Several studies were undertaken by various investigators during the last five decades to better understand the engineering behaviour of unsaturated soils. These studies are justified as more than 33% of soils worldwide are found in either arid or semi-arid regions with evaporation losses exceeding water infiltration. Due to this reason, the natural ground water table in these regions is typically at a greater depth and the soil above it is in a state of unsaturated conditions. Foundations of structures such as the housing subdivisions, multi-storey buildings, bridges, retaining walls, silos, and other infrastructure constructed in these regions in sandy soils are usually built within the unsaturated zone (i.e., vadose zone). Limited studies are reported in the literature to understand the influence of capillary stresses (i.e., matric suction) on the bearing capacity, settlement and liquefaction potential of unsaturated sands. The influence of matric suction in the unsaturated zone of the sandy soils is ignored while estimating or evaluating bearing capacity, settlement and liquefaction resistance in conventional engineering practice. The focus of the research presented in the thesis has been directed towards better understanding of these aspects and providing rational and yet simple tools for the design of shallow foundations (i.e., footings) in sands under both static and dynamic loading conditions. Terzaghi (1943) or Meyerhof (1951) equations for bearing capacity and Schmertmann et al. (1978) equation for settlement are routinely used by practicing engineers for sandy soils based on saturated soil properties. The assumption of saturated conditions leads to conservative estimates for bearing capacity; however, neglecting the influence of capillary stresses contributes to unreliable estimates of settlement or differential settlement of footings in unsaturated sands. There are no studies reported in the literature on how capillary stresses influence liquefaction, bearing capacity and settlement behavior in earthquake prone regions under dynamic loading conditions. An extensive experimental program has been undertaken to study these parameters using several specially designed and constructed equipment at the University of Ottawa. The influence of matric suction, confinement and dilation on the bearing capacity of model footings in unsaturated sand was determined using the University of Ottawa Bearing Capacity Equipment (UOBCE-2011). Several series of plate load tests (PLTs) were carried out on a sandy soil both under saturated and unsaturated conditions. Based on these studies, a semi-empirical equation has been proposed for estimating the variation of bearing capacity with respect to matric suction. The saturated shear strength parameters and the soil water characteristic curve (SWCC) are required for using the proposed equation. This equation is consistent with the bearing capacity equation originally proposed by Terzaghi (1943) and later extended by Meyerhof (1951) for saturated soils. Chapter 2 provides the details of these studies. The cone penetration test (CPT) is conventionally used for estimating the bearing capacity of foundations because it is simple and quick, while providing continuous records with depth. In this research program, a cone penetrometer was specially designed to investigate the influence of matric suction on the cone resistance in a controlled laboratory environment. Several series of CPTs were conducted in sand under both saturated and unsaturated conditions. Simple correlations were proposed from CPTs data to relate the bearing capacity of shallow foundations to cone resistance in saturated and unsaturated sands. The details of these studies are presented and summarized in Chapter 3. Standard penetration tests (SPTs) and PLTs were conducted in-situ sand deposit at Carp region in Ottawa under both saturated and unsaturated conditions. The test results from the SPTs and PLTs at Carp were used along with other data from the literature for developing correlations for estimating the bearing capacity of both saturated and unsaturated sands. The proposed SPT-CPT-based technique is simple and reliable for estimation of the bearing capacity of footings in sands. Chapter 4 summarizes the details of these investigations. Empirical relationships were proposed using the CPTs data to estimate the modulus of elasticity of sands for settlement estimation of footings in both saturated and unsaturated sands. This was achieved by modifying the Schmertmann et al. (1978) equation, which is conventionally used for settlement estimations in practice. Comparisons are provided between the three CPT-based methods that are commonly used for settlement estimations in practice and the proposed method for seven large scale footings in sandy soils. The results of the comparisons show that the proposed method provides better estimations for both saturated and unsaturated sands. Chapter 5 summarizes the details of these studies. A Flexible Laminar Shear Box (FLSB of 800-mm3 in size) was specially designed and constructed to simulate and better understand the behaviour of model surface footing under seismic loads taking account of the influence of matric suction in an unsaturated sandy soil. The main purpose of using the FLSB is to simulate realistic in-situ soils behaviour during earthquake ground shaking. The FLSB test setup with model footing was placed on unidirectional 1-g shake table (aluminum platform of 1000-mm2 in size) during testing. The resistance of unsaturated sand to deformations and liquefaction under seismic loads was investigated. The results of the study show that matric suction offers significant resistance to liquefaction and settlement of footings in sand. Details of the equipment setup, test procedure and results of this study are presented in Chapter 6. Simple techniques are provided in this thesis for estimating the bearing capacity and settlement behaviour of sandy soils taking account of the influence of capillary stresses (i.e., matric suction). These techniques are consistent with the methods used in conventional geotechnical engineering practice. The studies show that even low values of capillary stresses (i.e., 0 to 5 kPa) increases the bearing capacity by two to four folds, and the settlement of footings not only decreases significantly but also offers resistance to liquefaction in sands. These studies are promising and encouraging to use ground improvement techniques; such as capillary barrier techniques to maintain capillary stresses within the zone of influence below shallow foundations. Such techniques, not only contribute to the increase of bearing capacity, they reduce settlement and alleviate problems associated with earthquake effects in sandy soils.

