Dissertations / Theses on the topic 'Informatique – Défauts'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Informatique – Défauts.'
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.
Risch, Vincent. "Les Tableaux analytiques au service des logiques de défauts." Aix-Marseille 2, 1993. http://www.theses.fr/1993AIX22051.
Full textBentoumi, Mohamed. "Outils pour la détection et la classification : Application au diagnostic de défauts de surface de rail." Nancy 1, 2004. http://www.theses.fr/2004NAN10159.
Full textThe works concern with detection and classification problems for fault diagnosis. Two approaches are treated. The first one, where the K-classes global problem is splitted into sub problems, is called simultaneous detection and classification. Each sub problem is solved by a block that links together pre-processing phase, choice of the representation space, detection then decision. The resolution of the global problem is carried out by a sequential arrangement of the blocks or a parallel decision scheme. The second approach is the successive detection and classification approach. It consits of a first basic signal processing for alarm generation that indicates the possible existence of default. Then, high-level processings are activated in order to precisely analyze the default signature. Classification tools - linear classifiers, neural classifiers, SVM - are used. All these methods have been validated on a rail surface defect detection application in subway context
Kaftandjian, Valérie. "Reconnaissance automatique de défauts dans les produits métalliques par radioscopie numérique." Lyon, INSA, 1995. http://www.theses.fr/1995ISAL0056.
Full textThe present study refers to the Non Destructive Testing of metal products by digital radioscopy. Internal defects are automatically detected and their type identified, while the object is moving. The first part o( our study is dedicated to the characterization of the radioscopic image quality : the Signal to Noise ratio, the contrast, the sensitivity, the thickness resolution are measured on the digital image. Theo, the X-ray detector is investigated through its Detection Quantum Efficiency and its Transfer Function. First, the Contrast Transfer Function is measured and compared to a numerical simulation which confirmed that the detector resolution is determined by its geometrical aperture. Finally, the Modulation Transfer Function 1s studied in detail through several experimental methods. In the second part, our work is aimed at the radioscopic image analysis, in such a way to detect and recognize the internal defects automatically. This is applied to the inspection of aluminum ingots, where eight defect types are to be considered. The analysis consists of three parts : defect detection, feature extraction, recognition. The developed detection method is based on mathematical morphology. A study of the defects and the production process allowed us to model the defect types, and to gather as much a priori information as possible. The selected features for defect modeling are shape, orientation, width, contrast. Finally, our recognition method is based on an identification tree, whose inputs are the features, and outputs are the defect types. Once the objects are separated in shape, orientation and width families, a clustering of aligned and close objects is performed, in such a way to identify the objects that would have been disconnected by the segmentation process
Hecht, Geoffrey. "Détection et analyse de l'impact des défauts de code dans les applications mobiles." Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10133/document.
Full textMobile applications are becoming complex software systems that must be developed quickly and evolve continuously to fit new user requirements and execution contexts. However, addressing these constraints may result in poor low-level design choices, known as code smells. The presence of code smells within software systems may incidentally degrade their quality and performance, and hinder their maintenance and evolution. Thus, it is important to know this smells but also to detect and correct them. While code smells are well-known in object-oriented applications, their study in mobile applications is still in their infancy. Moreover there is a lack of tools to detect and correct them. That is why we present a classification of 17 code smells that may appear in Android applications, as well as a tool to detect and correct code smells on Android. We apply and validate our approach on large amounts of applications (over 3000) in two studies evaluating the presence and evolution of the number of code smells in popular applications. In addition, we also present two approaches to assess the impact of the correction of code smells on performance and energy consumption. These approaches have allowed us to observe that the correction of code smells is beneficial in most cases
Abramik, Stanisław. "Base de connaissance des défauts des systèmes électriques pour systèmes experts : contribution à l'étude du diagnostic de défaillance des convertisseurs statiques en temps réel." Toulouse, INPT, 2003. http://www.theses.fr/2003INPT030H.
Full textJacquemod, Gilles. "Detection de défauts dans des images radiographiques industrielles obtenues en temps reel par des détecteurs linéaires." Lyon, INSA, 1989. http://www.theses.fr/1989ISAL0089.
Full textBonté, Eric. "Calcul des extensions dans les théories de défauts en réseau : Application au raisonnement à profondeur variable." Paris 13, 1992. http://www.theses.fr/1992PA132015.
Full textForget, Lionel. "X-logiques : théorèmes de représentation pour les logiques préférentielles et la logique des défauts." Aix-Marseille 1, 1999. http://www.theses.fr/1999AIX11009.
Full textCyr, Charles. "Développement d'algorithmes de détection de défauts pour la maintenance prédictive de générateurs hydroélectriques." Thesis, Université Laval, 2004. http://www.theses.ulaval.ca/2004/21577/21577.pdf.
Full textDeguillemont, Denis. "Contribution de l'application de la transformation en ondelettes pour détecter et reconnaître des défauts sur des objets de nature complexe en production industrielle." Lille 1, 1998. http://www.theses.fr/1998LIL10131.
