HAL Arts et Métiers
Not a member yet
14127 research outputs found
Sort by
Thermodynamic modeling and experimental investigation of Ti PVD coatings as protective barriers against carbon diffusion during SPS
No full textInternational audienceThe fabrication of metallic parts through powder metallurgy techniques, such as spark plasma sintering (SPS), represents a relevant alternative to conventional manufacturing processes. This technique enables the production of dense, high-performance materials with<br /
New Industry 5.0 Techology Applied for a more flexible and Safier Grinding: Preserving Operator Expertise with an MR-Driven Robotic Solution
No full textInternational audienceGrinding operations in the metallurgy industry, particularly in small and medium-sized series present significant challenges due to the heavy and operator-intensive nature of the process on the operator. Conventional manual finishing exposes operators to hazardous conditions such as dust, noise and musculoskeletal disorders, making the job unattractive and leading to labor shortages. On the other hand, traditional automation methods often lack flexibility and find it difficult to adapt to different parts and varying defect characteristics. Hybrid methods such as collaborative robots (cobots) and exoskeletons, designed to assist workers, often restrict their natural movements, leading to new types of musculoskeletal disorders, and fail to protect against dust and noise. To overcome these limitations, an innovative approach is proposed, integrating mixed reality (MR) with industrial robotics to create a mechanized grinding system that enhances operator flexibility and safety while knowledge and decision-making stay within the human being.The presented system is based on an intuitive MR-based interface, implemented on a tablet, to enable the operator to maintain control over defect identification and treatment parameters. Using the tablet, the operator selects the surfaces requiring a finishing operation specifies defect types and assigns an intensity level to each defect. This information is then processed to automatically generate optimized robot trajectories. An industrial polyarticulated robotic arm performs the grinding operation based on predefined material behavior laws, control algorithms and real-time process adjustment to ensure optimum quality. Unlike conventional automated solutions solutions, this approach preserves the operator’s expertise and isolates him from the dangers associated to the process. Already examined and approved by a number of industrial companies, this research opens up a promising avenue for the future of finishing operations, particularly in companies requiring a high degree of adaptability and efficiency. By combining the advantages of human expertise with the strength and resilience of of robots, this approach paves the way for safer, more ergonomic, and more efficient processes in industrial environments.Les opérations de parachèvement dans la métallurgie, en particulier dans les entreprises travaillant en petites et moyennes séries, présentent des défis importants vis-à-vis de l’opérateur en raison de la nature lourde et exigeante du procédé mis en oeuvre. Le parachèvement manuel conventionnel expose les opérateurs à des conditions dangereuses telles que la poussière, le bruit et les troubles musculo-squelettiques rendant ainsi le métier peu attractif ce qui engendre des pénuries de main-d’œuvre. Les méthodes d’automatisation traditionnelles manquent souvent de flexibilité et ont du mal à s’adapter à diverses pièces et à des caractéristiques de défauts variables. Les méthodes hybrides telles que les robots collaboratifs (cobots) et les exosquelettes, conçus pour assister les travailleurs, restreignent souvent leurs mouvements naturels, ce qui provoque en outre de nouveaux types de troubles musculo-squelettiques et ne protège pas de la poussière et du bruit. Pour remédier à ces limitations, une approche innovante est proposée, intègrant la réalité mixte à la robotique industrielle pour créer un système de parachèvement mécanisé qui améliore la flexibilité et la sécurité de l’opérateur tout en maintenant le savoir faire et la prise de décision de l’être humain.Le système présenté s’appuie sur une interface intuitive basée sur la réalité mixte, implémentée sur une tablette, pour permettre à l’opérateur de garder le contrôle sur l’identification des défauts et les paramètres de traitement. Grâce à la tablette, l’opérateur sélectionne les surfaces qui nécessitent une opération de parachèvement, spécifie les types de défauts et attribue un niveau d’intensité à chaque défaut. Ces informations sont ensuite traitées pour générer automatiquement des trajectoires de robot optimisées. Un bras robotisé polyarticulé industriel exécute l’opération de meulage en s’appuyant sur des lois de comportement des matériaux prédéfinies, des algorithmes de contrôle et des mécanismes d’ajustement du procédé en temps réel pour assurer une qualité optimale. Contrairement aux solutions automatisées classiques, cette approche préserve l’expertise de l’opérateur et isole ce dernier des nuisances et dangers associés au procédé.Déjà examinées et approuvées par certaines entreprises industrielles, ces recherches ouvrent une voie prometteuse pour l’avenir des opérations de parachèvement, notamment dans les entreprises qui requièrent une grande adaptabilité. En combinant les atouts de l’expertise humaine avec la force des robots, cette approche ouvre la voie des procédés de parachèvement plus sûrs, plus ergonomiques et plus efficaces en milieu industriel.</p
Co-design of products and next-generation manufacturing systems – return on eight years of applied research
