Acoustic emission (AE) is one of many available technologies for condition health monitoring and diagnosis of rotating machines such as bearings. In recent years there have been many developments in the use of Acoustic Emission technology (AET) and its analysis for monitoring the condition of rotating machinery whilst in operation, particularly on high speed machinery. Unlike conventional technologies such as oil analysis, motor current signature analysis (MCSA) and vibration analysis, AET has been introduced due to its increased sensitivity in detecting the earliest stages of loss of mechanical integrity. This research presents an experimental investigation that is aimed at developing a mathematical model and experimentally validating the influence of operational variables such as film thickness, rotational speed, load, power loss, and shear stress for variations of load and speed conditions, on generation of acoustic emission in a hydrodynamic bearing. It is concluded that the power losses of the bearing are directly correlated with acoustic emission levels. With exponential law, an equation is proposed to predict power losses with reasonable accuracy from an AE signal. This experimental investigation conducted a comparative study between AE and Vibration to diagnose the rubbing at high rotational speeds in the hydrodynamic bearing. As it is the first known attempt in rotating machines. It has been concluded, that AE parameters such as amplitude, can perform as a reliable and sensitive tool for the early detection of rubbing between surfaces of a hydrodynamic bearing and high speed shaft. The application of vibration (PeakVue) analysis was introduced and compared with demodulation. The results observed from the demodulation and PeakVue techniques were similar in the rubbing simulation test. In fact, some defects on hydrodynamic bearings would not have been seen in a timely manner without the PeakVue analysis.In addition, the application of advanced signal processing and statistical methods was established to extract useful diagnostic features from the acquired AE signals in both time and frequency domain. It was also concluded that the use of different signal processing methods is often necessary to achieve meaningful diagnostic information from the signals. The outcome would largely contribute to the development of effective intelligent condition monitoring systems which can significantly reduce the cost of plant maintenance. To implement these main objectives, the Sutton test rig was modified to assess the capability of AET and vibration analysis as an effective tool for the detection of incipient defects within high speed machine components (e.g. shafts and hydrodynamic bearings). The first chapter of this thesis is an introduction to this research and briefly explains motivation and the theoretical background supporting this research. The second and third chapters, summarise the relevant literature to establish the current level of knowledge of hydrodynamic bearings and acoustic emission, respectively. Chapter 4 describes methodologies and the experimental arrangements utilized for this investigation. Chapter 5 discusses different NDT diagnosis. Chapter 6 reports on an experimental investigation applied to validate the relationship between AET on operational rotating machines, such as film thickness, speed, load, power loss, and shear stress. Chapter 7 details an investigation which compares the applicability of AE and vibration technologies in monitoring a rubbing simulation on a hydrodynamic bearing.

Made available in DSpace on 2016-01-13T13:26:49Z (GMT). No. of bitstreams: 0 Previous issue date: 2015-06-22. Added 1 bitstream(s) on 2016-01-13T13:32:07Z : No. of bitstreams: 1 000847056.pdf: 8844792 bytes, checksum: c194eab89a9feaffc22c568bf24f7741 (MD5) / A utilização de processos de usinagem para dar forma e acabamento é madeira tem sido ampliada muito nos últimos anos, levada pela necessidade de peças que contenham um melhor acabamento superficial atrelado à boa produtividade que é conseguida pelo aperfeiçoamento contínuo de equipamentos e de ferramentas para processar a madeira. Este trabalho utilizou um sistema experimental de torneamento com um novo cabeçote porta-ferramenta, contendo sensores de potência, vibração e emissão acústica acoplados em um torno EMCO Compact 5, muito comum para procedimentos didáticos, adaptados para o processamento de madeiras de pinus spp, a fim de demonstrar pela utilização da instrumentaçãodo processo, pode influenciar positivamente na análise de algum parâmetros como a rugosidade média (Ra) e rugosidade total (Rt), além dos valores de emissão acústica, potência de corte e vibração. A coleta dos dados se mostrou eficiente e precisa para os parâmetros principais de velocidade de avanço em 40 mm/min, 70 mm/min e 100 mm/min e variações de profundidade de corte em 0,5mm; 1,0 mm; e 1,5 mm; com velocidade de corte constante, e o uso de duas ferramentas de metal duro, uma constituída por metal duro polido e sem revestimento e outra ferramenta revestida com diaboreto de titânio. Entre os principais resultados obtidos, nota-se que a ferramenta com revestimento não influenciou na rugosidade média, e proporcionou menor rugosidade total e potência consumida, mas apresentou maiores valores para a vibração e emissão acústica captada durante os experimentos. Com relação ao tipo de madeira, os dados aquisitados estão coerentes, exceto pelos valores de potência consumida, que podem estar ligados a características próprias das peças de madeira usadas / The use of machining processes to shape and finish will wood has been greatly expanded in recent years, driven by the need for parts that contain a better surface finish tied will good productivity is achieved by continuous improvement of equipment and tools to process wood. This study used an experimental turning system with a new tool holder head, containing power sensors, vibration and acoustic emission engaged in a lathe EMCO Compact 5, very common for teaching procedures, adapted for the processing of wood pinus spp and eucalyptus spp, to demonstrate the use of the process instrumentation, can positively influence the analysis of some parameters such as the average roughness (Ra), total toughness (Rt), plus the acoustic emission values, cutting performance and vibration. Data collection was efficient and accurate for the main parameters of forward speeds at 40 mm/min, 70 mm/min and 100 mm/min and variations of the cutting depth of 0.5 mm; 1.0 mm; and 1.5 mm; with constant surface speed, and the use of two carbide tools, consisting of a polished carbide and uncoated and coated other tool with titanium diboride. The main results obtained, it is noted that the coating tool did not influence the average roughness, and provided lower total roughness and consumed power, but showed higher values for vibration and acoustic emission captured during the experiments. Regarding the type of wood, obtained data are consistent, except for the consumed power values, which may be linked to characteristics of used pieces of wood

