This study aimed at defining useful guidelines for the design of an armored truck cab. Particularly, the quality and performance of ballistic steel were defined, considering not too high demanding requirements. Two steel grades, corresponding to hardness of 500 HB and 600 HB, and three different suppliers were selected. After dividing the truck cab into main areas, a FEM simulation was performed. The investigated model considered a specific standardized threat to impact against a double layer steel protection. Due to the high strain rates involved, the formation of adiabatic shear bands has been observed. However, the main purpose of the model was to find an ultimate thickness protection layout of armored steel plates. Thereby, the thicknesses to be attributed to the various components of the main areas could be estimated. Two alternatives were therefore hypothesized: the first considering the critical thickness case of penetration, and the second, an oversized version, supposed to be more resistant to penetration. It was thus possible to estimate the weight of the truck cab for the two above mentioned alternatives, and verify such hypothesized weight complied with the assigned requirement. A possible cutting operation was then considered in order to optimize the use of the steel plates. An estimation of the pre-series production costs of was finally derived. / Denna studie gjorde det möjligt att definiera användbara indikatorer och följa en möjlig multitasking-strategi för utformningen av en bepansrad lastbilshytt. Fallstudien syftade till att definiera kvalitet och prestanda av stål, vilket är ett av de vanligaste materialen för skydd, främst i de fall där kraven inte är så stora. Två stålkvaliteter motsvarande hårdhetsvärdena 500 HB och 600 HB valdes från var och en av tre materialleverantörer. Efter att ha delat upp truckhytten i några huvudområden startades en FEM-simulering med programvara från Impetus Afea. Modellen behandlade ett specifikt hot som var större än det önskade kravet på ett dubbelskiktsskydd. På grund av de inblandade höga deformationshastigheterna har bildningen av adiabatiska skjuvband observerats. Huvudsyftet med modellen var emellertid att hitta en ultimat skyddslayout för tjockleken, så att de tjocklekar som skulle tillskrivas de olika komponenterna i huvudområdena kunde härledas. Två alternativ antogs därför, det ena med tanke på den ultimata tjockleken och den andra med tanke på en skyddslayout som antas vara mer motståndskraftig mot penetrering. Det var således möjligt att uppskatta hyttens vikt för de två alternativen och kontrollera att den uppfyllde de fastställda kraven. Därefter övervägdes en möjlig skäroperation för att optimera användandet av de tillgängliga stålplattorna.

As hard rock mining progresses into higher stress mining conditions through either late stage extraction or mining at depth, the rock mass is driven not just to the peak strength, but often well into the post-peak until complete ‘failure’ occurs and easier mining conditions become evident. Limited research has been accomplished in identifying the transition of the rock mass and its behaviour into the post-peak and this research investigates this behaviour in detail. As the rock mass progressively fails, fractures are initiated through intact rock and extension and shear failure of these and pre-existing features occurs. Associated with this failure are microseismic events, which can be used to give an indication of the strength state of the rock mass. Based on an analogy to laboratory testing of intact rock and measurement of acoustic emissions, the microseismicity can be used to identify, fracture initiation, coalescence of fractures (yield), localization (peak-strength), accumulation of damage (post-peak) and ultimate failure (residual strength) leading to aseismic behaviour. The case studies presented in this thesis provide an opportunity to examine and analyse rock mass failure into the post-peak, through the regional and confined failures at the Williams and the Golden Giant mines, both in the Hemlo camp in Northern Ontario, Canada. At the Williams mine, the progressive failure of a sill pillar region into the post-peak was analysed; relating the seismic event density, combined with numerical modelling and a spatial and temporal examination of the principal components analysis (PCA), to characterize the extent, trend and state of the yielding zone, which formed a macrofracture shear structure. Observations of conventional displacement instrumentation, indicates regional dilation or shear of the rock mass occurs at or prior to the point of ‘disassociation’ (breakdown of stable PCA trends) when approaching the residual strength. At the Golden Giant mine, the complete process from initiation to aseismic behaviour is monitored in a highly stressed and confined pendent pillar. The PCA technique, numerical modelling and focal mechanism studies are used to define significant stages of the failure process, in which a similar macrofracture structure was formed. Temporal observations of key source parameters show significant changes prior to and at the point of coalescence and localization.

