Search

International audience; Low back pain represents a major economic and societal challenge due to its high prevalence. Lumbar orthoses are one of the recommended treatments. Even if previous results showed their clinical effects, the detailed mode of action is still poorly known, making the device design difficult. A renewed instrumentation and experimental protocol should bring better insight into the lumbar brace–trunk mechanical interaction. This instrumentation should give detailed information on the basic physical or geometrical parameters: the pressure applied on the trunk, the body shape and the strain in the belt. The principal objective of this study was to propose and validate a new measurement protocol, based on pressure mapping systems and full-field shape and strain measurement. The feasibility of the protocol was tested along with its validity and repeatability. The influence of various parameters, which could cause changes in the measurements, was tested with six different belt configurations on one subject. Measurements were also performed to study the impact of posture on pressure and strain. Both pressure and strain appeared to be asymmetric from left to right. The pressure applied by the lumbar belt on the back varies with breathing and with posture. This study showed that full-field measurements were necessary to render the high variability of pressure or strain around the trunk, under recommendations of their use to guarantee a satisfying repeatability.

With the rising popularity of endovascular aortic repair (EVAR) for aortic aneurysms and dissections, there is a crucial need for investigating the delayed appearance of post-EVAR complications such as stent-graft kinking, fracture and migration respectively. These complications have been noted to be influenced by the radial stiffness and bending flexibility attributes of stent-grafts. Auxetic designs with negative Poisson's ratio offer interesting advantages such as enhanced fracture toughness, superior indentation resistance and adaptive stiffness in response to intricate morphology for stenting applications over conventional stent designs. The objective of this study is to propose different auxetic stent candidates and to compare their mechanical performance with two conventional stent candidates for endovascular applications using numerical simulation through crimp/crushing tests for their radial stiffness and three-point bending/kinking tests for their flexibility, respectively. The results demonstrate that the novel hybrid auxetic designs (CRE and CSTAR) possess the best trade-off between radial stiffness and bending flexibility characteristics among all candidates for stent-graft applications.

International audience; Objective: A numerical 3D model of the human trunk was developed to study the biomechanical effects of lumbar belts used to treat low back pain. Methods: This model was taken from trunk radiographies of a person and simplified so as to make a parametric study by varying morphological parameters of the patient, characteristic parameters of the lumbar belt and mechanical parameters of body and finally to determine the parameters influencing the effects of low back pain when of wearing the lumbar belt. The loading of lumbar belt is modelled by Laplace's law. These results were compared with clinical study. Results: All the results of this parametric study showed that the choice of belt is very important depending on the patient's morphology. Surprisingly, the therapeutic treatment is not influenced by the mechanical characteristics of the body structures except the mechanical properties of intervertebral discs. Discussion: The numerical model can serve as a basis for more in-depth studies concerning the analysis of efficiency of lumbar belts in low back pain. In order to study the impact of the belt's architecture, the pressure applied to the trunk modelled by Laplace's law could be improved. This model could also be used as the basis for a study of the impact of the belt over a period of wearing time. Indeed, the clinical study shows that movement has an important impact on the distribution of pressure applied by the belt.; Objectif : Un modèle numérique 3D du tronc humain a été développé pour étudier les effets biomécaniques des ceintures lombaires utilisées pour traiter les lombalgies. Méthodes : Ce modèle a été pris à partir de radiographies du tronc d'une personne et simplifié de manière à faire une étude paramétrique en faisant varier les paramètres morphologiques du patient, les paramètres caractéristiques de la ceinture lombaire et les paramètres mécaniques du corps et enfin à déterminer les paramètres influençant les effets de la lombalgie lors du port de la ceinture lombaire. La charge de la ceinture lombaire est modélisée par la loi de Laplace. Ces résultats ont été comparés à ceux d'une étude clinique. Résultats : Tous les résultats de cette étude paramétrique ont montré que le choix de la ceinture est très important en fonction de la morphologie du patient. De façon surprenante, le traitement thérapeutique n'est pas influencé par les caractéristiques mécaniques des structures corporelles, à l'exception des propriétés mécaniques des disques intervertébraux. Discussion : Le modèle numérique peut servir de base à des études plus approfondies concernant l'analyse de l'efficacité des ceintures lombaires dans les lombalgies. Afin d'étudier l'impact de l'architecture de la ceinture, la pression appliquée sur le tronc modélisée par la loi de Laplace pourrait être améliorée. Ce modèle pourrait également servir de base à une étude de l'impact de la ceinture sur une période de port. En effet, l'étude clinique montre que le mouvement a un impact important sur la répartition de la pression appliquée par la ceinture.

