Archive Ouverte INSA Lyon
Not a member yet
75807 research outputs found
Sort by
Vers une évaluation temps-réel des capacités de détection d’événements infrasons via l'estimation des pertes de transmission par méthodes d''apprentissage profond
No full textAccurate modeling of infrasound transmission loss is essential for evaluating the performance of the International Monitoring System, enabling the effective design and maintenance of infrasound stations to support compliance of the Comprehensive Nuclear-Test-Ban Treaty. State-of-the-art propagation modeling tools enable transmission loss to be finely simulated using atmospheric models. However, the computational cost prohibits the exploration of a large parameter space in operational monitoring applications. To address this, recent studies made use of a deep learning algorithm capable of making transmission loss predictions almost instantaneously. However, the use of nudged atmospheric models leads to an incomplete representation of the medium, and the absence of temperature as an input makes the algorithm incompatible with long range propagation. In this study, we address these limitations by using both wind and temperature fields as inputs to a neural network, simulated up to 130 km altitude and 4, 000 km distance. We also optimize several aspects of the neural network architecture. We exploit convolutional and recurrent layers to capture spatially and range-dependent features embedded in realistic atmospheric models, improving the overall performance. The neural network reaches an average error of 4 dB compared to full parabolic equation simulations and provides epistemic and data-related uncertainty estimates. Its evaluation on the 2022 Hunga Tonga-Hunga Ha’apai volcanic eruption demonstrates its prediction capability using atmospheric conditions and frequencies not included in the training. This represents a significant step towards near real-time assessment of International Monitoring System detection thresh olds of explosive sources.* We provide a convolutional recurrent neural network estimating in near real-time ground-level infrasound transmission loss.* The neural network exploits spatial and range-dependent features in atmospheric models, and makes predictions with an uncertainty estimate.* The network can be used as a tool for near real-time estimation of infrasound event detection capability at a global scale.* Nous proposons un réseau de neurones récurrent convolutif qui permet d'estimer en temps quasi-réel les pertes par transmission des infrasons au niveau du sol.* Le réseau de neurones exploite les caractéristiques spatiales et dépendante de la distance des modèles atmosphériques d'entrée, et fait des prédictions avec une estimation de l'incertitude.* Le réseau peut être utilisé comme outil pour l'estimation en temps quasi-réel de la capacité de détection d’événements infrasons à l'échelle mondiale
What pins edge dislocations in random alloys? Comparing size and elastic heterogeneities
No full textInternational audienceThis work compares the impact of size mismatch and elastic heterogeneities on solute pinning of edge dislocations in random body-centered cubic alloys -solid solutions in which different atomic species occupy lattice sites at random. Using atomic-scale modeling of both parametric alloys and more realistic binary systems (W-Ta and W-Nb) across the full composition range, we show that the mismatch of elastic constants between the alloy constituents can have an effect comparable to, or even greater than, the size mismatch. This contrasts with some solid solution strengthening theories, which assume a homogeneous elastic medium. Our results further demonstrate that the combined effects of size and modulus mismatches are well captured by a quadratic form, consistent with historical phenomenological approaches
Construire des systèmes soutenables : cadre et outils pour la conception en soutenabilité forte
No full textInternational audienc
Mieux tenir compte des données non quantifiées : une application de la méthode du maximum de vraisemblance pour étudier les tendances spatio-temporelles du glyphosate et de l’AMPA dans la Seine.
