Proteger as linhas de transmissão é uma das tarefas mais importantes dentro dos sistemas elétricos de potência. Faltas em linhas de transmissão devem ser localizadas precisamente e extintas o mais rápido possível. Para tal, o esquema de proteção de linhas utiliza valores amostrados de correntes e tensões para a execução das tarefas de detecção, classificação e localização da falta. Neste esquema, grandezas trifásicas de corrente (IA, IB, IC) e tensão (VA, VB, VC) compõem as entradas do sistema. Após a detecção e classificação da falta, o relé efetua o cálculo da impedância aparente para a verificação da zona de proteção na qual a falta se insere (localização). Dentro deste contexto, a rápida detecção e a correta classificação da falta são passos fundamentais para a lógica de controle de um relé. Para a utilização de sistemas de proteção com alta velocidade de operação, o conjunto detector + classificador deve realizar uma decisão precisa do tipo de falta envolvida em menos de 10 ms após a ocorrência desta. Alguns métodos convencionais têm lidado com este problema, porém os tempos para estimação do tipo de falta são algumas vezes excessivamente longos. Este trabalho apresenta um novo sistema que provê uma rápida e confiável detecção e classificação de faltas através das medidas de valores de correntes trifásicas. O novo método utiliza-se da teoria de Redes Neurais Artificiais, baseada em dois diferentes tipos de redes (MLP e RBF), para a tarefa de detecção e classificação de faltas nos níveis de tempo requeridos para um moderno sistema de proteção. Um estudo comparativo em relação ao desempenho das redes mencionadas também foi realizado. Os testes efetuados para as redes dos tipos MLP e RBF mostraram que o sistema proposto foi capaz de detectar e classificar corretamente 100% dos casos estudados. Deve ainda ser ressaltado, que na maior parte dos casos (93% para a rede MLP e 84% para a rede RBF), o processo de detecção e classificação foi completado com no máximo 5 amostras de pós-falta (5ms). Isto demonstra a rapidez na tarefa de detecção e classificação embutida no método proposto, principalmente levando-se em consideração os tempos apresentados pelos métodos convencionais. / Transmission line protection is one of the major tasks for a power system. Transmission line faults must be located accurately and isolated as fast as possible. In order to perform this task, the power system protection system uses the three-phase currents (IA, IB, IC) and voltages (VA, VB, VC) to detect, classify and locate the fault. After detecting and classifying the fault, the relay calculates the apparent impedance to verify in which protection zone the fault is located. Taking this into account, precise and fast detection and classification methods are fundamental steps for the relay control algorithm. The combination detection + classification must carry out the correct response in less than 10 ms after the fault for a high-speed protection system. Some conventional methods have treated this problem but the time for a correct classification is sometimes excessively long. This work presents a fast and reliable new system for fault detection and classification using the three-phase current measurements. This new system is based on Artificial Neural Networks (RBF and MLP) for the detection and classification tasks. A comparative study involving both types of neural networks was done. Tests showed that the proposed system was able to correctly detect and classify 100% of the studied cases where the majority (93% of the cases for MLP net and 84% for RBF net) of them was done in up to 5 post-fault samples (5 ms). The afore-mentioned demonstrates the high speed of the new method for the detection and classification tasks when compared to the conventional ones.

O trabalho proposto tem por objetivo avaliar a viabilidade de localização de falhas elétricas, por meio da análise de assinaturas magnéticas resultantes do sistema elétrico. Assim, diversos algoritmos - descritos na literatura científica atual e aplicáveis ao processamento de sinais conduzidos - foram analisados com o objetivo de verificar a possibilidade de complementar as metodologias já desenvolvidas para a detecção e classificação de falhas, de uma maneira não invasiva e via análise de assinaturas magnéticas resultantes. Diferentemente do processo de monitoração tradicional, cujos sensores precisam estar fisicamente conectados aos circuitos analisados, propõe-se estudar a viabilidade da análise dos sinais provenientes da assinatura magnética dos distúrbios para localizar falhas. Ressalta-se, assim, a característica não invasiva desse processo. A avaliação da metodologia proposta foi realizada através da escolha de casos de sensibilidade, cujos resultados teóricos e provenientes de simulações foram utilizados na análise de sua viabilidade. / The proposed study aims to evaluate the feasibility of determinig the fault location in electrical systems, by analyzing the resulting magnetic signatures. Thus, several algorithms - described in the current scientific literature and for the processing of conducted signals - will be analyzed in order to verify the possibility of complementing the methodologies already developed for fault detection and classification, a non-invasive way and by analysis magnetic signatures resulting. Differently from traditional monitoring process, in which the sensors need to be physically connected to the analyzed circuits. It should be noted, the characteristic of this methodology as a non-invasive procedure. The evaluation of the proposed methodology will be carried out through the choice of sensitivity cases, which theoretical results obtained from simulations will be used to analyze its viability.

