• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 212
  • 60
  • 38
  • 22
  • 8
  • 6
  • 4
  • 4
  • 4
  • 3
  • 2
  • 2
  • 1
  • Tagged with
  • 450
  • 450
  • 97
  • 96
  • 87
  • 80
  • 60
  • 51
  • 49
  • 48
  • 46
  • 45
  • 44
  • 44
  • 43
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
441

Fault detection of planetary gearboxes in BLDC-motors using vibration and acoustic noise analysis

Ahnesjö, Henrik January 2020 (has links)
This thesis aims to use vibration and acoustic noise analysis to help a production line of a certain motor type to ensure good quality. Noise from the gearbox is sometimes present and the way it is detected is with a human listening to it. This type of error detection is subjective, and it is possible for human error to be present. Therefore, an automatic test that pass or fail the produced Brush Less Direct Current (BLDC)-motors is wanted. Two measurement setups were used. One was based on an accelerometer which was used for vibration measurements, and the other based on a microphone for acoustic sound measurements. The acquisition and analysis of the measurements were implemented using the data acquisition device, compactDAQ NI 9171, and the graphical programming software, NI LabVIEW. Two methods, i.e., power spectrum analysis and machine learning, were used for the analyzing of vibration and acoustic signals, and identifying faults in the gearbox. The first method based on the Fast Fourier transform (FFT) was used to the recorded sound from the BLDC-motor with the integrated planetary gearbox to identify the peaks of the sound signals. The source of the acoustic sound is from a faulty planet gear, in which a flank of a tooth had an indentation. Which could be measured and analyzed. It sounded like noise, which can be used as the indications of faults in gears. The second method was based on the BLDC-motors vibration characteristics and uses supervised machine learning to separate healthy motors from the faulty ones. Support Vector Machine (SVM) is the suggested machine learning algorithm and 23 different features are used. The best performing model was a Coarse Gaussian SVM, with an overall accuracy of 92.25 % on the validation data.
442

Entwicklung und Validierung einer Simulationsbasis zum Test von Reglern raumlufttechnischer Anlagen

Le, Huu-Thoi 11 February 2004 (has links)
Heutzutage gewinnt die Simulation von Gebäuden und Anlagen zunehmend an Bedeutung, um die Betriebsweise der Anlagen zu diagnostizieren bzw. zu bewerten und den Energiebedarf vorherzusagen. Dabei hängt die erzielte Genauigkeit von dem Kompliziertheitsgrad des angewendeten Simulationsprogramms ab. Deshalb ist Modellbildung und -validierung ein sehr wichtiger Bestandteil eines Softwareentwicklungsprozesses, um die Zuverlässigkeit zu sichern. Am Institut für Thermodynamik und Technische Gebäudeausrüstung liegen zahlreiche Simulationsmodelle vor. Im Rahmen dieser vorliegenden Arbeit wurden weitere benötigte Modelle (hygrisches Verhalten der Wände (vereinfachtes Verfahren), Rippenrohrwärmeüberträger, Wärmeregenerator et al.) entwickelt und in das Programm TRNSYS eingefügt sowie die vorhandenen Modelle an ihre Genauigkeit angepasst. Insbesondere sind dies die Modelle für Splitsysteme bei stetiger und nichtstetiger Regelung mit der detaillierten Betrachtung des Anlagenverhaltens sowohl beim Voll- als auch beim Teillastbetrieb. Damit ist es erstmals gelungen, das gesamte Anlagensystem der Splittechnik ausführlich zu beschreiben. Um die analytische Validierung durchführen zu können, wurden die analytischen Modelle für eine Splitanlage bei stetiger und nichtstetiger Regelung unter den vordefinierten Randbedingungen entwickelt. Zur analytischen Validierung finden auch die vorhandenen Simulationsmodelle Anwendung, so dass sich die meisten Komponenten und das Simulationsprogramm TRNSYS verifizieren ließen. Diese Validierung erfolgte im Rahmen des IEA-SHC/HVAC BESTEST TASK 22. Da an diesem TASK verschiedene Forschungsinstitutionen mit jeweils unterschiedlichen Simulationsprogrammen teilnahmen, ergab sich die beste Möglichkeit, vergleichende Tests durchzuführen. Wenn dabei ein Programm signifikante Unterschiede zu den anderen liefert, liegt dies nicht immer an Programmfehlern. Aber kollektive Erfahrungen aus diesem TASK zeigen, dass bei Abweichungen meistens Fehler bzw. fragwürdige Algorithmen gefunden wurden. Nachdem das Simulationsprogramm TRNSYS validiert war, erfolgte die Erstellung eines Konzeptes zur Fehlererkennung und Diagnose der Regelstrategien von RLTA. Das Verfahren erlaubt sowohl die Beseitigung der möglichen Fehler in der Planungsphase beim Entwurf der Regelstrategien als auch den Test der vorhandenen Regelstrategien. Dies erhöht die Zuverlässigkeit und damit die Sicherheit beim Anlagenbetrieb. Schließlich dient das Verfahren als Werkzeug zur Optimierung der Betriebsweise von RLTA. Das Regelverhalten wurde anhand typischer Fälle vorgestellt und diskutiert. Mit Hilfe des Verfahrens zur Fehlererkennung und Diagnose der Betriebsweise von RLTA ließen sich vorhandene Regelstrategien testen und verbessern.
443

