O processo de soldagem por atrito linear com mistura (FSW) é uma técnica recente para a soldagem no estado sólido de materiais, em particular para o alumínio e suas ligas. O processo foi inventado na Inglaterra em 1991. Neste processo, as partes a serem soldadas são fixadas e uma ferramenta especial realiza a soldagem de forma contínua. A ferramenta possui uma velocidade de rotação e, durante a sua translação, o material é misturado no estado sólido e, conseqüentemente, soldando as duas partes. A junta soldada por FSW de alumínio AA 6063-T6, com espessura de 3 mm foi caracterizada. A soldagem foi realizada com uma rotação da ferramenta de 710 rpm e com uma velocidade de translação de 5,3 mm/s. A ferramenta empregada é do tipo three flats, com diâmetro do ombro de 14 mm, diâmetro do pino de 3 mm e com ângulo de 90° com relação à horizontal. Os corpos de prova foram soldados em três condições: um passe, dois passes e dois passes com inversão de rotação do pino. Após a soldagem foram realizados os seguintes tratamentos térmicos: solubilização, envelhecimento e recozimento. A junta soldada foi caracterizada por macrografias, micrografias, microdureza, ensaios de calorimetria diferencial e EBSD. Os resultados mostraram que existem ZTMAs diferentes conforme a condição dos de passes. O número de passes tem influência nas componentes da textura alterando de Cubo para Latão e para Goss + Cobre. Os tratamentos térmicos de envelhecimento e recozimento produziram as maiores e menores durezas do cordão, respectivamente. Foi observada a sinergia entre os fatores número de passes e região do cordão no tamanho de grão do cordão. O lado de retrocesso, após o tratamento térmico, apresentou os grãos mais finos. / Friction stir welding (FSW) is a recent process for aluminium welding in solid state. This process was invented in England in 1991. The welding process is done with a special rotating tool that travels along the joint while the parts are fixed. The tool has a speed and a rotation and during its translation the material mixtures in solid state and the joint occurs. The objective of this paper is to show the metallurgical and mechanical characteristics of a 3 mm thick Aluminum AA 6063 T6 plate welded joint. The tool rotation speed was 710 rpm and the translation speed was 5.3 mm/s. The type of the tool used was three flats, with a shoulder diameter of 14mm and pin diameter of 3mm and perpendicular to the plate. The samples were welded in three conditions: one pass, two passes and two passes with pin rotation inversion in the second pass. The welded samples were also submitted to solution heat treatment, solution heat treatment followed by aging and annealing heat treatments. The welded joint was studied with these main experimental techniques: optical and scanning electron microscopy, microhardness, differential scanning calorimetry and electron backscatter diffraction for texture analysis. The results showed different TAZs according to the welding conditions. The number of passes has influence over the texture components changing from Cube to Brass and to Goss + Copper. The aging and solution heat treatments showed the highest and the lowest hardness, respectively. Synergy between the welding conditions and weld region was observed for the grain size results. The retreating side produced the finest grains after heat treating.

