Aplicação de sistemas multi-classificadores no diagnóstico de falhas em motores de indução trifásicos

Equipment maintenance is the major cost factor in industrial plants, it is very important the development of fault predict techniques. Three-phase induction motors are key electrical equipments used in industrial applications mainly because presents low cost and large robustness, however, it isn t protected from other fault types such as shorted winding and broken bars. Several acquisition ways, processing and signal analysis are applied to improve its diagnosis. More efficient techniques use current sensors and its signature analysis. In this dissertation, starting of these sensors, it is to make signal analysis through Park s vector that provides a good visualization capability. Faults data acquisition is an arduous task; in this way, it is developed a methodology for data base construction. Park s transformer is applied into stationary reference for machine modeling of the machine s differential equations solution. Faults detection needs a detailed analysis of variables and its influences that becomes the diagnosis more complex. The tasks of pattern recognition allow that systems are automatically generated, based in patterns and data concepts, in the majority cases undetectable for specialists, helping decision tasks. Classifiers algorithms with diverse learning paradigms: k-Neighborhood, Neural Networks, Decision Trees and Naïves Bayes are used to patterns recognition of machines faults. Multi-classifier systems are used to improve classification errors. It inspected the algorithms homogeneous: Bagging and Boosting and heterogeneous: Vote, Stacking and Stacking C. Results present the effectiveness of constructed model to faults modeling, such as the possibility of using multi-classifiers algorithm on faults classification

Formas e função do discurso do outro no gênero monográfico

The present work investigates related discourse in rewriting discursive practices, at monographic works specifically at the theoretical foundation section. Focalizing some discursive strategies of voice management (direct and indirect discourse and modalization voice) we detach the introduction way and function of cited discourse. To do so, it were analyzed eighteen monographic works: nine of them final graduation works and other nine specialization works seeing that each works belonging to the same student, in two different stages, in the period from 2003 in graduation conclusion to 2005 in the end of specialization course. The data reveal that the monographic writer/student emphasizes the use of direct discourse in graduation works while in specialization works there was an emphasis at indirect speech. The analysis the way they introduce cited discourse pointed out that writer/student in graduation course such as specialization student make meaningless constructions when they do not use discendi verbs, they demonstrate difficulties inarticulate citing discourse with cited discourse. In what is related to functions of cited discourse we verify that the student/writer, in both stages or levels give emphasis to the function maintain an assertion, indicating that other s discourse serve mainly as a resource of authority just because that this function reveals the absence of a dialog between student writing and cited discourse. In a general way, the forms of other s discourse claim a form of writing that is found starting from a sequence of cited discourse in what student/writer voice in graduation and specialization comes to text surface just few times, but most of the times, the student takes other s words as they were themselves, every time there is an overlap of author/source

Ultrasonic density measurement cell design and simulation of non-ideal effects

This paper presents a theoretical analysis of a density measurement cell using an unidimensional model composed by acoustic and electroacoustic transmission lines in order to simulate non-ideal effects. The model is implemented using matrix operations, and is used to design the cell considering its geometry, materials used in sensor assembly, range of liquid sample properties and signal analysis techniques. The sensor performance in non-ideal conditions is studied, considering the thicknesses of adhesive and metallization layers, and the effect of residue of liquid sample which can impregnate on the sample chamber surfaces. These layers are taken into account in the model, and their effects are compensated to reduce the error on density measurement. The results show the contribution of residue layer thickness to density error and its behavior when two signal analysis methods are used. (c) 2006 Elsevier B.V. All rights reserved.

Robust fault diagnosis in power distribution systems based on fuzzy ARTMAP neural network-aided evidence theory

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

The photothermal transparent transducer method applied to the spectroscopy of liquids

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Razão eletromiográfica de músculos estabilizadores do ombro durante a execução de exercícios com haste oscilatória

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Análise quantitativa e qualitativa do nível de conhecimento dos alunos do curso de fisioterapia sobre a atuação do fisioterapeuta em saúde pública

Fisioterapia é a ciência da saúde que estuda, previne e trata os distúrbios cinéticos funcionais em órgãos e sistemas do corpo humano. O objetivo deste estudo é verificar a expectativa dos alunos do primeiro ao quinto semestre de fisioterapia sobre a atuação do fisioterapeuta em saúde pública e a expectativa desses alunos quanto à inserção do profissional de fisioterapia no Programa de Saúde da Família (PSF). Trata-se de estudo realizado na Faculdade Marechal Rondon, com 107 alunos, tendo como critério de exclusão os alunos do sétimo semestre. Foi usado um questionário, contendo onze questões, sendo oito de múltipla escolha e três dissertativas. Os dados passaram por tratamento estatístico, em que foram utilizadas a análise descritiva através do programa Microsoft Excel 2003 e análise do discurso do sujeito coletivo. A respeito da atuação do fisioterapeuta no Sistema Único de Saúde (SUS), 44% dos alunos consideram muito importante, 36% consideraram que o papel do fisioterapeuta no PSF é muito importante, enquanto 24% dos alunos consideram não saber informar quanto à eficácia de seu atendimento. Os alunos consideram importante a atuação do fisioterapeuta no SUS e PSF, mas pouco tem conhecimento sobre a atuação do fisioterapeuta em saúde pública.