Full textSabatier, Philippe. "Contrôle catastrophique d'un système a base de connaissances. : Conception, modélisation et application au diagnostic des défauts des fromages." Lyon, INSA, 1994. http://www.theses.fr/1994ISAL0055.
Full textIn AI research, we develop symbolic systems, which use representations considered as knowledge. In so doing we have confused the representations with the phenomenon we are modelling. But the map is not the territory. Today we consider that this confusion is the origin of the control problem of the knowledge bases. So we want emphasize the perspective that a knowledge representation is just that. Consequently we must develop inside each AI program a model of some system int he world. Our goal is to prove that Catastrophe Theory (CT) is a good way of modelling processes qualitatively (chapter 1). The use of CT allows describe the evolution of physical structures morphologically, and this is important for the biological diagnosis (chapter 2). The mathematical topic of the CT is exposed as a part of the system theory (chapter 3). To achieve one's end, we investigate an application of the CT to the cheese's « pathology » (chapter 4). Pathology can takes various forms and should be determined not by some external standard but to the individual in identical or varying circumstances. The evolution of the cheese, and the defects, may be determined from its localisation i a gradient : acidity, salt,… In all cases a frontier can be observed, which limits two zones, one with and another without the defect. This model is an internal representation of the physical reality. This is a vehicle that permit the interpretation of the phenomenological diversity (chapter 5)
Diouf, Saliou. "Contribution au diagnostic industriel de défauts de roulement et de balourd par techniques neuronales : application à la machine asynchrone à cage." Paris 12, 2007. http://www.theses.fr/2007PA120024.
Full textThis thesis is focused on vibratory signatures based faults detection and classification aiming computer aided diagnosis of defects relative to umbalanced rotation or dealing with rolling bearing's external ring's chipping defects appearing in turning machines. It relates the use of neural techniques for detection and computer aided diagnosis of such defects. We develop two approaches starting from the vibratory signatures obtained for the first one from a current sensor, and for the other one from an accelometer-like sensor. These signatures are initially pre-processed by PCA and after that they are classified by an artificial neural network (ANN). We use two types of neural classifiers, known for their classification capabilities : RBF and LVQ. The validation of these two approaches is carried out through the study of the pre-processing of the signatures and the parameters of different ANNs, using an experimental asynchronous electromotor
Casimir, Roland. "Diagnostic des défauts des machines asynchrones par reconnaissance des formes." Phd thesis, Ecole Centrale de Lyon, 2003. http://tel.archives-ouvertes.fr/tel-00139706.
Full textVu, Hien Duc. "Adaptation des méthodes d'apprentissage automatique pour la détection de défauts d'arc électriques." Electronic Thesis or Diss., Université de Lorraine, 2019. http://docnum.univ-lorraine.fr/ulprive/DDOC_T_2019_0152_VU.pdf.
Full textThe detection of electric arcs occurring in an electrical network by machine learning approaches represents the heart of the work presented in this thesis. The problem was first considered as a classification of fixed-size time series with two classes: normal and default. This first part is based on the work of the literature where the detection algorithms are organized mainly on a step of the transformation of the signals acquired on the network, followed by a step of extraction of descriptive characteristics and finally a step of decision. The multi-criteria approach adopted here aims to respond to systematic classification errors. A methodology for selecting the best combinations, transformation, and descriptors has been proposed by using learning solutions. As the development of relevant descriptors is always difficult, differents solutions offered by deep learning has also been studied. In a second phase, the study focused on the variable aspects in time of the fault detection. Two statistical decision paths have been explored, one based on the sequential probabilistic test (SPRT) and the other based on artificial neural networks LSTM (Long Short Time Memory Network). Each of these two methods exploits in its way the duration a first classification step between 0 and 1 (normal, default). The decision by SPRT uses an integration of the initial classification. LSTM learns to classify data with variable time. The results of the LSTM network are very promising, but there are a few things to explore. All of this work is based on experiments with the most complete and broadest possible data on the field of 230V alternative networks in a domestic and industrial context. The accuracy obtained is close to 100% in the majority of situations
Moha, Naouel. "DECOR : Détection et correction des défauts dans les systèmes orientés objet." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2008. http://tel.archives-ouvertes.fr/tel-00321081.
Full textDes techniques et outils ont été proposés dans la littérature à la fois pour la détection et la correction des défauts. Les techniques de détection proposées consistent principalement à définir des règles pour détecter les défauts et à les appliquer sur le code source d'un système. Quant aux techniques de correction, elles consistent à appliquer de façon automatique des refactorisations dans le code source du système analysé afin de le restructurer de manière à corriger les défauts. Cependant, la phase qui consiste à identifier les restructurations est réalisée manuellement par les
ingénieurs logiciels. Ainsi, il n'est pas possible de corriger
directement et automatiquement les défauts détectés. Ce problème est dû au fait que la détection et la correction des défauts sont traitées de façon isolée.