O problema da tolerância a falhas em robôs manipuladores cooperativos conectados rigidamente a um objeto indeformável é estudado nesta tese. A tolerância a falhas é alcançada através de reconfiguração do sistema de controle. Primeiro, a falha é detectada e isolada. Então, o sistema de controle é reconfigurado de acordo com a falha isolada. As falhas em robôs manipuladores são primeiramente estudadas de acordo com suas consequências no sistema cooperativo. Quatro tipos de falhas são identificados: juntas com balanço livre (sem atuadores ativos), bloqueadas, com informação incorreta de posição e com informação incorreta de velocidade. A detecção e a isolação dos dois primeiros tipos de falhas são alcançadas através de um sistema utilizando redes neurais artificiais. Redes do tipo MLP são empregadas para mapear a dinâmica dos robôs cooperativos sem falhas e uma rede RBF é utilizada para a classificação do vetor de resíduos. As falhas do tipo informação incorreta de posição ou velocidade das juntas são detectadas e isoladas através do uso das restrições impostas pela cadeia cinemática fechada presente no sistema cooperativo. Quando falhas do tipo juntas com balanço livre ou bloqueadas são isoladas, as leis de controle são reconfiguradas. Para estes casos, controladores híbridos de movimento e esmagamento do objeto são deduzidos. Quando falhas do tipo informação incorreta de posição ou velocidade das juntas são isoladas, as medidas afetadas são substituídas por valores estimados. Resultados obtidos em simulações e em robôs cooperativos reais mostram que a metodologia proposta é viável. / The problem of fault tolerance in cooperative manipulators rigidly connected to an undeformable load is addressed in this work. Fault tolerance is reached by reconfiguration of the control system. The faults are firstly detected and isolated. Then, the control system is reconfigured according to the isolated fault. Four faults are considered: free-swinging joint faults, locked joint faults, incorrectly measured joint position faults, and incorrectly measured joint velocity faults. Free-swinging and locked joint faults are detected and isolated by artificial neural networks. MLP’s are utilized to reproduce the dynamics of the fault-free system and an RBF is used to classify the residual vector. Incorrectly measured joint position and velocity faults are detected and isolated based on the kinematic constraints imposed on the cooperative system. When free-swinging and locked joint faults are isolated, the control laws are reconfigured. Control laws for motion and squeeze of the object are developed in these cases. When incorrectly measured joint position faults and incorrectly measured joint velocity faults are isolated, the faulty measurements are replaced by their estimates. Results obtained in simulations and in real cooperative robots indicate that the proposed methodology is viable.

Este trabalho apresenta o estado da arte das técnicas mais aplicadas para localização de faltas e modelagem de faltas de alta impedância e propõe a utilização de uma recente técnica baseada na decomposição dos sinais em componentes ortogonais. Este estudo avalia a aplicabilidade da técnica proposta utilizando dados reais de um sistema de distribuição de energia brasileiro, além de apresentar os conceitos teóricos sobre a decomposição em componentes ortogonais. / This work presents the state of the art of the most used techniques for locating and modelling high impedance faults and proposes the use of a recent technique based on the decomposition of the signals in orthogonal components. The objective of this study is to evaluate the application of the proposed technique using real data from a Brazilian distribution network, and presents the theory on orthogonal decomposition.

