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.

Driven by intense competition, increasing operational cost and strict environmental regulations, the modern process and energy industry needs to find the best possible way to adapt to maintain profitability. Optimization of control and operation of the industrial systems is essential to satisfy the contradicting objectives of improving product quality and process efficiency while reducing production cost and plant downtime. Use of optimization not only improves the control and monitoring of assets but also offers better coordination among different assets. Thus, it can lead to considerable savings in energy and resource consumption, and consequently offer a reduction in operational costs, by offering better control, diagnostics and decision support. This is one of the main driving forces behind developing new methods, tools and frameworks that can be integrated with the existing industrial automation platforms to benefit from optimal control and operation. The main focus of this dissertation is the use of different process models, soft sensors and optimization techniques to improve the control, diagnostics and decision support for the process and energy industry. A generic architecture for an optimal control, diagnostics and decision support system, referred to here as a learning system, is proposed. The research is centred around an investigation of different components of the proposed learning system. Two very different case studies within the energy-intensive pulp and paper industry and the promising micro-combined heat and power (CHP) industry are selected to demonstrate the learning system. One of the main challenges in this research arises from the marked differences between the case studies in terms of size, functions, quantity and structure of the existing automation systems. Typically, only a few pulp digesters are found in a Kraft pulping mill, but there may be hundreds of units in a micro-CHP fleet. The main argument behind the selection of these two case studies is that if the proposed learning system architecture can be adapted for these significantly different cases, it can be adapted for many other energy and process industrial cases. Within the scope of this thesis, mathematical modelling, model adaptation, model predictive control and diagnostics methods are studied for continuous pulp digesters, whereas mathematical modelling, model adaptation and diagnostics techniques are explored for the micro-CHP fleet. / FUDIPO – FUture DIrections for Process industry Optimization

Tato práce se zabývá vytvořením simulačního modelu dvojčinného pneumatického pístu s mechanickou sestavou, včetně modelů snímačů, s následujícím odhadem parametrů a aproximací chování demonstračního zařízení. Dalším cílem je prezentace různých přístupů prediktivní údržby na datové sadě měřené na demonstračním zařízení. Na měřený datový soubor se aplikovaly signal-based techniky bez použití simulačního modelu a model-based metody, které vyžadují použití simulačního modelu. Výsledkem této práce je ověření možnosti monitorování stavu zařízení pomocí nainstalovaných senzorů a vyhodnocení efektivity senzorů z hlediska přesnosti a finančních nákladů.

Sensors are highly essential components in most modern control systems and are used in increasingly complex ways to improve system precision and reliability. Since they are generally susceptible to faults it is common to perform on-line fault diagnostics on sensor data to verify nominal behavior. This is especially important for safety critical systems where it can be imperative to identify, and react to, a fault before it increases in severity. An example of such a safety critical system is the propulsion control of a vehicle. In this thesis, three different model-based methods for Fault Detection and Isolation (FDI) are developed and tested with the aim of detecting and isolating sensor faults in the powertrain of an electric, center articulated, four-wheel-drive vehicle. First, kinematic models are derived that combine sensor data from all sensors related to propulsion. Second, the kinematic models are implemented in system observers to produce fault sensitive zero-mean residuals. Finally, fault isolation algorithms are derived, which detect and indicate different types of faults via evaluation of the observer residuals. The results show that all FDI methods can detect and isolate stochastic faults with high certainty, but that offset-type faults are hard to distinguish from modeling errors and are therefore easily attenuated by the system observers. Faults in accelerometer sensors need extra measures to be detectable, owing to the environment where the vehicle is typically operated. A nonlinear system model shows good conformity to the vehicle system, lending confidence to its further use as a driver for propulsion control.

In this thesis, the red hot topic anomaly detection is studied, which is a subtopic in machine learning. The company, Procera Networks, supports several broadband companies with IT-solutions and would like to detected errors in these systems automatically. This thesis investigates and devises methods and algorithms for detecting interesting events in diagnostics data. Events of interest include: short-term deviations (a deviating point), long-term deviations (a distinct trend) and other unexpected deviations. Three models are analyzed, namely Linear Predictive Coding, Sparse Linear Prediction and Wavelet Transformation. The final outcome is determined by the gap to certain thresholds. These thresholds are customized to fit the model as well as possible. / I den här rapporten kommer det glödheta området anomalidetektering studeras, vilket tillhör ämnet Machine Learning. Företaget där arbetet utfördes på heter Procera Networks och jobbar med IT-lösningar inom bredband till andra företag. Procera önskar att kunna upptäcka fel hos kunderna i dessa system automatiskt. I det här projektet kommer olika metoder för att hitta intressanta företeelser i datatraffiken att genomföras och forskas kring. De mest intressanta företeelserna är framfärallt snabba avvikelser (avvikande punkt) och färändringar äver tid (trender) men också andra oväntade mänster. Tre modeller har analyserats, nämligen Linear Predictive Coding, Sparse Linear Prediction och Wavelet Transform. Det slutgiltiga resultatet från modellerna är grundat på en speciell träskel som är skapad fär att ge ett så bra resultat som mäjligt till den undersäkta modellen..