Le comportement à long terme des excavations souterraines est un enjeu social et économique majeur, en particulier dans le contexte du stockage en formation géologique profonde de déchets nucléaires à Haute Activité et Vie Longue (HAVL). Plusieurs galeries expérimentales ont été creusées dans le laboratoire de recherche souterrain de Meuse/Haute-Marne situé près de Bure en France, où des études sont menées pour comprendre le comportement global de la roche constitutive : l’argilite du Callovo-Oxfordien (COx).L’objectif de cette thèse est d’effectuer une modélisation avec Code_Aster d’ouvrages souterrains, et en particulier d’une des galeries du laboratoire de Meuse/Haute-Marne, en prenant en compte une approche non locale, l’effet du fluage et le couplage hydromécanique dans le cadre de la mécanique des milieux poreux, et de comparer les résultats numériques avec les données expérimentales disponibles.Le modèle élastoviscoplastique spécifiquement utilisé pour cette étude est le modèle L&K : il offre d’une part un couplage entre le comportement instantané et différé, et prend en compte d’autre part la dilatance, paramètre qui gouverne les déformations volumiques du matériau lors d’une sollicitation, et ses fortes variations, caractéristique essentielle des géomatériaux et en particulier du COx. La présence d’un fluide s’écoulant à travers le matériau va ajouter une composante hydraulique à la modélisation, dont le couplage avec la mécanique est assuré par les équations de Biot. Une autre nouveauté de ce travail concerne le couplage entre ce comportement rhéologique complexe et une approche non locale dans un cadre industriel. Parmi les méthodes de régularisation disponibles dans Code_Aster, la méthode second gradient de dilatation a été choisie parce que bien adaptée aux géomatériaux. Son but est de corriger la dépendance au maillage et les solutions numériques localisées.Une fois les outils numériques opérationnels et les paramètres du modèle L&K calés sur des essais effectués en laboratoire sur des échantillons, un bon accord général a été trouvé entre les résultats numériques et les mesures in situ, sans aucun recalage des paramètres. Les effets du temps observés expérimentalement sur l’évolution des déplacements et des pressions d’eau sont retrouvés au sein d’une même modélisation, validant ainsi la démarche prédictive suivie / The long term behaviour of underground excavations is a social and economic stake, in particular in the context of storage in deep geological formation of high activity and long life nuclear waste. Several experimental galleries have been dug in the underground research laboratory (URL) of Meuse/Haute-Marne located close to Bure in France, where studies are leaded in order to understand the global behaviour of the constitutive rock which is the Callovo-Oxfordian (COx) argillite.The purpose of this PhD Thesis is to establish a modelling with Code_Aster of underground excavations, and especially of a Meuse/Haute-Marne laboratory gallery, taking into account non local approach, creep effect and hydromechanical coupling in the framework of the mechanics of porous media, and then to compare numerical results with available experimental data.The specific elastoviscoplastic model used in this study is the L&K model: it offers a coupling between instantaneous and delayed behaviour, and it takes into account the dilation, parameter which governs the volume strains of the material during a solicitation, and its strong variation, a specificity of geomaterials and so of COx argillite. The fluid flowing through the material adds a hydraulic component to the modelling, which is coupled to mechanic component thanks to Biot’s equations.Another novelty of this work concerns the coupling between such complex rheological behaviour and a non local approach in an industrial way. Among methods of regularization available in Code_Aster, the second gradient of dilation is well fitted to geomaterials. Its aim is to correct mesh dependency and numerical localized solutions.After describing numeric tools and setting parameters of the L&K model on laboratory tests, a good general agreement was found between numeric results and in situ measures, without resetting parameters. Time effects experimentally measured on displacement and pore pressure evolution are observed in the same modelling, validating the followed predictive approach

En struktur analys på Ericssons MINILINK-6352 har utförts för att undersöka spänningar och deformationer på enheten, främst med fokus på de termiska gränskiktsmaterialen och buktningar av kretskortet. Dessa är viktiga aspekter när man överväger om enheten är termiska lämpad ur en mekanisk synvinkel, där god ytkontakt mellan de olika kropparna är avgörande för ordentlig kylning genom värmeledning. Analysen kräver tillräcklig materialdata till gränskiktsmaterialen och kretskortet för att kunna skapa lämpliga matematiska modeller. Enaxliga kompressionstester har genomförts för att karakterisera de hyperelastiska och viskoelastiska lagar för fyllda silikongummimaterial som används som termiska gränskiktsmaterial, som ibland kallas för gappad. Böjning av ett kretskort simulerades och jämfördes med ett tre--punkts böjtest för att verifiera om befintlig materialdata i beräkningsprogrammen var tillräcklig, jämförelsen visade god överensstämmelse. Kretskortet med dess komponenter, som modellerades som styva block, med gappads ovanpå som komprimeras av en platta simulerades och ett svagt område hittades. Detta område var sedan tidigare känt och har i ett senare skede eliminerats genom att tillsätta ytterligare