Portail HAL edf
Not a member yet
11029 research outputs found
Sort by
Multi-resource integration associated with the deployement of hydrogen and derivatives in an industrial hub.
No full textInternational audienceThe objective of this paper is to propose optimal strategy for the deployment of hydrogen in an industrialhub subject to several constraints and following an eco-industrial park (EIP) scheme. The study wasperformed using a multi-period MILP model that minimizes the total annualized cost of the overallsystem. In practice, the attractiveness of an industrial hub regarding the production of Fischer-Tropschfuels from electrolytic hydrogen and captured carbon dioxide from a nearby steel industry wasevaluated. Key results indicate that Fischer-Tropsch (FT) processing plant can be perfectly integratedin industrial hub via the design of EIP. In such scheme, global economic gain due to synthetic fuelssales, heat valorization and economies of scale regarding hydrogen production, can be obtained givenan appropriate market price of the commodities. When studying legislation-constrained scenarios,considering the European legislation on the production of renewable fuels of non-biological origin andthe current ETS market structure, the benefit of designing such EIP becomes controversial, unless FTfuels are supported on the market by environmental regulation. Biogenic source of carbon dioxideshould also be available in quantity. Otherwise, the current announced projects might be called off andother production routes of sustainable fuels might be favored
Un exemple de partage de l’espace : les loisirs nautiques sur la Loire
No full textInternational audienceAs France's longest river, the Loire in Nantes has acquired a maritime character that has forged much of its history and economic development for centuries. With 30 million tonnes of traffic per year, Nantes Saint-Nazaire Port is France's 4th largest seaport, and 28,500 jobs are generated by port activity (according to an Insee 2022 study based on 2018 data).Beyond this economic aspect, the Loire can be seen as the backbone of the Nantes conurbation, whose 24 communes are spread along its 2 banks. Well-known for its civil engineering structures, such as the Cheviré bridge, or its maritime heritage, such as the Canal Maritime de la Basse Loire (known as the Canal de la Martinière), the urban area is also home to over 650,000 inhabitants. And their leisure time is often spent on the water!In this presentation, we'll explain how sports associations have come together to form the Collectif des Association Nautiques de l'agglomération nantaise, to ensure that leisure boating, which was virtually absent from the Loire 30 years ago, regains its place alongside other activities on the Loire and its tributary, the Erdre, enabling residents to discover their city from the water, using gentle means of transport.Plus long fleuve de France, la Loire acquiert, à Nantes, un caractère maritime qui a forgé une bonne partie de son histoire et de son développement économique depuis des siècles. Avec un trafic de 30 millions de tonnes de trafic par an, Nantes Saint-Nazaire Port est le 4e grand port maritime français et 28 500 emplois sont générés par l'activité portuaire (selon une étude Insee 2022 sur données 2018).Au-delà de cet aspect économique, la Loire peut être représentée comme l’épine dorsale de l’agglomération nantaise dont les 24 communes se répartissent sur ses 2 rives. Bien connue en termes de génie civil pour ses ouvrages d’art comme le pont de Cheviré ou des éléments du patrimoine maritime comme le canal maritime de la Basse Loire (dit canal de la Martinière), l’agglomération est aussi le lieu de vie de plus de 650000 habitants. Et leurs loisirs regardent souvent vers l’eau !Dans cette communication, nous présenterons comment les associations sportives se sont structurées en un Collectif des Association Nautiques de l’agglomération nantaise pour permettre que la navigation de loisir, quasi absente sur la Loire il y a 30 ans, reprenne sa place, au même titre que les autres activités, sur la Loire et son affluent l’Erdre, en permettant aux habitants de découvrir leur ville à partir de l’eau, en mode déplacement doux
Multi-domain encoder–decoder neural networks for latent data assimilation in dynamical systems
No full textInternational audienceHigh-dimensional dynamical systems often require computationally intensive physics-based simulations, making full physical space data assimilation impractical. Latent data assimilation methods perform assimilation in reduced-order latent space for efficiency but struggle with complex, nonlinear state-observation mappings. Recent solutions like Generalized Latent Data Assimilation (GLA) and Latent Space Data Assimilation (LSDA) address heterogeneous latent spaces by incorporating surrogate mapping functions but introduce computational costs and uncertainties. Furthermore, current algorithms that integrate data assimilation and deep learning still face limitations when it comes to handling non-explicit mapping functions. To address these challenges, this paper introduces a novel deep-learning-based data assimilation scheme, named Multi-domain Encoder–Decoder Latent Data Assimilation (MEDLA), capable of handling diverse data sources by sharing a common latent space. The proposed approach significantly reduces the computational burden since the complex mapping functions are mimicked by the multi-domain encoder–decoder neural network. It also enhances assimilation accuracy by minimizing interpolation and approximation errors. Extensive numerical experiments from three different test cases assess MEDLA’s performance in high dimensional dynamical systems, benchmarking it against state-of-the-art latent data assimilation methods. The numerical results consistently underscore MEDLA’s superiority in managing multi-scale observational data and tackling intricate, non-explicit mapping functions
