O objetivo deste trabalho é estudar a identificação, classificação, localização e setorização de faltas em redes de distribuição radiais, verificar a maneira de aplicar e integrar diversas ferramentas numéricas convencionais, assim como ferramentas de sistemas inteligentes, visando identificar a ocorrência de uma falta, classificar as fases envolvidas com a mesma, e aplicar as diversas ferramentas existentes a fim de localizar em tempo real o eventual local onde houve a ocorrência da falta, permitindo que a mesma possa ser setorizada dentro do sistema da concessionária e informada ao centro de operações, usando uma nova abordagem inteligente baseada em análise espectral de sinais. / The aim of this work is study the identification, classification, location and sectorization of a fault in distribution radial networks, check how to implement and integrate various conventional numerical tools, as well as intelligent systems based tools, to identify the occurrence of a fault, classify the phases involved with it, and apply the various tools available to locate the place where a fault was occurred in real time, enabling it to be sectorized into the utility system and informed to operational center using a new intelligent approach based on spectral signals analysis.

Governmental and private institutions rely heavily on reliable computer networks for their everyday business transactions. The downtime of their infrastructure networks may result in millions of dollars in cost. Fault management systems are used to keep today’s complex networks running without significant downtime cost, either by using active techniques or passive techniques. Active techniques impose excessive management traffic, whereas passive techniques often ignore uncertainty inherent in network alarms,leading to unreliable fault identification performance. In this research work, new algorithms are proposed for both types of techniques so as address these handicaps. Active techniques use probing technology so that the managed network can be tested periodically and suspected malfunctioning nodes can be effectively identified and isolated. However, the diagnosing probes introduce extra management traffic and storage space. To address this issue, two new CSP (Constraint Satisfaction Problem)-based algorithms are proposed to minimize management traffic, while effectively maintain the same diagnostic power of the available probes. The first algorithm is based on the standard CSP formulation which aims at reducing the available dependency matrix significantly as means to reducing the number of probes. The obtained probe set is used for fault detection and fault identification. The second algorithm is a fuzzy CSP-based algorithm. This proposed algorithm is adaptive algorithm in the sense that an initial reduced fault detection probe set is utilized to determine the minimum set of probes used for fault identification. Based on the extensive experiments conducted in this research both algorithms have demonstrated advantages over existing methods in terms of the overall management traffic needed to successfully monitor the targeted network system. Passive techniques employ alarms emitted by network entities. However, the fault evidence provided by these alarms can be ambiguous, inconsistent, incomplete, and random. To address these limitations, alarms are correlated using a distributed Dempster-Shafer Evidence Theory (DSET) framework, in which the managed network is divided into a cluster of disjoint management domains. Each domain is assigned an Intelligent Agent for collecting and analyzing the alarms generated within that domain. These agents are coordinated by a single higher level entity, i.e., an agent manager that combines the partial views of these agents into a global one. Each agent employs DSET-based algorithm that utilizes the probabilistic knowledge encoded in the available fault propagation model to construct a local composite alarm. The Dempster‘s rule of combination is then used by the agent manager to correlate these local composite alarms. Furthermore, an adaptive fuzzy DSET-based algorithm is proposed to utilize the fuzzy information provided by the observed cluster of alarms so as to accurately identify the malfunctioning network entities. In this way, inconsistency among the alarms is removed by weighing each received alarm against the others, while randomness and ambiguity of the fault evidence are addressed within soft computing framework. The effectiveness of this framework has been investigated based on extensive experiments. The proposed fault management system is able to detect malfunctioning behavior in the managed network with considerably less management traffic. Moreover, it effectively manages the uncertainty property intrinsically contained in network alarms,thereby reducing its negative impact and significantly improving the overall performance of the fault management system.