Ainsi, nous proposons DECOR, une méthode qui englobe et définit toutes les étapes nécessaires pour la détection et la correction des défauts de code et de conception. Cette méthode permet de spécifier des règles de détection à un haut niveau d'abstraction et de suggérer des restructurations de code afin d'automatiser la correction des défauts.
Nous appliquons et validons notre méthode sur des systèmes libres orientés objet afin de montrer que notre méthode permet une détection précise et une correction adaptée des défauts.
Bouhours, Cédric. "Détection, Explications et Restructuration de défauts de conception : les patrons abîmés." Phd thesis, Toulouse 3, 2010. http://tel.archives-ouvertes.fr/tel-00457623.
Full textNoura, Hassan. "Contribution à la commande des systèmes : diagnostic et commande des systèmes soumis à des défauts." Nancy 1, 1993. http://www.theses.fr/1993NAN10208.
Full textFirmin, Christian. "Optimisation des réseaux de neurones à fonctions radiales de base par critères informationnels : application à la détection de défauts en production de bouteilles." Lille 1, 1997. http://www.theses.fr/1997LIL10048.
Full textBenoist, Bruno. "Système expert d'analyse automatique des signaux de défauts lors du contrôle non destructif des générateurs de vapeur des centrales nucléaires." Compiègne, 1990. http://www.theses.fr/1990COMPD304.
Full textAutomatic analysis, by computer, of defect signals in steam generator tubes, based on Eddy current multifrequency technique, is must often inefficient due to "pilgrim noise". The first step is to use a method that allows us to eliminate the noise : the adaptative interpolation. Thanks to this method, which ensures reliable data on each channel, the analysis can be realised by taking into account the data corresponding to each basic or mixed channel. By correlating these diverse data, we can class the signals according to three types of defects : single defects (symmetrical), multiple defects (several in the same place). The second step is to use an expert system which allows a reliable diagnosis for whatever family the defect belongs to. According to this classification, analysis is continued and results in the characterisation of the defect. The expert system has already been developed with the general purpose application expert system shell SUPER, which is briefly described. The knowledge base (SOCRATE) and the specific tools developed for this application are thoroughly described. The first results obtained with signals corresponding to real defects, that have been recorded in different places, are presented and discussed. The expert system is revealed efficient in aIl the studied cases, even with signals obtained in very noisy environments
Sallem, Fatma. "Détection et isolation de défauts actionneurs basées sur un modèle de l'organe de commande." Phd thesis, Université Paul Sabatier - Toulouse III, 2013. http://tel.archives-ouvertes.fr/tel-00865848.
Full textTounsi, Mohamed. "Cyclage actif en mode MLI des modules de puissance IGBT application des SVM pour le diagnostic des défauts de vieillissement thermique." Caen, 2011. http://www.theses.fr/2011CAEN2082.
Full textThe work concerns the implementation of an SVM classifier (Support Vector Machine) for an automatic diagnosis of power IGBT module thermal ageing defects, based on thermoelectric measurements. This has required the development of an active cycling test bench for the characterization of the thermal tiredness of IGBT modules, the analysis of the degradations generated in their assembly, and the classification of their operating mode : in " healthy " mode and " degraded " mode. The strategy adopted for the thermal cycling, aims to reproduce the constraints imposed on the components during their real operation. We followed the module static and dynamic parameters in order to detect possible drifts, and thermoelectric parameters used as criteria of thermo-mechanical degradations in their assembly. The thermal stress caused mainly, earlier switching-on and switching-off modes and appearance of a tile-current. A failure analysis has revealed, in the component assembly, bond-wire lift-off, solder layer degradation, and surface dislocation of the gate and emitter metallizations. The implemented SVM classifier, allows to distinguish between an IGBT degraded operating mode and a healthy operating mode, for a preventive maintenance in electric installations where the continuity of service and safety are very important
González-Contreras, Brian Manuel. "Contribution à la tolérance aux défauts des systèmes linéaires : Synthèse de méthodes d'accommodation fondée sur l'information du second ordre." Thesis, Nancy 1, 2009. http://www.theses.fr/2009NAN10008/document.