en stödpelare. Därav visar denna studie en metod för att hitta intressanta regioner tidigt i konstruktionsfasen som lätt kan ändras för att uppfylla nödvändiga krav och undvika brister i konstruktionen. Arbetet har visat sig användbart genom att hitta detta svaga område i exempel produkten, arbetet ger även tillräckligt med information och exempeldata för att ytterligare utreda liknande produkter. Kombinationen av erfarenhet och simulering möjliggör smartare designval. / A structural analysis on Ericssons MINILINK-6352 has been performed in order to investigate stresses and deformations of the unit, mainly focusing on the thermal interface materials and warpage of the printed circuit boards. These are important aspects when considering if the unit is thermally adequate from a mechanical point of view, where good surface contact between various bodies are critical for proper cooling through heat conductivity. The analysis requires sufficient materal data for the interface material and the circuit board in order to create suitable mathematical models. Uniaxial compression tests have been conducted to characterise the hyperelastic and viscoelastic constitutive laws of a filled silicone rubber material used as a thermal interface material, commonly referred to as a thermal pad. Bending of a printed circuit board was simulated and compared to a three-point bend test on the circuit board in order to verify material data already available in the computational software, which showed good agreement. The entire radio unit was mechanically analysed during its sealing process. The circuit board with attached components modelled as stiff blocks with thermal pads on top compressed by plates was simulated and a weak area was found. This area in question was already known and has in a later stage been eliminated by adding an additional supporting pillar. Hence this study shows a methodology to find regions of interest at an early design phase which can easily be altered to fulfil necessary requirements and eliminate design flaws. This work has proven useful in finding weak regions in the example product, it also provides enough information and example data to further investigate similar products. The combination of experience and simulation allows for smarter design choices.

Dans un contexte d’amélioration des dispositifs pour la réduction de bruit, l’étude sur le transfert d’énergie irréversible en utilisant des résonateurs purement acoustiques à comportement non linéaire a été réalisée. Les résonateurs acoustiques classiques en régime linéaire agissent comme un Amortisseur de Masse Accordée (TMD, en anglais) et ils sont efficaces pour une gamme de fréquence très étroite. Cependant, lorsqu’ils sont soumis à des excitations très fortes (régime non-linéaire) ils peuvent devenir efficaces pour une plus large gamme de fréquences si des termes non linéaires peuvent être activés. Dans un premier temps, une étude sur ce comportement non-linéaire d’un résonateur d’Helmholtz modifié a été réalisée expérimentalement. Ensuite, l’équation dynamique gouvernante de tels résonateurs ont été développées en prenant en compte les non-linéarités de la force de rappel et d’amortissement. Une approximation de la solution analytique de l’équation gouvernante du résonateur acoustique a été déterminée en utilisant les méthodes des échelles multiples du temps et de transformation du temps non régulière. Dans un deuxième temps, une étude du couplage entre un mode acoustique en basses fréquences et un résonateur (celui étudié précédemment) à comportement non-linéaire a été réalisée. Pour ce faire, des mesures expérimentales avec un montage du système couplé ont permis de vérifier l’atténuation acoustique produite par le résonateur en régime forcé et libre. Une modélisation analytique du couplage a permis d’identifier l’expression de la variété invariante lente, ce qui a permis d’étudier les possibles points d’équilibre et points singuliers du système. Les modèles analytiques développés ont également été vérifiés par des simulations numériques. / Nowadays, there is a need of new types of technologies for sound reduction because of the growing of different industries. In this context, we have studied the targeted energy transfer using a purely acoustic resonator. These acoustic resonators act, in the linear regime, as a Tuned Masse Damper (TMD) and they are efficient for a narrow frequency band. But, when they are excited with high forces, in the nonlinear regime, they are efficient for a wider frequency band if the nonlinear terms are activated. First, an experimental study about the nonlinear behavior of a modified Helmholtz Resonator was done. Then, the governing equation of such resonators were developed considering the nonlinearities in the restitution force and damping. An approximation of the analytical solution of the governing equation of the acoustical resonator is derived using the multiples scales of time method and the non-smooth time transformation method. In a second part, a study about the coupling between an acoustic mode in low frequencies and a resonator (the one studied in the previous part) with a nonlinear behavior is done. In order to do this, experimental measurements of the coupled system to confirm acoustic attenuation by the resonator in forced and free regime were done. Then, an analytical modelling of the coupled system allowed to derive the expression of the Slow Invariant Manifold (SIM), in order to identify the possible equilibrium points and singular points of the system. Derived analytical models were verified by numerical simulations.