International audience; Lower back pain is a major public health problem. Despite claims that lumbar belts change spinal posture due to applied pressure on the trunk, no mechanical model has yet been published to prove this treatment. This paper describes a first model for belt design, based on the one hand on the mechanical properties of the fabrics and the belt geometry, and on the other hand on the trunk geometrical and mechanical description. The model provides the estimation of the pressure applied to the trunk, and a unique indicator of the belt mechanical efficiency is proposed: pressure is integrated into a bending moment characterizing the belt delordosing action on the spine. A first in-silico clinical study of belt efficiency for 15 patients with 2 different belts was conducted. Results are very dependent on the body shape: in the case of high BMI patients, the belt effect is significantly decreased, and can be even inverted, increasing the lordosis. The belt stiffness proportionally increases the pressure applied to the trunk, but the influence of the design itself on the bending moment is clearly outlined. Moreover, the belt/trunk interaction, modeled as sticking contact and the specific way patients lock their belts, dramatically modifies the belt action. Finally, even if further developments and tests are still necessary, the model presented in this paper seems suitable for in-silico pre-clinical trials on real body shapes at a design stage.

Numerical simulation for Resin Transfer Molding manufacturing process is attempted by using theeXtended Finite Element Method (XFEM) combined with the level set method. XFEM allows to obtaining a goodnumerical precision of the pressure near the resin flow front, where its gradient is discontinuous. The enriched shapefunctions of XFEM are derived by using the level set values so as to correctly describe the interpolation with the resinflow front. In addition, the level set method is used to transport the resin flow front at each time step during themold filling. The level set values are calculated by an implicit characteristic Galerkin FEM. The multi-frontal solver ofIPSAP is adopted to solve the system. This work is validated by comparing the obtained results with analyticsolutions. Moreover, a localization method of XFEM and level set method is proposed to increase the computingefficiency. The computation domain is reduced to the small region near the resin flow front. Therefore, the totalcomputing time is strongly reduced by it. The efficiency test is made with a simple channel flow model. Severalapplication examples are analyzed to demonstrate ability of this method.초록

International audience; Numerical simulation for resin transfer moulding manufacturing process is attempted using the extended finite element method combined with the level set method. The level set method is used to transport the resin flow front at each time step during the mould filling. Extended finite element method allows to obtaining a good numerical precision of the resin pressure near the resin flow front, where the gradient of the pressure could be discontinuous. The enriched shape functions of extended finite element method are derived using the level set values so as to correctly describe these shape functions with the resin flow front. In addition, the position of the resin flow front at each time step is calculated by an implicit characteristic Galerkin finite element method. Some examples are presented to validate our study in comparison with analytical or experimental results and to illustrate its industrial applications.

Objective With cartilage physiopathology, knee stability and therefore the subsequent loadings on ligaments have a strong impact in the development of knee osteoarthritis. To understand this phenomenon, authors are proposing different knee models. Most of them are without any flexion, some have been introduced by controlling actively muscular groups. In this case, the forces used to achieve the flexion may not be physiological, and the pressure on the articulation can be questioned. We propose a mixed approach, both using medical imaging (MRI, EOS X-ray system) and force platform in conjunction with a finite element model. Material and methods One healthy volunteer underwent a MRI and an EOS imaging of the knee. EOS images gave the exact position of the bones for five flexion angles, the force platform ensuring consistent boundary conditions. A 3D geometrical model of the bones and cartilages was segmented from the MRI stacks; this model was meshed and smoothed for finite element analysis (FEA). The bones of the model were fitted to those of the EOS images to have the physiological positions of the knee. The FEA was carried out with the EOS imaging experimental boundary conditions, ensuring the global knee mechanical equilibrium. Results To validate this patient-specific model, its bony structure was confronted with the EOS images once the mechanical equilibrium reached. The difference of position was within the error range of the image registration, showing a satisfying equilibrium state. Last, this model gave us an estimation of the tension in the ligaments for every flexion as well as a pressure map on the cartilages. Discussion/Conclusion The FEA model main assumption lies on the absence of menisci and muscular active control. Nevertheless, the model is suitable to achieve a reasonable equilibrium. It has been used successfully in the study of ligament tension during the flexion cycle. It will then be used as a reference in studying arthritic knee joints for clinical cases.