Evaluation of machine learning methods for anomaly detection in combined heat and power plant

Carls, Fredrik January 2019 (has links)
In the hope to increase the detection rate of faults in combined heat and power plant boilers thus lowering unplanned maintenance three machine learning models are constructed and evaluated. The algorithms; k-Nearest Neighbor, One-Class Support Vector Machine, and Auto-encoder have a proven track record in research for anomaly detection, but are relatively unexplored for industrial applications such as this one due to the difficulty in collecting non-artificial labeled data in the field.The baseline versions of the k-Nearest Neighbor and Auto-encoder performed very similarly. Nevertheless, the Auto-encoder was slightly better and reached an area under the precision-recall curve (AUPRC) of 0.966 and 0.615 on the trainingand test period, respectively. However, no sufficiently good results were reached with the One-Class Support Vector Machine. The Auto-encoder was made more sophisticated to see how much performance could be increased. It was found that the AUPRC could be increased to 0.987 and 0.801 on the trainingand test period, respectively. Additionally, the model was able to detect and generate one alarm for each incident period that occurred under the test period.The conclusion is that ML can successfully be utilized to detect faults at an earlier stage and potentially circumvent otherwise costly unplanned maintenance. Nevertheless, there is still a lot of room for improvements in the model and the collection of the data. / I hopp om att öka identifieringsgraden av störningar i kraftvärmepannor och därigenom minska oplanerat underhåll konstrueras och evalueras tre maskininlärningsmodeller.Algoritmerna; k-Nearest Neighbor, One-Class Support Vector Machine, och Autoencoder har bevisad framgång inom forskning av anomalidetektion, men är relativt outforskade för industriella applikationer som denna på grund av svårigheten att samla in icke-artificiell uppmärkt data inom området.Grundversionerna av k-Nearest Neighbor och Auto-encoder presterade nästan likvärdigt. Dock var Auto-encoder-modellen lite bättre och nådde ett AUPRC-värde av 0.966 respektive 0.615 på träningsoch testperioden. Inget tillräckligt bra resultat nåddes med One-Class Support Vector Machine. Auto-encoder-modellen gjordes mer sofistikerad för att se hur mycket prestandan kunde ökas. Det visade sig att AUPRC-värdet kunde ökas till 0.987 respektive 0.801 under träningsoch testperioden. Dessutom lyckades modellen identifiera och generera ett larm vardera för alla incidenter under testperioden. Slutsatsen är att ML framgångsrikt kan användas för att identifiera störningar iett tidigare skede och därigenom potentiellt kringgå i annat fall dyra oplanerade underhåll. Emellertid finns det fortfarande mycket utrymme för förbättringar av modellen samt inom insamlingen av data.
444

