Search

Confronted with an increasingly uncertain and fluctuating economic environment characterized by a technological evolution driven by industry 4.0 concepts, companies have to adapt their supply chains in order to improve its competitiveness. They must also implement strategies that allows them to deal with these new challenges. Supply chain modularity offers the opportunity to change and reorganize the structure of the supply chain to meet changing market needs. In this paper, we propose a new framework to design a modular supply chain using Design Structure Matrix and a modularity measurement in Reconfigurable Supply Chain. We also show the importance of integrating the direction of flows, i.e. inbound and outbound flows in the calculation of modularity and, in particular, its impact on lead time. Finally, an illustrative example is developed to validate the proposed measurement.

International audience; Market disruptions and changes make the assessment process of supply chain performance more complex. Hence, the implementation of key indicators to evaluate the ability of the supply chain to cope with these disruptions has received increasing attention. Reconfigurability provide the ability to quickly change supply chain structure and functions by increasing its responsiveness and flexibility to deal with disruptions. This paper aims to analyze the contributions of researchers in the field of performance evaluation under supply chain disruption. A bibliometric analysis is conducted to identify publications dealing with performance evaluation under supply chain disruption field. The analysis started with the selection of articles using the following keywords (Reconfigurability, Evaluation, Supply Chain). Then, the selected articles were analyzed based on the criteria of the bibliometric analysis. The paper aims to capture the current research and potentialresearch related to the evaluation of performance in reconfigurable supply chain under disruption.

International audience; Reconfigurable Manufacturing Systems (RMS) are designed to cope with mass customization problem. Reconfigurable Assembly Systems (RAS) are a variant of this RMS concept in the assembly field. One of the most crucial issues in designing the RAS is the layout problem with human safety considerations. This paper’s main motivation is to propose a methodology to determine the most efficient and safe arrangement of facilities in a workshop. The objective is twofold: (i) to minimize the Material Handling Cost and (ii) to ensure human safety by applying "safety bubble". An implementation of the approach and an illustrative case study are presented.

International audience; We consider the observation problem for a particular class of bidimensional systems with scalar output which requires the consideration of an embedding in higher dimension for the usual high-gain observer synthesis. We propose a new approach that does not require any coordinates transformation. This approach is based on the design of a set of estimators running in parallel in the same dimension than the original system. Each estimator uses the knowledge of the first two derivatives of the output, and the further derivatives up to the m-th one (where m is the observability index over an invariant domain) are used to discriminate at any time among the different estimators. We give two examples showing the applicability of this approach with measurement noise. Biological systems used in batch bioprocess models are of particular motivation for this work.

International audience; Consider the observation problem of bidimensional systems which requires the construction of an embedding in higher dimension. A multi-observers approach has been recently introduced by the authors to deal with a class of such singularly observable systems. This approach does not require any coordinates transformation. In this work we illustrate through an academic example the applicability of this approach.

Titanium (Ti) and its alloys machining has been a long standing issue in the manufacturing industry. The extraordinary Ti machining costs restrict its used in the specialised applications. To achieve improved surface quality, researchers are working tirelessly to optimize various cutting parameters. In this study, high-power CO2 laser cutting of the 2 mm thick sheet of pure Ti is investigated. Additionally, effect of the input laser conditions on the machine surface quality is also analysed. The arithmetic roughness Ra and total roughness Rt of the cut work-pieces, produced by different cutting speed (V), laser power (Pu) and gas pressure (p) are measured by the roughness tester. The machined or cut surfaces were analyzed at several locations along the cutting surface. The Taguchi method was applied to minimize the surface roughness of the cut parts and to optimize the cutting parameters. It was observed that surfaces roughness increases with the evolution of the thermal energy by laser power. However, surface roughness decreases with faster cutting speeds. Finally, the optimum cutting conditions for the arithmetic roughness (Ra) was recorded at laser power Pu of 2 kW, V of 2400 mm/min and p of 14 bars. For total roughness (Rt), Pu, V and p were 2 kW, 2400 mm/min and 2 bars respectively.

The purpose of this thesis is to introduce a new co-design Systems Engineering (SE) methodology that puts stakeholders at the center of the design process by assuming that the ME-CPS can be correctly designed and operated if all stakeholders reach mutual agreements that satisfy their intrinsic objectives. These agreements progressively emerge throughout an iterative negotiation process between stakeholders and are formalized using “formal contracts” which can be simulated for verification and validation all along the development process and before signature. Stakeholders will agree to sign contracts when they are sure that contracts fulfill their objectives and that they can commit to their obligations as specified in the contracts. Models are used to structure the relationships between stakeholders at all stages of the system’s lifecycle: formal contracts are derived from high-level goals and are verified using “digital twins” that are composed of requirement models and behavioral models., L’objectif de cette thèse est d’introduire une nouvelle méthodologie de co-conception basée sur l’ingénierie système qui place les parties prenantes au centre du processus de conception en supposant que les systèmes multi-énergies seront correctement conçus et exploités seulement si toutes les parties prenantes parviennent à des accords mutuels qui satisfont leurs objectifs intrinsèques respectifs. Ces accords émergent progressivement à travers un processus de négociation itératif entre les différentes parties et sont formalisés à l’aide de "contrats formels" qui peuvent être simulés pour faire de la vérification et de la validation tout au long du processus de conception et avant la signature de tout engagement. Les parties prenantes seront prêtes à signer des contrats dès lors qu’elles seront sûres que ces derniers remplissent leurs objectifs et qu’elles peuvent s’engager à respecter leurs obligations telles qu’elles sont énoncées dans les contrats. Des modèles sont utilisés pour structurer les relations entre les parties prenantes à tous les stades du cycle de vie des systèmes : les contrats formels sont dérivés d’objectifs de haut niveau et sont vérifiés à l’aide de "jumeaux numériques" qui sont composés de modèles d’exigences et de modèles comportementaux.

International audience; Bonding techniques are increasingly used in many industrial fields. Modeling the under-load damaging behavior of hard structural adhesives is still an open challenge. This work proposes a new hard interface analytical model with evolutive micro-cracking damage. The model is obtained within a rigorous theoretical framework combining asymptotic theory and micromechanical homogenization. Main new features are: (i) the adoption of two dual homogenization approaches; (ii) the formulation of a thermodynamically-based damage evolution law for hard interfaces. The interface model is able to describe both ductile and brittle damage behavior of hard structural adhesives. Provided examples on the structural behavior, under several loads, suggest the suitability of the proposed interface model as a modeling strategy for hard structural adhesives with micro-cracking damage.

