HAL Arts et Métiers
Not a member yet
14127 research outputs found
Sort by
Accelerating cell topology optimisation by leveraging similarity in the parametric input space
No full textInternational audienceThe design of high-resolution topology-optimised (TO) structures is important for many industrialand medical applications because of their better mechanical performance underdifferent load conditions. Traditional density-based TO methods, like the Solid Isotropic Materialwith Penalisation (SIMP) method, can produce detailed designs but are very computationallyexpensive, especially for fine meshes. While surrogate models using neural networks can speedup the process, they often lack generality and can create discontinuities, making them lesseffective for solving new problems.This study addresses these issues by introducing a method to speed up cell-level TO withina 2-Level framework, where large structures are built by combining optimised square cells.A data-driven instance-based model provides a better starting point for the standard SIMPbasedoptimiser, placing it closer to a local minimum and reducing computation time. To avoidthe generality problems of other methods, the instance-based model uses a dataset expandedthrough two strategies: context-based data creation, which generates specific samples for theproblem, and data augmentation, which increases dataset size without extra computation.Two similarity metrics, vector-based and energy-based, are used to measure how close theinput parameters are. Both metrics are effective, but the energy-based metric is expected towork better in 3D cases, where higher-dimensional input spaces make other approaches less reliable.This methodology addresses important challenges associated with existing instance-basedmodels, enhancing the speed and applicability of high-resolution TO
Implicit learning of professional skills through immersive virtual reality: a media comparison study
No full textInternational audienceThis study investigates the effectiveness of Immersive Virtual Reality (IVR) compared to traditional slideshow lessons in teaching implicit knowledge. For this purpose, the research focuses on professional decision-making skills in viticulture. Most existing research on immersive learning concentrates on explicit learning strategies. In contrast, this study explores the potential of IVR to foster the transfer of implicit knowledge to real-world situations.Forty third-year engineering students were randomly assigned to an IVR or a traditional slideshow group. They learned to assess vine vigour through an implicit learning phase, followed by a real-world evaluation in an actual vineyard. Learning outcomes were measured by decision-making accuracy, response time, and intrinsic motivation.The findings show that the IVR group did not significantly outperform the slideshow group in decision-making accuracy. However, the IVR group took more time to make decisions. This observation suggests an impact of immersion during the transfer to real-world situations. Additionally, the IVR group showed a higher level of intrinsic motivation than the slideshow group.These results suggest that although the immersion effect does not directly enhance learning outcomes for this cognitive objective, it does affect how knowledge is transferred to the real world. They also confirm that the positive impact of immersion is difficult to generalize and may depend on the nature of the knowledge. Still, the immersion effect significantly improves learner motivation. This consistent finding could be a key factor in long-term educational success. Further research exploring the nuanced effects of immersion on different learning strategies and educational objectives could offer new practical perspectives for the future of educational technologies
Implementation of the twisting controller based on the deadbeat theorem
No full textThis work investigates the application of the deadbeat theorem to the implementation of the twisting controller. It will be shown that the twisting controller under the deadbeat implementation presents several advantages over the recently trending implementation method, i.e., implicit discretization, such as a straightforward and systematic procedure without requiring to handle backward discretization, generalized equations, and numerical solvers. Several properties of the twisting controller implemented based on the proposed deadbeat implementation including the chattering treatment, convergence analysis, robustness to parametric uncertainties, and matched disturbances are studied theoretically and validated based on numerical simulations leading to promising results
Bioinspired 4D Printed Tubular/Helicoidal Shape Changing Metacomposites for Programmable Structural Morphing
No full textInternational audienceBiological structures combine passive shape‐changing with force generation through intricate composite architectures. Natural fibers, with their tubular‐like structures and responsive components, have inspired the design of pneumatic tubular soft composite actuators. However, no development of passive structural actuation is available despite the recent rise of 4D printing. In this study, a biomimicry approach is proposed with inspiration from natural fiber architecture to create a novel concept of thermally active 4D printed tubular metacomposites. These metacomposites exhibit high mechanical performance and 3D‐to‐3D shape‐changing ability triggered by changes in temperature. A rotative printer is proposed for winding a continuous carbon fibers reinforced PolyAmide 6.I composite on a PolyAmide 6.6 polymer mandrel in a similar manner to the structure of cellulose microfibrils within the polysaccharide matrix of natural fiber cell‐walls. The resulting 4D printed tubular metacomposites exhibit programmable rotation and torque in response to thermal variations thanks to the control of their mesostructure and the overall geometry. Energy density values representing a trade‐off between the rotation and the torque are comparable to shape memory alloys when normalized by stiffness. Finally, a proof of concept for an autonomous solar tracker is presented, showcasing its potential for designing autonomous assemblies for structure morphing
“Mission ODD17” : une réflexion interdisciplinaire face aux défis environnementaux et sociétaux à travers la découverte des ODD.
