Dissertations / Theses on the topic 'Contrôle basé sur des règles'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 49 dissertations / theses for your research on the topic 'Contrôle basé sur des règles.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Cardot, Anaïs. "Rejeu basé sur des règles de transformation de graphes." Thesis, Poitiers, 2019. http://www.theses.fr/2019POIT2254.
Full textIn many modelling fields, such as architecture, archaeology or CAD, performing many variations of the same model is an expanding need. But building all those variations manually takes time. It is therefore needed to use automatic technics to revaluate some parts of a model, or even an entire model, after the user specifies the modifications. Most of the existing approaches dedicated to revaluating models are based on a system called parametric modelling. It is made of two parts, a geometric model and a parametric specification, which allows to record the series of operation that created the model, and the different parameters of those operations. This way, the user can change some parameters, or edit the list of operations to modify the model. To do so, we use a system called persistent naming, introduced during the 90ies, that allows us to identify and match the entities of an initial specification and the ones of a revaluated specification. In this thesis, our goal is to propose a persistent naming system that would be general, homogeneous and that would allow the user to edit a parametric specification (which means move, add, or delete some operations). We base our system on the Jerboa library, which uses graph transformation rules. This way, we will be able to use those rules to create our naming system, while also linking the notions of graph transformation rules and parametric specification. We will then describe how to use our naming method to revaluate or edit parametric specifications. Finally, we will compare our method with the other ones from the literature
Lavoie, Marc-André. "Développement et contrôle d'un bras robotique basé sur l'actionneur différentiel élastique." Mémoire, Université de Sherbrooke, 2009. http://savoirs.usherbrooke.ca/handle/11143/1494.
Full textOuedraogo, Sarah. "Développement de Stratégies Optimisées de Gestion de l’Energie Intermittente dans un Micro Réseau Photovoltaïque avec Stockage." Electronic Thesis or Diss., Corte, 2023. http://www.theses.fr/2023CORT0008.
Full textMicrogrids are considered as the future of energy production and distribution in electrical grid. Many of them incorporate photovoltaic generation and storage, mostly in the form of batteries, to power various loads. The main objective of this thesis is to propose energy management strategies designed to optimize the operating costs of a photovoltaic microgrid with battery while respecting specific constraints. This microgrid powers residential buildings and electric vehicles.To achieve this, five energy management strategies based on rules, with increasing complexity, were developed. These strategies were compared to an optimization using linear programming in terms of energy and economic performance. The results indicate that the most optimal strategy achieved a performance level close to the linear programming, which is considered "optimal." However, some limitations were observed for the initial strategies, including power cuts, which are not acceptable. To improve these strategies, the seasonal effect, particularly in photovoltaic production, was taken into account, eliminating power cuts. Depending on the chosen strategy, the batteries are more or less stressed, so it was necessary to consider the varying battery aging and its impact on performance. Suitable battery aging models were thus implemented. The results showed that the profitability of batteries depends on their installation cost and they remain economically viable for costs below approximately 175 €/kWh. The most effective rule-based control strategy considers variations in electricity costs, photovoltaic production forecasting, seasonal variation in PV production, and battery degradation in its decision-making process. This strategy improves financial gain by approximately 68 % compared to the simplest rule-based strategy, which is similar to a self-consumption strategy.An analysis of the influence of different parameters, such as electricity purchase tariffs, battery capacity, power exchanged with the main grid and consumption profiles was conducted through simulations. It was found that the electricity pricing model has a significant effect on energy distribution and financial gain. The influence of battery size, limitation of power exchange with the main grid, and consumption profile strongly depends on the strategy used, as well as the electricity pricing model.This work highlights the importance of integrating the characteristics of photovoltaic energy into energy management strategies through the use of various tools such as photovoltaic production forecasting. This information is valuable for investment and operational decision-making
Ghadi, Abderrahim. "Modèle hiérarchique de contrôle d'accès d'UNIX basé sur un graphe de rôles." Strasbourg, 2010. http://www.theses.fr/2010STRA6005.
Full textConcerning access control, can the following question be addressed : "Is the access control system decidable ?". In other words : is it true that starting from a safe state of protection, we can assume at any time that is no intrusion which will endanger our system ?. In order to answer this question, we propose to model the access control system in the form of a graph of roles. The roles, which represent the vertices of graph contain, according to the security-policy, certain number of privileges. Every privilege represents one or several access rights on a given object. We presented two methods of use of this graph : The first consists in using an algorithm, which we developed by basing itself on the algorithms of the theory of the graphs, permit to search all over the path of the graph in order to find illicit privilege transfer. The second consists in storing our graph in a directory LDAP, this which brings us to develop a new plan LDAP to represent our graph of roles
Knotek, Michal. "Diagnostic de fautes basé sur l'analyse temporelle." Phd thesis, Université Joseph Fourier (Grenoble), 2006. http://tel.archives-ouvertes.fr/tel-00110429.
Full textun rôle primordial dans l'amélioration de la disponibilité opérationnel le
des équipements. Dans les systèmes industriels, une part importante
(jusqu'à 80%) est consacrée à la maintenance, test et diagnostic. Pour
des systèmes complexes, la résolution des problèmes liés au diagnostic
et d'une manière générale de la supervision nécessite la mise en oeu-
vre d'une approche générique. Le diagnostic concerne les deux phases
indissociables de détection et de localisation. Dans cette thèse nous
proposons une approche dynamique de diagnostic pour les systèmes à
événements discrets. L'approche proposée basée sur l'exploitation du
temps, est applicable à tout système dont l'évolution dynamique dépend
non seulement de l'ordre des événements discrets mais aussi de la durée
des tâches associées comme pour les processus de communication ou les
processus batch. Dans cette thèse, le diagnostic des fautes est réalisé
grâce à l'implémentation d'un modèle basé sur l'utilisation des auto-
mates temporisés. L'objectif est de concevoir un observateur pour un
système donné, qui permet de détecter et localiser les éventuel les dé-
fail lances du procédé. Cet observateur est appelé “diagnoser”. Une
défail lance est constatée lorsque le séquencement temporel en sortie est
incorrect. Nous présentons donc les différentes étapes de la démarche
de diagnostic : la construction du diagnoser, la vérification du modèle
ainsi qu'une l'application de la démarche sur un exemple réel avec son
extension aux systèmes hybrides.
Maurice, François. "Un modèle d'évaluation et d'amélioration d'entités logicielles basé sur l'utilisation de métriques." Toulouse 3, 1996. http://www.theses.fr/1996TOU30192.
Full textKomarova, Maryna. "Authentification rapide et contrôle d'accès basé sur la confiance dans les réseaux sans fil hétérogène." Phd thesis, Télécom ParisTech, 2008. http://pastel.archives-ouvertes.fr/pastel-00003793.
Full textKomarova, Maryna. "Authentification rapide et contrôle d'accès basé sur la confiance dans les réseaux sans fil hétérogènes." Paris, ENST, 2008. http://pastel.archives-ouvertes.fr/pastel-00003793/fr/.