The integration of mechatronic systems generates many multi-physical disturbances (thermal, electromagnetic and dynamic) making their choice of architecture complex. Indeed, the increasing introduction of electronic and electrical (E/E) components in most of the current systems, increases the risk of occurrence of many electromagnetic interferences (EMI) that can strongly degrade their behavior. While these electromagnetic compatibility (EMC) problems are usually addressed in the detailed design phase, where the possibilities of compromise are limited to a few positioning adjustments or costly protection solutions, one solution is to propose a collaborative framework for the evaluation, from the early design phases, of physical design architectures taking into account these electromagnetic (EM) constraints. Actually, it is important at this stage that all the multidisciplinary actors involved can define, modify/update, add their knowledge and constraints and exchange their data while continuing to work in their usual digital environment. In addition, these system engineering activities must be supported with "Model-Based System Engineering" (MBSE) approaches, to support the digital continuity, consistency and traceability of the models and data required for this evaluation process.To meet this need, this thesis is based on the MBSE SAMOS (Spatial Architecture based on Multi-physics and Organization of Systems) collaborative approach to support, from the preliminary design phase, the evaluation of the 3D concept architecture under electromagnetic constraints. In this context, we have initially developed a SysML extension called EMILE (ElectroMagnetic Interactions Layout Extension) to formalize and model, as early as possible, the EM constraints in the system model.This extension notably includes the definition of EM requirements, the description of electromagnetic coupling modes and the specification of simulation configurations allowing the further verification and validation of the requirements, thanks to the development of a Human-Machine Interface. Our research work then focused on an evaluation methodology combining a topological approach with EM modeling, in order to support the process of qualitative and quantitative evaluation of electromagnetic interference (EMI). Indeed, for a given type of EMI, the topological analysis of the system architecture allows to qualitatively identify the existence of the victim components and their associated potential aggressors. Once these potential EMIs have been identified, a quantitative evaluation can then be performed, for example based on the physical equations and laws of the identified coupling, and on the electromagnetic and geometric requirements predefined with the EMILE extension. As a result, this approach ensures the relevant choice of a 3D physical architecture of the concept under EM constraints. The proposed approaches have been illustrated on a case study of an electric vehicle power train, based on various software implementation scenarios (SysML, Modelica, Matlab, FreeCAD) within the Sketcher 3D EM tool.; L’intégration des systèmes mécatroniques génère de nombreuses perturbations multi-physiques (thermiques, électromagnétiques et dynamiques) rendant leur choix d’architecture complexe. En effet, l’introduction croissante de composants électroniques et électriques (E/E) dans la plupart des systèmes actuels, augmente le risque d’occurrence de nombreuses interférences électromagnétiques (IEM) pouvant dégrader fortement leur fonctionnement. Tandis que ces problèmes de compatibilité électromagnétique (CEM) sont généralement traités en phase de conception détaillée, où les possibilités de compromis se limitent à quelques ajustements de positionnement ou à des solutions de protection coûteuses, une solution consiste à proposer un cadre collaboratif pour l’évaluation, dès les premières phases de conception, d’architectures physiques de concept prenant en compte ces contraintes électromagnétiques (EM). En effet, il est important à cette étape, que l’ensemble des acteurs multidisciplinaires impliqués puissent définir, modifier/mettre à jour, ajouter leurs connaissances et contraintes et échanger leurs données tout en continuant à travailler dans leur environnement numérique habituel. Par ailleurs, ces activités d’ingénierie système doivent être outillées avec des approches d’« Ingénierie Système basée sur des modèles » (MBSE), pour supporter la continuité numérique, la cohérence et la traçabilité des modèles et des données nécessaires à ce processus d’évaluation.Pour répondre à ce besoin, cette thèse s’appuie sur l’approche collaborative MBSE SAMOS (Spatial Architecture based on Multi-physics and Organization of Systems) afin de supporter l'évaluation de l'architecture 3D de concept en phase amont sous contraintes électromagnétiques. Dans ce contexte, nous avons, dans un premier temps, développé une extension SysML appelée EMILE (ElectroMagnetic Interactions Layout Extension) pour formaliser et modéliser, au plus tôt, les contraintes EM dans le modèle système.Cette extension inclut notamment la définition des exigences EM, la description des modes de couplage électromagnétiques et la spécification des configurations de simulation permettant la vérification et la validation ultérieures des exigences, grâce au développement d’une interface homme-machine. Notre travail de recherche a ensuite porté sur une méthodologie d’évaluation combinant une approche topologique avec une modélisation EM., afin de supporter le processus d'évaluation qualitative et quantitative des interférences électromagnétiques (IEM). En effet, pour un type d’IEM donné, l'analyse topologique de l’architecture système permet d'identifier qualitativement l'existence des composants victimes et de leurs agresseurs potentiels associés. Une fois ces IEM potentielles identifiées, une évaluation quantitative peut alors être réalisée, par exemple en se basant sur les équations et lois physiques du couplage identifié, et sur les exigences électromagnétiques et géométriques prédéfinies avec EMILE. Cette approche permet ainsi de garantir le choix pertinent d’une architecture physique 3D de concept sous contraintes EM. Les approches proposées ont été illustrées sur une étude de cas d'une chaîne de traction du véhicule électrique, en s’appuyant sur plusieurs scénarios d’implémentation logicielle (SysML, Modelica, Matlab, FreeCAD) au sein de l’outil Sketcher 3D EM.

The integration of mechatronic systems generates many multi-physical disturbances (thermal, electromagnetic and dynamic) making their choice of architecture complex. Indeed, the increasing introduction of electronic and electrical (E/E) components in most of the current systems, increases the risk of occurrence of many electromagnetic interferences (EMI) that can strongly degrade their behavior. While these electromagnetic compatibility (EMC) problems are usually addressed in the detailed design phase, where the possibilities of compromise are limited to a few positioning adjustments or costly protection solutions, one solution is to propose a collaborative framework for the evaluation, from the early design phases, of physical design architectures taking into account these electromagnetic (EM) constraints. Actually, it is important at this stage that all the multidisciplinary actors involved can define, modify/update, add their knowledge and constraints and exchange their data while continuing to work in their usual digital environment. In addition, these system engineering activities must be supported with "Model-Based System Engineering" (MBSE) approaches, to support the digital continuity, consistency and traceability of the models and data required for this evaluation process.To meet this need, this thesis is based on the MBSE SAMOS (Spatial Architecture based on Multi-physics and Organization of Systems) collaborative approach to support, from the preliminary design phase, the evaluation of the 3D concept architecture under electromagnetic constraints. In this context, we have initially developed a SysML extension called EMILE (ElectroMagnetic Interactions Layout Extension) to formalize and model, as early as possible, the EM constraints in the system model.This extension notably includes the definition of EM requirements, the description of electromagnetic coupling modes and the specification of simulation configurations allowing the further verification and validation of the requirements, thanks to the development of a Human-Machine Interface. Our research work then focused on an evaluation methodology combining a topological approach with EM modeling, in order to support the process of qualitative and quantitative evaluation of electromagnetic interference (EMI). Indeed, for a given type of EMI, the topological analysis of the system architecture allows to qualitatively identify the existence of the victim components and their associated potential aggressors. Once these potential EMIs have been identified, a quantitative evaluation can then be performed, for example based on the physical equations and laws of the identified coupling, and on the electromagnetic and geometric requirements predefined with the EMILE extension. As a result, this approach ensures the relevant choice of a 3D physical architecture of the concept under EM constraints. The proposed approaches have been illustrated on a case study of an electric vehicle power train, based on various software implementation scenarios (SysML, Modelica, Matlab, FreeCAD) within the Sketcher 3D EM tool., L’intégration des systèmes mécatroniques génère de nombreuses perturbations multi-physiques (thermiques, électromagnétiques et dynamiques) rendant leur choix d’architecture complexe. En effet, l’introduction croissante de composants électroniques et électriques (E/E) dans la plupart des systèmes actuels, augmente le risque d’occurrence de nombreuses interférences électromagnétiques (IEM) pouvant dégrader fortement leur fonctionnement. Tandis que ces problèmes de compatibilité électromagnétique (CEM) sont généralement traités en phase de conception détaillée, où les possibilités de compromis se limitent à quelques ajustements de positionnement ou à des solutions de protection coûteuses, une solution consiste à proposer un cadre collaboratif pour l’évaluation, dès les premières phases de conception, d’architectures physiques de concept prenant en compte ces contraintes électromagnétiques (EM). En effet, il est important à cette étape, que l’ensemble des acteurs multidisciplinaires impliqués puissent définir, modifier/mettre à jour, ajouter leurs connaissances et contraintes et échanger leurs données tout en continuant à travailler dans leur environnement numérique habituel. Par ailleurs, ces activités d’ingénierie système doivent être outillées avec des approches d’« Ingénierie Système basée sur des modèles » (MBSE), pour supporter la continuité numérique, la cohérence et la traçabilité des modèles et des données nécessaires à ce processus d’évaluation.Pour répondre à ce besoin, cette thèse s’appuie sur l’approche collaborative MBSE SAMOS (Spatial Architecture based on Multi-physics and Organization of Systems) afin de supporter l'évaluation de l'architecture 3D de concept en phase amont sous contraintes électromagnétiques. Dans ce contexte, nous avons, dans un premier temps, développé une extension SysML appelée EMILE (ElectroMagnetic Interactions Layout Extension) pour formaliser et modéliser, au plus tôt, les contraintes EM dans le modèle système.Cette extension inclut notamment la définition des exigences EM, la description des modes de couplage électromagnétiques et la spécification des configurations de simulation permettant la vérification et la validation ultérieures des exigences, grâce au développement d’une interface homme-machine. Notre travail de recherche a ensuite porté sur une méthodologie d’évaluation combinant une approche topologique avec une modélisation EM., afin de supporter le processus d'évaluation qualitative et quantitative des interférences électromagnétiques (IEM). En effet, pour un type d’IEM donné, l'analyse topologique de l’architecture système permet d'identifier qualitativement l'existence des composants victimes et de leurs agresseurs potentiels associés. Une fois ces IEM potentielles identifiées, une évaluation quantitative peut alors être réalisée, par exemple en se basant sur les équations et lois physiques du couplage identifié, et sur les exigences électromagnétiques et géométriques prédéfinies avec EMILE. Cette approche permet ainsi de garantir le choix pertinent d’une architecture physique 3D de concept sous contraintes EM. Les approches proposées ont été illustrées sur une étude de cas d'une chaîne de traction du véhicule électrique, en s’appuyant sur plusieurs scénarios d’implémentation logicielle (SysML, Modelica, Matlab, FreeCAD) au sein de l’outil Sketcher 3D EM.

