HAL-MINES ParisTech
Not a member yet
27348 research outputs found
Sort by
Controlling the properties of polymeric biomaterials via macromolecular engineering and processing parameters
No full textImpact of polystyrene re-processing on Non-Intentionally Added Substances (NIAS)
No full textInternational audienc
Natural vs Synthetic Refrigerants in Geothermal Heat Pumps: A Life Cycle Assessment for Sustainable Decision-Making
No full textNational audienc
Recovering sparse DFT from missing signals via interior point method on GPU
No full textWe propose a method to recover the sparse discrete Fourier transform (DFT) of a signal that is both noisy and potentially incomplete, with missing values. The problem is formulated as a penalized least-squares minimization based on the inverse discrete Fourier transform (IDFT) with an -penalty term, reformulated to be solvable using a primal-dual interior point method (IPM). Although Krylov methods are not typically used to solve Karush-Kuhn-Tucker (KKT) systems arising in IPMs due to their ill-conditioning, we employ a tailored preconditioner and establish new asymptotic bounds on the condition number of preconditioned KKT matrices. Thanks to this dedicated preconditioner -- and the fact that FFT and IFFT operate as linear operators without requiring explicit matrix materialization -- KKT systems can be solved efficiently at large scales in a matrix-free manner. Numerical results from a Julia implementation leveraging GPU-accelerated interior point methods, Krylov methods, and FFT toolkits demonstrate the scalability of our approach on problems with hundreds of millions of variables, inclusive of real data obtained from the diffuse scattering from a slightly disordered Molybdenum Vanadium Dioxide crystal
Sensitivity analysis for parametric nonlinear programming: A tutorial
No full textThis tutorial provides an overview of the current state-of-the-art in the sensitivity analysis for nonlinear programming. Building upon the fundamental work of Fiacco, it derives the sensitivity of primal-dual solutions for regular nonlinear programs and explores the extent to which Fiacco's framework can be extended to degenerate nonlinear programs with non-unique dual solutions. The survey ends with a discussion on how to adapt the sensitivity analysis for conic programs and approximate solutions obtained from interior-point algorithms
New approach for 3D printing of cellulose solutions and making aerogels from them
No full textInternational audiencePolysaccharide-based 3D printed objects are promising materials with customized shapes for life science applications. A classical approach is to use gelation to stabilize the shape of printed solution. To vary solution rheological properties, nanocellulose is often used. In this work, we developed a new approach allowing printing dissolved cellulose with no crosslinking and no additives, only by varying solution rheological properties and adjusting cellulose solubility. These printed cellulose objects were then transformed into aerogels, lightweight and nanostructured materials. In addition, we demonstrated that this approach can be applied not only to neat reference cellulose, but also to dissolved cellulose textile.Microcrystalline cellulose and viscose textile were dissolved in ionic liquid. By adjusting cellulose solubility, solution viscoelastic properties were varied with the goal to obtain yield-stress fluid. The printed objects were of various shapes, composed of many layers and self-standing. The solvent was then washed out by a non-solvent, and cellulose was dried with supercritical CO2. 3D printed cellulose aerogels obtained from MCC and viscose possess porosity higher than 90%, a specific surface area around 250 m²/g and a low density (< 0.2 g/cm3).The developed approach opens new easy ways of printing cellulose solutions and valorizing waste textile.AcknowledgmentsWe are grateful to Institut Carnot MINES and PSL for the financial support of this work. We thank Julien Jaxel (PERSEE, MINES Paris) for supercritical drying
STARCH AND STARCH-HYDROXYAPATITE AEROGELS AND XEROGELS: HIERARCHICALLY STRUCTURED BIO-BASED MATERIALS FOR POTENTIALCANCELLOUS BONE REPAIR
No full textInternational audiencePorous starches are promising bio-based materials for biomedical use, notably as scaffolds and drug carriers. Potato starch gels were obtained via starch dissolution, retrogradation, solvent exchange and either supercritical drying CO₂ (aerogels) or evaporative drying (xerogels). Incorporating 100-200 µm hydroxyapatite (HA) based-beads produced porous starch-HA xerogels with preserved nanostructure, suitable for bone regeneration. Cytotoxicity tests confirmed material safety. 3Dprinting enabled hierarchical control, making these constructs viable candidates for cancellous bone substitutes
Quality Control of Tilted Solar Irradiance: A Systematic and Generalizable Procedure
