HAL Université de Toulouse, et Toulouse INP
Not a member yet
159308 research outputs found
Sort by
Heat transfer modelling of radiant flux from a halogen lamp for enhancing Infrared thermography simulation
No full textInternational audienceInfraRed Thermography (IRT) is widely used in Non-Destructive Evaluation (NDE) for its ability to provide real-time, two-dimensional, non-contact measurements of heat distribution. Enhancing the analysis of thermal results requires a comprehensive understanding of the entire measurement chain from the heat source, through propagation, to detection, signal processing and data interpretation, which demands an effective combination of simulation and experimental approaches. This study presents the modelling of heat transfer, with particular emphasis on accurately characterising the radiant heat flux emitted by a halogen lamp. A fluxmeter sensor was employed to measure the radiant heat flux at different distances and spatial locations. Subsequently, 3D heat transfer models incorporating these heat sources were established and applied to an Acrylonitrile Butadiene Styrene plate to investigate thermal behaviour and the influence of factors within the measurement chain. Critical parameters were also considered, including thermal conductivity, convective heat transfer coefficients, fluxmeter sensor sensitivity, heat flux characteristics and measurement methods. Simulation results were validated against experimental data using both an infrared camera and a pyrometer and demonstrated strong agreement. Relative errors were below 4.2 % for pyrometer measurements, whereas slightly higher errors, up to 5.9 % for IRT method, which is mainly attributed the influence of environmental factors on this technique. These findings confirm the accuracy and reliability of the calibrated heat source and modelling parameters. Integrating experimental data into the thermal simulation enhances both accuracy and consistency, thereby establishing a more robust framework for the application of numerical methods throughout the IRT measurement chain in NDE applications
Dyadic Ru-based nanomaterials for visible light-driven photocatalytic hydrogen evolution
No full textInternational audienceVisible light-driven water splitting is an appealing strategy to store renewable energy in the chemical bonds of molecular hydrogen. In this regard, the development of photocatalytic architectures where charge transfer and recombination can be controlled represents a key challenge. The surface functionalization of Ru/RuO 2 nanoparticles (NPs) with the [Ru(2,2′-bpy) 2 (qpy)](PF 6 ) 2 photosensitizer (PS), yielding PS-NPs "dyadic" hybrid nanomaterials, represents a promising strategy. Four HER photocatalysts with different PS:NPs ratios are synthesized and thoroughly characterized by analytical and spectroscopic techniques. X-ray photoelectron spectroscopy (XPS) reveals the covalent binding of the PS to the NPs surface. Analysis of the photocatalytic performance in aqueous triethanolamine (TEOA) shows that the activation of the nanocatalyst (RuO 2 reduction) and the hydrogen evolution rate improves when the PS loading increases. Under visible-light irradiation, the nanomaterials with higher PS loading show sustained production of hydrogen for at least 80 h. The morphological and compositional evolution of the hybrid nanomaterials under photocatalytic conditions is studied and correlated with hydrogen production rates over time, pointing to a sequential leaching of PS from the nanomaterials surface. Additionally, photophysical experiments allow attaining an insight into the photochemical mechanism, which involves oxidative quenching with a fast electron injection, but also fast back electron transfer.</div
Swarm dynamics for global optimisation on finite sets
No full textInternational audienceConsider the global optimisation of a function defined on a finite set endowed with an irreducible and reversible Markov generator.By integration, we extend to the set of probability distributions on and we penalise it with a time-dependent generalised entropy functional.Endowing with a Maas' Wasserstein-type Riemannian structure, enables us to consider an associated time-inhomogeneous gradient descent algorithm.There are several ways to interpret this \cP(V)-valued dynamical system as the time-marginal laws of a time-inhomogeneous non-linear Markov process taking values in , each of them allowing for interacting particle approximations.This procedure extends to the discrete framework the continuous state space swarm algorithm approach of Bolte, Miclo and Villeneuve \cite{Bolte}, but here we go further by considering more general generalised entropy functionals for which functional inequalities can be proven.Thus in the full generality of the above finite framework, we give conditions on the underlying time dependence ensuring the convergence of the algorithm toward laws supported by the set of global minima of .Numerical simulations illustrate that one has to be careful about the choice of the time-inhomogeneous non-linear Markov process interpretation
Optimised amine density via plasma treatment for covalent immobilisation of AMPs to promote osteoblasts and enhance antimicrobial activity
No full textInternational audienc
Harnessing the potential of medicinal plants in the malaria fighting: In vitro and in vivo antiplasmodial activities of Feretia apodanthera Delille (Rubiaceae)
No full textInternational audienc
Certified Enumeration of AI Explanations: A Focus on Monotonic Classifiers
No full textInternational audienceThe theory of minimal explanations offers a rigorous, model-based solution to the problem of producing explanations for the decisions of AI models. In some high-stakes contexts, there is a need to generate all possible explanations for a particular decision using certified programs, whose output can be trusted. We used the proof assistant Coq to certify a recently proposed algorithm for the enumeration of explanations in the case of monotonic classifiers. Our experimental results on the extracted code showcase the scalability of this approach, underscoring its potential for improving trust and reliability in AI systems
