HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
    11652 research outputs found

    Editorial on the Molten Salt and Deposit-Induced Corrosion

    No full text
    International audienceCorrosion mechanisms in the presence of molten salts and under deposits have been studied for over 50 years. Still, they remain an important topic because of their relevance in various technological sectors, including energy production and storage, aeronautics, the glass industry, etc. The degradation in this form of corrosion is often quite significant. It occurs over a wide temperature and pressure range with very complex phenomena clearly influenced by the presence of (even at the ppm level) corrosive species

    Editorial on oxidation in complex atmospheres

    No full text
    International audienceThe atmospheres surrounding metal parts in many industrial processes rarely comprise a single oxidant. Whether in fossil energy conversion plants, nuclear power plants, jet engines, chemical or petrochemical production routes, several molecules, sometimes present as main components, sometimes at the impurity level, play a role in corrosion mechanisms. Current research into the development of low-emission technologies also clearly demonstrates the importance of considering, in addition to O 2 , the effects of H 2 , H 2 O, NH 3 , etc. in the development and selection of alloys This collection of papers considers many situations where oxidation competeswith, for example, carburizing, nitriding, and sulfdizing reactions and where water vapor can have a signifcant impact in cases where, at frst glance, it would not even be considered. The papers include studies on the efect of H2 and/or H2O on pure titanium, on chromia scale, but also the benefts of alumina-forming alloys for many complex environments, as well as the subtle efect of Mn and Fe on the metal dusting resistance of NiCu alloys, highlighting the relevant assistance of thermodynamicand ab initio calculation to address these complex corrosion issues. The following papers are contributions from the “Oxidation in Complex Atmospheres” session of the 11th International Conference on High-Temperature Corrosion and Protection of Materials (HTCPM 2024), held at Les Embiez Island, France, on 9–14 June 202

    Self-Learning from Pairwise Credal Labels

    No full text
    International audienceSelf-learning is a popular machine learning approach to compensate for the lack of labeled data, in instancesassociated with “pseudo-labels” produced by the trainedmodel are used. Such an approach can degrade the solutionwhen the provided pseudo-labels are of a scarce quality. Asolution is to take into account the labeling uncertainty. Weadvocate the use of rich uncertainty models, which make itpossible to better account for the lack of information or imprecision attached with pseudo labels. We address the specificcase where the model is obtained via a one-versus-rest decomposition of the set of classes

    Optimisation des problèmes de planification des processus, d'implantation d'atelier et d'ordonnancement, dans le contexte RMS