Clouds of points optimization in convex polygons with evolutionary algorithms based on Wasserstein barycenters
No full textInternational audienceWe consider the optimization of functions where the variables are clouds of points (equivalently bags of vectors). The clouds can have different sizes and the objective functions are invariant under arbitrary permutation of the points within the cloud. Furthermore, no information related to the convexity and/or smoothness of the functions is available. Such functions are of practical interest as they appear, for example, when studying networks of sensors or actuators. The design of a wind farm where the variables are a bag of turbine positions belongs to this family of problems.The above characteristics make it hard to use off-the-shelf algorithms such as gradient-based methods. We introduce a generic approach for such black-box optimization problems where the cloud is modeled as a discrete uniform measure supported by the cloud points. This allows to define stochastic, evolutionary, optimization algorithms whose transition operators (crossover and mutation) use Wasserstein barycenters (see [1] for a review). The crossover operator interpolates between two clouds of points, by calculating their average in the Wasserstein sense. We give an illustration of such an interpolation on clouds in Figure 1. We prove that the Wasserstein barycenter has a contracting effect, which can lead to a premature convergence to degenerated solutions. Specific mutations, once again based on Wasserstein barycenters, are designed to counteract this effect.We first investigate the performance of variants of our Wasserstein-based optimizers by comparing them to classical evolutionary algorithms on a family of test functions including wind farm proxies. The point domains are convex polygons such as squares and trapezes. Second, the points optimization domain is generalized to involve exclusion zones, defined as subdomains which must not include any point of the optimal solution. Exclusion zones constitute a specific type of constraint which is handled by penalization
Experimental investigation of a new CO2 refrigeration system arrangement for supermarket applications
No full textInternational audienceA new position of an internal heat exchanger within the thermodynamic cycle is proposed to improve the performance of CO2 refrigeration systems in warm climates, specifically in supermarket applications. The proposed configuration uses the saturated vapor from the liquid receiver to recover the heat that is rejected by a subcooler positioned after the gas cooler. This system results in a reduced evaporator superheat. This configuration is compared with a reference system.The experimental study is conducted on a CO2 laboratory cooling system reproducing a commercial plant. The cooling capacity of the system is 30 kW. The water inlet temperature in the gas cooler (hot source) varies from 15 °C to 35 °C, while the mono-ethylene glycol outlet temperature of the evaporator (cold source) is fixed at -8 °C. The setup is validated through an energy balance.Using a subcooler in this new position improves the coefficient of performance by 10.2 % under transcritical conditions and 6.3 % under subcritical conditions when compared with the reference system
La méthodologie ACIMOV pour l'intégration agile et continue des modules ontologiques
No full textInternational audienceCe travail décrit la méthodologie d’ingénierie d’ontologie Agile and Continuous Integration for Modular Ontologies and Vocabularies (ACIMOV) pour développer des ontologies et des vocabulaires. ACIMOV étend la méthodologie agile SAMOD pour mieux prendre en compte l’aspect modulaire des ontologies actuelles et le développement collaboratif entre différents types d’experts. ACIMOV adopte les principes de développement agile et DevOps. ACIMOV a été conçue pour être opérationnalisée à l’aide de plateformes de développement logiciel collaboratif Git et dotées de workflows d’intégration et de déploiement continus. Ces travaux ont été publiés dans un atelier international [4]
Signaling through cAMP-Epac1 induces metabolic reprogramming to protect podocytes in glomerulonephritis
No full textInternational audienceUnlike classical protein kinase A, with separate catalytic and regulatory subunits, EPACs are single chain multi-domain proteins containing both catalytic and regulatory elements. The importance of cAMP-Epac-signaling as an energy provider has emerged over the last years. However, little is known about Epac1 signaling in chronic kidney disease. Here, we examined the role of Epac1 during the progression of glomerulonephritis (GN). We first observed that total genetic deletion of Epac1 in mice accelerated the progression of nephrotoxic serum (NTS)-induced GN. Next, mice with podocyte-specific conditional deletion of Epac1 were generated and showed that NTS-induced GN was exacerbated in these mice. Gene expression analysis in glomeruli at the early and late phases of GN showed that deletion of Epac1 in podocytes was associated with major alterations in mitochondrial and metabolic processes and significant dysregulation of the glycolysis pathway. In vitro, Epac1 activation in a human podocyte cell line increased mitochondrial function to cope with the extra energy demand under conditions of stress. Furthermore, Epac1-induced glycolysis and lactate production improved podocyte viability. To verify the in vivo therapeutic potential of Epac1 activation, the Epac1 selective cAMP mimetic 8-pCPT was administered in wild type mice after induction of GN. 8-pCPT alleviated the progression of GN by improving kidney function with decreased structural injury with decreased crescent formation and kidney inflammation. Importantly, 8-pCPT had no beneficial effect in mice with Epac1 deletion in podocytes. Thus, our data suggest that Epac1 activation is an essential protective mechanism in GN by reprogramming podocyte metabolism. Hence, targeting Epac1 activation could represent a potential therapeutic approach
