Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Jusqu’au bout de l’humain. Critique raisonnée de l’expansionnisme spatial
No full textInternational audienceCet article présente une critique des revendications éthiques qui sous-tendent ce qu’il appelle le nouveau futurisme. L’auteur suggère (sans établir ni nier la possibilité de tels cas) que tout argument légitime en faveur de l’expansion de l’habitat humain au-delà de la planète devrait rester humain, c’est-à-dire qu’il devrait montrer la valeur éthique de l’expansion au-delà de la Terre en termes de qualité de vie qu’elle pourrait offrir à un nombre fini d’êtres qui vivent eux-mêmes le type de vie limitée et écologiquement vulnérable que nous vivons aujourd’hui sur la planète Terre
Analytical and numerical study of the lateral buckling of an imperfect pipe accounting for torsional effects
No full textInternational audienceThe stability problem of pipelines under high temperature and high pressure in terms of lateral buckling has been largely concerned recently. Owing to the complex conditions encountered in the sea bottom, many factors affect the structural behavior of pipelines resting on the seabed, such as initial imperfections and unexpected torsion. This paper thus focuses on the mechanical response of imperfect pipes (with a fixed length) under uniaxial compression, in presence of torsional effects. The problem is addressed first in an analytical way. The classical governing equations of a beam lying on the seabed are adjusted, introducing extra terms accounting for initial imperfections and twisting effects. The cases of imperfect pipes with free and fixed torsion at both ends are investigated, and closed-form analytical solutions are obtained in most cases. Furthermore, non-linear finite element computations are performed. Results include force–displacement response curves and buckled deformed shapes. A detailed comparison is conducted between numerical results and analytical solutions, accounting for various lateral stiffness densities, and different imperfection types and amplitudes. A general good agreement is achieved, at least within a small deformation domain. However, the numerical calculations sometimes lead to unexpected phenomena that analytical solutions cannot capture due to geometric non-linearities
Système Mondial de Navigation par Satellite Embarqué sur Porteur Marin pour l’Observation de la Vapeur d’Eau Atmosphérique en Mer
No full textGlobal Navigation Satellite System (GNSS) has unfolded as a ground-based atmosphere monitoring tool. Some studies aiming to compensate for the lack of atmosphere sensing above the oceans began to apply this method to shipborne GNSS antennas. The undertaken work through this thesis put forward processing automation of longlasting shipborne GNSS antenna datasets. To this end, we first established a processing strategy suitable for shipborne GNSS datasets by testing several processing configurations on actual and simulated data. The estimates are analyzed to provide a best-suited processing configuration for water vapor retrieval at sea. We applied the recommended optimal strategy to the long-lasting dataset processing of the campaigns from the French Oceanographic Fleet between 2015 and 2022. We compared the extracted precipitable water vapor to external datasets before assessing their potential use in climate studies. This automated processing strategy adapted to routine measurements could lead to a numerical weather prediction application.Le Système de Navigation Mondial par Satellite (GNSS) s’est peu à peu imposé comme outil de mesure atmosphérique, au service de l’étude du climat et de la prévision météorologique à terre. Afin de pallier le manque d’observation de la vapeur d’eau atmosphérique en mer, différentes études l’ont appliqué à des antennes GNSS embarquées sur porteur marin. Les travaux de recherche menés durant cette thèse proposent une automatisation du traitement de données GNSS d’antennes embarquées sur porteur marin, appliquée à des données acquises sur une longue période. Pour cela, nous établissons une stratégiede traitement adaptée aux données GNSS d’antennes embarquées en mer, en appliquant différentes configurations de traitement des données réelles et simulées. Les solutions estimées sont évaluées afin de recommander une stratégie adaptée à l’estimation de vapeur d’eau en mer.La configuration optimale retenue est utilisée pour post-traiter les jeux de données binaires acquis lors des campagnes océanographiques effectuées entre 2015 et 2022 par la Flotte Océanographique Française. La vapeur d’eau précipitable ainsi extraite est évaluée par comparaison à d’autres sources de données, puis une étude approfondie du résultat est présentée afin d’en montrer l’apport pour l’étude du climat. L’adaptation de ce traitement à du temps réel pourrait permettre d’utiliser ces données pour la prévision météorologique