Full textIn this thesis, we study the current situation and trends in wireless technologies development. We discuss the problems related to security mechanisms specific to each technology, and in particular the possibilities for integration and interworking. We first study the possibility of authentication latency decreasing in a scenario where the network access authentication is decoupled from the service access authentication. An authorized user is granted network and service access as a result of a single authentication process that combines 802. 1X and PANA operations. Then we introduce the Fast re-Authentication Protocol (FAP) for inter-domain roaming, which aims to reduce the authentication delay for a mobile user in a visited administrative domain. The approach eliminates the need for communication between the target and the user's home networks for credentials verification. We develop the Fast re-Authentication Protocol by suggesting a ticket distribution scheme for inter-domain roaming. This method decreases the number of tickets sent and consequently the overhead and delay of the ticket acquisition phase of the protocol. Numerical results obtained from experiments on a test-bed and a series of simulations show that the proposed scheme enhances inter-domain handover parameters such as authentication latency and signalling cost. To improve the access control to network resources we propose the adjustable trust model. The purpose of this work is to provide the network with the opportunity to react to user behaviour. Users are motivated to gain higher trust because trusted users have access to a larger set of services with higher quality of service. Validation of the proposed trust-based access control method has been done via simulations. Finally, we discuss how the proposed solutions can be implemented in a single framework
Moussa, Kaouther. "Estimation de domaines d'attraction et contrôle basé sur l'optimisation : application à des modèles de croissance tumorale." Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALT078.
Full textThe main objective of this thesis is to propose frameworks and algorithms that are based on advanced control approaches, in order to guide cancer treatments scheduling. It also aims at pointing out the importance of taking into account the problem of stochastic uncertainties handling in the drug scheduling design, since cancer dynamical systems are considered to be highly uncertain phenomena.Cancer dynamical interactions are still an open research topic which is not fully understood yet. The complexity of such dynamics comes from their partially unknown behavior and their uncertain nature. Additionally, they are often described by nonlinear complex dynamics and require taking into consideration many constraints related to physiology as well as biology.In terms of control design, this topic gathers many complexity ingredients such asnonlinear dynamics, constraints handling and optimality issues. Therefore, in this thesis, we propose to use a recent optimal control approach that is based on moment optimization. This framework has the advantage of considering all the state and input variables as probability densities, allowing therefore to explicitly consider parametric as well as initial state uncertainties in the optimal control problem. We use this framework in Part II, in order to design robust optimal control schedules that represent cancer drugs injection profiles.The second problem that we address in Part III consists in the estimation of regionsof attraction for cancer interactions models. This problem is interesting in the context of cancer treatment design, since it provides the set of all possible initial conditions (tumor and patient health indicators), that can be driven to a desired targeted safe region, where the patient is considered to be healed. Furthermore, we focus on the assessment of methodologies that take into consideration the parametric uncertainties that can affect the dynamical model
Voiry, Matthieu. "Étude et conception d'un système automatisé de contrôle d'aspect des pièces optiques basé sur des techniques connexionnistes." Thesis, Paris Est, 2008. http://www.theses.fr/2008PEST0077/document.
Full textIn various industrial fields, the problem of diagnosis is of great interest. For example, the check of surface imperfections on an optical device is necessary to guarantee its operational performances. The conventional control method, based on human expert visual inspection, suffers from limitations, which become critical for some high-performances components. In this context, this thesis deals with the design of an automatic system, able to carry out the diagnosis of appearance flaws. To fulfil the time constraints, the suggested solution uses two sensors working on different scales. We present one of them based on Normarski microscopy, and the image processing methods which allow, starting from issued data, to detect the defects and to determine roughness in a reliable way. The development of an operational prototype, able to check small optical components, validates the proposed techniques. The final diagnosis also requires a classification phase. Indeed, if the permanent defects are detected, many “false” defects (dust, cleaning marks. . . ) are emphasized as well. This complex problem is solved by a MLP Artificial Neural Network using an invariant description of the defects. This representation, resulting from the Fourier-Mellin transform, is a high dimensional vector, what implies some problems linked to the “curse of dimensionality”. In order to limit these harmful effects, various dimensionality reduction techniques (Self Organizing Map, Curvilinear Component Analysis and Curvilinear Distance Analysis) are investigated. On one hand we show that CCA and CDA are more powerful than SOM in terms of projection quality. On the other hand, these methods allow using more simple classifiers with equal performances. Finally, a modular neural network, which exploits local models, is developed. We proposed a new classification problems decomposition scheme, based on the intrinsic dimension concept. The obtained data clusters of homogeneous dimensionality have a physical meaning and permit to reduce significantly the training phase of the classifier, while improving its generalization performances
Fosse, Pascal. "Une expérience sur le raisonnement en langage naturel : une étude de l'effet de contraintes syntaxico-sémantiques sur un mécanisme bi-directionnel basé sur des règles de production contextuelles." Paris 11, 1985. http://www.theses.fr/1985PA112334.
Full textThis thesis presents a natural-language question-answering system. We attempt to establish a link between a question and a knowledge base: taken from an encyclopedia. This link is made of a sequence of elementary steps of inference, each of which is performed by means of a "contextual production rule", acting on surface-level texts. The reasoning module basically operates as an inference engine. We explain here how we improved a previous version of this module, and particularly describe the implementation of two crucial points: first, an efficient pattern¬matching algorithm, especially designed for natural language strings, and second, the evaluation of syntactic and semantic functions. The efficiency of this new version is evaluated through a series of experiments. These experiments allow us: - to justify our choices corresponding to local strategies - to estimate the limits of the method - to measure the improvement we obtained, compared with the level of performance of the version we started with
Tik, Sovanna. "Contrôle d'un système intégré d'assainissement urbain basé sur la qualité de l'eau vers des stratégies tolérantes aux fautes." Doctoral thesis, Université Laval, 2020. http://hdl.handle.net/20.500.11794/40351.