International audience; Knee degradation and pain when developing osteoarthritis are strongly related not only to the pressure on the cartilage, but also to the knee stability and to the subsequent loadings on the ligaments. Authors proposed the knee models without any flexion. Knee flexion had been introduced by controlling actively muscular groups or by using a full displacement-controlled model. But in both cases, the pressure applied on the articulation has been questioned, the forces used to achieve the flexion being far from the physiology in the former and the bone positioning uncertainties leading to high variations in the cartilage pressure and ligament loads in the latter. Here, we propose a mixed approach, both using medical imaging (MRI, EOS X-ray system) and force platform in conjunction with a finite element model. A healthy volunteer underwent a MRI and an EOS imaging of the knee. EOS images gave the exact position of the bones for five flex-ion angles of the knee, ranging from 0° to 85°. The subject's knee was loaded with a specific force on his foot during all acquisitions to keep consistent boundary conditions on the knee. A 3D geometrical model of the bones and cartilages was segmented from the MRI stacks; this model was meshed and smoothed for Finite Element

International audience; The main goal of this study is to optimize the structural thickness distribution for improving the mechanical performance of composite structures manufactured by compression molding processes. During the preliminary preparation for optimization, a processing simulation program was developed to predict the fiber volume fraction and orientation using a Control Volume/Finite Element method. To verify the effects of the processing conditions on the mechanical properties, a few simple geometries and loading conditions were investigated. Considering the fiber state changes, the optimal thickness distributions of a plate in one- and two-dimension have been searched. Box's complex method was applied to solve the non-linear inequality constrained problem. (C) 2009 Elsevier Ltd. All rights reserved.

International audience; We suggest new approaches to reduce the number of fitness function evaluations in genetic algorithms (GAs) applied to multidisciplinary optimization of composite laminates. In the stacking sequence design of laminated structures, the design criteria are classified into two groups, which are layer combination dependent criteria and layer sequence dependent criteria. The memory approach is employed to lessen the number of fitness function evaluations for the identical design individuals that appear during the search. The permutation operator with local learning or random shuffling is applied to the same design individual to improve the fitness for layer sequence dependent criterion, while maintaining the same performance for layer combination dependent criterion. The numerical efficiency of the present method is validated by the sample problem of weight minimization of composite laminated plate under multiple design constraints.

International audience; The knee is the joint that support the most stress in the human body. It is not uncommon to see osteoarthritis develop in this joint. A survey lead in 18 different countries showed that 175 out of 100 000 persons needed a total knee arthroplasty [Kurtz, 2011], which became a common procedure against the knee osteoarthritis. But with a bad position, the prosthesis ages faster and has to be replaced [Liau, 2002]. Any new intervention is a hard task; it is very painful for the patient, he will lose mobility with the new joint replacement, and this intervention is very expensive. To avoid that, it is important that the position of the knee prosthesis is correct at the first intervention. This study aims to build a Finite Element model of the knee specific-patient allowing us to predict the biomechanical effects of the surgery and to propose an optimal position for the knee implant.

A structure-process integrated optimization methodology is proposed considering both the structural properties and manufacturing process for the weight minimization of composite laminated plates manufactured by Resin Transfer Molding (RTM) process. In the optimization problem, the structural weight of composite structures is minimized under the stiffness constraint and the mold filling time constraint which is a part of process cycle time. With the results obtained, it is investigated how the weight and material cost are traded-off. The optimization methodology suggests a guide of material selection in the preliminary conceptual design stage.