No description available.

Les piles à combustible constituent une alternative aux moteurs thermiques utilisés dans le cadre d’applications transport ou dans le cadre d’applications stationnaires. Cependant, il existe encore aujourd’hui des verrous technologiques limitant leur développement à l’échelle industrielle. Un des verrous importants est la détermination de leur état de santé en temps réel permettant un diagnostic voire un pronostic de leurs dysfonctionnements ou de leur temps de bon fonctionnement. Dans ce travail, deux approches différentes ont été abordées en vue de proposer des outils de diagnostic pour des empilements de taille industrielle : une approche électrochimique et une approche acoustique. Dans un premier temps, un modèle simple de spectroscopie d’impédance électrochimique et un algorithme d’identification des paramètres ont été développés dans l’optique de proposer un outil de diagnostic rapide et facile à implémenter. Dans un second temps, deux méthodes basées sur la propagation des ondes sonores (l’émission acoustique et les acousto-ultrasons) ont été testées et évaluées dans des conditions de fonctionnement optimales et dégradées. / Fuel cells are a promising alternative to classical fossil energy based engines used in transport applications and electro-generators. However, there still exist limitations preventing it to be used at an industrial scale. In particular, the real-time determination of the cell’s state of health, allowing one to establish a diagnosis or a prognosis of the malfunctions and failures In this document, two different approaches are tested in order to propose real-diagnostic tools for industrial scale stacks. First, a simple electrochemical impedance spectroscopy model is developed in order to create a fast and easily embeddable diagnostic tool. Finally, two techniques based on ultrasound propagation (acoustic emission and acousto-ultrasound).are tested and evaluated in optimal and degraded conditions.

Orientador: Paulo Roberto de Aguiar / Banca: Fabricio Guimarães Baptista / Banca: Danilo Hernane Spatti / Resumo: A dressagem é um processo essencial para a usinagem, pois impacta diretamente na qualidade superficial de peças retificadas. A dressagem consiste em recompor a característica cortante dos grãos abrasivos da superfície do rebolo após sucessivas retificações, a partir da remoção de material pela ponta do dressador de diamante. O dressador pode ser monitorado em tempo real utilizando sensores, a fim de garantir a qualidade no processo da dressagem e fornecer informações da condição da ferramenta, sem precisar interromper o processo de usinagem. Assim, este trabalho propõe um método para predição da largura de atuação do dressador de diamante, utilizando sinais de emissão acústica e redes neurais artificiais. Sabe-se que a largura de atuação do dressador impacta diretamente na dressagem, pois determina a agressividade que é gerada no rebolo. Por isso, um sistema de monitoramento online que seja capaz de predizer um passo à frente a condição do dressador seria de grande valia para o processo, pois extinguiria a necessidade de paradas para monitoramento direto, reduziria a atuação dos operadores e evitaria a subutilização ou sobreutilização da ferramenta. Para este fim, os experimentos foram realizados em uma retificadora plana equipada com rebolos abrasivos de óxido de alumínio e dressadores piramidais de ponta única, do tipo natural. Os sinais de emissão acústica foram aquisitados a uma frequência de 2 MHz e processados utilizando filtros digitais e diferentes estatísticas, como ... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Dressing is an important process in machining, because it directly impacts on the quality surface of the workpieces. Dressing consists of restoring the cutting characteristics of the abrasive grains on wheel surface after several grinding passes through material removal by the tip of the diamond dresser. The dresser can be monitored in real time using sensors in order to guarantee the quality of the dressing process and provide information about tool condition, without the need of interrupting machining process. Then, this work proposes a methodology to predict the width of dresser at the dressing depth using acoustic emission signals and artificial neural networks. It is know that the width of dressing at the dressing depth directly impacts in dressing, because it defines the wheel sharpness. Therefore an online monitoring system able to predict one step forward the condition of the dresser would be of great value for the process because it would extinghish the need of stops for the direct monitoring, would reduce the function of operators and would avoid the sub-utilization and over over-utilization of the tool. For this purpose, experiments were performed using a horizontal surface machine equipped with aluminum oxide grinding wheels and natural single-point pyramidal dressers. Acoustic emission signals were acquired at 2 MHz and processed using digital filters and some statistics as RMS, ROP, power law, CFAR and MVD. Two kinds of neural networks were developed, onde for estimulation and other for prediction, wherein the output of first were used as input of second. For estimation MLP neural models were develope, being that the statistics of sensor signals were used as their inputs. For prediction, TDNN neural models were developed, inserting time delay. Results obtained indicate that acoustic emission is efficient to predict width of dresser using the best statistics and selected width bands. Utilization of neural networks performed... / Mestre