Investigation of thermo-mechanical shape memory signatures of 3D printed and Injection molded polymers
No full textInternational audienceThis research paper investigates the shape memory signatures (SMSs) of SMPs fabricated by two different manufacturing processes, namely Fused Filament Fabricated (FFF) and Injection-molding (IM). The SMSs of each manufacturing route is first investigated by rating the thermophysical and thermodynamic properties of amorphous and semicrystalline-type shape memory polymers. Secondly, the capturing of SMSs is completed by a series of thermomechanical shape programming/recovery cycles on FFF printed specimens, and the results are correlated with the IM-produced samples of similar nature. Furthermore, the thermomechanical cycles are programmed at varying heating rates (1, 2, 3, and 5 °C min−1) to analyze their influence on shape memory behavior and cycling time. Results depict that an increase in induced pre-strain driven by small to large deformations (10% → 25%) in the elastic range leads to maximum shape memory performance and the highest performance was reported while heating at 2 and 3 °C min−1, respectively. Finally, the IM-produced samples showed slightly better shape recovery compared to the 3D-printed samples. The shape memory signatures of 3DP specimens are indeed up to 86% like the ones produced by injection molding, which is mostly governed by the distinct residual stress/strain induced during the two different manufacturing processes
Quelle place pour les enjeux environnementaux dans les formations d’ingénieurs en Algérie ? Le cas de l’École Nationale Polytechnique d’Alger
No full textInternational audienceAlors que les questions environnementales résonnent comme des incontournables dans les formations d’ingénieurs contemporaines en France et en Europe, qu’en est-il ailleurs, dans des pays nord-africains confrontés encore davantage, et plus brutalement encore, aux impacts du réchauffement climatique ? Est ici proposée une étude de la formation des ingénieurs en Algérie, et de la place qui y est réservée aux questions environnementales. Elle est basée sur une analyse des programmes de formation de l’Ecole Nationale Polytechnique d’Alger et sur une enquête par entretiens avec des enseignants menée de 2018 à 2020. Cet article montre comment se réalise la production curriculaire dans une grande école d’ingénieurs, par qui et avec quels débats. S’en dégage une nette autonomie des enseignants, lesquels sont peu bridés par des référentiels et des prescriptions mais confrontés à d’autres types de contraintes, en termes d’organisation institutionnelle et de ressources. Les enseignants introduisent dans les programmes des cours relatifs à l’environnement mais de manière réduite par rapport à leurs ambitions de formation, tout en s’accordant à dire qu’il y a multiplicité des urgences, l’urgence environnementale n’en étant qu’une parmi d’autre
Long‐term acoustic monitoring of nonstereotyped blue whale calls in the southern Indian Ocean
No full textInternational audienceMonitoring the presence of blue whale (Balaenoptera musculus ssp.) stereotyped calls has been a widely used method to assess the different populations' distribution worldwide. All blue whale populations also produce nonstereotyped vocalizations, or D-calls. Here, we monitored the presence of D-calls in long-term records from a large hydrophone array located in the open southern Indian Ocean, using an automated detection method and manual validation of the detections. D-calls were detected at all sites of the array, which extends from 24°S to 56°S, but the majority of them were detected at the two southernmost sites. We observed a latitudinal shift in their seasonal occurrence, with more D-calls in the north during austral autumn and winter and more in the south during austral spring. The geographical occurrence of D-calls compared to that of songs indicates that blue whale acoustic behavior switches from a song-intensive and sparse-D-call emission in the north to song-moderate and more intensive D-call emissions in the south. These findings support the hypothesis that both call types are used for different purposes, as D-calls are mainly detected around foraging grounds and songs in wintering grounds. Monitoring both call types might therefore be a relevant acoustic indicator of blue whale behavior
A systemic analysis of the technological trajectory at company level based on patent data: The case of Sanofi's vaccine technology