Full textRegulation concerning urban wastewater management is evolving towards a holistic approach which leads stakeholders to increasingly consider the integrated urban wastewater system(IUWS), including the sewer system and the water resource recovery facility (WRRF), as a whole. This change of paradigm opens doors to new collaborations between the traditionally separated fields of urban drainage and wastewater treatment operation and research. In this context of innovation and integration, modelling is going to be a key tool, granting a better understanding of the complexity of interactions between the different parts of the IUWS. At the same time, the IUWS has to become more resilient to face the impacts of climate change, the long-term effects of which are difficult to quantify. Still, climate change experts agree on the increasing probability of occurrence of high intensity rain events, reaching the limits of current wastewater infrastructure operational capacities. In order for these infrastructures to continue to provide adequate performance with respect to current norms and regulations, while limiting costs and budgetary needs, flexible solutions, such as implementing real-time control (RTC) systems, should be considered. Especially since recent technological developments bring new and more reliable tools to measure water quality, and also to analyse and manage large quantities of data.This study focuses on the behaviour of the particulate pollution in the IUWS, an important water quality indicator, commonly measured as the total suspended solids (TSS) concentrationand highly correlated with turbidity measurement. As turbidity sensors are readilyavailable with increasing reliability, their operational use can be envisioned. The objective of this thesis is to develop and assess water quality-based control strategies for the IUWS. An integrated - sewer and WRRF - model, using the particle settling velocity distribution (PSVD) approach, has been proposed. This approach improves the description of particle settling and resuspension in the whole system, allowing a better forecast of TSS concentration dynamics, especially during rain events. Based on this integrated model, innovative water quality-based control strategies, aiming at reducing overall discharge of particulate pollutants (and the heavy metals, pathogens, etc. that come with it) to the receiving water during wet weather, have been proposed and successfully tested. Experimental studies demonstrated that turbidity sensors installed in the IUWS are subject to harsh conditions, leading to faulty sensor signals. The performance loss due to the use of water quality-based control strategies affected by a faulty turbidity signal has been evaluatedin detail and this has demonstrated the need for fault detection tools. When a faulty signal is detected, alarms will allow operators to decide which operating mode should be used instead of the one using the now faulty sensor. Alternatively, automatic responses and changes could be implemented to counteract the effects of a faulty signal. This last option, called downgraded operating mode, which implies reverting to a control strategy that does not use the faulty signal, has been simulated and tested. If the fault is detected within reasonable time (which depends on the system dynamics), these simulations have shown that part of the increase of performance gained with the water quality-based strategy using the good signal, can be maintained. The results of this study thus present a strong case for optimizing wastewater infrastructure operations, especially when they are considered as an integrated system, instead of autonomous and mutually exclusive systems. Furthermore, the developed integrated model may be use as a decision support tool to help determine the best possible (integrated) controlstrategy. Keywords : chemically enhanced primary treatment (CEPT); integrated modelling; integratedurban wastewater system (IUWS); real-time control (RTC); stormwater management.
Campo, Clément. "Conception d'un système de contrôle d’antennes basé sur la radio logicielle pour réception et émission améliorées de données." Thesis, Poitiers, 2020. http://www.theses.fr/2020POIT2270.
Full textAs a wireless way to exchange information, electromagnetic waves are more omnipresent in our environment than ever. The ever increasing number of connected devices calls for a better use of the available spectrum. In the particular case of telecommunications with a projectile, which is the case of study in this thesis, communications must also be discreet and reliable, even in a hostile environment. In the general framework of telecommunications as well as in this particular field of application, antenna arrays and the dynamic spatial filtering they allow offer multiple advantages for present and future challenges. Antenna array steering requires phase coherent and phase aligned functioning from the control electronics. In a previous PhD thesis, an analog system allowed beam steering of the array embedded in a projectile towards a base station at all times during projectile flight. However, this system was only able to switch between 16 different configurations for the embedded array radiation pattern and was functional only around a 5.2 GHz working frequency. On the other hand, Software Defined Radio (SDR) uses wide-band programmable components thanks to which received or generated signals can be processed in digital baseband. Therefore, using SDR would allow for a more precise control of the radiation pattern over large frequency bandwidths. Despite these promises, this technology remains rarely used for phase coherent applications. This work hence studies possibilities provided by commercial SDR for phase coherent applications. Telecommunications with a projectile, which also require phase alignment, constitute the considered application. Linear and planar antenna arrays are studied. An antenna weighting system of 4 channels for both data reception and transmission is assembled using commercial SDR. Distinct solutions are developed for data reception or transmission in order to automate phase shift compensation between channels. Several antenna weighting and Direction of Arrival (DOA) algorithms are implemented in C++. As the available equipment does not allow the automated measurement of the radiation pattern of antenna arrays when steered by SDR, a dedicated experimental setup is proposed. The developed system performance is then quantified in an anechoic environment for arrays of different geometries, and working frequencies from 2.3 to 5.2 GHz. Depending on the measured array, the main lobe or null can be steered within 60 to more than 100° along 1 or 2 dimensions. The implemented algorithms are also used to develop a projectile tracking station based on DOA estimation of the transmitter embedded in the projectile. The resulting station is tested with several projectiles flying at a speed close to Mach 1. The projectiles are electronically followed by the system as expected from simulations. The signal to noise ratio of the station combined signals is superior to that of a single element signal by more than 5 dB, and transmitted flight data is correctly decoded
Bouloc, Jeremy. "Système de contrôle pour microscope à force atomique basé sur une boucle à verrouillage de phase entièrement numérique." Thesis, Aix-Marseille, 2012. http://www.theses.fr/2012AIXM4307/document.
Full textAn atomic force microscope (AFM) is used to characterize insulating materials or semiconductors with a resolution up to the atomic length scale. The microscope includes a force sensor linked to a control electronic in order to properly characterize these materials. Among the various modes (static and dynamic), we focus mainly on the dynamic mode and especially on the frequency modulation mode (FM-AFM). In this mode, the force sensor is maintained as a harmonic oscillator by the servo system. The research project ANR Pnano2008 entitled: "metal piezoresistivity silicon carbide cantilever for very high frequency dynamic AFM" aims to significantly increase the performance of a FM-AFM by developing new very high frequency force sensors. The goal is to increase the sensitivity of the sensor and to decrease the time necessary to obtain topography images of the material. The control system of this new sensor must be able to detect frequency variations as small as 100mHz for cantilevers with resonance frequencies up to 50MHz. Since the state-of-the-art systems doe not present these performances, the objective of this thesis was to develop a new control system. It is fully digital and it is implemented on a FPGA based prototyping board. In this report, we present the system overall functioning and its main features which are related to the cantilever resonant frequency detection. This detection is managed by a phase locked loop (PLL) which is the key element of the system
Deslandres, Véronique. "Contribution à l'amélioration de la qualité dans les systèmes de production par un système basé sur la connaissance." Lyon 1, 1993. http://www.theses.fr/1993LYO10177.
Full textKanta, Abdoul-Fatah. "Développement d'un système expert basé sur l'intelligence artificielle pour la prédiction et le contrôle du procédé de projection plasma." Besançon, 2007. http://www.theses.fr/2007BESA2013.
Full textAtmospheric plasma spraying (APS) is a process in which the coating properties are indirectly linked to the selection of the operating parameters (power, injection kinematic and environmental parameters). This process is characterized by several parametric drifts and fluctuations at different characteristic times. These drifts and fluctuations originate especially from the electrode wear and intrinsic plasma jet instabilities. The objective of this work is to develop an expert system which could adjust in real time the operating parameters as a function, in a first time, of the measured in-flight particles to manufacture a coating. Due to the large amplitudes of these drifts an fluctuations, the strategy to adopt will depend on the required corrections to apply to the particle characteristics. Artificial intelligence base on artificial neural networks and fuzzy logic controller appeared as pertinent tools to reach this objective. Fuzzy logic permits to define parametric correction rules and the artificial neural networks permits to predict in-flight particle and coating characteristics. Coupling neural protocols to fuzze logic permit to develop an independent controller. This system offers the possibility to adjust the process parameters in order to improve the reproducibility of the process and also, to optimize the coatings development conditions by taking into account the effects of the electrode wear. The system is flexible in order to to permit a full control based on pre-defined rules aiming and maintaining at constant value the in-flight particle characteristics (surface temperature and velocity) by adjusting in real time pwer parameters whatever the fluctuations
Tran, Trung Thien. "Extraction robuste de primitives géométriques 3D dans un nuage de points et alignement basé sur les primitives." Doctoral thesis, Université Laval, 2016. http://hdl.handle.net/20.500.11794/26744.
