Tesi sul tema "Fault decomposition"

As digital control systems are used in life-critical applications, assessment of the safety of these control systems becomes increasingly important. One means of formally performing this assessment is through fault tree analysis. Software Fault Tree Analysis (SFTA) starts with a system-level hazard that must be decomposed in a largely-human-intensive manner until specific modules of the software system are indicated. These modules can then be formally analyzed using statement templates. The focus of this thesis is to approach the decomposition of a system-level hazard from a formalized standpoint. Decomposition primarily proceeds along two distinct but interdependent dimensions, specificity of event and subsystem size. The Specificity-of-Event dimension breaks abstract or combined events into the specific system events that must be analyzed by the fault tree. The Subsystem-Size dimension deals with the scope of the hazard, and itemizes the subsystems where localized events may lead to the hazard. Decomposition templates are developed in this thesis to provide a framework for decomposing a system-level hazard to the point at which line-by-line code analysis can be conducted with existing statement templates. These templates serve as guides for conducting the decomposition, and ensure that as many as possible of all the applicable decomposition aspects are evaluated

Dans ce travail de doctorat, une étude théorique et expérimentale systématique a été menée sur l'isolation du mélange de gaz C4F7N-CO2-O2 respectueux de l'environnement et ses caractéristiques de décomposition et de biosécurité sous défauts électriques et thermiques. Sur la base de la méthode de dynamique moléculaire ReaxFF, le processus de décomposition thermique du mélange gazeux sous différentes teneurs en O2 et températures est simulé. En combinant les résultats simulés avec des essais de décomposition thermique, le processus cinétique de décomposition thermique du mélange gazeux et le mécanisme d'évolution de ses sous-produits dans différentes conditions sont révélés. En même temps, le mécanisme d'influence de la teneur en O2 sur la tension de claquage et les valeurs caractéristiques statistiques du mélange C4F7N-CO2-O2 pour des décharges partielles sont analysés et le mécanisme d'influence de différents facteurs sur la génération et l'inhibition des gaz et des sous-produits solides au cours du processus de décomposition par décharge du mélange gazeux est clarifié. En conclusion, sur la base des résultats de simulation et expérimentaux, nous proposons la quantité optimale d'additif O2 et les composants caractéristiques de diagnostic du mélange gazeux C4F7N-CO2-O2 pour les défauts des équipements isolés au gaz moyenne tension ; nous testons la biosécurité du C4F7N et de ses produits de décomposition après l'exposition à l'arc, puis évaluons la faisabilité, la sécurité de l'application du mélange de gaz C4F7N-CO2-O2 dans l'équipement en combinant avec les caractéristiques isolantes et de décomposition électrique et thermique du mélange de gaz C4F7N-CO2-O2 et, les résultats de la biosécurité
In this doctoral work, a systematic theoretical and experimental study has been carried out on the insulation of environmentally friendly C4F7N-CO2-O2 gas mixture and on its decomposition characteristics and biosafety under electrical and thermal faults. Based on the ReaxFF molecular dynamics method, the thermal decomposition process of the gas mixture under different O2 contents and temperatures is simulated. The kinetic process of the thermal decomposition of the gas mixture and the evolution mechanism of its by-products under different conditions are revealed by combining with thermal decomposition tests. Meanwhile, the influence mechanism of O2 content on the breakdown voltage and partial discharge statistical characteristic values of the C4F7N-CO2-O2 mixture is analyzed, and the influence mechanism of different factors on the generation and inhibition of gas and solid by-products during the discharge decomposition process of the gas mixture is clarified. In conclusion, based on the simulation and experimental results, we propose the optimal O2 additive amount and fault diagnosis characteristic components of C4F7N-CO2-O2 gas mixture for medium-voltage gas-insulated equipmentwe test the biosafety of C4F7N and its arc decomposition products, and then evaluate the feasibility and safety of applying C4F7N-CO2-O2 gas mixture in equipment by combining with the insulating and electrical and thermal decomposition characteristics of C4F7N-CO2-O2 gas mixture and the results of the biosafety

