Portail HAL edf
Not a member yet
11029 research outputs found
Sort by
Shifts and Changes of changes in Public Opinion: Lessons from Recent Years on Nuclear Energy in France and Europe
No full textInternational audienc
Adaptive Forecasting of Extreme Electricity Load
No full textElectricity load forecasting is a necessary capability for power system operators and electricity market participants. Both demand and supply characteristics evolve over time. On the demand side, unexpected or extreme events as well as longer-term changes in consumption habits affect demand patterns. On the production side, the increasing penetration of intermittent power generation significantly changes the forecasting needs. We address this challenge in three ways. First, we consider probabilistic (quantile) rather than point forecasting; indeed, uncertainty quantification is required to operate electricity systems efficiently and reliably. The probabilistic forecasts are generated using both linear and non-linear quantile regressions applied to the residuals of the mean forecasting model. Second, our approach is adaptive; we have developed models that incorporate the most recent observations to automatically respond to changes in the underlying process. Our adaptation methodology leverages the Kalman filter, which has previously been successfully employed for adaptive load forecasting, as well as Online Gradient Descent - a combination of an incremental strategy and pinball loss. Third, we extend the adaptive setting to extreme scenarios by using the aforementioned methods to compute an adaptive threshold used as a reference in recently developed machine learning models targeting extreme values. Finally, we apply our different approaches on the french daily electricity consumption as use case
Développement d'une modélisation thermomécanique du meulage pour l'estimation de l'impact sur les contraintes résiduelles de soudage
No full textNational audienceThis study addresses the numerical modeling of manual grinding, focusing on residual stresses. It proposes a macroscopic method based on thermal and thermomechanical simulations in 3D on Code_Aster to simulate grinding. The approach has been validated using experimental data collected via thermocouples placed on an inconel 600 mock-up. Although the initial model offers a promising approximation of temperatures, expanding the model to include shearing and experimental residual stress analyses could improve its accuracy.Cette étude aborde la modélisation numérique du meulage manuel, en se concentrant sur les contraintes résiduelles. Elle propose une méthode macroscopique basée sur des simulations thermiques et thermomécaniques sur Code_Aster en 3D pour simuler le meulage. L'approche a été validée à l'aide de données expérimentales recueillies via des thermocouples placés sur une maquette en inconel 600. Bien que le modèle initial offre une approximation prometteuse des températures, une expansion du modèle pour inclure le cisaillement et des analyses de contraintes résiduelles expérimentales pourrait améliorer sa précision
A decentralized algorithm for a Mean Field Control problem of Piecewise Deterministic Markov Processes
No full textInternational audienceThis paper provides a decentralized approach for the control of a population of agents to minimize an aggregate cost. Each agent evolves independently according to a Piecewise Deterministic Markov dynamics controlled via unbounded jumps intensities. The N-agent high dimensional stochastic control problem is approximated by the limiting mean field control problem. A Lagrangian approach is proposed. Although the mean field control problem is not convex, it is proved to achieve zero duality gap. A stochastic version of the Uzawa algorithm is shown to converge to the primal solution. At each dual iteration of the algorithm, each agent solves its own small dimensional sub problem by means of the Dynamic Programming Principal, while the dual multiplier is updated according to the aggregate response of the agents. Finally, this algorithm is used in a numerical simulation to coordinate the charging of a large fleet of electric vehicles in order to track a target consumption profile
Protective role of the podocyte IL-15 / STAT5 pathway in focal and segmental glomerulosclerosis
No full textInternational audienc
'Aimez-nous, on se charge du reste': De l'"agribashing" à la "communication positive", les métamorphoses de l'anti-environnementalisme agricole
No full text'Aimez-nous, on se charge du reste'De l'"agribashing" à la "communication positive", les métamorphoses de l'anti-environnementalisme agricole
Fluorescence spectroscopy for tracking microbiological contamination in urban waterbodies