No full textInternational audienceIn a rapidly changing industrial context, manufacturing companies must ensure high flexibility while maintaining performance. Since 2017, applied research in collaboration with an automotive supplier has focused on co-designing products and assembly processes to develop modular and reconfigurable production systems. A new modelling approach has been introduced to stabilise product architectures, enabling product family identification and optimised multi-product assembly systems. Key findings highlight the importance of positioning and locating in assembly, along with an optimization tool addressing resource selection and task allocation. Practical applications include modular production lines and adaptable tools, successfully tested in industrial settings, reducing reconfiguration times and enhancing efficiency. Knowledge transfer activities have strengthened regional expertise, promoting innovation in standardization and modularity. These efforts demonstrate how collaborative research drives industrial competitiveness and fosters local innovation in the factory of the future.Dans un contexte industriel en mutation, les entreprises manufacturières doivent allier flexibilité et performance. Depuis 2017, une recherche appliquée en partenariat avec un équipementier automobile a permis de co-concevoir des produits et des processus d’assemblage pour développer des systèmes de production modulaires et reconfigurables. Une nouvelle approche de modélisation stabilise les architectures produits, facilitant l’identification de familles de produits et l’optimisation des systèmes d’assemblage multiproduits. L’importance du positionnement et du guidage dans l’assemblage a été mise en avant, ainsi qu’un outil d’optimisation pour la sélection des ressources et l’affectation des tâches. Des solutions concrètes, telles que des lignes de production modulaires et des outils adaptables, ont été testées avec succès, réduisant les temps de reconfiguration et améliorant l’efficacité. Le transfert de connaissances a renforcé l’expertise régionale, favorisant l’innovation et la compétitivité dans l’usine du futur
Smart disassembly cell for circularity: Technologies de l'industrie 4.0 pour le désassemblage des DEEE et la récupération des composants.
No full textInternational audienceSmart disassembly cell for circularity: Technologies de l'industrie 4.0 pour le désassemblage des DEEE et la récupération des composants
Digital twins for predictive monitoring and anomaly detection: application to seismic and railway infrastructures
No full textInternational audienceThis paper presents two case studies on the application of digital twins for predictive monitoring and anomaly detection in critical infrastructures: seismic monitoring for dams and sinkhole detection for railway tracks. The first case study focuses on the development of a surrogate model using neural networks to simulate seismic wave propagation and quantify uncertainties related to seismic hazards in dam infrastructure. By training a neural network on a dataset generated through the spectral element method, the model significantly reduces computational time and enables efficient global sensitivity analysis. This approach enhances the understanding of how key parameters influence seismic responses, supporting more reliable and responsive seismic monitoring systems for dams. The second case study addresses railway infrastructure, where sinkholes can lead to dangerous track deformations. Using high-resolution LiDAR data, we develop an automated system to detect ground deformations near railway tracks. By extracting ground points from LiDAR point clouds and generating Digital Elevation Models (DEMs), sinkholes are detected through a segmentation-based approach. Comparing different filtering algorithms, such as Voxel Based Ground Filtering (VBGF) and Cloth Simulation Filter (CSF), allows for optimizing the detection process and enhancing safety. Both case studies illustrate how digital twins can be employed for real-time monitoring, leveraging advanced data processing and predictive models to improve the safety and resilience of critical infrastructure
Mechanical instability lasts longer than the backward centre of pressure shift during gait initiation
No full textInternational audienceIntroduction: Ensuring balance is vital for everyday movement. During gait initiation, a common paradigm to assess balance-movement coordination in humans, mechanical instability must be generated during the anticipatory period to initiate movement. A first phase of the anticipatory period contributing to generate mechanical instability was described, with a backward centre of pressure (CoP) shift, accelerating the whole-body centre of mass (WBCoM) forward [1]. However, this analysis does not consider the rotational contribution of the free segments. The internal whole-body angular momentum (HM) around the mediolateral axis also exhibited a first phase where HM is directed forward, followed by a second one where some segments rotated in the opposite direction to limit the mechanical instability and the increase in HM magnitude, interpreted as a stabilising mechanism [2]. The current study aims to determine whether the forward directed HM only corresponds to the mechanical instability generated by the backward CoP shift, or whether the mechanical instability lasts longer than the backward CoP shift.Methods: Thirteen healthy participants (23.5 ± 4.2 years, 1.72 ± 0.1 m, 65.5 ± 8.8 kg) performed three gait initiation trials. Participants were asked to voluntarily initiate gait and walk for five meters from a hip-width upright standing posture. Ground reaction forces and moments were recorded by four force plates (Sensix) at 2000 Hz, lowpass