La diagnosi di difetti in macchine rotanti basata sull’analisi vibrazionale ha raggiunto una soddisfacente fase di maturità, essendo disponibili numerose metodologie consolidate per la rilevazione e l’identificazione di difetti. Tuttavia, diverse problematiche restano ancora aperte; questa tesi ne prende in considerazione due. Da un lato, la ciclostazionarietà non è stata ancora utilizzata esplicitamente per progettare criteri di deconvoluzione cieca per la diagnosi di macchine rotanti, sebbene l'importanza di applicare la ciclostazionarietà per scopi diagnostici sia stata ampiamente riconosciuta. Dall’altro, la localizzazione di un difetto che si verifica in una ruota dentata installata in un albero intermedio di un riduttore a più stadi, particolarmente complessa per la sovrapposizione di più sorgenti di vibrazione, non è stata ancora oggetto di studi. In questo contesto, basandosi sulla teoria dei processi ciclostazionari, la tesi affronta questi due aspetti, differenti ma correlati e complementari, relativi all'identificazione di difetti localizzati in ingranaggi e cuscinetti volventi. La prima parte della tesi propone un metodo di deconvoluzione cieca, basato sul quoziente di Rayleigh generalizzato, risolto mediante un algoritmo iterativo di decomposizione agli autovalori. Questo approccio è caratterizzato dalla presenza di una matrice di pesatura che guida il processo di deconvoluzione, grazie alla quale il metodo può essere facilmente adattato a criteri arbitrari. Un nuovo criterio basato sulla massimizzazione della ciclostazionarietà del secondo ordine viene proposto e confrontato con altri metodi di deconvoluzione cieca esistenti in letteratura. Il confronto, effettuato su segnali simulati e segnali sperimentali, ha dimostrato che l’algoritmo è efficace nella stima delle eccitazioni ciclostazionarie a partire da risposte vibratorie sia a regimi stazionari sia a regimi non stazionari. Questo metodo è validato attraverso due diversi casi sperimentali relativi ad un rotismo ordinario a due stadi e ad un cuscinetto volvente. L'originalità di questa parte riguarda l'introduzione di un nuovo algoritmo di deconvoluzione cieca basato su di un criterio ciclostazionario che consente l'estrazione di sorgenti ciclsotazionarie aventi una determinata frequenza ciclica. Sulla base di questo metodo, sono proposti inoltre due procedure originali per la diagnosi di cuscinetti e ingranaggi. In particolare, queste procedure si basano sul criterio ciclostazionario massimizzato mediante il metodo di deconvoluzione cieca che consente la diagnosi del difetto in termini di tipologia e di severità. La seconda parte riguarda lo sviluppo e la validazione di un metodo per l'identificazione di difetti localizzati presenti in una ruota dentata calettata su un albero intermedio di un rotismo ordinario multi-stadio. In questo contesto, si propone una metodologia che combina la Empirical Mode Decomposition e la media sincrona per separare il segnale ciclostazionario del primo ordine relativo alle ruote dentate sincrone, montate sul medesimo albero, in un insieme di segnali rappresentativi relativi alle singole ruote dentate. I modi oscillatori fisicamente significativi sono selezionati attraverso un criterio basato sui coefficienti di correlazione di Pearson. Il rilevamento dei guasti viene eseguito successivamente mediante indicatori di condizione dedicati. In aggiunta agli indicatori di condizione standard, sono proposti due nuovi indicatori di condizione sensibili alle variazioni di energia del segnale sul passo della ruota, che si sono dimostrati particolarmente efficaci per il rilevamento dei difetti localizzati. L’efficacia della metodologia proposta è esaurientemente discussa mediante l’applicazione a segnali simulati e da due set di dati sperimentali. In tutti i casi esaminati, i risultati mostrano la capacità di identificare con successo la ruota difettosa nei casi di più ruote calettate sullo stesso albero.
Vibration analysis provides a useful aid for monitoring many mechanical systems and industrial processes. In recent years, the vibration-based diagnosis of machines and mechanical systems has reached a satisfactory stage of maturity. Several established signal processing methodologies are now available for detecting and identifying localized faults, especially for gears and bearings. However, several questions are still open. Among them, this thesis addresses two correlated issues. On the one hand, cyclostationarity has not been explicitly used to design blind deconvolution criteria for machine diagnosis before now, although the importance to take advantage of cyclostationarity for diagnostics purpose has been widely recognized. Concurrently, the localization of a gear fault occurring in a gear located in an intermediate shaft of a multi-stage gearbox can be particularly complex due to the superposition of vibration signatures of different synchronous wheels. Nevertheless, this issue has not been investigated yet. On these grounds, this thesis has been focused on these two different but complementary facets about impulsive fault identification in rotating machines both rooted in the cyclostationary framework. The first part of the thesis focuses on a blind deconvolution method based on the generalized Rayleigh quotient and solved by means of an iterative eigenvalue decomposition algorithm. This approach is characterized by the presence of a weighting matrix that drives the deconvolution process, whereby it can be easily adapted to arbitrary criteria. A novel criterion based on the maximization of the cyclostationarity of the signal is proposed and compared with the other blind deconvolution methods existing in the literature. The proposed algorithm is extensively compared taking into account cyclostationary synthetic signals and real ones, demonstrating superior capability to recover cyclostationary sources both in stationary regimes and non-stationary regimes. This method is successfully validated for diagnostic purposes through two different experimental cases consisting of a gear tooth spall and an outer race bearing fault. The originality of this part mainly regards the introduction of a novel blind deconvolution algorithm based on a cyclostationary criterion that allows for the extraction of cyclostationary sources having a given cyclic frequency. Two original and consistent diagnostic protocols for bearing and gear diagnosis are proposed as well. In particular, these diagnostic procedures take advantage of the maximized cyclostationary criterion computed by way of the proposed blind deconvolution method allowing the diagnosis in terms of fault type and severity. The second part addresses a method for the identification of gear tooth faults occurring in a wheel located in the intermediate shaft of multi-stage gearboxes. In this context, this part introduces a methodology which combines the Empirical Mode Decomposition and the Time Synchronous Average in order to separate the first-order cyclostationary signal of the synchronous gears mounted on the same shaft into a set of representing signals of the single gears. The physical meaningful modes are selected by means of a criterion based on Pearson’s correlation coefficients and the fault detection is performed by dedicated condition indicators. The proposed methodology is exhaustively discussed and supported by simulated examples as well as two experimental datasets. This original strategy successfully identifies the faulty gear in both the experimental tests and therefore can be considered reliable for the identification of a faulty gear when the fault occurs in a shaft with multiple gears. Furthermore, two novel condition indicators sensitive to signal energy variations on the circular pitch are proposed and proved to be effective for the local gear fault detection.

