HAL-Ecole des Ponts ParisTech
Not a member yet
42743 research outputs found
Sort by
Explicit Runge--Kutta schemes with hybrid high-order and weakGalerkin methods for the wave equation in first-order form
No full textWe derive energy error estimates in a fully-discrete setting for the first-order wave equation in its Friedrichs formulation, using third- and fourth-order explicit Runge--Kutta (ERK3 and ERK4) schemes in time combined with hybrid high-order (HHO) and weak Galerkin (WG) methods in space. We establish the key properties to address the static coupling between the cell and the face unknowns inherent in hybrid methods. On simplicial meshes, we prove optimal convergence rates in time and in space for both ERK3 and ERK4 under the usual CFL condition. The two key arguments in the proof are the special interpolation operator devised for hybridizable discontinuous Galerkin (HDG) methods, and recent bounds on the operator norm of Taylor polynomials applied to the discrete differential operator. On general polytopal meshes, using different arguments, we prove optimal convergence rates in time and quasi-optimal in space for ERK3 under the usual CFL condition. Numerical experiments illustrate the theoretical findings
Record-breaking high temperature amplifies the negative anomaly of tropical net land carbon sinks in the 2023-2024 El Niño
No full textInternational audienc
Convolution Block Extension of DCNN for Retinal Vascular Segmentation: Taxonomy and Discussion
No full textInternational audienceThe retinal vascular tree (RVT) segmentation is a main step for diagnosing several ocular diseases. Higher accurate segmentation remains crucial to ensure a reliable disease detection and hence clinical treatment. Numerous standard deep learning (DL) architectures have been employed to segment the RVT regardless of the image field However, due to the intricate morphologies of vascular trees comprising fine and complex structures, those DL architectures failed to achieve high accuracy in retinal vessel segmentation. Therefore, several promising solutions have been developed to overcome these limitations, where their main contributions rely on adapting the convolution processing of deep convolutional neural networks (DCNNs) blocks with respect to the retinal vessels characteristics. In this paper, we present a review of extended convolution blocks within DCNNs for RVT segmentation from fundus images. Our main contributions remain on (1) Identifying the different principles extension of convolution blocks; (2) Proposing a taxonomy of convolution block extension, and (3) Analyzing and discussing the strengths and weaknesses of each extension type with respect to segmentation quality and database characteristics. The presented study allows a valuable recommendation for future research in the field of RVT segmentation based on DCNN
Learning elastoplasticity with implicit layers
No full textInternational audienceThis work presents a novel machine learning approach for learning elastoplasticity directly from stress-strain data. Data-driven plasticity modeling is challenging due to the non-smooth transitions induced by the yield criterion and the complex nature of multi-dimensional yield surfaces. To address these difficulties, we propose an implicit layer-based architecture that formulates the elastoplastic constitutive update as a convex optimization problem with learnable parameters, representing general plastic yield surfaces as parametrized convex sets. The proposed approach exhibits strong generalization capabilities due to its embedded convex structure, achieving high performance even with limited data and in the presence of noise
Some results and comments on a class of non-equilibrium multiphase flow models
No full textInternational audienc
Une histoire économique et sociale : la France, de la Préhistoire à nos jours
No full textNational audienceRaconter et comprendre l’histoire économique et sociale de la France, des origines de son peuplement jusqu’aux développements les plus récents, tel est l’objectif qui a rassemblé les quelque soixante-dix spécialistes – historiens, économistes, mais aussi archéologues, préhistoriens, géographes, politistes et sociologues – qui signent collectivement ce texte. De la Préhistoire à nos jours, les auteurs envisagent les activités et les échanges, mais aussi les institutions politiques et religieuses en ce qu’elles ont influencé et structuré l’économie et la société française, sur la longue durée, et jusque dans son quotidien
A stabilized hybridized Nitsche method for sign-changing elliptic PDEs
No full textInternational audienceWe present and analyze a stabilized hybridized Nitsche method for elliptic problems with sign-changing coefficients without imposing symmetry assumptions on the mesh around the material interfaces. The use of a stabilized primal-dual formulation allows us to cope with the sign-changing nature of the problem and to prove optimal error estimates under two assumptions on the continuous problem, namely that it admits a unique solution and that the contrast at the sign-changing interface lies outside a certain critical interval. The method can be used on arbitrary shape-regular meshes (fitted to material interfaces) and yields optimal convergence rates for smooth solutions. As an illustration, the method is applied to simulate a realistic acoustic cloaking device
L’investissement citoyen : levier de territorialisation juste de la transition énergétique ? Étude d’une dynamique d’autoconsommation collective à Rennes, France