A multi-objective optimization methodology is proposed which simultaneously considers the mechanical performance and the manufacturing cost from the early stage of design of composite laminated plates. Resin transfer molding process is chosen as the manufacturing process for its wide acceptance among industries. Interrelations between the structural parameters and the process parameters are identified. With proper assumptions, the multi-objective problem is formulated into a problem with a single objective function. As design variables, the stacking sequence of layers is investigated which minimizes the maximum displacement and the mold filling time. Injection gate location is also considered in order to minimize the mold filling time for a given structure geometry, material properties and loading condition. The genetic algorithm is adopted as the optimization scheme. The results from the sample problems illustrate usefulness of the developed methodology.

A multi-constraint optimization methodology for the design of composite laminated plates manufactured by Resin Transfer Molding (RTM) process is presented. As design constraints, both the manufacturing process requirement and the structural requirement were considered. The stacking sequence of layers and the thickness of the structure were optimized for a preassigned geometry of the structure, constituent materials and loading condition. As a structural requirement, either the stiffness constraint or strength constraint was selected. Such constraint was used as basis in the enhancement for the specific mechanical performance of composite structures. Maximum allowable mold filling time was assigned as a process requirement in the effort to increase productivity. The objective function was newly defined without introduction of any auxiliary parameters, such as scaling parameters, weighting coefficients and penalty parameters which are usually involved in multi-constraint or multi-objective optimization problems. As an optimization method, the genetic algorithm was used. The results for a sample problem show the validity of the proposed optimization methodology.

This paper presents and illustrates a straightforward method to determinate the concentration of the humidity in non steady and transient states in the core of the resin matrix composite panel exposed to cyclic temperature and moisture environmental conditions during a long time. In order to complete the previous studies, an analytical solution in transient state is proposed. Several examples are presented to demonstrate the assessment and performance in this method compared to the classical step-by-step integration.

International audience; The main goal of the present study is to optimise the precharge conditions such as the precharge location and dimensions that give significant effects on the mechanical performance of composite structures manufactured by the compression moulding process. As preliminary step of optimisation, we developed a manufacturing simulation program to predict the fibre volume fraction and fibre orientation. And coupled with this simulation program and a structural analysis program, a genetic algorithm (GA) is implemented to optimise the precharge conditions. The penalty function method and the repair algorithm are modified for handling constraints. The repair algorithm is applied to a symmetric structure and an arbitrary shape structure to find optimal precharge conditions. (C) 2011 Elsevier Ltd. All rights reserved.

International audience; Low back pain is a major public health problem in European Countries. In France, about 50% of population is suffering of this pathology every year [Fassier, 2011]. Because of health care cost and sick leave [Fassier, 2011], [Leclerc, 2009], low back pain as both societal and economic adverse consequences. Concerning medical devices, such as lumbar belts, several clinical trials have shown their efficiency [Calmels, 1999]. It might be essential to know which parameters are important so as to increase the efficiency of their medical effects on patients. Nevertheless, both mechanical and physiological effects of lumbar belts remain unclear.In this study, a parametrical method is used in order to identify which parameters are more important than others through numerical simulation. This study is indispensable to design lumbar belts.

An integrated optimisation methodology is proposed to optimise the manufacturing cost as well as the structural performance and the weight of composite laminated plates manufactured by the resin transfer moulding (RTM) process. In the present approach, the fibre type, the number of fabrics, the layer stacking sequence and the fibre volume fraction are optimised to minimise the structural weight and the material cost of composite structure under the stiffness constraint and the mould filling time constraint which is a part of process cycle time. With the results obtained, it is investigated how the weight and the material cost are traded-off. The optimisation methodology suggests a guide to cost-effective material selection in the preliminary conceptual design stage.