Esta tese tem por objetivo a aplicação do processamento por atrito linear na liga de titânio Ti-6Al-4V. Derivado da solda por atrito linear, é um processo recente desenvolvido na década de 90 para união de alumínio. Sua aplicação em outros tipos de materiais como aços e ligas de alto desempenho, em especial o titânio, tem interessado a industria. A metodologia utilizada nesta tese para avaliar o processamento por atrito linear, consistiu na execução de ensaios mecânicos de tração em condições mistas em chapas da liga de titânio Ti-6Al-4V. A máquina utilizada para o processamento das chapas foi um centro de usinagem CNC convencional, adaptado com dispositivos especiais. Além dos ensaios de tração em condições mistas, foram executadas medições de microdurezas nas regiões atingidas pelo processo, avaliação das microestruturas resultantes e medições de tensão residual para uma caracterização mais ampla do processo. As microestruturas na região processada são caracterizadas por uma estrutura totalmente transformada. As temperaturas de pico na região processada excederam a temperatura -transus durante o processamento e a transformação da fase + ocorreu durante a fase de resfriamento. A transformação da fase para resultou na formação de agulhas de fase nos contornos e pelo interior dos grãos da fase . Pequenas regiões com estrutura equiaxial de grãos ( globular) foram observados na zona de processamento. A abordagem dos resultados quantitativos foi feita de forma estatística, visando identificar os parâmetros de maior interação com os resultados observados. Foi identificado nesta tese que a rotação da ferramenta apresentou a maior influência nos resultados de tensão residual, microdureza e tensão de escoamento. Uma importante contribuição à modelagem da tensão de escoamento para materiais anisotrópicos é proposta, baseado em um critério de escoamento ortotrópico. Equações complementares baseadas nos testes mistos de tração e cisalhamento são propostas para modificar o modelo ortotrópico. O intuito deste modelo é indicar em que condições o material tem seu regime de escoamento atingido, podendo servir de base para simulações práticas de peças em condições similares. / This thesis aims at the application of friction stir processing (FSP) in Ti-6Al-4V titanium alloy. Derived from friction stir welding (FSW), it is a recent process developed in the 90s for aluminum joining. Its application to other types of materials such as steel and high performance alloys, in particular titanium, has interested industry. The methodology applied in this thesis to evaluate FSP, consisted in the execution of tensile test at mixed conditions to Ti-6Al-4V sheets 4. The machine used for processing the sheet was a conventional CNC milling machine, assembled with special fixture devices. In addition to tensile tests, measurements have been performed to the regions affected by the process such as evaluation of microhardness, microstructure and residual stress condition. The microstructures at the processed region are characterized by a transformed structure. The peak temperatures, in the processed region, exceeded the -transus temperature during the processing and transformation of the phase + occurred during the cooling phase. This transformation resulted in the formation of boundary and intergranular phase (Widmanstätten) at the grains. Small regions of equiaxed grain structure (globular ) were observed in the processed zone. The approach to the quantitative results was made in statistical form aiming to identify the parameters interaction with the observed results. It was identified in this thesis that the tool spinning rotation showed the highest influence on the results of residual stress, hardness and yield strength. An important contribution to the modeling of anisotropic materials yield stress is proposed based on an orthotropic yield criterion. Additional equations based on the mixed tests for tensile and shear are proposed to modify the orthotropic model. The purpose of this model is to indicate the conditions under which the material has reached its yield regime, and may be a basis for practical simulations in similar conditions.

Orientador: Janito Vaqueiro Ferreira / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-23T16:12:48Z (GMT). No. of bitstreams: 1 ArcilaGago_ManuelFelipe_M.pdf: 7192169 bytes, checksum: 7b23a08769656a07765344e20d1f6ad4 (MD5) Previous issue date: 2013 / Resumo: A variedade de materiais no setor aeronáutico para redução de peso e custo tem se proliferado a um grau intensivo, onde têm sido revisadas diferentes pesquisas para encontrar outros tipos de materiais de fácil maneabilidade para construção de peças que satisfazem as restrições impostas. Assim, existe uma procura constante de soluções para facilitar a produção, e ao mesmo tempo aumentar a segurança das aeronaves levando em consideração pontos importantes como a fadiga e ruptura do material. Um material frequentemente utilizado que atende a estes requisitos devido a suas propriedades de densidade e resistência é o alumínio, e é neste ambiente que existe um processo de manufatura utilizado para a soldagem conhecido como "Friction Stir Welding" (FSW). No presente momento, estudos para criação de modelos que representem características mecânicas utilizadas em projetos em função de parâmetros do processo tem sido pesquisados. Embora este processo seja de difícil modelagem devidos as suas complexidades, tem sido estudado e utilizado diferentes algoritmos que possibilitem o melhoramento da representação do modelo, tais como os relacionados com máquinas de aprendizagem (ML) e suas diferentes otimizações. Neste contexto, a presente pesquisa tem seu foco na obtenção de um modelo baseado no algoritmo de aprendizagem de Máquina de Vetores de Suporte (SVM), e também com outros algoritmos tais como Regressão Polinomial (RP) e Rede Neural Artificial (RNA), buscando encontrar modelos que representem o processo de soldagem por FSW através das propriedades mecânicas obtidas pelos ensaios de tração e por análise de variância (ANOVA), entendendo suas vantagens e, posteriormente, recomendar