    No full text
    Reconfigurable Manufacturing Systems (RMS) have emerged as a strategic response to the dynamic and uncertain market conditions driven by the increasing demand for mass-customized products. Companies are focusing on cost-effective approaches to provide a diverse product variety efficiently. The effective production in RMS, aimed at lowering costs, is intricately connected to process planning and layout problems. These optimization problems are interdependent and might benefit from an integrated approach to increase the RMS performance, including its efficiency. This study emphasizes the importance of integrating process planning, layout, and scheduling problems by proposing and analyzing different models, including sequential, partially sequential, and integrated approaches, to optimize resource allocation and scheduling for RMS. The objective here is to minimize the system’s total costs, encompassing production, material handling, machine reconfiguration, layout reconfiguration, and tardiness. To tackle this challenge, this research work started with developing three exact solution methods and assessing their computation time for small instances. Due to the complexity of the problem, especially for larger instances, we introduced a genetic algorithm adapted to our specific problem. Additionally, we incorporated into the genetic algorithm a local search technique also adapted to our problem to enhance results. Finally, a case study was introduced to extend our investigation. While most of the work on Reconfigurable Manufacturing Systems (RMS) typically focuses on the production or assembly of products, we developed a case study to demonstrate that the flexibility of these systems is also well-suited for disassembly problems. This emphasis on disassembly directly influences the life cycle of products, highlighting that RMS can play a pivotal role in addressing environmental concerns. This underscores the idea that RMS can contribute to environmental sustainability, not only by optimizing gas and waste emissions (as shown in several works) but also by actively participating in a circular economy.Les systèmes de production reconfigurables (RMS) sont apparus comme une réponse stratégique aux marchés dynamiques et incertains, motivés par la demande croissante de produits personnalisés en masse. Les entreprises se concentrent stratégiquement sur des approches rentables pour fournir efficacement une grande variété de produits. La production dans les RMS, visant à réduire les coûts, est étroitement liée aux problèmes de planification des processus et d’implantation d’atelier. Ce travail de thèse souligne l’importance de l’intégration des problèmes de planification des processus, d’implantation d’atelier et d’ordonnancement en proposant et en analysant différents modèles, y compris des approches séquentielles, partiellement séquentielles et intégrée, afin d’optimiser l’allocation des ressources et l’ordonnancement pour les RMS. L’objectif ici est de minimiser les coûts totaux du système, englobant les coûts de production, de transport, de reconfiguration de machines, de reconfiguration d’atelier et des pénalités de retards. Pour relever ce défi, notre travail a commencé par le développement de trois méthodes de résolution exactes, en évaluant leur temps de calcul pour de petites instances. En raison de la complexité du problème, notamment pour les instances plus grandes, nous avons introduit un algorithme génétique adapté à notre problème. De plus, nous avons incorporé à l’algorithme génétique une technique de recherche locale également adaptée à notre problème pour améliorer encore les résultats. Enfin, une étude de cas a été introduite pour étendre notre recherche. Alors que la majorité des travaux sur les systèmes de production reconfigurables se concentrent généralement sur la production ou l’assemblage de produits, nous avons développé une étude de cas pour démontrer que la flexibilité de ces systèmes est également bien adaptée aux problèmes de démontage de produits pour la remanufacture. Le démontage influence directement le cycle de vie des produits, soulignant que le RMS peut jouer un rôle central dans la réponse aux préoccupations environnementales. Cela souligne également l’idée selon laquelle RMS peut contribuer à la durabilité environnementale, non seulement en optimisant les émissions de gaz et de déchets (comme le montrent plusieurs travaux), mais également en participant activement à une économie circulaire

    Scaffold électrospinné triphasique avec co-culture dynamique pour l'ingénierie tissulaire des jonctions os-tendon-muscle

    No full text
    International audienceMicrostructures in electrospun scaffolds are known to guide the differentiation of stem cells. [1-3]. In order to study the bone-tendon-muscle junctions, we propose to generate micrometric walls on a part of the aligned area of a formerly electrospun bi-phasic scaffold (honeycomb / aligned fibres). On this triphasic structure, C2C12 cells were located between the microwalls and cocultured in dynamic condition with C3H10T1/2 seeded on both the honeycombs and the aligned fibers.Les microstructures matériaux éléctrofilés sont connues pour guider la différenciation des cellules souches. [1-3]. Afin d'étudier les jonctions os-tendon-muscle, nous proposons de générer des parois micrométriques sur une partie de la zone alignée d'un matériau bi-phasique (nid d'abeille/fibres alignées) précédemment électrospinné. Sur cette structure triphasique, des cellules C2C12 ont été placées entre les micro-parois et co-cultivées en condition dynamique avec des C3H10T1/2 ensemencées à la fois sur les nids d'abeilles et les fibres alignées

    Combination of Dependent Gaussian Random Fuzzy Numbers

    No full text
    International audienceGaussian random fuzzy numbers are random fuzzy sets generalizing Gaussian random variables and possibility distributions. They define belief functions on the real line that can be conveniently combined by the product-intersection rule under the independence assumption. In this paper, we provide formulas for the combination of an arbitrary number of Gaussian random fuzzy numbers whose dependence is described by a correlation matrix

    Polymère à empreinte moléculaire pour la reconnaissance et l'immobilisation des bactéries photosynthétiques

