Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Improvement of the Adjustable Localization Operator identification based on the confined plasticity zone shape for multiaxial elasto-plastic cyclic predictions
No full textInternational audienceSimplified methods for elasto-plastic calculations are a cost and time savings solution to obtain mechanical state at the notch tip. The Adjustable Localization Operator (ALO) method has been proven to be a robust and inherently multiaxial method. This study aims to propose an alternative strategy for identifying its parameters. The principal idea is to consider the shape of the confined plastic zone in order to adapt the various parameters of the adjustable localization operator. The proposed approach has been validated on notched structures with plane strain assumption as well as on a new design structure for multiaxial proportional and non-proportional loadings. The results obtained have shown the interest of considering the shape of the confined plastic
A General Comparison Principle for Hamilton Jacobi Bellman Equations on Stratified Domains
No full textInternational audienceThis manuscript aims to study finite horizon, first order Hamilton Jacobi Bellman (HJB) equations on stratified domains. This problem is related to optimal control problems with discontinuous dynamics. We use nonsmooth analysis techniques to derive a strong comparison principle as in the classical theory and deduce that the value function is the unique viscosity solution. Furthermore, we prove some stability results of the Hamilton Jacobi Bellman equation. Finally, we establish a general convergence result for monotone numerical schemes in the stratified case
Quantifying mixing in arbitrary fluid domains: a Padé approximation approach
No full textInternational audienceWe consider the model problem of mixing of passive tracers by an incompressible viscous fluid. Addressing questions of optimal control in realistic geometric settings or alternatively the design of fluid-confining geometries that successfully effect mixing requires a meaningful norm in which to quantify mixing that is also suitable for easy and efficient computation (as is needed, e.g., for use in gradient-based optimization methods). We use the physically inspired reasonable surrogate of a negative index Sobolev norm over the complex fluid mixing domain Ω, a task which could be seen as computationally expensive since it requires the computation of an eigenbasis for L2(Ω) by definition. Instead, we compute a representant of the scalar concentration field in an appropriate Sobolev space in order to obtain an equivalent definition of the Sobolev surrogate norm. The representant, in turn, can be computed to high-order accuracy by a Padé approximation to certain fractional pseudo-differential operators, which naturally leads to a sequence of elliptic problems with an inhomogeneity related to snapshots of the time-varying concentration field. Fast and accurate potential theoretic methods are used to efficiently solve these problems, with rapid per-snapshot mix-norm computation made possible by recent advances in numerical methods for volume potentials. We couple the methodology to existing solvers for Stokes and advection equations to obtain a unified framework for simulating and quantifying mixing in arbitrary fluid domains. We provide numerical results demonstrating the convergence of the new approach as the approximation order is increased
Verifying Provable Stability Domains for Discrete-Time Systems Using Ellipsoidal State Enclosures
No full textInternational audienceStability contractors, based on interval analysis, were introduced in recent work as a tool to verify stability domains for nonlinear dynamic systems. These contractors rely on the property that-in case of provable asymptotic stability-a finitely large domain in a multi-dimensional state space is mapped into its interior after a certain integration time for continuous-time processes or after a certain number of discretization steps in a discrete-time setting. However, a disadvantage of the use of axis-aligned interval boxes in such computations is the omnipresent wrapping effect. As shown in this contribution, the replacement of classical interval representations by ellipsoidal domain enclosures reduces this undesirable effect. It also helps to find suitable ratios for the edge lengths if interval-based domain representations are investigated. Moreover, ellipsoidal domains naturally represent the possible regions of attraction of asymptotically stable equilibrium points that can be analyzed with the help of quadratic Lyapunov functions, for which stability criteria can be cast into linear matrix inequality (LMI) constraints. For that reason, this paper further presents possible interfaces of ellipsoidal enclosure techniques with LMI approaches. This combination aims at the maximization of those domains that can be proven to be stable for a discrete-time rangeonly localization algorithm in robotics. There, an Extended Kalman Filter (EKF) is applied to a system for which the dynamics are characterized by a discrete-time integrator disturbance model with additive Gaussian noise. In this scenario, the measurement equations correspond to the distances between the object to be localized and beacons with known positions
A 1-D/3-D coupling approach for compressible non-equilibrium two-phase flows using the Baer-Nunziato model based on the Finite-Volume framework
No full textInternational audienc
Interface propagation and energy dissipation in Shape Memory Alloys
No full textInternational audienc
Stability of the P1nc-(P0+P1) element
No full textInternational audienceWe solve the Stokes problem numerically. We analyse the P1nc-(P0+P1) mixed finite element method which exhibits interesting numerical features. However, only an incomplete proof of the inf-sup condition is available. We prove here this condition and the stability of the method
Les communications furtives embarquées dans un drone aérien à l’aide de la radio intelligente