quais dos algoritmos de aprendizagem tem maior beneficio / Abstract: In the aerospace industry to reduce weight and cost, a great quantity of materials has been used, which has generated research to find types of materials, that have been better maneuverability and to guarantee the properties required to development of pieces for the industry. Thus, the studies look for optimize between production easiness and increase the aircraft safety, taking into consideration important issues such as fatigue and fracture of the materials. One of the most common approach used is aluminum by their mechanical properties (density and strength), although it has many problems to be welding with the traditional methods. Currently, the Friction Stir Welding (FSW) process is used in the industry, as well in the academy. However, the FSW is difficult to model by the complexities in the physical phenomenal occurred during the weld process, as result, has been studied and used different algorithms that allow enhance the model representation. The Machine Learning (ML) is a methodology studied to obtain the model optimized. In this context, the present research focus by to obtain a model-based in learning algorithm using Support Vector Machine (SVM). Although comparisons were made with other algorithms such as Polynomial Regression (PR) and Artificial Neural Network (ANN), searching to find models that represent the FSW process weld using the mechanical properties obtained by tensile tests and analysis of variance (ANOVA). Finally, conclusions to understand the advantages learning algorithms are presented / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestre em Engenharia Mecânica

Le soudage par friction malaxage « Friction Stir Welding » est un procédé d'assemblage de pièces en phase semi solide. Le cordon de soudure est obtenu grâce à un outil de révolution composé d'un épaulement et d'un pion. Ce procédé utilise le principe de la conversion de l'énergie mécanique en énergie thermique par frottement de l'outil avec les pièces à assembler. Ce travail de thèse est une contribution à la modélisation expérimentale et à la simulation numérique de ce procédé permettant de fournir des modèles pour aider à la compréhension des phénomènes thermiques et mécaniques ainsi que les interactions entre les paramètres de soudage. Les études expérimentales sont principalement orientées vers la caractérisation de l'écoulement de matière à l'aide de marqueurs et vers l'optimisation des paramètres du procédé. L'effet de la géométrie de l'outil (outil à pion cylindrique ou outil avec méplats) sur l'écoulement de la matière au cours du soudage est étudié. Pour représenter la géométrie de l'écoulement et prédire les champs thermiques et mécaniques à l'état stationnaire, des modèles formulés sur la base de la dynamique des fluides sont adoptés dans la présente thèse. Deux modèles thermomécaniques sont développés: (1) un premier modèle numérique construit sous Fluent permet d'étudier le comportement thermomécanique et l'écoulement au cours du soudage FSW. Une loi de comportement dépendante de la température et de la vitesse de déformation est utilisée et une discussion sur les conditions de contact entre l'outil et les plaques à souder est présentée. Les résultats de l'écoulement sont comparés avec ceux obtenus expérimentalement dans le cas de suivi des trajectoires de particules de cuivre. (2) un deuxième modèle original basé sur une procédure itérative est mis en œuvre permettant le soudage de plaque de grandes dimensions. En effet, pour une meilleure prise en compte des conditions aux limites thermiques, un modèle thermomécanique construit autour de l'outil de soudage et couplé avec un modèle thermique pour tout le reste du domaine étudié. Ce modèle permet de prendre en compte le transfert de chaleur dans l'outil et dans la plaque support. Les cycles thermiques et la plage de viscosité pour deux alliages d'aluminium (AA7020-T6 et AA6061-T3) sont analysés et comparés avec succès aux résultats expérimentaux. Les efforts et le couple de soudage calculés numériquement sont validés par rapport à la littérature. / The friction stir welding is a process for assembling a semisolid phase parts. The weld seam is achieved by a revolution tool consists of the shoulder and the pin. This process converts the mechanical energy into heat energy by friction of the tool with the parts to be joined.This thesis is a contribution to the experimental modeling and numerical simulation of this process in order to provide models to assist in understanding the thermal and mechanical phenomena and interactions between the welding parameters. Experimental studies are used for the characterization of the material flow with markers and to optimize the process parameters. The effect of tool geometry (cylindrical pin or tool with flats) on the material flow during welding is studied. To represent the flow geometry and predict the thermal and mechanical fields in the steady state, CFD models are adopted in this thesis. Two thermomechanical models are developed: (1) first numerical model is used to study the thermomechanical behavior and flow during FSW. FLUENT is employed to solve the coupled thermal and fluid flow equations. A behavior law depending on temperature and strain rate is used and a discussion on the contact conditions between the tool and the workpieces is presented. The results of the flow are compared with those obtained experimentally in the followed case of trajectories copper particles. (2) A second original model based on iterative procedure is implemented to welding large plates. To take account the correctly thermal boundary conditions, a thermomechanical model built around the welding tool and coupled with a thermal model for the rest of the area studied. This model allows taking into account the transfer of heat in the tool and in the backing-bar. Thermal cycles and the viscosity range for two aluminum alloys (AA7020-T6 and AA6061-T3) are successfully analyzed and compared with experimental results. The loads and torque welding are calculated numerically and validated in the literature.

