Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Toughening induced by the formation of facets in mode I+III brittle fracture: experiments versus a two-scale Cohesive Zone model
No full textInternational audienceWhen subjected to some anti-plane shear mode III loading, segmentation of the crack front frequently occurs during propagation: even if the crack is initially planar, propagation produces facets/segments rotated toward the shear free direction. Here, we examine, both experimentally and theoretically, the effect of this microstructure on the effective macroscale brittle fracture toughness. Experiments performed on PMMA beams reveal that the critical load leading to abrupt rupture increases with mode III to mode I ratio. This apparent macroscopic toughening is usually taken into account by invoking a specific mode III toughness in addition to the mode I one. By applying thoroughfully a micro/macroscale Cohesive Zone (CZ) model that we have recently developed, we demonstrate that an additional material constant is useless here since this toughness increase can be attributed mainly to the presence of the facets at the microscale, whose geometry can be anticipated to depend on the classical mode I material constants. More precisely, two related physical mechanisms are generated due to the formation of a disconnected crack front: (i) changes in fractured surface area in comparison to a straight propagation, and (ii) crack shielding caused by the facets that reduce the effective crack opening. While the first effect is obvious to quantify, we show that the second plays an essential role but is more complex to take into account: it depends on the solution of the three-dimensional elasticity problem in presence of the facets, that is considered in the CZ model. We illustrate on the experiments how to use this approach in practice to determine the critical fracture threshold
MBDA’s Industrial Model and European Defence
No full textInternational audienceAmong European arms producers, MBDA represents the most advanced example of industrial integration and cross-border specialisation, thanks to a unique model of French–British centres of competence. However, it constitutes an exception inside the European defence technological and industrial base. It is therefore relevant to analyse why MBDA was able to achieve such consolidation process from fragmented competences of the European missile industry and to merge once competing companies despite the lack of a unique regulatory framework at the European level. This article presents the main features of MBDA today. It analyses the consolidation process in the European missile industry that created MBDA. It underlines how cooperative programmes provided a critical impulse in favour of industrial consolidation and the specialisation of MBDA sites across Europe. It explains how technological and industrial assets were reshuffled across European borders. It focuses on exports as a key dimension for MBDA’s model. Finally, it concludes by exploring the limits that result from the lack of a unified regulatory framework in Europe
Modelling and identification of fatigue load spectra: Application in the automotive industry
No full textInternational audienceThis paper is focused on variable amplitude loading spectra applied in the automotive industry, more specifically for the chassis system parts with respect to high cycle fatigue design. A first analytical model referred to as Heuler's model is considered. It is of the simplest use, with one parameter to be identified. A whole identification process, based on road measurement realized on carmaker's proving ground with instrumented vehicles, is developed and discussed. Once the model is identified, the cycles influence on damage production is also investigated. In the light of the obtained results with this model, a second one, noted modified Heuler's model, is introduced and investigated. This model, similar to the first, requires the identification of two parameters. Once again, the whole model identification process is applied, and the lifetime accuracy is assessed. A new tool is presented, enabling the visualisation on the loading spectrum representation of the most and less damaging cycles. Hence, a methodology is set to legitimize a gate implementation when studying a variable amplitude loading spectrum
All-in-one: Toward hybrid data collection and energy saving mechanism in sensing-based IoT applications
No full textInternational audienceBig data collection and storage have become one of the most obvious challenge in this era. Indeed, much of that data is collected thanky to a huge number of connected devices in sensing-based IoT applications. Thus, in order to deal with data growth in such applications, researchers have focused on data reduction approach as an efficient solution for minimizing the amount of data collection and saving the limited sensor energy in such networks. Mainly, data reduction approach relies on various kinds of data processing techniques such that aggregation, compression, prediction, clustering, sensing frequency adaptation and spatial-temporal correlation. However, each of those techniques has its own advantages and disadvantages regarding sensor energy saving, data reduction ratio, data accuracy, complexity, etc. In this paper, we propose a hybrid data collection and energy saving mechanism, called All-in-One, for sensing-based IoT applications. The proposed mechanism takes advantages from existing data reduction techniques while optimizing various performance metrics. All-in-One relies on the cluster network architecture and works on three main phases: on-period, in-period and in-node. The first phase, e.g. on-period, allows each sensor node to search the similarity among its periodic collected data then to reduce its data transmission to the Cluster-Head (CH) by applying either data aggregation, compression or prediction technique. The second phase, e.g. in-period, allows each sensor to study the variation of the monitored condition then to reduce its data collection according to two techniques, on-off transmission or adapting sensing frequency. The last phase, e.g. in-node, is applied at the CH level and aims to remove the redundancy among data collected by neighboring nodes, based on in-network correlation or data clustering techniques, before sending the data to the sink. We conducted simulations on real sensor data in order to evaluate the efficiency of our mechanism, in terms of several performance metrics, compared to other exiting techniques
Cardiac Abnormality Detection Based on an Ensemble Voting of Single-Lead Classifier Predictions
No full textInternational audienc
Techniques numériques rétrogrades de type McKean pour les EDPs et une application à la gestion de l'énergie.