Several studies were undertaken by various investigators during the last five decades to better understand the engineering behaviour of unsaturated soils. These studies are justified as more than 33% of soils worldwide are found in either arid or semi-arid regions with evaporation losses exceeding water infiltration. Due to this reason, the natural ground water table in these regions is typically at a greater depth and the soil above it is in a state of unsaturated conditions. Foundations of structures such as the housing subdivisions, multi-storey buildings, bridges, retaining walls, silos, and other infrastructure constructed in these regions in sandy soils are usually built within the unsaturated zone (i.e., vadose zone). Limited studies are reported in the literature to understand the influence of capillary stresses (i.e., matric suction) on the bearing capacity, settlement and liquefaction potential of unsaturated sands. The influence of matric suction in the unsaturated zone of the sandy soils is ignored while estimating or evaluating bearing capacity, settlement and liquefaction resistance in conventional engineering practice. The focus of the research presented in the thesis has been directed towards better understanding of these aspects and providing rational and yet simple tools for the design of shallow foundations (i.e., footings) in sands under both static and dynamic loading conditions. Terzaghi (1943) or Meyerhof (1951) equations for bearing capacity and Schmertmann et al. (1978) equation for settlement are routinely used by practicing engineers for sandy soils based on saturated soil properties. The assumption of saturated conditions leads to conservative estimates for bearing capacity; however, neglecting the influence of capillary stresses contributes to unreliable estimates of settlement or differential settlement of footings in unsaturated sands. There are no studies reported in the literature on how capillary stresses influence liquefaction, bearing capacity and settlement behavior in earthquake prone regions under dynamic loading conditions. An extensive experimental program has been undertaken to study these parameters using several specially designed and constructed equipment at the University of Ottawa. The influence of matric suction, confinement and dilation on the bearing capacity of model footings in unsaturated sand was determined using the University of Ottawa Bearing Capacity Equipment (UOBCE-2011). Several series of plate load tests (PLTs) were carried out on a sandy soil both under saturated and unsaturated conditions. Based on these studies, a semi-empirical equation has been proposed for estimating the variation of bearing capacity with respect to matric suction. The saturated shear strength parameters and the soil water characteristic curve (SWCC) are required for using the proposed equation. This equation is consistent with the bearing capacity equation originally proposed by Terzaghi (1943) and later extended by Meyerhof (1951) for saturated soils. Chapter 2 provides the details of these studies. The cone penetration test (CPT) is conventionally used for estimating the bearing capacity of foundations because it is simple and quick, while providing continuous records with depth. In this research program, a cone penetrometer was specially designed to investigate the influence of matric suction on the cone resistance in a controlled laboratory environment. Several series of CPTs were conducted in sand under both saturated and unsaturated conditions. Simple correlations were proposed from CPTs data to relate the bearing capacity of shallow foundations to cone resistance in saturated and unsaturated sands. The details of these studies are presented and summarized in Chapter 3. Standard penetration tests (SPTs) and PLTs were conducted in-situ sand deposit at Carp region in Ottawa under both saturated and unsaturated conditions. The test results from the SPTs and PLTs at Carp were used along with other data from the literature for developing correlations for estimating the bearing capacity of both saturated and unsaturated sands. The proposed SPT-CPT-based technique is simple and reliable for estimation of the bearing capacity of footings in sands. Chapter 4 summarizes