Les défauts d’arcs électriques sont souvent la cause du déclenchement d’un incendie dans l’habitat. Les détecter constitue une amélioration de la sécurité électrique domestique. La contribution de cette thèse porte sur la problématique de l’identification et de la détection de défauts d’arcs électriques sur le réseau basse tension domestique (charge simple et combinée). Dans ce travail de thèse, le premier axe de recherche porte sur la mise en œuvre d’un modèle de simulation de réseau. Sur le modèle de réseau l’objectif est d’introduire à volonté et de manière graphique des charges de nature différentes associées à un défaut d’arc électrique série. La modélisation porte sur la source de tension, des charges domestiques, le défaut d’arc et le bloc d’analyse numérique du courant et de la tension pour la commande d’un disjoncteur. Le second axe de recherche porte la mise au point de méthodes de détection robuste d’un défaut d’arc électrique Quatre méthodes originales de détection portant sur la tension et le courant de ligne ont été développées dans le cadre de ce travail de thèse. Elles sont toutes basées sur l’analyse des caractéristiques fréquentielles et temporelles du courant de ligne. Toutes les méthodes produisent une information dont l’amplitude tend à s’accroître en présence d’un arc électrique. Toutefois, le caractère non déterministe et très variable des signaux produits par un arc électrique ainsi que la grande variabilité des charges et des perturbations qu’elles introduisent rendent la détection très difficile. L’ensemble des études s’est donc systématiquement composé d’une comparaison entre la situation avec et sans arcs. Le résultat a abouti à une approche combinée de plusieurs méthodes et d’un modèle simple de décision basé sur un simple seuil. Un prototype implémentant une partie de l’algorithme a montré la faisabilité d’un circuit en temps réel / The electrical arc faults are the cause of electrical home fire. Detecting them is an improvement of domestic electrical safety. The thesis contribution focuses on the problem of the electrical arc fault identification and detection in the domestic low voltage network. In this thesis, the first research objective is the modeling of a domestic electrical network. The aim of the modeling is to recreate graphically the loads and the topology of the real domestic electrical network associated with a series electrical arc fault. The modeling is made on the AC voltage source, domestic loads, arc fault model and a block of current and voltage analysis to control a circuit breaker. The second research objective is the development of a robust method for detecting arcing fault. Four detecting methods were developed as part of this thesis; they are based on the current analysis of frequency and time characteristics. All methods generate information whose amplitude tends to increase in the presence of an electric arc. However, the non-deterministic and random feature of the signals produced by the electrical arc fault as well as the large number and variability of loads and the disturbances which they introduce make the arc detection very difficult. All the studies consistently are composed of a comparison between the normal operation and arc fault signals. The result has resulted in a combined approach of some methodologies and a simple decision model based on a threshold. A prototype which contains a part of the algorithm was shown the feasibility of a circuit in real time

Les arcs électriques ont été à l'origine de plusieurs cas d'accidents aériens graves par incendie ou par destruction d'organes électriques fondamentaux (ex : actionneurs, panneau supérieur,...) (TWA 800 en 1996 et Vol 11 Swiss Air en 1998). Les prévenir et les détecter constituent une amélioration notable de la sécurité en aéronautique. Ainsi, bien que le problème soit connu depuis longtemps, les premiers systèmes de détection d'arcs pour l'avionique sont très récents. Dans un avion, ces défauts ont pour causes principales le vieillissement et la détérioration des câbles. Afin d'étudier et de proposer un système de détection d'arc, les trois phases d'étude suivantes ont été suivies chronologiquement dans ce travail de thèse : premièrement, la réalisation d'un banc de production et de caractérisation des grands types d'arcs électriques (AC et DC). Ce système permet de produire des arcs contrôlés en intensité et en durée selon les modes série, parallèle, DC et AC. L'amorçage des arcs est obtenu par surtension. Cette surtension est produite par commutation inductive. Deuxièmement, la recherche de modèles électriques fiables expliquant les phénomènes d'arcs. Nous proposons donc un meta modèle associant les caractéristiques de différents modèles comportementaux et impédencemétriques.Cette association permet d'améliorer significativement les performances globales de la modélisation dans les circuits. Troisièmement, l'étude et la proposition d'algorithmes de détection en vue de l'implantation sur circuit électronique suffisamment réduit pour être implanté dans un disjoncteur. La détection de la signature d'arc se rapproche d'un problème de détection d'informations fortement bruitées pour lequel peu de solutions ont été jusqu'à présent proposées car cette recherche est relativement nouvelle.Nous proposons et testons différentes méthodes de détection. / The 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.

