Portail HAL edf
Not a member yet
11029 research outputs found
Sort by
Évolution de la ressource en eau et de la demande en eau dans une station de sports d’hiver dans le contexte du changement climatique
No full textInternational audienceWater is a key factor for adapting to climate change. Its role is particularly crucial in ski resorts, where water use for snowmaking has become one of the main adaptation strategies to cope with worsening snow conditions on the ski slopes. This study presents a method for comparing the water availability with the evolution of its demand for snowmaking (estimated using the ClimSnow climate service), while taking into account other water uses. The results of this approach, based on the MORDOR-TS hydrological model, are illustrated for the Valloire ski resort (Savoie, France). By midcentury, mean annual water flows are expected to decrease only slightly, but the hydrological regime of rivers would remain largely unchanged. By the end of the century and for the RCP 8.5 scenario, however, average annual flows are expected to fall by around −20%, with a shift from a nival to a nivopluvial regime, an earlier melting wave, an increase in winter flows and a significant reduction in late summer flows. Regarding snowmaking, a detailed analysis of the number of times the water reservoir can be filled reveals an increasing pressure on the reservoir’s filling capacity between July and September, and a pressure that is projected to decrease in winter.L’eau constitue un facteur central pour l’adaptation au changement climatique. C’est le cas pour la production de neige utilisée par les domaines skiables en réponse à la dégradation de leurs conditions d’exploitation. Cette étude présente une méthode permettant de confronter la disponibilité de la ressource en eau à l’évolution de la demande pour la production de neige, estimée avec le service climatique ClimSnow, en tenant compte d’autres usages de l’eau. Les résultats de cette approche, s’appuyant sur le modèle hydrologique MORDOR-TS, sont illustrés sur le cas de Valloire (Savoie, France). D’ici le milieu du siècle, les apports annuels moyens devraient diminuer légèrement, mais le régime hydrologique des rivières resterait globalement inchangé. En fin de siècle et pour le scénario RCP 8.5, les apports annuels moyens devraient diminuer d’environ −20 %, passant d’un régime nival à nivo-pluvial, avec une onde de fonte plus précoce, une augmentation des débits hivernaux et une aggravation significative des étiages de fin d’été. En ce qui concerne la production de neige, une analyse du nombre de remplissages de la retenue d’eau révèle une pression croissante sur la capacité de remplissage de la retenue entre juillet et septembre, et une pression en baisse en hiver
Analyse mathématique et numérique des modes d'un guide d'ondes électromagnétiques hétérogène.
No full textInternational audienceIn the homogeneous case, i.e. with constant epsilon and mu, the modes (E_n, H_n, \beta_n) are easily obtained by solving scalar problems in the section S of the guideand are pairwise orthogonal in L^2(S). They are either propagating (\beta in R) or purely evanescent (\beta in iR) and they have phase and group velocities of the same sign.For heterogeneous guides, i.e. with varying epsilon and mu in the section, these properties are generally not true and the mathematical analysis of the modes is much more delicate.In this talk, we present different formulations to study them and discuss their respective advantages.For strong variations of epsilon and/or mu, we show numerically that inverse modes, with group and phase velocities of opposite sign, can exist.Such cases for which PMLs fail to capture the outgoing solution are one of the reasons why we develop modal transparent conditions.Dans le cas homogène, c'est-à-dire avec epsilon et mu constants, les modes (E_n, H_n, \beta_n) s'obtiennent facilement en résolvant des problèmes scalaires dans la section S du guide et sont deux à deux orthogonaux dans L^2(S). Ils sont soit propagatifs (\beta dans R), soit purement évanescents (\beta dans iR) et ont des vitesses de phase et de groupe de même signe.Pour des guides hétérogènes, c'est-à-dire avec epsilon et mu variables dans la section, ces propriétés sont généralement fausses et l'analyse mathématique des modes est beaucoup plus délicate.Dans cet exposé, nous présentons différentes formulations pour les étudier et discutons de leurs avantages respectifs.Pour de fortes variations d'epsilon et/ou de mu, nous montrons numériquement que des modes inverses, avec des vitesses de groupe et de phase de signe opposé, peuvent exister.Les cas pour lesquels les PML ne parviennent pas à capturer la solution sortante sont l'une des raisons pour lesquelles nous développons des conditions de transparence modale
An entropy penalized approach for stochastic optimization with marginal law constraints. Complete version
No full textInternational audienceThis paper focuses on stochastic optimal control problems with constraints in law, which are rewritten as optimization (minimization) of probability measures problem on the canonical space. We introduce a penalized version of this type of problems by splitting the optimization variable and adding an entropic penalization term. We prove that this penalized version constitutes a good approximation of the original control problem and we provide an alternating procedure which converges, under a so called "Stability Condition", to an approximate solution of the original problem. We extend the approach introduced in a previous paperof the same authors including a jump dynamics, non-convex costs and constraints on the marginal laws of the controlled process. The interest of our approach is illustrated by numerical simulations related to demand-side management problems arising in power systems