I och med ökad konkurrens och ökade kundkrav behöver företag bedriva produktutveckling med en rapid approach. ESAB:s sätt att svara upp mot dessa krav har resulterat i ett brett produktsortiment av friktionssvetshuvuden. Deras långsiktiga mål är därför att inrätta en ny standard för att reducera kostnader. Planen är därför att aktivt bedriva produktutveckling tillsammans med universitet och högskolor runt om i landet. Detta produktutvecklingsprojekt kan beskrivas som en första etapp i ett hållbart produktutvecklingsarbete på ESAB mot ett långsiktigt mål om att sätta en ny standard inom FSW. Syftet med detta examensarbete är att analysera två friktionssvetshuvuden: FSW LEGIO 4UT och FSW LEGIO 5UT genom K-FMEA och DFA2. Därigenom kartlägga vilka potentiella brister som finns i de befintliga konstruktionerna och belysa var bristerna finns. Analysresultatet ska tillsammans med produktutvecklingsteori, forma ett informationsunderlag för bedrivande av fortsatt produktutvecklingsarbete. Tillvägagångssättet har bedrivits genom tre faser: 1) uppstart/förarbete, 2) genomförande och 3) avslut/presentation. Vetenskapssynen har präglats av positivism med vissa inslag av hermeneutik. Forskningsmetoden har både vad gäller datainsamling, analysmetodik och utvärderingsmetodik varit kvalitativ. Undersökningsansatsen har varit av abduktiv art medan undersökningsinriktningen har varit kombinerad explorativ och deskriptiv. Först analyserades ritningsunderlag, komponentlistor, STEP-filer och produktspecifikationer för sammanlagt åtta svetshuvuden. Ett strategiskt urval av tre gjordes för att sedan ytterligare exkludera ett av dessa. En djupdykning gjordes i det empiriska underlaget som gav upphov till två hypoteser: 1) komponenter som under drift är i rörelse torde vara högst belägna att utsättas för fel, 2) svetshuvudena kan möjligen komprimeras volymmässigt och komponenter reduceras till antalet. En semi-strukturerad fokusgruppintervju genomfördes för fyra respondenter med teknisk expertis på området. Resultatet av denna gjorde att hypoteserna kunde bekräftas. Detta empiriska datamaterial ställdes mot befintlig teori rörande K-FMEA och DFA2. Vidare utfördes K-FMEA enligt teorin i sju steg på såväl systemnivå som komponentnivå. Sedan gjordes detsamma för DFA2 vars arbetssätt sker i fyra steg. Resultatet av denna studie påvisar att de allra mest kritiska komponenterna rör spindeln, spindelhuset och lager. Påfrestningar finns också i hydrauliken som har tilldelats höga kriticitetstal. Det finns även feltyper med hög risk relaterat till tillverkning och montering. Vad gäller DFA2 visade resultatet att de befintliga konstruktionerna har låg grad av monteringsvänlighet. Detta då flera monteringsriktningar används, flera komponenter är stora, tunga och svårhanterade. En del komponenter har svåråtkomlig monteringsplats. Vidare behövs för vissa komponenter extra hjälpmedel och verktyg för att kunna genomföra monteringen. Rekommendationen till ESAB i en potentiell andra etapp av detta produktutvecklingsarbete, vore att komplettera denna studie med andra produktutvecklingsverktyg och kostnadskalkyler. Detta för att se till att göra rätt från början och eliminera risker för eventuella efterföljande kostsamma justeringar. / With increased competition and increased customer requirements, companies need to pursue product development with a rapid approach. ESAB’s way of responding to these demands has resulted in a wide product range of friction welding heads. Their long-term goal is therefore to set a new standard to reduce costs. The plan is to actively conduct product development together with universities in Sweden. This product development project can be described as a first step in sustainable product development work at ESAB towards a long-term goal of setting a new standard within FSW. The purpose of this thesis is to analyze two friction welding heads: FSW LEGIO 4UT and FSW LEGIO 5UT through K-FMEA and DFA2. Thereby, identify the potential shortcomings found in the existing constructions and highlight where the shortcomings exist. The analysis results, together with product development theory, will form an information base for the pursuit of continued product development. The approach has been conducted through three phases: 1) start-up/pre-work, 2) implementation, and 3) completion/presentation. The science view has been characterized by a positivistic nature with elements of hermeneutic in some respects. The research method has been qualitative in terms of data collection, analysis methodology and evaluation methodology. The research effort has been abductive, while the research focus has been a combination of exploratory and descriptive. First, drawing documents, component lists, STEP files and product specifications were analyzed for a total of eight friction welding heads. A strategic