No full textThis thesis concerns McKean Stochastic Differential Equations (SDEs) to representpossibly non-linear Partial Differential Equations (PDEs). Those depend not onlyon the time and position of a given particle, but also on its probability law. In particular, we treat the unusual case of Fokker-Planck type PDEs with prescribed final data. We discuss existence and uniqueness for those equations and provide a probabilistic representation in the form of McKean type equation, whose unique solution corresponds to the time-reversal dynamics of a diffusion process.We introduce the notion of fully backward representation of a semilinear PDE: thatconsists in fact in the coupling of a classical Backward SDE with an underlying processevolving backwardly in time. We also discuss an application to the representationof Hamilton-Jacobi-Bellman Equation (HJB) in stochastic control. Based on this, we propose a Monte-Carlo algorithm to solve some control problems which has advantages in terms of computational efficiency and memory whencompared to traditional forward-backward approaches. We apply this method in the context of demand side management problems occurring in power systems. Finally, we survey the use of generalized McKean SDEs to represent non-linear and non-conservative extensions of Fokker-Planck type PDEs.Cette thèse s'intéresse aux équations différentielles stochastiques de type McKean(EDS) et à leur utilisation pour représenter des équations aux dérivées partielles (EDP) non linéaires. Ces équations ne dépendent pas seulement du temps et de la position d'une certaine particule mais également de sa loi. En particulier nous traitons le cas inhabituel de la représentation d'EDP de type Fokker-Planck avec condition terminale fixée. Nous discutons existence et unicité pour ces EDP et de leur représentation sous la forme d'une EDS de type McKean, dont l'unique solutioncorrespond à la dynamique du retourné dans le temps d'un processus de diffusion.Nous introduisons la notion de représentation complètement non-linéaire d'une EDP semilinéaire. Celle-ci consiste dans le couplage d'une EDS rétrograde et d'un processus solution d'une EDS évoluant de manière rétrograde dans le temps. Nous discutons également une application à la représentation d'une équation d'Hamilton-Jacobi-Bellman (HJB) en contrôle stochastique. Sur cette base, nous proposonsun algorithme de Monte-Carlo pour résoudre des problèmes de contrôle. Celui ciest avantageux en termes d'efficience calculatoire et de mémoire, en comparaisonavec les approches traditionnelles progressive rétrograde. Nous appliquons cette méthode dans le contexte de la gestion de la demande dans les réseaux électriques. Pour finir, nous faisons le point sur l'utilisation d'EDS de type McKean généralisées pour représenter des EDP non-linéaires et non-conservatives plus générales que Fokker-Planck
Education numérique - Manuel Cycle 1 (1-2°) - Collection Décodage 2024
No full textInternational audienceLe manuel CODAGE d’Éducation Numérique pour le Cycle 1 a été réalisé dans le cadre du projet cantonal d’introduction d’un enseignement de l’éducation numérique dans le cursus scolaire vaudois.Il est le fruit d’une collaboration entre la Direction pédagogique de la DGEO (Direction Générale de l’Enseignement Obligatoire et de la pédagogie spécialisée) du canton de Vaud, le Centre LEARN-EPFL (École Polytechnique Fédérale de Lausanne), la HEP (Haute École Pédagogique) Vaud, avec l’expertise d’Inria (Institut français de recherche en sciences du numérique)
Dispositif de roulement et véhicule muni de celui-ci
No full textLa présente invention concerne un dispositif de roulement, comportant un moyeu rotatif autour d’une direction (X), des éléments annulaires, disposés autour du moyeu et portant chacun une surface périphérique de roulement, des tiges suivant une deuxième direction longitudinale (L), un support radial autour du moyeu, les tiges traversant des premiers évidements du support radial et étant inclinables par rapport au support radial (103). L’invention est caractérisée en ce que chaque élément annulaire a des deuxièmes évidements, au travers desquels les tiges sont aptes à coulisser en translation suivant la direction longitudinale (L) lorsque les tiges sont inclinées, les éléments annulaires ayant chacun des guidages en translation dans des plans perpendiculaires à la direction (X)
Stochastic incremental mirror descent algorithms with Nesterov smoothing
No full textInternational audienceWe propose a stochastic incremental mirror descent method constructed by means of the Nesterov smoothing for minimizing a sum of finitely many proper, convex and lower semicontinuous functions over a nonempty closed convex set in a Euclidean space. The algorithm can be adapted in order to minimize (in the same setting) a sum of finitely many proper, convex and lower semicontinuous functions composed with linear operators. Another modification of the scheme leads to a stochastic incremental mirror descent Bregman-proximal scheme with Nesterov smoothing for minimizing the sum of finitely many proper, convex and lower semicontinuous functions with a prox-friendly proper, convex and lower semicontinuous function in the same framework. Different to the previous contributions from the literature on mirror descent methods for minimizing sums of functions, we do not require these to be (Lipschitz) continuous or differentiable. Applications in Logistics, Tomography and Machine Learning modelled as optimization problems illustrate the theoretical achievements
Modeling the formation of social conventions from embodied real-time interactions
No full textInternational audienceWhat is the role of real-time controland learning in the formation of social conventions? To answer this question, we propose a computational model that matches human behavioral data in a social decision-making game that was analyzed both in discrete-time and continuous-time setups. Furthermore unlike previous approaches,our model takes into account the role of sensorimotor control loops in embodied decision-making scenarios.For this purpose, we introduce the Control-based Reinforcement Learning (CRL) model. CRL is grounded in the Distributed Adaptive Control (DAC) theory of mind and brain, where low-level sensorimotor control is modulated through perceptual and behavioral learningin a layered structure. CRL follows these principles by implementing a feedback control loop handling the agent’s reactive behaviors(pre-wired reflexes), along with an Adaptive Layer that uses reinforcement learning,to maximize long-term reward.We test our model in a multi-agent game-theoretic task in which coordination must be achieved to find an optimalsolution.We show that CRL is able to reach human-level performance on standard game-theoreticmetrics such as efficiency in acquiring rewards and fairness in reward distribution