CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / O monitoramento e o diagnÃstico de falhas incipientes em transformadores de potÃncia imersos em Ãleo estÃo diretamente relacionados à avaliaÃÃo das condiÃÃes do sistema de isolamento. Neste estudo, estabelece-se o conceito de monitoramento e diagnÃstico, e em seguida tÃcnicas de monitoramento on-line sÃo discutidas. Um sistema de prÃ-diagnÃstico à elaborado baseado na utilizaÃÃo de um dispositivo on-line de monitoramento, Hydran da GE, para classificar a gravidade da situaÃÃo de falha detectada. Uma vez detectada uma situaÃÃo de falha, mÃdulos inteligentes de diagnÃstico de falhas incipientes, via redes neurais, podem ser utilizados para identificaÃÃo da falha interna do equipamento. Para completar a verificaÃÃo da qualidade dielÃtrica do lÃquido isolante, tambÃm à descrito um algoritmo inteligente, baseado em redes neurais, para diagnÃstico do estado do Ãleo atravÃs das grandezas fÃsico-quÃmicas. A relaÃÃo entre os atributos fÃsico-quÃmicos e as grandezas cromatogrÃficas referente ao Ãleo mineral tambÃm foram averiguadas. Foi desenvolvida, entÃo, a estimaÃÃo dos gases dissolvidos atravÃs das caracterÃsticas fÃsico-quÃmicas. Os mÃdulos de monitoramento on-line, diagnÃsticos do estado do Ãleo e de falhas incipientes, alÃm da estimaÃÃo dos gases dissolvidos, perfazem um sistema computacional de auxÃlio à operaÃÃo e manutenÃÃo. O sistema implementado apresenta resultados satisfatÃrios na implantaÃÃo em uma planta de usina termelÃtrica. / The monitoring and diagnosis of incipient fault in power transformers immerses in oil are directly related to the assessment of the isolation system conditions. In this research, it is established the concept of monitoring and diagnosis, after that, on-line monitoring techniques are discussed. A pre-diagnosis system is elaborated based on use of a monitoring on-line device, Hydran GE, to classify the situation gravity of the detected fault. Once detected a fault situation, intelligent modules of incipient fault diagnosis, by neural networks, can be used to identification of internal fault of the equipment. To complete the checking of the dielectric quality of the isolate liquid, it is also described an intelligent algorithm, based on neural networks, to diagnosis of the oil estate through physical-chemical attribute. The relation between physical-chemical attributes and chromatographic ones regarding to mineral oil were also verified. It was developed, then, the dissolved gases esteem through physicalchemical characteristics. The on-line monitoring modules, diagnosis of oil estate and incipient fault, besides dissolved gases esteem, constitute a computation aid system to operation and maintenance. The implemented system presents satisfied results in a thermoelectric power plant.

Orientador: Robson Pederiva / Dissertação (mestrado) - Universidade Estadual de Campinas. Faculdade de Engenharia Mecanica / Made available in DSpace on 2018-08-07T00:56:52Z (GMT). No. of bitstreams: 1 Silva_GilbertoMachadoda_M.pdf: 1128289 bytes, checksum: e5a6e2d7aeb16be17a0a3d45dd834193 (MD5) Previous issue date: 2006 / Resumo: Aplica-se neste trabalho a metodologia de diagnóstico de falhas em sistemas mecânicos rotativos sustentados por mancais magnéticos ativos em conjunto com um sistema de controle ativo de vibração, excitados por forças de desbalanço e ruído branco. Este diagnóstico baseia-se no emprego das equações de correlações, através da formulação matricial de Ljapunov, para sistemas lineares estacionários juntamente com redes neurais artificiais. Este procedimento utiliza apenas as variáveis de estado medidas, através da correlação das variáveis de saída. É gerado um conjunto de relações envolvendo os parâmetros físicos do sistema juntamente com as matrizes de correlações das variáveis medidas. As falhas no sistema são detectadas através do monitoramento da variação dos parâmetros físicos e da comparação das funções de correlação teóricas e estimadas. As redes neurais artificiais são usadas para mapear as correlações que envolvem estados que não são medidos. Dado ao grande número de equações de compatibilidade resultantes, é proposta uma metodologia para selecionar as equações relacionadas com as falhas propostas. Com o método de diagnóstico de falhas proposto é possível detectar e discernir as falhas tanto mecânicas quanto elétricas, bem como sua localização no sistema / Abstract: This work applies the methodology of fault diagnosis in rotating machinery supported by active magnetic bearings and active con1rol systems, excited by unbalance and white noise. This diagnostic applies the correlation matrices based on the Ljapunov matrix formulation and artificial neural network for linear stationary systems. The procedure uses only measured state variables, computing the correlation between the output variables. It is possible to derive specific relations involving the physical parameter of the system and the correlation matrices of the measured variables. Faults in the system can be detected by monitoring the variation of the physical parameter through a comparison of theoretical and estimated correlation functions. Artificial neural networks are used to map the correlations involving the variables, which are difficult to be measured. There is a large number of resultant compatibility equations and it is proposed a methodology to select the equations that establish relationships with the faults. The proposed fault diagnosis method can detect the fault present in the system and it is also possible to distinguish between mechanical and electrical fault as well as their location in the system / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestre em Engenharia Mecânica