Axlebox bearings are one of the most critical components of a rail vehicle with regard to safety. An axlebox bearing that breaks during operation can be dangerous for the passengers and expensive for the operator. In-service failure of axlebox bearings has been the cause of many catastrophic accidents. Thus, it is of utmost importance to predict bearing failures as early as possible. This will increase reliability and safety of the vehicle as well as reduce the vehicle maintenance cost. Monitoring of roller bearings is an active research eld, and many methods have been proposed by other researchers. Many of these methods employ complex algorithms to make the most use of the given measurements. The algorithms often lack interpretability and have high computational costs, making them dicult to employ in an on-board system. This thesis proposes an interpretable and transparent algorithm that predicts bearing damages with high accuracy. Meanwhile, it tries to retain interpretability as much as possible. The algorithm is based on Empirical Mode Decomposition (EMD) and Singular Value Decomposition (SVD). These two techniques extract essential and meaningful information from the axlebox accelerations. The algorithm is benchmarked on two benchmark datasets, and the results are compared to the respective literature. Then the algorithm is employed on the railway axlebox acceleration measurements that were taken on an axlebox test bench available at SWJTU. The proposed algorithm can be extended to incorporate additional measurements of dierent types, e.g. sound or temperature measurements. The incorporation of other types of measurements will improve the performance of the algorithm even further.
Axelbox lager är en av de viktigaste komponenterna i ett järnvägsfordon när det berör säkerheten. Ett axelbox lager som havererar under drift kan vara farligt for passagerarna och även dyrt för operatören. Driftfel av lagren har varit orsaken till många katastrofala olyckor. Därför är det av yttersta vikt att förutsäga lagerfel så tidigt som möjligt. Detta ökar fordonets tillförlitlighet och säkerhet samt minskar underhållskostnaderna. Mycket forskning har utförts inom övervakning av rullager. Många metoder använder komplexa algoritmer för att maximalt utnyttja matningarna. Algoritmerna saknar ofta tolkbarhet och har höga beräkningskostnader, vilket gör dem svåra att använda i ett integrerat system. Denna avhandling kombinerar era metoder för databehandling och maskininlärning till en algoritm som kan förutsäga lagerskador med hög precision, samtidigt som tolkningsförmågan bibehalls. Bland andra välkända metoder sa använder algoritmen Empirical Mode Decomposition (EMD) och Singular Value Decomposition (SVD) för att extrahera väsentlig information for vibrationsmätningarna. Algoritmen testas sedan med tre olika vibrationsdatamängder, varav en mättes specikt med tanke på simulering av axelbox lager. Ett annat mål med algoritmen är att göra den tillämpad för ytterligare mätningar. Det bör vara möjligt att inkludera mätningar av olika slag, dvs ljud- eller temperaturmätningar, och därigenom förbättra resultaten. Detta skulle minska implementeringskostnaden avsevärt eftersom befintliga sensorer används för detta ändamål. I händelsen av att de föreslagna metoderna inte fungerar med nya mätningar är det även möjligt att integrera ytterligare funktioner i algoritmen.