Grâce à leurs avancées techniques en termes de poids, performances et fiabilité, les machines synchrones sont de plus en plus utilisées dans le domaine de transport et en particuliers dans l’aéronautique. Les stratégies de maintenances de ces systèmes électriques sont alors indispensables afin d'éviter des surcouts liés à des temps d'arrêt non planifiés. Ce document propose une analyse des conséquences d'un défaut inter-spires intermittent naissant dans l'enroulement statorique d'une Machine Synchrone à Aimants Permanents (MSAP). Ce type de défaut correspond à l'état peu avancé d'un futur court-circuit permanent. Jusqu'à présent, les études menées se sont limitées à la détection de courts-circuits inter-spires permanents. L'objectif de cette analyse est de définir une méthode de détection de ce type de défaut qui soit facile à mettre en œuvre. A partir d'une étude analytique du courant statorique d'une MSAP contrôlée en courant en présence de court-circuit intermittent, nous avons étudié l'impact des différentes grandeurs influençant la perturbation du courant. Nous avons constaté que la forme de la perturbation créée par le défaut était toujours la même et qu'elle était la signature du défaut intermittent dans le courant. Par la suite cette étude analytique a été validée expérimentalement. Dans la partie suivante nous avons étudié la sensibilité des méthodes de détection des courts-circuits inter-spires permanents appliquées au cas du court-circuit intermittent. Ces méthodes se sont révélées inadaptées pour la détection du défaut étudié dans ce travail. Nous avons donc proposé une méthode dédiée qui est basée sur la détection de la signature du défaut et qui utilise une transformation en ondelette adaptée. Il s'agit d'une méthode de détection de forme qui permet non seulement de détecter le défaut intermittent mais aussi de le distinguer des autres types de défauts. La performance de la méthode a été validée par les résultats de simulation et de manipulation. Dans une dernière partie, une étude plus générale sur le suivi de vieillissement des enroulements est proposée. Elle est basée sur le suivi de l'évolution des courbes d'admittance hautes fréquences d'un bobinage au cours du temps en utilisant les fonctions de transfert de ce dernier. / Thanks to technical advances in terms of weight, performance and reliability, synchronous machines are increasingly used in the transport field and especially in aeronautics. The maintenance strategies of these electrical systems are essential to avoid extra costs associated with unscheduled downtime. This document offers a study on the intermittent inter-turn fault occurring in the stator winding of a Permanent Magnet Synchronous Machine (PMSM) and its consequences. This type of fault correspond to the emerging state for a future permanent short circuit condition. So far, studies have been limited to the detection of continuous inter-turn short circuits. The main purpose of this analysis is to define a detection method for this type of fault easy to implement. Based on the stator current analytical study of a PMSM current controlled in presence of intermittent short circuit, we had studied the impact of different variables influencing the current disturbance. We had found that the shape of the disturbance created by the fault was always the same and that it was the fault signature in the current signal. Later this analytical study was validated experimentally. In the next part we had studied the sensitivity of continuous short circuits detection methods applied in the case of intermittent short circuit. These methods have been proved unsuitable to detect the defect studied in this work. Therefore, we had proposed a dedicated method based on the fault signature identification using an adapted wavelet transform. It is a pattern detection method able to detect the intermittent fault and to distinguish it from other types of defects. The performance of the method was validated by simulation and experimental results. In the last part, a more general study concerning the winding health monitoring is proposed. It uses transfer functions and it is based on the monitoring over time of the winding high frequencies admittance curves evolution.