selection of three was made to further exclude one more. An in-depth dive into the empirical evidence was made which gave rise to two hypotheses: 1) Components that move relative to others in welding operation, are most likely to be subject to errors; 2) These friction welding heads may possibly be compressed in terms of volume, and the components reduced to the number. A semi-structured focus group interview was conducted with a total of four respondents who have technical expertise in the field. The result of the interview allowed the hypothesis to be confirmed. The empirical data was put against the existing theory regarding K-FMEA and DFA2. Furthermore, K-FMEA was performed according to the theory in seven stages at both system level and component level. Then the same was done for DFA2, whose mode of operation takes place in four stages. The result of this thesis shows that the most critical components relate to the spider, spider house and the bearings. There are also stains in hydraulics that have been awarded high risk priority numbers. Along with that, failure modes with high risk frequencies related to manufacturing and assembly are identified. When it comes to DFA2, the result shows that the existing designs has low degree of assembly ease. This is because several mounting directions are used, several components are large, heavy and difficult to handle. Some components also have difficult access point. In addition, external equipment such as fixtures and tools, are needed to install some components. The recommendation to ESAB in a potential second step of their product development work, would be to complement this study with other product development tools and cost calculations. This to make sure to do right from the start and eliminate risks for any subsequent costly adjustments.

Dans le domaine aéronautique, le soudage par friction-malaxage (FSW – Friction Stir Welding) apparait comme un procédé innovant d’assemblage des fuselages, allégeant la structure avion en remplaçant la technique actuelle de rivetage. La simulation numérique est un support permettant de mieux comprendre et maîtriser ce procédé. L’alliage d’aluminium étudié dans ce projet est de type AA2024-T3 et tire principalement ses propriétés du durcissement structural. Modéliser l’évolution de la précipitation s’avère essentiel pour définir les propriétés finales de la soudure. Le modèle choisi doit prendre en considération les familles de précipités formés et les processus de germination, croissance et coalescence. Il doit être précis, robuste et rapide pour être applicable au procédé FSW. Un modèle de classe couplé à des calculs d’équilibre thermodynamiques a été choisi dans cette étude. Pour définir la cinétique de croissance d’un précipité, une cinétique de croissance initialement établie pour un alliage binaire a été étendue à un alliage multicomposé. Connaissant la distribution en taille des familles de précipités, les propriétés mécaniques sont définies selon un modèle empirique. La calibration anisotherme a été réalisée via un essai de DSC où signaux expérimentaux et simulés ont été comparés pour déterminer la teneur initiale des phases en présence et définir les paramètres matériaux de l’alliage. L’essai isotherme a établi le lien entre état de précipitation et propriétés mécaniques résultantes. Le modèle est appliqué à la simulation des évolutions microstructurales au cours d’une soudure FSW afin de prédire les propriétés finales du joint soudé. Les évolutions thermiques sont déterminées via l’utilisation d’un modèle macroscopique développé parallèlement dans une thèse, également portée par la Chaire Daher. Les données numériques obtenues sont comparées à des expériences instrumentées, montrant une bonne estimation des duretés. Les profils expérimentaux sont retrouvés, de même que les caractéristiques des différents domaines, validant l’approche et sa capacité à simuler efficacement les évolutions des processus de précipitation. / In the aeronautic industry, the friction stir welding (FSW) process is seen as an interesting option to lighten aircraft structure by replacing the standard riveting technology used to join parts. Numerical simulation is chosen to improve understanding of the different mechanisms occurring during FSW. The aluminum alloy studied is an AA2024-T3 grade. Its mechanical properties mainly derive from structural hardening mechanisms. An accurate model of precipitate evolution is essential to define hardness profile of the weld. The chosen simulation has to be robust and time-efficient in order to be suitable for the FSW process modeling. It must consider the two families of precipitates (GPB zones and S phase) and model nucleation, growth and coarsening phenomena. A PSD model is chosen and coupled with thermodynamic equilibrium calculations. To define the growth