Intelligent fault detection systems using machine learning can be applied to learn to spot anomalies in signals sampled directly from machinery. As a result, expensive repair costs due to mechanical breakdowns and potential harm to humans due to malfunctioning equipment can be prevented. In recent years, Autoencoders have been applied for fault detection in areas such as industrial manufacturing. It has been shown that they are well suited for the purpose as such models can learn to recognize healthy signals that facilitate the detection of anomalies. The content of this thesis is an investigation into the applicability of Autoencoders for fault detection in mobile robotics by assigning anomaly scores to sampled torque signals based on the Autoencoder reconstruction errors and the Mahalanobis distance to a known distribution of healthy errors. An experiment was carried out by training a model with signals recorded from a four-wheeled mobile robot executing a pre-defined diagnostics routine to stress the motors, and datasets of healthy samples along with three different injected faults were created. The model produced overall greater anomaly scores for one of the fault cases in comparison to the healthy data. However, the two other cases did not yield any difference in anomaly scores due to the faults not impacting the pattern of the signals. Additionally, the Autoencoders ability to isolate a fault to a location was studied by examining the reconstruction errors faulty samples determine whether the errors of signals originating from the faulty component could be used for this purpose. Although we could not confirm this based on the results, fault isolation with Autoencoders could still be possible given more representative signals.

Closed-loop system identification and fault detection and isolation are the two fundamental building blocks of process monitoring. Efficient and accurate process monitoring increases plant availability and utilisation. This dissertation investigates a subspace system identification and fault detection methodology for the Benfield process, used by Sasol, Synfuels in Secunda, South Africa, to remove CO2 from CO2-rich tail gas. Subspace identification methods originated between system theory, geometry and numerical linear algebra which makes it a computationally efficient tool to estimate system parameters. Subspace identification methods are classified as Black-Box identification techniques, where it does not rely on a-priori process information and estimates the process model structure and order automatically. Typical subspace identification algorithms use non-parsimonious model formulation, with extra terms in the model that appear to be non-causal (stochastic noise components). These extra terms are included to conveniently perform subspace projection, but are the cause for inflated variance in the estimates, and partially responsible for the loss of closed-loop identifiably. The subspace identification methodology proposed in this dissertation incorporates two successive LQ decompositions to remove stochastic components and obtain state-space models of the plant respectively. The stability of the identified plant is further guaranteed by using the shift invariant property of the extended observability matrix by appending the shifted extended observability matrix by a block of zeros. It is shown that the spectral radius of the identified system matrices all lies within a unit boundary, when the system matrices are derived from the newly appended extended observability matrix. The proposed subspace identification methodology is validated and verified by re-identifying the Benfield process operating in closed-loop, with an RMPCT controller, using measured closed-loop process data. Models that have been identified from data measured from the Benfield process operating in closed-loop with an RMPCT controller produced validation data fits of 65% and higher. From residual analysis results, it was concluded that the proposed subspace identification method produce models that are accurate in predicting future outputs and represent a wide variety of process inputs. A parametric fault detection methodology is proposed that monitors the estimated system parameters as identified from the subspace identification methodology. The fault detection methodology is based on the monitoring of parameter discrepancies, where sporadic parameter deviations will be detected as faults. Extended Kalman filter theory is implemented to estimate system parameters, instead of system states, as new process data becomes readily available. The extended Kalman filter needs accurate initial parameter estimates and is thus periodically updated by the subspace identification methodology, as a new set of more accurate parameters have been identified. The proposed fault detection methodology is validated and verified by monitoring process behaviour of the Benfield process. Faults that were monitored for, and detected include foaming, flooding and sensor faults. Initial process parameters as identified from the subspace method can be tracked efficiently by using an extended Kalman filter. This enables the fault detection methodology to identify process parameter deviations, with a process parameter deviation sensitivity of 2% or higher. This means that a 2% parameter deviation will be detected which greatly enhances the fault detection efficiency and sensitivity. / Dissertation (MEng)--University of Pretoria, 2008. / Electrical, Electronic and Computer Engineering / unrestricted

