Portail HAL IMT Mines Albi
Not a member yet
5440 research outputs found
Sort by
Multi-Physics Simulation of Void Evolution in Thermoset Prepreg UD Laminates
No full textInternational audienc
Usinabilité d'un composite à matrice céramique oxyde-oxyde par jet d'eau abrasif
No full textInternational audienceMachining ceramic matrix composites (CMC) represents a challenge in manufacturing structural aerospace components. Abrasive water jet (AWJ) machining is a relatively underexplored but up-and-coming solution, as it induces less defects compared to the conventional machining. This work aims to demonstrate the feasibility of this machining on an oxide-oxide CMC (OCMC) and develop an empirical model to control and predict the machined depth. First, a study of some AWJ machine parameters was conducted, which allowed for the selection of the parameters to focus on. A design of experiments was constructed to investigate the influence of water jet pressure, feed rate, and abrasive flow rate on the machined depth. Various characterization steps were implemented to achieve this, including confocal and scanning electron microscopy, followed by image processing. Following the design of experiments, the ability to machine an OCMC by AWJ has been demonstrated and a model has been developed to predict the machined depth with an error of less than 10 %.L’usinage des composites à matrice céramique (CMC) représente un challenge pour la fabrication de pièces structurelles aéronautiques. Le jet d’eau abrasif (JEA) est une solution d’usinage peu étudiée mais très prometteuse, puisqu’elle limite fortement la dégradation structurale du matériau usiné. Ces travaux ont pour but de démontrer la faisabilité de ce type d’usinage sur un CMC oxyde-oxyde puis de développer un modèle empirique dans le but de contrôler et prédire la profondeur usinée. Tout d’abord, une étude de certains paramètres de la machine de JEA a été réalisée, permettant de sélectionner les paramètres sur lesquels l’étude se focalisera. Un plan d’expérience a été construit pour étudier l’influence de la pression du jet d’eau, la vitesse d’avance et le débit d’abrasif sur la profondeur usinée. Différentes étapes de caractérisations sont mises en œuvre notamment par microscopie confocale et électronique à balayage, suivies par du traitement d’image. Suite au plan d’expérience, la capacité d’usiner un CMC oxyde-oxyde par JEA a été démontrée et un modèle permettant de prédire la profondeur usinée avec une erreur inférieure à 10 % a été développé
Sovereignty and usages of Generative Artificial Intelligence in Academic Work
No full textInternational audienceGenerative Artificial Intelligence (AI), particularly Large Language Models, are increasingly adopted within academic settings to accelerate research and enhance or evolve pedagogical practices. While initially serving as an assistant to lecturersresearchers,questions arise regarding their legitimacy and potential biases. This article aims to uncover the reality of AI integration through a simplified questionnaire. Based on our findings, we propose a locally implemented and sovereign approach to generative AI solutions to address concerns such as data research sovereignty
Reimagining Supply Chain Learning with Realistic Data-Driven Gaming Simulator
No full textInternational audienceAs supply chains grow increasingly complex, educational tools must evolve to better prepare students and practitioners for managing global networks. Traditional models like the Beer Game illustrate fundamental supply chain dynamics but fail to capture modern challenges such as multi-tier fulfillment, demand variabilities, and disruptions. Our interactive supply chain simulator addresses this gap by integrating a configurable multi-facility network, autonomous decision-making, and demand-driven replenishment, implemented with progressive complexity. The simulator integrates descriptive, predictive, and prescriptive analytics, enabling users to experiment with strategies and analyze real-time impacts on key performance metrics. Statistical testing from usage by master's students reveals a significant correlation between simulator engagement and improved team performance, reinforcing its value as a learning aid. User feedback further indicates that the simulator enhances analytical thinking and decision-making skills, equipping students to confidently navigate real-world supply chain complexities
Investigation of cooling efficiency and structural characteristics of supercritical CO2 jets for machining assistance
No full textInternational audienceIn a context of safer, more cost-effective, and environmentally friendly machining processes, supercritical carbon dioxide (scCO2) emerges as a promising alternative to conventional cutting fluids. However, a consolidated understanding of the optimal scCO2 jet parameters for machining and its impact on the tool and the material remains unavailable. This work intends to address the behavior of a free and impinged scCO2 jet structure and its geometrical evolution in function of the upstream jet conditions. The supersonic structure and parameters were analyzed under varying initial pressures and temperatures in both near and far fields. The obtained images showed the detailed jet structure and highlighted the presence of three different zones within the flow: laminar, transitional then turbulent. To investigate this behavior and to better understand the jet’s cooling ability in these different zones, an in-depth analysis of the jet’s complex structure was carried out using high-speed imaging combined with an optical imaging known as Schlieren