kinetics of precipitates, an exact analytical solution is extended to a multi-component alloy. Knowing the distribution of precipitates size, the mechanical properties are defined based on an empirical model. The amount and properties of phases are initialized through non-isothermal DSC calibration and comparison between experimental heat flux and simulated one. Isothermal test is selected to establish the link between precipitation state and mechanical properties. The model is applied to the simulation of microstructural evolution in FSW in order to predict the final properties of the weld. Thermal changes are determined through the use of a macroscopic model developed during a twin project within the Chair Daher. Numerical results are compared with instrumented experiments and show a good estimate of hardness. The experimental profiles are found, as well as the characteristics of the different areas. This validates the approach and its efficiency to simulate the evolution of the precipitation process.

Friction stir spot welding (FSSW) is a solid-state welding process which is especially useful for joining precipitation-hardened aluminium alloys that undergo adverse property changes during fusion welding. It also has potential as an effective method for solid-state joining of dissimilar alloys. In FSSW, heat generation and plastic flow are strongly linked, and the scale of the process in time and space is such that it is difficult to separate and control the influence of all the relevant input parameters. The use of modelling is well-established in the field of welding research, and this thesis presents an analysis of the thermal and mechanical aspects of FSSW, principally using the finite element (FE) technique. Firstly, a thermal FE model is shown, which is subsequently validated by reference to experimental temperature data in both aluminium-to-aluminium and aluminium-to-steel welds. Correlations between high-quality welds and temperature fields are established, and predictions are made for peak temperatures reached under novel welding conditions. Deformation and heating are strongly linked in FSSW, but existing modelling tools are poorly suited to modelling flow processes in the conditions extant in FSSW. This thesis discusses the development and optimisation of two novel techniques to overcome the limitations of current approaches. The first of these uses greatly simplified constitutive behaviour to convert the problem into one defined purely by kinematics. In doing so, the boundary conditions reduce to a small number of assumptions about the contact conditions between weld material and tool, and the model calculation time is very rapid. This model is used to investigate changes in the slip condition at the tool to workpiece interface without an explicit statement of the friction law. Marker experiments are presented which use dissimilar composition but similar strength alloys to visualise flow patterns. The layering behaviour and surface patterns observed in the model agree well with observations from these experiments. The second approach extends the FE method to include deformation behaviour without the need for a fully-coupled approach, guided by the kinematic model. This is achieved using an innovative sequential small-strain analysis method in which thermal and deformation analyses alternate, with each running at a very different timescale. This technique avoids the requirement to either remesh the model domain at high strains or to use an explicit integration scheme, both of which impose penalties in calculation time and model complexity. The method is used to relate the purely thermal analysis developed in the work on thermal modelling to welding parameters such as tool speed. The model enables predictions of the spatial and temporal evolution of heat generation to be made directly from the constitutive behaviour of the alloy and the assumed velocity profile at the tool-workpiece interface. Predictions of the resulting temperature history are matched to experimental data and novel conditions are simulated, and these predictions correlate accurately with experimental results. Hence, the model is used to predict welding outcomes for situations for which no experimental data exists, and process charts are produced to describe optimum welding parameters. The methods and results presented in this thesis have significant implications for modelling friction stir spot welding, from optimising process conditions, to integration with microstructural models (to predict softening in the heat-affected zone, or the formation of intermetallics at the interface in dissimilar welds). The technique developed for sequential small strain finite element analysis could also be investigated for use in other kinematically constrained solid-state friction joining processes.