No full textInternational audienceThis work investigates the spatial and temporal trends of glyphosate and one of its main degradation products, aminomethylphosphonic acid (AMPA), at several river monitoring points within the greater Paris area (Marne and Seine Rivers, upstream and downstream of the urban areas). To handle datasets containing up to ~50% left-censored data, the maximum likelihood estimation (MLE) method was applied, thus avoiding distorting the information through common practices such as ignoring these data points or substituting them with half the reporting limit. We aimed to illustrate a practical approach to handle left-censored data in environmental studies using glyphosate and AMPA as examples. We modeled the variability of their concentrations with a lognormal distribution, and found a significant contribution of the urban areas to pollution levels in surface waters. Notably, AMPA concentrations downstream of Paris were an order of magnitude higher compared to rural sections of the Marne river. Temporal trend analyses showed a general decline in glyphosate concentrations, as well as AMPA to a lesser extent, following the “Labbé” law prohibiting the use of pesticides in urban areas. While our findings suggest an influence of the Greater Paris urban areas to pollution levels, the study of underlying processes were beyond the scope of this work. Broader, more comprehensive analyses with additional data are necessary to confirm and consolidate these results.Le travail présenté s’intéresse aux tendances spatio-temporelles des concentrations de glyphosate et de l'un de ses principaux produits de dégradation, l'acide aminométhylphosphonique (AMPA), en plusieurs points de surveillance de deux grandes rivières de l'agglomération parisienne (la Marne et la Seine, en amont et en aval de Paris). Pour traiter les données non quantifiées (jusqu’à ~50% des valeurs), nous avons appliqué la méthode du maximum de vraisemblance, afin d’illustrer une manière de ne pas dénaturer l’information contenue dans des séries statistiques contenant de telles données. Nous avons modélisé les distributions des concentrations en glyphosate et AMPA par une loi log-normale, et constaté une contribution significative des zones urbaines aux niveaux de contamination dans les rivières. Notamment, les concentrations en AMPA en aval de Paris étaient supérieures d'un ordre de grandeur par rapport aux sections rurales de la Marne. Les analyses des tendances temporelles ont montré un déclin général des concentrations de glyphosate, et d'AMPA dans une moindre mesure, suite à l’entrée en vigueur de la loi Labbé qui interdit l’usage des pesticides en milieu urbain. Bien que nos résultats mettent en évidence des tendances globales, l’étude des processus sous-jacents dépasse le cadre de ce travail : des analyses plus larges et plus systémiques avec des données supplémentaires sont nécessaires pour confirmer et consolider ces résultats
L'observatoire des Sédiments du Rhône (OSR) : une plateforme de recherche multidisciplinaire pour des recherches fondamentales et appliquées sur la vallée du Rhône (France)
No full textInternational audienceThe Rhône Sediment Observatory, created in 2009, involves an interdisciplinary research team (hydrology, hydraulics, geomorphology, geochemistry, geomatics, sedimentology) in strong interaction with river stakeholders. It aims to provide scientific knowledge for promoting a sustainable management of channel forms and sediment processes of this river, all along the main channel. A summary of the major findings after 16 years of research is proposed. They allowed to better characterize and explain the channel evolution of the last century, highlight aquatic and riparian habitat conditions, and assess channel sensitivity to changes. A network of monitoring stations is operating to continuously survey suspended particulate matter (SPM) concentrations, and to collect SPM samples for the analyses of their inorganic or organic contaminant contents. Fluxes of sediment and contaminants are calculated automatically from an open source database and an associated website accessible to scientific and stakeholders. Large events, including flash floods and sediment flushing, are also monitored at the river scale. Specific tools and protocols have been developed for monitoring sediment transfers, archiving data and sediment samples, and informing practitioners.L'Observatoire des Sédiments du Rhône, créé en 2009, regroupe des équipes de recherche multidisciplinaires (hydrologie, hydraulique, géomorphologie, géochimie, géomatique, sédimentologie) en interactions fortes avec les gestionnaires du fleuve. Son objectif est de fournir des connaissances scientifiques pour promouvoir une gestion durable du lit, des berges et canaux, ainsi que des processus sédimentaires du fleuve, l'un des plus importants de la mer Méditerranée. L'Observatoire couvre l'ensemble du tronçon français du fleuve. Cette présentation propose une synthèse des avancées majeures après 16 années de recherche. Ces travaux ont permis de mieux expliquer l'évolution du chenal au cours du dernier siècle, de caractériser les habitats aquatiques et ripariens, et d’évaluer la sensibilité du chenal aux changements. Un réseau de stations permet de mesurer en continu les concentrations de matières en suspension (MES) et les contaminants inorganiques et organiques associés. Les flux de MES et de contaminants sont calculés automatiquement à partir d'une base de données et d’un outil de diffusion accessible aux scientifiques et gestionnaires. Des événements majeurs tels que les crues et des évacuations de sédiments de barrage sont suivis à l’échelle du fleuve. Plusieurs outils et protocoles ont été développés pour caractériser les processus sédimentaires, accompagner les acteurs du fleuve dans sa restauration, archiver les données et les échantillons, et diffuser ces informations aux gestionnaires
