Portail HAL edf
Not a member yet
11029 research outputs found
Sort by
Some Comments on Cavitating Flow Simulations in Venturi and Hydrofoil Geometries
No full textInternational audienceSince the first studies published by [1,2], many numerical and physical models have been applied to simulate and analyze cavitating flow simulations in simple geometries (as hydrofoils, injectors or Venturi devices, for example), as well as in more complex geometries of hydraulic machines (pumps and turbines). Throughout these 30 years, the LEGI laboratory (Laboratoire des Ecoulements Géophysiques et Industriels) has actively participated in several research works applying different approaches and geometries. Homogeneous [3-6] and two-phase flow models [7] have been developed and evaluated for several turbulent cavitating flows, taking into account thermodynamic effects or not. This paper presents a synthesis of the most important results obtained in the laboratory on Venturi and hydrofoils configurations for cold water. It also describes the problems encountered, lessons learned and experiences acquired. The review highlights some conclusions on the influence of cavitation, turbulence and numerical models on the prediction of unsteady behavior of cavitating flows
Short-term investigation of interactions of plain concrete, marine microorganisms, and sea water for the application of Floating Offshore Wind Turbines (FOWTs)
No full textInternational audienceFOWTs operate in deep waters. A durable concrete cover in the submerged zone is critical in immersed components because it prevents the transport of aggressive ions, and thus protects the reinforcement from corrosion. The aim of this interdisciplinary study is to identify the surface interactions between cementitious materials, biofouling at their surface, and seawater with emphasis on, (i) the influence of the material on the composition and structure of biofilm, and (ii) the influence of biofilm and seawater on mineralogical and chemical changes within cementitious matrix. Mortars and concrete are prepared using two low-CO2 binders incorporating supplementary cementitious materials, CEM III and CEM V, as well as a conventional binder, CEM I. The specimens were submerged at a depth of 27m at SOLA station, Banyuls-sur-mer, France. Biofouling was characterized using Environmental Scanning Electron Microscope (ESEM) and environmental DNA analysis, whilst, microstructure, mineralogical, and chemical changes in cementitious materials were determined using SEM coupled to Energy Dispersive Spectroscopy (SEM-EDS), X-Ray Diffraction (XRD), and Electron Probe Micro Analysis (EPMA) respectively. Our results suggest that irrespective of the binder-type and age, earlyage exposure (after 30- and 90 days) leads to the precipitation of calcium carbonate, mainly in the form of aragonite at seawater/biofilm-concrete interface. EPMA findings on CEM I concrete indicate calcium leaching and Mg-precipitate on the outer layers. After three months, sulfur-rich and chloride-rich zones followed the Mg-rich zone. The microbial diversity in marine biofilms was determined as a function of binder-type, the differences in composition fading over time on concrete surface
Analyse comparative des modèles champs de phase pour la simulation de la propagation de fissure en milieux anisotropes
No full textInternational audienceL'étude de la fissuration dans les milieux anisotropes est de plus en plus importante en mécanique des structures.Notamment, les procédés de fabrication additive, appréciés pour leur capacité de réparabilité, sont en plein essor.Afin de garantir que leur emploi est sécurisé, il est nécessaire de comprendre et de modéliser leur dégradation mécanique.Cependant, la modélisation de la fissuration dans ces matériaux reste partielle.Bien que certains modèles puissent représenter qualitativement la fissuration, peu d'études quantitatives sont disponibles à ce jour.Cette étude propose une analyse comparative de différents modèles de simulation par champs de phase pour la propagation des fissures en milieux anisotropes.L'objectif est de développer des simulations prédictives précises et efficaces pour ces phénomènes.Une première analyse qualitative des modèles existants est réalisée pour identifier leurs similarités et différences, en termes de modélisation mécanique et d'implémentation numérique.Ensuite, différents modèles sont calibrés sur des données expérimentales existantes obtenues sur des éprouvettes fabriquées par dépôt de fil fondu.Les prédictions des différents modèles sont alors comparées aux résultats expérimentaux.Les résultats attendus de cette étude incluent une meilleure compréhension des performances relatives des différents modèles de simulation par champs de phase, ainsi que des recommandations pour leur utilisation dans des contextes pratiques. Cette recherche pourrait ainsi contribuer à améliorer la conception et la durabilité des structures produites par fabrication additive
Vers une transformation de modèle pour l'analyse de la fiabilité d'architecture de réseaux électriques
No full textInternational audienceElectrical energy is at the heart of our daily use and economic activities. As the demand for electrical energy is constantly increasing, the system that supplies it must be safe. Thus, the electricity system has evolved towards the concept of Smart Grid to cope with these various needs. As a result, the definition of standardized electrical system models is necessary. However, a challenge for industrial actors remains : How to ensure compliance of a more reliable electrical system, interoperable and integrating the different engineering fields requirements, standards, and technologies ? The adoption of electricity standards requires establishing an efficient framework for network data exchange at different levels of the electrical infrastructure. The safety level of installations must be guaranteed, whatever the configurations. Electricité De France (EDF) has developed the FIGARO