Full textThis thesis is devoted to the synthesis of accommodation methods founded on the second order information (SOI) assignment in the context of fault tolerance for deterministic linear systems. The major contribution of this research concerns using this information in the reconfigurability analysis (capability of the system to respond to faults) and developing strategies for fault accommodation in order to recover nominal performances in terms of system dynamics and also to guarantee the assigned second order information. Firstly, approaches for measuring the SOI using the system's input/output data are proposed. A first approach based on the initial response is considered. An interesting alternative to this approach, in considering the problem as one of identification, is proposed as an indirect computation of the SOI but online and using input/output data. An index based on reconfigurability, which is directly related to the SOI, is also proposed. Based on this online SOI computation, the index is applied to networked control systems affected by network induced delays in order to calculate their impact over the system. Secondly, fault accommodation strategies for loss of effectiveness type faults are proposed under the feedback SOI synthesis. SISO systems are first considered, approach founded on the modified pseudo inverse method. On the other hand, a strategy for MIMO systems founded on the pseudo inverse method is taken into account. Examples illustrating the application of the approaches are also presented. All these developed approaches are applied and illustrated through the well known process benchmark: the three tank hydraulic system. The simulations show up and notice the results obtained, and bring out the contribution of the developed approaches
Shamsi, Nejad Mohammad Ali. "Architectures d'alimentation et de commande d'actionneurs tolérants aux défauts : régulateur de courant non linéaire à large bande passante." Thesis, Vandoeuvre-les-Nancy, INPL, 2007. http://www.theses.fr/2007INPL046N/document.
Full textThe objective of this work is to study different structures of supply and control of actuators for the embarked systems, motorized by permanent magnet synchronous machines (PMSM) supplied with voltage source inverters (VSI). By analyzing the advantages and the disadvantages of these structures, we choose two power segmented structures: a three-phase PMSM supplied with two three-leg VSI and a double-star PMSM supplied by two three-leg VSI. The other aspect is the fault detection and the reconfiguration of the supply and command to make possible the operation in degraded mode. For the open-circuit or short-circuit fault of an inverter switch, one or several strategies in degraded operating mode are proposed. The simulation and experimentation results validate the effectiveness of the proposed methods. The last part of this work is devoted to the study and modeling of a large band-width hybrid current controller named "modulated hysteresis". The performances of this controller are compared with the other regulators
Andrea, Jonathan. "Génération, modélisation et détection des défauts d'arc électrique : application aux systèmes embarqués aéronautiques." Thesis, Nancy 1, 2011. http://www.theses.fr/2011NAN10017/document.
Full textThe arcs have been the source of several aircraft accidents by fire or serious destruction of the electrical fundamentals (eg actuators, top panel ,...) (TWA 800 in 1996 and Swiss Air Flight 11 1998). Prevent and detect constitute a significant improvement in aviation safety. Thus, although the problem has been known for a while, detection systems for avionics are very recent. In an aircraft, these defects are caused by the aging and the deterioration of the cables. To study and propose a system for arc detection, the following three phases were followed chronologically in this thesis: first, the developement of a test bench to characterize the main types of arcs (AC and DC). This system can produce arcs controlled intensity and duration depending on the mode serial, parallel, DC and AC. The initiation of arcs is obtained by overvoltage. This surge is produced by inductive switching. Second, the search for reliable electric models explaining the phenomena of arcs. We therefore propose a meta model combining the characteristics of different behavioral models and impedancemetry.This combination significantly improves the overall performance of the modeling in the circuits. Thirdly, the study of detection algorithms for implementation on an electronic circuit, small enough to be implanted in a circuit breaker.We propose and test different methods of detection
Homssi, Louay. "Outils de modélisation, identification et traitement du signal pour le diagnostic des défauts dans les systèmes continus." Toulouse, INSA, 1992. http://www.theses.fr/1992ISAT0017.
Full textMtimet, Habib. "Analyse fréquentielle d'image pour une résolution subpixel : Etude d'un système nanolidar appliqué à la localisation des défauts dans les composants III-V." Montpellier 2, 1994. http://www.theses.fr/1994MON20018.
Full textSoldani, Siegfried. "Vers le diagnostic embarqué de défaillances dans les systèmes à événements discrets : application au domaine automobile." Toulouse 3, 2008. https://hal.laas.fr/tel-01591892.
Full textThis thesis deals with the diagnosis of behavior functions in the embedded network systems and especially in the automotive systems. It is focused on the localization of intermittent faults in discrete events systems. The proposed methods are based on a discrete events model representing the normal functioning of the observable behavior of the monitored system. The detection step consists of a comparison between the flow of observable events emitted by the monitored system and the flow foreseen by the model. A localization mechanism, based on the diagnoser approach (Sampath et al. , 1998), points out the set of events potentially responsible for the faults. The aim of this approach is to get the diagnosis information into a data structure (off-line) which efficiently joins the observations with the faults during the on-line diagnosis. This thesis have lead to the building of many applications for the intermittent fault diagnosis. This works have been validated on a test bench representing the network used in the city transports (bus, subways. . . )
Sene, Moustapha. "Contribution à la mise au point de techniques de détection et de diagnostic de défauts de roulements basées sur la tranformée en ondelettes et les réseaux de neurones artificiels." Paris 12, 2007. http://www.theses.fr/2007PA120075.