Current-based Techniques for Condition Monitoring of Pumps

Becker, Vincent 12 December 2022 (has links)
[ES] Las bombas hidráulicas son el núcleo de muchos procesos en la industria y el sector servicios. Conviene tener en cuenta que los motores eléctricos son responsables del 69% del consumo de energía eléctrica en la industria, siendo en torno a un 22% de motores utilizados para el accionamiento de bombas. Los fallos de estas bombas pueden provocar averías en el proceso y, por lo tanto, implican altos costes económicos para el operador de la planta. Además, un funcionamiento defectuoso de las bombas conlleva una reducción de la eficiencia energética de la planta. De forma habitual, se utilizan principalmente dos tipos de estrategias orientadas al mantenimiento de maquinaria. Una estrategia de mantenimiento (mantenimiento preventivo) consiste en la sustitución de las piezas desgastadas en un intervalo de tiempo fijo. Este tipo de estrategia presenta muchas desventajas asociadas a la escasa optimización en el uso de los recursos y al consiguiente impacto económico. Por otro lado, la estrategia basada en la condición del equipo (mantenimiento basado en la condición) liga el reemplazo de las piezas desgastadas al estado del equipo, el cual es monitorizado a través de señales adquiridas mediante sensores. Sin embargo, el uso de sensores tiene algunos inconvenientes, como costes de inversión adicionales, posibles problemas en el montaje del sensor y posibles fallos del mismo. El análisis de la señal de corriente no se ha utilizado de forma habitual en la práctica para evaluar el estado de la bomba, aunque en muchas aplicaciones se dispone de sensores de corriente ya instalados que se podrían utilizar a tal fin. Se ha demostrado que técnicas basadas en el análisis de la corriente resultan de gran utilidad para diagnosticar varios tipos de fallos en motores eléctricos. De hecho, el análisis de la firma de corriente del motor se utiliza hoy en día ampliamente en la industria, especialmente para el diagnóstico de fallos en motores de inducción. En la presente tesis, se evalúa la utilización de la técnica de análisis de corrientes para el diagnóstico de fallos típicos relacionados con las bombas en diferentes aplicaciones. Se investigan tres tipos de bombas diferentes: bombas en línea de rotor húmedo, bombas de rotor seco y bombas sumergibles. En la tesis se han adaptado diversas técnicas, previamente empleadas para la detección de fallos en motores, al diagnóstico de fallos en la propia bomba. Los resultados indican que fallos como obstrucción de la bomba, fisura del impulsor y desgaste de los cojinetes influyen especialmente en dos frecuencias del espectro de corriente, las cuales pueden utilizarse como base de estrategias de mantenimiento basadas en la condición. En concreto, en las bombas de rotor húmedo, estos dos indicadores de fallo varían sensiblemente en función del punto de carga hidráulica de la bomba. Con la ayuda de un método de extracción de características basado en la motor reference frame theory, se demuestra que las mencionadas frecuencias pueden analizarse en tiempo real en un entorno industrial. Además, se presentan directrices para la monitorización en la nube y se valida con la ayuda de ensayos de laboratorio. Adicionalmente, se demuestra que los fallos son también detectables al analizar la corriente de arranque mediante herramientas de descomposición tiempo-frecuencia. Este hito no se había abordado anteriormente en la literatura técnica del área en lo referente a la detección de fallos en bombas. En conclusión, los resultados de este trabajo demuestran que los métodos de diagnóstico basados en la corriente pueden detectar con éxito diversos tipos de fallo en bombas, lo cual constituye un punto de gran interés para las industrias que utilicen estos activos en sus procesos. / [CA] Les bombes hidràuliques són el nucli de molts processos en la indústria i en el sector dels serveis. Cal mencionar que els motors elèctrics són responsables del 69% del consum de la energia elèctrica en la indústria, sent al voltant del 22% dels motors utilitzats per l'accionament de bombes. Les fallades d'aquestes bombes poden causar avaries en els processos, i per tant, representen un alt cost econòmic per a l'operador de la planta. A més a més, un funcionament defectuós en les bombes representa una reducció de l'eficiència energètica de la planta. De manera habitual, s'utilitzen principalment dos tipus d'estratègies orientades al manteniment de la maquinària. Una estratègia de manteniment (manteniment preventiu) consisteix en la canvi de les peces desgastades en un interval fixe de temps. Aquest tipus d'estratègia presenta molts desavantatges associats a la reduïda optimització en el ús dels recursos i el seu impacte econòmic. D'altra banda, la estratègia basada en la condició dels equipaments (manteniment basat en la condició) enllaça la substitució de les peces desgastades al estat de l'equip, el qual es monitoritzat per mig de senyals adquirides per sensors. No obstant això, el ús de sensors té alguns inconvenients com costos d'inversió addicionals, possibles problemes al muntatge i possibles fallades. L'anàlisi dels senyals de