Portail des publications scientifiques IMT Mines Alès
Not a member yet
5198 research outputs found
Sort by
Level Finder – Prévision des crues par Intelligence artificielle (réseaux de neurones) pour la mise en œuvre de plans d’intervention graduée et la gestion de crise à l’échelle (inter)communale
No full textThis thesis aims to propose an approach and tools for developing flash flood forecast bulletins that are tailored to the needs of crisis managers and support decision-making in crisis situations. To achieve this goal, a characterization of the link between crisis management and hydrological forecasting is conducted through a scientific review. This review highlights the lack of knowledge of hydrological functioning among crisis management stakeholders, which prevents them from taking full advantage of flood forecasts. Conversely, hydrologists show little interest in the specific variables that can be applied directly to crisis management. They also do not sufficiently consider its organization at a local level. To bridge this gap, a workshop methodology was developed to deepen the understanding of how crisis managers interpret hydrometeorological forecast bulletins and to identify their needs. Neural networks were employed to forecast hydrological variables and to classify crisis management variables. The evaluation of the forecast quality focused on their relevance in crisis situations. The workshop helped define the desired and useful forecast bulletins for managers at various levels. Regarding hydrological forecasts, better quality is observed in level forecasts compared to discharge forecasts, both from a hydrological and crisis management perspective. The evaluation of forecast quality in relation to crisis management demonstrated that these forecasts are inseparable from the objectives of the managers. Finally, the feasibility of predicting crisis management variables through classification using artificial neural networks was demonstrated. The results of this work are promising, as they show that it is possible to apply simple, yet highly useful, models that are directly exploitable in crisis management. This is made possible by the properties of neural networks. This work also proposes an approach to defining the needs of managers. In the future, these models and approaches can be improved and applied to other regions. Research efforts to explore the links between hydrology and crisis management should continue due to the complexity of these links.L’objectif de ces travaux de thèse est de développer des prévisions de crues éclair, basées sur des modèles de réseaux de neurones, facilement interprétables par les gestionnaires de crise, et adaptées à leurs besoins pour une prise de décision éclairée et rapide en situation de crise. Pour atteindre cet objectif, une caractérisation du lien entre la gestion de crise et la prévision hydrologique est réalisée à travers une revue scientifique. Elle met en évidence la méconnaissance du fonctionnement hydrologique par les acteurs de la gestion de crise, ce qui les empêche de tirer pleinement profit des prévisions de crues. À l’inverse, les hydrologues s’intéressent très peu aux variables spécifiques pouvant s’appliquer directement à la gestion de crise et ne considèrent pas suffisamment l’organisation celle-ci au niveau local. Pour combler ce fossé, une méthodologie d’atelier est élaborée afin d’approfondir la compréhension qu’ont les gestionnaires du bulletin de prévision hydrométéorologique, en vue de définir leurs besoins. Les réseaux de neurones sont employés pour prévoir les variables hydrologiques et pour prévoir des variables de gestion de crise par classification. L’évaluation de la qualité des prévisions réalisées est focalisée sur leur pertinence en situation de crise. L’atelier a permis de définir les bulletins de prévision adaptés aux gestionnaires sur plusieurs niveaux. En termes de prévisions hydrologiques, une meilleure qualité de la prévision de la hauteur par rapport au débit est constatée. L’évaluation de la qualité des prévisions du point de vue de la gestion de crise a montré qu'elles sont indissociables des objectifs des gestionnaires. Enfin, la faisabilité de la prévision de variables de gestion de crise par classification effectuée par réseaux de neurones artificiels a été démontrée. Les résultats de ces travaux sont prometteurs car ils montrent qu’il est possible d’appliquer des modèles simples, mais très utiles, et directement exploitables en gestion de crise. Ceci est possible grâce aux propriétés des réseaux de neurones. Ces travaux proposent aussi une approche pour définir les besoins des gestionnaires qui pourront être améliorés et appliqués à d’autres contextes. Les efforts de recherche pour explorer les liens entre hydrologie et gestion de crise devront se poursuivre en raison de la complexité de ces liens
Choix de fournisseurs sous une demande dynamique et des délais de livraison stochastiques
No full textNational audienc
Développement d'une méthodologie d'évaluation environnementale, sociale et économique de pratiques généralisées de réutilisation des eaux usées traitées