In June 2009, a 2.2 square mile 3-D high resolution seismic reflection survey was shot in southeastern Illinois in the Phillipstown Consolidated oilfield. A well was drilled in the 3-D survey area to tie the seismic to the geological data with a synthetic seismogram from the sonic log. The objectives of the 3-D seismic survey were three-fold: 1.) To image and interpret faulting of the Herald-Phillipstown Fault using drillhole-based geological and seismic cross-sections and structural contour maps created from the drillhole data and seismic reflection data, 2.) To test the effectiveness of imaging the faults by selected seismic attributes, and 3.) To compare spectral decomposition amplitude maps with an isochron map and an isopach map of a selected geologic interval (VTG interval). Drillhole and seismic reflection data show that various formation offsets increase near the main Herald-Phillipstown fault, and that the fault and its large offset subsidiary faults penetrate the Precambrian crystalline basement. A broad, northeast-trending 10,000 feet wide graben is consistently observed in the drillhole data. Both shallow and deep formations in the geological cross-sections reveal small horst and graben features within the broad graben created possibly in response to fault reactivations. The HPF faults have been interpreted as originally Precambrian age high-angle, normal faults reactivated with various amounts and types of offset. Evidence for strike-slip movement is also clear on several faults. Changes in the seismic attribute values in the selected interval and along various time slices throughout the whole dataset correlate with the Herald-Phillipstown faults. Overall, seismic attributes could provide a means of mapping large offset faults in areas with limited or absent drillhole data. Results of the spectral decomposition suggest that if the interval velocity is known for a particular formation or interval, high-resolution 3-D seismic reflection surveys could utilize these amplitudes as an alternative seismic interpretation method for estimating formation thicknesses. A VTG isopach map was compared with an isochron map and a spectral decomposition amplitude map. The results reveal that the isochron map strongly correlates with the isopach map as well as the spectral decomposition map. It was also found that thicker areas in the isopach correlated with higher amplitude values in the spectral decomposition amplitude map. Offsets along the faults appear sharper in these amplitudes and isochron maps than in the isopach map, possibly as a result of increased spatial sampling.

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

In this thesis we present new approaches in the domains of soft fault detection and location in complex wire networks, based on the properties of time reversal. When addressing the detection of soft faults, the idea is to adapt the testing signal to the network under test, instead of being predefined for all the tested networks, as opposed to standard reflectometry techniques. We prove that this approach, which we name the Matched Pulse approach (MP), is beneficial whenever the system is more complex, i.e., its response is richer in echoes, which is opposed to common understanding. The MP analysis is conducted via a formal mathematical analysis, followed by simulation and experimental results validating the proposed approach. In the domain of soft fault location, and based on the DORT (Décomposition de l'Opérateur de Retournement Temporel) properties, we derive a distributive non-iterative method able to synthesize signals that focus on the fault position. Through a statistical study we analyze some of the influencing parameters on the performance of the method, and then simulation and experimental results show that the method is able to synthesize signals directly focalizing on the soft fault position, without the need for iterations.

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.

In order to identify the weakest bus in a power system so that the Unified Power Flow Controller could be connected, an investigation of static and dynamic voltage stability is presented. Two stability indices, static and dynamic, have been proposed in the thesis. Multi-Input Multi-Output (MIMO) analysis has been used for the dynamic stability analysis. Results based on the Western System Coordinate Council (WSCC) 3-machine, 9-bus test system and IEEE 14 bus Reliability Test System (RTS) shows that these indices detect with the degree of accuracy the weakest bus, the weakest line and the voltage stability margin in the test system before suffering from voltage collapse. Recently, Flexible Alternating Current Transmission systems (FACTs) have become significant due to the need to strengthen existing power systems. The UPFC has been identified in literature as the most comprehensive and complex FACTs equipment that has emerged for the control and optimization of power flow in AC transmission systems. Significant research has been done on the UPFC. However, the extent of UPFC capability, connected to the weakest bus in maintaining the power flows under fault conditions, not only in the line where it is installed, but also in adjacent parallel lines, remains to be studied. In the literature, it has normally been assumed the UPFC is disconnected during a fault period. In this investigation it has been shown that fault conditions can affect the UPFC significantly, even if it occurred on far buses of the power system. This forms the main contribution presented in this thesis. The impact of UPFC in minimizing the disturbances in voltages, currents and power flows under fault conditions are investigated. The WSCC 3-machine, 9-bus test system is used to investigate the effect of an unsymmetrical fault type and position on the operation of UPFC controller in accordance to the G59 protection, stability and regulation. Results show that it is necessary to disconnect the UPFC controller from the power system during unsymmetrical fault conditions.
Libyan Government