the details of these investigations. Empirical relationships were proposed using the CPTs data to estimate the modulus of elasticity of sands for settlement estimation of footings in both saturated and unsaturated sands. This was achieved by modifying the Schmertmann et al. (1978) equation, which is conventionally used for settlement estimations in practice. Comparisons are provided between the three CPT-based methods that are commonly used for settlement estimations in practice and the proposed method for seven large scale footings in sandy soils. The results of the comparisons show that the proposed method provides better estimations for both saturated and unsaturated sands. Chapter 5 summarizes the details of these studies. A Flexible Laminar Shear Box (FLSB of 800-mm3 in size) was specially designed and constructed to simulate and better understand the behaviour of model surface footing under seismic loads taking account of the influence of matric suction in an unsaturated sandy soil. The main purpose of using the FLSB is to simulate realistic in-situ soils behaviour during earthquake ground shaking. The FLSB test setup with model footing was placed on unidirectional 1-g shake table (aluminum platform of 1000-mm2 in size) during testing. The resistance of unsaturated sand to deformations and liquefaction under seismic loads was investigated. The results of the study show that matric suction offers significant resistance to liquefaction and settlement of footings in sand. Details of the equipment setup, test procedure and results of this study are presented in Chapter 6. Simple techniques are provided in this thesis for estimating the bearing capacity and settlement behaviour of sandy soils taking account of the influence of capillary stresses (i.e., matric suction). These techniques are consistent with the methods used in conventional geotechnical engineering practice. The studies show that even low values of capillary stresses (i.e., 0 to 5 kPa) increases the bearing capacity by two to four folds, and the settlement of footings not only decreases significantly but also offers resistance to liquefaction in sands. These studies are promising and encouraging to use ground improvement techniques; such as capillary barrier techniques to maintain capillary stresses within the zone of influence below shallow foundations. Such techniques, not only contribute to the increase of bearing capacity, they reduce settlement and alleviate problems associated with earthquake effects in sandy soils.

Performance of buildings and bridges during past earthquakes has indicated that many of these structures are vulnerable to seismic damage and structural collapse. The deficiencies in pre-1970s design codes have resulted in poor performance of reinforced concrete structures during seismic excitations. The Richter Magnitude 6.6 - 1971 San Fernando Earthquake raised awareness for seismic retrofit needs of existing buildings for the first time. The majority of deficiencies of vulnerable concrete columns can be overcome through seismic retrofits that involve additional transverse reinforcement. This can be done either by providing reinforced concrete, steel, or fibre-reinforced polymer (FRP) jackets around existing columns; or by applying transverse prestressing to columns (RetroBelt System). The research project presented in this thesis involves a seismic retrofit methodology for seismically deficient building and bridge columns, utilizing the use of high-strength packaging straps as external reinforcement for transverse prestressing. The emphasis in the project is placed on experimental research. Three seismically deficient full-size reinforced concrete columns, with a circular, a square and a rectangular cross- section, either critical in shear or flexure, were designed, built and tested under simulated seismic loading. The results indicate that external prestressing of columns in transverse direction with high-strength steel straps improves ductility and energy dissipation capacity of seismically deficient columns. They further indicate that current analytical techniques can be used to predict the force-displacement relationships of columns. A design approach is presented for the retrofit methodology investigated.