For bipedal robots, how to realize natural, versatile and dynamic motions like human is still a challenging topic. In order to achieve such high performance by bipedal robots, one useful approach is to develop joint mechanism with high back-drivability, that makes the dynamic coupling forces such as inertia force, centrifugal force available for dynamic motions and results reduction of necessary capacity for joint actuators leading to the reduction of weight of the robot. For this purpose, in this paper, we propose a novel knee joint mechanism consisting of a crossed four-bar linkage with variable transmission (CFLVT) and a joint stop inspired from human structure. CFLVT behaves as a transmission with variable reduction ratio according the knee joint angle to smooth the static torque so that the knee joint can be driven by a small actuator with low reduction ratio. In addition, a joint stop at the maximum extending angle is introduced in the knee joint which is useful for force assist just like that of a human. The characteristic and design of CFLVT and joint stop are described first. And a prototype of the bipedal robot implemented with CFLVT and joint stop is developed. The feasibility of the proposed mechanism is shown by experimental results for walking and kicking motions.

International audience; Sound attenuation performances of a duct silencer made of a metamaterial composed of a set of poroelastic lamellas, are investigated numerically and experimentally. The lamella arrangement is designed so that skeleton elastic resonances appear at low frequency. In addition, the air gap between the lamellas induces pressure diffusion effects commonly encountered in highly contrasted double porosity media. The configuration of the silencer also exhibits strong anisotropy. A 3D finite element model of the silencer has been developed in order to identify the different mechanisms taking place in the silencer. Numerical results are shown to match well with experimental data for different lamella orientations. By exploiting the geometrical periodicity of the silencer, a Bloch waves analysis is realized and a parametric study is conducted. This reveals strong model dispersion and interaction due to lamella elastic resonances. It is shown that certain design parameters can be tuned in order to achieve large attenuation and absorption at low frequency.

Structural materials should be stronger, tougher, and lighter to withstand the extreme conditions in the aeronautic and automotive applications. In this work, alternative intermetallic based composites (IMBCs) were designed from Mgx (1-x) Ti-Al recycled intermetallic alloy reinforced with the different percentages of CNT/Al2O3 fiber and Boron/Al2O3 fiber. Aluminum powder (~ 5 wt %) obtained from fresh scrap was added to the compositions to improve the homogenization of the mixture. Then, final composites were produced through sintering, combined method of sintering followed by forging and/or sintering followed by thixoforming which are low cost and efficient methods to manufacture light, multifunctional materials. Two different composites were designed by using these manufacturing processes: First group was considered as matrix +0.5 wt % CNT +1% Al2O3 fiber and the second group was designed as matrix +0.5wt % Boron +1% Al2O3 fiber. Quasi-static compression tests, nanoindentation (wear, modulus, hardness) test were performed to study the effect of the reinforcements and the manufacturing methods. The microstructure analyses (matrix/interface) have been carried out by Scanning Electron Microscope (SEM).

Aluminum based hybrid composites were produced from recycled AA7075 chips with the addition of TiC (d ≤ 3–5 micron), MoS2 and Al2O3 fiber. In the two groups of composites produced, the content of MoS2 and Al2O3 were fixed as 2 wt % and 3 wt % respectively, whereas TiC content was at two levels (5–10%). The combined method of powder metallurgy route, sintering followed by forging, was used to manufacture these composites. These composites are targeted for aeronautical and automotive industries for components subjected to static as well as cyclic and dynamic loading. In addition to mechanical properties, machinability of these composites is of importance hence, MoS2 was included in the formulation. Micro hardness, 3 point bending, low velocity impact and nanoindentation (creep and wear) tests were performed on samples manufactured by just sintering and sintering followed by forging. The results showed that, in generals, the samples that were forged after sintering yielded better properties. The microstructure analyses (matrix/interface) have been carried out by Scanning Electron Microscope (SEM).

This work aims at the unification of the thermodynamically consistent representationof the micromorphic theory and the microdamage approach for the purpose of modeling crack growthand damage regularization in crystalline solids. In contrast to the thermodynamical representationof the microdamage theory, micromorphic contribution to flow resistance is defined in a dual fashionas energetic and dissipative in character, in order to bring certain clarity and consistency to themodeling aspects. The approach is further extended for large deformations and numericallyimplemented in a commercial finite element software. Specific numerical model problems arepresented in order to demonstrate the ability of the approach to regularize anisotropic damage fieldsfor large deformations and eliminate mesh dependency.

International audience

International audience

In this paper, we propose a new method to measure the vibrations of a 3D mechanical structure. It requires only a video sequence taken by a single camera and a Finite Element model of the structure. First, the intrinsic and extrinsic parameters of the camera are identified. Then, a collection of numerical deflection shapes and normal modes are projected in the image frame of the camera. This is done thanks to a linearization of the perspective projection model. Comparing the motion of targets seen by the camera and the motion of Finite Element model at the same location gives the time evolution of modes amplitude. Finally the motion of the model is reconstructed in 2D in the image frame and in 3D in the world frame. By following this procedure, vibration amplitude can be magnified applying a scale factor on a modal basis covering a large frequency bandwidth. A video graphical abstract can be found on youtube, see [1].This paper is the preprint of the MSSP article that can be found at https://doi.org/10.1016/j.ymssp.2021.108032

International audience; The diagnosis procedure in satellite operations is constrained by strict time limits. If a failure occurs, operators must be efficient and proactive, as diagnosis must be completed within the smallest number of visibility windows (pass duration over the ground station which are usually very short and separated by long periods where the ground cannot communicate with the satellite). Improving the diagnosis procedure and tools, time and precision-wise, is therefore essential. The proposal presented in this position paper is to provide the operators with operation-dedicated models to help them in reacting quickly and in finding a suitable repair solution as fast as possible. These models will gather system architecture, functional and dysfunctional data taken from system engineering and safety analysis models. The paper presents the framework of the proposed solution and discusses different implementation options.

One of the most critical issues that can affect the useful life of a product is the obsolescence of its components or functionalities. To minimize its effects and bring long-term benefits to systems, obsolescence must be proactively managed. A critical step in the process of proactive obsolescence management is the obsolescence risk analysis of critical components or functionalities. However, estimating the degree of obsolescence of multi-component systems is an area that is still under-explored, particularly when interdependencies exist between components. This estimation can be more complicated where there is no prior knowledge about the interaction between components. We used Weibull’s distribution to model the components’ interaction and calculate the obsolescence degree of the global system. This approach is evaluated using a numerical example based on a meteorological data acquisition system. The obsolescence of main components of this system is modeled while taking into account the interaction between them. Results are presented and discussed while clarifying the scientific issues that should be tackled in future works.