No full textInternational audienceEnvironmental and societal issues are complex and require a thorough understanding and interdisciplinary approach. In recent years, the latest ministerial directives have encouraged our teachers (experts in their field) to integrate sustainable development and social responsibility objectives into their teaching. However, a strictly disciplinary approach to these issues is often insufficient. To obtain a more global and effective vision of these challenges, it is essential to develop and promote interdisciplinarity, in particular by bringing teachers together to establish the possible connections and interactions between their disciplines. With this in mind, the Racine ParisTech network has designed a workshop to (re)discover the objectives and challenges of sustainable development, with the aim of giving participants concrete ways of changing their practice and incorporating (more) socio-environmental issues. This funworkshop is broken down into several tasks that gradually lead participants to think more deeply about the Sustainable Development Goals (SDGs) and how to integrate them into their professional practice. This is the workshop we are presenting here.Les enjeux environnementaux et sociétaux sont complexes et nécessitent une compréhension approfondie ainsi qu'une approche interdisciplinaire. Depuis quelques années, les dernières directives ministérielles encouragent nos enseignants (experts disciplinaires) à intégrer des objectifs de développement durable et responsabilité sociétale dans leurs enseignements. Cependant, une approche strictement disciplinaire est souvent insuffisante face à ces problématiques. Pour obtenir une vision plus globale et efficace de ces défis, il est essentiel de développer et promouvoir l'interdisciplinarité, notamment en réunissant les enseignants pour établir les connexions et interactions possibles entre leurs disciplines. Dans cette dynamique, le réseau Racine ParisTech a conçu un atelier sur la (re)-découverte des objectifs de développement durable et leurs enjeux, visant à donner des pistes concrètes aux participants pour modifier leur pratique et y intégrer (davantage) les enjeux socio-environnementaux. Cet atelier ludique se décompose en plusieurs missions qui mènent progressivement les participants à approfondir leur réflexion sur les Objectifs de Développement Durable (ODD) ainsi que leur intégration dans leur pratique professionnelle. C'est l'atelier que nous vous présentons dans le cadre de cette communication
La fresque de l’entrepreneuriat innovant et collectif : un outil de sensibilisation aux enjeux et aux risques entrepreneuriaux
No full textInternational audienc
Leveraging Augmented Reality for Enhanced Smart and Connected Product Design: An Experimental Approach
No full textInternational audienceThe Internet of Things (IoT) is transforming all sectors by enabling traditional products to become smart and connected, capable of collecting, analyzing, and exchanging data. In parallel, Augmented Reality (AR) has emerged as a promising technology to support product design by offering immersive and interactive ways to visualize and iterate ideas. This paper presents an AR-based application developed to assist designers during the early ideation phase of IoT product conception. The system enables visualization of virtual objects and interaction with 14 sensor capability cards and 12 user experience elements. An experimental study involving 14 undergraduate engineering students was conducted to compare the proposed AR tool with a traditional 2D paper-based method, using two design cases: a connected bicycle and a smart window. Quantitative results from NASA-TLX and SUS questionnaires indicate that the AR method maintained or reduced perceived workload, particularly in terms of complexity and time pressure, while achieving usability scores comparable to or better than the traditional approach. These findings demonstrate the potential of AR as an effective and cognitively sustainable tool for enhancing creativity in early-stage product design
Étude et compréhension des interactions gazeuses moule métal en fonderie d'acier
No full textThe composition of the mold atmosphere plays a crucial role in mold-metal interactions during steel casting and is a key contributor to several casting defects, including metal penetration, subsurface porosity, and blowhole formation – issues that often require extensive repair or lead to part rejection. While the basic mechanisms of atmosphere formation in sand molds are described in the literature, current understanding – particularly in the context of steel casting – remains limited. Experimental studies involving direct measurement of the chemical composition of mold atmospheres are scarce and often date back to the previous century. This thesis aims to address this gap by experimentally investigating the influence of process parameters on atmosphere formation in sand molds poured with steel, and evaluating its impact on casting soundness. One major challenge is the difficulty of direct gas measurement due to the harsh conditions and technical constraints of the casting process. To address this, a custom gas analysis system was developed during the project, enabling real-time, in-situ monitoring of the mold atmosphere. The design and validation of this system are provided in the thesis. The experimental campaign included online gas measurements and gas sampling using Tedlar bags. The results revealed that pouring temperature and minor variations in melt composition had negligible effects on atmosphere formation. Sand humidity and reclamation state showed limited influence, while sand additives had the most significant impact on atmospheric composition. Metallographic analysis of castings provided insights into the relationship between atmosphere and defect formation. Additionally, statistical analysis using a custom Python