by Trafikverket, consideration is taken to horizontal track displacement for a minimum displacement of 0.1 m if there doesn’t exist data suggesting that a larger displacement is prevalent on the bridge. However, Trafikverket has received data which suggest that a considerable number of bridges could experience load eccentricities which exceed the standard minimum value. This raises the question whether or not 0.1m is the most optimal limit value for load eccentricity on railway bridges. For RC trough bridges, a larger load eccentricity may result in one main beam carrying a larger portion of the load which will decrease the axle load which the bridge can carry. It is therefore important to evaluate the influence of larger horizontal displacements than what is currently is considered as a preventive action.   In addition, several studies on Swedish concrete bridges constructed during the 20th century have pointed to a significant increase in concrete compressive- and tensile strength over time. This suggests that it is possible that a considerable amount of RC trough bridges have a higher capacity than what was originally intended, and further research is required in order to understand the behaviour of these bridges when key material parameters are altered.        There are three main tasks which this master thesis seeks to complete. The first part is a detailed analysis of a database named BaTMan (Bridge and Tunnel Management) that belongs to Trafikverket. In this analysis parameters such as span length, age, material type and damages for every identified railway bridge is extracted and further processed in Microsoft Excel in order to gain a clear overview of the RC trough bridge population. The second task regards the development of a non-linear finite element model of a typical RC trough bridge named Lautajokki. The model is analysed using ATENA Science and its behaviour is verified against test results obtained during a full-scale test of the bridge performed by Paulsson et al. (1996). The last task is to use the devolved model to perform a parametric study where the effects of changes in load eccentricity, compressive strength as well as tensile strength is studied.