Physics‐Based Machine Learning Electroluminescence Models for Fast yet Accurate Solar Cell Characterization
No full textInternational audienceElectroluminescence analyses of solar cells and modules allow for fast, cost‐effective, and nondestructive spatial characterization of devices at different stages of their development and use. Voltage‐dependent electroluminescence (ELV) measurements have been shown to mimic diode voltage–current characteristics. A derived physical model enables the determination of two local pseudoparameters from ELV data measured on silicon solar cells: a pseudorecombination current and a pseudoseries resistance . Local characteristics of the solar cells, such as the series resistance or the dark saturation current , can be deduced from these pseudoparameters. ELV measurements are stored in large data cubes, typically containing a few hundred thousand pixels. Pixel‐wise regression is commonly achieved through nonlinear least squares (NLLS) minimization; knowing that a luminescence image of a 6 ′ ′ silicon solar cell contains about 1 Mpix, this method is time‐consuming, necessitating a trade‐off between sample size, spatial resolution, fitting accuracy, and computation duration. We hence propose to replace NLLS fitting with machine learning (ML) techniques, known for their efficiency in rapidly processing large datasets. We compare the regression performances of a multilayer perceptron (MLP) with the ones of a convolutional neural network (CNN) called modified U‐NET (mU‐NET). The first ML model conducts a pixel‐wise analysis of the data cube and the second processes the entire data cube in a single step. We present a comprehensive characterization of prediction accuracy, objectively assessing the advantages and limitations of the proposed techniques. Our first step is to ensure that the prediction precision is sufficient for a valid comparison of the analysis duration. The deviation of accuracy of these models compared to NLLS is almost negligible for MLP and of 3.1 % when employing mU‐NET, demonstrating their relevancy for operational application. Both ML models are fast and efficient: the time required for regression decreases by a factor of 240 with the MLP and by a factor of 1200 with the mU‐NET, compared to the NLLS method
Reduced-order modeling for nonlinear vibrations of structures
No full textInternational audienceThis chapter is devoted to the presentation of model-order reduction techniques that are used in the field of structural vibrations. A special emphasis is placed on substructuring methods for localized nonlinearities and on nonlinear normal modes defined via invariant manifolds for distributed smooth nonlinearities, as key tools to perform efficient yet accurate dimensional reductions. Other reduction techniques such as proper orthogonal decomposition, implicit condensation, and modal derivatives are also briefly covered at the end of the survey. The contents of this chapter were written for the Handbook of Nonlinear Dynamics during the Summer of 2024
Influence of the mesh on the crack path in phase-field fracture simulations
No full textMeeting of the 10th GAMM workshop on phase-field modeling and the workshop Materials/Microstructure modelling: Analytics & Benchmarks organized and hosted by KIT with support by the DGM.International audienceOver the past 25 years, phase-field fracture models [1, 2] have become increasingly popular for modeling crack propagation. In particular, their (Γ-)convergence towards the Linear Elastic Fracture Mechanics (LEFM) provides strong theoretical foundations. Despite this popularity, limitedresearch has been conducted on how spatial discretization (e.g., mesh size, structure, and element geometry) affects the predicted crack path. This study addresses this gap from the perspective of the mechanical engineering community. We employ a benchmark problem inspired by the PureShear test [3] (also called strip specimen), involving an infinite strip with an initial horizontal edge crack located above the specimen center and subjected to tensile loading. The crack path is expected to deviate towards the center of the specimen exponentially. This result has been recovered using an incremental crack propagation solver based on LEFM, which serves as our reference. Phase-field fracture simulations, performed using the Finite Element Method, are then carried out. Different meshes (varying mesh size, structured/unstructured, and element geometry) are used in the simulations to assess their influence on the crack path. The bias induced by the mesh is evaluated by comparing the phase field simulation results with the reference. The final goal of this study is to provide recommendations to avoid, or at least mitigate, any bias induced by spatial discretization
Enhancement of hot carrier effect and signatures of confinement in terms of thermalization power in quantum well solar cell