Influence of Cell Position on the Capacity of Retired Batteries: Experimental and Statistical Studies
No full textInternational audienceUnderstanding how batteries perform after automotive use is crucial to determining their potential for reuse. This article presents experimental results aimed at advancing knowledge of retired battery performance. Three modules extracted from electric vehicles were tested. Their performance was assessed, and the results were analyzed statistically using analysis of variance (ANOVA). The 36 retired cells exhibited a high level of performance, albeit with significant variation. On average, the cells had a 95% state of health capacity with a dispersion of 2.4%. ANOVA analysis suggests that cell performance is not correlated with their position inside the module. These results demonstrate the need to evaluate dispersion within retired batteries and to develop thermal management and balancing systems for second-life batteries
Penser la généralisation quand il est question de restauration écologique : évolution des pratiques scientifiques dans les études menées sur le Rhône
No full textInternational audienceAs part of the ecological restoration of the Rhône River which started in the late 1990s, a group of French and Swiss scientists set up the RhônEco program to assess its impact. After more than 20 years of research described as interdisciplinary and co-constructed between scientists and managers, the program was renewed in 2024, with a major thematic and methodological extension, accompanied by a significant change in its members. These changes seem particularly interesting to study from a Science and Technology Studies (STS) perspective, which has highlighted the importance of questioning scientific production in relation to the social, political and economic contexts in which it is embedded. The aim is to examine the potential evolution of scientific practices and discourses in this changing context. A campaign combining participant observation and semi-structured interviews (n=22) was carried out with scientists involved in the RhônEco program. The survey highlighted the emergence of a nomothetic epistemology, which relies mainly on bibliographical and theoretical contributions in the construction of its questions and methodologies, and aims to provide generalizing answers. This differs from the idiographic epistemology already present in the program, which is mainly based on an inductive approach associated with empirical observations, emphasizing the specificity of the Rhone River case study and aiming at a holistic understanding of the system.Dans le cadre des mesures de restauration écologique initiées sur le fleuve Rhône à la fin des années 1990, un collectif de scientifiques franco-suisse s’est structuré au sein du programme RhônEco pour évaluer leurs effets. Après plus de 20 ans de recherches décrites comme interdisciplinaires et co-construites entre scientifiques et gestionnaires, le programme fait l’objet en 2024 d’un renouvellement générant un fort élargissement thématique et méthodologique associé à un turnover important de ses membres. Ces changements semblent particulièrement intéressants à étudier du point de vue des Sciences and Technology Studies (STS) ayant montré l’importance d’interroger la production scientifique au regard des contextes sociaux, politiques, et économiques dans lesquels elle s’inscrit. Il s’agit ainsi de questionner la potentielle évolution des pratiques et des discours scientifiques dans ce contexte en mouvement. Une campagne mêlant observation participante et entretiens semi-directifs (n=22) a été menée auprès de scientifiques impliqués dans différentes générations du programme RhônEco. L’enquête met en évidence l’affirmation d’une épistémologie nomothétique mobilisant principalement les apports bibliographiques et théoriques dans la construction de ses questions, de ses méthodologies et visant in fine à fournir des réponses généralisantes. Celle-ci se différencie de l’épistémologie idiographique majoritaire par le passé et fondée principalement sur une approche inductive associée à des observations empiriques mettant l’accent sur la spécificité de l’étude de cas rhodanienne, tout en développant une compréhension holistique du système fluvial
DGtal 2.0
No full textWe are happy to announce the major release 2.0 of DGtal. This big update makes the library more modern, in both how it is built and the tools it uses. This release also includes new features in both the C++ library and its python binding. Let us take a look at the big changes (see more in the Changelog section).* New Polyscope-based viewerThe new version of DGtal (2.0) includes a brand new viewer that is based on Polyscope. This modern viewer lets you customise things like colors, transforms, clipping planes and more while you’re using it. This is also the perfect time to make the viewer’s code simpler and easier to use. A guide is available to help you adapt old codes to the new viewer.* DGtal Paraview PluginDGtal now has a plugin for paraview which is available at https://github.com/DGtal-team/DGtal-paraview. The main motivation is to use DGtal geometry processing tools in a VTK/Paraview pipeline, widely used in scientific data visualization (medical imagine, material sciences…). For now, this plugin binds most of the Shortcuts and Geometry Shortcuts with a mapping between DGtal and vtk/paraview voxel-based objects.