Full textIn this research project, we address reverse engineering and quality control problems that play significant roles in industrial manufacturing. Reverse engineering attempts to rebuild a 3D model from the scanned data captured from a object, which is the problem similar to 3D surface reconstruction. Quality control is a process in which the quality of all factors involved in production is monitored and revised. In fact, the above systems currently require significant intervention from experienced users, and are thus still far from being fully automated. Therefore, many challenges still need to be addressed to achieve the desired performance for automated production. The first proposition of this thesis is to extract 3D geometric primitives from point clouds for reverse engineering and surface reconstruction. A complete framework to extract multiple types of primitives from 3D data is proposed. In particular, a novel validation method is also proposed to assess the quality of the extracted primitives. At the end, all primitives present in the point cloud are extracted with their associated data points and descriptive parameters. These results could be used in various applications such as scene and building reconstruction, constructive solid geometry, etc. The second proposition of the thesis is to align two 3D datasets using the extracted geometric primitives, which is introduced as a novel and robust descriptor. The idea of using primitives for alignment is addressed several challenges faced by existing registration methods. This alignment problem is an essential step in 3D modeling, registration and model retrieval. Finally, an automatic method to extract sharp features from 3D data of man-made objects is also proposed. By integrating the extracted sharp features into the alignment framework, it is possible implement automatic assignment of primitive correspondences using attribute relational graph matching. Each primitive is considered as a node of the graph and an attribute relational graph is created to provide a structural and relational description between primitives. We have experimented all the proposed algorithms on different synthetic and real scanned datasets. Our algorithms not only are successful in completing their tasks with good results but also outperform other methods. We believe that the contribution of them could be useful in many applications.
Truong, Hien Thi Thu. "Un modèle de collaboration basé sur les contrats et la confiance." Phd thesis, Université de Lorraine, 2012. http://tel.archives-ouvertes.fr/tel-00769076.
Full textThibault, Robert. "Contrôle de l'énergie injectée dans un réseau électrique par un convertisseur triphasé utilisant un régulateur basé sur un modèle interne sinusoïdal." Mémoire, École de technologie supérieure, 2006. http://espace.etsmtl.ca/511/1/THIBAULT_Robert.pdf.
Full textGrout, Olivier. "Conception d'un système de management de l'environnement basé sur les normes ISO 14000 au sein de l'industrie pharmaceutique." Bordeaux 2, 1999. http://www.theses.fr/1999BOR2P004.
Full textGeveaux, Emmanuel. "Conception d'un environnement de développement des applications de contrôle de procédé basé sur le modèle formel GRAFCET et fondé sur un langage graphique flot de données." Poitiers, 1998. http://www.theses.fr/1998POIT2298.
Full textKendoul, Farid. "Modélisation et commande de véhicules aériens autonomes, et développement d'un pilote automatique basé sur la vision." Compiègne, 2007. http://www.theses.fr/2007COMP1694.
Full textIn this thesis, we address the 3D flight control problem for small unmanned aerial vehicles. The first step in achieving such UA V autonomy is basic stability. In doing so, we investigate control techniques which would allow a small rotorcraft UA V to perform autonomously basic tasks such as take-off, hovering and landing. The proposed flight controllers consider actuator saturation and guarantee system stability and good flight performance. One of the main contributions of this thesis is also the development of a vision-based autopilot for self localization, obstacles detection and control of small autonomous helicopters using optic flow and inertial data. The proposed autopilot is inspired by insects and represents a major step toward our goal of designing autonomous small flying machines capable of navigating within urban and unknown environments. It is composed of a vision module for localization and perception, and an adaptive control system for navigation and trajectory tracking
Guyot, Dimitri. "Evaluation sur modèle de simulation thermique dynamique calibré des performances d’un contrôleur prédictif basé sur l’utilisation de réseaux de neurones." Thesis, Paris, HESAM, 2020. http://www.theses.fr/2020HESAC022.
Full textThe development of machine learning techniques, particularly neural networks, combined with the development of new information and communication technologies, is shaking up our societies through technological advances in a variety of sectors. The building sector is not spared, so these techniques may represent an interesting opportunity in a context where greenhouse gas emissions must be drastically reduced. The objective of this work is to assess the interest of these techniques in the field of building energy, with the aim of reducing energy consumption and improving thermal comfort. In addition, we ensure that this evaluation is carried out with a global vision, by placing the possible advantages in front of the different needs relating to the development of these technologies. This thesis work is organized in three parts preceded by a detailed introduction intended to give the reader an overview of the various contextual elements, thus allowing the thesis work to be placed in perspective. We then give in the first part the theoretical framework needed to understand the problems encountered during the elaboration and creation of neural networks for building energy applications. Then, a bibliographical study giving the reader a broad overview of the various applications of neural networks in the field of building energy is presented. The second part is devoted to the calibration of the building model that is then used to test and evaluate a predictive controller implementing neural networks. After an explanation of the method used and a detailed presentation of the model, a complete analysis of the calibration results is carried out. We conclude this part with observations and recommendations regarding the standard calibration guidelines recommended by three international organizations. Finally, a practical application using neural networks for the predictive control of indoor temperature is presented in the third part. After a theoretical introduction concerning predictive control, we detail the method employed to train the neural networks used. The results obtained in simulation with a predictive controller are then analyzed and compared with those obtained with two reference controllers for various simulation hypothesis. The predictive controller is thus tested in several scenarios, ranging from an ideal situation to more realistic operating conditions, including two different types of heat emitters, namely radiant ceilings and underfloor heating
Bruni, Christelle. "Dynamique longitudinale et transverse d'un laser à électrons libres basé sur un anneau de stockage de deuxième et troisième génération." Paris 11, 2004. http://www.theses.fr/2004PA112165.
Full textThis work aims at improving the understanding of the dynamics of a Free Electron Laser (FEL) based on a storage ring. A new phenomena of saturation by the electron bunch transverse dimensions is pointed out, during the FEL operation with a magnetic optics ''low emittance''. A part of this thesis work concerns the study of the longitudinal dynamics of the FEL depending on a parameter : the detuning between the pass frequency of the electrons and the back and return frequency of the laser pulse in the optical cavity. This study shows that the FEL presents a pulsed or a steady state behavior depending on the detuning. Besides, a spectro-temporal behavior of the FEL is pointed out showing the FEL pulse substructures and holes in the spectrum. Another part of this work concerns the response of the resonant FEL system, when a perturbation is applied on the gain. Under the effect of a modulation, the FEL can present a periodic or chaotic regime. The analysis of these regimes shows that a period doubling cascade leads to chaos. After having generated a pulsed or chaotic temporal dynamical behavior of the FEL thanks to a modulation, a feedback is realized to suppress the natural pulsed regimes of the FEL in favor of a steady state regime. In addition, supplementary electronic loops allow to damp longitudinal instabilities (50 Hz and its harmonics), affecting particularly the ELETTRA FEL
El, Baida Rania. "Gestion des incohérences dans les systèmes de contrôle d'accès." Artois, 2004. http://www.theses.fr/2004ARTO0401.