filtered at 20 Hz and used to compute the CoP. The trajectories of 54 markers placed on anatomical landmarks were recorded with 15 infrared cameras (Vicon) at 200 Hz, lowpass filtered at 10 Hz. HM was computed as the sum of the segment angular momenta transferred to the WBCoM based on a 13 segments whole-body model. The coefficient of cancellation, defined as the amount of segment angular momentum compensating each other divided by the total magnitude of HM was then computed [3]. The duration of the mechanical instability phase was computed using the CoP and HM methods. For both methods, the phase began when the WBCoM acceleration was larger than the mean +3 standard deviations measured during quiet standing. The instability phase ended when the CoP was the most backward and toward the swing limb [1], and when the time derivative of HM became positive (i.e., when some segments started to rotate in the opposite direction), respectively. We compared the phase durations computed by each method using a paired-sample t-test.Results: The duration of the mechanical instability phase based on the CoP was significantly shorter than the one based on HM (260 ± 73 vs. 368 ±101 ms, p<0.001).[Figure 1: Average and standard deviation of internal whole-body angular momentum (HM, top) and coefficient of cancellation (bottom) during the anticipatory period, ending at foot-off. The black and red vertical lines represent the phase delimitation based on the CoP and HM methods, respectively.]Discussion: Our results suggest that the whole-body continues to rotate forward after the end of the backward CoP shift (after the black vertical line in Figure 1), without any regulation from opposite segment rotation, as indicated by a low coefficient of cancellation until change in the time derivative of HM. This suggests that mechanical instability is not large enough to initiate gait at the end of the backward CoP shift, and that the whole-body must continue to rotate forward before any regulation appears. This phase of mechanical instability would not have been detected by analysing only the CoP. These results being obtained on young healthy participants, future studies should assess if this longer period of mechanical instability is also present in older participants and/or participants with balance deficits.References:1. Crenna et al, Exp Brain Res, 172 :519-532, 20062. Bechet et al, MBJ, 1, 20243. Bennet et al, Hum. Mov. Sci, 29 :114-124 : 201
Enhanced terahertz nondestructive evaluation and stratigraphic reconstruction with filtered deconvolution
No full textInternational audienceTerahertz pulsed imaging (TPI) is a valuable tool for nondestructive evaluation (NDE), enabling high-resolution stratigraphic characterization and structural inspection of various materials. However, its accuracy -defined by both the precise localization of peaks and the reduction of spurious peaks -can be limited by challenges such as low signal-to-noise ratio (SNR) and temporally overlapping echoes when layers are thin. While advanced signal-processing methods such as sparse deconvolution (SD) and orthogonal matching pursuit (OMP) have shown promise in addressing these issues, their effectiveness often hinges on careful parameter selection, which typically requires significant expertise. This paper presents a novel approach that enhances the accuracy and simplifies parameter determination in THz TPI by applying SD to a filtered impulse response function in the time domain. This method also aids in OMP parameter determination, further advancing the precision and usability of THz TPI for NDE applications. We demonstrate the effectiveness of the proposed approach by applying it to experimental data obtained from Willow glass on silicon and mill scale on hot-rolled steel strip, showcasing its versatility across different material systems.</div
Etude des cinétiques d'oxydation et de nitruration appliquées aux aciers électriques à grains orientés
No full textIn order to meet the various environmental challenges, electrical steels for electrical conversion have undergone a number of developments in recent years. This is reflected in an optimization of their Goss texture orientation {110} . The latter is obtained following numerous processing stages, including the two thermochemical treatments studied in this thesis: decarburization in an oxidizing atmosphere and high-temperature nitriding.In addition to removing carbon and forming an internal silicon oxide layer, decarburization recrystallizes the material, eliminating the effects of rolling on atomic diffusion. A study of the influence of the various parameters on the microstructure is carried out and reveals oxygen diffusion following the fundamental principles of diffusion. Oxygen diffusion is not limited by oxide precipitation, and oxide morphology has no influence on oxygen diffusion kinetics.The oxide layer has an influence on nitriding, allowing the surface to behave like an Fe-N system and leading to increased nitrogen adsorption at the surface, which diffuses underneath and then precipitates as nitrides. When decarburization is ineffective, high carbon content is observed which, combined with nitrogen adsorption, leads to austenitic transformation of the matrix, potentially altering processing kinetics. A parametric study of the nitriding of electrical steels was also carried out. Among the parameters studied, only temperature has a significant influence in limiting nitrogen adsorption. Parameters can then be optimized according to the desired microstructural properties.Afin de répondre aux différents enjeux environnementaux, les aciers électrique assurant la conversion électrique ont connu de nombreux développements ces dernières années. Ceci se