Les problématiques de sureté nucléaire passent par l’évaluation des installations existantes et notamment des enceintes de confinement. L’enceinte est un double sarcophage en béton armé. Elle constitue une barrière protégeant à la fois le réacteur d’une agression extérieure mais aussi protégeant l’extérieur d’une éventuelle fuite en cas d’accident. Le bon fonctionnement de l’enceinte est assuré lorsqu’elle remplit ses fonctions de résistante et d’étanchéité. L’évaluation de ces deux fonctions se fait notamment par le suivi de l’état de fissuration de l’enceinte.La fissuration du béton est associée au développement de micro-fissures qui apparaissent lorsque les contraintes locales du matériau sont dépassées. Il résulte de ces micro-fissurations la libération d'énergie qui se propage dans le matériau sous la forme d'ondes mécaniques. La détection de ses ondes permettra de remonter à leur source et de caractériser l’état d’endommagement du matériau.L’émission acoustique (EA) permet, à l’aide d’un réseau de capteurs piézoélectriques placés sur la surface, d’enregistrer les signaux issus de la libération d’énergie mécanique émise par les microfissures au sein du matériau. L’analyse des temps de vol de ces ondes permet entre autre de localiser les micro-fissurations au sein du matériau. Cette particularité de l’EA est donc parmi les techniques passives d’inspection pour le suivi de l’endommagement et de la fissuration du béton.Cependant, la précision des outils de localisation existants est fortement dépendante des méthodes d’analyse des signaux d’EA enregistrés. Cette dépendance constitue un verrou important qu’il est nécessaire de lever afin d’assurer le suivi spatial de la propagation de fissures le plus fiable possible. Par ailleurs l’identification des mécanismes sources d’EA est un enjeu crucial car il permet de caractériser les différents phénomènes qui interviennent lors de la propagation de fissure.Sur la base d’essais de fissuration et d’endommagement thermique du béton nous avons montré l’effet des stratégies de localisation sur la représentation de l’endommagement. A partir de ces résultats nous avons montré qu’une analyse spatiale basée sur la densité d’événements acoustique et l’énergie libérée permettait de suivre la propagation de l’endommagement thermique mais aussi de la fissuration dans les éprouvettes. Une analyse originale des signaux par inter-corrélation est proposée afin de regrouper les signaux en classes. Enfin les méthodes développées ont été appliquées pendant les essais d’ouverture et de refermeture d’une fissure sur la souflerie de l’Onera ainsi que sur la maquette « Vercors » d’EDF qui représente une enceinte de confinement de centrale nucléaire à l’échelle 1/3. / Nuclear safety issues involve the evaluation of existing installations, and in particular of containment systems. The enclosure is a double sarcophagus of reinforced concrete. It constitutes a protective barrier to both the reactor of an external aggression but also an agent of protection of the exterior of a possible leak in case of accident. The correct functioning of the enclosure is ensured in full filling of the functions of resistance and sealing. The evaluation of these two functions in particular for monitoring the state of cracking of the enclosure.The cracking of the concrete is associated with the development of micro-cracks which appear when the local stresses of the material are exceeded. As a result of these micro-cracks, the release of energy is propagated in the material in the form of mechanical waves. This phenomenon is called "acoustic emission".Acoustic emission (EA) is also the name of the technique used to directly measure the manifestation of these micro-cracks. The EA consists of recording, by means of a network of piezoelectric sensors placed on the surface, the ultrasonic waves emitted by the micro-cracks within the material. This particularity of the AE places it at the forefront of inspection techniques for monitoring the damage and cracking of concrete.The accuracy of the existing location tools is highly dependent on the methods of analyzing the recorded EA signals. This dependence constitutes an important lock that must be lifted in order to ensure the spatial follow-up of the most reliable crack propagation possible. Moreover, the identification of the source mechanisms of EA is a crucial stake because it makes it possible to characterize the different phenomena that intervene during the propagation of crack.On the basis of tests of cracking and thermal damage of the concrete we shown the impact of the strategies of localization on the representation of the damage. Based on our results, we proposed a spatial analysis based on the acoustic event density and the released energy allowing to follow the propagation of the thermal damage also of the cracking in the specimens. An original analysis of the signals by cross-correlation is proposed in order to group the signals into classes. Finally, the methods developed were applied during crack opening and referencing trials on the Onera blower and on a 1/3 scale model of a nuclear power plant confinement enclosure called “Vercors”.

