HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Degradation of pesticide atrazine in water by high voltage electrical discharges
No full textInternational audienceIn recent years, the widespread use of pesticides hasincreased agricultural output, but it has also caused many problems.For instance, atrazine, the most commonly used herbicide, has beenreported to not only pollute the water environment, but also affect thereproductive system of humans and other animals as an endocrinedisrupting chemical (EDC). It is thus urgent to find an efficient wayto degrade atrazine. Here, we investigated the degradation of atrazineby high voltage electrical discharges and compared it with traditionalFenton oxidation and ultrasound treatment
Notebooks et science ouverte : FAIR mieux
No full textLes notebooks sont aujourd'hui largement adoptés dans les pratiques numériques de recherche. Malgré leur omniprésence croissante, leurs caractéristiques, les rôles et usages associés aux notebooks ont pour le moment donné lieu à peu d'investigations dans une perspective d'études des sciences et des techniques (STS). Dans cet article, nous proposons une synthèse de travaux empiriques menés sur les notebooks afin d'identifier les principaux résultats existants, que cela concerne la classification de types de notebooks, les pratiques installées, les limites et améliorations proposées. Dans la continuité de cette synthèse qui souligne surtout l'existence de travaux dans le domaine de la science des données (data science) et non pas des pratiques de recherche en contextes académiques, nous discutons le rôle des notebooks comme vecteur et levier des principes FAIR (Findable, Accessible, Interoperable, Reusable) associés à la science ouverte
Application de la technologie des fluides supercritiques pour la formulation de dispersions solides amorphes de quercétine avec du chitosan
No full textInternational audienc
Modelling dense powder flow with CFD inside a rotating drum and a screw reactor
No full textInternational audienc
Primal–dual on-the-fly reduced-order modeling for large-scale transient dynamic topology optimization
No full textInternational audienceManaging large-scale topology optimization under dynamic loading poses significant computational and storage challenges as against static loading. In order to fill this gap in computational cost, this paper proposes a reduced order modeling strategy that involves constructing discrete basis functions (modes) in adaptive fashion for both the primal as well as dual problem in topology optimization of transient dynamic systems. The projection bases are enriched based on the residual threshold of the corresponding systems. We address the computational cost and scalability of the ROM learning and updating phases. The approach is validated using 2D and 3D benchmark problems, by comparing alternative reduced-order-sensitivity formulations and projection basis update schemes
Objets numériques, objets sociaux. Dix ans d'enquête en sciences sociales sur internet
No full textTo celebrate its tenth anniversary, the journal RESET is publishing a special issue entitled “Digital objects, social objects”, providing an opportunity to look back at the journal's history, its publication process and a (provisional) assessment of its scientific output.Pour célébrer ses dix ans d’existence, la revue RESET publie un numéro spécial « Objets numériques, objets sociaux », l’occasion de revenir sur l’histoire de cette revue, son processus de publication, et de tirer un bilan (provisoire) de sa production scientifique
Formal stabilization of a coupled ODE-PDE switched system
No full textInternational audiencePartial Differential Equations (PDEs) are a ubiquitous model that describes a wide range of dynamical systems. While stabilization of systems involving PDEs is an important problem, little work has been done on formal verification and synthesis for such systems. In this paper, we explain how a coupled ODE-PDE control problem can be formally stabilized using a tiling based control synthesis algorithm associated to set based reachability, which is usually used on finite dimensional problems. To formally prove the stabilization of the PDE, the original infinite dimensional problem is transformed into a finite dimensional one using, among other tools, model order reduction. The strength of our approach relies on the fact that we never explicitly discretize the PDE state using e.g. a finite element approximation, and consequently, we provide stability guarantees directly on the infinite dimensional state.</div
How to deal with unsuspected source of annoyance in electric vehicle cabin using the "sonic augmented" concept?