No full textSDR (Software Defined Radio) created illegal spectrum’s usage opportunities. It is possible to design a low-cost stealth network embedded in low-power platform that communicates in the unused users downlink side of the LTE-A (LTE-Advanced). The stealth parallel network needs reliable detectors of unused times to avoid jam of the LTE-A network. We created three such detectors. They work after the synchronization and can detect the free spaces with only one symbol data. The detection process is shorter than one symbol period and works for any frequency bandwidth. Therefore, this minimises the delays and maximises the usable space. We also modelized the eNodeB (evolved Node B), the stealth receiver and the propagation channel between them to find the statistical features of the signal analysed by the detectors. We studied the detectors and did simulations on full and partial LTE-A bandwidths. Next, we optimised and implemented them on a lowcost FPGA (Field-Programmable Gate Array) board with a HLS (High-level synthesis) tool. Finally, we looked to necessary steps to get a whole stealth network and to adapt the detecors to 5G networks.La SDR (software defined radio) a créé des opportunités d’exploitation frauduleuse du spectre. Il est possible de concevoir un réseau de communications furtives bon marché, embarqué sur des plateformes légères et qui émet ses trames dans les parties inutilisées par les usagers du spectre duLTE-A (LTE-advanced) dans la bande descendante. Afin de ne pas perturber le réseau principal, il faut que l’émetteur furtif dispose d’un détecteur fiable d’instants inexploités du LTE-A. Nous avons donc développé trois détecteurs de signaux capables, après synchronisation, de détecter un signal OFDM (orthogonal frequency division multiplexing) à partir d’un seul symbole, pour n’importe quelle largeur fréquentielle et avec une latence inférieure à un symbole LTE-A. Nous minimisons ainsi les délais et nous maximisons l’espace temps-fréquence exploitable par le réseau furtif. Une modélisation statistique de l’eNodeB (evolved node b), du récepteur furtif et de leur canal de propagation a permis de préciser les entrées des détecteurs. Les performances des détecteurs ont été étudiées et simulées pour des bandes LTE-A partielles et totales. Nous les avons optimisés et implémentés sur une carte FPGA (field-programmable gate array) à bas coût via une description d’un haut niveau d’abstraction, la HLS (high-level synthesis). Enfin, nous avons étudié les travaux nécessaires à la création d’un furtif complet ainsi que les adaptations possibles du modèle à la norme 5G
Simulation numérique d'impacts hydrodynamiques obliques de géométries représentatives de structures aéronautiques
No full textInternational audienceThis communication deals with the emergency water landing of an aircraft (ditching). During an aircraft ditching, the structure experiences critical loads and eventually fails. The appropriate design allows minimizing the risks of occupants’ injuries. Therefore, the behaviour of an aircraft in a ditching situation has been thoroughly studied and demonstrated using scale model tests, comparison with existing designs, analytical investigations and numerical simulations. The recent progress in computation capabilities led to the increased use of numerical simulations in the certification process of aircraft. However, modelling such complex hydrodynamic impacts is challenging, even with state-of-the-art numerical approaches. For example, during oblique and high-velocity hydrodynamic impacts, complex hydrodynamic phenomena tend to develop. Among these phenomena, we can cite air cushion, suction forces, cavitation, ventilation, and aeration. These phenomena result from the interaction between the structure and the fluids, they occur simultaneously and influence each other. The objective of this work is to demonstrate the capability of the present numerical method to model hydrodynamic impacts in realistic ditching conditions and involving suction forces, cavitation, and ventilation. In the present work, numerical simulations are carried out using the explicit Finite Elements solver Radioss, an Eulerian formulation for the fluid and a Lagrangian formulation for the structure. The fluid-structure interaction is based on an immersed contact interface (penalty method). The numerical results are compared to the experimental results of Iafrati and Grizzi (2019) in terms of hydrodynamic force and pressure. The numerical method showed encouraging results regarding the study of hydrodynamic impacts in conditions representative of an aircraft ditching
FML : un langage de fédération de modèles pour l’interopérabilité sémantique de sources d’information hétérogènes
No full textModelling is a universal practice at the root of human thought, for designing, understanding, calculating, imagining, analyzing and communicating. Model Driven Engineering (MDE) aims to systematize the use of models in all activities related to the whole life cycle of systems and software (development, deployment, integration, maintenance and evolution). From this context emerges the need to coherently manage multiple models (or information sources interpreted as models), where each model comes with its own autonomy and lifecycle. The core contribution of this thesis is founded on the formalization of an approach called "model federation", which aims to explicitly reify the links between federated models, while associating behaviour with those links. More specifically, we propose a modelling language called FML, which is both a language for conceptualization and reification of interpretation, but also features a referencing mechanism that enables federation links to point on heterogeneous data sources. FML allows both to react to the behaviour and evolution of federated models and to program behaviours acting on these models. We propose a tooling environment for the language within a software infrastructure called Openflexo. Finally, we validate the approach on four use cases, research prototypes and industrial experiments.La modélisation est une pratique universelle à la base de la pensée humaine, pour concevoir, comprendre, calculer, imaginer, analyser, communiquer. L'Ingénierie Dirigée par les Modèles (IDM) se propose de systématiser l'utilisation de modèles dans toutes les tâches liées au cycle de vie des systèmes et des logiciels (développement, déploiement, intégration, maintenance et évolution). De ce contexte émerge le besoin de gérer en cohérence de multiples modèles (ou des sources d'information interprétées comme des modèles), qui ont chacun leur autonomie et leur propre cycle de vie. Le coeur de la contribution de cette thèse repose sur la formalisation d'une approche que nous nommons "fédération de modèles" et qui se propose de réifier explicitement les liens entre les modèles fédérés et de leur associer un comportement. Plus précisément, nous proposons un langage de modélisation appelé FML, qui est à la fois un langage permettant la conceptualisation et la réification de l'interprétation, mais qui est également doté d'un mécanisme de désignation qui permet l'établissement de liens de fédération vers des sources de données hétérogènes. FML permet, à la fois, de réagir aux comportements et évolutions des modèles fédérés et de programmer des comportements agissant sur ces modèles. Nous proposons un environnement outillé du langage au sein d'une infrastructure logicielle appelée Openflexo. Enfin, nous validons l'approche sur quatre cas d'utilisation, prototypes de recherche et expérimentations industrielles