Modélisations simplifiées pour l’analyse du risque sismique de bâtiments en béton armé. Résumé de la thèse en français (1800 signes max.) : La thèse s’inscrit dans le cadre du projet MARS (Méthodes Avancées pour le Risque Sismique, EDF R&D). Elle concerne plus particulièrement certaines tâches sur le développement des méthodes simplifiées et robustes de calcul pour permettre la simulation intensive et table de la réponse sismique de bâtiments en béton armé. En effet, |’analyse de risque nécessite un très grand nombre de calculs pour tenir compte des incertitudes, tant sur le chargement (aléa sismique) que sur le comportement non linéaire des structures. Dans la première partie de ce travail, nous effectuerons une étude bibliographique sur les modèles de résolution sismique pour les bâtiments en béton armé. Cette étape va nous permettre de rassembler le maximum d’éléments nécessaires permettant de comprendre et d’identifier tous les paramètres, les avantages, les inconvénients et la limite d’utilisation de chaque procédure de calcul numérique par éléments finis. Dans la deuxième partie, on développe un macro-élément de poteau-poutre, associé â un modèle de comportement non linéaire afin de traduire la réponse de la structure sous les sollicitations sismiques. Des hypothèses cinématiques ont été adoptées pour limiter le nombre de degrés de liberté. La loi de comportement globale en cisaillement est décrite dans le cadre delà plasticité. Nous avons choisi un modèle à écrouissage cinématique pour prendre en compte la dissipation due à la fissuration. Les paramètres sont identifiés à partir de résultats expérimentaux ou bien pré-calculés par des analyses â une échelle locale (calculs 3D par éléments finis ou calcul simplifié type « Modified Compression Field Theory >>). Des analyses numériques ont été réalisées afin de valider le modèle proposé comparant à des essais expérimentaux disponibles dans la littérature. / This PhD is part of the MARS project (Advanced Methods for Seismic Risk, EDF R&D). It relates particularity to the development of simplified and robust calculation. The overall aim is to significantly reduce the intensive computation time without loosing a reliable simulation of the seismic response of reinforced concrete buildings methods. Seismic risk analysis requires a very large number of repeated calculations to account for uncertainties of both the loading (seismichazard) and the nonlinear behaviour of structures. ln the first part of this work, a bibliographic study on seismic resolution models for reinforced concrete buildings is provided. This step allows collecting the maximum of necessary elements to understand and identify all the parameters, advantages, disadvantages and limits of use of each finite element calculation method. In the second part, a macro—elements for beam—column joint associated to a nonlinear behavior to reflect the response to the structure under seismic loads ls developed. Kinematic assumptions have been adopted to limit the number of degrees of freedom. The law of global shear behavior is described in the context of plasticity. A model with kinematic hardening is chosen to account for the dissipation due to cracking. Model parameters are identified from experimental results or pre-calculated by analysis on a local scale vla 3D finite element calculation or the implied "Modihed Compression Field Theory Numerical analyses were performed to validate the proposed approach against experimental tests available in the literature.

Reinforced concrete frames, infilled with brick/concrete block masonry, are the most common type of structures found in multi-storeyed constructions, especially in developing countries. Usually, the infill walls are considered as non-structural elements even though they alter the lateral stiffness and strength of the frame significantly. Approximately 80% of the structural cost from earthquakes is attributable to damage of infill walls and to consequent damages of doors, windows and other installations. Despite the broad application and economical significance, the infill walls are not included in the analysis because of the design complexity and lack of suitable theory. But in seismic areas, ignoring the infill-frame interaction is not safe because the change in the stiffness and the consequent change in seismic demand of the composite structural system is not negligible. The relevant experimental findings shows a considerable reduction in the response of infilled frames under reverse cyclic loading. This behaviour is caused by the rapid degradation of stiffness, strength, and low energy dissipation capacity resulting from the brittle and sudden damage of the unreinforced masonry infill walls. Though various national/international codes of practice have incorporated some of the research outcomes as design guidelines, there is a need and scope for further refinement. In the initial part of this work, a numerical modelling and linear elastic analysis of masonry infilled RC frames has been done. A multi-storey multi-bay frame infilled with masonry panels, is considered for the study. Both macro modelling and micro modelling strategies are adopted. Seismic loading is considered and an equivalent static analysis as suggested in IS 1893, 2002 is done. The results show that the stiffness of the composite structure is increased due to the obvious confinement effects of infill panels on the bounding frame. A parametric study is conducted to investigate the influence of size and location of openings, presence/absence of