Le procédé de soudage par friction et malaxage est un procédé récent qui a été développé au sein de l'institut de soudure britannique "The Welding Institute" au début des années 90. Ce procédé, utilisé généralement en aéronautique, est sans apport de matière et permet de souder principalement des alliages d'aluminium difficilement soudables par les procédés classiques de soudage. Il consiste à malaxer le matériau de base à l'aide d'un outil constitué d'un pion et d'un épaulement frottant sur les faces supérieures des tôles à souder. La modélisation de ce procédé est très complexe puisque ce dernier implique des couplages entre des phénomènes mécaniques, thermiques et métallurgiques. Le malaxage dans le procédé de soudage FSW est difficile à simuler à l'aide de la méthode des éléments finis (en lagrangien) puisque la zone proche de l'outil de soudage est le siège de grandes déformations. Donc le remaillage est nécessaire. Cependant, le remaillage est cher et très difficile pour les problèmes tridimensionnels. Par ailleurs, après un remaillage, il est nécessaire d'interpoler les champs (vitesses, contraintes,...) correspondant à la solution courante, ce qui peut introduire des erreurs supplémentaires dans le calcul (on parle de diffusion numérique). Nous proposons dans ce travail des modèles basés sur la méthode sans maillage dite "Smoothed Particle Hydrodynamics SPH" et la méthode des moindres carrés mobiles (Moving Least Square MLS) pour la simulation de ce procédé. Ces modèles sont formulés dans le cadre lagrangien et utilisent la forme forte des équations aux dérivées partielles. Le premier modèle basé sur SPH considère la zone de soudure comme un fluide non newtonien faiblement compressible et dont la viscosité dépend de la température. Ce modèle est proposé pour la simulation numérique du comportement thermomécanique d'un matériau soudé par le procédé FSW. Dans le deuxième modèle, un algorithme itératif implicite de premier ordre a été proposé, pour simuler le malaxage de la matière dans le cas d'un matériau viscoplastique, en utilisant la méthode MLS et la technique de collocation. Le troisième modèle est un algorithme implicite d'ordre élevé basée sur le couplage de la méthode MLS et la Méthode Asymptotique Numérique MAN. Cet algorithme permet de réduire le temps de calcul par rapport à l'algorithme itératif implicite de premier ordre. La validation de ces trois modèles proposés a été faite par le code industriel Fluent / Friction stir welding is a recent process that has been developed by the British Welding Institute TWI "The Welding Institute" since 1990s. This process, generally used in aerospace, does not need additional material and allows mainly joining plates of aluminum alloys which are difficult to weld by the classical welding processes. It consists in mixing the base material using a tool comprising a pin and a shoulder which heats the plates to be welded by friction. The modeling of this process is very complex since it involves the coupling between mechanical, thermal and metallurgical phenomena. The mixing in welding process FSW is difficult to simulate using finite element method in lagrangian framework since the area near the welding tool is submitted to large deformations. So remeshing procedure is often required. However, remeshing can be very expensive and difficult to perform for three-dimensional problems. Moreover, after remeshing step, it is necessary to interpolate the fields (velocities, constraints ...) corresponding to the current solution, which may lead to additional errors in the calculation (called numerical diffusion). We propose in this work models based on meshless methods called "Smoothed Particle Hydrodynamics SPH" and Moving Least Square method for the simulation of this welding process. These models are formulated in lagrangian framework and use the strong form of partial differential equations. The first model based on SPH considers the welding zone as a weakly compressible non-newtonian fluid and whose viscosity depends on the temperature. This model is proposed for the numerical simulation of thermo-mechanical behavior of a welded material by FSW process. The second model is a first order implicit iterative algorithm proposed to simulate material mixing in the case a visco-plastic behavior using the MLS method and the collocation technique. The third model is a high order implicit algorithm based on the coupling of MLS method and Asymptotic Numerical Method (ANM). This algorithm allows reducing the computation time by comparison with the first order implicit iterative algorithm. The validation of these three proposed models was done by the industrial code Fluent