No full textInternational audienceThe disappearance of the main source of noise with electric vehicles resulted in a breakthrough in car cabin sound experience. Despite significantly lowering the sound level pressure, one cannot claim perfect quietness is achieved. Indeed, as the auditory masking produced by the internal combustion engine is not effective anymore, unsuspected sounds such as electric whistling, whining or switching are now part of this sound environment and may be identified as unpleasant sounds. However, drivers expectations are to have audible feedback indicative of vehicle operation that matches their expectations. Therefore, the presence of unwanted and unpleasant sounds, on the one hand, and the absence of desired and informative sounds, on the other hand, is an opportunity for sound design to create a new sound environment inside car cabins. The present work proposes a "sonic augmented" concept based on the idea of adding designed sounds, called "masks", participating both in the overall soundscape augmentation and in the attenuation of unpleasant sounds, called "defects".</div
Structural Health Monitoring (SHM) of Polymer-Matrix Composites (PMC) with the Insertion of Piezoelectric Transducers and a Data Fusion Approach
No full textInternational audienceIn this article, the results obtained from an experimental campaign, performed under self-heating fourpoint bending tests, on smart Polymer-Matrix Composite (PMC) are presented. The smart capabilities of these PMC specimens came from the insertion of in-situ piezoelectric transducers (PZT) within the material. These transducers were used both for evaluating the rapid fatigue limit and to monitor the health of the structure, known as Structural Health Monitoring (SHM). As the self-heating test can introduce damage in the material, it was possible to follow its evaluation through the Non-Destructive Testing (NDT) (Ultrasonic testing, Acoustic Emission, Capacitance measurement) methods used for SHM. In this study, the fatigue limit was evaluated from the capacitance measurements of the embedded transducers, and validated from the stabilized temperature measurements. So, this paper showed an original way of evaluating the fatigue limit and a multi-acquisition procedure for SHM, from in-situ piezoelectric transducers
Architecture de prise de décision/commande sûre et économe en énergie pour une formation de systèmes multi-véhicules évoluant dans un environnement dynamique
No full textThe widespread use of Intelligent Transportation Systems (ITS) primarily stems from the imperative to mitigate human errors contributing to accidents. However, addressing the “individual” Automated Vehicles (AVs) alone remains insufficient, given that a number of situations require the “coordination” of the relative movements of AVs. Within the Multi-Vehicle System (MVS) paradigm, AVs derive advantages from connectivity and road preview information. Consequently, they can sense more accurately, process more information, and can be more tightly controlled. The collaborative assessment of safety within MVS allows for the establishment of advanced collision-avoidance strategies, particularly in challenging scenarios such as intersection crossings and onramp mergings. Moreover, MVS technology facilitates the reduction of gaps between vehicles, enhancing road capacity and traffic flow. The MVS’s shorter response time enables improved control of AV dynamics, paving the way for promising energy-centric strategies. The main aim of the research done in this Ph.D. manuscript is to propose a safe and energy efficient decision/control architecture for MVS that navigates in dynamic and complex environment such as on-ramp merging and multi-road navigation in highway. Based on the multi-controller foundational control architecture a Cooperative Multi-Controller Architecture (C-MCA) is proposed. The first part of the proposed C-MCA deals with the decision-making level. This level involves a multi-behavior decision-making strategy, responsible for activating the MVS’ suitable behavior based on the safety metric. Two distinct behaviors are proposed: (a) The nominal behavior, designed for executing the merging scenario while adhering to the individual goals of the MVS vehicles. It is activated when no collision risk is detected, (b) The cooperative behavior, activated by the decision-making level when both of the safety and the energy efficiency requirement are not satisfied by the nominal behavior. Its objective is to solve the merging conflict by generating a safe and energy efficient passing order of the MVS vehicles in the merging zone. The second part of the proposed