infill panels in a particular storey and elevation irregularity in terms of floor height. The results show that the interaction of infill panel changes the seismic response of the composite structure significantly. Presence of openings further changes the seismic behaviour. Increase in openings increases the natural period and introduce newer failure mechanisms. Absence of infill in a particular storey (an elevation irregularity) makes it drift more compared to adjacent storeys. Since the structural irregularities influence the seismic behaviour of a building considerably, we should be cautious while construction and renovation of such buildings in order to take the advantage of increased strength and stiffness obtained by the presence of infill walls. A nonlinear dynamic analysis of masonry infilled RC frames is presented next. Material non linearity is considered for the finite element modelling of both masonry and concrete. Concrete damage plasticity model is employed to capture the degradation in stiffness under reverse cyclic loading. A parametric study by varying the same parameters as considered in the linear analysis is conducted. It is seen that the fundamental period calculation of infilled frames by conventional empirical formulae needs to be revisited for a better understanding of the real seismic behaviour of the infilled frames. Enhancement in the lateral stiffness due to the presence of infill panel attracts larger force and causes damage to the composite system during seismic loading. Elevation irregularities included absence of infill panels in a particular storey. Soft storey shows a tendency for the adjacent columns to fail in shear, due to the large drift compared to other storeys. The interstorey drift ratios of soft storeys are found to be larger than the limiting values. However this model could not capture the separation at the interfaces and related failure mechanisms. To improve the nonlinear model, a contact surface at the interface is considered for a qualitative analysis. A one bay one storey infilled frame is selected. The material characteristics were kept the same as those used in the nonlinear model. Contact surface at the interface was given hard contact property with pressure-overclosure relations and suitable values of friction at the interface. This model could simulate the compressive diagonal strut formation and the switching of this compressive strut to the opposite diagonal under reverse cyclic loading. It showed an indication of corner crushing and diagonal cracking failure modes. The frame with central opening showed stress accumulation near the corners of opening. Next, the micro modelling strategy for masonry suggested by Lourenco is studied. This interface element can be used at the masonry panel-concrete frame interface as well as at the expanded masonry block to block interface. Cap plasticity model (modified Drucker – Prager model for geological materials) can be used to describe the behaviour of masonry (in terms of interface cracking, slipping, shearing) under earthquake loading. The blocks can be defined as elastic material with a potential crack at the centre. However, further experimental investigation is needed to calibrate this model. It is required to make use of the beneficial effects and improve upon the ill-effects of the presence of infills. To conclude, infill panels are inevitable for functional aspects such as division of space and envelope for the building. Using the lateral stiffness, strength contribution and energy dissipation capacity, use of infill panels is proposed to be a wiser solution for reducing the seismic vulnerability of multi-storey buildings.

Performance of buildings and bridges during past earthquakes has indicated that many of these structures are vulnerable to seismic damage and structural collapse. The deficiencies in pre-1970s design codes have resulted in poor performance of reinforced concrete structures during seismic excitations. The Richter Magnitude 6.6 - 1971 San Fernando Earthquake raised awareness for seismic retrofit needs of existing buildings for the first time. The majority of deficiencies of vulnerable concrete columns can be overcome through seismic retrofits that involve additional transverse reinforcement. This can be done either by providing reinforced concrete, steel, or fibre-reinforced polymer (FRP) jackets around existing columns; or by applying transverse prestressing to columns (RetroBelt System). The research project presented in this thesis involves a seismic retrofit methodology for seismically deficient building and bridge columns, utilizing the use of high-strength packaging straps as external reinforcement for transverse prestressing. The emphasis in the project is placed on experimental research. Three seismically deficient full-size reinforced concrete columns, with a circular, a square and a rectangular cross- section, either critical in shear or flexure, were designed, built and tested under simulated seismic loading. The results indicate that external prestressing of columns in transverse direction with high-strength steel straps improves ductility and energy dissipation capacity of seismically deficient columns. They further indicate that current analytical techniques can be used to predict the force-displacement relationships of columns. A design approach is presented for the retrofit methodology investigated.