Full textModelling information security policies is being an important task in many domains. We proposed a new access control system called OrBAC (Organization based access control). This system brings many solutions to the existing access control systems. For instance, OrBAC takes into account the context and is able to represent various kinds of privileges : permission, prohibition and obligation. However, this system does not deal with conflicts due to the joint handling of permission or obligation and prohibition policies. We dealt with the problem of handling conflicts in the OrBAC system, modelled by first order logic knowledge bases. We showed that approaches suggested for handling conflicts in propositionnal knowledge bases are not adapted for inconsistent first order knowledge bases. We proposed an approach in which we weak first order formulas responsible of conflicts
Marceau, Anne. "Les règles et principes constitutionnels relatifs au contrôle des collectivités territoriales : étude sur l'article 72 alinéa 3 de la Constitution française du 4 octobre 1958." Dijon, 2000. http://www.theses.fr/2000DIJOD015.
Full textPialasse, Jean-Philippe. "Évaluation du contrôle sensorimoteur chez les patients ayant une scoliose idiopathique de l'adolescent : vers un biomarqueur des troubles sensorimoteur basé sur la stimulation vestibulaire galvanique." Doctoral thesis, Université Laval, 2015. http://hdl.handle.net/20.500.11794/26968.
Full textScoliosis is the most frequent spinal deformity in adolescence. In 80% of the cases, it is idiopathic, meaning that no cause has been associated with the patient's case. Idiopathic scoliosis seems to respond to a multifactorial model including genetic, environmental, neurological, hormonal, biomechanical and skeletal growth factors. A neurological assumption is that an anomaly of the vestibular system would cause asymmetrical activation of the vestibulospinal pathway and of paraspinal muscles. This cascade would generate the scoliotic deformity. Animal models have demonstrated this possibility. In addition, many vestibular related anomalies are observed in adolescents with scoliosis as vestibulo-ocular reflex abnormalities or balance control disorders. Galvanic vestibular stimulation allows exploring sensorimotor control by faltering the vestibular afferents. The objective of this thesis is to explore the sensorimotor control through vestibular-evoked postural response in patients with scoliosis and healthy controls. The results of the first study show that the vestibular-evoked postural response is larger in patients compared to controls. Moreover, the amplitude of the postural response is not scaled to the spinal deformation amplitude. In a second study, through a neuromechanical feedback control model, we demonstrate that patients assigned a larger weight to vestibular signal compared to controls. Results of the third study reveal that young adults with idiopathic scoliosis, compared to controls, have a larger postural response. This observation excludes a transient response due to the maturation of the nervous system. Then, balance control impairment seems secondary to a neurosensory phenomenon as balance control dysfunction is observed in patients who had surgery reducing spine deformation. Ultimately, an algorithm has been developed to distinguish patients with or without sensorimotor control problems compared to healthy adolescents. Remarkably, the amplitude of the feedforward vestibular response of these patients is larger and they assign a larger weight to vestibular than proprioceptive information. Overall, this thesis proposes a procedure to identify patients with scoliosis having sensorimotor control impairment. In the end, it is believed that the classification procedure may help future clinical studies as patients with sensorimotor dysfunction could be identified. Hopefully, future research will enhance this procedure and lead to an efficient biomarker.
Tapsoba, René. "Ciblage d'inflation et règles budgétaires dans les pays en développement: Interactions et Conséquences macroéconomiques." Phd thesis, Université d'Auvergne - Clermont-Ferrand I, 2012. http://tel.archives-ouvertes.fr/tel-00734755.
Full textHuafeng, Yu. "Un Modèle Réactif Basé sur MARTE Dédié au Calcul Intensif à Parallélisme de Données : Transformation vers le Modèle Synchrone." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2008. http://tel.archives-ouvertes.fr/tel-00497248.
Full textLes travaux de cette thèse s'inscrivent dans le cadre de la validation formelle et le contrôle réactif de calculs à haute performance sur systèmes-sur-puce (SoC).
Dans ce contexte, la première contribution est la modélisation synchrone accompagnée d'une transformation d'applications en équations synchrones. Les modéles synchrones permettent de résoudre plusieurs questions liées à la validation formelle via l'usage des outils et techniques formels offerts par la technologie synchrone. Les transformations sont développées selon l'approche d'Ingénierie Dirigé par les Modèles (IDM).
La deuxième contribution est une extension et amélioration des mécanismes de contrôle pour les calculs à haute performance, sous forme de constructeurs de langage de haut-niveau et de leur sémantique. Ils ont été défini afin de permettre la vérification, synthèse et génération de code. Il s'agit de déterminer un niveau d'abstraction de représentation des systèmes où soit extraite la partie contrôle, et de la modéliser sous forme d'automates à états finis. Ceci permet de spécifier et implémenter des changements de modes de calculs, qui se distinguent par exemple par les ressources utilisées, la qualité de service fournie, ou le choix d'algorithme remplissant une fonctionnalité.
Ces contributions permettent l'utilisation d'outils d'analyse et vérification, tels que la vérification de propriétés d'assignement unique et dépendance acyclique, model checking. L'utilisation de techniques de synthèse de contrôleurs discrets est également traitée. Elles peuvent assurer la correction de faˆ on constructive: à partir d'une spécification partielle du contrôle, la partie manquante pour que les propriétés soient satisfaites est calculée. Grâce à ces techniques, lors du développement de la partie contrôle, la spécification est simplifiée, et le résultat est assuré d'être correct par construction.
Les modélisations synchrone et de contrôle reposes sur MARTE et UML. Les travaux de cette thèse sont été partiellement implémentés dans le cadre de Gaspard, dédié aux applications de traitement de données intensives. Une étude de cas est présentée, dans laquelle nous nous intéressont à une application de système embarqué pour téléphone portable multimédia.
Ghassemi, Elham. "Modèle computationnel du contrôle auto-adaptatif cérébelleux basé sur la Logique Floue appliqué aux mouvements binoculaires : déficit de la coordination binoculaire de la saccade horizontale chez l’enfant dyslexique." Thesis, Paris 5, 2013. http://www.theses.fr/2013PA05L001.
Full textThis thesis focuses on the cerebellum. We follow two main lines: in terms of cerebellar functions, we are interested in learning and adaptation motor control ; in terms of cerebellar dysfunctions, we are interested in developmental dyslexia.We focus on learning motor control in order to provide a functional computational model applied to voluntary eye movements. To this end, Fuzzy Logic is one of our valuable tools. We proposed two models. The former is AFCMAC (Auto-adaptive Fuzzy Cerebellar Model Articulation Controller), the result of the integration of Fuzzy Logic in CMAC (Cerebellar Model Articulation Controller) architecture, in order to improve learning speed/time and memory requirements compared to the CMAC. The latter is CMORG (fuzzy logiC based Modeling for Oculomotor contRol LearninG), whose structure is also based on Fuzzy Logic, and in which, the neural network is used as the memory to handle Fuzzy rules. The evaluation results of the proposed (AFCMAC and CMORG) and studied (CMAC and FCMAC – Fuzzy Cerebellar Model Articulation Controller) models via oculomotor data of dyslexic and control groups while reading show that CMORG is the most efficient both in terms of learning speed/time and also memory consumption. Another main advantage of CMORG over the other models is its interpretability by experts. Regarding the developmental dyslexia, we conducted an experimental study on binocular motor control deficits during saccades in six dyslexic children while two different tasks (text reading and character string scanning) and in two viewing distances (40 cm and 100 cm). We corroborate and adhere to the idea that the (bad) quality of binocular coordination of saccades in dyslexic children is independent of reading difficulties, maybe associated with magnosystem and cerebellar deficit hypothesis
Attemene, N'guessan Stéphane. "Optimisation temps réel des flux énergétiques au sein d'un système multi-sources multi-charges basé sur les énergies d'origine renouvelable." Thesis, Bourgogne Franche-Comté, 2019. http://www.theses.fr/2019UBFCD044.