No full textInternational audienceWhereas most of technological trajectory research in evolutionary economics focuses on the sectoral level, ourobjective is to propose an analysis at the scale of the company. We are studying the leading pharmaceuticalcompany Sanofi and focusing on the development of its vaccine technologies, a very topical issue in the Covid-19context. What are the characteristics of Sanofi’s vaccine technology trajectory and how can they explain thecompany’s position in the Covid-19 crisis, i.e. its ability (or not) to innovate and respond to social demand for avaccine? We contribute to technological trajectory literature by integrating a systemic dimension of the linksbetween source and destination technology. Our ‘Knowledge Capital Approach’ is based on patent citation dataand network analysis. It has allowed us to identify the trajectory of Sanofi’s vaccine technical system and to showits central position in its knowledge capital, where the technologies mastered by the company converge. Theseresults can be viewed as paradoxical when considering Sanofi’s lag in the Covid-19 race for a new vaccine. Ourhistorical and qualitative approach adds to these findings by showing the role of strategic and institutionalfactors in explaining this situation
VIBR: Learning View-Invariant Value Functions for Robust Visual Control
No full textInternational audienceEnd-to-end reinforcement learning on images showed significant progress in the recent years. Data-based approach leverage data augmentation and domain randomization while representation learning methods use auxiliary losses to learn task-relevant features. Yet, reinforcement still struggles in visually diverse environments full of distractions and spurious noise. In this work, we tackle the problem of robust visual control at its core and present VIBR (View-Invariant Bellman Residuals), a method that combines multi-view training and invariant prediction to reduce out-of-distribution (OOD) generalization gap for RL based visuomotor control. Our model-free approach improve baselines performances without the need of additional representation learning objectives and with limited additional computational cost. We show that VIBR outperforms existing methods on complex visuo-motor control environment with high visual perturbation. Our approach achieves state-of the-art results on the Distracting Control Suite benchmark, a challenging benchmark still not solved by current methods, where we evaluate the robustness to a number of visual perturbators, as well as OOD generalization and extrapolation capabilities
Cultural evolution in populations with heterogeneous and variable preferences
No full textTheoretical and empirical work in cultural evolution often assume populations in which individuals all agree on the quality of a given cultural trait (i.e. they have homogeneous preferences), and where those preferences are stable in time. Yet, these assumptions are not always met: for example, an uneven distribution of information in a population could lead to heterogeneous preferences; moreover, in some cultural domains (e.g. aesthetic culture), diverse preferences may be the norm rather than the exception. We present an agent-based model in which we can control the heterogeneity of preferences, as well as the effect of cultural traits on the evolution of preferences. We find that assuming homogeneous or heterogeneous preferences leads to different predictions on several outcomes. First, populations with greater heterogeneity of preferences converge toward greater cultural diversity. Second, while we replicate the classical result that increasing opportunities to learn socially leads to less diversity in homogeneous populations, we find that this relationship is reversed in heterogeneous populations. We show that this happens because increasing social learning opportunities leads the distribution of cultural traits to converge toward the distribution of preferences. We also look at the consequences of allowing cultural traits to modify the preferences of individuals that possess them. This can for example capture self-reinforcing beliefs, or traits where the acquisition costs make individuals less likely to switch to another trait after possessing them for some time. We find that such “attractive” cultural traits naturally emerge in our model, and that they tend to decrease cultural diversity when preferences are not homogeneous. Overall, by showing that the effect of different parameters on cultural diversity are dependent on the assumed distribution of preferences, we highlight the importance of taking into account the possible heterogeneity of preferences when making predictions about cultural dynamics