In this work, the effect of a turning process on fatigue performance of a Ti-6.5Al-3.5Mo-1.5Zr-0.3Si (TC11) titanium alloy is studied in the high cycle fatigue (HCF) and very high cycle fatigue (VHCF) regimes. For this purpose, the surface characteristics including surface morphology, surface roughness and residual stress were investigated. Moreover, axial fatigue tests were conducted with an ultrasonic fatigue testing system working at a frequency of 20 kHz. The results show that the turning process deteriorated the fatigue properties in both HCF and VHCF regimes. The fatigue strength at 1 &times, 108 cycles of turned samples is approximately 6% lower than that of electropolished ones. Fracture surface observations indicate that turning marks play a crucial role in the fatigue damage process, especially in the crack initiation stage. It was observed that the crack of all the turned samples originated from turning marks. In addition, the compressive residual stress induced by the turning process played a more effective role in resisting crack propagation in the VHCF regime than in the HCF regime.

International audience

International audience; With the increasing complexity of systems, engineers have to design an increasing number of models to perform simulation of the product. In this work we intend to compare the system engineering and safety models of a system, and establish a typology of the differences between those models.; Avec l'explosion de la complexité des systèmes, les ingénieurs doivent concevoir un grand nombre de modèles afin de les représenter et simuler. Dans cet article, nous comparons les modèles d'ingénierie système et de sûreté de fonctionnement d’un système et établissons une typologie des différences entre ces modèles.

This thesis is part of work related to the problems of assembled structures. After an analysis and a synthesis of the different modeling scales involved in order to determine the damping in the joints, the manuscript highlights the scales changes, i.e. model reductions. Many numerical methods are used to solve nonlinear vibration problems. The framework of the thesis being steady-state vibrations, the Harmonic Balance Method is commonplace. Here it is coupled with an original fixed point algorithm. Depending on the case study, three resolution paths are proposed. If we know everything about the behavior, the study of the complete structure can be summarized to the resolution of a differential system. The question is "how to solve it efficiently?" A comparison of four different formulations of the same problem in the time and frequency domains, with or without the regularization of hysterical forces, provides answers to this question. If this is not possible or unreasonable, then one must try to decompose the problem. One way to speed up the process is to reduce the model. For this purpose a new basis for reducing the non-linear part is introduced. Its construction is based on an energy indicator and its use is based on a chart. Finally, if the construction of this chart is impossible, it is then necessary to have a complete computation with sequential generations of dynamic charts of the sub-structure. This adaptive methodology alternates the time and frequency resolutions respectively on non-linear and linear domains in a non incremental way., Cette thèse s'inscrit dans des travaux liés aux problématiques des structures assemblées. Après une analyse et une synthèse des différentes échelles de modélisation mises en jeu afin de déterminer l'amortissement dans les liaisons, le manuscrit met en évidence les changements d'échelles, soient les réductions de modèle. La résolution des problèmes de vibration non-linéaire mettent en jeu de nombreuses méthodes numériques. Le cadre de la thèse étant le régime permanent, l'équilibrage harmonique est la méthode clé, couplée à une résolution originale de type point fixe. En fonction du cas d'étude, trois pistes de résolution sont proposées. Si on connait tout du comportement, l'étude de la structure complète se résume à la résolution d'un système différentiel. La question est "comment le résoudre de manière efficace ?" Une comparaison de quatre formulations différentes du même problème dans les domaines temporel et fréquentiel, avec ou sans régularisation des forces hystérétiques, permet d'apporter des réponses à cette question. Si cela n'est pas possible ou déraisonnable, il faut alors essayer de décomposer le problème. Une des solutions afin d'accélérer le processus est de réduire le modèle. Pour cela une nouvelle base de réduction de la partie non-linéaire est introduite. Sa construction prend appui sur un indicateur énergétique et son utilisation sur un abaque. Enfin, si la construction de cet abaque est impossible, il est alors nécessaire d'avoir un calcul complet avec générations séquentielles d'abaques dynamiques de la sous-structure. Cette méthodologie adaptative alterne les résolutions temporelle et fréquentielle respectivement sur les domaines non-linéaire et linéaire sans recours à une méthode incrémentale type AFT.

This thesis is part of work related to the problems of assembled structures. After an analysis and a synthesis of the different modeling scales involved in order to determine the damping in the joints, the manuscript highlights the scales changes, i.e. model reductions. Many numerical methods are used to solve nonlinear vibration problems. The framework of the thesis being steady-state vibrations, the Harmonic Balance Method is commonplace. Here it is coupled with an original fixed point algorithm. Depending on the case study, three resolution paths are proposed. If we know everything about the behavior, the study of the complete structure can be summarized to the resolution of a differential system. The question is "how to solve it efficiently?" A comparison of four different formulations of the same problem in the time and frequency domains, with or without the regularization of hysterical forces, provides answers to this question. If this is not possible or unreasonable, then one must try to decompose the problem. One way to speed up the process is to reduce the model. For this purpose a new basis for reducing the non-linear part is introduced. Its construction is based on an energy indicator and its use is based on a chart. Finally, if the construction of this chart is impossible, it is then necessary to have a complete computation with sequential generations of dynamic charts of the sub-structure. This adaptive methodology alternates the time and frequency resolutions respectively on non-linear and linear domains in a non incremental way.; Cette thèse s'inscrit dans des travaux liés aux problématiques des structures assemblées. Après une analyse et une synthèse des différentes échelles de modélisation mises en jeu afin de déterminer l'amortissement dans les liaisons, le manuscrit met en évidence les changements d'échelles, soient les réductions de modèle. La résolution des problèmes de vibration non-linéaire mettent en jeu de nombreuses méthodes numériques. Le cadre de la thèse étant le régime permanent, l'équilibrage harmonique est la méthode clé, couplée à une résolution originale de type point fixe. En fonction du cas d'étude, trois pistes de résolution sont proposées. Si on connait tout du comportement, l'étude de la structure complète se résume à la résolution d'un système différentiel. La question est "comment le résoudre de manière efficace ?" Une comparaison de quatre formulations différentes du même problème dans les domaines temporel et fréquentiel, avec ou sans régularisation des forces hystérétiques, permet d'apporter des réponses à cette question. Si cela n'est pas possible ou déraisonnable, il faut alors essayer de décomposer le problème. Une des solutions afin d'accélérer le processus est de réduire le modèle. Pour cela une nouvelle base de réduction de la partie non-linéaire est introduite. Sa construction prend appui sur un indicateur énergétique et son utilisation sur un abaque. Enfin, si la construction de cet abaque est impossible, il est alors nécessaire d'avoir un calcul complet avec générations séquentielles d'abaques dynamiques de la sous-structure. Cette méthodologie adaptative alterne les résolutions temporelle et fréquentielle respectivement sur les domaines non-linéaire et linéaire sans recours à une méthode incrémentale type AFT.

The aim of this paper is to investigate the effect of variant sandblasting parameters such as pressure, angle and sand granulometry on galvanized steel surface chemical composition, roughness and free energy for the purpose of increasing its adhesion properties for further use in structural bonding applications. Two sandblasting pressures (5 × 10 5 Pa, 6 × 10 5 Pa), two sandblasting angles (75°, 90°), and three glace sand granulometries (200/300, 90/150, 70/110) were tested. Surface chemical composition was evaluated using a scanning electron microscope, surface roughness was measured using a mechanical surfscan and surface free energy was determined using the contact angle method along with the two component Owens-Wendt model. Trials revealed that the sandblasting method has an effect on the surface free energy, wettability, chemical composition and roughness. As the sandblasting pressure and glace sand granulometry increased, the surface average roughness (Ra) increased. Specimens treated with the lowest glace sand granulometry had the highest surface free energy ( γ s t ). Glace sand granulometry had a more significant effect on the galvanized steel surface properties than that of the sandblasting pressure and angle. Sandblasting angle variation affected the surface chemical composition and wettability rather than the roughness. Sand residue presented at the level of the treated surface varied according to the sandblasting setting which might have affected the surface free energy and wettability along with the roughness.