C-MCA focuses on the local trajectory planning level. Each behavior is assigned a dedicated controller tailored to meet its specific requirements. For instance, the cooperative behavior controller has the responsibility of translating the vehicle’s passing order into feasible dynamics (i.e., trajectory, velocity, etc.). The translation task is facilitated through the proposed dynamic formation reconfiguration strategy. In essence, the approach leverages the multi-vehicle system’s formation navigation capabilities to conceptualize the merging challenge as a formation reconfiguration problem. A formalization of the formation reconfiguration problem is presented, employing a flexible and generic formal approach. The proposed dynamic formation reconfiguration strategy, employs virtual dynamic targets to ensure a secure reshaping of the formation toward the desired configuration based on the selected passing order. The overall architecture’s performance is assessed through co-simulation using Matlab/Simulink and SCANeR studio.L’engouement important pour les systèmes de transport intelligent est justifié principalement par l’impératif de réduire, voire annihiler, les erreurs humaines induisant les accidents. Cependant, s’attaquer uniquement aux Véhicules Autonomes (VAs) “individuels” demeure insuffisant, étant donné que plusieurs situations nécessitent la “coordination” des mouvements relatifs des VAs. Dans le paradigme du Système Multi-Véhicules (SMV), les VAs bénéficient de l’information issue de leur connectivité. Par conséquent, ils peuvent détecter plus précisément, traiter davantage d’informations et être contrôlés de manière plus précise. L’évaluation collaborative de la sécurité au sein du SMV permet l’établissement de stratégies avancées en matière d’évitement de collision, en particulier dans des scénarios complexes tels que les croisements au sein d’intersections et les insertions sur les entrées d’autoroute. De plus, la technologie SMV facilite la réduction des espaces entre les véhicules, améliorant ainsi la capacité et la fluidité du trafic routier. Le temps de réponse plus court du SMV permet un meilleur contrôle de la dynamique des VAs, ouvrant la voie à des stratégies énergétiques prometteuses. L’objectif principal des travaux de recherche constituant ce manuscrit de doctorat est de proposer une architecture de décision/contrôle sûre et peu énergivore pour le SMV naviguant dans des environnements dynamiques et complexes. Inspirée des architectures multi-contrôleurs, une Architecture Multi-Contrôleurs Coopérative est proposée. La première partie de l’architecture proposée concerne le niveau de prise de décision. Ce niveau, impliquant une stratégie de prise de décision à plusieurs comportements, est responsable de l’activation du comportement approprié du SMV en fonction de la métrique de sécurité. Deux comportements distincts sont proposés : (a) Le comportement nominal, conçu pour réaliser le scénario d’insertion tout en respectant les objectifs individuels des véhicules formant le SMV, est activé lorsqu’aucun risque de collision est détecté, (b) Le comportement coopératif est activé par le niveau de prise de décision lorsque l’exigence de sécurité n’est pas satisfaite par le comportement nominal. Son objectif est de résoudre le conflit lors de l’insertion en générant un ordre de passage sûr et économe énergétiquement pour les véhicules composant le SMV dans la zone d’insertion. La deuxième partie de l’architecture proposée se concentre sur le niveau de planification de trajectoire locale. Chaque comportement se voit attribuer un contrôleur dédié, conçu pour répondre à ses besoins spécifiques. Par exemple, le contrôleur du comportement coopératif est chargé de traduire l’ordre de passage des véhicules en dynamiques réalisables (trajectoire, vitesse, etc.). Cette tâche d’obtention de dynamiques réalisables est facilitée par la stratégie de reconfiguration dynamique de la formation proposée. En substance, l’approche tire parti des capacités de navigation en formation du SMV pour conceptualiser des manœuvres coopératives ayant trait d’insertion en milieu autoroutier, comme un problème de reconfiguration de formation. Une formalisation du problème de reconfiguration de la formation est présentée, utilisant une approche formelle, flexible et générique. La stratégie proposée utilise des cibles dynamiques virtuelles pour garantir une reconfiguration sécurisée de la formation, ceci, de sa forme initiale vers la configuration souhaitée par rapport à l’ordre de passage sélectionné. La performance de l’architecture globale a été évaluée par co-simulation en utilisant Matlab/Simulink et SCANeR studio