HAL Portal UBS (Université Bretagne Sud)
Not a member yet
14201 research outputs found
Sort by
La carte, outil pour l'accès, le contrôle et la gestion des ressources naturelles des îles Kerguelen des années 1860 aux années 1920
No full textInternational audienceL’analyse des commanditaires et des usages des cartes de l’archipel des Kerguelen produites entre les années 1860, règne des baleiniers américains, et les années 1920, dernière décennie de l’exploitation baleinière par des sous-concessionnaires étrangers, permet d’interroger le statut de la ressource (res nullius ou sous concession), ses enjeux (scientifiques, économiques, stratégiques) et les limites du pouvoir de ceux qui y ont accès, l’exploitent, la gèrent. L’article interroge ainsi la place des cartes comme outil de pouvoir pour l’accès, le contrôle et la gestion des ressources naturelles insulaires
New First Order Necessary Conditions for Free End-Time Problems with Measurably Dependent Data
No full textInternational audienceFirst order necessary conditions of optimality for free end-time optimal control problems with measurably time-dependent data include boundary conditions on the maximized Hamiltonian function evaluated along the optimal state trajectory and costate trajectory. Since pointwise evaluation of an a.e. defined, measurable function is meaningless, boundary values of the maximized Hamiltonian must be appropriately interpreted. In earlier papers, this difficulty was overcome by expressing the necessary conditions in terms of elements in the essential values of the maximized Hamiltonian. Necessary conditions involving essential values have been derived for dynamic optimization problems with dynamic constraints, either controlled differential equations or differential inclusions. They have been extended to cover problems with pathwise state constraints. This paper shows that, throughout this literature, available first order necessary conditions can be improved by replacing the essential value of the maximized Hamiltonian by two, more refined, constructs, namely the sub- and superessential values of the maximized Hamiltonian. We give an example of an optimal control problem in which the new conditions, involving sub- and superessential values, can be used to show that a certain admissible process is not a minimizer, while the former conditions, involving essential values, fail to do so. The search for the appropriate definition of boundary value of the maximized Hamiltonian is intimately connected with the representation and estimation of limiting sub- and superdifferentials of indefinite integral functions; the paper includes a section on this topic. The improvements to available first order necessary conditions provided by super- and subessential values is connected with the fact that these sets capture precisely the differential properties of indefinite integral values, whereas essential values only provide estimates, which may not be exact
La folie du pouvoir : Hérodote et l’invention d’un motif historiographique
No full textInternational audienc
Whole Cells of Microorganisms—A Powerful Bioanalytical Tool for Measuring Integral Parameters of Pollution: A Review
No full textInternational audienceMicrobial biosensors are bioanalytical devices that can measure the toxicity of pollutants or detect specific substances. This is the greatest advantage of microbial biosensors which use whole cells of microorganisms as powerful tools for measuring integral parameters of environmental pollution. This review explores the core principles of microbial biosensors including biofuel devices, emphasizing their capacity to evaluate biochemical oxygen demand (BOD), toxicity, heavy metals, surfactants, phenols, pesticides, inorganic pollutants, and microbiological contamination. However, practical challenges, such as sensitivity to environmental factors like pH, salinity, and the presence of competing substances, continue to hinder their broader application and long-term stability. The performance of these biosensors is closely tied to both technological advancement and the scientific understanding of biological systems, which influence data interpretation and device optimization. The review further examines cutting-edge developments, including the integration of electroactive biofilms with nanomaterials, molecular biology techniques, and artificial intelligence, all of which significantly enhance biosensor functionality and analytical accuracy. Commercial implementations and improvement strategies are also discussed, providing a comprehensive overview of the state-of-the-art in this field. Overall, this work consolidates recent progress and identifies both the potential and limitations of microbial biosensors, offering valuable insights into their future development for environmental monitoring
Bilan 2020-24 de la campagne de suivi des Espèces Exotiques marines dans le cadre du Réseau Alien Grand Ouest
No full textDataset of hygrothermal and energy measurements for a raw compressed earth brick house of Sense-City equipment during different seasons
No full textInternational audienceA non-insulated raw compressed earth brick test house was built and deeply instrumented in Sense-City equipment at Champs sur Marne, France. A wide variety of sensors were deployed in the room, inside the walls and outside to monitor the hygrothermal behaviour, the thermal comfort and the heating energy consumption. The measurement campaigns were performed over several weeks during winter, spring and summer seasons in 2024. The test house was unoccupied and exposed to natural weather conditions. During winter tests, different heating scenarios were considered. The provided dataset of sensor outputs can be useful for a better understanding of earthen construction and for the experimental validation of building physics models at both the wall and building scales
First-order two-scale simulations for charged particle dynamics in inhomogeneous strong magnetic fields
No full textWe develop a multi-scale numerical approach to solve the Newton-Lorentz equations for the dynamics of a charged particle in inhomogeneous in space and time electromagnetic fields with a strong magnetic component. Based on the general setting of [11], we consider zero-order and first-order two-scale models derived from asymptotic expansions, to approximate the initial model. Then, we implement a two-scale numerical method as follows: first, with the help of symbolic software we obtain explicit forms of the right-hand side terms in the two-scale equations, and second, we classically use a Runge-Kutta integrator to solve the equations. The numerical experiments demonstrate the relevance of the first-order two-scale solution for various simulations of multi-scale plasma. Indeed, the first-order solution proves to be highly accurate when approximating the original model and computationally efficient thanks to its large time step compared to the cyclotron period
Jean-François Bert, Le corps qui pense: Une anthropologie historique des pratiques savantes Basel: Schwabe Verlag, 2023. Pp. 176. ISBN 978-3-7965-4873-4.
No full textInternational audienc
Probabilistic models for permutations and dependence
No full textInternational audienceIn order to improve the financial performance of a company, we must classify all their commercialized products according to their interests using financial criteria. The criteria and the final classification can be modeled using permutations. We consider Mallows' models defined in the space of permutations. We are particularly interested in the question of dependence in Mallows' models. Here, we introduce new machine learning approaches based on a cost function minimizing the impact of the dependence between criteria. In the multi criteria aggregation model, we consider some of the criteria as a permutation generated using a Mallows' model, in which the modal permutation is the permutation we want to find, while some other criteria are simulated using other criteria. Finally, the methodology is illustrated with a simulation study that compares the performances of the approaches
On the speed of convergence of the pressure function at zero temperature
No full textWe prove here that the pressure function cannot converge to the limit entropy at zero temperature faster than some exponential rate. Furthermore, we characterize this limit rate via an expression involving the Peierls barriers between the irreducible components of the Aubry set. This extends and completes results from [8] and [7].In the first one, an exact exponential speed of convergence was proved, under the assumption that the Aubry set is a subshift of finite type. In the later one, a rate was given but without interpretation in term of Thermodynamical quantities of the system.</div