    No full text
    Molecularly imprinted polymers (MIPs) are receptor-like biomimetic materials. They are cross-linked polymers synthesized in the presence of a molecular template, which induces three-dimensional binding sites in the polymer that are complementary to the template in size, shape and position of chemical groups. In the present work, they were used to recognize and immobilize photosynthetic bacteria on polymeric scaffolds. Due to the challenges of imprinting whole bacteria, cell surface proteins were more reasonable candidates for cell recognition. Thus, the Slr1270 protein from the cyanobacterium Synechocystis sp. PCC 6803 was chosen as the target protein for imprinting using protein structure data banks and bioinformatics tools. Two epitope sequences were selected and validated by an in silico rational approach, where after a template peptide was chemically synthesized. MIPs targeting the Slr1270 peptide epitopes were then synthesized by a solid-phase approach. N-propylacrylamide was incorporated as the main monomer into the polymerization mixture to generate thermoresponsive MIPs that feature a lower critical solution temperature around 25ºC. The MIP nanogels were then characterized using dynamic light scattering and scanning electron microscopy. The resulting MIPs showed the ability to recognize and bind their template epitope peptides as well as the whole bacterium as shown by fluorescent equilibrium binding assays, flow cytometry and fluorescence microscopy. Various approaches were developed for interfacing the MIPs with the nanocellulose matrix, based on physisorption and chemisorption. To measure the interactions between cells and the nanocellulose matrix, various techniques were employed, including QCM-D, electron and light microscopy techniques.Les polymères à empreintes moléculaires (MIPs) sont des matériaux biomimétiques de type récepteur. Ce sont des polymères réticulés synthétisés en présence d'un gabarit moléculaire, qui induit des sites de liaison tridimensionnels dans le polymère qui sont complémentaires au garbarit en termes de taille, de forme et de position des groupes chimiques. Dans le cadre de ce travail, ils ont été utilisés pour reconnaître et immobiliser des bactéries photosynthétiques sur des échafaudages polymériques. En raison des défis liés à l'impression de bactéries entières, les protéines de la surface cellulaire étaient des meilleurs cibles pour la reconnaissance cellulaire. Ainsi, la protéine Slr1270 de la cyanobactérie Synechocystis sp. PCC 6803 a été choisie comme cible pour l'empreinte en utilisant des banques de données de structures protéiques et des outils de bioinformatique. Deux séquences d'épitopes ont été sélectionnées et validées par une approche rationnelle in silico, après quoi un peptide garbarit a été synthétisé chimiquement. Les MIPs ciblant les épitopes peptidiques Slr1270 ont ensuite été synthétisés par une approche en phase solide. Le N-propylacrylamide (NPAm) a été incorporé comme monomère principal dans le mélange de polymérisation pour générer des MIPs thermosensibles présentant une température critique de solution inférieure (LCST) autour de 25ºC. Les MIP-NGs ont ensuite été caractérisés en utilisant la diffusion de la lumière dynamique (DLS) et la microscopie électronique à balayage (SEM). Les MIPs résultants ont montré la capacité de reconnaître et de lier leur peptide garbarit et la bactérie entière comme le montrent des essais de liaison en équilibre par fluorescence, la cytométrie en flux et la microscopie de fluorescence. Diverses approches ont été développées pour interfacer les MIPs avec la matrice de nanocellulose, basées sur la physisorption et la chimisorption. Pour mesurer les interactions entre les cellules et la matrice de nanocellulose, diverses techniques ont été utilisées, notamment la QCM-D et des techniques de microscopie électronique et de fluorescence

    Application of Vacuum Freezing to Improve the Quality of Thawed Carrot Discs

    No full text
    International audienc

    Limit theorems for local polynomial estimation of regression for functional dependent data

    No full text
    International audienceLocal polynomial fitting exhibits numerous compelling statistical properties, particularly within the intricate realm of multivariate analysis. However, as functional data analysis gains prominence as a dynamic and pertinent field in data science, the exigency arises for the formulation of a specialized theory tailored to local polynomial fitting. We explored the intricate task of estimating the regression function operator and its partial derivatives for stationary mixing random processes, denoted as (Yi,Xi) (Y_i, X_i) , using local higher-order polynomial fitting. Our key contributions include establishing the joint asymptotic normality of the estimates for both the regression function and its partial derivatives, specifically in the context of strongly mixing processes. Additionally, we provide explicit expressions for the bias and the variance-covariance matrix of the asymptotic distribution. Demonstrating uniform strong consistency over compact subsets, along with delineating the rates of convergence, we substantiated these results for both the regression function and its partial derivatives. Importantly, these findings rooted in reasonably broad conditions that underpinned the underlying models. To demonstrate practical applicability, we leveraged our results to compute pointwise confidence regions. Finally, we extended our ideas to the nonparametric conditional distribution, and obtained its limiting distribution.</p

    0

    full texts

    11,652

    metadata records
    Updated in last 30 days.
    HAL Portal UTC Université de Technologie de Compiègne
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