Hydroxyapatite‐Supported Ruthenium Catalysts in Ammonia Synthesis: Impact of Ba and Cs as Catalyst Promoters
No full textInternational audienceThis work is devoted to the synthesis, the characterization, and the evaluation of hydroxyapatite‐supported ruthenium catalysts, with or without Ba and/or Cs promotion. Thus, a series of catalysts containing Ru, Cs, and Ba was synthesized by the incipient wetness impregnation method. Such catalysts are characterized by different physicochemical methods, providing insights into their properties. These catalysts are evaluated in the ammonia synthesis reaction at 350–500 °C and 10–25 bar. Sample 1Ru/hydroxiapatite (HAP), without promoter, shows a negligible catalytic activity, due to the formation of large Ru nanoparticles, which are not favorable for the formation of ammonia. On the other hand, the addition of Cs and Ba improves the catalytic performance, and Ba is found to be better than Cs. The pretreatment of the barium‐containing catalysts under Ar flow at 600 °C is also found to be crucial for the decomposition of barium nitrate into barium oxide, thereby enhancing catalytic activity
Utility-Driven Demand Estimation Framework for Regional Transport Systems
No full textInternational audienceRegional transport systems face unique challenges, such as low population density and limited infrastructure, complicating accurate demand estimation, yet essential to any approach to designing or managing transport activities. This paper introduces a demand estimation framework tailored to regional contexts, integrating both passenger and freight needs into a unified, utility-based model that dynamically adjusts to user expectations and transport mode characteristics over time. The model incorporates an original utility function that weights key transport features based on specific user profiles. An illustrative case allows validating the framework’s applicability, illustrating its ability to map demand flows, supporting it as an effective decision-support tool for regional transport operators
Engineering core–shell-structured BaAl2O4 overlaid Ni catalyst with strong metal-support interaction for durable and efficient CH4 dry reforming
No full textInternational audienceDry reforming of methane (DRM) over Ni-based catalysts is an economically reasonable technology forlarge-scale CO 2 utilization. However, prolonged Ni sintering and carbon deposition reduce the durabilityand efficiency of DRM, hindering its engineering application. Herein, we propose a facile approach bycombining continuous microscale coprecipitation with solid-state reactions to construct a BaAl 2O 4-overlayer-confined Ni catalyst. The 5- wt%-Ni@BaAl 2O4 catalyst exhibited advanced CO 2 and CH4 conver-sions of 96% and 86% at 800 °C and a GHSV of 144 L gcat−1 h−1. Moreover, the kd-CO 2 and kd-CH 4 ofNi@BaAl2O 4 were 0.0063 and 0.0029 h−1; which are approximately half and one-thirds of those of Ni/BaAl2O 4 and slightly better than those of Ni@MgAl 2O4 , underscoring the versatility of the proposed syn-thesis protocol for constructing core–shell structures. XAS, HAADF–STEM–EDS, and CO transmission-IRcharacterizations confirmed the SMSI of ∼2-nm amorphous BaAl 2O 4-overlaid ∼10 nm Ni with an overallmesoporous structure. After a long-term test, the sintering and coking inhibition effects of Ni@BaAl 2O4(10 → 11 nm, 0.55 mg C gcat−1 h−1) outperformed Ni/BaAl 2O4 (13 → 22 nm, 1.90 mg C g cat−1 h−1 ) andNi@MgAl2O 4. In situ time-resolved CH 4 → CO 2 transient response, DRIFTS experiments, and DFT calcula-tions suggested that Ni@BaAl 2O 4 and Ni/BaAl 2O 4 followed the Mars–van Krevelen and Langmuir–Hinshelwood redox mechanisms, respectively. The functional interfacial lattice oxygen promoted theremoval of Cads* on Ni and core–shell structure induced fast CO 2 adsorption and CO desorption. The pre-sent study provides a facile approach for constructing a stable and active Ni-based core − shell catalyst.Furthermore, it offers novel insights into the functionalities of non-reducible spinel overlayers in theDRM process
Impression 3D de gels pour la médecine personnalisée : imprimabilité et propriétés pharmaceutiques
No full textNational audienc
Numerical modelling of thermoplastic CO2 assisted extrusion foaming: Exploring the process-microstructure relationship
No full textInternational audienceThis study presents a comprehensive numerical modeling approach for thermoplastic foams processing, produced via supercritical CO₂-assisted extrusion foaming. The model focuses on the nucleation and growth of bubbles within the polymer matrix, employing the influence volume approach to simulate these dynamics. Key phenomena such as gas diffusion, nucleation rate, bubble growth, and the evolution of influence and non-influence volumes are captured to predict foam microstructure and cellular density. Results highlight the model’s capability to predict equilibrium bubble sizes ranging from micrometers to hundreds of micrometers, offering insights into foam behavior and microstructure control. The final size distribution will be used to create on Paraview® the microstructure that will be compared to experimental observations in the future