corrent no s'utilitzen de manera habitual en la pràctica per avaluar l'estat de la bomba, encara que en moltes aplicacions, estos sensors es troben instal·lats i es podrien fer servir per a aquesta finalitat. Ha estat demostrat que les tècniques basades en l'anàlisi de la corrent són de gran utilitat per el diagnosi de diversos tipus de fallades en motors elèctrics. De fet, l'anàlisi de la firma de la corrent del motor s'utilitza àmpliament en l'indústria, especialment per el diagnosi de fallades en motors d'inducció. En la present tesi, s'avalua l'utilització de la tècnica d'anàlisi de corrents per el diagnosi de fallades típiques relacionades en bombes per a diferents aplicacions. Se investiguen tres tipus de bombes diferents: bombes en línia de rotor humit, bombes de rotor sec i bombes submergibles. En aquesta tesi se han adaptat diverses tècniques, prèviament utilitzades en el diagnosi de màquines elèctriques, per al diagnosi de la pròpia bomba. Els resultat indiquen que les fallades com obstrucció de la bomba, la fissura de l'impulsor i el desgast dels coixinets influeixen especialment en dos freqüències de l'espectre de la corrent, les quals es poden utilitzar com a base per a una estratègia de manteniment basada en la condició. Particularment, en les bombes de rotor humit, aquestos dos indicadors de fallada varíen sensiblement en funció del punt de càrrega hidràulica de la bomba. En l'ajuda de un mètode d'extracció de característiques basat en la "motor reference frame theory", es demostra que les mencionades freqüències es poden analitzar en temps real en un entorn industrial. A més a més, es presenten directrius per la monitorització en el núvol i es valida en l'ajuda de assajos en el laboratori. Addicionalment, es demostra que les fallades són també detectables quan s'analitza la corrent d'arrancada mitjançant ferramentes de descomposició temps-freqüència. Aquest fet no ha estat analitzat prèviament en cap tipus de literatura tècnica dins del camp de detecció de fallades en bombes. En conclusió, els resultats d'aquest treball demostren que els mètodes de diagnosi basats en la corrent poden detectar en èxit diversos tipus de fallades en bombes, el qual constitueix un punt d'interés per a l'indústria que utilitzen aquest tipus de actiu en els seus processos. / [EN] Pumps are the heart of many processes in industry and service sector. Electric motors are responsible for 69% of electric energy consumption in industry, with 22% of them being used for the operation of pumps. Pump faults can lead to process breakdowns and are thus related to high costs for the plant operator. Furthermore, faulty operation of pumps reduces the energy efficiency of the plant. In many cases, a time-based maintenance strategy is applied, which means that typical wear parts are replaced within defined time cycles, which comes with some drawbacks such as poor resource efficiency and high costs. Condition-based maintenance strategies - meaning that the replacement of parts is planned based on the condition of the pump - are often based on the evaluation of sensor signals like vibration or noise. However, the use of sensors also has some drawbacks, such as additional investment costs, frequent problems with the sensor mounting, and possible sensor faults. There is no widespread use of the current signal to make statements about the pump condition, although current sensors are installed in many applications anyway. As for motor fault diagnosis, different current-based techniques have demonstrated their function. Today, motor current signature analysis is used in industry, especially for the diagnosis of induction motors. In this thesis, the current-based diagnosis of typical pump-related faults in different applications is evaluated. In total, three different pump types are investigated: a wet-rotor pump, a dry-runner inline pump, and a submersible pump. The techniques used for motor fault detection are adapted for the diagnosis of pump-related faults. The results indicate that the faults clogging, impeller crack, and bearing wear, in particular, influence two frequencies in the current spectrum, which can be used as a basis for a condition-based maintenance strategy. Especially in wet-rotor pumps, these two fault indicators strongly vary depending on the hydraulic load point of the pump. With the help of a feature extraction method based on the adapted reference frame theory, this work demonstrates that the two frequencies can be analyzed in real time in a field environment. Furthermore, a concept for cloud monitoring is presented and validated with the help of a laboratory test. Additionally, it is demonstrated that the faults are visible if the starting current is evaluated in a time-frequency map, which has not been considered before in the literature on pump-related faults. In summary, the findings of this work indicate that current-based diagnosis methods can successfully detect typical faults in pumps, a fact that is of high interest for companies using these assets in their industrial processes. / Becker, V. (2022). Current-based Techniques for Condition Monitoring of Pumps [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/190630
445