A new control methodology derived from higher order sliding mode control (HOSMC) strategy has been recently presented. It ensures high accuracy and requires reduced control effort and chattering; however, only finite time convergence has been proven. In this paper, this controller is revisited and domains of convergence are given for some cases. The second objective is to implement this controller to produce the maximal amount of energy that can be produced by a twin wind turbines (TWT) system. The performance of the proposed control strategy is evaluated and compared to HOSMC.

A multibody system can be considered as an assembly of interconnected components: flexible bodies, rigid bodies, joint and unit control. This paper describes an analytical methodology for the predesign of a multibody system mixing rigid and flexible structure and regarding the vibrational aspect. Indeed, one of the principal aspects that limit performance in mechanical systems is the presence of vibrations. This contribution is based on the object-oriented modelling with Modelica language. Structural engineers frequently encountered problems arising from deflection of flexible structure such as plate and beam under moving load. In this paper, we study the interaction between a dynamic exciter moving with a constant velocity over a flexible beam resting on an elastic foundation by using the theory of dynamic response of Euler Bernoulli beam. The study of the moving load is of great importance in the transportation field. This methodology will be illustrated to the railway system. In fact, the dynamic response of a railway track subjected to a moving train load may be simplified as a beam traversed by a moving load. An analytical based approach is considered for the dynamic simulation of the flexible multibody system. The effect of some parameters on the dynamic response of the system has been studied.

International audience; The concept of healthy twin wind turbines isrecently studied in the literature. Control in the dq-frame,based on the symmetric of the healthy system, has been proposed.Nevertheless the robustness of such control with respect toasymmetric faultshas not been studied. In this paper, is proposeda new asymmetric model in abc-frame, taking into considerationan electrical fault. The healthy control is tested in theframework of the faulty model, and simulations show that somerobustifications are necessary.

International audience; A numerical method is proposed to approximate the solution of parametric eigenvalue problem when the variability of the parameters exceed the radius of convergence of low order perturbation methods. The radius of convergence of eigenvalue perturbation methods, based on Taylor series, is known to decrease when eigenvalues are getting closer to each other. This phenomenon, knwon as veering in structural dynamics, is a direct consequence of the existence of branch point singularity in the complex plane of the varying parameters where some eigenvalues are defective. When this degeneracy, referred to as Exceptional Point (EP), is close to the real axis, the veering becomes stronger. The main idea of the proposed approach is to combined a pair of eigenvalues to remove this singularity. To do so, two analytic auxiliary functions are introduced and are computed through high order derivatives of the eigenvalue pair with respect to the parameter. This yields a new robust eigenvalue reconstruction scheme which is compared to Taylor and Puiseux series. In all cases, theoretical bounds are established and all approximations are compared numerically on a three degrees of freedom toy model. This system illustrate the ability of the method to handle the vibrations of a structure with a randomly varying parameter. Computationally efficient, the proposed algorithm could also be relevant for actual numerical models of large size, arising from other applications involving parametric eigenvalue problems, e.g., waveguides, rotating machinery or instability problems such as squeal or flutter.

This paper investigates high power CO2 laser cutting of 5mm-thick Ti-6Al-4V titanium alloy sheets, aiming to evaluate the effects of various laser cutting parameters on surface roughness. Using multiple linear regression, a mathematical model based on experimental data was proposed to predict the maximum height of the surface Sz as a function of two laser cutting parameters, namely cutting speed and assist-gas pressure. The adequacy of the proposed model was validated by Analysis Of Variance (ANOVA). Experimental data were compared with the model’s data to verify the capacity of the proposed model. The results indicated that for fixed laser power, cutting speed is the predominant cutting parameter that affects the maximum height of surface roughness.

Recycled AA7075 aluminum alloy and pure electrolytic copper were used as the matrix to manufacture new aluminum-copper metal matrix composites (ACMMCs). Powder metallurgy methods were used and the green compacts were finalized by sintering only and sinter + forging. Experimental and numerical investigation of recycled hybrid metal matrix composites manufactured by the two methods was performed. Al–Cu matrix combined with the reinforcements ZnO, Nb2Al and SiC. Two basic formulations were used where the contents Nb2Al and SiC was kept constant and the content of ZnO were 15 wt% and 30 wt%, respectively. The effects of these reinforcements used in the hybrid metal matrix composite structure on the mechanical and physical properties were investigated. The Nb2Al, SiC ratios used in the structure (chemical interaction in the internal structure, the effects on static-dynamic compression and wear behaviors) were kept constant, in particular the ZnO component (interactions with other components and their effects on electrical properties) were investigated. Micro-hardness analyses, surface scratch tests, quasi-static and dynamic compression tests were conducted. Also, electrical conductivity of the composites were determined. The effect of the composite's formulations and production method on the results were investigated. It was found when ZnO content was reduced the yield stress and ultimate strength values increased, but their resistance to impact loading reduced. Also, sinter + forged samples exhibited higher yield stress and ultimate strength than the just sintered samples. The damage and microstructural analyses were performed by Scanning Electron Microscope (SEM). Moreover, a non-linear numerical model was utilized to simulate quasi-static compression and dynamic compression (low velocity impact) behaviors of the composites for both formulations and manufacturing methods. Finite element simulations were performed using the ABAQUS™/Explicit dynamic finite element software. It was determined that there was a satisfactory agreement between experimental and finite element model results.

Companies always search for alternative solutions for designing mechatronic systems in more efficient ways to cope with the high-performance standards, competitive market, and design time reduction. In this article, we present a complete multiphysical approach to design and validate an electronic throttle body (ETB) system in a concurrent way. Our goal is to reduce the modeling and test efforts by using an enriched multiphysical model in the preliminary phase. This model is the main stone that connects different fields and ensures efficient collaboration. This article is the result of a close collaboration between laboratories and industrials in the automotive field and aims at developing a methodology that takes into account collaboration constraints during design. Thanks to this approach, industrial requirements of the ETB are achieved at an early stage. An experimental test bench is presented to validate the simulation results.

International audience

Part 7: New Product Development; International audience; Obsolescence issues are one of the main costs in the life-cycle of sustainment-dominated systems, those that require support for many decades.Obsolescence not only occurs when a system element becomes no longer available due to manufacturing updates or production interruption. It also includes the deterioration of the system or component capacity to operate as intended, since it is no longer suitable to fulfil its function, even if it still operates and can be manufactured and supported. As such, it also impacts on the “-ilities” (reliability, functionality, etc. …) which can be even traced to legislation changes, for instance, the anti-pollution legislation updates for automotive.Obsolescence has so far been mainly addressed from a reactive point of view. Around 70% of the total product cost is committed as a result of decisions in the early design stages. Late changes induce delays in the life-cycle plan and large increases in cost due to re-design and rework. Knowing the impact of the conceptual design phase on life-cycle cost, this paper intends to present a model-based systems engineering approach to proactively assess obsolescence risks.The approach uses obsolescence considerations as constraints during the early design phases, to allow the identification of components potentially at risk of obsolescence. Presenting designers with this information allows them to improve the design, by making it robust and resilient, or to accept the risk and develop an obsolescence mitigation plan.It will be illustrated with an example from the automotive industry.

