Portail HAL edf
Not a member yet
11029 research outputs found
Sort by
Towards diffusion models for large-scale sea-ice modelling
No full textWe make the first steps towards diffusion models for unconditional generation of multivariate and Arctic-wide sea-ice states. While targeting to reduce the computational costs by diffusion in latent space, latent diffusion models also offer the possibility to integrate physical knowledge into the generation process. We tailor latent diffusion models to sea-ice physics with a censored Gaussian distribution in data space to generate data that follows the physical bounds of the modelled variables. Our latent diffusion models reach similar scores as the diffusion model trained in data space, but they smooth the generated fields as caused by the latent mapping. While enforcing physical bounds cannot reduce the smoothing, it improves the representation of the marginal ice zone. Therefore, for large-scale Earth system modelling, latent diffusion models can have many advantages compared to diffusion in data space if the significant barrier of smoothing can be resolved
Assessment of indium-free transparent conductive oxide back contacts for high efficiency ultra-thin Cu(In,Ga)Se2 solar cells down to 250 nm
No full textInternational audienceThis work examines the feasibility and performance impact of replacing the usual molybdenum backcontact with indium-free transparent conductive oxides (TCOs) like fluorine-doped tin oxide (SnO2:F) and aluminum-doped zinc oxide (ZnO:Al) for ultra-thin Cu(In,Ga)Se2 (CIGS) solar cells (250-450 nm).Motivated by indium scarcity and cost reduction, these TCOs are evaluated for their figure of merit, stability under Se atmosphere, Na diffusion permeability, and band alignment with CIGS absorbers.Using simulations, prototype fabrication, and comprehensive characterizations, the compatibility of these TCOs with CIGS absorbers is assessed. Solar cells with thicknesses of 450 nm and 250 nm are fabricated. Their performance was compared under both rear and front illumination, as well as with the use of reflectors. A record efficiency of 8.6% with front illumination is achieved for a 250 nm CIGS absorber using a gold back reflector with SnO2:F, single-step CIGS deposition, and no heavy alkalines doping. The best rear-illuminated efficiencies are obtained with ZnO:Al back contacts, reaching 6% for a 250 nm CIGS, with only a 9% loss in Jsc compared to front illumination confirming a lower surface recombination rate at the ZnO:Al/CIGS interface compared to Mo/CIGS or SnO2:F/CIGS interfaces.</div
Generative AI models capture realistic sea-ice evolution from days to decades
No full textSea ice plays an important role in stabilising the Earth system. Yet, representing its dynamics remains a major challenge for models, as the underlying processes are scale-invariant and highly anisotropic. This poses a dilemma: physics-based models that faithfully reproduce the observed dynamics are computationally costly, while efficient AI models sacrifice realism. Here, to resolve this dilemma, we introduce GenSIM, the first generative AI model to predict the evolution of the full Arctic sea-ice state at 12-hour increments. Trained for sub-daily forecasting on 20 years of sea-ice-ocean simulation data, GenSIM makes realistic predictions for 30 years, while reproducing the dynamical properties of sea ice with its leads and ridges and capturing long-term trends in the sea-ice volume. Notably, although solely driven by atmospheric reanalysis, GenSIM implicitly learns hidden signatures of multi-year ice-ocean interaction. Therefore, generative AI can extrapolate from sub-daily forecasts to decadal simulations, while retaining physical consistency
Non-destructive measurement of fuel rods inner pressure and composition: Numerical consideration of different spring dimensions
No full textInternational audienceThe technique presented in this paper enables non-destructive measurement of both the internal pressure and gas composition within fuel rods used in pressurised water reactors. The sensor consists of a piezoelectric element shaped as a tile, which can be applied directly on the fuel rod cladding in the plenum region, where fission gases accumulate. Acoustic waves generated by the sensor propagate inside the cladding and the reflections occurring within the gas are detected. The time-of-flight between measured echoes is used to retrieve the composition of the internal gas mixture. Pressure is related to the amplitude of the acoustic signal and is determined through a calibration process that links theoretical amplitude to experimental measurements. The developed theoretical model significantly reduces the calibration time by taking into account the losses and signal attenuation caused by the presence of a spring located within the measurement zone. This paper presents the construction of a theoretical model that considers the effects of the pressure and gas composition, as well as the shadowing and diffraction phenomena induced by the spring, all of which influence the amplitude of the measured acoustic signal
Bees as an example for operationalizing the ecosystem services approach into environmental radiological protection: updates from TG 125
No full textInternational audienceThe mandate of the International Commission on Radiological Protection (ICRP) Task Group (TG) 125 is to explore if and how the ecosystem services approach can supplement the current framework of environmental radiological protection (ERP). Ecosystem services, the benefits people derive from nature, provides a structure by which to explicitly characterize and protect the coupled human-nature relationships which are central to human well-being, the integrity of ecosystems, and sustainable development. This presentation provides updates from the TG's most recent discussions, which have focused on the importance of (1) considering if and to what extent the ecosystem services approach provides "added value" and (2) acknowledgement that ecosystem services cannot provide a "magic number" for protection. The TG currently considers that the ES approach can promote support transparency in decision-making and is likely to often be useful for existing exposure scenarios and to facilitate the protection principle of optimization. However, the TG also acknowledges that ES may not be the right tool in every situation. Feedback received thus far suggests that operationalizing the ecosystem services approach is a notable challenge for implementation. Therefore, this presentation will also examine bees as a representative species for demonstrating the applicability of the ecosystem services approach across geographies, cultures, and exposure scenarios as a complement to the ICRP's current conservation-based (i.e., ecocentric) approach to ERP, given they are one of the ICRP's 12 Reference Animals and Plants (RAPs). Bees are a widely-distributed insect family which provide ecosystem services such as pollination, production of honey and beeswax, and desirable aesthetic inspiration. They directly contribute to maintenance of biodiversity in wild habitats and human well-being through provision of these various ecosystem services. We describe how including the impacts of radiological contamination and associated decision-making on bees and their ecosystem services has the potential to improve radiological protection through adoption of a One Health perspective, enhanced stakeholder engagement, and improved decision-making. We provide specific case studies demonstrating how to conceptually apply the ecosystem services approach for different exposure scenarios (e.g., existing and planned) using bees as a model species