O objetivo deste trabalho é o desenvolvimento de um sistema de monitoramento capaz de identificar em processo o mecanismo de desgaste ao qual o rebolo está sendo submetido durante a retificação. Assim, inicialmente foram pesquisados os meios físicos, os tipos de sensores e estratégias utilizadas nos sistemas de monitoramento, o estágio atual de desenvolvimento e suas aplicações. Para que este sistema seja confiável e eficiente foi necessário estudar detalhadamente a fenomenologia dos mecanismos de desgaste do rebolo. Este estudo permite definir parâmetros para identificação do tipo de desgaste que está ocorrendo, baseado em indicações fornecidas pelo monitoramento de variáveis cujo comportamento é conhecido, como potência consumida pelo motor do rebolo e vibração ou ainda por medições metrológicas. Dois sistemas ópticos de mapeamento da superfície do rebolo são apresentados e estes sistemas serviram como base para o desenvolvimento de um novo sistema de mapeamento da Distribuição da Energia Acústica de Retificação no rebolo, DEAR. Realizaram-se experimentos com o novo sistema, monitorando simultaneamente outros meios físicos. Procurou-se desta forma relacionar o padrão gráfico do mapa gerado pelo novo sistema com o tipo de desgaste que estava ocorrendo, baseado nas informações obtidas no estudo dos mecanismos de desgaste. A análise dos resultados mostraram que o mapa DEAR pode ser utilizado para identificar o tipo de mecanismo de desgaste predominante que está ocorrendo no rebolo, em processo. Esta informação é muito útil na escolha de ações corretivas para problemas da retificação. / The objective of this work is the development of a monitoring system able to identify in process the grinding wheel wear mechanism. Thus, initially the physical quantities, sensors types and monitoring systems strategies, developing status and their applications were researched. To get a reliable and efficient system was necessary a detailed study of the grinding wheel wear mechanism phenomenology. This study allows to define parameters in order to identify which type of wear is occurring, using indication obtained by the monitoring variables whose behavior is known, such as grinding power, vibration or metrological measurement. Two grinding wheel surface optical mapping systems are presented, those systems were utilized as base to developing a new mapping system of Grinding Acoustic Energy Distribution (DEAR). Experiments have been done with the new system, monitoring simultaneously other physical quantities, looking for a way to link the mapping graphic patterns generates by the new system with the wear type which was occurring, based on information obtained from the wear mechanism study. The results and analysis have shown that the DEAR map can be used to identify the predominant wear mechanism type which is occurring on a grinding wheel, in process. This information is very useful in the corrective actions when facing a grinding problem.