La liquéfaction des sols saturés lors des séismes est l’un des problèmes les plus importants auxquels sont confrontés les ingénieurs. Il n’y a guère eu de séisme majeur sans au moins quelques cas de liquéfaction. Des tassements, des basculements de bâtiments, des écoulements latéraux, des cônes de liquéfaction et des instabilités de pentes, sont certaines de ses manifestations. La conception sismique des centrales nucléaires et autres installations critiques comprend systématiquement une évaluation du risque de liquéfaction.Dans ce cadre, des méthodes de nature entièrement empirique sont couramment utilisées en ingénierie. Ces approches procurent des marges à la conception et des limitations d’utilisation. Pour exploiter ces marges en situation de réévaluation, on a recours à des calculs transitoires non linéaires avancés dans lesquels on doit modéliser finement la loi de comportement du sol pour mettre en évidence les montées de pression interstitielle.Ces derniers calculs sont coûteux en termes de temps et de compétences numériques. L’objectif de ce travail de recherche est notamment de réduire les conservatismes en vigueur lors de l’utilisation de la méthode simplifiée sans pour autant mettre en œuvre d’emblée les méthodes les plus sophistiqués. On propose pour cela une méthode de complexité intermédiaire qui élargit l’applicabilité des modèles semi-empiriques pour une analyse plus fine du risque sismique.Dans un premier temps, en partant d’un calcul linéaire équivalent conventionnel, une nouvelle approche pour la prise en compte de la montée de pression interstitielle est proposée sous le nom « X-ELM ». Le modèle de comportement employé est basé sur la relation entre la pression interstitielle et les déformations volumiques plastiques. La nouvelle approche «X-ELM » est utilisée pour modéliser la réponse des sols pour le séisme de Tōhoku (Mw=9.0) à la ville d’Urayasu au Japon. Le modèle a été appliqué sur douze profils de sols différents. L’étude de ces cas rend possible la validation du modèle par comparaison des résultats des calculs aux observations in situ. Le modèle peut donc être considéré comme un outil fiable pour la prédiction de déclenchement de liquéfaction des sols saturés.Ensuite, un outil de prédiction rapide a été conçu en se basant sur des approximations de processus aléatoire, sur les propriétés mécaniques de base du sol et sur les caractéristiques du chargement sismique. Outre son ampleur, une caractéristique importante du signal sismique d’entrée est sa durée qui peut conduire à de fortes non linéarités et à un état de liquéfaction étendu. En considérant donc la durée de phase forte, le spectre de réponse, la fréquence propre du modèle et les caractéristiques de densification du sol, l’outil de prédiction proposé procure des estimations rapides du taux de montée de pression interstitielle et du tassement pré-liquéfaction sans devoir exécuter des calculs transitoires.Enfin, un modèle 2D de barrage est étudié, en examinant l’influence de la montée de pression interstitielle et celle des déformations de cisaillement sur la réponse sismique de l’ouvrage. Un calcul linéaire équivalent adapté aux situations bidimensionnelles est élaboré et le prédicteur est employé pour évaluer la montée de pression interstitielle. On compare les résultats de la simulation aux observations in situ, piézométriques et accélérométriques.En conclusion, ce travail de recherche fournit des méthodes et outils de calculs numériques performants et accessibles aux ingénieurs pour l’évaluation sismique des profils de sols et des ouvrages en terre tels que digues ou barrages / The liquefaction of saturated soils during earthquakes is one of the most important problems facing engineers. There has hardly been a major earthquake without at least some cases of liquefaction. Settlements, tilting of buildings, lateral flows, sand boilings and slope instabilities have been some of its manifestations. The seismic design of nuclear power plants and other critical facilities systematically includes a liquefaction risk assessment.In this context, fully empirical methods are commonly used in engineering. These approaches provide design margins and limitations of use. To exploit these margins in a re-evaluation situation, we use advanced nonlinear transient calculations in which the soil behavior must be finely modeled to highlight the pore-water pressure build-up. These last calculations are expensive in terms of time and numerical skills. The objective of this research work is to reduce the conservatisms in force when using the simplified method without necessarily implementing the most sophisticated methods from the outset. To this end, we propose a method of intermediate complexity that broadens the applicability of semi-empirical models for a more detailed analysis of seismic risk.First, starting from a conventional equivalent linear calculation, a new approach for taking into account excess pore pressure is proposed under the name "X-ELM". The behavioral model employed is based on the relationship between pore pressure and plastic volumetric deformations. The new "X-ELM" approach is used to model soil response in the city of Urayasu,Japan during the Tohoku earthquake (Mw = 9.0). The model has been applied to twelve different soil profiles. The study of these cases makes possible the validation of themodel by comparing the results of the calculations with the observations in situ. The model can therefore be considered as a reliable tool for the prediction of liquefaction triggering of saturated soils.Then, a prediction tool was designed based on random process approximations, the basic mechanical properties of the soil and the