* New Geometry ShortcutsShortcuts are now available for Corrected Normal Current curvature tensor estimators estimators.The Geometry Shortcuts now contain volumetric processing tools (Distance transformation and Voronoi maps)Updated Python BindingsShortcuts are also available in the Python wrapper.* Building the libraryDGtal now uses more modern CMake scripts, which involves creating sublibraries in a top-down structure.All DGtal CMake variables now have the prefix “DGTAL_”.Most dependencies are now available through FetchContent, so the user doesn’t need to pre-install them at the correct location or specify any path. This version includes:BoostEigenPoncaPolyscope (see viewer section below)Patate was upgraded to Ponca for point cloud based differential estimators.GMP is no longer used; instead, we only rely on Boost Multiprecision (with its own backend) for arbitrary precision integers (enabled by default).Qt and libqglviewer are no longer required for the viewer.CI and CD are now built on top of actions that can be reused. These actions are used with side repositories.Breaking ChangesThe new minimal standard is now C++20, which replaced the previous standard of C++14. This has a few impacts. Some of the side effects of DGtal that you might notice are:You can no longer convert enums to int.Old std::allocator API is being removed (this was used in MPolynomial).New code might not work with older versions because it uses new libraries and language features.Removal of the deprecated namespace. Classes, functions and variables inside this namespace are no longer accessible. This is mostly about convolution on surfaces, which is now replaced with LocalEstimatorFromSurfelFunctorAdapter.The Viewer is now based on Polyscope (see the section above for more details). The stream API was kept the same, but most of the modifiers were taken out. See the Migrating viewer page
Evolutions technologiques et méthodologiques dans la caractérisation de la température par infrarouge thermique aéroporté
No full textInternational audienceThere is growing interest in using airborne TIR as a thermal diagnostic tool for rivers, in particular to assess the presence of potential thermal refuges. While some aspects of the technology are evolving and providing easier access to low-cost equipment, there are still limits as to how results can be used to address some of the issues (e.g. absolute temperatures). After a short description of the latest advances, the article presents the results of both technological (testing of a drone sensor) and methodological comparisons (semi-automated detection of cool-water patches). The main aim of this work is to move away from the more subjective methods used to date, especially the visual detection of thermal refuges, and their classification according to existing typologies. The paper ends with a proposal for a simpler and more parameter-based classification of cool-water patches, which remains comparable to the visual method used so far and so will open the door to large-scale analyses of data from different period, different rivers and even different countries.L’utilisation de l’IRT-a comme outil de diagnostic thermique des rivières fait l’objet d’un intérêt croissant, notamment pour évaluer la présence de refuges thermiques potentiels. Si certains aspects de la technologie évoluent et permettent un accès facilité à du matériel à bas coût, des limites subsistent sur l’utilisation des résultats pour répondre à certaines problématiques (ex. températures absolues). Après une description des dernières avancées, l’article présente les résultats issus de comparatifs à la fois technologiques (test d’un capteur drone) et méthodologiques (détection semi-automatique des taches froides). L’objectif principal de ce travail est de s’éloigner des méthodes jugées subjectives utilisées jusqu’ici, notamment la détection visuelle des refuges thermiques, puis leur classification suivant des typologies existantes. L’article se termine par une proposition de classification des taches d'eau froide plus simple et davantage basée sur leurs paramètres, qui reste comparable à la méthode visuelle utilisée jusqu'à présent, ouvrant alors la porte à des analyses à grande échelle de données provenant de périodes différentes, de rivières différentes voire même de pays différents
Design and validation of an individualizable thermophysiological model to assess thermal, hydric and cardiovascular strain
No full textInternational audienceAs heatwaves become more frequent, severe and longer, populations, especially urban population, become more vulnerable in part because of aging. Understanding heat-related symptoms and identifying vulnerability factors are crucial for risk prevention. While thermophysiological simulations enable the study of effects of heat stress at an individual scale, existing models lack the ability to simulate thermal, hydric and cardiovascular strain considering vulnerability factors. Therefore, this paper proposes an adaptation of the JOS-3 multi-segment multi-node model. The adadpted model, called aJOS-3, incorporates water balance, heart rate calculation, and the modeling of thermoregulatory limitations linked to dehydration and cadiovascular capacity. The adapted model is validated through comparison with simulation and experimental data from the literature. A sensitivity analysis is also carried out to address concerns regarding the individualization of the adapted model, aJOS-3. This model proves to be relevant for comparing different situations to assess heat strain responsible for heat-related symptoms, although estimating water intake and cardiac index remains a challenge for heat strain prediction