Activation function study for wavelet network

The main purpose of this paper is to investigate theoretically and experimentally the use of family of Polynomial Powers of the Sigmoid (PPS) Function Networks applied in speech signal representation and function approximation. This paper carries out practical investigations in terms of approximation fitness (LSE), time consuming (CPU Time), computational complexity (FLOP) and representation power (Number of Activation Function) for different PPS activation functions. We expected that different activation functions can provide performance variations and further investigations will guide us towards a class of mappings associating the best activation function to solve a class of problems under certain criteria.

False alarms in damage detection approaches

Structural Health Monitoring (SHM) denotes a system with the ability to detect and interpret adverse changes in structures in order to improve reliability and reduce life-cycle costs. The greatest challenge for designing a SHM system is knowing what changes to look for and how to classify them. Different approaches for SHM have been proposed for damage identification, each one with advantages and drawbacks. This paper presents a methodology for improvement in vibration signal analysis using statistics information involving the probability density. Generally, the presence of noises in input and output signals results in false alarms, then, it is important that the methodology can minimize this problem. In this paper, the proposed approach is experimentally tested in a flexible plate using a piezoelectric (PZT) actuator to provide the disturbance.

Fuzzy logic to predict thermal damages of ground parts

One of the critical problems in implementing an intelligent grinding process is the automatic detection of workpiece surface burn. This work uses fuzzy logic as a tool to classify and predict burn levels in the grinding process. Based on acoustic emission signals, cutting power, and the mean-value deviance (MVD), linguistic rules were established for the various burn situations (slight, intermediate, severe) by applying fuzzy logic using the Matlab Toolbox. Three practical fuzzy system models were developed. The first model with two inputs resulted only in a simple analysis process. The second and third models have an additional MVD statistic input, associating information and precision. These two models differ from each other in terms of the rule base developed. The three developed models presented valid responses, proving effective, accurate, reliable and easy to use for the determination of ground workpiece burn. In this analysis, fuzzy logic translates the operator's human experience associated with powerful computational methods.

Parkinson's disease identification through Optimum-Path Forest

Artificial intelligence techniques have been extensively used for the identification of several disorders related with the voice signal analysis, such as Parkinson's disease (PD). However, some of these techniques flaw by assuming some separability in the original feature space or even so in the one induced by a kernel mapping. In this paper we propose the PD automatic recognition by means of Optimum-Path Forest (OPF), which is a new recently developed pattern recognition technique that does not assume any shape/separability of the classes/feature space. The experiments showed that OPF outperformed Support Vector Machines, Artificial Neural Networks and other commonly used supervised classification techniques for PD identification. © 2010 IEEE.

ECG arrhythmia classification based on optimum-path forest

An important tool for the heart disease diagnosis is the analysis of electrocardiogram (ECG) signals, since the non-invasive nature and simplicity of the ECG exam. According to the application, ECG data analysis consists of steps such as preprocessing, segmentation, feature extraction and classification aiming to detect cardiac arrhythmias (i.e.; cardiac rhythm abnormalities). Aiming to made a fast and accurate cardiac arrhythmia signal classification process, we apply and analyze a recent and robust supervised graph-based pattern recognition technique, the optimum-path forest (OPF) classifier. To the best of our knowledge, it is the first time that OPF classifier is used to the ECG heartbeat signal classification task. We then compare the performance (in terms of training and testing time, accuracy, specificity, and sensitivity) of the OPF classifier to the ones of other three well-known expert system classifiers, i.e.; support vector machine (SVM), Bayesian and multilayer artificial neural network (MLP), using features extracted from six main approaches considered in literature for ECG arrhythmia analysis. In our experiments, we use the MIT-BIH Arrhythmia Database and the evaluation protocol recommended by The Association for the Advancement of Medical Instrumentation. A discussion on the obtained results shows that OPF classifier presents a robust performance, i.e.; there is no need for parameter setup, as well as a high accuracy at an extremely low computational cost. Moreover, in average, the OPF classifier yielded greater performance than the MLP and SVM classifiers in terms of classification time and accuracy, and to produce quite similar performance to the Bayesian classifier, showing to be a promising technique for ECG signal analysis. © 2012 Elsevier Ltd. All rights reserved.

Análise de distúrbios de tensão em sistemas de distribuição de energia elétrica baseada em sistemas imunológicos artificiais

Pós-graduação em Engenharia Elétrica - FEIS

Análise do princípio centrando do método pilates e de indicadores de resistência e estabilidade da coluna lombar

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Implementação de técnicas de processamento de sinais para o monitoramento da condição de mancais de rolamento

Pós-graduação em Engenharia Mecânica - FEG