Governmental and private institutions rely heavily on reliable computer networks for their everyday business transactions. The downtime of their infrastructure networks may result in millions of dollars in cost. Fault management systems are used to keep today’s complex networks running without significant downtime cost, either by using active techniques or passive techniques. Active techniques impose excessive management traffic, whereas passive techniques often ignore uncertainty inherent in network alarms,leading to unreliable fault identification performance. In this research work, new algorithms are proposed for both types of techniques so as address these handicaps. Active techniques use probing technology so that the managed network can be tested periodically and suspected malfunctioning nodes can be effectively identified and isolated. However, the diagnosing probes introduce extra management traffic and storage space. To address this issue, two new CSP (Constraint Satisfaction Problem)-based algorithms are proposed to minimize management traffic, while effectively maintain the same diagnostic power of the available probes. The first algorithm is based on the standard CSP formulation which aims at reducing the available dependency matrix significantly as means to reducing the number of probes. The obtained probe set is used for fault detection and fault identification. The second algorithm is a fuzzy CSP-based algorithm. This proposed algorithm is adaptive algorithm in the sense that an initial reduced fault detection probe set is utilized to determine the minimum set of probes used for fault identification. Based on the extensive experiments conducted in this research both algorithms have demonstrated advantages over existing methods in terms of the overall management traffic needed to successfully monitor the targeted network system. Passive techniques employ alarms emitted by network entities. However, the fault evidence provided by these alarms can be ambiguous, inconsistent, incomplete, and random. To address these limitations, alarms are correlated using a distributed Dempster-Shafer Evidence Theory (DSET) framework, in which the managed network is divided into a cluster of disjoint management domains. Each domain is assigned an Intelligent Agent for collecting and analyzing the alarms generated within that domain. These agents are coordinated by a single higher level entity, i.e., an agent manager that combines the partial views of these agents into a global one. Each agent employs DSET-based algorithm that utilizes the probabilistic knowledge encoded in the available fault propagation model to construct a local composite alarm. The Dempster‘s rule of combination is then used by the agent manager to correlate these local composite alarms. Furthermore, an adaptive fuzzy DSET-based algorithm is proposed to utilize the fuzzy information provided by the observed cluster of alarms so as to accurately identify the malfunctioning network entities. In this way, inconsistency among the alarms is removed by weighing each received alarm against the others, while randomness and ambiguity of the fault evidence are addressed within soft computing framework. The effectiveness of this framework has been investigated based on extensive experiments. The proposed fault management system is able to detect malfunctioning behavior in the managed network with considerably less management traffic. Moreover, it effectively manages the uncertainty property intrinsically contained in network alarms,thereby reducing its negative impact and significantly improving the overall performance of the fault management system.

Changes in work, along with improvements in techniques to statistically model uncertainty, have resulted in a class of groupware tools able to forecast the activities and/or attentional state of their users. This thesis represents an exploration into the design, development, and use of one such system. I describe the design and development of a groupware calendar system called Augur that is augmented with the ability to predict the attendance of its users. Using Bayesian networks, Augur models the uncertain problem of event attendance, drawing inferences based on the attributes of calendar events as well as a history of attendance provided by each user. This system was deployed to an academic workgroup and studied over the course of a semester. To more deeply explore the social implications of Augur and systems like it, I conducted a structured privacy analysis of Augur to examine the vulnerabilities inherent in this type of forecasting groupware system. I present an architecture, user interface, and probabilistic model for Augur. This work also addresses the feasibility of such a system and the challenges faced when deploying it to an academic workgroup. I also report on an exploration of the systems use by individuals, its effects on communication within working relationships, and its effectiveness with respect to the presence of domestic calendars. Finally, I present a set of implications for the workplace social environment with the introduction of Augur. Specifically, I show how the integrity of predictions generated by Augur can have consequences for the privacy of users and their representations through the shared calendar. Overall, this thesis is presented as an early exploration into the potential for a new class of forecasting groupware applications. It offers guidance and lessons learned for both designers and researchers seeking to work in this area. It also presents a complete calendar application as an example for building and studying such systems.

Tecnologia de montagem de superfície (SMT) é um método para construção de circuitos eletrônicos, nos quais os componentes são montados diretamente sobre a superfície da placa de circuito impresso. Tais dispositivos eletrônicos são chamados de dispositivos de montagens de superfície ou SMDs. Paralelamente, as vantagens oferecidas pelo componente eletrônico LED SMD têm causado uma grande aplicação desse dispositivo em substituição ao LED convencional. O presente trabalho apresenta um sistema inteligente baseado em sistemas de inferência fuzzy para estimar valores de intensidade luminosa de equipamentos automotivos a partir de dados de projeto. Embora o trabalho esteja direcionado para a aplicação de LEDs SMD em lanternas traseiras, o trabalho aqui desenvolvido pode ser generalizado e usado em outras aplicações industriais, tais como semáforos de trânsito, painéis eletrônicos de mensagens ou qualquer outra aplicação onde use LEDs SMD em conjunto. Resultados de protótipos são apresentados para validar a técnica proposta. Por meio desses resultados, pode-se observar que a aplicação de sistemas inteligentes é uma abordagem atrativa para este tipo de problema. / Surface mount technology (SMT) is a method for making electronic circuits in which the components are mounted directly onto the surface of printed circuit boards. Such electronic devices are called surface-mount devices or SMDs. The advantages offered by the electronic component SMD LED (Light Emitting Diode) have caused a wide application of this device in replacement of conventional LEDs. This work shows an intelligent system using fuzzy interference systems to estimate values of luminous intensity in automotive equipments from design data. Although this work is aimed to the application of SMD LEDs in rear lights, methods hereby developed and described can also be used in other applications, such as traffic lights, electronic panels of messages or any other application where SMD LEDs are used in groups. Results of prototypes are presented to validate the proposed technique. From these results, it can be observed that the application of intelligent systems is an attractive approach to this type of problem.