Full textThis thesis concerns faults diagnosis based on analysis of mechanical vibrations. Detection and faults diagnosis are focused. Two computational techniques are used : wavelet transform for defect's detection and artificial neural networks for classification and diagnosis. Application of the studies concepts leads to early detection of unbalanced force defects and flaking paths defects of the outer ring of bearing. In this frame, classification approaches based on probability estimation and neural techniques (RBF, LVQ) have been implemented. Validation of these approaches is achieved using indicators (kurtosis, crest factor, root mean square, etc. ) extracted from the same vibratory signatures
Rocha, barbosa Cassandra. "Coordination et ordonnancement de tâches à grains fins entre environnements d'exécution HPC." Electronic Thesis or Diss., Reims, 2023. http://www.theses.fr/2023REIMS016.
Full textSupercomputers are becoming more and more complex to use. This is why the use of so-called hybrid programming models, MPI + X, are being implemented in applications. These new types of models allow a more efficient use of a supercomputer, but also create new problems during the execution of applications. These problems are of different types.More specifically, we will study three problems related to MPI + X programming. The progression of non-blocking MPI communications within the X environment. Then two types of possible imbalance in MPI+X applications. The first being between MPI processes and the second within an MPI process, i.e., imbalance within X.A solution in the case of an X environment in recursive tasks will first be presented for the MPI communication progress problem using progress task insertion in the X environment. For the imbalance between MPI processes, a solution for resource rebalancing within a node will be presented. Finally, for the imbalance in the X environment, a solution to use the imbalance to run a second application will also be presented
Jalabert, Antoine. "Architecture de mémoire haute densité à base d'électronique moléculaire tolérante à un très grand nombre de défauts." Phd thesis, Télécom ParisTech, 2006. http://pastel.archives-ouvertes.fr/pastel-00002014.
Full textLin-Kwong-Chon, Christophe. "Approches neuronales adaptatives pour le contrôle tolérant aux défauts de systèmes pile à combustible." Thesis, La Réunion, 2020. http://www.theses.fr/2020LARE0008.
Full textThe proton exchange membrane fuel cell is a promising electrochemical converter for production of electricity from the decarbonated hydrogen carrier. However, some technological challenges limit its deployment, such as durability, reliability or financial cost. The active fault-tolerant control strategy is one of the solutions to mitigate any system fault according to three actions: diagnosis, decision and control. This study proposes to develop a generic controller module adaptive to health states through neural networks. Dynamic programming controller, reinforcement learning, and echo-state models are combined for the design of the adaptive controller. This controller employs three neural models with specific roles: an actor, a predictor and a critic. Flooding and membrane drying faults are considered in this study. The proposed controller was able to demonstrate interesting capabilities on a simulation fuel cell model in multi-variable regulation for oxygen stoichiometry, membrane pressure difference and temperature. The results show superior performance of the proposed controller compared to a proportional integral derivative controller. Stability analyses were conducted to prove the continuity of the adaptive controller. The controller has been validated experimentally on a single cell test-bench. The configuration of the test-bench imposed constraints specific to an on-line and real-time application. The generic nature of the controller offers the possibility to switch from one configuration to another without having to design another controller. Several tests are carried out for regulation of the zero-pressure difference at the membrane. The controller was validated on the occurrence of flooding and membrane dryness faults, including actuator and water purging disturbances. The approach and the generic controller adaptive to the states of health proposed in this thesis allow to satisfy control requirements regarding the fault-tolerant control strategy. The first interest lies in the compensation of the multilateral effects of faults that lead to unwanted dynamic changes. Another interest is to be able to modify in situ operating conditions, components or even auxiliaries while being able to ensure a stable and optimal control
Boutayeb, Mohamed. "Contribution à l'estimation de l'état et des paramètres des systèmes singuliers : application au diagnostic." Nancy 1, 1992. http://www.theses.fr/1992NAN10004.
Full textDieng, Abdoulaye. "Modélisation dynamique et commande d'un ensemble « génératrice synchrone pentaphasée à FEM non sinusoïdale – convertisseur AC/DC » tolérant aux défauts." Nantes, 2014. http://archive.bu.univ-nantes.fr/pollux/show.action?id=f111b637-afbe-49f3-b3a3-033b65a51a64.
Full textIn the context of exploitation of marine renewable energy sources, the access difficulties to the installations requires to explore fault-tolerant energy conversion systems. This Doctorate thesis deals with the dynamical modeling and control of an innovative energy conversion chain. The studied conversion chain consists of a non-sinusoidal EMF 5-phase permanent magnet synchronous generator which delivers the energy to the DC bus via an AC/DC converter. Two AC/DC topologies are investigated: a 5-phase PWM rectifier and a 5-phase VIENNA rectifier. The energy transfer optimization needs optimal control strategies using all the EMF harmonics in normal mode and in specific fault mode due to the opening of one phase of the generator. Specific control algorithms are developed based on the possible connection between the neutral of the machine and the midpoint of the DC bus. Particular profiles of the current references imposed by the control strategies require robust and high dynamic performance controllers. For the inner current loops, the linear PIα fractional controller and a specific non-linear phase-shift self-oscillating controller are synthesized. For the outer voltage loop, control strategies are developed and regulators are synthetized. Software and experiment test benches are built. Results illustrate the working and performances of the control strategies developed for different scenarii with both converters
Sidibé, Yaya Yannick. "Aide à la décision pour la détection et l’analyse des défauts de surface dans les structures immergées." Thesis, Le Havre, 2015. http://www.theses.fr/2015LEHA0006/document.
