HAL-MINES ParisTech
Not a member yet
27348 research outputs found
Sort by
δ13C, HCO3- and Ca2+ dynamics to assess inorganic carbon production and transport in a small carbonaceous catchment (Orgeval, France)
No full textInternational audienc
Suppression of measurement-induced state transitions in cosϕ-coupling transmon readout
No full textDrive-induced unwanted state transitions (DUST) are limiting both for microwave readout and parametric operations of superconducting qubits. Among them, measurement-induced state transitions (MIST) are due to intrinsic resonances described by the readout Hamiltonian. They were previously studied with a qubit linearly coupled to its readout mode, which constitutes the usual readout Hamiltonian. Since MIST can appear even at moderate powers, they limit the readout SNR and the QND readout fidelity. In this work, we study the high-power readout regime in a different transmon readout scheme, implementing a nonlinear coupling called the cosφ-coupling. This coupling stems from a transmon molecule circuit and has symmetry properties that suppress nonparity-conserving MIST. We succeed in performing multi-state single-shot readout up to the fifth excited state of the transmon, which enables us to identify leakage pathways from the computational subspace. The measurements indicate that the system is free of MIST up to high powers, with more than 300 photons in the readout mode. The MIST can be controllably turned on by breaking the parity symmetry of the coupling using flux-tuning. These experimental results are corroborated by branch analysis and simulations of the classical chaotic dynamics, showing that the cosφ-coupling is very robust to readout photons compared to the usual transverse coupling
HARP-NeXt: High-Speed and Accurate Range-Point Fusion Network for 3D LiDAR Semantic Segmentation
No full textInternational audienc
HD-OOD3D: Supervised and Unsupervised Out-of-Distribution object detection in LiDAR data
No full textInternational audienc
Scenario for the formation of fretting obtained TTS in TA6V from detailed microstructural analysis
No full textInternational audienceFretting is defined as surface degradation caused by small oscillatory movements. Fretting in the gross slip regime produces high local stresses at the surface. This results in wear and subsurface plastic deformation, which can create a nanosized-grained structure known as a Tribologically Transformed Structure (TTS). If the condition of TTS formation in the fretted interface is well documented in the literature, the metallurgical transformation leading to this structure is not well known. This article aims to study this phenomenon thanks to an innovative in-depth texture analysis of the TTS which is carried out using a TEM crystal plane indexing technique. Several TTS from different steps in the TTS formation are studied. Special care is taken in determining their texture, and chemical analysis is carried out. Considering the temperatures involved and the successive structures analyzed, a new model of continuous dynamic recrystallization via the apparition of shear bands is proposed to explain the heterogeneity of TTS and its microstructural characteristics
Contributions à l'étude des Systèmes Logistiques Souterrains
No full textThis work focuses on Underground Logistics Systems (ULS) in urban logistics. Underground freight transport holds significant potential to reduce road congestion, pollution, and logistics costs, while enhancing transport reliability. Although several systems showed early promise as far back as the 19th century, many were eventually decommissioned due to the decline of transported goods and competition from road freight. Today, interest in ULS persists, particularly in China, yet their adoption is hindered by economic, regulatory, and logistical barriers. The first part of this research reviews the historical development and current challenges of ULS. The second part explores the conditions under which such systems could minimize the energy losses associated with urban freight distribution. The third part examines the temporal patterns of goods deliveries and assesses the extent to which ULS can meet recipients' time constraints. The entire study uses the city of Paris and the Île-de-France region as its application field.Cette thèse porte sur les Systèmes Logistiques Souterrains (SLS) appliqués à la logistique urbaine. Le transport souterrain de marchandises présente un potentiel important pour réduire la congestion routière, la pollution et les coûts logistiques, tout en renforçant la fiabilité des flux. Malgré des débuts prometteurs dès le XIXᵉ siècle, de nombreux systèmes ont été abandonnés face au déclin des marchandises transportées et à la concurrence du transport routier. Aujourd'hui, l'intérêt pour ces systèmes subsiste, notamment en Chine, mais leur déploiement se heurte à des freins économiques, réglementaires et logistiques. La première partie de ce travail étudie l'histoire et les enjeux actuels des SLS. La deuxième partie explore les critères sous lesquelles de tels ces systèmes contribueraient à minimiser les pertes énergétiques liées à la distribution urbaine. Enfin, la troisième partie examine la répartition temporelle des livraisons et évalue la capacité de tels systèmes à répondre aux exigences temporelles des destinataires. L'ensemble de ces travaux prend la ville de Paris et l'Île-de-France comme terrain d'application
Caractérisation et simulation de la croissance de grains en 3D d'un superalliage base nickel polycristallin