The problem of fault diagnosis consists of detecting and isolating faults present in a system. As technical systems become more and more complex and the demands for safety, reliability and environmental friendliness are rising, fault diagnosis is becoming increasingly important. One example is automotive systems, where fault diagnosis is a necessity for low emissions, high safety, high vehicle uptime, and efficient repair and maintenance. One approach to fault diagnosis, providing potentially good performance and in which the need for additional hardware is minimal, is model-based fault diagnosis with residuals. A residual is a signal that is zero when the system under diagnosis is fault-free, and non-zero when particular faults are present in the system. Residuals are typically generated by using a mathematical model of the system and measurements from sensors and actuators. This process is referred to as residual generation. The main contributions in this thesis are two novel methods for residual generation. In both methods, systems described by Differential-Algebraic Equation (DAE) models are considered. Such models appear in a large class of technical systems, for example automotive systems. The first method consider observer-based residual generation for linear DAE-models. This method places no restrictions on the model, such as e.g. observability or regularity, in comparison with other previous methods. If the faults of interest can be detected in the system, the output from the design method is a residual generator, in state-space form, that is sensitive to the faults of interest. The method is iterative and relies on constant matrix operations, such as e.g. null-space calculations and equivalence transformations. In the second method, non-linear DAE-models are considered. The proposed method belongs to a class of methods, in this thesis referred to as sequential residual generation, which has shown to be successful for real applications. This method enables simultaneous use of integral and derivative causality, and is able to handle equation sets corresponding to algebraic and differential loops in a systematic manner. It relies on a formal framework for computing unknown variables in the model according to a computation sequence, in which the analytical properties of the equations in the model as well as the available tools for equation solving are taken into account. The method is successfully applied to complex models of an automotive diesel engine and a hydraulic braking system.

The increasing numbers and complexity of spacecraft is driving a growing need for automated fault detection, isolation, and recovery. Anomalies and failures are common occurrences during space flight operations, yet most spacecraft currently possess limited ability to detect them, diagnose their underlying cause, and enact an appropriate response. This leaves ground operators to interpret extensive telemetry and resolve faults manually, something that is impractical for large constellations of satellites and difficult to do in a timely fashion for missions in deep space. A traditional hurdle for achieving autonomy has been that effective fault detection, isolation, and recovery requires appreciating the wider context of telemetry information. Advances in machine learning are finally allowing computers to succeed at such tasks. This dissertation presents an architecture based on machine learning for detecting, diagnosing, and responding to faults in a spacecraft attitude determination and control system. Unlike previous approaches, the availability of faulty examples is not assumed. In the first level of the system, one-class support vector machines are trained from nominal data to flag anomalies in telemetry. Meanwhile, a spacecraft simulator is used to model the activation of anomaly flags under different fault conditions and train a long short-term memory neural network to convert time-dependent anomaly information into a diagnosis. Decision theory is then used to convert diagnoses into a recovery action. The overall technique is successfully validated on data from the LightSail 2 mission. <br>

Recent refinements in Industry 4.0 and Machine Learning demonstrate the positive effects of using deep learning models for intelligent maintenance. The primary benefit of Deep Learning (DL) is its capability to extract attributes and make fast, accurate, and automated predictions without supervision. However, DL requires high computational power, significant data preprocessing, and vast amounts of data to make accurate predictions for intelligent maintenance. Given the considerable obstacles, meta-learning has been developed as a novel way to overcome these challenges. As a learning technique, meta-learning aims to quickly acquire knowledge of new tasks using theminimal available data by learning through meta-knowledge. There has been less research in the area of using meta-learning for Predictive Maintenance (PdM) and we considered it necessary to conduct this review to understand the applicability of meta-learning’s capabilities and functions to PdM since the outcomes of this technique seem to be rather promising. The review started with the development of a methodology and four research questions: (1) What is the taxonomy of meta-learning for PdM?, (2) What are the current state-of-the-art methodologies? (3) Which datasets are available for meta-learning in PdM?, and (4) What are the open issues, challenges, and opportunities of meta-learning in PdM?. To answer the first and second questions, a new taxonomy was proposed and meta-learnings role in predictive maintenance was identified from selected 55 papers. To answer the third question, we determined which types of datasets and their characteristics exist for this domain. Finally, the challenges, open issues, and opportunities of meta-learning in predictive maintenance were examined to answer the final question. The results of the research questions provided suggestions for future research topics.

No description available.

No description available.

No description available.

No description available.

No description available.