traduit par une optimisation de leur texture de Goss d'orientation {110} . Cette dernière est obtenue à la suite de nombreuses étapes de traitements comprenant les deux traitements thermochimique étudiés dans cette thèse : une décarburation sous atmosphère oxydante et une nitruration haute température.En plus d'assurer une élimination du carbone et la formation d'une couche d’oxyde interne de silicium, la décarburation permet une recristallisation du matériau, éliminant alors les effets du laminage sur la diffusion atomique. Une étude de l'influence des différents paramètres sur la microstructure est réalisé et révèle une diffusion de l'oxygène suivant les principes fondamentaux de la diffusion. Cette dernière n'est pas limitée par la précipitation d'oxydes et leur morphologie n'intervient pas dans les cinétiques de diffusion de l'oxygène.La couche d'oxyde présente une influence sur la nitruration, en permettant à la surface de se comporter comme un système Fe-N et entraînant l'augmentation de l'adsorption d'azote à la surface qui diffuse sous celle-ci pour ensuite précipiter en nitrures. Lors d'une décarburation peu effective, d'importante teneur en carbone sont relevée qui, associées à l'adsorption d'azote entraîne une transformation austénitique de la matrice, altérant potentiellement les cinétiques de traitement. Une étude paramétrique de la nitruration des aciers électriques a également été faite. Parmi les paramètres étudiés, seule la température a une influence notable en limitant l'adsorption de l'azote. Une optimisation des paramètres est alors possible en fonction des propriétés microstructurales souhaitées
Degenerative cervical myelopathy: timing of surgery
No full textInternational audienceBackground Despite the growing burden of degenerative cervical myelopathy (DCM), consensus on the optimal timing of surgical intervention remains lacking, especially for patients with mild symptoms or asymptomatic cord compression or in the context of recent trauma. Different scores, such as the mJOA, Nurick scale and NDI are commonly used to classify disease severity, but guidelines for managing these patients do not provide a clear framework for intervention timing. Materials and methods We conducted a narrative review of the literature on the optimal timing of surgical intervention for DCM, using PubMed to identify relevant studies. The search was focused on surgical and non-operative management, clinical and radiological assessments, biomarkers and emerging technologies. The selected papers were reviewed for relevance and quality, with guidance from a senior author. Results The initial search identified 6,705 articles, which were narrowed down to 136 relevant studies after applying filters for study type and clinical focus. A final selection of 87 papers was categorized by topics and the findings were synthesized to highlight trends, challenges and knowledge gaps in surgical timing for DCM. Focus of the study This review article examines strategies for determining the optimal timing for surgery in DCM. It explores how radiological signs, clinical indicators and other markers may help identify patients at risk of rapid neurological deterioration, particularly in the ‘grey-zone’ population (mild symptoms or asymptomatic disease), enabling clinicians to assess correctly different clinical scenarios and to indicate timely surgical intervention
A climatology of local hourly wet spells across the tropics
No full textInternational audienceWe explore the relationships amongst duration, total amount, mean and maximum hourly intensity of rainfall and spatial scale for more than a thousand rain gauges covering a diverse range of tropical and subtropical (30°N-30°S) climates, from arid (< = 200 mm year⁻1) to very wet (> 3000–4000 mm year⁻1). We find that the interannual variation of seasonal (3-month) amounts is primarily driven by wet hour frequency rather than the mean intensity of rainfall. A total of 3.5 million local wet spells (WS: consecutive wet hours with at least 1 mm of rain) are then systematically analyzed. WS lasting 5 h or less account for 80% of wet hours and of total rainfall. The amount of rainfall during a local WS is most strongly controlled by its duration and then its mean hourly intensity, but these WS characteristics are nearly independent. The 3.5 million local WS are then grouped into a reduced set of 7 “canonical” wet spells (cWS) from the 24 hourly rainfall from their start (dry hours < 1 mm are reset to 0), firstly square-rooted and then simply normalized by the 24 hourly means across all WS. Despite some relatively small differences amongst national networks, the seven cWS offer a general framework for tropical rainfall variability seen through the prism of the local WS. About 72% of the WS included in two categories (cWS#1, 51,8% of the WS, and #3, 20% of the WS) are short and result in negligible amounts of rainfall. Two other categories (cWS#2, 8.8% of the WS, and #4, 4.9% of the WS) are short but with very high mean intensity and have the smallest spatial scales. Additionally, two categories (cWS #6, 3.9% of the WS, and cWS #7, 0.9% of the WS) are long-lasting, moderately intense and cover the largest areas. The final category (cWS#5, 9.4% of the WS) is moderately long in duration and less intense compared to cWS#2, #4, #6 and #7. Hourly extremes tend to occur primarily during cWS#2 and cWS#4, contributing the least to the total number of wet hours, and then during cWS#6. Daily extremes tend to occur more in cWS#7 and #6, which are the longest cWS, but with a significant contribution from the short intense cWS#4. A larger contribution of cWS#2 and cWS#4 to 3-month amounts tends to decrease the spatial scale of their interannual variations