Historicamente, as atividades de manutenção têm sido caracterizadas como um problema dentro das organizações. Entretanto, nos últimos anos, esta atitude tem sido revista e tem levado a um reconhecimento da manutenção como uma função estratégica nas empresas. Esta mudança do papel da manutenção é fruto da globalização de mercados, que forçam as organizações a competir não somente em qualidade, diferenciação ou preço, mas também, em inovações tecnológicas, redução de tempo de atendimento e tecnologia de informação. A informação é um recurso essencial para direcionar e atingir os objetivos da gestão. No caso da manutenção, a coleta e análise de dados são recursos indispensáveis para o monitoramento e controle de suas atividades. O sistema proposto visa o atendimento dessas expectativas, através da aquisição de dados por sensores de emissão acústica em equipamentos rotativos. A pesquisa buscou a confirmação de que é possível a detecção de falhas em mancais de rolamentos, a partir do monitoramento dos níveis de emissão acústica, e assim conseguir uma forma pró-ativa de gerir a manutenção. A análise dos dados coletados fornece parâmetros para que se possam gerar informações que disparem ações, tanto do departamento de manutenção como dos demais setores envolvidos (direta ou indiretamente) com suas atividades, de maneira integrada. / Historically, the maintenance activities have been characterized as a problem for the organizations. However, in the last years, maintenance has been recognized as strategic for the enterprises. This change in the role played by maintenance is a result of the globalization of markets that force the organizations to compete not only in quality, prices, but also in technology innovation, time to market and information technology. Information is an essential resource for achieving the objectives of managing. Regarding to maintenance, the gathering and analysis of data are of fundamental importance for the monitoring and control of the related activities. The system proposed in this research has the objective of improving the managing of maintenance through the data acquisition of acoustic emission in rotary equipments. This research aims at the confirmation of the possibility of monitoring failure in ball bearings though the monitoring of acoustic emission and from this information to manage the maintenance processes. The analysis of the collected data offered parameter for the generation of a list of integrated actions for the maintenance department as well other involved sectors of the company.

Abstract: The process of inspecting welds done in production at Volvo Aero in Trollhättan is timeconsuming and a lot of this time goes into examining faulty objects. The aim of this thesis is to startdevelopment of a system that analyses acoustic emission from cooling welds to determine the qualityof the weld. Our aim is to be able to detect cracks in the material and to give information on thecracks using the data gathered by our sensors. To do this we will use methods to locate sound sourcesand then rate our findings and do some simplifications on the result of our calculations. We willanalyze our calculated data to find crack signatures and classify our findings and give alarms if wefind cracks that are considered too big for comfort. We will also give insight in to future aspects of ourwork and look at ways to improve on our proposed methods. We will discuss our systems pros, consand what things have been taken into consideration during design, and what strategies we proposeto handle the results from the system.

Le contrôle non-destructif (CND) par émission acoustique (EA) permet la détection de la création ou de l’évolution de défauts. Il repose sur l’utilisation d’un ou plusieurs capteurs pouvant être installés de façon permanente à l’écoute d’événements pouvant se dérouler au sein de structures de grandes dimensions. Dans le cadre du développement d’un simulateur de CND/EA, cette thèse s’intéresse à la modélisation de la mesure par un capteur d’EA d’un champ d’ondes élastiques quelconque. Une formulation exacte, puis une formulation approchée qui en dérive, sont obtenues à partir du principe de réciprocité électromécanique, de la réponse d’un capteur d’EA à une onde. On démontre que le calcul précis de ces réponses ne peut pas être effectué sous les hypothèses simplificatrices couramment faites conduisant naturellement à un modèle unidimensionnel. Au contraire, on démontre que pour les capteurs courants, du fait de leur conception visant à les rendre sensibles sur une large gamme de fréquences, les modes complexes de vibration de l’élément piézoélectrique et la propagation des ondes dans le boîtier du capteur influencent fondamentalement sa réponse. Un modèle éléments finis de capteur est proposé, validé expérimentalement et utilisé pour mener à bien des études paramétriques utiles à l’opérateur pour choisir un capteur, ou au fabriquant de capteur pour en optimiser la conception. Une représentation de la sensibilité des capteurs d’EA est proposée beaucoup plus complète que ce qui est généralement connu des capteurs. Une stratégie de simulation reposant sur la constitution d’une base de données de ces sensibilités est développée. / Nondestructive testing (NDT) by acoustic emission (AE) allows the detection of the formation or the evolution of defects. It relies on the use of one or several sensors that can be permanently attached to a large structure to listen to AE events that can happen. In the framework of the development of a tool for simulating NDT/AE, this thesis aims at modeling the measurement of arbitrary elastic wave field by AE sensor. An exact formulation, then an approximate one which derives from it, are obtained from the electromechanical reciprocity principle, of the response of an AE sensor to elastic waves. It is demonstrated that the accurate calculation of these responses cannot be made under the usual simplifying hypotheses which lead naturally to one-dimensional model. Rather the opposite, it is demonstrated that for standard sensors conceived to be highly sensitive over a large bandwidth, complex vibration modes together with the sensor case influence fundamentally their response. A finite element model is proposed, experimentally validated and used to conduct parametric studies helpful for an operator choosing a sensor, for a manufacturer to optimally conceive sensors. A representation of AE sensor sensitivity is proposed, far more complete than what is usually known about sensors. A simulation strategy is developed relying on the formation of a database of sensor sensitivities.