Towards the Implementation of Condition-based Maintenance in Continuous Drug Product Manufacturing Systems

Rexonni B Lagare (8707320) 12 December 2023 (has links)
<p dir="ltr">Condition-based maintenance is a proactive maintenance strategy that prevents failures or diminished functionality in process systems through proper monitoring and management of process conditions. Despite being considered a mature maintenance management strategy in various industries, condition-based maintenance remains underutilized in pharmaceutical manufacturing. This situation needs to change, especially as the pharmaceutical industry continues to shift from batch to continuous manufacturing, where the implementation of CBM as a maintenance strategy assumes a greater importance.</p><p dir="ltr">This dissertation focused on addressing the challenges of implementing CBM in a continuous drug product manufacturing system. These challenges stem from the unique aspects of pharmaceutical drug product manufacturing, which includes the peculiar behavior of particulate materials and the evolutionary nature of pharmaceutical process development. The proposed solutions to address these challenges revolve around an innovative framework for the practical development of condition monitoring systems. Overall, this framework enables the incorporation of limited process knowledge in creating condition monitoring systems, which has the desired effect of empowering data-driven machine learning models.</p><p dir="ltr">A key feature of this framework is a formalized method to represent the process condition, which is usually vaguely defined in literature. This representation allows the proper mapping of preexisting condition monitoring systems, and the segmentation of the entire process condition model into smaller modules that have more manageable condition monitoring problems. Because this representation methodology is based on probabilistic graphical modelling, the smaller modules can then be holistically integrated via their probabilistic relationships, allowing the robust operation of the resulting condition monitoring system and the process it monitors.</p><p dir="ltr">Breaking down the process condition model into smaller segments is crucial for introducing novel fault detection capabilities, which enhances model prediction transparency and ensures prediction acceptance by a human operator. In this work, a methodology based on prediction probabilities was introduced for developing condition monitoring systems with novel fault detection capabilities. This approach relies on high-performing machine learning models capable of consistently classifying all the initially known conditions in the fault library with a high degree of certainty. Simplifying the condition monitoring problem through modularization facilitates this, as machine learning models tend to perform better on simpler systems. Performance indices were proposed to evaluate the novel fault detection capabilities of machine learning models, and a formal approach to managing novel faults was introduced.</p><p dir="ltr">Another benefit of modularization is the identification of condition monitoring blind spots. Applying it to the RC led to sensor development projects such as the virtual sensor for measuring granule flowability. This sensor concept was demonstrated successfully by using a data-driven model to predict granule flowability based on size and shape distribution measurements. With proper model selection and feature extraction guided by domain expertise, the resulting sensor achieved the best prediction performance reported in literature for granule flowability.</p><p dir="ltr">As a demonstration exercise in examining newly discovered faults, this work investigated a roll compaction phenomenon that is usually concealed from observation due to equipment design. This phenomenon results in the ribbon splitting along its thickness as it comes out of the rolls. In this work, important aspects of ribbon splitting were elucidated, particularly its predictability based on RC parameters and the composition of the powder blend used to form the ribbon. These findings have positive ramifications for the condition monitoring of the RC, as correspondence with industrial practitioners suggests that a split ribbon is desirable in some cases, despite being generally regarded as undesirable in the limited literature available on the subject.</p><p dir="ltr">Finally, this framework was primarily developed for the pharmaceutical dry granulation line, which consists of particle-based systems with a moderate level of complexity. However, it was also demonstrated to be feasible for the Tennessee Eastman Process (TEP), a more complex liquid-gas process system with a greater number of process faults, variables, and unit operations. Applying the framework resulted in machine learning models that yielded one of the best fault detection performances reported in literature for the TEP, while also introducing additional capabilities not yet normally reported in literature, such as fault diagnosis and novel fault detection.</p>
446

Software Fault Detection in Telecom Networks using Bi-level Federated Graph Neural Networks / Upptäckt av SW-fel i telekommunikationsnätverk med hjälp av federerade grafiska neurala nätverk på två nivåer