Le Canal de Corse est un bassin confiné asymétrique localisé entre l’Île de Corse et l’Archipel de la Toscane, dont le flanc ouest est dominé par des processus turbiditiques et hémipélagiques et le flanc est par des mouvements en masse et des processus contouritiques. Le présent projet de doctorat a pour objectif de comprendre plus précisément les mécanismes contrôlant la formation des glissements sous-marins dans les contourites vaseuses (dépôts sédimentaires formés par les courants) pendant la période Plio-Quaternaire. Le vaste jeu de données disponible pour ce projet de doctorat inclut : la bathymétrie multifaisceaux, la sismique réflexion, les mesures géotechniques in situ, les mesures de vitesse de courant et les résultats d’un modèle hydrodynamique.Les contourites du Canal de Corse sont principalement composées de vase avec la présence de couches de sable formées par de forts courants de fond pendant les périodes de baisse du niveau marin. La croissance des dépôts contouritiques dépend de la disponibilité de sédiment fourni par le système turbiditique. Ainsi, cette croissance est lente pendant les périodes interglaciaires de haut niveau marin et rapide pendant les bas niveaux marins. Les courants contrôlent la morphologie du fond et génèrent les plastered drifts de forme convexe avec des pentes plus raides dans la partie avale, limités par une incision créée par les courants (moat). Le Pianosa Slump a été initié dans cette partie basse du plastered drift. Les moats pourraient être érodés préférentiellement pendant les périodes froides passées déclenchant ainsi certains glissements observés. Un autre facteur prédisposant l’instabilité de pente sur la Ride de Pianosa est la faiblesse d’une couche dont le comportement mécanique se caractérise par du radoucissement (perte de résistance avec le cisaillement). Cette propriété particulière est due à la présence de zéolites (produit de l’altération des roches volcaniques). Cette couche a formé la surface basale de rupture du Pianosa Slump. En conclusion, les deux principaux facteurs prédisposant la formation de glissements sous-marins sur la Ride de Pianosa sont : la morphologie du plastered drift avec une pente plus raide en aval et la couche faible composée de sédiment vaseux riche en zéolites. Le principal facteur déclenchant semble être l’érosion basale. / The Corsica Trough is an asymmetric confined basin located between the Corsica Island and the Tuscan Ar-chipelago, with the western flank dominated by turbiditic and hemipelagic processes and the eastern flank by mass transport and contouritic processes. The present PhD project aims to develop our understanding of the mechanisms that control the formation of submarine landslides within muddy contourites (sediment deposits related to bottom currents) during the Plio-Quaternary. The broad data set available for this PhD project includes: multibeam bathymetry, seismic reflection data, sediment cores, in situ geotechnical measurements, current ADCP measurements and results of a hydrodynamic model.The contourites of the Corsica Trough are mainly composed of mud with sandy layers formed by enhanced bottom currents during periods of sea level fall. The contourite drifts grow slowly during sea level high-stands and rapidly during sea level low-stands due to the high sediment availability provided by an active turbidite sys¬tem. Bottom currents control the seafloor morphology and generate plastered drifts on the slope. This is a con¬vex-shaped contourite with steep slope gradients in the lower part limited by a moat (incision created by bottom currents). The Pianosa Slump was initiated in this lower part of the plastered drift. The occurrence of continuous erosive processes during cold periods could undercut the slope and trigger submarine landslides. Another predis¬posing factor for slope instability identified is the presence of a potential weak layer with a post-peak strain soften¬ing behaviour (strength loss with increasing strain). This particular property is caused by the presence of zeolites (product of the alteration of volcanic rocks). This layer originated the basal failure surface of the Pianosa Slump.In summary, the two main factors predispose the formation of submarine landslides in the Pianosa Ridge are: the morphology of the plastered drift with steep slopes in the lower part and a potential weak layer composed of zeolitic muddy sediment. The main triggering factor seems to be undercutting by bottom currents.