Full textThis work is focused on the real-time optimal control of a stand-alone system consisting of a photovoltaic generator, a PEM fuel cell, an alkaline electrolyzer, a battery and supercapacitor pack for a stationary application. The coupling of these different sources aims to improve performance, the availability of the resulting electrical grid, the supply of electricity over much longer periods, and especially the satisfaction of the load by using each source in a controlled way.First, a thorough study of the feasibility of the system from a technical, energetic, economic and environmental point of view is carried out. As a result, an optimal sizing method is proposed. A sensitivity analysis to evaluate the influence of subsystems cost and the size respectively on the overall energy cost and the equivalent CO2 emitted by the system is also discussed. Then, a model enabling easy scaling of components to achieve the capacity required for the system is developped. The global model of the system is obtained by exploiting the modularity of the formalism used for modeling (the Energetic Macroscopic Representation). Finally, an energy management method based on Energy consumption Minimization Strategy (ECMS) is proposed. A comparative study of the results obtained by the ECMS and those obtained by dynamic programming has enabled the validation of the optimal control strategy developed
Gaillard, Arnaud. "Système éolien basé sur une MADA : contribution à l'étude de la qualité de l'énergie électrique et de la continuité de service." Thesis, Nancy 1, 2010. http://www.theses.fr/2010NAN10023/document.
Full textWind Energy Conversion Systems (WECS) will probably be intended, in a short term, to provide ancillary services such as reactive power compensation, as done by conventional generators of power plants and/or participate in improving electric power quality by filtering harmonic currents. Furthermore, by considering the increased installed wind power, connected to the grid, WECS will certainly ensure, in long term, the continuity of service following an electrical fault on the grid or on one of elements of the electromechanical conversion (power switches, sensors…) to ensure their reliability. In this thesis, we have studied in the first part a WECS based on Doubly Fed Induction Generator (DFIG) with an amplification gain between the stator and rotor currents. We have demonstrated that such a WECS can efficiently improving the electric power quality by compensating simultaneously reactive power and harmonic currents on the grid, without requiring any oversized elements in the electromechanical conversion chain. Then, we studied fault tolerant converter topologies and associated controls, ensuring the continuity of service in the presence of possible faults of a semiconductor or a current sensor. To minimize the time fault detection, we used an FPGA (Field Programmable Gate Array) to implement the fault tolerant controller. Simulation, “FPGA in the loop” prototyping and experimental results have validated the performances of the proposed methods and fault tolerant topologie
Cadorette, Jules. "Étude des paramètres de contrôle d'un système de détection basé sur des détecteurs à photodiode à avalanche : application à la tomographie par émission de positrons à haute résolution spatiale." Mémoire, Université de Sherbrooke, 1994. http://hdl.handle.net/11143/12057.
Full textBlagouchine, Iaroslav. "Modélisation et analyse de la parole : Contrôle d’un robot parlant via un modèle interne optimal basé sur les réseaux de neurones artificiels. Outils statistiques en analyse de la parole." Thesis, Aix-Marseille 2, 2010. http://www.theses.fr/2010AIX26666.
Full textThis Ph.D. dissertation deals with speech modeling and processing, which both share the speech quality aspect. An optimum internal model with constraints is proposed and discussed for the control of a biomechanical speech robot based on the equilibrium point hypothesis (EPH, lambda-model). It is supposed that the robot internal space is composed of the motor commands lambda of the equilibrium point hypothesis. The main idea of the work is that the robot movements, and in particular the robot speech production, are carried out in such a way that, the length of the path, traveled in the internal space, is minimized under acoustical and mechanical constraints. Mathematical aspect of the problem leads to one of the problems of variational calculus, the so-called geodesic problem, whose exact analytical solution is quite complicated. By using some empirical findings, an approximate solution for the proposed optimum internal model is then developed and implemented. It gives interesting and challenging results, and shows that the proposed internal model is quite realistic; namely, some similarities are found between the robot speech and the real one. Next, by aiming to analyze speech signals, several methods of statistical speech signal processing are developed. They are based on higher-order statistics (namely, on normalized central moments and on the fourth-order cumulant), as well as on the discrete normalized entropy. In this framework, we also designed an unbiased and efficient estimator of the fourth-order cumulant in both batch and adaptive versions
Nguyen, Lam Hung. "Contrôle non linéaire et guidage visuel de véhicules sous-marins." Thesis, Université Côte d'Azur (ComUE), 2019. http://theses.unice.fr/2019AZUR4095.
Full textRecent advances in technology have led to a growing interest in underwater robotic applications.Historically, payload constraints have severely limited the development of scale-model underwatervehicles. They are small; their dynamics are highly nonlinear, subjected to complex and imprecisehydrodynamic effects and to disturbances commensurable with the control actions. Consequently, theyprovide an ideal testing ground for sophisticated nonlinear control techniques. Classical controltechniques applied to these vehicles seem to be inadequate in dealing with their highly nonlineardynamics, model uncertainties and external disturbances. Another key arising issue is the difficulty ofnavigation in cluttered environments and close to obstacles without precise positioning with respect tothe environment and without velocity measurements.In the present thesis, two research topics have been considered. Firstly, two novel dynamic visualservoing controllers exploiting the so-called homography matrix for fully-actuated vehicles (withoutmeasuring the linear velocity) have been proposed. These two approaches are derived for two differentcamera configurations: downward-looking and forward-looking. The key contribution here is thenon-requirement of costly Doppler Velocity Log (DVL) velocity sensor. The control performance hasbeen validated over simulations and via practical trials in a real challenging environment. Theexperimental platform as well as the software was built over the period of the thesis. Secondly, a noveltrajectory tracking algorithm for a class of underactuated vehicles with axisymmetric slender body hasbeen developed. The proposed approach makes use of a new nonlinear dynamic model of the vehicle,and exploits the symmetry of the vehicle in control design. The control performance and its robustnesshave been validated via simulations using a realistic model of an underwater vehicle
Liu, Mingxing. "Personnage virtuel : contrôleur hybride couplant commande dynamique multi-objectifs et mouvements capturés." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2012. http://tel.archives-ouvertes.fr/tel-00825375.
Full textSavourey, David. "Ordonnancement sur machines parallèles : minimiser la somme des coûts." Phd thesis, Université de Technologie de Compiègne, 2006. http://tel.archives-ouvertes.fr/tel-00156405.