language and the KB3-K6 tool that uses a reliability approach based on Markov models, to verify electrical system reliability and availability performances. To enhance data exchange between stakeholders in electrical domains, the Common Information Model standard (CIM) has been defined on an international scale in UML (Unified Modelling Language). In this paper, the main objective is to define needs in terms of smart grid data to perform safety studies in relation to standards such as the CIM. The methodology adopted is based on CIM concepts analysis and dependability concepts described in FIGARO language. This paper aims to define an approach to build reliability models from data and architecture elements of existing systems, based on a standard description of these models. The main result is a set of transformation rules between a CIM model concepts for dependability field and a FIGARO model concepts for reliability assessment.L'énergie électrique est essentielle pour nos usages quotidiens et industriels. Les besoins en énergie électrique augmentant constamment, le système qui assure sa fourniture doit être sûr. Ainsi, le système électrique a évolué vers le concept de Réseau Intelligent pour faire face à ces divers usages. Dès lors, la définition de modèles de systèmes électriques se standardise. Cependant, un défi pour les acteurs industriels persiste : comment assurer la cohérence d'un système électrique plus fiable, interopérable et intégrant les exigences liées aux métiers, normes et technologies du secteur de l'énergie électrique ? Cette cohérence questionne les mécanismes d'échange de données facilitant les interactions à différents niveaux de l'infrastructure. Pour le métier de la sûreté de fonctionnement, EDF a développé le langage FIGARO et l'outil KB3-K6 utilisant une approche basée sur des modèles de Markov, pour vérifier les exigences de fiabilité et de disponibilité de ses réseaux. Dans les domaines du transport et de la distribution de l'électricité, le standard CIM (Common Information Model) a été développé afin d'échanger des informations de réseaux électriques. Dans cet article, nous proposons une démarche pour identifier les correspondances entre des concepts du modèle CIM décrit en UML (Unified Modelling Language) et les concepts de la sûreté de fonctionnement décrits en langage FIGARO afin de définir une future transformation de modèles. Pour cela, il est nécessaire d'identifier le niveau sémantique des concepts des deux domaines. L'objectif est de définir une approche pour construire des modèles de fiabilité à partir de données et éléments d'architecture de systèmes existants, sur la base de modèles standards de description de réseaux électriques. Le principal résultat est un ensemble de règles de transformation entre les concepts d'un modèle CIM et les concepts d'un modèle FIGARO
An Algorithmic Framework for the Optimization of Deep Neural Networks Architectures and Hyperparameters
No full textInternational audienc
Thermodynamic architecture for steam generation using heat pumps system valorising industrial waste heats
No full textInternational audienc
Modélisation des contraintes résiduelles dans les soudures circonférentielles multi-passes bout à bout de tuyauterie en acier inoxydable austénitique pour la classification du risque de corrosion sous contrainte en milieu primaire
No full textNational audienceLa prévision des contraintes résiduelles dans les soudures est essentielle pour évaluer l'intégrité d'un composant soumis à des mécanismes de dégradation tels que la corrosion sous contrainte (CSC). Une vaste campagne de simulation de soudures circonférentielles multipasses bout à bout de tuyaut erie en acier inoxydable austénitique a été menée afin de couvrir différentes configurations de soudage. Un abaque numérique est construit sur la base d'un plan d'expériences de 100 simulations numériques d u soudage par couple diamètre-épaisseur. Cet ensem ble de simulations montre la présence systématique d'une zone de compression sur la moitié de l'épaisseur à partir de la surface interne, avec une certaine variabilité en fonction des conditions de soudage et de la géométrie. La classification du risque de CSC est alors définie par un ensemble de deux indicateurs qui peuvent être liés à l'initiation de la fissure et à sa propagation. La prise en compte des charge ments en service en plus des contraintes résiduelles de soud age réalistes rend l'analyse du fact eur d'intensité de contrainte réalisable en tenant compte de la position et de l'orientation idéales de la fissure
Flexible, Transparent, and Bifacial Perovskite Solar Cells and Modules Using the Wide-Band Gap FAPbBr 3 Perovskite Absorber
No full textInternational audiencePerovskite solar cells (PSCs) offer impressive performance and flexibility thanks to their simple, lowtemperature deposition methods. Their bandgap tunability allows for a wide range of applications, transitioning from opaque to transparent devices. We present the first demonstration of flexible, bifacial PSCs using the wide bandgap FAPbBr3 perovskite. The role of optimization for both electron and hole transport layers on bifaciality, transparency, and stability have been studied. PSCs achieved a maximum power conversion efficiency (PCE) of 6.8% and 18.7% under 1-Sun and under indoor light conditions, respectively, showing up to 98% bifaciality factor and an average visible transmittance (AVT) of 55%. Additionally, P1-P2-P3 laser ablation scheme has been developed on flexible PET substrate for perovskite solar modules showing PCE of 4.7% PCE and high geometrical fill factor (97.8%). These findings highlight the potential of flexible, bifacial PSCs for diverse applications like building-integrated PV, agrivoltaics, automotive tech, wearable sensors and IoT, and more. TOC GraphicWith growing global energy demand, it is necessary to move from fossil fuels to renewable energy resources. Solar energy is a promising solution and photovoltaics technology has made significant progress