No full textInternational audienceLe projet TERR’REUSE propose d’analyser les effets de la réallocation de l’eau par la REUSE au sein de deux territoires/bassins versants (BV) contrastés d’Occitanie (Atlantique et Méditerranée), incluant plusieurs Living Labs du projet WOC, à travers divers scénarios de REUSE et de changement global (volet prospective). On visera, en s’appuyant sur une équipe pluridisciplinaire, à analyser les impacts dans le temps et dans l’espace de ces réutilisations (hydrologiques, économiques, environnementaux, etc.) dans l’optique de mettre en exergue les bénéfices et/ou limites de la pratique (pertinence du choix de la REUSE vis-à-vis d’autres alternatives) et à produire des indicateurs utiles à la prise de décision. Un volet du projet visera ainsi à évaluer la réception politique et sociale de la REUSE et identifier les freins et des leviers à son acceptabilité, mais aussi à estimer l’émergence d’une valeur collective économique, sociale et environnementale de la REUSE favorable à son déploiement. Enfin, un dernier volet envisage des échanges/débats avec les acteurs du territoire afin de nourrir le projet et assurer la diffusion de ses résultats
Impaired perception of a partner's synchronizing behavior reduces positive attitude toward humanoid robot in schizophrenia patients
No full textInternational audienc
The Effects of Replacing Sand with Glass Fiber-Reinforced Polymer (GFRP) Waste on the Mechanical Properties of Cement Mortars
No full textInternational audienceThe aim of this study is to examine the effect of the partial replacement of sand by Glass Fiber-Reinforced Polymer (GFRP) waste on the mechanical properties of cement mortars. Compressive and flexural tests were carried out on mortars containing 0, 3, 5, 10, and 15% (by volume) of GFRP waste. It appears that the incorporation of 3% GFRP waste did not significantly affect the mechanical strength. However, further increasing the GFRP waste content led to a reduction in the mechanical strength. The flexural strength seemed less affected than the compressive strength, since the decrease in flexural strength at a 10% replacement was only 37%, while it was 54% for the compressive strength. However, an improvement in the toughness of the mortar with an increase in the substitution rate was observed. The reference sample displayed a flexural toughness of 0.351 N·m, while the mortar incorporating 15% of GFRP exhibited a flexural toughness of 0.642 N·m. The reuse of GFRP waste in cementitious materials, therefore, constitutes an interesting recycling solution
An Application for Predicting Phishing Attacks: A case of Implementing a Support Vector Machine Learning Model
No full textInternational audienceThe imminent threat that phishing websites poses is a major concern for internet users worldwide. These fraudulent websites are crafted by cyber attackers to appear trustworthy and deceive vulnerable users into divulging confidential data like medical health records, credit card details, passwords, and Personal Identifiable information (PII). To bait their victims, cybercriminals employ tactics such as social engineering, spear-phishing attacks, and email phishing scams. As a result, unsuspecting individuals may be enticed to visit these websites, putting their sensitive information at risk. This work presents an application designed to predict phishing attacks after comparing polynomial and radial basis function of support vector machine (SVM). The proposed application leverages a dataset of known legitimate, suspicious and phishing attacks stored in a database and employs an SVM algorithm for classification based on user input. The application provides a user-friendly graphical user interface (GUI) that allows reporting of new phishing incidents based on the features that have strong relationship in determining if a website is phishing or not. The proposed application utilizes the inherent scalability of database technology to support record expansion whenever there is an instance of a user initiating phishing prediction thereby, making it suitable for use in a wide range of organizational settings
Agir en écologue de l'action
No full textInternational audienceIn the wake of the United Nations' Sustainable Development Goals (SDG) and the reports of the Intergovernmental Panel on Climate Change (IPCC), the pressure is increasing to respond to the urgent need to combat global warming and preserve biodiversity. Drawing on complex thinking (Morin, 1977-2004) and an interdisciplinary approach (epistemology, philosophy, sociology, psychology, mathematical, etc.), this review will attempt to understand what is the ecological of action (Morin, 1980) in all its complexity, and will address the conditions, obstacles and changes of an epistemological, ontological and educational nature that this action requires. It will not to deal with the question of proposing concrete actions as of reflecting on the transformations in thinking that these actions require. The concepts of complexity, ecology of action, eco-ontology, resonance and capability, all of which contribute to understanding the world in all its complexity, will guide us as we sketch out what an ecologist of action might be.Dans le sillage des Objectifs de Développement Durable (ODD) de l'ONU, et des rapports du Groupe d'experts Intergouvernemental sur l'Evolution du Climat (GIEC), les incitations à répondre à l'urgence de la lutte contre le réchauffement climatique et à la préservation de la biodiversité, se font de plus en plus pressantes. A partir de la pensée complexe (Morin, 1977-2004) mais aussi d'un traitement se voulant interdisciplinaire (épistémologie, philosophie, sociologie, psychologie, mathématiques ...), cette revue de question tentera d'appréhender ce qu'est l'écologie de l'action (Morin, 1980) dans toute sa complexité et abordera les conditions, les obstacles et les changements d'ordre épistémologique, ontologique et éducationnel que nécessite cette action. Il ne sera pas tant question de proposer des actions concrètes que de réfléchir aux transformations de la pensée que nécessitent ces actions. Ainsi les concepts de complexité, d'écologie de l'action, d'écoontologie, de résonance et de capabilité, tous concourant à appréhender le monde dans toute sa complexité, nous guiderons pour esquisser ce que pourrait être un écologue de l'action