script with a machine learning classification algorithm linked the presence of inclusions to the oxidizing potential of the atmosphere. Elevated hydrogen concentrations at the start of pouring were found to promote porosity defects.La composition de l’atmosphère du moule joue un rôle crucial dans les interactions moule-métal lors de la coulée des aciers, et contribue de manière significative à plusieurs défauts de coulée, tels que l’abreuvage, la porosité gazeuse et la formation de soufflures – des défauts qui nécessitent souvent des réparations importantes, voire entraînent le rebut de la pièce. Bien que les mécanismes fondamentaux de formation de l’atmosphère dans les moules en sable soient décrits dans la littérature, la compréhension actuelle – en particulier dans le contexte de la coulée des aciers – reste limitée. Les études expérimentales impliquant des mesures directes de la composition chimique de l’atmosphère dans les moules sont rares et, dans de nombreux cas, datent du siècle dernier. Cette thèse vise à apporter une meilleure compréhension sur les phénomènes en étudiant expérimentalement l’influence des paramètres du procédé sur la formation de l’atmosphère dans les moules en sable utilisés en fonderie des aciers, et en évaluant son impact sur la qualité des pièces. L’un des principaux défis réside dans la difficulté de mesurer directement la composition des gaz, en raison des conditions extrêmes et des contraintes techniques du procédé de la fonderie. Pour y répondre, un système d’analyse des gaz sur-mesure a été développé au cours du projet, permettant un suivi en temps réel et in-situ de l’atmosphère du moule. La conception et la validation de ce système sont présentées dans le manuscrit. La campagne expérimentale a consisté à mesurer le gaz en continu lors de la coulée et refroidissement de la pièce. En parallèle le gaz a été prélevé dans des sacs Tedlar pour être analysé à posteriori. Les résultats ont montré que la température de coulée et les variations mineures de la composition du métal n’avaient que peu d’effet sur la formation de l’atmosphère. L’humidité du sable et son état de régénération ont montré une influence limitée, tandis que les additifs du sable ont eu l’impact le plus significatif sur la composition de l’atmosphère formée dans le moule. Des observations métallographiques des pièces coulées ont permis de mieux comprendre la relation entre l’atmosphère et la formation des défauts. De plus, une analyse statistique, réalisée à l’aide d’un script Python personnalisé intégrant un algorithme de « machine learning », a permis de relier la présence d’inclusions au potentiel oxydant de l’atmosphère. Une concentration élevée d’hydrogène au début de la coulée a été identifiée comme favorisant l’apparition de défauts de porosité
Low tip-speed ratio design of horizontal-axis wind turbines
No full textInternational audienceThis chapter aims at exploring the potential of horizontal-axis wind turbines at small scales, paving the way for promising energy harvesting applications. In this paper, we focus on the conversion of wind kinetic energy into mechanical energy at the rotor shaft. A bibliographic review of the performance, geometry, and sizing methods for small-scale HAWTs is presented. We highlight the limitations of the classical Blade Element / Momentum Theory (BEM) for designing such turbines. We explore another approach of design, based on ”blade cascade”. Preliminary parametric experimental studies examining the effects of Reynolds number, solidity, blade count, and velocity distributions yield encouraging results for optimizing the performance of small-wind turbines, offering high energy efficiency and low starting speeds to address the challenges of small scale wind energy
High-throughput determination of crystallite size and microstrain from X-ray diffraction data with deep neural networks
No full textThe development of high-throughput X-ray diffraction (XRD) techniques presents significant challenges to traditional data analysis frameworks. The present work demonstrates that deep convolutional neural networks (CNNs) can reliably quantify key microstructural parameters of polycrystalline materials, specifically crystallite size and heterogeneous strain (microstrain) extracted from powder XRD patterns, within milliseconds. A CNN was trained using synthetic data generated with a dedicated code based on the GSAS-II software. We employ the example of monoclinic ZrO2 to critically examine the theoretical performances of the CNN. The accuracy of the predictions was systematically investigated within a crystallite size range of 5 – 1000 nm and a microstrain range of 0.05-2%. In this context, the role of the resolution of the diffractometer and the XRD peak profile shape is discussed in details. For instance, on a low-resolution laboratory diffractometer it is possible to determine the crystallite size within a 5–100 nm range with 99.7% accuracy in the absence of microstrain, and 97.3% accuracy when microstrains are present. Higher accuracies are obtained with a high-resolution diffractometer and over an extended range of size and microstrain. The potential of the CNN is demonstrated by the analysis of the crystallite size and microstrains in MgAl2O4 formed by a solid-state reaction between MgO and Al₂O₃ using in situ XRD at 1200 °C performed at the ESRF. The CNN can determine the crystallite size and microstrain in 0.004 s per diagram with values that deviate by ~7% and ~8% from a conventional Rietveld refinement, respectively, which is contained within the statistical uncertainties of these parameters. We further demonstrate that the same training dataset can be used for other regression problems, for instance phase fraction quantification, with no additional computational cost and only minor modifications to the network architecture. This work paves the way for real-time data analysis at synchrotron facilities