in recent years. Perovskite solar cells (PSCs) have achieved impressive power conversion efficiencies (PCEs) of up to 26% in a short timeframe 1,2 . PSCs are ideal for flexible photovoltaics, as they can be deposited using simple and low-temperature solution processing methods 3-6 . Another notable characteristic of PSCs is their remarkable band-gap tunability, which can be achieved through composition engineering 7,8 . The incorporation of bromide and chloride into the perovskite composition allows for the creation of highly transparent perovskites with band gaps as high as 2.39 eV 8 . Furthermore, PSCs exhibit excellent performance in the blue-light region, making them well suited as light harvesters for diffused sunlight on cloudy days and low-intensity indoor lighting during nights 9-11 . This, combined with flexibility provide an easy installation option to seamlessly integrate PVs into surfaces such as building facades, windows and walls, sensors, electric vehicles, wearable devices, and smart displays and act as replacements for conventional energy sources 7,9,12-15 .Unlike opaque photovoltaics, ST-PVs aim to maximize both average visible transmittance (AVT) and PCE simultaneously, considering the trade-off between AVT and PCE. 14,16,17 . Consequently, a new metric, called Light Utilization Efficiency (LUE), has emerged, which represents the product of PCE and AVT 14 . LUE quantifies the potential of emergent ST-PV technologies in converting incoming visible light into electricity while maintaining rather high AVT. A suitable for applications such as smart windows, low-power displays, and automotive industry ST-PV technology should guarantee AVT greater than 50-80% 14 . The optimal bandgap range to achieve AVT values of 50-80% is typically between 2.18 and 2.32 eV 9 . Formamidinium lead bromide (FAPbBr3) perovskite, which has a band gap of 2.23 eV, falls well within the optimal band gap range for achieving high AVT, and has already demonstrated AVT values exceeding 50% 18,19 . The PCE of opaque FAPbBr3 solar cells typically reaches around 10% 20-22 . However, when using semitransparent back contact, PCE decreased to approximately 8% 18,19,23 . This reduction in efficiency is primarily attributed to the lack of back-reflection of the metallic back contact, which results in the reduced collection of backscattered light for harvesting. Furthermore, semitransparent solar cells often employ a thinner absorber layer to enhance the AVT of the cell. FAPbBr3-based semitransparent PSCs exhibit minimal angular dependence of PCE, making them highly attractive for window-integrated applications 19 . When deposited on flexible substrates, these cells also experience a decrease in PCE, with the highest reported PCE of 5.0% for flexible opaque cells 24 and no flexible semitransparent PSCs (ST-PSCs) reported using this perovskite to the best of our knowledge. In this study, we fabricated flexible, bifacial PSCs using FAPbBr3 perovskite as absorber. The devices have architecture of PET/ITO/SnO2/FAPbBr3/HTL/ITO (Figure 1a). First, we examined the impact of potassium (K) treatment on the SnO2 used as electron transport layer (ETL). Secondly, we studied the effect of hole transport layer (HTL) on the device bifaciality factor and long-term stability. Lastly, we fabricated flexible, bifacial perovskite solar modules for the first time with high geometrical fill factor of 97.8%. The fabricated flex-ST-PSCs demonstrated high bifaciality factor achieving similar PCEs from both sides.For SnO2-based PSCs, treatment with potassium ionic salts has become a very common passivation method to improve device performance 25-28 . The cations and/or anions of these salts play various roles in increasing the performance of PSCs. For example, they can passivate defects in SnO2 or perovskite layer, as well as act as preferred nuclei for perovskite formation by forming ionic bonds with perovskite precursors 25 . We performed potassium treatment to investigate its effectiveness in wide-bandgap bromide-based perovskite, FAPbBr3, using potassium chloride (KCl) salt. Figure 1b presents the statistical</div
Toward Agile Interaction Model based ontology development Methodology (AIME) for FAIR European data spaces
No full textInternational audienceThe European Union's initiative on European data spaces aims to support innovation and economic growth by facilitating secure, interoperable data sharing across key sectors such as health, energy, and mobility. Within European data spaces, high priority is given to adherence to FAIR data principles. Thus, data and services should be findable, accessible, interoperable, and reusable. Ontologies as semantic models are fundamental in achieving semantic interoperability, as they provide a common framework for data and service access, discovery, understanding and reuse. However, the inherent complexity of data space initiatives, coupled with their underlying domains diversity are key challenge facing optimal FAIRness. This motivates the need for an agile methodology that aligns with data space principles, integrates existing domain standards, consolidates the use of metadata for ontology's FAIRness, and engages various involved actors (domain experts, data service providers, ontology engineers) towards the development of common data space ontology. In the frame of the OMEGA-X project, aiming to build an energy data space, this paper presents AIME, an ontology development methodology integrating the reuse of both standards and reference ontologies to enable the development of modular ontologies for data spaces. It leverages agile principles to strengthen communication between various stakeholders through different steps: reference standards and ontologies selection, selection of use cases, design and selection of interaction models capturing data exchanges, ontology modules creation, automation and continuous integration features to support the entire original ontology engineering phaseand reuse by data space users.</div