Sandwich materials are currently much valued in industry, especially in transport (automotive, aeronautics, shipbuilding and railroads) and civil engineering. Because of this rise of interest, it becomes more and more important to develop analysis tools able to take their specificities into account. Herein, we introduce a very simple and efficient post-processing method especially developed for heterogeneous materials such as sandwich structures. The method, based on Reissner principle, permits to fulfill the force equilibrium at interfaces between different layers, the skins and the core whose mechanical properties are very different, and yields accurate results for continuous stress components even with a coarse meshing through the thickness of the structure. Only small matrices are involved in the computation, leading to an easy programming and implementation even in existing finite element packages. Moreover, the problem caused by nodal extrapolation of stress components from Gauss points is removed. The present formulation, detailed in 2D for a simple illustration, is also valid in 3D. Finally, the method is assessed in the classical case of a simply supported beam under uniform pressure as well as in the non-standard case like a U-beam. Copyright © 2007 John Wiley & Sons, Ltd.

International audience; Une méthodologie d'optimisation est proposée en vue de minimiser la masse de la structure composite en tenant compte à la fois la performance mécanique et le coût de fabrication. Le couplage des paramètres structuraux et ceux du procédé de fabrication, en l'occurrence le procédé RTM dans cette étude, nécessite une optimisation simultanée de la conception structurale et du procédé de fabrication. Les interactions entre les paramètres sont identifiées et une nouvelle méthode d'optimisation est proposée pour déterminer l'ordre d'empilement, le nombre de plis, la localisation des points d'injection et l'épaisseur de la structure. Une nouvelle approche est proposée en vue de prendre en compte les critères de conception sans paramètres de pénalité, qui sont souvent utilisés pour des problèmes d'optimisation multicritère. Afin de faire face avec les difficultés dues à la variation du nombre de plis, quelques opérateurs génétiques comme le croisement et la mutation sont modifiés.

International audience; A simultaneous optimization methodology is proposed considering both the structural and the process requirements for the weight minimization of laminated plates manufactured by the Resin Transfer Molding (RTM) process. The coupling between the structural performance and the process cost is considered through the investigation of state parameters in the RTM process. As the structural constraint, maximum allowable displacement or maximum allowable failure index of the structure under load is assigned. For the process constraint, maximum allowable mold filling time is given. A new approach to deal with design constraints without having to concern about the penalty parameters is suggested. The influence of design constraints on the optimal design configuration is investigated through parametric studies. As the optimization scheme, Genetic Algorithm (GA) is employed. To take into account the layer number variation, genetic operators such as the crossover and the mutation are modified.

International audience; In this article, nite elements are developed on the basis of the displacements and the Pian and Tong functionals with help of Lagrange multipliers. These elements represent sandwich structure in an accurate way, especially at interfaces, where the equilibrium state is satisfied. Developed elements are assessed and compared on several examples of static linear problems.

International audience; In this paper, we present a simple and efficient post-processing method in finite element computations for the analysis of sandwich structures. This method is based on Reissner's functional which is applied locally at interfaces between layers those mechanical properties are very different. The method, valid in 2D as well as in 3D, is assessed in 2D for the computation of an U-type sandwich beam for which both the ratio of Young's modulus of the skin to that of the core and the radius of curvature vary. This example points out that the present method can be used for the design of complex structures based on finite element analysis owing to its numerical stability, i.e. its results remain accurate even using coarse meshes.

National audience; L'intérêt grandissant pour les matériaux sandwich implique de disposer d'outils d'analyse capables de prendre en compte leurs spécificités. Dans ce papier, nous introduisons une méthode de post-traitement simple et efficace spécialement développée pour les matériaux hétérogènes comme les structures sandwich. La méthode, basée sur le principe de Reissner, permet de vérifier l'état d'équilibre des forces aux interfaces entre les différentes couches et donne des résultats corrects pour les composantes continues des contraintes même pour des maillages grossiers suivant l'épaisseur de la structure. Cette méthode, détaillée en 2D mais valide en 3D, est illustrée dans le cas d'une poutre sandwich simplement supportée et soumise à une pression uniforme.