Negative emotions disrupt intentional synchronization during group sensorimotor interaction
No full textInternational audienceEmotions play a fundamental role in human interactions and trigger responses in physiological, psychological,and behavioral modalities. Interpersonal coordination often entails attunement between individuals in variousmodalities. Previous research has elucidated the mechanisms of interpersonal synchronization and the emo-tions aroused by joint action: cardiac activity aligns in disputing marital couples, spectators share enjoymentin observing live dance performances, and joint finger-tapping evokes positive emotions. However, little isknown about the impact of emotions on intentional interpersonal synchronization. To address this problem,we conducted an experiment in 2022 asking 60 participants to engage in a three-way finger-tapping synchro-nization task. We systematically induced emotional states (positive, neutral, and negative) with social compar-ison feedback using success–failure manipulations. An analysis of behavior synchronization using theKuramoto order parameter revealed that negative emotion induction significantly diminished time spent in syn-chrony compared to positive induction. Moreover, the results exposed incremental struggles in attaining higherlevels of synchronization (Q2–Q3) after the induction of negative emotions. These outcomes further substan-tiate the necessity of integrating the indices of agents’ emotions into interpersonal synchronization and coor-dination models. We discuss the implications of this work for research on interpersonal emotion in joint actionand applied outcomes in emotion-aware technologies and intervention
Flame-retardant performance of phosphorylated furan-containing alkyd resins
No full textInternational audienceOver the past decade, with the emergence of environmental and human health concerns, research has focused on developing new solutions to replace fossil resources with more sustainable biobased resources. Furan derivatives can be obtained from plants and their structures provide high resistance to high temperatures. Therefore, these furan derivatives have been functionalized with phosphorus to increase their fire resistance properties. Thus, a new biobased bisfuran containing nitrogen and phosphorus (HMF-MXDA-DEP) was synthesized and used for the first time in polymerization to develop alkyd resins. The flame retardant properties of these resins have been studied. These alkyd resins are generally used as a binder in paint or varnish formulations. The use of a reactive flame retardant during the polycondensation reaction of alkyd resins makes it possible to form covalent bonds and offer better durability over time. The impact of HMF-MXDA-DEP on film properties such as drying time, flexibility, adhesion and color was studied and the thermal and flame retardant properties by differential scanning calorimeter, thermogravimetric analysis and calorimeter pyrolysis-combustion fluxes were evaluated. Increased thermal stability and good flame retardant (FR) properties of alkyd resins have been demonstrated. In the cone calorimeter test, the lowest peak in heat release rate (pHRR) was obtained with a 2 wt % coating of P HMF-MXDA-DEP on wood and a 49 % decrease in pHRR compared to FR-free alkyd resins has been demonstrated
A new methodology for the mechanical behavior analysis of polyvinyl chloride and recycled polyethylene terephthalate foams: Finite element model updating using mechanical field measurements
No full textInternational audienceThis paper aims to characterize the intrinsic mechanical behavior of an eco-friendly foam core used in sandwich materials dedicated to the maritime sector. A thorough understanding of intrinsic properties is essential for accurate dimensional structure characterization. The model setup is based on the choice of materials that meet the criteria of sustainable development, mechanical performance, and durability to promote ecological practices in the nautical industry. Experimental tests, instrumented by a digital image correlation technique, were carried out on two types of foams: polyvinyl chloride (PVC), which is the reference foam traditionally used in marine applications, and recycled polyethylene terephthalate (rPET), the substitute foam. The sample sizes recommended by compression test standards often lead to complex stress states that do not allow direct determination of the foam's intrinsic mechanical properties. Therefore, a finite element model updating method was required to identify the anisotropic elastic behavior of these two types of foams