In order to identify the weakest bus in a power system so that the Unified Power Flow Controller could be connected, an investigation of static and dynamic voltage stability is presented. Two stability indices, static and dynamic, have been proposed in the thesis. Multi-Input Multi-Output (MIMO) analysis has been used for the dynamic stability analysis. Results based on the Western System Coordinate Council (WSCC) 3-machine, 9-bus test system and IEEE 14 bus Reliability Test System (RTS) shows that these indices detect with the degree of accuracy the weakest bus, the weakest line and the voltage stability margin in the test system before suffering from voltage collapse. Recently, Flexible Alternating Current Transmission systems (FACTs) have become significant due to the need to strengthen existing power systems. The UPFC has been identified in literature as the most comprehensive and complex FACTs equipment that has emerged for the control and optimization of power flow in AC transmission systems. Significant research has been done on the UPFC. However, the extent of UPFC capability, connected to the weakest bus in maintaining the power flows under fault conditions, not only in the line where it is installed, but also in adjacent parallel lines, remains to be studied. In the literature, it has normally been assumed the UPFC is disconnected during a fault period. In this investigation it has been shown that fault conditions can affect the UPFC significantly, even if it occurred on far buses of the power system. This forms the main contribution presented in this thesis. The impact of UPFC in minimizing the disturbances in voltages, currents and power flows under fault conditions are investigated. The WSCC 3-machine, 9-bus test system is used to investigate the effect of an unsymmetrical fault type and position on the operation of UPFC controller in accordance to the G59 protection, stability and regulation. Results show that it is necessary to disconnect the UPFC controller from the power system during unsymmetrical fault conditions.

Os sistemas elétricos de potência modernos apresentam inúmeros desafios em sua operação. Nos sistemas de distribuição de energia elétrica, devido à grande ramificação, presença de extensos ramais monofásicos, à dinâmica das cargas e demais particularidades inerentes, a localização de faltas representa um dos maiores desafios. Das barreiras encontradas, a influência da impedância de falta é uma das maiores, afetando significativamente a aplicação dos métodos tradicionais na localização, visto que a magnitude das correntes de falta é similar à da corrente de carga. Neste sentido, esta tese objetivou desenvolver um sistema inteligente para localização de faltas de alta impedância, o qual foi embasado na aplicação da técnica de decomposição por componentes ortogonais no pré-processamento das variáveis e inferência fuzzy para interpretar as não-linearidades do Sistemas de Distribuição com presença de Geração Distribuída. Os dados para treinamento do sistema inteligente foram obtidos a partir de simulações computacionais de um alimentador real, considerando uma modelagem não-linear da falta de alta impedância. O sistema fuzzy resultante foi capaz de estimar as distâncias de falta com um erro absoluto médio inferior a 500 m e um erro absoluto máximo da ordem de 1,5 km, em um alimentador com cerca de 18 km de extensão. Tais resultados equivalem a um grau de exatidão, para a maior parte das ocorrências, dentro do intervalo de ±10%.
Modern electric power systems present numerous challenges in its operation. Fault location is a major challenge in Power Distribution Systems due to its large branching, presence of single-phase laterals and the dynamic loads. The influence of the fault impedance is one of the largest, significantly affecting the use of traditional methods for its location, since the magnitude of the fault currents is similar to the load current. In this sense, this thesis aimed to develop an intelligent system for location of high impedance faults, which was based on the application of the decomposition technique of orthogonal components in the pre-processing variables and fuzzy inference to interpret the nonlinearities of Power Distribution Systems with the presence of Distributed Generation. The data for training the intelligent system were obtained from computer simulations of an actual feeder, considering a non-linear modeling of the high impedance fault. The resulting fuzzy system was able to estimate distances to fault with an average absolute error of less than 500 m and a maximum absolute error of 1.5 km order, on a feeder about 18 km long. These results are equivalent to a degree of accuracy for the most occurrences within the ± 10% range.

Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro
A análise de gases dissolvidos tem sido aplicada há décadas como a principal técnica de manutenção preditiva para diagnosticar defeitos incipientes em transformadores de potência, tendo em vista que a decomposição do óleo mineral isolante produz gases que permanecem dissolvidos na fase líquida. Entretanto, apesar da importância desta técnica, os métodos de diagnóstico mais conhecidos são baseados em constatações de modelos termodinâmicos e composicionais simplificados para a decomposição térmica do óleo mineral isolante, em conjunto com dados empíricos. Os resultados de simulação obtidos a partir desses modelos não reproduzem satisfatoriamente os dados empíricos. Este trabalho propõe um modelo termodinâmico flexível aprimorado para mimetizar o efeito da cinética de formação de sólidos como restrição ao equilíbrio e seleciona, entre quatro modelos composicionais, aquele que apresenta o melhor desempenho na simulação da decomposição térmica do óleo mineral isolante. Os resultados de simulação obtidos a partir do modelo proposto apresentaram uma melhor adequação a dados empíricos do que aqueles obtidos a partir dos modelos clássicos. O modelo propostofoi, ainda, aplicado ao desenvolvimento de um método de diagnóstico com base fenomenológica.Os desempenhos desta nova proposta fenomenológica e de métodos clássicos de diagnóstico por análise de gases dissolvidos foram comparados e discutidos; o método proposto alcançou desempenho superior a vários métodos usualmente empregados nessa área do conhecimento. E, ainda, um procedimento geral para a aplicação do novo método de diagnóstico é descrito
The dissolved gas analysis has been applied for decades as the main predictive maintenance technique for diagnosing incipient faults in power transformers since the decomposition of the mineral insulating oil produces gases that remain dissolved in the liquid phase. Nevertheless, the most known diagnostic methods are based on findings of simplified thermodynamic and compositional models for the thermal decomposition of mineral insulating oil, in addition to empirical data. The simulations results obtained from these models do not satisfactorily reproduce the empirical data. This work proposes a flexible thermodynamic model enhanced to mimic the kinetic effect of solid formation as a restriction to equilibrium and selects, among four compositional models, the one offering the best performance on the simulation of the thermal decomposition of mineral insulating oil. The simulation results obtained from the proposed model showed better adequacy to reported data than the results obtained from the classical models. The proposed model was also applied in the development of a diagnostic method with a phenomenological basis. The performances of this new phenomenological proposition and of classical dissolved gas analysis diagnostic methods are compared and discussed; the proposed method achieved a performance superior to several methods usually employed in this area of knowledge.Also, a general procedure for the application of the new diagnostic method is described

O objetivo principal desta tese consiste no desenvolvimento de ferramentas matemáticas e computacionais dedicadas a um sistema de diagnóstico de barras quebradas no rotor de Motores de Indução Trifásicos. O sistema proposto é baseado em um método matemático de decomposição de sinais elétricos, denominado de Decomposição em Componentes Ortogonais, e ferramentas de aprendizagem de máquinas. Como uma das principais contribuições desta pesquisa, realizou-se um aprofundamento do entendimento da técnica de Decomposição em Componentes Ortogonais e de sua aplicabilidade como ferramenta de processamento de sinais para sistemas elétricos e eletromecânicos. Redes Neurais Artificiais e Support Vector Machines, tanto para classificação multi-classes quanto para detecção de novidades, foram configurados para receber índices advindos do processamento de sinais elétricos de motores, e a partir deles, identificar os padrões normais e os padrões com falhas. Além disso, a severidade da falha também é diagnosticada, a qual é representada pelo número de barras quebradas no rotor. Para a avaliação da metodologia, considerou-se o acionamento de motores de indução pela tensão de alimentação da rede e por inversores de frequência, operando sob diversas condições de torque de carga. Os resultados alcançados demonstram a eficácia das ferramentas matemáticas e computacionais desenvolvidas para o sistema de diagnóstico, sendo que os índices criados se mostraram altamente correlacionados com o fenômeno da falha. Mais especificamente, foi possível criar índices monotônicos com a severidade da falha e com baixa variabilidade, demonstrando-se que as ferramentas são eficientes extratores de características.
This doctoral thesis consists of the development of mathematical and computational tools dedicated to a diagnostic system for broken rotor bars in Three Phase Induction Motors. The proposed system is based on a mathematical method for decomposing electrical signals, named the Orthogonal Components Decomposition, and machine learning tools. As one of the main contributions of this research, an in-depth investigation of the decomposition technique and its applicability as a signal processing tool for electrical and electromechanical systems was carried-out. Artificial Neural Networks and Support Vector Machines for multi-class classification and novelty detection were configured to receive indices derived from the processing of electrical signals and then identify normal motors and faulty motors. In addition, the fault severity is also diagnosed, which is represented by the number of broken rotor bars. Experimental data was tested in order to evaluate the proposed method. Signals were obtained from induction motors operating with different torque levels and driven either directly by the grid or by frequency inverters. The results demonstrate the effectiveness of the mathematical and computational tools developed for the diagnostic system since the indices created are highly correlated with the fault phenomenon. More specifically, it was possible to create monotonic indices with the fault severity and with low variability, what supports that the solution is an efficient fault-specific feature extractor.