大部的類神經網路均為解決特定問題而設計，並非真正去模擬人腦的功能 ，在本論文中介紹一個模擬人類學習方式的類神經網路，稱為推理類神經 網路（The Reasoning Neural Network），其主要兩個組成為強記（ cram -ming）及推理（reasoning）部份，透過彈性的組合這兩個部份可 使類神經網路具有類似人類的學習程序。在本論文中介紹其中一個學習程 序並用四個實驗來評估推理類神經網路的績效，從實結果得知，推理類神 經網路能以合理的隱藏節點數（hidden nodes）達到學習的目標，並建立 一個網路內部表示方式（internal representation），及具有好的推理 能力（g eneralization ability）。 / Most of artification Neural Networks are designed to resolve spe -cific problems, rather than to model the brain. The Reasoning N -eural Network (RNN) that imitates the way of human learning is presented here. Two key components of RNN are the cramming and t -he reasoning. These components coulds be arranged flexibly to a -chieve the human-like learning procedure. One edition of the RNN used in experiments is introduces, and four different proble -ms are used to evaluate the RNN's performance. From simulation results, the RNN accomplishes the goal of learning with a reason -able number of hidden nodes, and evolves a good internal repres -entation and a generalization ability.

Cost-effective machining of hardened steel (>60 HRC) components such as a large wind turbine bearing poses a significant challenge. This thesis investigates a new laser tempering based hybrid turning approach to machine hardened AISI 52100 steel parts more efficiently and cost effectively. The approach consists of a two step process involving laser tempering of the hardened workpiece surface followed by conventional machining at higher material removal rates using lower cost ceramic tooling to efficiently cut the laser tempered material. The specific objectives of this work are to: (a) study the characteristics of laser tempering of hyper-eutectoid 52100 hardened steel, (b) model the laser tempering process to determine the resulting hardness, and (c) conduct machining experiments to evaluate the performance of the laser tempering based hybrid turning process in terms of forces, tools wear and surface finish. First, the microstructure alterations and phase content in the surface and subsurface layers are analyzed using metallography and x-ray diffraction (XRD) respectively. Laser tempering produces distinct regions consisting of - a tempered white layer and a dark layer- in the heat affected subsurface region of the workpiece. The depth of the tempered region is dependent on the laser scanning conditions. Larger overlap of laser scans and smaller scan speeds produce a thicker tempered region. Furthermore, the tempered region is composed of ferrite and martensite and weak traces of retained austenite (~ 1 %). Second, a laser tempering model consisting of a three dimensional analytical model to predict the temperature field generated by laser scanning of 52100 hardened steel and a phase change based hardness model to predict the hardness of the tempered region are developed. The thermal model is used to evaluate the temperature field induced in the subsurface region due to the thermal cycles produced by the laser scanning step. The computed temperature histories are then fed to the phase change model to predict the surface and subsurface hardness. The laser tempering model is used to select the laser scanning conditions that yield the desired hardness reduction at the maximum depth. This model is verified through laser scanning experiments wherein the hardness changes are compared with model predictions. The model is shown to yield predictions that are within 20 % of the measured hardness of the tempered region. Using the laser scanning parameters determined from the laser tempering model, cutting experiments using Cubic Boron Nitride (CBN) tools and low cost alumina ceramic tools are conducted to compare the performance of laser tempering based hybrid turning with the conventional hard turning process. The machining experiments demonstrate the possibility of higher material removal rates, lower cutting forces, improved tool wear behavior, and consequently improved tool life in the laser tempering based process. In addition, the laser tempered based hybrid turning process produce is shown to yield lower peak-to-valley surface roughness height than the conventional hard turning process. Furthermore, it is found that lower cost ceramic tools can be used in place of CBN tools without compromising the material removal rate.