Can subfossil insects complement pedoanthracology in reconstructing the past trajectories of old-growth forests? A study case from the Northern Central Pyrenees (France)
No full textInternational audienceThe presence of insect remains preserved in soils has the potential to serve as a complementary proxy to charcoal, facilitating the reconstruction of Holocene forest trajectories at high spatial resolution. Six pits were dug at three old-growth forest sites (two per site) in the Central Pyrenees (France). Insect remains and charcoal were collected in each soil layer, following the pedoanthracological method usually conducted in similar mountain contexts. Radiocarbon dating was performed on a selection of both insects and charcoal, and a time-since-death index was developed to evaluate the degradation stage and relative age of the insect remains. Insect remains were present in most layers, but were more abundant in the upper ones, as with charcoal. Whereas radiocarbon dating did not work on individual insect remains, the time-since-death index showed a consistent relationship between increasing degradation and increasing depth. Saproxylic beetles, which are key indicators of the maturity of old-growth forests, were poorly preserved in the soils studied, but some of the other beetles identified at genus or species level provided useful information on past forest openness
Decentralized multi-agent multi-armed bandits for smart electric vehicles charging
No full textInternational audienceSmart charging of electrical vehicles can help in avoiding congestion and peak load demands in an electrical distribution network. On the consumer's side, the advantage lies in minimizing the daily charging cost. He may also benefit from cheap photovoltaic electricity from local sources and therefore reduce his environmental impact. However, this cheaper electricity is variable and uncertain. In many research works, this has been formulated and solved as a centralized or hierarchical optimization problem. However, such systems may suffer from lack of scalability, single point of failures, and privacy breaches. We propose a fully decentralized and fair multi-agent system combined with reinforcement learning called "Decentralised multi-armed bandit (2-armed bandit) based on Thompson sampling"(D-MAB2AB-TS) to control the charging of electrical vehicles under uncertainties. The problem under consideration is formulated as a two-armed bandit (charging or not) for each instant. The proposed algorithm, based on Thompson Sampling, takes into account the uncertainties in the choice of arms combination of other players. The proposed algorithm finds the best combination of arms to play with a computational complexity O(m) linear with the number of arms. The suggested system is also model-free, as it does not assume the model of the environment to be perfectly known, which is a common assumption in many of the existing centralized optimization strategies for smart charging.</div
Feed restriction has no effect on milk lipolysis, although it results in a marked reduction in milk lipoprotein lipase activity in dairy goats
No full textInternational audienceClimate change pressures livestock feeding systems by reducing the availability and nutritional quality of forages and pasture. Previous studies have shown that dietary restriction impacts milk lipolysis (i.e. the hydrolysis of milk fat by lipoprotein lipase (LPL)) in cows and ewes. However, data regarding the effects of feed restriction on milk lipolysis in goats are scarce. Therefore, the present study aims to investigate the impact of feed restriction on goat milk composition, with a specific focus on lipolysis, LPL activity and milk fat characteristics. For that purpose, two groups of 12 dairy goats (121 ± 7 days in milk) balanced with regard to the αs1-casein (CSN1S1) genotype (7 A/A goats, corresponding to strong CSN1S1 genotypes and 5 O/O goats, corresponding to null CSN1S1 genotypes) received either a control diet (100 % of the dry matter intake (DMI) ad libitum: non-restricted; NR) or the experimental diet (65 % of the DMI ad libitum: restricted; R) according to a 2 × 2 cross-over design. Dietary restriction did not significantly affect either lipolysis levels in milk or milk fat globule size, whereas a sharp decrease was observed in milk LPL activity within the short-term restriction window tested. A genotype effect was evidenced for milk lipolysis (copper soap method), milk fat content and milk fat yield. No genotype × diet effect was observed with regard to lipolysis or LPL activity under feed restriction. This work provides a solid framework for follow-up lipidomic and proteomic studies to gain a comprehensive understanding of the regulation of the lipolytic system in goats
Simultaneous estimation of radiance and its sensitivities to radiative properties in a spherical-heterogeneous atmospheric radiative transfer model by Monte Carlo method: Application to Titan
No full textInternational audienceWe propose a control variates technique to reduce the variance of null-collision Monte Carlo algorithms used for solving the Radiative Transfer Equation (RTE) in highly heterogeneous media. The method complements the classical spatially partitioned overestimate approach by additionally recording the minimum absorption coefficient within each voxel during preprocessing. During path tracing, the attenuation due to this minimum absorption is evaluated analytically, while the residual part is handled by path-samplings. This analytical treatment significantly improves convergence particularly in strongly absorbing media such as the planetary atmospheres in infrared absorbing band. The mathematical equivalence between the original and control-variates estimators is demonstrated, and numerical applications for Earth's and Titan's atmospheres confirm the expected variance reduction.</div