Nesta dissertação apresentaremos um m´método para detecção automática da quantidade de carga adicionada em uma lavadora de roupas para que esta possa adequar o seu nível de água e ciclo de lavagem. A máquina na qual este algoritmo foi desenvolvido é uma máquina de eixo vertical (abertura na parte de cima) e utiliza um motor síncrono trapezoidal (BPM do inglês Brushless Permanent Magnet). O algoritmo que será descrito aqui utiliza uma rede neural para inferir a quantidade de carga baseado em informações disponíveis nesta m´máquina como, corrente do motor, velocidade do cesto e tensão de alimentação, entre outros, essas informações estão disponíveis na maioria dos modelos de máquinas de lavar roupas que utilizam esse tipo de motor. A utilização de um algoritmo para detectar automaticamente e de forma precisa a quantidade de roupas é muito importante, pois dessa forma evita-se o desperdício de insumos e, principalmente, água no processo de lavagem. Além disso apresentaremos os resultados que mostram a diferença entre o uso da rede neural e o método linear chamado planejamento de experimento (DOE do inglês Design of Experiments). / In this dissertation a method for automatic load size detection will be presented, so the water level and the washing cycles can be chosen by a washing machine. The machine where this algorithm was developed is a top load washing machine that uses a brushless permanent magnet motor (BPM motor). The algorithm that is going to be described here uses a neural network to deduce the load size based on information available on this machine such as, motor current, basket speed, power supply voltage and others. These signals are available on most washing machines that uses this kind of motor. The use of an algorithm that detects automatically and precisely the load amount is very important in order to avoid the waste of soap, bleach and softner and, more importantly, water during the wash task. Moreover the use of the neural network will be compared with a linear methods called DOE (design of experiment). Finally, results showing the difference between both methods are presented.

Submitted by Vanessa Christiane (referencia@ufersa.edu.br) on 2018-06-15T00:08:00Z No. of bitstreams: 1 IngridyMPB_DISSERT.pdf: 7607381 bytes, checksum: c08a68ab4cbff8326651716b5e44e1d8 (MD5) / Approved for entry into archive by Vanessa Christiane (referencia@ufersa.edu.br) on 2018-06-18T16:55:37Z (GMT) No. of bitstreams: 1 IngridyMPB_DISSERT.pdf: 7607381 bytes, checksum: c08a68ab4cbff8326651716b5e44e1d8 (MD5) / Approved for entry into archive by Vanessa Christiane (referencia@ufersa.edu.br) on 2018-06-18T16:56:28Z (GMT) No. of bitstreams: 1 IngridyMPB_DISSERT.pdf: 7607381 bytes, checksum: c08a68ab4cbff8326651716b5e44e1d8 (MD5) / Made available in DSpace on 2018-06-18T16:56:45Z (GMT). No. of bitstreams: 1 IngridyMPB_DISSERT.pdf: 7607381 bytes, checksum: c08a68ab4cbff8326651716b5e44e1d8 (MD5) Previous issue date: 2018-02-02 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The use of mobile devices for the continuous monitoring of patients with a particular pathology may benefit in their prevention, diagnosis and treatment. The objective of this work is the construction of a Medical Assistance Equipment (EMA) for the monitoring of patients with pathology related to di culty in the food intake process and/or Oropharyngeal Dysphagia. The EMA works by capturing the movements and sound signals generated during the chewing and swallowing process and, from this, identifies the physical type of the food, classifying it as: i) liquid; ii) pasty; iii) solid. It is important to emphasize the requirements for a construction of EMA were taken from the context of Dysphagia. After a classification, the information is stored generating a food history, with detailed information about the meals performed and the distribution of Dysphagia that the patient is. To this end, a domain ontology was implemented with logical axioms capable of classifying the type of physical material of the swallowed food based on the analysis of the data captured during the swallowing process. In order to analyze the results generated by the EMA, experiments were carried out in a real environment with 10 participants, authorized by the ethics committee under the following opinion number: 2.332.026. Each participant was invited to swallow liquid, pasty and solid foods. The data generated by the participants were analyzed and classified by the developed ontology. At the end, the results presented 100% of correct answers in relation to experiments with solid foods, 80% of correct answers in experiments with liquid foods and 75% of correct answers in relation to experiments with pasty foods. A general analysis of the EMA presented the safety area of 85%. Finally, the EMA provided relevant results regarding correct classification of data. Thus, a