As hard rock mining progresses into higher stress mining conditions through either late stage extraction or mining at depth, the rock mass is driven not just to the peak strength, but often well into the post-peak until complete ‘failure’ occurs and easier mining conditions become evident. Limited research has been accomplished in identifying the transition of the rock mass and its behaviour into the post-peak and this research investigates this behaviour in detail. As the rock mass progressively fails, fractures are initiated through intact rock and extension and shear failure of these and pre-existing features occurs. Associated with this failure are microseismic events, which can be used to give an indication of the strength state of the rock mass. Based on an analogy to laboratory testing of intact rock and measurement of acoustic emissions, the microseismicity can be used to identify, fracture initiation, coalescence of fractures (yield), localization (peak-strength), accumulation of damage (post-peak) and ultimate failure (residual strength) leading to aseismic behaviour. The case studies presented in this thesis provide an opportunity to examine and analyse rock mass failure into the post-peak, through the regional and confined failures at the Williams and the Golden Giant mines, both in the Hemlo camp in Northern Ontario, Canada. At the Williams mine, the progressive failure of a sill pillar region into the post-peak was analysed; relating the seismic event density, combined with numerical modelling and a spatial and temporal examination of the principal components analysis (PCA), to characterize the extent, trend and state of the yielding zone, which formed a macrofracture shear structure. Observations of conventional displacement instrumentation, indicates regional dilation or shear of the rock mass occurs at or prior to the point of ‘disassociation’ (breakdown of stable PCA trends) when approaching the residual strength. At the Golden Giant mine, the complete process from initiation to aseismic behaviour is monitored in a highly stressed and confined pendent pillar. The PCA technique, numerical modelling and focal mechanism studies are used to define significant stages of the failure process, in which a similar macrofracture structure was formed. Temporal observations of key source parameters show significant changes prior to and at the point of coalescence and localization.

L’état de contrainte des galeries et tunnels souterrains évolue dans le temps, tout d’abord en raison de l’avancement du front de taille lors de la construction, mais aussi en raison des changements graduels des propriétés du terrain ou des caractéristiques mécaniques des matériaux constitutifs de l’ouvrage (réduction de la cohésion, vieillissement). Ces changements sont causés par des actions combinées comme: les processus d’altération dus aux infiltrations des eaux météoriques et de réseaux concessionnaires, mais aussi par l’apparition des fissures qui induisent par ailleurs des redistributions de contraintes et de déformations, ainsi que par le fluage des matériaux. Dans cette étude, un travail préliminaire, nous permet de quantifier l’état de contrainte initial lors de la construction des ouvrages. Notre approche basée sur la modélisation éléments finis 2D permet de simuler l’état d’équilibre interne des excavations, nos résultats sont en adéquation avec ceux issus des méthodes analytiques de la littérature. Cet état initial est ensuite calibré via une modélisation 3D. Puis, une macro-modélisation à base d’éléments finis nous permet d’envisager différents scénarios de détérioration des soutènements, notamment à l’extrados des galeries de couloirs d’accès et tunnels de ligne de métro Parisien. Un des scénarios étudié prend en considération la réduction de la résistance mécanique du piédroit et de la zone interface entre le sol et la paroi, ainsi que le fluage de l’encaissant et du revêtement. En utilisant une loi qui permet de simuler la perte de cohésion, paramètre qui évolue en fonction de la déformation inélastique et en fonction du temps, le comportement mécanique des zones dégradées est ainsi reproduit par un modèle viscoplastique de type adoucissant ou à écrouissage négatif. Nous mettons bien en exergue les processus de dégradation observés sur ces galeries durant leur exploitation. Les différentes modélisations permettent de mieux évaluer l’état des différents composants de l’ouvrage selon le type de dégradation et selon le temps traduisant ainsi son vieillissement. Elles permettent également de jauger la sensibilité et la pertinence de différents paramètres. In fine, nous évaluons les pathologies de ces structures, ce qui permet de fournir des indicateurs quant à la gestion de ce patrimoine. / The stress state of underground tunnels and galleries evolves over time , firstly because of the advancement of the working face during construction , then due to gradual changes in soil properties and mechanical properties of materials component of the infrastructures (reduction of cohesion, aging). These changes are caused by combined actions such as: weathering processes due to infiltration of meteoric waters and concessionaires networks, but also by the cracks appearance that induce further redistribution of stress and strain, and the creep. A particular attention is paid to the determination of the initial state of stresses applied on the structure right after the construction. In order to approach this initial stress state which corresponds to the construction phase of underground structures, we propose a methodology based on comparing the numerical model with analytical methods taken from literature. Then we compare 2D and 3D numerical models. Thereafter, a macro- modeling based on the finite elements method allows considering different scenarios of the tunnel-linings deterioration, specifically at the extrados of the galleries and Paris metro line tunnels. One of the studied cases takes into account the mechanical strength loss of the abutment and the interface between the ground and the wall area including their creeps. Therefore, Elasto-viscoplastic with strain softening constitutive model is used in this study to reproduce the degraded zone behavior. The numerical models show that we can reproduce the underground structure behavior during degradation and assess the state of different structure components. The approach used here for the Parisian metro tunnels may be used by the other underground structures managers.