No full textInternational audienceDissolved organic matter (DOM) plays a crucial role in freshwater ecosystem function. Monitoring of DOM in aquatic environments can be achieved by using fluorescence spectroscopy. Particularly, DOM fluorescence can constitute a signature of microbiological contamination with a potential for high frequency monitoring. However, limited data are available regarding urban waterbodies. This study considers fluorescence data from field campaigns conducted in the Paris metropolitan region: two watercourses (La Villette basin and the river Marne), two stormwater network outlets (SO), and a wastewater treatment plant effluent (WWTP-O). The objectives of the study were to characterize the major fluorescence components in the studied sites, to investigate the impact of local rainfall in such components and to identify a potential fluorescence signature of local microbiological contamination. The components of a PARAFAC model (C1-C7), corresponding to a couple of excitation (ex) and emission (em) wavelengths, and the fluorescence indices HIX and BIX were used for DOM characterization. In parallel, fecal indicator bacteria (FIB) were measured in selected samples. The PARAFAC protein-like components, C6 (ex/em of 280/352 nm) and C7 (ex/em of 305/340 nm), were identified as markers of microbial contamination in the studied sites. In the La Villette basin, where samplings covered a period of more than 2 years, which also included similar numbers of wet and dry weather samples, the protein-like components were significantly higher in wet weather in comparison to dry weather. A positive relationship was obtained between C6 and FIB. In urban rivers, the high frequency monitoring of C6 levels would support the fecal contamination detection in rivers. In addition, it could help targeting specific field campaigns to collect comprehensive dataset of microbiological contamination episodes
Representation learning with unconditional denoising diffusion models for dynamical systems
No full textInternational audienceWe propose denoising diffusion models for data-driven representation learning of dynamical systems. In this type of generative deep learning, a neural network is trained to denoise and reverse a diffusion process, where Gaussian noise is added to states from the attractor of a dynamical system. Iteratively applied, the neural network can then map samples from isotropic Gaussian noise to the state distribution. We showcase the potential of such neural networks in proof-of-concept experiments with the Lorenz 1963 system. Trained for state generation, the neural network can produce samples that are almost indistinguishable from those on the attractor. The model has thereby learned an internal representation of the system, applicable for different tasks other than state generation. As a first task, we fine-tune the pre-trained neural network for surrogate modelling by retraining its last layer and keeping the remaining network as a fixed feature extractor. In these low-dimensional settings, such fine-tuned models perform similarly to deep neural networks trained from scratch. As a second task, we apply the pre-trained model to generate an ensemble out of a deterministic run. Diffusing the run, and then iteratively applying the neural network, conditions the state generation, which allows us to sample from the attractor in the run's neighbouring region. To control the resulting ensemble spread and Gaussianity, we tune the diffusion time and, thus, the sampled portion of the attractor. While easier to tune, this proposed ensemble sampler can outperform tuned static covariances in ensemble optimal interpolation. Therefore, these two applications show that denoising diffusion models are a promising way towards representation learning for dynamical systems
Domaine de stabilité - Capacité portante
No full textSous la présidence de la Dr. Nathalie DUFOUR (Cerema Meditrerranée), cette session comporte les présentations : Introduction - Principes (Panagiotis KOTRONIS, Centrale Nantes) Numerical modelling of rigid inclusions foundations subjected to seismic loading (Panagiotis KOTRONIS, Centrale Nantes) Détermination du diagramme de stabilité via une approche analytique basée sur le calcul à la rupture (Yuxiang SHEN, Terrasol) Benchmark ICEDA (Matthieu JACQUET, EDF)Cette vidéo a été captée le 17 décembre 2024 à Aix-en-Provence sur le Campus Méditerranée du Cerema, lors de la journée de clôture du projet ANR ASIRIplus-SDS : Amélioration des Sols par Inclusions Rigides, Sollicitations Dynamiques et Sismiques (https://anr.fr/Projet-ANR-19-CE22-0015)L'ensemble de la journée a été filmée. Les vidéos sont disponibles sur https://www.youtube.com/playlist?list=PLwDLuT_SpnNTrmoV61iQajdWXFxnNfdFp, mais aussi sur HAL : https://hal.science/hal-05426459v1, https://hal.science/hal-05520272v1, https://hal.science/hal-05520383v1, https://hal.science/hal-05520223, https://hal.science/hal-05426490v