medical team can monitor, from a distance, the patient’s evolution onwards, the detailed information available, not EMA, facilitating the monitoring process and improving a quality of life of patients requiring remote monitoring / O uso de dispositivos móveis para o monitoramento contínuo de pacientes com uma determinada patologia pode beneficiar significativamente em sua prevenção, diagnóstico e tratamento. O objetivo deste trabalho é a construção de um Equipamento Médico Assistencial (EMA) para o monitoramento de pacientes com alguma patologia relacionada à dificuldade no processo de ingestão de alimentos e/ou Disfagia Orofaríngea. O EMA funciona através da captura dos movimentos e dos sinais sonoros gerados durante o processo de mastigação e deglutição, e, a partir disso, identifica o tipo físico do alimento, classificando-o em: i) líquido; ii) pastoso; iii) sólido. É importante ressaltar que os requisitos definidos para a construção do EMA foram retirados do contexto da Disfagia. Após a classificação, essas informações são armazenadas gerando um histórico alimentar com informações detalhadas sobre as refeições realizadas e o nível de Disfagia que o paciente se encontra. Para este fim, foi implementada uma ontologia de domínio com axiomas lógicos capazes de classificar o tipo físico do alimento deglutido tendo por base a análise dos dados capturados durante o processo de deglutição. Com o intuito de analisar os resultados gerados pelo EMA, foram realizados experimentos em ambiente real com 10 participantes, autorizado pelo comitê de ética sob o seguinte número de parecer: 2.332.026. Cada participante foi convidado a deglutir alimentos líquidos, pastosos e sólidos. Os dados gerados pelos participantes foram analisados e classificados pela ontologia desenvolvida. Ao final, os resultados apresentam 100% de acerto em relação aos experimentos realizados com alimentos sólidos, 80% de acerto em relação aos experimentos realizados com alimentos líquidos e 75% de acerto em relação aos experimento realizados com alimentos pastosos. A análise geral do EMA apresentou o nível de acurácia de 85%. Por fim, o EMA proporcionou resultados relevantes quanto à classificação correta de dados. Assim, a equipe médica pode acompanhar, a distância, a evolução do paciente diante das informações detalhadas disponíveis no EMA, facilitando o processo de monitoramento e melhorando a qualidade de vida dos pacientes que necessitam de acompanhamento a distância / 2018-06-14

Made available in DSpace on 2014-06-11T19:22:34Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-12-13Bitstream added on 2014-06-13T20:09:49Z : No. of bitstreams: 1 geronimo_tm_me_bauru.pdf: 1704386 bytes, checksum: 1a6dd612ef17da8f95f721e29761eddc (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O monitoramento de processos de fabricação pro usinagem tem se mostrado de extrema importância nas empresas que buscam um nível de excelência em qualidade, auxiliando na melhor alocação de recursos e redução de desperdícios oriundos de peças com problemas de qualidade. Abordagens multisensoriais têm sido empregadas no monitoramento desses processos com o objetivo de utilizar os sinais no treinamento de sistemas de inteligência artificial na tarefa de indicar desvios nas ferramentas ou no produto sendo fabricado. Neste trabalho, três sistemas de inteligência artificial foram utilizados com o o objetivo de prover estimativas para o diâmetro de furos obtidos por processo de furação de precisão. Uma rede neural artificial perceptron de múltiplas camadas (RNA MLP), um sistema de inerferência adaptável neuro-fuzzy (ANFIS) e uma rede neural artificial com função de base radial (RBF) foram treinados com os dados obtidos com os sensores para estimar os diâmetros dos furos para cada material de corpo-de-prova. A definição da melhor configuração para cada sistema de inteligência artificial foi obtida através de algoritmos desenvolvidos para verificar a influência dos sinais e dos parâmetros particulares de cada sistema sobre o resultado final da estimativa. Os resultados obtidos indicam que a RNA MLP apresenta maior robutez perante a variação nos dados apresentados. O sistema ANFIS e a rede RFB mostraram que seu resultado varia acentuadamente quando há variações nos dados apresentados no treinamento, requerendo que estes sistemas sejam treinados sempre com o conjunto de dados apresentados na mesma ordem. A análise de influência dos sinais mostrou que, embora a abordagem multisensorial apresente bons resultados na rede MLP, o mesmo não se repetiu para os demais sistemas... / The supervision of manufacturing process by machining has been extremely important in companies which aim an excellence level in quality, helping on best assets allocation and waste reduction originated from pieces with quality problems. Multi-sensory approaches have been employed in the supervision