characteristics of the seismic loading. In addition to its magnitude, an important feature of the input seismic signal is its duration which can lead to strong nonlinearities and an extended liquefaction state. Considering the strong phase duration, the response spectrum, the natural frequency of the model and the characteristics of soil compaction, this tool provides fast estimations of the rate of pore pressure build-up and pre-liquefaction settlement without having to perform transient calculations.Finally, a 2D dam model is studied, by examining the influence of excess pore pressure and that of the shear strains on the seismic response of the structure. An equivalent linear computation adapted to two-dimensional situations is elaborated and the predictor is used to evaluate pore pressure increase. Simulation results are compared with in situ, piezometric and accelerometric observations.In conclusion, this research work provides methods and tools of numerical computation that are efficient and accessible to engineers for the seismic evaluation of soil profiles and earth structures such as dikes or dams

[pt] Enfoca-se a análise sísmica de usinas termonucleares sob metodologia no domínio da frequência utilizando-se como modelo, o do prédio do reator da usina Angra 3 e como principal ferramenta o programa de análise de problemas com interação solo-estrutura, SASSI. São avaliadas alternativas para três fases distintas do projeto: geração de sismos artificiais, análise do sistema principal e desenvolvimento de espectros para análise do sistema secundário. Na geração de sismos artificiais, a introdução de critérios de compatibilização complementares ao do espectro de resposta é avaliada considerando-se as consequências sobre as respostas em um ponto da estrutura. No sistema principal, sujeito a excitações sísmicas nas direções horizontal e vertical, comparam-se soluções no domínio do tempo e da freqüência e se reconhece a metodologia na frequência como uma alternativa conveniente à análise no tempo para cálculo de respostas em termos de acelerações , deslocamentos e tensões máximas e de espectros de acelerações. São ainda, desenvolvidos espectros de resposta acoplados para a análise de sistemas secundários segundo sistemática já proposta, para cálculo no domínio do tempo, que comparados com estes, alcançam amplitudes da mesma ordem, porém com tempo de processamento inferior. A metodologia de análise no domínio da frequência apresenta- se como um recurso mais preciso, pois permite a consideração de parâmetros do solo não constantes com a frequência; no que diz respeito ao programa utilizado, mostra-se flexível e acessível. / [en] The seismic analysis of nuclear power plant structures and components is considered to evaluate the convenience to update the brazilian design routine through the use of frequency domain techniques provided by codes such as SASSI. For this purpose one uses a FEM model of a reactor building which has been already analyzed by current techniques in time domain, and comparisons are made with results obtained in the frequency domain, with special attention to the following points: - analysis of the main structures using foundation impedance functions variable with the excitation frequency; - generation of coupled floor response spectra and their application to the analysis of secondary systems; - consideration of complementary criteria based on a minimum seismic spectral density function to generate artificial ground motion accelerograms compatible with a prescribed design response spectrum. As a first step, an exploratory work on the usage of SASSI program is made running a couple of simple examples. In the continuation of the work, a series of runs is made using the above-mentioned structural model under currently prescribed ground motions in the design of brazilian NPP structures. The results are compared with those from previous analyses using different methodologies and some interesting conclusions are brought up. / [es] En el presente trabajo se enfoca el análisis sísmico de plantas termonucleares bajo la metodología en el dominio de la frecuencia utilizando como modelo, el del edificio del reactor de la planta Angra 3 y como principal herramienta el programa de análisis de problemas con interacción suelo-extructura, SASI. Son evaluadas alternativas para tres fases distintas del proyecto: generación de sismos artificiales, análisis del sistema principal y desarrollo de espectros para análisis del sistema secundario. En la generación de sismos artificiales, se evalúa la introducción de criterios de compatibilización complementares al del espectro de respuesta. En el sistema principal, sujeto a excitaciones sísmicas en las direcciones horizontal y vertical, se compararon soluciones en el dominio del tiempo y de la frecuencia. Se reconoce la metodología en la frecuencia como una alternativa conveniente para cálculo de respuestas en términos de aceleraciones, deslocamientos y tensiones máximas y de espectros de aceleraciones. Se desarrollan, además, espectros de respuesta acoplados para el análisis de sistemas secundarios según sistemática ya propuesta, para cálculo en el dominio del tiempo, que comparados con éstes, alcanzan amplitudes de la misma orden, pero con tiempo de procesamiento inferior. La metodología de análisis en el dominio de la frecuencia se presenta como un recurso más preciso, pués permite la consideración de parámetros del suelo no constantes con la frequencia; con respecto al programa utilizado, éste se muestra flexible y accesible.