Bourgerie, Rémi January 2023 (has links)
The increasing complexity of telecom networks, induced by the recent development of 5G, is a challenge for detecting faults in the telecom network. In addition to the structural complexity of telecommunication systems, data accessibility has become an issue both in terms of privacy and access cost. We propose a method relying on bi-level Federated Graph Neural Networks to identify anomalies in the telecom network while ensuring reduced communication costs as well as data privacy. Our method considers telecom data as a bi-level graph, where the highest level graph represents the interaction between sites, and each site is further expanded to its software (SW) performance behaviour graph. We developed and compared 4G/5G SW Fault Detection models under 3 settings: (1) Centralized Temporal Graph Neural Networks model: we propose a model to detect anomalies in 4G/5G telecom data. (2) Federated Temporal Graph Neural Networks model: we propose Federated Learning (FL) as a mechanism for privacy-aware training of models for fault detection. (3) Personalized Federated Temporal Graph Neural Networks model: we propose a novel aggregation technique, referred to as FedGraph, leveraging both a graph and the similarities between sites for aggregating the models and proposing models more personalized to each site’s behaviour. We compare the benefits of Federated Learning (FL) models (2) and (3) with centralized training (1) in terms of SW performance data modelling, anomaly detection, and communication cost. The evaluation includes both a scenario with normal functioning sites and a scenario where only a subset of sites exhibit faulty behaviour. The combination of SW execution graphs with GNNs has shown improved modelling performance and minor gains in centralized settings (1). In a normal network context, FL models (2) and (3) perform comparably to centralized training (CL), with slight improvements observed when using the personalized strategy (3). However, in abnormal network scenarios, Federated Learning falls short of achieving comparable detection performance to centralized training. This is due to the unintended learning of abnormal site behaviour, particularly when employing the personalized model (3). These findings highlight the importance of carefully assessing and selecting suitable FL strategies for anomaly detection and model training on telecom network data. / Den ökande komplexiteten i telenäten, som är en följd av den senaste utvecklingen av 5G, är en utmaning när det gäller att upptäcka fel i telenäten. Förutom den strukturella komplexiteten i telekommunikationssystem har datatillgänglighet blivit ett problem både när det gäller integritet och åtkomstkostnader. Vi föreslår en metod som bygger på Federated Graph Neural Networks på två nivåer för att identifiera avvikelser i telenätet och samtidigt säkerställa minskade kommunikationskostnader samt dataintegritet. Vår metod betraktar telekomdata som en graf på två nivåer, där grafen på den högsta nivån representerar interaktionen mellan webbplatser, och varje webbplats utvidgas ytterligare till sin graf för programvarans (SW) prestandabeteende. Vi utvecklade och jämförde 4G/5G SW-feldetekteringsmodeller under 3 inställningar: (1) Central Temporal Graph Neural Networks-modell: vi föreslår en modell för att upptäcka avvikelser i 4G/5G-telekomdata. (2) Federated Temporal Graph Neural Networks-modell: vi föreslår Federated Learning (FL) som en mekanism för integritetsmedveten utbildning av modeller för feldetektering. I motsats till centraliserad inlärning aggregeras lokalt tränade modeller på serversidan och skickas tillbaka till klienterna utan att data läcker ut mellan klienterna och servern, vilket säkerställer integritetsskyddande samarbetsutbildning. (3) Personaliserad Federated Temporal Graph Neural Networks-modell: vi föreslår en ny aggregeringsteknik, kallad FedGraph, som utnyttjar både en graf och likheterna mellan webbplatser för att aggregera modellerna. Vi jämför fördelarna med modellerna Federated Learning (FL) (2) och (3) med centraliserad utbildning (1) när det gäller datamodellering av SW-prestanda, anomalidetektering och kommunikationskostnader. Utvärderingen omfattar både ett scenario med normalt fungerande anläggningar och ett scenario där endast en delmängd av anläggningarna uppvisar felaktigt beteende. Kombinationen av SW-exekveringsgrafer med GNN har visat förbättrad modelleringsprestanda och mindre vinster i centraliserade inställningar (1). I en normal nätverkskontext presterar FL-modellerna (2) och (3) jämförbart med centraliserad träning (CL), med små förbättringar observerade när den personliga strategin används (3). I onormala nätverksscenarier kan Federated Learning dock inte uppnå jämförbar detekteringsprestanda med centraliserad träning. Detta beror på oavsiktlig inlärning av onormalt beteende på webbplatsen, särskilt när man använder den personliga modellen (3). Dessa resultat belyser vikten av att noggrant bedöma och välja lämpliga FL-strategier för anomalidetektering och modellträning på telekomnätdata.
447

PCBA verification and fault detection using a low-frequency GMR-based near-field probe with magnetic closed-loop feedback compensation : A non-contact alternative to physical probing / Verifiering och feldetektering av kretskort mha en lågfrekvent närfältssond baserad på en GMR-sensor med magnetisk återkopplingskrets med sluten kompensationsslinga : Ett kontaktlöst alternativ till fysisk sondering