of these processes aiming the use of signals in the training of artificial intelligence systems in order to indicate deviations in tools or in product being manufactured. Turning, grinding, milling and drilling have benn the target of the application of these supervision intelligence systems. In this work these artificial intelligence systems were applied in order to provide estimations for the diameters of the holes obtained by precision drilling process. A Multilayer Perceptron Neural Network (ANN MLP), and adaptive neuro-fuzzy inference system (ANFIS) and an artificial neural network with radial basis function (RBF) were trained with the data obtained from the sensors to estimate the hole diameters for each material of the test pieces. The definition of the best configuration for each artificial intelligence system was obtained through algorithms developed to verify the signals influence and particular parameters of each system concerning the final estimation result. The tests results were analyzed under three criteria: the absolute and medium errors, the system capacity of obtaining correct results for each estimation - classifying them as approved or rejected - and the error analysis regarding the necessary tolerance classes to maintain process within the limits of precision mechanics. The results obtained indicate that the ANN MLP presents higher robustness before variation in the data presented. The ANFIS system and RFB network have shown that their result vary sharply when there are data variations presented in training... (Complete abstract click electronic access below)

Submitted by DANILO DE MELO BRUNINI null (dbrunini@gmail.com) on 2017-07-14T02:28:32Z No. of bitstreams: 1 Dissertacao_Final_DaniloBrunini.pdf: 4098580 bytes, checksum: f552cf864fb92ba3889ebf437d9c11d7 (MD5) / Approved for entry into archive by Monique Sasaki (sayumi_sasaki@hotmail.com) on 2017-07-14T19:08:26Z (GMT) No. of bitstreams: 1 brunini_dm_me_bauru.pdf: 4098580 bytes, checksum: f552cf864fb92ba3889ebf437d9c11d7 (MD5) / Made available in DSpace on 2017-07-14T19:08:26Z (GMT). No. of bitstreams: 1 brunini_dm_me_bauru.pdf: 4098580 bytes, checksum: f552cf864fb92ba3889ebf437d9c11d7 (MD5) Previous issue date: 2017-05-31 / Os transformadores são equipamentos importantes do sistema elétrico de potência, possuem alto custo e suas falhas tem influência direta na qualidade da energia entregue aos consumidores. Uma das principais causas de falhas em transformadores imersos em líquido isolante, as descargas parciais, advém da degradação física e química do sistema de isolação devido à diversos fatores tais como sobrecarga, cargas não-lineares, chaveamento e superaquecimento. Essas descargas parciais aceleram a degradação do dielétrico do transformador e podem levar à destruição do equipamento, ocasionando elevado prejuízo financeiro. Dessa forma, são necessárias ações de prevenção de falhas causadas por descargas parciais em transformadores, através de métodos de monitoramento e localização. Este trabalho teve por objetivo apresentar um método de localização de descargas parciais em transformadores de potência imersos em óleo mineral isolante utilizando o método de emissão acústica e sistemas inteligentes do tipo redes neurais artificiais e algoritmos genéticos. Foram aplicadas métricas de processamento de sinais aos sinais acústicos gerados a partir de descargas parciais e obtidos através de sensores piezelétricos de baixo custo instalados no lado externo do tanque do transformador. Estas métricas foram utilizadas para treinamento das redes neurais a fim de obter a distância euclidiana entre os sensores e as descargas parciais. Essas distâncias euclidianas foram utilizadas em um sistema não-linear de localização o qual foi solucionado através de um algoritmo genético a fim de obter as coordenadas tridimensionais da descarga parcial. A análise dos resultados obtidos nas etapas de treinamento das redes neurais e execução do algoritmo genético foi realizada através de critérios estatísticos como erro absoluto médio, erro relativo percentual, desvio padrão e coeficiente de correlação. Esta técnica de localização mostrou resultados promissores uma vez que as coordenadas tridimensionais de duas descargas parciais, escolhidas para validação, apresentaram erros absolutos médios inferiores a 3 cm. / Transformers are important devices of the electric power system, which have high cost and their failures have a direct influence on the power quality delivered to the consumers. One of the main causes of failures in oil-immersed transformers, the partial discharges, comes from the physical and chemical degradation of the insulation system due to several factors such as overload, non-linear loads, switching and overheating. These partial discharges accelerate the degradation of the transformer dielectric and they can lead to the destruction