Full textThis study concerns the damages detection and diagnosis for immersed structure. The structures are metallic plates. The proposed method focuses on the analysis of ultrasonic acoustic measurements obtained by submarine echography. It combines signal processing tools and Gaussian neural networks for classification purpose. Methods with and without reference models are proposed. The usual detection technics with contact are not applicable for the considered systems like stream turbines. This research consists to use a single and a single transducer under different incidence angles opposed to others technics using numerous sensors and their accurate location. The present research use Lamb wave according to their sensibility to the structural damages. The different stages are the following : - 1. Experimental setup for Lamb wave generation and acquisition. - 2. Study of the Lamb wave processing on immersed structures, in particular in metallic plate immersed in water. - 3 .Signal characterization for different types of damages. - 4. Estimation of the angle and lift-off distance
Maazi, Mostafa. "Conception et réalisation de systèmes en gamme millimétrique pour l'évaluation non destructive : application à la restauration de profils et d'images d'objets enfouis." Lille 1, 2005. https://pepite-depot.univ-lille.fr/RESTREINT/Th_Num/2005/50376-2005-9.pdf.
Full textToma, Samuel. "Méthodologie de détection et d'identification des défauts multiples dans les systèmes complexes à partir d'évènements discrets et de réseaux de neurones : applications aux aérogénérateurs." Thesis, Corte, 2014. http://www.theses.fr/2014CORT0006/document.
Full textThis thesis deals with the time-domain analysis of the electrical machines fault diagnosis due to early short-circuits detection in both stator and rotor windings. It also introduces to the Discrete EVent system Specification (DEVS) a generic solution to enable concurrent and comparative simulations (CCS). The DEVS-based CCS is an extension introduced using an aspect-oriented programming (AOP) to interact with the classic DEVS simulator. A new DEVS-based artificial neural network (ANN) is also introduced with a separation between learning and calculation models. The DEVS-based CCS is validated on the proposed ANN DEVS library inside the DEVSimPy environment. The concurrent ANN contributes in the time-domains analysis for the electrical machine fault diagnosis. This new method is based on data coming directly from the sensors without any computation but with a new dedicated preprocessing technique. Later, some enhancements are brought to the artificial neural network based on a new multistage architecture reducing the training time and errors compared to the single ANN. The new architecture and techniques has been validated on real data sixteen non-destructive windings faults analysis and localization
Gassara, Hamdi. "Analyse, commande et observation des modèles flous de type Takagi-Sugeno à état retardé." Amiens, 2011. http://www.theses.fr/2011AMIE0124.
Full textLallbeeharry, Navish. "Réseaux d'énergie embarqués dans les systèmes de transport : communication et diagnostic passif par Courant Porteur en Ligne (CPL)." Thesis, Lille 1, 2018. http://www.theses.fr/2018LIL1I070/document.
Full textThe research presented in this thesis focuses firstly on the reliability of PLC technology and secondly on the diagnosis of wired networks using PLC in the aeronautical sector. First of all, we studied the OFDM synchronization in a PLC communication chain for data transmission on avionic systems. The objective of the study is to make the PLC link more reliable by respecting the latency constraints set in the aeronautical domain. We introduced sampling frequency errors in order to analyze the effects of sampling frequency desynchronization on the bit error rate. We then proposed an optimized algorithm for error estimation. The second contribution of the thesis is devoted to the monitoring of embedded wired networks and the objective being to use PLC communication as a mean of detecting a fault. We focused our study on the detection of a fault following an electric arc, modeled by a parallel resistance, or a connection fault, modeled by a series resistance on a wire of the two-wired cable. In order to have the channel characteristics with and without faults, a test bench of a multi-branch harness has been designed with different resistive fault samples. We proposed several correlation-based fault detection coefficients, Euclidean distance, and a blind algorithm based on received symbols. In order to avoid false alarms or no detection, detection thresholds were discussed. The different detection coefficients and the thresholds were compared and tested in the presence of white noise in terms of the transmitted power and the detection rate
Landon, Yann. "Contribution à l'amélioration de la qualité d'usinage en fraisage 3 axes sur machines-outils à commande numérique par la mise en place d'un compagnon virtuel." Toulouse 3, 2003. http://www.theses.fr/2003TOU30150.
Full textCheik, Njifon Ibrahim. "Modélisation des modifications structurales, électroniques et thermodynamiques induites par les défauts ponctuels dans les oxydes mixtes à base d'actinides (U,Pu)O2." Thesis, Aix-Marseille, 2018. http://www.theses.fr/2018AIXM0356/document.