For a mechatronic system, the lack of robustness is mainly due to an ignorance of variability which is unavoidable in every designed system at all stages of its development and during its entire life cycle. In fact, the inherent imperfections of manufacturing processes and operational variations such as material wear or multi-physical effects (thermal, vibrations, etc.) involve parametric variations, which can degrade the proper system function. To ensure a high level of quality and to improve design robustness, the deviations between actual and target definition should be restricted by specified tolerances. Moreover, it is important to understand the relationship between parameters deviations and system behavior. Thus, in this thesis, a variational approach is proposed to specify the admissible parameters deviations for which the given specifications can be guaranteed. Furthermore, this approach is used to identify performances sensitivity to parameters deviations in a mechatronic system and to determine influential parameters that affect significantly the proper functioning of the system and its behavior. The complete approach is applied to an example in the aeronautic field: an electromechanical actuator driving an aircraft primary flight control surface (aileron)., Pour un système mécatronique, le manque de robustesse est principalement dû à une ignorance de la variabilité qui est inévitable dans tout système conçu, à toutes les étapes de son développement et tout au long de son cycle de vie. En effet, les imperfections inhérentes aux processus de fabrication et les variations en fonctionnement telles que l'usure des matériaux ou les effets physiques multiples (thermiques, vibrations, etc.) impliquent des variations paramétriques susceptibles de dégrader le bon fonctionnement du système. Pour garantir un niveau de qualité élevé et pour améliorer la robustesse de la conception, les déviations entre les valeurs réelles et les valeurs nominales des paramètres et des variables doivent être limitées par des tolérances bien spécifiées.Pour ce faire, il est important de comprendre la relation entre les écarts des paramètres et le comportement du système. Ainsi, dans cette thèse, une approche variationnelle est proposée pour spécifier les tolérances admissibles des paramètres pour lesquels le comportement reste conforme aux spécifications. De plus, cette approche est utilisée pour identifier la sensibilité des performances du système aux écarts des paramètres dans un système mécatronique. Cette analyse de sensibilité vise à déterminer les paramètres agissant de manière significative sur le bon fonctionnement du système et son comportement. L’approche proposée est appliquée à un exemple dans le domaine de l’aéronautique : un actionneur électromécanique entraînant une surface de contrôle de vol primaire d’un aéronef (aileron).

For a mechatronic system, the lack of robustness is mainly due to an ignorance of variability which is unavoidable in every designed system at all stages of its development and during its entire life cycle. In fact, the inherent imperfections of manufacturing processes and operational variations such as material wear or multi-physical effects (thermal, vibrations, etc.) involve parametric variations, which can degrade the proper system function. To ensure a high level of quality and to improve design robustness, the deviations between actual and target definition should be restricted by specified tolerances. Moreover, it is important to understand the relationship between parameters deviations and system behavior. Thus, in this thesis, a variational approach is proposed to specify the admissible parameters deviations for which the given specifications can be guaranteed. Furthermore, this approach is used to identify performances sensitivity to parameters deviations in a mechatronic system and to determine influential parameters that affect significantly the proper functioning of the system and its behavior. The complete approach is applied to an example in the aeronautic field: an electromechanical actuator driving an aircraft primary flight control surface (aileron).; Pour un système mécatronique, le manque de robustesse est principalement dû à une ignorance de la variabilité qui est inévitable dans tout système conçu, à toutes les étapes de son développement et tout au long de son cycle de vie. En effet, les imperfections inhérentes aux processus de fabrication et les variations en fonctionnement telles que l'usure des matériaux ou les effets physiques multiples (thermiques, vibrations, etc.) impliquent des variations paramétriques susceptibles de dégrader le bon fonctionnement du système. Pour garantir un niveau de qualité élevé et pour améliorer la robustesse de la conception, les déviations entre les valeurs réelles et les valeurs nominales des paramètres et des variables doivent être limitées par des tolérances bien spécifiées.Pour ce faire, il est important de comprendre la relation entre les écarts des paramètres et le comportement du système. Ainsi, dans cette thèse, une approche variationnelle est proposée pour spécifier les tolérances admissibles des paramètres pour lesquels le comportement reste conforme aux spécifications. De plus, cette approche est utilisée pour identifier la sensibilité des performances du système aux écarts des paramètres dans un système mécatronique. Cette analyse de sensibilité vise à déterminer les paramètres agissant de manière significative sur le bon fonctionnement du système et son comportement. L’approche proposée est appliquée à un exemple dans le domaine de l’aéronautique : un actionneur électromécanique entraînant une surface de contrôle de vol primaire d’un aéronef (aileron).

In this paper, a control based on singular perturbation approach is designed to solve the problem of power management for a serial hybrid electric vehicle. In this case, the power supply system consists of an Internal Combustion Engine (ICE) mechanically bound to a Permanent Magnet Synchronous Machine (PMSM) through a flywheel. Both sources are torque-controlled and connected to the traction motor by simple (without battery) DC link. The contribution of this paper is to apply the singular perturbation approach to design a control, which takes into account the different dynamic of the considered power topology. The proposed strategy is based on two-closed loops cascade control (fast and slow), which allows to impose the constraints of slower (faster respectively) ICE torque dynamics (PMSM current dynamics respectively) and to protect the power devices.

With the ever increase of the complexity and the importance of industrial systems, many research works are looking for developing different techniques that allowto detect, locate and identify any abnormalities in the system as early as possible. In this thesis, we present a new diagnosis method, called sparse recovery, that allows to diagnose several faults in dynamical systems. The sparse recovery method is based on a dynamical algorithm that estimates a sparse fault vector from few system measurements. The term sparse means that many faults can be considered but only few of them can occur simultaneously. Those faults will be called "activated node" and their number will be denoted by s. From a theoretical point of view, the faults number s depends on the available information given by the sensor measurements and on the condition that 2s Å1 must be lower or equal than the number of measurements. Other theoretical conditions of the sparse recovery method will be discussed in the manuscript. In order to illustrate the performances of the proposed diagnosis method, we apply it to a reduced gearbox model. Different defects that disrupt its operating are considered, for example mechanical defects such as eccentricity, cracks and sensor defects (bias). The diagnosis of these defects can be carried out based on the knowledge of some characteristics of the system (e.g., temporal and frequency characteristics of the gear element). In this thesis, it is shown that with the help of the sparse recovery method, it is often possible to detect the presence of the fault(s) and it(s) time of appearance and to locate the defective element(s) as well as to identify it(s) magnitude(s). Sparse recovery is an effective method for the diagnostic of dynamical systems, however, it present some limitations related to the number of defects that may occur simultaneously. This number of simultaneous faults depends on the dimension of the measurement vector, in other words, it depends on the number of sensors placed in the system. If the number of sensors is large then the possiblility to detect faults increases. However, the installation of many sensors is often very expensive. Thus, in order to reduce the number of sensors, the estimation of state variables of the dynamical system using observer’s techniques is a good solution that provides a trade-off between the financial impact and the capacity to monitor the dynamical behaviour of systems. Theimplementation of observers becomes more and more complex as the complexity of the system increases, namely in the case of mechatronic systems that merging different physical quantities. In this thesis, we consider the homogeneous differentiator with variable exponent. We apply several differentiators in order to estimat system states and analyze the dynamical behavior of a simple stage gear driven by an asynchronous machine. To the best of our knowledge, it is the first time that this differentiator type is applied to mechatronic system. This mechatronic mechanism is widely used in many industrial applications. An important problem of the modelization step is the coupling between the mechanical part and the electrical part.Many research works present several ideas of the coupling between the asynchronous motor and the gear element. In this thesis, a new coupling of the system is presented. It takes into account the transient state of the electrical part and the torsional effect of a simple stage gear element., Avec l’augmentation continue de la complexité des systèmes industriels et de l’importance à maintenir leurs sécurité, de nombreux travaux de recherche visent à développer différentes méthodes permettant de détecter, localiser et identifier les anomalies d’un système le plus tôt possible. Dans ce contexte, nous présentons une nouvelle méthode de diagnostic, appelée la reconstruction parcimonieuse, afin d’analyser et diagnostiquer différents défauts intervenants dans des systèmes dynamiques. La méthode de la reconstruction parcimonieuse est fondée sur un algorithme dynamique qui estime un vecteur de défaut parcimonieux à partir de quelques mesures du système. Le terme "parcimonieux" signifie que plusieurs défauts sont pris en compte dans la modélisation, mais seulement un nombre restreint peut se produire simultanément. Ces défauts seront appelés "nœuds ou défauts actifs" et leur nombre sera noté s. D’un point de vue théorique, le nombre de défauts s dépend des informations disponibles provenant des capteurs (des mesures) et de la condition que 2s+1 soit inférieur ou égal au nombre de mesures. D'autres conditions théoriquesde cette méthode de diagnostic seront discutées dans ce manuscrit. Afin d’illustrerles performances de la méthode de diagnostic proposée, nous l’appliquons à un modèlesimplifié d’un système de transmission par engrenages. Différents défauts qui perturbent son fonctionnement sont pris en compte, à savoir des défauts mécaniques tels que défaut d’excentricité et défaut de fissure et des défauts de capteurs (biais). Le diagnostic de ces défauts est effectué en se basant sur des connaissances de certaines caractéristiques du système (par exemple les caractéristiques temporelles et fréquentielles de l’élément d’engrenages).Dans cette thèse, il est montré qu’à l’aide de la reconstruction parcimonieuse,il est souvent possible de détecter la présence de(s) défaut(s) ainsi que son(leurs) temps d’apparition et de localiser le (les) élément(s) défectueux ainsi que d’identifier son (leurs) amplitude(s).La reconstruction parcimonieuse permet le diagnostic en ligne des défauts dans lessystèmes dynamiques mais elle présente également certaines limitations liées au nombre de défauts qui se produit simultanément. Ce nombre dépend de la dimension du vecteur de mesures, autrement dit, il dépend du nombre de capteurs placés dans le système. Si ce nombre est important, alors la possibilité de détecter des défauts augmente mais l’installation de nombreux capteurs est souvent très coûteuse. Afin de réduire le nombre de capteurs utilisés, l’estimation d’états par la technique des observateurs est une solution privilégiée qui assure un compromis entre l’impact financier et la capacité à prédire le comportement dynamique des systèmes. L’implémentation des observateurs devient de plus en plus délicate lorsque la complexité du système augmente, notamment dans le cas des systèmes mécatroniques regroupant différentes grandeurs physiques. Dans cette thèse, nous considérons le différenciateur homogène avec un exposant variable. Nous appliquons plusieurs différenciateurs pour estimer les états du système et analyser le comportementdynamique d’une transmission par engrenages simple étage entraînée par unmoteur électrique. A notre connaissance, c’est la première fois que ce type d’observateur est appliqué à un système mécatronique. Ce dispositif mécatronique est largement utilisé dans de nombreuses applications industrielles.Un problème important de la modélisation est le couplage entre la partie mécanique et la partie électrique du système mécatronique. Plusieurs activités de recherche présentent des idées de couplage entre ces deux éléments. En se basant sur les travaux récents, une nouvelle approche du couplage de l’ensemble moteur-engrenage est présentée. Ce couplage prend en compte le régime transitoire de la partie électrique et l’effet torsionnel du réducteur à engrenages simple étage.