of the equipment, causing high financial losses. Thus, actions are necessary to prevent faults caused by partial discharges in transformers, through monitoring and locating methods. The aim of this work was to present a method for locating partial discharges in oil-immersed power transformers using the acoustic emission method and intelligent systems such as artificial neural networks and genetic algorithms. Signal processing metrics were applied to the acoustic signals generated from partial discharges and obtained by low-cost piezoelectric sensors installed on the external side of the transformer tank. These metrics were used to train the neural networks in order to obtain the euclidean distance between the sensors and the partial discharges. These euclidean distances were used in a nonlinear location system, which was solved through a genetic algorithm in order to obtain the three-dimensional coordinates of the partial discharge. The analysis of the results obtained from the stages of neural networks training and genetic algorithm execution was performed through statistical criteria such as mean absolute error, percentage relative error, standard deviation and correlation coefficient. This localization technique showed promising results since the three-dimensional coordinates of two partial discharges, chosen for validation, presented mean absolute errors of less than 3 cm.

Esta tese propõe a utilização de sistemas inteligentes para, automaticamente, realizar diagnósticos e localizar falhas na instalação e na operação de redes de comunicação industrial que utilizam o protocolo Profibus DP. Para tais tarefas, uma série de análises é realizada a partir dos sinais transmitidos pela camada física, de telegramas transmitidos pela camada de enlace e de funções da camada de usuário do protocolo Profibus DP. Para a análise da camada física, amostras dos sinais elétricos transmitidos são processadas e apresentadas a algumas Redes Neurais Artificiais para que sejam classificadas de acordo com a sua forma de onda. Caso estes sinais apresentem alguma deformação, o sistema indica uma provável causa para o problema, afinal, os problemas das redes Profibus originam padrões específicos e característicos impressos nas formas de onda do sinal digital. Ainda através da análise das amostras dos sinais oriundos da camada física, algumas fontes de problemas são detectadas a partir da análise do nível médio de tensão do sinal que um determinado dispositivo está transmitindo. Tal análise é realizada a partir de um Sistema Especialista. Também utilizando Sistemas Especialistas, os telegramas transmitidos pela camada de enlace deste protocolo são analisados e a partir destes, falhas de configuração são detectadas. Por fim, é proposto um sistema nebuloso responsável por indicar ao usuário um valor próximo ao ideal para a variável de tempo denominada target rotation time. A proposta foi testada e validada a partir de dados obtidos de redes Profibus estabelecidas em laboratório e de alguns dados sintéticos originados por software. Os resultados obtidos foram suficientes para a comprovação da tese de que sistemas computacionais inteligentes podem contribuir de maneira efetiva no diagnóstico de problemas em redes Profibus DP e até mesmo em outros tipos de rede. / This thesis proposes the use of intelligent systems to automatically perform diagnostics and locate faults during the installation and operation of industrial communication networks that use the Profibus DP protocol. For such tasks, some analyzes are performed from the signals transmitted by the physical layer, from telegrams transmitted by the data link layer and from some user layer functions of the Profibus DP protocol. For physical layer analysis, the transmitted electrical signals samples are processed and submitted for some artificial neural networks that classifies each signal according to its waveshape. If these signals have some deformation, the system indicates a probable cause for the problem, after all, the Profibus problems originate specific and characteristic patterns printed on the digital signal waveform. Still analyzing the physical layer signal samples, some problems sources are detected from the signal voltage analysis. Such analysis is performed from an Expert System. Also using expert systems, the data link layer telegrams are analyzed and configuration faults are detected. Finally, it is proposed a fuzzy system responsible for specify a value close to ideal for the target rotation time variable. The proposal has been tested and validated with data from Profibus networks established in laboratory. Besides, some synthetic data were generated by software. The results were sufficient to prove the thesis that intelligent computational systems can contribute effectively to diagnose problems in Profibus DP networks and even in other types of networks.