Le soudage FSW (Friction Stir Welding) est un procédé de soudage en phase solide pressenti pour des applications de transport en générale aérospatial et naval. Malgré le nombre considérable d’études qui ont été réalisées depuis son avènement en 1991, le contrôle du procédé n’est pas encore effectif.Ce travail a consisté en une partie expérimentale visant à la génération, par un outil trigone, de joints soudés dont la microstructure a été corrélée à l’écoulement de matière pendant le procédé. La connaissance de cet écoulement de matière a permis dans la deuxième partie d’enrichir le modèle thermofluide développé en périodique pour prédire la microstructure des joints de soudure FSW, notamment les "onion rings". Finalement, l’occurrence des "onion rings" a été corrélée à la vitesse de déformation maximale atteinte par les particules de la zone soudée, prédite par le modèle. Par ailleurs, un travail d’affinement du champ de vitesse en voisinage du pion est réalisé en modélisant l’outil trigone. Ce qui permet en plus de l’interaction (entrainement) outil/matière par frottement, d’intégrer une interaction par obstacle. Cette approche devrait permettre, en perspectives de ce travail, une meilleur description thermomécanique locale et par voie de conséquence microstructurale. / Friction Stir Welding is a solid state joining process developed for transport applications as aerospace and naval. Since its introduction, a large number of investigations have been carried out but the process is not fully controlled. This work including experimental section in which welds have been generated by trigonal tool. The microstructure of these welds has been correlated with the material flow during the process. By understanding the material flow, the transient thermofluid model developed in the second section has been significantly enriched. This modeled has been developed for predicting the microstructure of the weld, especially, the "onion rings". Finally, the occurrence of "onion rings" has been correlated with the maximal strain rate reached by any particle in the weld seam, simulated by the model. However, the velocity has been refined at the vicinity of the tool through the trigonal pin modelling. This was helpful to move the material not only by friction but also by obstacle at the interaction tool/material. The above approach should enable, in this work layout, a better local thermomechanical description and consequently microstructural.

<div>Aluminum 6061 is extensively used in industry and welding and additive manufacturing (AM) of Al6061 offer flexibility in manufacturing. Solid-state welding and AM processes can overcome the shortcomings of fusion-based processes, such as porosity and hot cracking. In this thesis, friction stir welding and friction stir additive manufacturing, which are both based on the concepts of friction stir processing (solid-state), were studied. The welding parameters for a sound weld during friction stir welding of Al6061-T6 alloy were determined based on the experimental and numerical analysis. Formation of tunnel defects and cavity defects was also studied. A Coupled Eulerian-Lagrangian (CEL) finite element model was established to analyze the process, where the workpiece was modeled as an Eulerian body, and the tool as Lagrangian. The model was validated by conducting experiments and correlating the force measured by a three-axis dynamometer. The experimentally validated simulation model was used to find an optimum parameter set for the sound weld case.</div><div>To demonstrate the friction stir additive manufacturing process, a 40 mm × 8 mm × 8 mm (L×B×H) material was fabricated by adding five 1.6 mm thick plates. A similar coupled Eulerian-Lagrangian based finite element model was used to predict the effects of sound process parameters, such as the tool’s rotational speed and the translational speed. The temperature predicted by the model was used to predict the microhardness distribution in the sample and to further elucidate the hardness change in the weld zone, which showed a good agreement with the experimental results. The microstructure of the samples was analyzed, and the mechanical properties of the additive manufactured samples were characterized and compared with those of other AM techniques via tensile tests and tensile shear tests.</div>