Full textinférieures a également été réalisée. Enfin, nous avons proposé une méthode de résolution exacte utilisant les règles de dominance ainsi que les bornes inférieures.
Mihaita, Adriana. "Approche probabiliste pour la commande orientée évènement des systèmes stochastiques à commutation." Thesis, Grenoble, 2012. http://www.theses.fr/2012GRENT036/document.
Full textHybrid systems are dynamical systems, characterized by a dual behaviour, a continuousinteraction between a discrete and a continuous functioning part. The center ofour work is represented by a particular class of hybrid systems, more specific by thestochastic switching systems which we model using continuous time Markov chains anddifferential equations.The random behaviour of such systems requires a special command which adapts tothe arbitrary events that can completely change the evolution of the system. We chose anevent-based control policy which is triggered only when it’s necessary (on an unforeseenevent - for example when a threshold that is reached), and until certain functioningconditions are met (the system returns in the normal state).Our approach aims to develop a probabilistic model that calculates a performancecriterion (in this case the energy of the system) for the proposed control policy. We startby proposing a discrete event simulation for the controlled stochastic switching system,which gives us the opportunity of applying a direct optimisation of the control command.It also allows us to compare the results with the ones obtained by the analytical modelswe have built when the event-based control is applied.An analytical model for computing the energy consumed by the system to apply thecontrol is designed by using the exit probabilities of the control region, respectively, thesojourn times of the Markov chain before and after reaching the control limits. The lastpart of this work presents the results we have obtained when comparing the analyticaland the simulation method
Mihaita, Adriana, and Adriana Mihaita. "Approche probabiliste pour la commande orientée évènement des systèmes stochastiques à commutation." Phd thesis, Université de Grenoble, 2012. http://tel.archives-ouvertes.fr/tel-00770325.
Full textCheaito, Marwan. "Un cadre de spécification et de déploi[e]ment de politiques d'autorisation." Toulouse 3, 2012. http://thesesups.ups-tlse.fr/1617/.
Full textOur work proposes a methodology for the conception and the development of an adaptable authorization system that can cover the requirements of access control in organizations such as the heterogeneity of organizational practices, the technologies used and the contexts to consider. To answer, two approaches of management have guided our study: the attribute based access control and the policy based management. * The attribute based access control permits to specify permissions related to any security characteristics of users, actions, resources and environment. This approach addresses the problems related to the expressivity of languages of authorization policies. * The policy based management, for its part, aims to enable the dynamic adaptability behavior of the system through the management of policies. Roughly, the architecture supporting this approach consists of two main entities: a policy decision point (PDP), and a policy enforcement point (PEP). The PDP is independent of the managed element, which interprets the policy and takes management decisions based on it. The PEP is the interface between the PDP and the managed element. It compels the managed system to execute the management decisions taken by the PDP. This architecture supports the integration of the management system in a managed environment. Therefore, we have chosen the standard XACML as a target technology because it supports both approaches. If in a theoretical point of view, XACML seems to address the issues of adaptability. However, systems designed according to this standard are limited to a given situation. In fact, a developed XACML system that satisfies the given requirements and a particular technological environment cannot be easily reused in another context. Therefore, it is necessary to define a methodology in order to render such authorization system adaptable in practice. Our approach is to define a minimal authorization system (i. E. A core that only supports the standard XACML) that can be easily extended for specific requirements in a given situation (ex. , the ability to express and implement policies including complex constraints, adaptation to a particular technological environment, etc. ). This raises the following questions: * What extensions that should be imported to the minimal authorization system? Is it possible to reuse the extensions for different situations? For example, the number of technologies used to store the user credentials is limited (ex. , LDAP, MySQL, Active Directory). However, the user credentials and the structuring of the data of user credentials may differ from one organization to another. * How to manage the lifecycle of the extensions in the authorization system ? In fact, there is a strong link between the use of an extension and the requirements expressed in an authorization policy. For example, the existence in the authorization system of an extension to retrieve the role of the user in a MySQL database is required only if the authorization system must evaluate at least one policy that includes constraints on the role of users. Otherwise, this extension does not happen to be present in the minimal authorization system. Similarly, if a policy including constraints on user roles is loaded into the authorization system, the associated extension must be added also. Based on the various studies from the field of software engineering related to the concept of adaptability, we have identified the requirements of adaptation of authorization systems. This work allows us to determine a classification of extensions according to the requirements of adaptability in authorization systems of type XACML. We have also proposed a methodology to facilitate the creation of extensions and promoting their reusability in the authorization system. Finally, we have addressed the problem of managing the lifecycle of the extensions in the authorization system by proposing the concept of "self-contained policy". A self-contained policy includes the authorization rules and any necessary information in order the authorization system interprets and runs it (i. E. , the extensions and their configurations). Also, we have described the architecture for specifying and deploying self-contained policies using the approach components-oriented services. Our architecture is flexible and dynamically adaptable. We have implemented it using the OSGi framework in order to validate our work in both terms of feasibility, and performance. In addition, we have conducted a self-contained policy editor that facilitates the writing of authorization rules and the configuration of the used extensions
Kortebi, Abdesselem. "Evaluation d' une architecture IP orientée flot." Paris 6, 2005. http://www.theses.fr/2005PA066517.
Full textZhao, Zilong. "Extracting knowledge from macroeconomic data, images and unreliable data." Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALT074.
Full textSystem identification and machine learning are two similar concepts independently used in automatic and computer science community. System identification uses statistical methods to build mathematical models of dynamical systems from measured data. Machine learning algorithms build a mathematical model based on sample data, known as "training data" (clean or not), in order to make predictions or decisions without being explicitly programmed to do so. Except prediction accuracy, converging speed and stability are another two key factors to evaluate the training process, especially in the online learning scenario, and these properties have already been well studied in control theory. Therefore, this thesis will implement the interdisciplinary researches for following topic: 1) System identification and optimal control on macroeconomic data: We first modelize the China macroeconomic data on Vector Auto-Regression (VAR) model, then identify the cointegration relation between variables and use Vector Error Correction Model (VECM) to study the short-time fluctuations around the long-term equilibrium, Granger Causality is also studied with VECM. This work reveals the trend of China's economic growth transition: from export-oriented to consumption-oriented; Due to limitation of China economic data, we turn to use France macroeconomic data in the second study. We represent the model in state-space, put the model into a feedback control framework, the controller is designed by Linear-Quadratic Regulator (LQR). The system can apply the control law to bring the system to a desired state. We can also impose perturbations on outputs and constraints on inputs, which emulates the real-world situation of economic crisis. Economists can observe the recovery trajectory of economy, which gives meaningful implications for policy-making. 2) Using control theory to improve the online learning of deep neural network: We propose a performance-based learning rate algorithm: E (Exponential)/PD (Proportional Derivative) feedback control, which consider the Convolutional Neural Network (CNN) as plant, learning rate as control signal and loss value as error signal. Results show that E/PD outperforms the state-of-the-art in final accuracy, final loss and converging speed, and the result are also more stable. However, one observation from E/PD experiments is that learning rate decreases while loss continuously decreases. But loss decreases mean model approaches optimum, we should not decrease the learning rate. To prevent this, we propose an event-based E/PD. Results show that it improves E/PD in final accuracy, final loss and converging speed; Another observation from E/PD experiment is that online learning fixes a constant training epoch for each batch. Since E/PD converges fast, the significant improvement only comes from the beginning epochs. Therefore, we propose another event-based E/PD, which inspects the historical loss, when the progress of training is lower than a certain threshold, we turn to next batch. Results show that it can save up to 67% epochs on CIFAR-10 dataset without degrading much performance. 3) Machine learning out of unreliable data: We propose a generic framework: Robust Anomaly Detector (RAD), The data selection part of RAD is a two-layer framework, where the first layer is used to filter out the suspicious data, and the second layer detects the anomaly patterns from the remaining data. We also derive three variations of RAD namely, voting, active learning and slim, which use additional information, e.g., opinions of conflicting classifiers and queries of oracles. We iteratively update the historical selected data to improve accumulated data quality. Results show that RAD can continuously improve model's performance under the presence of noise on labels. Three variations of RAD show they can all improve the original setting, and the RAD Active Learning performs almost as good as the case where there is no noise on labels
Tapsoba, René. "Inflation targeting and fiscal rules in developing countries : interactions and macroeconomic consequences." Thesis, Clermont-Ferrand 1, 2012. http://www.theses.fr/2012CLF10394/document.