No full textInternational audienceThis article explores the contribution of an urban collective self-consumption project to understanding the new dynamics of French electrification through the lens of energy justice. It analyzes the socio-spatial, legal, and financial framework of an initiative based in Rennes and examines the role of citizen finance as a lever for a territorialized energy transition—territorialization being considered here as essential for a just transition. As such, how does citizen investment contribute to territorializing the energy transition? Does this territorialization help reduce energy injustices, and if so, how? To address these questions, the article draws on a geolegal fieldwork conducted over one year. The findings demonstrate how the studied project effectively localizes practices of electricity valuation. Then, it enhances three dimensions of justice with a particular focus on an innovative flat rate electricity package sale, including an “eco” tariff designed for populations experiencing energy precarity, and the enhancement of procedural justice. Finally, the article explores governance implications, particularly regarding what “taking stock” means for the local deployment of energy systems. This work first addresses a key need, expressed in both social sciences and law, to clarify the concept of territorialization. By mobilizing a case study of limited scale (spatial embedment, number of participants, financial flow complexity), it contributes to shedding light on investment logics in electrical infrastructure and their impacts on energy access. Lastly, this article serves a collective effort to document the dynamics of collective self-consumption as implementing political ambitions for energy decentralization in the French context.Cet article explore la contribution d’un projet d’autoconsommation collective (ACC) urbaine à la compréhension des nouvelles dynamiques d’électrification française, sous l’angle de la justice énergétique. Il analyse le montage socio-spatial, juridique et financier d’une initiative rennaise et observe le rôle de la finance citoyenne comme levier de transition territorialisée — la territorialisation étant ici envisagée comme nécessaire à une transition juste. L’article pose deux questions : comment l’investissement citoyen territorialise-t-il la transition ? Cette territorialisation participe-t-elle à réduire les injustices énergétiques et, le cas échéant, comment ? Pour y répondre, l’article s’appuie sur un terrain géo-légal d’un an. Les résultats montrent comment l’opération étudiée localise les pratiques de valorisation de l’électricité. Trois dimensions de justice sont ensuite mises en avant en insistant sur l’innovation d’un système forfaitaire comprenant un tarif « éco » destiné aux populations en précarité énergétique et l’amélioration de la justice procédurale. Enfin, l’article explique les implications en termes de gouvernance autour de ce que « prendre action » signifie pour un déploiement électrique local. Cet article sert tout d’abord le besoin, exprimé tant en sciences sociales qu’en droit, de clarifier la notion de transition territorialisée. En mobilisant un cas d’étude aux dimensions restreintes (ancrage spatial, nombre de participants, complexité des flux financiers), il contribue également à éclairer les logiques d’investissements dans les infrastructures électriques et leurs impacts sur les formes d’accès à l’énergie. Ce travail s’inscrit enfin dans un effort collectif de documentation des dynamiques d’autoconsommation électrique, modalité de traduction en actes des ambitions politiques de décentralisation énergétique
The end of Low Emission Zones or the impossibility of restricting cars in cities without social justice
No full textInternational audienceLow emission zones are a response to air quality issues. However, they appear to be central to urban decarbonisation policies and a justification for promoting modal shift policies and supporting alternative modes to the car. Yet their ambition is limited by the social inequalities they raise. By studying the example of the Greater Paris low-emission zone in France, we will examine the notion of environmental justice in this public policy.Research on the subject focuses on different aspects of social inequalities in the face of this environmental policy, but without taking an integrated view of inequalities as a function of use and location. They point to a number of inequalities in isolation without succeeding in showing the overlapping inequalities that certain households may suffer. The aim of this research is to question the notion of justice surrounding this environmental policy. Benefits, constraints and capabilities will therefore be studied.On the one hand, we will be looking at the beneficiaries of exposure for resident and non-resident users of low-emission zones. It would appear, however, that if these restriction policies are to be accepted, they must also benefit the populations whose trips are restricted. We will also assess the social impact of this measure and evaluate which households are affected, not just in terms of ownership but also in terms of use. Finally, we are going to analyse the current modal shift potential as a function of people's place of activity. This will reveal more clearly the constraints on car use as a function of the modes available, and thus reveal territorial and social inequalities.Finally, a broader perspective will show why policies that target the greatest number of people, and especially the poorest, while excluding the richest from restrictions, threaten the place of environmental issues in society in general
Fate of a Water Drop in a Cellulosic Material
No full textInternational audienceThe long-term evolution of an aqueous droplet containing pigments, viruses, ions or polymers, and reaching the surface of a wood piece, a paper, a natural fabric or a mask, governs the structure durability, the user's comfort, the printing quality or the bacteria or viruses development. It is generally believed that this droplet somewhat spreads, penetrates the structure, stabilizes and eventually dries, following standard wetting and fluid mechanics. Here, from nuclear magnetic resonance relaxometry and magnetic resonance imaging allowing to distinguish water in its different states, we show that, instead of drying, the water is absorbed in nanoscale pores in the cellulose fibers and tends to diffuse throughout the entire structure in this particular state (bound water). Thus, the liquid water disappears from the porosity before drying, while the non-absorbed solute or particles remain stuck to the solid surfaces in the initial region of liquid penetration. We develop a model predicting the coupled transport of vapor and bound water through the structure, which predicts quantitatively the observations without any parameter fitting. This suggests to change the way of considering the fate of sweat, mist, ink, or rain on natural fabric clothing or biobased materials, and consider this diffusion as a potential major effect