The importance of strongly coupling human activity, appliance use and electricity tariffs for load curve simulation in the residential sector
No full textInternational audienceActivity-based approaches for modeling residential energy consumption are actively being developed. However, these approaches rarely consider the impact of electricity tariffs on occupant behavior and energy use. We present an agent-based model of activity and energy consumption in which electricity tariffs influence behaviors, appliance use, and appliance control. Our results show that the model adequately reproduces load curve dynamics when compared to real consumption data from households with electric domestic hot water. We then illustrate the model’s capabilities through a prospective case study examining changes in the temporal placement of off-peak hours
A spatio-temporal weather generator for the temperature over France
No full textInternational audienceStochastic weather generators are efficient statistical models producing synthetic weather series by replicating key statistical properties without the computational cost of physical models. However, for applications requiring temperature simulation over a large area, challenges arise due to non-stationarity over time and spatial-temporal dependencies. This paper introduces a daily stochastic weather generator for temperature with arbitrary spatial resolution. The non-stationarity issue is addressed using a decomposition method to separate deterministic terms (trends and seasonality), from the stochastic part representing the underlying climate variability. We extend the existing local decomposition method to extrapolate to any point in space.The spatial-temporal dependence is modeled as a Gaussian field with a non-separable covariance function, accommodating complex interactions between time and space. Our generator, calibrated on a few French weather stations, is validated using several spatio-temporal indicators. First, we evaluate the generator's performance at the fitting stations, comparing simulated and observed indicators. Subsequently, we compare our spatial simulations to a high resolution gridded observation datasets.Results demonstrate that the proposed generator accurately captures the observed spatio-temporal statistics, even for extremes such as large scale persistent heat waves
Exploring the shear behavior of masonry triplets via digital image correlation, damage quantification and Mohr-Coulomb criterion identification
No full textInternational audienceShear properties are essential for evaluating the strength of masonry infill walls. Traditionally, shear strength parameters are experimentally determined through triplet shear tests. While extensive research exists on these tests, few studies have combined full-field measurement techniques such as Digital Image Correlation (DIC) with mesoscopic damage analysis and shear strength quantification. Leveraging the detailed displacement fields provided by DIC and force measurements, the study seeks to enhance the characterization of the Mohr-Coulomb criterion in mortar joints. An experimental campaign on eight specimens yields valuable insights through force-displacement curves and DIC post-processing in a monoview configuration. This analysis proposes a novel method to achieve a more precise estimation of triplet shear strength
Futurs débits de la partie française de la rivière Meuse – un examen plus approfondi des incertitudes
No full textInternational audienceIt is essential to assess the impact of future climate change on catchment hydrology in a rigorous manner, accounting for uncertainties. In this study, we assess the impact of climate change on the natural streamflow of the French part of the Meuse catchment. Climate projections from two Representative Concentration Pathways (RCPs) and five General Circulation Model/Regional Climate Model (GCM/RCM) couples were retrieved to feed four hydrological models run with several parameter sets to assess future streamflow. A variance analysis tool was employed to partition the sources of uncertainty. Although an increase in air temperature is expected in the future, particularly with the RCP 8.5, the projected trend in precipitation remains uncertain. A slight increase in annual precipitation, an increase in winter precipitation and an uncertain signal for summer precipitation are anticipated. These developments will result in an increase in winter streamflow and an uncertain signal for summer streamflow, although these evolutions may not be homogeneous along the Meuse course. Regarding uncertainties, internal variability represents the greatest uncertainty in the near future, while climate models account for the highest uncertainty in the mid and far futures. The results were found to be comparable to those of the Francewide Explore2 project
Hyperelastic nature of Hoek-Brown criterion
No full textWe propose a nonlinear elasto-plastic model, for which a specific class of hyperbolic elasticity arises as a straight consequence of the yield criterion invariance on the plasticity level. We superimpose this nonlinear elastic (or hyperelastic) behavior with plasticity obeying the associated flow rule. Interestingly, we find that a linear yield criterion on the thermodynamical force associated with plasticity results in a quadratic yield criterion in the stress space. This suggests a specific hyperelastic connection between Mohr–Coulomb and Hoek–Brown (or alternatively between Drucker–Prager and Pan–Hudson) yield criteria. We compare the elasto-plastic responses of standard tests for the Drucker–Prager yield criterion using either linear or the suggested hyperbolic elasticity. Notably, the nonlinear case stands out due to dilatancy saturation observed during cyclic loading in the triaxial compression test. We conclude this study with structural finite element simulations that clearly demonstrate the numerical applicability of the proposed model