Sundh, Joacim January 2022 (has links)
As electronics are getting both smaller and more advanced, the need to verify and validate remains and the means are getting more complex the more functions and components are added. Traditionally, in-circuit tests (ICTs) are performed by probing dedicated test points on the Printed Circuit Board Assembly (PCBA) in a test sequence that is unique to each product. But as the density of components increases, the choice between component and test point must be considered. Instead of decreasing the reliability during verification by having to remove less system-critical test points, this thesis suggests the use of a near-field probe (NFP) based around a Giant Magneto-Resistance (GMR) sensor to possibly replace the need for a physical test point by instead performing contactless testing. The use of a GMR sensor allows for bandwidth from 0 Hz up to the MHz range, whereas commercial NFPs are based on a different technique and are operational from the MHz range and up. The goal of this project was to improve the non-linearity of typically 15% present in the AAH002-02 model from NVE by the use of an analogue closed-loop magnetic feedback circuit. The project successfully improved the linearity to 99.8% by the use of an instrumentation amplifier, a subtractor and a push-pull amplifier in conjunction with a 3x30 turn planar coil embedded in a PCB, located beneath the sensor Integrated Circuit (IC). The resulting linearity was verified by a Helmholtz coil where a uniform magnetic field was produced with linearly increased field strength, and calculated using the R2 value from a linear regression analysis on the acquired data. In the future, the data acquired from this kind of NFP could be used together with a Machine Learning (ML) model to remove the manual labour required when constructing these product-unique test sequences. / Dagens elektronik blir både mindre och mer avancerad, men behovet av verifiering och validering av dessa kvarstår och metoderna för detta ökar i komplexitet ju fler funktioner och komponenter som läggs till. Dagens kretskortstester genomförs genom att sondera dedikerade testpunkter strategiskt utplacerade på kretskortet enligt en testsekvens som är unikt skapad för varje produkt. Men med att densiteten av komponenter ökar måste valet mellan komponent och testpunkt tas i beaktning. Instället för att minska tillförlitligheten vid validering genom att ta bort mindre kritiska testpunkter föreslår denna avhandling användandet av en närfältssond baserad runt en Giant Magneto-Resistance (GMR)-sensor för att möjligen ersätta behovet av en fysisk testpunkt genom att istället genomföra kontaktlös testning. Användandet av en GMR-sensor tillåter en bandbredd från 0 Hz upp till MHzområdet, där kommersiella närfältssonder är baserade på annan teknik och är funktionsdugliga från MHz-området och uppåt. Målet med detta projekt var att förbättra olinjäriteten på typiskt 15% som är närvarande hos en sensor av modell AAH002-02 från NVE genom en analog magnetisk återkopplingskrets med sluten slinga. Projektet lyckades förbättra linjäriteten till 99.8% genom användandet av en intrumentförstärkare, en subtraherare och en push-pull-förstärkare i samverkan med en plan spole på 3x30 varv inbyggd i ett mönsterkort placerd under sensorns integrerade krets. Den resulterande linjäriteten validerades med hjälp av en Helmholtz-spole där ett uniformt magnetfält producerades med linjärt ökande fältstyrka och beräknades genom R2 -värdet från en linjär regression-analys på den inhämtade datan. I framtiden kan datan som inhämtats från den här sortens närfältssond kunna användas tillsammans med en maskininlärningsmodell för att ersätta det manuella arbetet som idag krävs för att konstruera dessa produktunika testsekvenser
448

Real-time detection of stator resistance unbalances in three phase drives / Realtids detektering av obalanser i statorsmotstånd i trefasiga enheter