In this paper, we study the problem of constructing flat inputs for multi-output dynamical systems. The notion of flat inputs has been introduced by Waldherr and Zeitz in [30, 31] and can be seen as dual to that of flat outputs. In the single-output case, a flat input can be constructed if and only if the original dynamical system together with its output is observable. In the multi-output case, the observability is not necessary for the existence of flat inputs. The observable case has been treated in [31], where a system of linear algebraic equations has been proposed in order to determine the control vector fields associated to the flat inputs. The goal of this paper is to treat the unobservable case for multi-output dynamical systems. We start by discussing the case when the dynamical system together with the given output is observable and we present a generalization of the results of [31] by relating them with the notion of minimal differential weight. Then we give our main results. We consider the unobservable case for which locally, on an open and dense subset of $\mathbb R^n$ , we construct control vector fields $g_1 , \ldots, g_m$ such that the associated control system is flat. Finally, we explain how our results can be applied to private communication.

International audience; Optimization metaheuristics solve the complex facility layout problems in a reasonable time while browsing large spaces of solutions. The objective of this article is to compare the performance of two methods vs in solving a constrained facility layout problem. We consider their speed in relation to the performance of the obtained solutions, after their convergence, in terms of the total distance traveled by the parts in the workshop. The two chosen metaheuristics have been successfully applied in many search problems: the Genetic Algorithm (GA) and the Particle Swarm Optimization (PSO). In order to determine the shortest path between equipment in a given irregular area (with walls, obstacles, or fixed installations), the A* algorithm was combined with them. The comparison therefore concerns the two methods and . GA and PSO algorithms generate potential solutions for which the shortest path for any couple of machines is determined through the A* search algorithm by bypassing the obstacles. The mathematical model used and the parameters of the genetic algorithm are those developed in [1]. The numerical experiments demonstrate the feasibility and effectiveness of both approaches. The results show that GA provides better solutions in terms of fitness values while PSO is faster.

This work focused on a numerical modeling of innovative non-conventional energy dissipating systems developed and tested experimentally in [1] to enhance their energy absorption capacity. The basic concept utilized the axial plastic buckling of right-circular mild steel like-composite tubes with various case-hardened patterns. Such a heat treatment has been applied on 15% of the outer surface with a depth of 0.5 mm. In this work, a nonlinear finite element modeling was conducted describing the response of such tubes targeting particularly their behavior along the tube thickness. The S4R shell element of 1 mm size was used looking for an accurate mesh convergence. A contact with a dry friction penalty was highlighted by the contact between the tube surface-to-surface and the tube-to-rigid bodies. The mass scaling technique was also considered to increase the stability time. Four case-hardened patterns of 2, 3, 4 and 5 rings and two other vertical strip patterns of 2 and 3 strips were modeled, simulated and compared with the experimental data. Moreover, a numerical design platform with several new case-hardened patterns was proposed. Among the latter, two patterns (12C and 3R3V) showed better energy absorption capacities than the others even for those observed experimentally. Actually, they offered a gain in the energy absorption capacity of 42% compared to the conventional case.

Parametric study is a crucial step in the design process of a mechatronic system. In fact, it helps designers to describe the model parameters, the physical phenomena and to modify the design parameters to reach more satisfactory results in the predesign phase. The main idea of this paper is to develop a novel pre-designing approach to perform parametric studies concerning complex interactions among the different internal components of a system. In this paper, we are interested only in the vibration constraint between the different components embedded in a mechatronic system. To illustrate our approach, we used a specific case study: the vibration interaction between a dynamic excitation and a flexible beam as a support structure, having simply supported boundary conditions is investigated. The position of the perturbation source over the beam is parametrized and the feedback effect between components is taken into consideration. Our methodology is based on a pure analytical approach with Modelica/Dymola as an object oriented modeling language. This model can help designers to perform a parametric study and to examine the impact of the design variables on the mechatronic system response at an early stage of design. Object-oriented approach is used to produce a reusable component-based architecture.

Dynamics of complex structures with dissipative materials is an high stake for mechanical design and optimization. Polymeric materials are commonly used in this way. These materials are effective in various industry, but mastering their use comes up against technological limitations. One of these limitation is the characterization of mechanical properties as a function of frequency and this thesis deals with it. Two contributions are provided in this experimental research work. The first one concerns the time–temperature superposition principle. This principle makes it possible to extend the mechanical properties measured over a restricted frequency range from data obtained at different temperatures. This widespread method is unfortunately not valid for many polymeric materials. It is the reason why innovative structural adhesives are studied in order to characterize the limits of the principle of time-temperature equivalence. The approach is carried out using data from a dynamic mechanical analysis (DMA) conducted on a commercial test bench. The second contribution are part of the first one and represents the main objective of this thesis. An innovative DMA bench is developed for the characterization of polymers over wide frequency ranges without requirements of the principle of time-temperature equivalence. The bench performances are precisely assessed through a characterization campaign conducted on polymer adhesives.; La prédiction des ambiances dynamiques des structures complexes intégrant des matériaux dissipatifs est un enjeu important pour le dimensionnement et l’optimisation des structures. Les matériaux communément utilisés dans ce but sont les polymères. Ils sont particulièrement efficaces dans diverses applications, mais la maîtrise de leur usage se heurte à certains verrous technologiques. L’un de ces verrous est la caractérisation des propriétés mécaniques en fonction de la fréquence et la thèse porte sur ce point. Deux contributions principales sont apportées dans ce travail de recherche expérimentale. La première concerne le principe d’équivalence temps-température. Ce principe permet d’étendre les propriétés mécaniques mesurées sur une plage fréquentielle restreinte à partir de données obtenues à différentes températures. Bien que répandue, cette méthode n’est malheureusement pas valide pour beaucoup de matériaux polymères. C’est pour cette raison que des adhésifs structuraux innovants sont étudiés plus en détails afin de caractériser les limites du principe de l’équivalence temps-température. La démarche est réalisée à partir de données issue d’une campagne d’Analyse Mécanique Dynamique (AMD ou DMA) conduite sur un banc d’essai du commerce. La seconde contribution s’inscrit dans la continuité de la première et représente l’objectif principal de cette thèse. Un dispositif DMA innovant est développé pour la caractérisation des polymères sur de larges plages fréquentielles sans recourir au principe d’équivalence temps-température. Les performances du banc conçu sont évaluées avec précision au travers d’une campagne de caractérisation conduite sur des adhésifs polymères.