Full textThis thesis is concerned with the role of Inflation Targeting (IT) and Fiscal Rules (FRs), as well as of their interactions, on macroeconomic environment. After laying the conceptual and empirical backgrounds of both these rules-Based policy frameworks (Chapter 1), the dissertation highlights new evidence on their macroeconomic consequences. First, IT adoption does help attracting more Foreign Direct Investment into Developing Countries (Chapter 2). Second, inthese countries, IT adoption provides strong incentives for governments to undertake reforms designed to improve the quality of institutions (Chapter 3). Third, the introduction of national-Level numerical FRs does stand as a credible remedy against fiscal indiscipline in these countries, all the more in politically stable economies and where the numberof FRs in place is important. But this discipline-Enhancing effect of FRs decreases with the time length since their adoption, and is weakened in the presence of supranational rules and in countries with more fragmented government. Interestingly, this effect varies with the type of rules: while Budget Balance Rules and Expenditure Rules have significant disciplineenhancing effects, the effect of Debt Rules proved not significantly different from zero (Chapter 4). The last three chapters of the thesis focus on the role of the interaction between IT and FRs, and to a broader extent, on the interplay between monetary and fiscal authorities. The first result that emerges is that IT, which is a framework for conducting monetary policy, proves also sufficiently binding for fiscal authorities to providing them with strong incentives for improving fiscal discipline, notably in developing countries (Chapter 5). In addition, on the one hand, IT and FRs act complementarily, as adopting both IT and FRs leads to better results in terms of running fiscal surpluses and in terms of bringing down average inflation than adopting only one of these two frameworks. On the other hand, the sequence which consists of introducing FRs first before adopting IT yields better inflationary and fiscal performances than the opposite sequence (Chapter 6). Finally, better Policy Mix coherence, that is, better coordination between monetary and fiscal policies, is conducive to higher economic growth in the Economic Community of West African States (ECOWAS) (Chapter 7)
El, Maarabani Mazen. "Verification and test of interoperability security policies." Phd thesis, Institut National des Télécommunications, 2012. http://tel.archives-ouvertes.fr/tel-00717602.
Full textJemel, Mayssa. "Stockage des données locales : sécurité et disponibilité." Electronic Thesis or Diss., Paris, ENST, 2016. http://www.theses.fr/2016ENST0053.
Full textDue to technological advancements, people are constantly manipulating multiple connected and smart devices in their daily lives. Cross-device data management, therefore, remains the concern of several academic and industrial studies. The proposed frameworks are mainly based on proprietary solutions called private or closed solutions. This strategy has shown its deficiency on security issues, cost, developer support and customization. In recent years, however, the Web has faced a revolution in developing standardized solutions triggered by the significant improvements of HTML5. With this new version, innovative features and APIs are introduced to follow business and user requirements. The main purpose is to provide the web developer with a vendor-neutral language that enables the implementation of competing application with lower cost. These applications are related neither to the used devices nor to the installed software. The main motivation of this PhD thesis is to migrate towards the adoption of standardized solutions to ensure secure and reliable cross-device data management in both the client and server side. There is already a proposed standardized Cloud Digital Safe on the server side storage that follows the AFNOR specification while there is no standardized solution yet on the client-side. This thesis is focused on two main areas : 1) the proposal of a standardized Client Digital Safe where user data are stored locally and 2) the synchronization of these data between the Client and the Cloud Digital Safe and between the different user devices. We contribute in this research area in three ways. First, we propose a Client Digital Safe based on HTML5 Local Storage APIs. We start by strengthening the security of these APIs to be used by our Client Digital Safe. Second, we propose an efficient synchronization protocol called SyncDS with minimum resource consumption that ensures the synchronization of user data between the Client and the Cloud Digital Safe. Finally, we address security concerns, in particular, the access control on data sharing following the Digital Safe requirements
Berrebi, Johanna. "Contribution à l'intégration d'une liaison avionique sans fil. L'ingénierie système appliquée à une problématique industrielle." Phd thesis, Ecole Polytechnique X, 2013. http://pastel.archives-ouvertes.fr/pastel-00800141.
Full textCardot, Anais. "Rejeu basé sur des règles de transformation de graphes." Thesis, 2019. http://www.theses.fr/2019POIT2254/document.
Full textIn many modelling fields, such as architecture, archaeology or CAD, performing many variations of the same model is an expanding need. But building all those variations manually takes time. It is therefore needed to use automatic technics to revaluate some parts of a model, or even an entire model, after the user specifies the modifications. Most of the existing approaches dedicated to revaluating models are based on a system called parametric modelling. It is made of two parts, a geometric model and a parametric specification, which allows to record the series of operation that created the model, and the different parameters of those operations. This way, the user can change some parameters, or edit the list of operations to modify the model. To do so, we use a system called persistent naming, introduced during the 90ies, that allows us to identify and match the entities of an initial specification and the ones of a revaluated specification. In this thesis, our goal is to propose a persistent naming system that would be general, homogeneous and that would allow the user to edit a parametric specification (which means move, add, or delete some operations). We base our system on the Jerboa library, which uses graph transformation rules. This way, we will be able to use those rules to create our naming system, while also linking the notions of graph transformation rules and parametric specification. We will then describe how to use our naming method to revaluate or edit parametric specifications. Finally, we will compare our method with the other ones from the literature
Bendakir, Narimel. "RARE : un système de recommandation de cours basé sur les régles d'association." Thèse, 2006. http://hdl.handle.net/1866/16732.
Full textBaseri, Yaser. "Contrôle d’Accès Sécurisé dans l’Info-Nuage Mobile (Secure Access Control in Mobile Cloud)." Thèse, 2017. http://hdl.handle.net/1866/20486.
Full text