Full text(U,Pu)O2 (commonly called MOX) is currently used as nuclear fuel in pressurized water reactors with a Pu content of around 10 wt.%, and is envisaged as the reference fuel in Generation IV sodium fast reactors (SFR) with a Pu content of around 25 wt.%. Under operation, (U,Pu)O2 is submitted to fission reactions which generate a large quantity and variety of point defects, as well as fission products. By migrating, point defects and gaseous fission products can aggregate into nano-voids, dislocations and fission gas bubbles, which lead to the modification of the fuel microstructure. Therefore, a better description of the fuel behaviour at the atomic scale, and especially of the elementary mechanisms involved in the diffusion of point defects and fission products, is necessary to refine the models used in the fuel performance codes used to simulate the behaviour of fuels at the macroscopic scale. We use electronic structure calculations based on the DFT+U method combined with the occupation matrix control scheme (OMC) to investigate (U,Pu)O2 properties for various Pu contents. Static energy minimizations and ab initio molecular dynamics were used. We have first determined bulk structural, electronic and thermodynamics properties of (U,Pu)O2. We then studied the stability of point defects in (U,Pu)O2 and (U,Ce)O2, as well as the structural and electronic modifications induced by these point defects, in (U,Pu)O2 and the common experimental surrogate (U,Ce)O2. Finally, the fission gas (Kr and Xe) and helium (He) trapping and solubility in (U,Pu)O2 matrix are investigated
Langeron, Yves. "Modélisation stochastique pour la sûreté de fonctionnement des systèmes commandés." Thesis, Troyes, 2015. http://www.theses.fr/2015TROY0001/document.
Full textIn the context of control systems, the research effort is focused on how to reconfigure the control law upon the occurrence of a faulty situation. The reconfiguration procedure aims to overcome the fault and thus to maintain system performances. The main issue of this thesis is to study these systems in terms of their dependability by questioning the causes that generate a fault. Then it is assumed a close relationship between the control of a system, its degradation and its faults. A stochastic modelling framework is proposed combining the use of the system and the various modes of deterioration. The actuator is assumed to be the most critical part of a system. The prognosis of its remaining life RUL is derived from these models. This RUL is then used as a tool for reconfiguring the LQR law (Linear Quadratic Regulator) of a system with a single actuator in the context of a predictive maintenance. The impact of this new maintenance policy on static and dynamic performances is assessed. Finally the stochastic behavior of a fault tolerant control system is studied by means of the achieved models
Bacchus, Alexandre. "Représentativité de la modélisation aux éléments finis pour le diagnostic de machines synchrones de grande puissance." Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10014/document.
Full textAn identification method of rotor inter-turn short-circuits and static eccentricities of an operating turboalternator is developed. This approach provides the type of the machine functional state such as its seriousness and location. The electromotive force obtained from a radial flux sensor is considered in order to identify the machine faults. Some previous works have shown that learning methods are efficient to characterize precisely the fault of a machine. Thus, a fault signatures data set, i.e. prototypes matrix, is built thanks to simulations using finite element method applied to the machine considering a great number of functional states. Therefore, the goal of this work is to identify the class (fault) of experimental measurements using the simulated output. To do that, a small scale alternator is modelled and simulation outputs are compared to the experimental measurements. The application of supervised classification method chosen beyond the shape of data shows that a good classification rate of 79 % for short-circuits and 93 % for eccentricities can be achieved thanks to specific features and simulation output treatments. An identification rule of the fault type is designed by considering a hierarchical classification approach. It achieves excellent results since all experimental measurements are assigned to the right type of fault. Finally, the use of a finite state automaton achieves better results for short-circuits identification by taking into account the temporal evolution of the machine functional state
Sid, Mohamed Amine. "Co-conception diagnostic et ordonnancement des mesures dans un système contrôlé en réseau." Thesis, Université de Lorraine, 2014. http://www.theses.fr/2014LORR0003/document.
Full textThe works developed in this thesis deal with 'fault diagnosis and sensor scheduling co-design' in networked control system. This multidisciplinary subject requires theoretical knowledge in both fault diagnosis and communication networks. Our contribution consists in developing a co-design approach that integrates in the same framework the characteristics of fault diagnosis performance and real time sensor scheduling. The main benefit of this approach is minimizing the required network resources for attending acceptable fault diagnosis performances. We are interested in the development of more efficient and more adapted for real time implementation fault diagnosis algorithms while taking into account different types of communication constraints. In conjunction with these algorithms, two sets of sensor scheduling techniques are used : - Off-line scheduling - On-line scheduling (event triggered sampling) For off-line scheduling, the communication sequences are designed before the implementation of the diagnostic algorithm. In this context, we proposed several techniques for scheduling with different spatial and temporal complexity and adapted to different operating condition for the detection and the isolation of faults based on the information provided by the selected communication sequences. Moreover, we deal also with on-line scheduling techniques based on the event triggered sampling developed during the last decade. In This approach measurement packets are transmitted if and only if the transmission condition (event) is verified. This saves resources provided by the network while maintaining acceptable performance of fault diagnosis. The objective of these algorithms is to minimize the number of transmitted information which means less energy consumption and has a major importance in wireless networked control systems
Lelong, Adrien. "Méthodes de diagnostic filaire embarqué pour des réseaux complexes." Thesis, Lille 1, 2010. http://www.theses.fr/2010LIL10121/document.