In this article results of tribological and structural properties comparison of flame sprayed, by SupJet-S-Eutalloy® multipurpose system, two self-fluxing alloys of Ni-B-Si type in the form of powder for spraying with remelting are presented. Layers were produced on EN-GJL-250 grade grey cast iron with flake graphite in an austenitic matrix. Abrasive wear resistance examination acc. to ASTM G65-00 and erosive wear resistance examination acc. to ASTM G76-2 were performed. Obtained results were compared with abrasion-resistant steel Hardox® 400. Macro- and microscopic examination were carried enabling layer quality evaluation and structure of examined layer and base material determination. X-ray microanalysis was performed to obtain qualitative and quantitative data about microareas chemical composition. The Eutalloy® BronzoChrom 10185 powder flame sprayed layer with higher Boron and Silicon content was characterized by 13% lower wear intensity under mineral abrasive conditions and about 30% higher erosion wear resistance than Eutalloy® NiTec 10224 powder flame sprayed layer.

International audience

International audience

International audience; Failure Modes, Effects and Criticality Analysis (FMECA) is one of the well-known methods of quality management that is used for continuous improvement in product or process design. This method uses linguistic expressions and has good information about cause-effect chains. However, it lacks probabilistic information. Transforming it into a Bayesian Network (BN) makes it possible to be used in maintenance for both diagnosis and prediction. The purpose of this paper is to develop a method that uses as much information as possible from FMECA, including frequency and detection to precisely make the configuration of the BN. To build a BN's structure from FMECA, we elaborate a tool to do it systematically. Moreover, we develop an algorithm to set the parameters of a BN obtained. Elicitation methods based on expert knowledge are used when data is not sufficient. A case study of FMECA in the automotive industry is introduced to verify the applicability of the proposed method in an industrial environment.

Reluctance coil guns are electromagnetic launchers having a good ratio of energy transmitted to actuator volume, making them a good choice for propelling objects with a limited actuator space. In this paper, we focus on an application, which is launching real size soccer balls with a size constrained robot. As the size of the actuator cannot be increased, kicking strength can only be improved by enhancing electrical to mechanical energy conversion, compared to existing systems. For this, we propose to modify its inner structure, splitting the coil and the energy storage capacitor into several ones, and triggering the coils successively for propagating the magnetic force in order to improve efficiency. This article first presents a model of reluctance electromagnetic coil guns using a coupled electromagnetic, electrical and mechanical models. Four different coil gun structures are then simulated, concluding that splitting the kicking coil into two half size ones is the best trade-off for optimizing energy transfer, while maintaining an acceptable system complexity and controllability. This optimization results in robust enhancement and leads to an increase by 104 % of the energy conversion compared to a reference launcher used. This result has been validated experimentally on our RoboCup robots. This paper also proves that splitting the coil into a higher number of coils is not an interesting trade-off. Beyond results on the chosen case study, this paper presents an optimization technique based on mixed mechanic, electric and electromagnetic modelling that can be applied to any reluctance coil gun.

In the frame of the growing trend of promoting the carbon-free mobility, RedHV+ project is carried out in order to investigate whether or not it is possible to manufacture a high-speed and high performance gearbox compatible with car costs. As part of this project, a new dynamic model for very high speed rotors working at non-stationary operating conditions is proposed.First, the different assumptions used for rotordynamics modeling are separated into different categories and classified based on the order of complexity they add to the model. Then, according to the operating conditions of the RedHV+ test bed, a choice of the most realistic modeling assumptions is done. As a first step of the dynamic modeling, the work focused on monorotors. In this study, we suggest not to make any assumption on the angular speed. The rotor shaft is flexible in traction, bending and torsion. The energy source is assumed to be non-ideal and the angular displacement including both the rigid body motion and the torsional deformation is assessed. The dynamic model leads to a novel expression of the gyroscopic effect terms. It shows its capacity, through both its analytical expression as well as through numerical results, to accurately introduce the coupling between the flexural and torsional degrees of freedom. The results are presented at a first step, for a simple rotor made of a shaft, a disk and flexible support. Another degree of complexity is introduced later, by considering a geared parallel-rotor system. The aim is to improve the accuracy of simulations for the rotor dynamics in non-stationary conditions especially when getting through critical speeds.The study of geared systems mainly focuses on the dynamic modeling of a simplified architecture of the RedHV+ test bed. The results obtained using the new dynamic model show interesting conclusions regarding the vibration behavior of the rotor when crossing the critical speeds. The reached vibration levels, in the presence of elastic coupling, highlight the importance of a correct dimensioning of a rotating machinery going through critical speeds.Finally, an identification tool based on the Unscented Kalman Filter is developed in order to enrich the dynamic models of the rotating machineries with more realistic values of its uncertain parameters. The identification tool showed interesting results and it is shown that its performance increases when the rotor goes through critical speeds.; Avec les mesures de plus en plus exigeantes visant à réduire l’impact environnemental des transports routiers, des nouvelles technologies de motorisations électriques sont explorées. Notamment, le projet RedHV+ porte sur la possibilité de construire, à coût automobile, des réducteurs haute vitesse et haut rendement pour les véhicules hybrides. Dans le cadre de ce projet, une modélisation originale du comportement dynamique des réducteurs très hautes vitesses et en régime non-stationnaire est proposée.Dans un premiers temps, les hypothèses de modélisation sont classées par catégories et par ordre de complexité. Ensuite, les hypothèses traduisant au mieux les conditions de fonctionnement des réducteurs de vitesse du projet RedHV+ sont retenues. Le travail de modélisation qui est fait dans une première étape s’intéresse aux monorotors flexibles en flexion et en torsion, fonctionnant à très hautes vitesses et en régime non-stationnaire. Ceci donne lieu, conjointement à l’hypothèse de source non-idéale d’énergie, à une formulation originale de l’équation de la dynamique du rotor. Notamment, la nouveauté dans le modèle dynamique réside dans l’expression des termes résultant de la prise en compte des effets gyroscopiques. Le modèle dynamique, à travers son expression analytique ainsi qu’à partir des résultats de simulations numériques, montre sa capacité à bien prendre en compte et représenter le couplage flexion-torsion. Grace à sa prise en considération du phénomène de Sommerfeld, phénomène observable au niveau du passage par des vitesses critiques, une meilleure précision des niveaux vibratoires latéraux est obtenue.Le nouveau modèle dynamique du monorotor est ensuite étendu à un étage d’engrenages. L’architecture du réducteur utilisée dans le banc RedHV+ est simplifiée pour faire l’étude de son comportement dynamique. Quoique la modélisation du banc est simpliste par rapport à la vraie architecture, elle permet de tirer des conclusions intéressantes par rapports à des éventuels risques lié à un fonctionnement voisinant les vitesses critiques.Finalement, un outil d’identification est développé afin d’enrichir les modèles dynamiques avec des valeurs plus réalistes des paramètres incertains d’une machine tournante en régime non-stationnaire. L’outil associant le modèle dynamique à des observations est mis en œuvre sur un exemple de rotor rigide et montre sa performance à estimer les paramètres incertains, surtout lors du passage par des vitesses critiques.