No full textThis PhD thesis focuses on the development and application of advanced 3D characterization and simulation techniques for Inconel 718, a nickel-based superalloy used in aircraft engines. Two metallurgical states are studied: an equiaxed microstructure and an additively manufactured microstructure after thermal treatment. A dual-beam PFIB-SEM is used to develop and optimize serial-sectioning protocols for the equiaxed state, while mechanical polishing coupled with EBSD is employed for the additively manufactured state. The acquired EBSD maps are processed for 3D reconstruction, and guidelines for this workflow are outlined. Reconstructed microstructures are integrated into a finite element framework to study the effects of meshing strategies on grain boundary topology, emphasizing voxelization limitations. Isotropic and heterogeneous grain growth simulations are conducted using different grain boundary properties models, even incorporating twin boundaries. Finally, 2D and 3D simulations are compared, analyzing differences in grain growth topology and grain size distribution evolutions.Cette thèse de doctorat se concentre sur le développement et l'application de techniques avancées de caractérisation et de simulation 3D pour l'Inconel 718, un superalliage base nickel utilisé dans les turboréacteurs. Deux états métallurgiques sont étudiés : une microstructure équiax et une microstructure fabriquée de manière additive après traitement thermique. Un microscope PFIB-SEM à double faisceau est utilisé pour développer et optimiser les protocoles de coupes en série pour l'état équiaxe, tandis que le polissage mécanique couplé à l'EBSD est employé pour l'état fabriqué de manière additive. Les cartes EBSD obtenues sont traitées pour une reconstruction 3D, et des méthodes pour cette transition sont détaillées. Les microstructures reconstruites sont intégrées dans des simulations éléments finis afin d'étudier les effets des stratégies de maillage sur la topologie des joints de grains, en mettant l'accent sur les limitations de la voxelisation. Des simulations en croissance de grains isotropes et hétérogènes sont réalisées en utilisant différents modèles de propriétés des joints de grains, incluant même les joints de macles. Enfin, des simulations 2D et 3D sont comparées, en analysant les différences de topologie de croissance des grains et l'évolution de la distribution de la taille des grains
Predicting thrombosis in cerebral aneurysms using computational methods and graph neural networks
No full textInternational audienc
Efficient embedding initialization via dominant eigenvector projections
No full textInternational audienceThe embedding layer is essential in deep learning, transforming high-dimensional data into compact representations. However, growing datasets and model sizes pose challenges in training time, memory, and generalization. We propose a scalable method for embedding initialization via spectral dimensionality reduction using dominant eigenvector projections.The proposed approach leverages on MIRAMns, multiple implicitly restarted Arnoldi method with nested subspaces, to extract most informative directions from large and potentially sparse data representations. Unlike traditional embeddings or autoencoders, this proposed approach requires few tunable parameters and is inherently parallel. We apply MIRAMns to matrix representations such as covariance and co-occurrence matrices to compute lowdimensional embeddings that preserve data structure and variance. Experiments across diverse datasets show that the proposed method achieves comparable or better accuracy with significantly reduced dimensionality, enabling smaller, faster deep networks. Additionally, our parallel implementation scales efficiently on HPC platforms, making it well-suited for large-scale scientific and AI workloads.</p
Molecular analysis highlights TREM2 as a discriminating biomarker for patients suffering from pancreatic ductal adenocarcinoma
No full textInternational audiencePancreatic cancer is projected to become the second leading cause of cancer-related deaths by 2030, with its mortality continuing to rise, unlike other common cancers such as breast or colorectal. Late-stage diagnosis, often accompanied by metastatic dissemination, drastically impairs patient survival and underscores the urgent need for improved biomarkers to guide therapeutic strategies. While molecular signatures have been proposed to stratify pancreatic cancer patients, their ability to predict outcomes remains limited. In this study, we applied established molecular signatures to our in-house transcriptomic data from a cohort of pancreatic cancer patients. We took advantage of published datasets to construct comprehensive atlases of cells present in primary and metastatic pancreatic cancers. The atlas of metastasis samples, representative of routinely harvested patient biopsies, revealed that monocyte/macrophage signatures provided superior discriminatory power compared to existing molecular classifications. Notably, the abundance of TREM2-expressing macrophages emerged as a significant parameter for stratifying patients. Our findings position TREM2+ macrophages as a promising biomarker for pancreatic cancer, with potential to enhance patient stratification and inform the development of targeted therapies. This work highlights the critical role of tumor-associated macrophages in pancreatic cancer progression and lays the groundwork for further functional and translational studies