No full textInternational audienceSolar irradiance on tilted surfaces is a key input for photovoltaic energy production. Yet, most existing quality-control procedures have focused almost exclusively on horizontal measurements, largely overlooking tilted-plane data. In this study, two novel methods are proposed to assess the quality of solar irradiance on tilted surfaces. The first relies solely on solar and surface geometry, combined with the historical maximum ground albedo at the site, to establish acceptable upper and lower limits. The second is based on a clear-sky model, accounting for its known uncertainties and the historical maximum ground albedo, to define similar limits. Both methods are validated using four years of one-minute data from surfaces with various tilt angles in southern France. Results indicate that the geometry-based method proves simple and effective, while the clear-sky–based method provides tighter envelopes at the cost of requiring more detailed inputs. Together, these tools enhance the reliability of solar data for photovoltaic applications
Les marchés publics : un outil de politique industrielle ? Réindustrialiser par les achats
No full textLa France investit des milliards dans la réindustrialisation, tout en attribuant ses marchés publics àdes entreprises installées à l’étranger.Cette contradiction, souvent passée sous silence, est pourtant au coeur d’un paradoxe structurelde notre action publique : la puissance d’achat de l’État — près de 300 milliards d’euros par an, soitenviron 10% du PIB — reste largement pensée comme une simple fonction administrative, alors mêmequ’elle pourrait devenir un levier central de politique industrielle.Pourrait-on utiliser les achats de l’État comme un outil de politique industrielle ?Nous défendons que, dans un contexte marqué par les relocalisations critiques, les ruptures d’approvisionnement,la transition écologique et la nécessité de sécuriser des filières stratégiques, la commandepublique offre un champ d’action décisif. Elle peut orienter les chaînes de valeur, soutenir les technologiesémergentes, renforcer les bassins d’emploi industriels. Mais pour cela, encore faut-il lui donner unedirection claire, des outils adaptés, et des règles d’action compatibles avec l’objectif poursuivi.Ce mémoire explore cette opportunité. Il part d’un double constat : d’une part, les outils juridiques etopérationnels existent — ou pourraient exister — pour articuler commande publique et ambition industrielle; d’autre part, ces outils sont aujourd’hui sous-utilisés, dispersés ou percutés par des contraintescontradictoires (juridiques, organisationnelles, culturelles).Notre démarche s’appuie sur une cinquantaine d’entretiens menés avec des acheteurs, juristes, industriels,acteurs territoriaux et institutionnels. Elle propose une lecture renouvelée du droit, des pratiqueset des marges de manoeuvre, avec une ambition : faire de la commande publique un levier stratégiquepleinement assumé, piloté et professionnalisé, au service de la réindustrialisation du pays.Nous avons choisi de structurer l’analyse selon les grandes étapes du processus achat — de ladéfinition de la politique achat à l’évaluation des résultats — en ciblant les moments-clés où des choixstructurants peuvent être opérés. À chaque étape, nous formulons des propositions concrètes, réalisteset juridiquement fondées pour aligner la commande publique sur les priorités industrielles de la Nation.Car la commande publique n’est pas un simple acte de dépense. Elle est un acte de puissance. Et dans une période où l’on souhaite réarmer l’économie française, il est temps de la considérer comme telle
ATOMSURF: Surface representation for learning on protein structures
No full textInternational audienceWhile there has been significant progress in evaluating and comparing different representations for learning on protein data, the role of surface-based learning approaches remains not well-understood. In particular, there is a lack of direct and fair benchmark comparison between the best available surface-based learning methods against alternative representations such as graphs. Moreover, the few existing surface-based approaches either use surface information in isolation or, at best, perform global pooling between surface and graph-based architectures. In this work, we fill this gap by first adapting a state-of-the-art surface encoder for protein learning tasks. We then perform a direct and fair comparison of the resulting method against alternative approaches within the Atom3D benchmark, highlighting the limitations of pure surface-based learning. Finally, we propose an integrated approach, which allows learned feature sharing between graphs and surface representations on the level of nodes and vertices across all layers. We demonstrate that the resulting architecture achieves state-of-the-art results on all tasks in the Atom3D benchmark, while adhering to the strict benchmark protocol, as well as more broadly on binding site identification and binding pocket classification. Furthermore, we use coarsened surfaces and optimize our approach for efficiency, making our tool competitive in training and inference time with existing techniques. Code can be found online: github.com/Vincentx15/atomsur