L’histoire des événements accidentels dans les industries chimiques montre que leurs conséquences sont souvent graves sur les plans humain, environnemental et économique. Cette thèse vise à proposer une approche de détection et de diagnostic des défauts dans les procédés chimiques afin de prévenir ces événements accidentels. La démarche commence par une étude préalable qui sert à identifier les causes majeures responsables des événements industriels chimiques en se basant sur le retour d’expérience (REX). En France, selon la base de données ARIA, 25% des évènements sont dus à l’emballement thermique à cause d’erreurs d’origine humaine. Il est donc opportun de développer une méthode de détection et de diagnostic précoce des défauts dus à l’emballement thermique. Pour cela nous développons une approche qui utilise des seuils dynamiques pour la détection et la collecte de mesures pour le diagnostic. La localisation des défauts est basée sur une classification des caractéristiques statistiques de la température en fonction de plusieurs modes défectueux. Un ensemble de classificateurs linéaires et de diagrammes de décision binaires indexés par rapport au temps sont utilisés. Enfin, la synthèse de l'acide peroxyformique dans un réacteur discontinu et semi-continu est considérée pour valider la méthode proposée par des simulations numériques et ensuite expérimentales. Les performances de détection de défauts se sont révélées satisfaisantes et les classificateurs ont démontré un taux de séparabilité des défauts élevés
The history of accidental events in chemical industries shows that their human, environmental and economic consequences are often serious. This thesis aims at proposing an approach of detection and diagnosis faults in chemical processes in order to prevent these accidental events. A preliminary study serves to identify the major causes of chemical industrial events based on experience feedback. In France, according to the ARIA database, 25% of the events are due to thermal runaway because of human errors. It is therefore appropriate to develop a method for early fault detection and diagnosis due to thermal runaway. For that purpose, we develop an approach that uses dynamical thresholds for the detection and collection of measurements for diagnosis. The localization of faults is based on a classification of the statistical characteristics of the temperature according to several defectives modes. A multiset of linear classifiers and binary decision diagrams indexed with respect to the time are used for that purpose. Finally, the synthesis of peroxyformic acid in a batch and semi batch reactor is considered to validate the proposed method by numerical simulations and then experiments. Faults detection performance has been proved satisfactory and the classifiers have proved a high isolability rate of faults

En tant que géoscientifique dans le domaine de l’Exploration pétrolière et gazière depuis une vingtaine d’années, mes fonctions professionnelles m’ont permis d’effectuer différents travaux de recherche sur la thématique de la gestion des risques et des incertitudes. Ces travaux de recherche se situent sur l’ensemble de la chaîne d’analyse Exploration, traitant de problématiques liées à l’acquisition et au traitement sismique, jusqu’au placement optimal de forages d’exploration. Un volet plus poussé de mes travaux s’est orienté sur la gestion des incertitudes géophysiques en Exploration pétrolière, là où l’incertitude est la plus importante et paradoxalement la moins travaillée.On peut regrouper mes travaux de recherche en trois grands domaines qui suivent les grandes étapes du processus Exploration : le traitement sismique, leur interprétation, et enfin l'analyse et l'extraction des différentes incertitudes qui vont nous permettre de calculer les volumes d’hydrocarbures en place et récupérables, ainsi que l’analyse de ses risques associés. L’ensemble des travaux de recherche ont été appliqués avec succès sur des cas d’études opérationnelles. Après avoir introduit quelques notions générales et détaillé les grandes étapes du processus Exploration et leur lien direct avec ces problématiques, je présenterai quatre grands projets de recherche sur un cas d’étude algérien
In the last 20 years, I have been conducting various research projects focused on the management of risks and uncertainties in the petroleum exploration domain. The various research projects detailed in this thesis are dealing with problematics located throughout the whole Exploration and Production chain, from seismic acquisition and processing, until the optimal exploration to development wells placement. Focus is made on geophysical risks and uncertainties, where these problematics are the most pronounced and paradoxically the less worked in the industry. We can subdivide my research projects into tree main axes, which are following the hydrocarbon exploration process, namely: seismic processing, seismic interpretation thanks to the integration with various well informations, and eventually the analysis and extraction of key uncertainties, which will be the basis for the optimal calculation of in place and recoverable volumes, in addition to the associated risk analysis on a given target structure. The various research projects that are detailed in this thesis have been applied successfully on operational North Africa and North Sea projects. After introducing risks and uncertainty notions, we will detail the exploration process and the key links with these issues. I will then present four major research projects with their theoretical aspects and applied case study on an Algerian asset