International audience; Les matériaux sandwich, pourtant couramment utilisés, souffrent encore d'un manque de précision dans leur modélisation, même en statique linéaire, et ce spécialement le long des interfaces entre les couches, où l'état d'équilibre des forces n'est généralement pas vérifié. Dans ce papier, contrairement aux méthodes généralement utilisées pour recouvrer les contraintes à partir des déplacements nodaux et basées sur un processus de minimisation de la différence entre les formulations en déplacements et mixte, nous proposons ici d'utiliser une méthode de post-traitement, basée sur le principe d'Hellinger-Reissner (Washizu, 1992), mais que nous n'appliquons que localement le long des interfaces, i.e. de modifier le calcul des contraintes, mais seulement sur un nombre très restreint d'éléments. La validation et l'efficacité de la méthode sont montrées par deux études sur des poutres sandwich en statique linéaire. Les résultats sont comparés aux solutions analytiques et EF d'ANSYS.

National audience; Nous présentons une série de modélisations par éléments finis qui représentent correctement le comportement mécanique des plaques sandwich, notamment la répartition des contraintes dans l'épaisseur. Les formulations présentées, valables en 2D comme en 3D, font intervenir des multiplicateurs de Lagrange le long des interfaces entre les couches. Ces multiplicateurs sont identifiés comme les composantes des contraintes devant être continues aux interfaces. Les couches sont modélisées indépendamment les unes des autres, tout en respectant leurs caractéristiques géométriques et matérielles. Finalement, les modèles présentés sont validés non seulement sur des structures sandwich, mais aussi sur des assemblages collés.

National audience; Ce papier présente une série d'études comparatives menées sur une poutre sandwich simplement supportée soumise à une pression uniforme à l'aide de différents éléments proposés dans ANSYS 5.2 : l'influence du raffinement du maillage ainsi que celle du rapport des modules d'Young entre les couches sont étudiées. À travers ces études, nous constatons une nécessité du développement d'une méthode de post-traitement, appelée Reissner local, qui permet d'améliorer la précision des contraintes aux interfaces pour les grands rapports entre modules d'Young des peaux et âme. La validité et l'efficacité de cette méthode est montrée au travers de ces études.

International audience; Cet article présente une courte revue des méthodes disponible pour le calcul des contraintes aux interfaces entre des matériaux ayant des propriétés mécaniques extrêment différentes : éléments finis classiques, mixtes ou mixtes hybrises, méthodes de post-processing.

International audience; As sandwich plates become more and more an important structural component, it would be essential to develop analysis tools taking their speci cities into acount. The present work concerns the development of hybrid sandwich nite elements modelling the mechanical behaviour of sandwich plates in shear in a more realistic way, and respecting physical phenomena at the interfaces between the skins and the core which have in general very heterogeneous mechanical properties. These elements are initially composed of three sub-elements in the same manner as the sandwich plate. The hybrid formulation has been retained because it permits to de ne, in a simple way, a modelling corresponding to all required specifications with the minimum number of variables. In the direction of the thickness, stress and displacement elds are both linear in each layer. They both are quadratic in the others directions to take exural e ects into account. We present in this paper several results and comparisons of static linear problems.