La dangerosité des défauts électriques et notamment des défauts d’arcs série dans les installations basse tension est connue depuis longtemps et représente une problématique d’actualité. La détection et la localisation de ces défauts constituent ainsi le sujet d’étude de cette thèse. Notons également, qu’à l’heure actuelle, aucun disjoncteur pour la détection des défauts d’arcs n’est équipé de la fonction localisation d’un arc sur la ligne électrique. Plusieurs méthodes de localisation des défauts d’arcs électriques séries ont été proposées dans le travail présenté. La première méthode est basée sur les paramètres d’impédance obtenus à partir des lois de Kirchhoff et ceci sur une ligne expérimentale de 49 m de longueur. La seconde méthode utilise la modélisation de ligne pour obtenir différents vecteurs de signatures utilisés pour entrainer un réseau de neurones. La troisième méthode par transformée en ondelettes est basée sur l’identification des ondes haute fréquence qui apparaissent en présence d’un défaut d’arc série. L’autre contribution majeure de cette thèse est la mise au point d’un algorithme performant de détection de la présence d’un défaut d’arc électrique par analyse du courant de ligne. L’algorithme est conçu pour détecter de manière fiable les défauts d'arc dans les modes de fonctionnement stationnaires et transitoires des appareils ménagers puis dans des configurations complexes de masquage de charges et d'appareils perturbateurs. L’algorithme repose sur l’analyse du courant de ligne par un filtre de Kalman associé à une logique de décision. La technique mise en œuvre, portant sur un seuillage adaptatif à base de logique floue (Fuzzy Logic), entraîne une réduction significative des faux déclenchements / The dangerousness of electrical defects and in particular serial arcing ones in low-voltage installations is well known and represents a topical research issue. The detection and localization of these defects is therefore the subject of this thesis. It should also be noted that, at present time, no circuit-breaker for arc fault detection is equipped with the arc location function on the power line. Several methods for locating series arc faults have been proposed in this work. For the first method, a model based on the impedance parameters of the experimental power line (length 49 meters) based on Kirchhoff's laws was developed. The second method uses line modeling to obtain different signature vectors used to train a neural network. The third wavelet transform method is based on the identification of high frequency waves that occur in the presence of a series arc fault. The other major contribution of this thesis is the development of an efficient algorithm for detecting the presence of an electrical arc fault by the line current analysis. The algorithm is designed to reliably detect series arcing faults in stationary and transient operating modes of household appliances and then in complex load masking and with disturbance device configurations. The algorithm is based on the analysis of the line current by a Kalman filter associated with a decision logic block. The technique used based on adaptive fuzzy logic thresholding logic, allows significant reduction in false triggering

La problématique traitée dans cette thèse vise à développer une stratégie efficace de détection et de classification des impacts en présence d'incertitudes de modélisation du robot et de son environnement et en utilisant un nombre minimal de capteurs, notamment en l'absence de capteur d’effort.La première partie de la thèse porte sur la détection d'un impact pouvant avoir lieu à n'importe quel endroit du bras robotique et à n'importe quel moment de sa trajectoire. Les méthodes de détection d’impacts sont généralement basées sur un modèle dynamique du système, ce qui les rend sujettes au compromis entre sensibilité de détection et robustesse aux incertitudes de modélisation. A cet égard, une méthodologie quantitative a d'abord été mise au point pour rendre explicite la contribution des erreurs induites par les incertitudes de modèle. Cette méthodologie a été appliquée à différentes stratégies de détection, basées soit sur une estimation directe du couple extérieur, soit sur l'utilisation d'observateurs de perturbation, dans le cas d’une modélisation parfaitement rigide ou à articulations flexibles. Une comparaison du type et de la structure des erreurs qui en découlent et de leurs conséquences sur la détection d'impacts en a été déduite. Dans une deuxième étape, de nouvelles stratégies de détection d'impacts ont été conçues: les effets dynamiques des impacts sont isolés en déterminant la marge d'erreur maximale due aux incertitudes de modèle à l’aide d’une approche stochastique.Une fois l'impact détecté et afin de déclencher la réaction post-impact du robot la plus appropriée, la deuxième partie de la thèse aborde l'étape de classification. En particulier, la distinction entre un contact intentionnel (l'opérateur interagit intentionnellement avec le robot, par exemple pour reconfigurer la tâche) et un contact non-désiré (un sujet humain heurte accidentellement le robot), ainsi que la localisation du contact sur le robot, est étudiée en utilisant des techniques d'apprentissage supervisé et plus spécifiquement des réseaux de neurones feedforward. La généralisation à plusieurs sujet humains et à différentes trajectoires du robot a été étudiée. / The present thesis aims to develop an efficient strategy for impact detection and classification in the presence of modeling uncertainties of the robot and its environment and using a minimum number of sensors, in particular in the absence of force/torque sensor.The first part of the thesis deals with the detection of an impact that can occur at any location along the robot arm and at any moment during the robot trajectory. Impact detection methods are commonly based on a dynamic model of the system, making them subject to the trade-off between sensitivity of detection and robustness to modeling uncertainties. In this respect, a quantitative methodology has first been developed to make explicit the contribution of the errors induced by model uncertainties. This methodology has been applied to various detection strategies, based either on a direct estimate of the external torque or using disturbance observers, in the perfectly rigid case or in the elastic-joint case. A comparison of the type and structure of the errors involved and their consequences on the impact detection has been deduced. In a second step, novel impact detection strategies have been designed: the dynamic effects of the impacts are isolated by determining the maximal error range due to modeling uncertainties using a stochastic approach.Once the impact has been detected and in order to trigger the most appropriate post-impact robot reaction, the second part of the thesis focuses on the classification step. In particular, the distinction between an intentional contact (the human operator intentionally interacts with the robot, for example to reconfigure the task) and an undesired contact (a human subject accidentally runs into the robot), as well as the localization of the contact on the robot, is investigated using supervised learning techniques and more specifically feedforward neural networks. The challenge of generalizing to several human subjects and robot trajectories has been investigated.