Os motores de indução trifásicos são os principais elementos de conversão de energia elétrica em mecânica motriz aplicados em vários setores produtivos. Identificar um defeito no motor em operação pode fornecer, antes que ele falhe, maior segurança no processo de tomada de decisão sobre a manutenção da máquina, redução de custos e aumento de disponibilidade. Nesta tese são apresentas inicialmente uma revisão bibliográfica e a metodologia geral para a reprodução dos defeitos nos motores e a aplicação da técnica de discretização dos sinais de correntes e tensões no domínio do tempo. É também desenvolvido um estudo comparativo entre métodos de classificação de padrões para a identificação de defeitos nestas máquinas, tais como: Naive Bayes, k-Nearest Neighbor, Support Vector Machine (Sequential Minimal Optimization), Rede Neural Artificial (Perceptron Multicamadas), Repeated Incremental Pruning to Produce Error Reduction e C4.5 Decision Tree. Também aplicou-se o conceito de Sistemas Multiagentes (SMA) para suportar a utilização de múltiplos métodos concorrentes de forma distribuída para reconhecimento de padrões de defeitos em rolamentos defeituosos, quebras nas barras da gaiola de esquilo do rotor e curto-circuito entre as bobinas do enrolamento do estator de motores de indução trifásicos. Complementarmente, algumas estratégias para a definição da severidade dos defeitos supracitados em motores foram exploradas, fazendo inclusive uma averiguação da influência do desequilíbrio de tensão na alimentação da máquina para a determinação destas anomalias. Os dados experimentais foram adquiridos por meio de uma bancada experimental em laboratório com motores de potência de 1 e 2 cv acionados diretamente na rede elétrica, operando em várias condições de desequilíbrio das tensões e variações da carga mecânica aplicada ao eixo do motor. / The three-phase induction motors are the key elements of electromechanical energy conversion in a variety of productive sectors. Identify a defect in an operating motor can provide, before it fails, greater safety for decision making on machine maintenance, reduce costs and increase process availability. This thesis initially presents a literature review and the general methodology for reproduction of defects in the motors and the application of discretization technique of current and voltage signals in the time domain. It was also developed a comparative study of methods of pattern classification for the identification of defects has been developed in these machines, such as Naive Bayes, k-Nearest Neighbor, Support Vector Machine (Sequential Minimal Optimization), Artificial Neural Network (Multilayer Perceptron), Repeated incremental Pruning to Produce Error Reduction and C4.5 Decision Tree. Also applied the concept of Multi-Agent Systems (MAS) to support the use of multiple competing methods in a distributed manner to pattern recognition of faults in bearings, broken rotor bars and stator short-circuit in induction motors. Additionally, some strategies for the definition of the severity of the aforementioned defects in engines have been explored, including making an investigation of the influence of voltage unbalance in the machine feed for the determination of these anomalies. Experimental data are acquired from 1 and 2 cv motors under sinusoidal supply, operating in various unbalance conditions and under a wide range of mechanical load applied to the motor shaft.