碩士
國立中央大學
機械工程研究所
96
This thesis applies empirical mode decomposition in fault diagnosis of turbo-pump bearings. In the initial stage design simulation signals of fault bearing and do technical development as well as verify, next by electrical discharge machining and electrical discharge machining of deep hole in inner-ring and outer-ring of bearing to make defects technique for realizing the technique of breakdown diagnosis on two-plane rotor system and turbo-pump platform. Empirical mode decomposition no need to define basic function and transformation, which cause the defect of conventional envelope analysis with band-pass filtering could be improved during the research. With this character could the following phenomena be avoid: first, the hard obtainment about resonance frequency of subject; second, range of band-pass filtering could be easily influenced by subject in the time of choosing and result in the difference to result of envelope analysis. In the end compare the analysis from envelope analysis algorithm of empirical mode decomposition and the technique of conventional envelope analysis, then to discuss the advantage and deficiency between them. This thesis explains the practicality in development of on-line diagnosis technique of bearing monitoring, and in the future could be applied to other rotating machines for fault diagnosis.

碩士
國立勤益科技大學
電機工程系
107
This study proposes combining an extension neural network with the Chaos Theory and Empirical Mode Decomposition for power capacitor fault recognition, where the current data are measured and diagnosed for a power capacitor bank running at low voltage, and the capacitor current measurement is tested by a power testing machine. Afterwards, the Empirical Mode Decomposition is combined with the chaos synchronization detection method to analyze the voltage and current signals extracted by the high frequency oscillograph, and the dynamic chaos error scatter map using chaos eyes as the fault diagnosis feature is established. Finally, the extension neural network algorithm is used for capacitor fault detection, and the real -time status of the power capacitor is monitored by the developed human - machine interface. The advantage of the proposed method is that big data are compressed and meaningful eigenvalues are extracted, in order to effectively detect subtle changes in the power capacitor current signals, and diagnose the faults in the operating state of the power capacitor. According to the actual measurement result, the accuracy of the proposed method is as high as 95%, which is better than the extension theory (84%) and the multilayer artificial neural network (91%), proving this method is applicable to power capacitor discharge detection.

碩士
國立中央大學
機械工程研究所碩士在職專班
100
In this paper, the novel approach combining Hilbert-Huang Transform (HHT) and the multi-scale entropy (MSE) analysis is utilized for diagnosing the roller bearing faults, such as inner race defect, outer race defect and roller defect, under the operating conditions of variable rotation speeds. The vibration signals are first measured through the order tracking technique, so that the signals are sampled with identical angle increment and thus the vibration signals are stationary without the factor of shaft rotation speed. The vibration signals are then decomposed into a number of Intrinsic Mode Functions (IMFs) by using the Empirical Mode Decomposition (EMD) method. The envelope analysis is employed to the IMFs that have amplitude modulation phenomenon. The envelope signals are transformed to the series of different scales by course-grained process and MSE of the series can be calculated. With the extracted features of the MSEs, the decision tree algorithm is utilized to classify the different faulted bearing types and faulted levels.

碩士
國立中興大學
機械工程學系所
106
This study investigates the diagnosis of multiple faults that occur concurrently in the bearing through empirical mode decomposition and constrained independent component analysis. The vibration measurements are first decomposed into several intrinsic modal functions through the empirical mode decomposition method. The intrinsic mode functions that present obvious amplitude modulation phenomenon are selected to synthesize a new signal. The constrained independent component analysis is employed to extract the signal component which is highly correlated to the bearing fault features. The fast Fourier transform is utilized to obtain the frequency-domain features of the faulted signal, and the extracted features are compared with the one derived from the theoretical characteristics. The time-domain and frequency-domain characteristics of this independent component are quantified for the intelligent diagnosis through the support vector machine classifier.

碩士
國立中央大學
光機電工程研究所
97
Post processing of Ensemble Empirical Mode Decomposition (EEMD) can be utilized to decompose the vibration signals of rotating machinery into finite number of Intrinsic Mode Functions (IMFs) without mode mixing problem. The basis of the post processing of EEMD will satisfy the well-defined conditions of IMF. The Autoregressive (AR) model of information-contained IMFs can be used to predict the unmeasured vibration signal, and the coefficients of AR model represent the feature of systematic dynamic behavior. In this paper, the post-processing of EEMD combining the AR model is proposed for diagnosing the looseness faults at different conponents of rotating machinery. The information-contained IMFs are selected to build the AR model. The looseness types are identified by analyzing the coefficients of AR model. The effectiveness of the proposed method is validated through the analysis of the experimental data.