Numerical simulation for Resin Transfer Molding (RTM) manufacturing process is attempted by using the eXtended Finite Element Method (XFEM) combined with the level set method. XFEM allows to obtaining a good numerical precision of the pressure near the resin flow front, where its gradient is discontinuous. The enriched shape functions of XFEM are derived by using the level set values so as to correctly describe the interpolation with the resin flow front. In addition, the level set method is used to transport the resin flow front at each time step during the mold filling. The level set values are calculated by an implicit characteristic Galerkin FEM. The multi-frontal solver of IPSAP is adopted to solve the system. This work is validated by comparing the obtained results with analytic solutions.Moreover, a localization method of XFEM and level set method is proposed to increase the computing efficiency. The computation domain is reduced to the small region near the resin flow front. Therefore, the total computing time is strongly reduced by it. The efficiency test is made with simple channel or radial flow models. Several application examples are analyzed to demonstrate ability of this method. A wind turbine blade is also treated as industrial application. Finally, a Graphic User Interface (GUI) tool is developed so as to make easy the pre/post-processing of the simulation.; Le procédé de fabrication par RTM (Resin Transfer Molding) a été étudié numériquement à l’aide de la méthode XFEM (eXtended Finite Element Method) combinée avec la méthode Level set. La méthode XFEM permet d’obtenir une bonne précision numérique de la pression près du front d’écoulement, où son gradient est discontinu. Les fonctions de forme enrichies de la méthode XFEM sont proposées à l’aide des valeurs de Level set en vue de décrire correctement l’interpolation avec le front d’écoulement. En plus, la méthode de Level set est utilisée pour transporter le front d’écoulement à chaque pas de temps durant le remplissage du moule. Les valeurs de Level set sont calculées à l’aide d’une méthode de Galerkin implicite. Le solveur multi-frontal d’IPSAP a été utilisé pour la résolution du système. Cette étude a été validée en comparaison avec les solutions analytiques.En outre, une méthode de localisation avec XFEM et la méthode Level set a été proposée afin d’améliorer l’efficacité de calcul. Elle permet de réduire le domaine de calcul près du front d’écoulement. Par conséquent, le temps de calcul est fortement réduit grâce à cette méthode. Un test d’efficacité a été fait avec des modèles simples en écoulement laminaire ou radial.Quelques exemples d’application sont présentés pour illustrer la capacité de cette méthode. Une pale d’éolienne a également traitée comme application industrielle. Enfin, une interface d’utilisateur graphique a été développée en vue de fournir une facilité des pré- et post-processus.

Numerical simulation for Resin Transfer Molding (RTM) manufacturing process is attempted by using the eXtended Finite Element Method (XFEM) combined with the level set method. XFEM allows to obtaining a good numerical precision of the pressure near the resin flow front, where its gradient is discontinuous. The enriched shape functions of XFEM are derived by using the level set values so as to correctly describe the interpolation with the resin flow front. In addition, the level set method is used to transport the resin flow front at each time step during the mold filling. The level set values are calculated by an implicit characteristic Galerkin FEM. The multi-frontal solver of IPSAP is adopted to solve the system. This work is validated by comparing the obtained results with analytic solutions.Moreover, a localization method of XFEM and level set method is proposed to increase the computing efficiency. The computation domain is reduced to the small region near the resin flow front. Therefore, the total computing time is strongly reduced by it. The efficiency test is made with simple channel or radial flow models. Several application examples are analyzed to demonstrate ability of this method. A wind turbine blade is also treated as industrial application. Finally, a Graphic User Interface (GUI) tool is developed so as to make easy the pre/post-processing of the simulation., Le procédé de fabrication par RTM (Resin Transfer Molding) a été étudié numériquement à l’aide de la méthode XFEM (eXtended Finite Element Method) combinée avec la méthode Level set. La méthode XFEM permet d’obtenir une bonne précision numérique de la pression près du front d’écoulement, où son gradient est discontinu. Les fonctions de forme enrichies de la méthode XFEM sont proposées à l’aide des valeurs de Level set en vue de décrire correctement l’interpolation avec le front d’écoulement. En plus, la méthode de Level set est utilisée pour transporter le front d’écoulement à chaque pas de temps durant le remplissage du moule. Les valeurs de Level set sont calculées à l’aide d’une méthode de Galerkin implicite. Le solveur multi-frontal d’IPSAP a été utilisé pour la résolution du système. Cette étude a été validée en comparaison avec les solutions analytiques.En outre, une méthode de localisation avec XFEM et la méthode Level set a été proposée afin d’améliorer l’efficacité de calcul. Elle permet de réduire le domaine de calcul près du front d’écoulement. Par conséquent, le temps de calcul est fortement réduit grâce à cette méthode. Un test d’efficacité a été fait avec des modèles simples en écoulement laminaire ou radial.Quelques exemples d’application sont présentés pour illustrer la capacité de cette méthode. Une pale d’éolienne a également traitée comme application industrielle. Enfin, une interface d’utilisateur graphique a été développée en vue de fournir une facilité des pré- et post-processus.