Reinforced concrete (RC) structures constitute a significant portion of the building inventory in earthquake-prone regions of the United States. Accurate analysis tools are necessary to allow the quantitative assessment of the performance and safety offered by RC structures. Currently available analytical approaches are not deemed adequate, because they either rely on overly simplified models or are restricted to monotonic loading. The present study is aimed to establish analytical tools for the accurate simulation of RC structures under earthquake loads. The tools are also applicable to the simulation of reinforced masonry (RM) structures. A new material model is formulated for concrete under multiaxial, cyclic loading conditions. An elastoplastic formulation, with a non-associative flow rule to capture compression-dominated response, is combined with a rotating smeared-crack model to capture the damage associated with tensile cracking. The proposed model resolves issues which characterize existing concrete material laws. Specifically, the newly proposed formulation accurately describes the crack opening/closing behavior and the effect of confinement on the strength and ductility under compressive stress states. The model formulation is validated with analyses both at the material level and at the component level. Parametric analyses on RC columns subjected to quasi-static cyclic loading are presented to demonstrate the need to regularize the softening laws due to the spurious mesh size effect and the importance of accounting for the increased ductility in confined concrete. The impact of the shape of the yield surface on the results is also investigated. Subsequently, a three-dimensional analysis framework, based on the explicit finite element method, is presented for the simulation of RC and RM components under cyclic static and dynamic loading. The triaxial constitutive model for concrete is combined with a material model for reinforcing steel which can account for the material hysteretic response and for rupture due to low-cycle fatigue. The reinforcing steel bars are represented with geometrically nonlinear beam elements to explicitly account for buckling of the reinforcement. The strain penetration effect is also accounted for in the models. The modeling scheme is validated with the results of experimental static and dynamic tests on RC columns and RC/RM walls. The analyses are supplemented with a sensitivity study and with calibration guidelines for the proposed modeling scheme. Given the computational cost and complexity of three-dimensional finite element models in the simulation of shear-dominated structures, the development of a conceptually simpler and computationally more efficient method is also pursued. Specifically, the nonlinear truss analogy is employed to capture the response of shear-dominated RC columns and RM walls subjected to cyclic loading. A step-by-step procedure to establish the truss geometry is described. The uniaxial material laws for the concrete and masonry are calibrated to account for the contribution of aggregate interlock resistance across inclined shear cracks. Validation analyses are presented, for quasi-static and dynamic tests on RC columns and RM walls. / Ph. D.

Use of soil retaining walls for roads, embankments and bridges is increasing with time and reinforced soil retaining walls are found to be very efficient even under critical conditions compared to unreinforced walls. They offer competitive solutions to earth retaining problems associated with less space and more loads posed by tremendous growth in infrastructure, in addition to the advantages in ease and cost of construction compared to conventional retaining wall systems. The study of seismic performance of reinforced soil retaining walls is receiving much attention in the light of lessons learned from past failures of conventional retaining walls. Laboratory model studies on these walls under controlled seismic loading conditions help to understand better how these walls actually behave during earthquakes. The objective of the present study is to investigate the seismic response of geosynthetic reinforced soil wall models through shaking table tests. To achieve this, wrap faced and rigid faced reinforced soil retaining walls of size 750 × 500 mm in plan and 600 mm height are built in rigid and flexible containers and tested under controlled dynamic conditions using a uni-axial shaking table. The effects of frequency and acceleration of the base motion, surcharge pressure on the crest, number of reinforcing layers, container boundary, wall structure and reinforcement layout on the seismic performance of the retaining walls are studied through systematic series of shaking table tests. Results are analyzed to understand the effect of each of the considered parameters on the face displacements, acceleration amplifications and soil pressures on facing at different elevations of the walls. A numerical model is developed to simulate the shaking table tests on wrap faced reinforced soil walls using a computer program FLAC (Fast Lagrangian Analysis of Continua). The experimental data are used to validate the numerical model and parametric studies are carried out on 6 m height full-scale wall using this model. Thus, the study deals with the shaking table tests, dynamic response of reinforced walls and their numerical simulation. The thesis presents detailed description of various features and various parts of the shaking table facility along with the instrumentation and model containers. Methodology adopted for the construction of reinforced soil model walls and testing procedures are briefly described. Scaling and stability issues related to the model wall size and reinforcement strength are also discussed. From the study, it is observed that the displacements are decreasing with the increase in relative density of backfill, increase in surcharge pressure and increase in number of reinforcing layers; In general, accelerations are amplified to the most at the top of the wall; Behaviour of model walls is sensitive to model container boundary. The frequency content is very important parameter affecting the model response. Further, it is noticed that the face displacements are significantly affected by all of the above parameters, while the accelerations are less sensitive to reinforcement parameters. Even very low strength geonet and geotextile are able to reduce the displacements by 75% compared to unreinforced wall. The strain levels in the reinforcing elements are observed to be very low, in the order of ±150 micro strains. A random dynamic event is also used in one of the model tests and the resulted accelerations and displacements are presented. Numerical parametric studies provided important insight into the behaviour of wrap faced walls under various seismic loading conditions and variation in physical parameters.