Singh, Bhanu Pratap January 2020 (has links)
An estimated 30% of the faults in Induction Machine (IM) are related to its stator. These faults are mostly in the form of an Inter-Turn Short Circuit (ITSC) fault i.e., when two winding inside the stator of IM are shorted due to insulation failure. However, ITSC fault can be avoided by detecting them in advance and then scheduling the maintenance of the IM. This thesis studies two methods for detecting this incipient ITSC fault in a three-phase IM and then estimating the stator resistance unbalance due to the ITSC fault. The first method is based on the asymmetry caused in the IM by the ITSC fault. As a result of this asymmetry, the negative sequence components of the stator voltages and the stator currents are generated inside the IM. A healthy IM also have these negative sequence components due to the manufacturing process and the supply voltage unbalances. The characteristics and the compensation methods of these negative sequence components in a healthy IM are discussed. The results show that after compensating the negative sequence components in a healthy machine, they can be used for detecting an ITSC fault and then to calculate the fault quantities as well as the stator resistance unbalances. The second method for detecting an ITSC fault is based on analysing the stator resistance unbalances. A three-phase drive is used to inject DC voltage in the stationary reference frame. The DC current generated by this DC voltage is measured and then by applying Ohm’s law stator phase resistances are calculated. In a healthy IM, the phase resistances are balanced. However, in case of ITSC fault in any of the phases, the phase resistance of that phase deviates from those of the other two phases which can be utilized for detecting ITSC fault. / Uppskattningsvis 30% av alla fel i induktionsmaskiner (IM) är kopplad till dess stator. Dessa fel är i huvudsak Inter-Turn Short Circuit (ITSC)-fel, dvs. två lindningar inom IM:ens stator blir kortsluta pga. ett isoleringsfel. Emellertid kan man undvika ITSC-fel genom att detektera dem i förhand och planera underhåll. Det här examensarbetet undersöker två metoder för att detektera ett förestående ITSC-fel i en tre-fas IM. Den första metoden är baserad på asymmetrin i IM:er pga. ITSC-felet. Resultatet av den här asymmetrin är att en negativ sekvens genereras i IM:ens statorspänning och statorström. En oskadad IM kan också visa dessa negativa sekvenser pga. tillverksprocessen och statorspänningsobalanser. Egenskaperna och kompensationsmetoderna för dessa negativa sekvenser i en oskadad IM kommer att diskuteras. Resultaten visar att efter kompenseringen av de negativa sekvenserna i en oskadad IM, kan de användas för att detektera ITSC-fel och efteråt för att beräkna felstorheter och även statormotståndobalanser. Den andra metoden för att detektera ITSC-fel är baserad på en undersökning av statormotståndobalanser. Ett tre-fas-drivsystem används för att injektera likspänning i den stationära referensramen. Likströmmen som följer av denna likspänning mäts och statorfasmotstånden beräkna efteråt med Ohms lag. I en oskadad IM är fasmotstånden balanserade. Däremot, när ett ITSC-fel uppstår i en fas, avviker fasmotståndet i den felaktiga fasen från de andra två fasernas, vilket kan användas för att detektera ITSC-fel.
449

Fault Location Algorithms in Transmission Grids

Harrysson, Mattias January 2014 (has links)
The rapid growth of the electric power system has in recent decades resulted in an increase of the number of transmission lines and total power outage in Norway. The challenge of a fast growing electrical grid has also resulted in huge increases of overhead lines and their total length. These lines are experiencing faults due to various reasons that cause major disruptions and operating costs of the transmission system operator (TSO). Thus, it’s important that the location of faults is either known or can be estimated with reasonably high accuracy. This allows the grid owner to save money and time for inspection and repair, as well as to provide a better service due to the possibility of faster restoration of power supply and avoiding blackouts.  Fault detection and classification on transmission lines are important tasks in order to protect the electrical power system. In recent years, the power system has become more complicated under competitive and deregulated environments and a fast fault location technique is needed to maintain security and supply in the grid. This thesis compares and evaluates different methods for classification of fault type and calculation of conventional one-side and two-side based fault location algorithms for distance to fault estimation.  Different algorithm has been implemented, tested and verified to create a greater understanding of determinants facts that affect distance to faults algorithm’s accuracy.  Implemented algorithm has been tested on the data generated from a number of simulations in Simulink for a verification process in implemented algorithms accuracy. Two types of fault cases have also been simulated and compared for known distance to fault estimation.
450

Užití programovatelných hradlových polí v systémech průmyslové automatizace / Field Programmable Gate Arrays Usage in Industrial Automation Systems

Nouman, Ziad January 2016 (has links)
Tato disertační práce se zabývá využitím programovatelných hradlových polí (FPGA) v diagnostice měničů, využívajících spínaných IGBT tranzistorů. Je zaměřena na budiče těchto výkonových tranzistorů a jejich struktury. Přechodné jevy veličin, jako jsou IG, VGE, VCE během procesu přepínání (zapnutí, vypnutí), mohou poukazovat na degradaci IGBT. Pro měření a monitorování těchto veličin byla navržena nová architektura budiče IGBT. Rychlé měření a monitorování během přepínacího děje vyžaduje vysokou vzorkovací frekvenci. Proto jsou navrhovány paralelní vysokorychlostní AD převodníky (> 50 MSPS). Práce je zaměřena převážně na návrh zařízení s FPGA včetně hardware a software. Byla navržena nová deska plošných spojů s FPGA, která plní požadované funkce, jako je řízení IGBT pomocí vícenásobných paralelních koncových stupňů, monitorování a diagnostiku, a propojení s řídicí jednotkou měniče.

Page generated in 0.0939 seconds