Ce travail de thèse aborde le problème de la détection et de l’isolation des pannes à base de modèle du système dynamique non linéaire. Les techniques de détection et d’identification de pannes sont déjà appliquées aux systèmes industriels et elles jouent un rôle important pour assurer les performances attendues des systèmes automatiques. Les différentes approches du diagnostic des systèmes dynamiques semblent être souvent le résultat de contextes différents notamment en ce qui concerne les applications visées et le cahier des charges qui en résulte. Ainsi, la nature des informations disponibles sur le système ou le type de défauts à détecter conduit à la mise en œuvre de stratégies spécifiques. Dans cette étude on suppose disposer d’un modèle de fonctionnement du système et les pannes considérées sont celles qui conduisent le système à ne plus suivre ce modèle. Après avoir introduit la notion de platitude différentielle pour un système dynamique non linéaire continu, plusieurs exemples de systèmes dynamiques différentiellement plats sont introduits. Les redondances analytiques mises en évidence par cette propriété sont dans une première étape utilisées pour détecter des pannes. Ceci conduit à développer des estimateurs d’ordre supérieurs pour les dérivées des sorties plates du système et des estimateurs non linéaires de l’état du système. Cette approche est mise en œuvre dans le cadre de la détection de pannes des moteurs d’un Quadri-Rotor.La notion de platitude pour les systèmes dynamiques discrets est alors introduite. Il est alors possible de développer une nouvelle approche pour la détection des pannes, fondée sur la redondance temporelle entre les informations résultant des mesures directes de composantes du vecteur d’état du Quadri-Rotor et les estimations des sorties plates à chaque instant d’échantillonnage. Cette approche qui est illustrée ici aussi dans le cas du Quadri-Rotor, permet aussi de développer une méthode d’identification en ligne des pannes en se basant sur la chronologie de la propagation de leurs effets / This PhD is submitted in model-based faults detection and isolation in nonlinear dynamic system. The techniques of faults detection and isolation are already being applied to industrial systems and have played an important role to ensure the expected performance of automated systems. The differences in approaches to diagnosis of dynamic systems often seem to be the result of different contexts including in respect of applications and referred the specification that follows. Thus, the nature of information available on the system or the type of fault detection leads to the implementation of specific strategies. In this study we have assumed a model of system operation and faults considered are those that lead the system to no longer follow this model.After introducing the concept of differential flatness for a nonlinear dynamical system continued, several examples of differentially flat systems dynamics are introduced. The analytical redundancy highlighted by this property is a first step used to detect faults. This leads to develop estimators for higher order derivatives of the outputs flat of the system and estimator plate for nonlinear system state. This approach is implemented in the context of fault detection engine of a Quadri-Rotor.The notion of flatness for discrete dynamical systems is introduced. It is then possible to develop a new approach for fault detection based on temporal redundancy between the information from direct measurements of components of the state vector of Quadri-Rotor and estimates of output flat at each sampling instant. This approach is illustrated here as in the case of the Quadri-Rotor, can also develop a method for online identification of fault based on the chronology of the spread of their effects