No full textInternational audienceA theoretical model using electron–phonon scattering rate equations is developed for assessing carrier thermalization under steady-state conditions in two-dimensional systems. The model is applied to investigate the hot carrier effect in III–V hot-carrier solar cells with a quantum well absorber. The question underlying the proposed investigation is: what is the power required to maintain two populations of electron and hole carriers in a quasi-equilibrium state at fixed temperatures and quasi-Fermi level splitting? The obtained answer is that the thermalization power density is reduced in two-dimensional systems compared to their bulk counterpart, which demonstrates a confinement-induced enhancement of the hot carrier effect in quantum wells. This power overall increases with the well thickness, and it is moreover shown that the intra-subband contribution dominates at small thicknesses while the inter-subband contribution increases with thickness and dominates in the bulk limit. Finally, the effects of the thermodynamic state of phonons and screening are clarified. In particular, the two-dimensional thermalization power density exhibits a non-monotonic dependence on the thickness of the quantum well layer, when both out-of-equilibrium longitudinal optical phonons and screening effects are taken into account. Our theoretical and numerical results provide tracks to interpret intriguing experimental observations in quantum well physics. They will also offer guidelines to increase the yield of photovoltaic effect based on the hot carrier effect using quantum well heterostructures, a result critical to the research toward high-efficiency solar cell devices
KEMASS: Knowledge-Enhanced Multi-Agent simulation for energy Scheduling Support
No full textInternational audienceThe transition to decentralized energy distribution, where any node can function as a consumer and/or producer, presents challenges in the design and testing of control algorithms, particularly in maintaining production. The existing energy scheduling model, assuming uniformity, struggles to capture the unique dynamics and constraints of individual production units. This paper introduces KEMASS, a method and system for generating a Multi-Agent System using Ontologies and Knowledge Graphs to tailor optimization algorithms for power plants. Implemented in a specific energy production valley, KEMASS closely simulates the actual system, optimizing energy schedule while considering local constraints. Although not yet a complete Digital Twin for Energy Scheduling Support, KEMASS, with its dynamic Knowledge Graphs and Ontologies, is more adaptable to evolving into one compared to othersystems. The use of Knowledge Representation technologies makes it suitable for various applications
An overview of variance-based importance measures in the linear regression context: comparative analyses and numerical tests
No full textInternational audienceOne of the most fundamental issues in many socio-environmental studies is the identification of causal effects and influential variables related to phenomena of interest. In the context of regression analysis, importance measures are effective tools for feature selection and model interpretation, allowing for the ranking of the most influential regressors. In particular, variance-based importance measures (VIMs) are a prominent topic in the field of statistics, as well as in the emerging field of global sensitivity analysis. This is due to their accessible interpretation as variance shares of the explained variable. This work focuses on the linear regression model and aims to provide an updated overview of the most well-founded methods, mainly from comparative analyses and numerical tests on various toy cases. The paper also addresses some of the practical challenges that arise, including the case of dependent inputs and high input dimensionality. The practical relevance of these tools is demonstrated through empirical studies on simulated data and public datasets. The Supplementary Material also presents the use of VIMs in a classification context, specifically via the logistic linear regression model
Radiation Accidents and Malicious Events – Scenarios and Scope of the Work of ICRP Task Group 120
No full textInternational audienceThe International Commission on Radiological Protection (ICRP) Task Group 120 (TG120) isdeveloping ICRP recommendations for radiological protection for a wide range of radiationaccidents and malicious events, complementing those given in ICRP Publication 146 (2020) forlarge nuclear accidents. The scope includes accidents involving criticalities, operating faults, andfires and explosions in nuclear facilities, inadvertent damage to sealed radiation sources, as wellas malicious events, such as sabotage of nuclear facilities or materials, use of radiologicaldispersal devices, the contamination of food and drinking water supplies, and the deploymentof nuclear weapons. A template has been designed to collate relevant information on a widerange of case studies and hypothetical malicious scenarios to ensure that the recommendationsdeveloped are broadly applicable and comprehensive. For all scenarios, a graded approach toprotection is being taken, accepting that specific guidance may be required for some distinctiveaspects, for example, protection during times of armed conflict. This paper provides an overviewof the scenarios and scope of the work of TG120, including some of the radiological and nonradiological impacts of radiation emergencies, along the response and recovery timeline