Dans la majorité des secteurs industriels, les coûts de maintenance des systèmes de production représentent une partie non négligeable de l’ensemble de ces coûts. Pour améliorer les performances et la fiabilité des systèmes, différentes stratégies de maintenance sont utilisées. L’objectif est d’être capable de détecter toutes défaillances potentielles susceptibles d’affecter la production. L’analyse vibratoire est aujourd’hui un des outils les plus performants en termes de détection et de suivi de ces défaillances. Cependant, l’application de cette technique reste difficile dans le cas de machines fonctionnant avec des vitesses de rotation faibles et variables. Dans la présente thèse, nous proposons une technique alternative à l’analyse vibratoire pour surveiller l’état de fonctionnement de ces machines. Nous nous intéresserons plus particulièrement au suivi des roulements. Cette technique alternative se base sur l’analyse des variations de la vitesse angulaire instantanée qui sont générées par un défaut de roulement. Cette technique sera testée sur un banc d’essai et comparée avec d’autres techniques telles que l’analyse des signaux vibratoires et ultrasonores pour différentes conditions de chargement et pour différents types de défauts de roulement. Cette étude montre une grande efficacité de cette méthode dans le cas de très basses vitesses de rotation, constantes ou variables. L’ensemble de cette méthodologie a ensuite été implémentée sur un four de diffusion d’une unité de production sucrière. Les mesures réalisées montrent que cette technique est une technique prometteuse pour le suivi de systèmes fonctionnant avec des vitesses de rotation faibles et variables. / In the majority of industrial sectors, production systems’ maintenance costs represent an important part of the whole process’s cost. In order to improve systems’ reliability and performance, different maintenance strategies are used. The goal is to be able to detect all potential malfunction that may affect the production. Nowadays, vibration analysis is one of the best tools in term of detection and defect monitoring. However, the application of these vibration based techniques remain difficult in the case of machines functioning at low speeds. In this thesis, we propose an alternative technique of vibration analysis for the surveillance of these machines that operate on very low speeds and variable speeds. We are specifically interested in bearings’ monitoring. This alternative technique is based on instantaneous angular speed’s variations analysis that are induced by a bearing defect. This technique is tested on a test bench and compared with other techniques such as vibration and ultrasound analysis for different loading conditions and for different bearing fault types. This study shows a high efficiency of this method in the case of very low speeds, constant and variable. This methodology has then been implemented on sugar diffusion oven in a sugar production factory. The measurements shows that this technique is a promising technique for machines’ surveillance in very low and variable speeds working conditions.

La thèse est dédiée à l'analyse et à la conception de la commande tolérante aux défauts (fault tolerant control - FTC) en se fondant sur des méthodes ensemblistes. Nous étudions l'apparition des défauts pour les systèmes multi-capteurs, et les modes de détection, ainsi que la conception de lois de commande qui assurent la stabilité en boucle fermée. L'utilisation des ensembles invariants/contractifs permet la caractérisation des signaux résiduels, qui sont utilisés par la suite dans le processus de détection et d'isolement des défauts. La décision est fondée sur le positionnement par rapport à des hyperplans de séparation avec des importantes réductions de temps de calcul. Un mécanisme dual mis en œuvre par un bloc de récupération, permet la certification de la récupération des capteurs précédemment affectés par ces défauts.Dans une perspective théorique, nous soulignons les conditions qui permettent l'inclusion du bloc FDI (fault detection and isolation) et sa raison d'être dans la conception des lois de commande. Cela conduit par exemple à la synthèse des gains de retour d'état statique, par résolution de problèmes d'optimisation efficace (linéaire/convexe).Selon les paramètres choisis pour le réglage, la conception de la FTC peut être complétée par un superviseur de référence ou d'une loi de commande prédictive, qui adapte la trajectoire d'état et l'action de commande par retour d'état, afin d'assurer l'identification et la détection des défauts. Les questions spécifiques à l'utilisation de méthodes ensemblistes sont détaillées et des améliorations diverses sont proposées, par exemple : la construction des ensembles invariants, des formulations moins complexes des problèmes de type Mixed Integer Programming (MIP), l'analyse de la stabilité des systèmes commutés (notion de ``dwell-time''). / The scope of the thesis is the analysis and design of fault tolerant control (FTC) schemes through the use of set-theoretic methods. In the framework of multisensor schemes, the faults appearance and the modalities to accurately detect them are investigated as well as the design of control laws which assure the closed-loop stability. By using invariant/contractive sets to describe the residual signals, a fault detection and isolation (FDI) mechanism with reduced computational demands is implemented based on set-separation. A dual mechanism, implemented by a recovery block, which certificates previously fault-affected sensors is also studied. From a broader theoretical perspective, we point to the conditions which allow the inclusion of {FDI} objectives in the control law design. This leads to static feedback gains synthesis by means of numerically attractive optimization problems. Depending on the parameters selected for tuning, is shown that the FTC design can be completed by a reference governor or a predictive control scheme which adapts the state trajectory and the feedback control action in order to assure {FDI}. When necessary, the specific issues originated by the use of set-theoretic methods are detailed and various improvements are proposed towards: invariant set construction, mixed integer programming (MIP), stability for switched systems (dwell-time notions).