Development of a method for acoustic monitoring of the pressure and composition of the gas in the fuel rods of a PWR assembly in a nuclear spent fuel pool
No full textInternational audienceThe internal pressure and molar composition of the fission gases (xenon/krypton and helium) present within the free volumes of the fuel rod after irradiation is of major interest to nuclear energy suppliers. This parameter is a fuel behavior indicator and reflects the overall fuel performance in operation during shipping and long-term storage. Rod internal pressure is one criterion amongst others, like cladding corrosion, against which the acceptable fuel burn-up limit is set. IES, EDF and SONAXIS SA are developing a non-destructive acoustic process to perform measurements, directly on fuel assemblies in the spent fuel pools of nuclear power plants. This will make it possible to carry out extensive measurement campaigns to support and improve thermodynamic evolution models, and to obtain more consistent measurement statistics. The principle of the measurement is to determine the speed and attenuation of the acoustic waves propagating in the gas at the level of the tube plenum. A piezoelectric transducer, driven by a pulse generator, generates the acoustic waves in the fuel rod which acts as a resonant cavity. The signal by the transducer, of a reflected sound wave through the gas, is then analysed. The composition is determined by measuring the time of flight of the acoustic signal and the pressure can be estimated by a calibrated process analyzing the amplitude of the signal. The aim of this paper is to present the latest work which has enabled to develop operational sensors which are sufficiently miniaturized to be used in the bundle of a fuel assembly. A specific signal processing software has been developed and it was demonstrated that acoustic non-destructive measurement of pressure and composition of fission gas in a configuration close to that of an assembly in a spent fuel pool can be achieved
Wave–structure interaction by a two–way coupling between a fully nonlinear potential flow model and a Navier–Stokes solver
No full textInternational audienceA two-way domain decomposition coupling procedure between a fully nonlinear potential flow model and a Navier–Stokes solver capturing the free surface with a Volume of Fluid method is used to study wave–structure interaction applied to offshore wind turbines. Away from the structure, the large-scale inviscid wave field is modeled by the potential code. Wave generation and absorption in this 3D hybrid model take place in the outer potential domain. The codes exchange data in the region around their common boundaries. Through the two-way coupling, waves propagate in and out of the viscous subdomain, making the hybrid algorithm suitable to study wave diffraction on marine structures, while keeping the viscous subdomain small. Each code uses its own mesh and time step. Subdomains are overlapping, therefore continuity conditions on velocity and free surface have to be verified on two distinct coupling surfaces at any time. Parallel implementation with communications between the models relying on the Message Passing Interface library allows calculations on large spatial and temporal scales. The coupling algorithm is first tested for regular nonlinear waves and then applied to simulate wave loads exerted on a vertical monopile in 3D. Attention is paid to the high-order components of the horizontal force
Secondary organic aerosol formed by Euro 5 gasoline vehicle emissions: chemical composition and gas-to-particle phase partitioning
No full textInternational audienceIn this study we investigated the photo-oxidation of Euro 5 gasoline vehicle emissions during cold urban, hot urban and motorway Artemis cycles. The experiments were conducted in an environmental chamber with average OH concentrations ranging between 6.6 × 105–2.3 × 106 molec. cm−3, relative humidity (RH) between 40 %–55 % and temperatures between 22–26 °C. A proton-transfer-reaction time-of-flight mass spectrometer (PTR-ToF-MS) and the CHemical Analysis of aeRosol ON-line (CHARON) inlet coupled with a PTR-ToF-MS were used for the gas- and particle-phase measurements respectively. This is the first time that the CHARON inlet has been used for the identification of the secondary organic aerosol (SOA) produced from vehicle emissions. The secondary organic gas-phase products ranged between C1 and C9 with one to four atoms of oxygen and were mainly composed of small oxygenated C1–C3 species. The SOA formed contained compounds from C1 to C14, having one to six atoms of oxygen, and the products' distribution was centered at C5. Organonitrites and organonitrates contributed 6 %–7 % of the SOA concentration. Relatively high concentrations of ammonium nitrate (35–160 µg m−3) were formed. The nitrate fraction related to organic nitrate compounds was 0.12–0.20, while ammonium linked to organic ammonium compounds was estimated only during one experiment, reaching a fraction of 0.19. The SOA produced exhibited log C∗ values between 2 and 5. Comparing our results to theoretical estimations for saturation concentrations, we observed differences of 1–3 orders of magnitude, indicating that additional parameters such as RH, particulate water content, aerosol hygroscopicity, and possible reactions in the particulate phase may affect the gas-to-particle partitioning