Full textResearch works presented in this thesis rely on on line diagnosis of wire networks. It consists indetecting and locating intermittent or permanent electrical faults, on a system's network while this system is running. Such a diagnosis is based on the principle of reflectometry which is used for off line diagnosis until then. The aim is the analysis and improvement of reflectometry methods and the implementation of related processing in order to automate and to embed it in the target system for a real time execution. The first contribution refers to the use of multicarrier signals so as to minimize interferences between the running target system and the reflectometry module. Pulse deconvolution algorithms are required for this purpose. These algorithms are also used for high resolution processing described subsequently. A low computational cost semi-blind deconvolution method is proposed among others. Distributed reflectometry, consisting in the simultaneous injection of signals at several points of the network, is then studied. An innovative filtering method called "selective average" is proposed as a solution to the problem of interferences due to the simultaneous injection of the modules. Finally several considerations on the implementation and automation are studied. An innovative intermittent fault detection algorithm for noisy environment is also proposed
Asnaashari, Ahmad. "Modélisation de la défaillance des réseaux d'eau : approches statistique, réseau de neurones et survie." Lille 1, 2007. https://pepite-depot.univ-lille.fr/LIBRE/Th_Num/2007/50376-2007-Asnaashari.pdf.
Full textAguiar, Cota Braulio Norberto. "Pilotage automatique robuste de train sous contrainte d'enrayage-patinage." Thesis, Valenciennes, 2019. http://www.theses.fr/2019VALE0022.
Full textMillions of people use the trains every day. Therefore, rail transport and infrastructure efficiency and safety is critical, for customers and companies. A major challenge nowadays is to climb automation levels for trains, from Grade of Automation (GoA) “0” which is basically on-sight train operation all the way to GoA “4” where train operations are unattended. GoA levels are defined by the International Association of Public Transport (UITP—French acronym), and aim to provide a roadmap for the development and the integration of the so-called automatic train operation (ATO) systems. ATO are operational safety enhancement devices that help the driver by automating some operations on the train, helping him/her to be more attentive and to focus on possible safety issues or unexpected situations. Currently, the level of automation that is reached for commercial rolling stock is GoA “2”, semi-automatic train operation, when starting and stopping operations are automated. Most of the existing ATO systems are GoA 2. As within any modern vehicle, on-board systems on a train are linked, and the ATO is working together with automatic train protection (ATP) and automatic train supervision (ATS), to ensure the respect of the speed restrictions and stop the train at the station with accuracy and within an acceptable tolerance of its timetable. The stopping task is carried out by automatic train stop control (ATSC), and all of the devices together form a package called automatic train control (ATC) (Dong et al. 2010). All automatic systems rely on two main functions: perception and decision. The ATO device will rely heavily on the available information to ensure an accurate perception of its environment and of the operational situation, in order to carry out the right decisions. This information acquired by sensors of different technologies and made available through the communication network (bus) of the train. Obviously, sensors are limited by technology, the communication bandwidth is not infinite; and all technical systems can experience faults and failures. Those are major challenges to design efficient and robust ATO devices, because the usual way to deal with these issues is to use sensors of different technologies for each information of interest. This makes such a system more complex, possibly costly, and it increases the amount heavily of transmitted data and its supporting infrastructure. There is a promising solution to this challenge, and it is called virtual sensors, or observers that are developed by automatic control researchers to supplement the sensors at a fraction of the cost, and embeds knowledge of the system in the automated device through analytical models of the environment. The design of a particular type of observer, and its practical exploitation for automatic train operations is the main contribution of this PhD work
Khelassi, Ahmed. "Nouvelle méthodologie de synthèse de lois de commande tolérante aux fautes garantissant la fiabilité des systèmes." Phd thesis, Université Henri Poincaré - Nancy I, 2011. http://tel.archives-ouvertes.fr/tel-00618655.
Full textBaghli, Lotfi. "Contribution à la commande de la machine asynchrone, utilisation de la logique floue, des réseaux de neurones et des algorithmes génétiques." Phd thesis, Université Henri Poincaré - Nancy I, 1999. http://tel.archives-ouvertes.fr/tel-00356891.
Full textBottcher, Claudia. "Un cadre pour guider la localisation automatique de défauts en utilisant des préférences stratégiques locales." Paris 13, 1996. http://www.theses.fr/1996PA132037.
Full textChabi, Djaafar. "Architectures de circuits nanoélectroniques neuro-inspirée." Phd thesis, Université Paris Sud - Paris XI, 2012. http://tel.archives-ouvertes.fr/tel-00679300.
Full text