Please use this identifier to cite or link to this item:
http://repositorio.ugto.mx/handle/20.500.12059/10213
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.rights.license | http://creativecommons.org/licenses/by-nc-nd/4.0 | es_MX |
dc.contributor | DAVID CAMARENA MARTINEZ | es_MX |
dc.creator | DANIEL ALEJANDRO ZAMBRANO ROMÁN | es_MX |
dc.date.accessioned | 2023-12-04T19:51:11Z | - |
dc.date.available | 2023-12-04T19:51:11Z | - |
dc.date.issued | 2023-10 | - |
dc.identifier.uri | http://repositorio.ugto.mx/handle/20.500.12059/10213 | - |
dc.language.iso | eng | en |
dc.publisher | Universidad de Guanajuato | es_MX |
dc.rights | info:eu-repo/semantics/openAccess | es_MX |
dc.subject.classification | CIS- Maestría en Ingeniería Eléctrica (Instrumentación y Sistemas Digitales) | es_MX |
dc.title | Short circuit detection in electrical transformers through statistical analysis of vibration signals in the time and frequency domains | en |
dc.type | info:eu-repo/semantics/masterThesis | es_MX |
dc.creator.id | info:eu-repo/dai/mx/cvu/1161100 | es_MX |
dc.subject.cti | info:eu-repo/classification/cti/7 | es_MX |
dc.subject.cti | info:eu-repo/classification/cti/33 | es_MX |
dc.subject.cti | info:eu-repo/classification/cti/3306 | es_MX |
dc.subject.keywords | Vibration signals | en |
dc.subject.keywords | Electric Transformers | en |
dc.subject.keywords | Machine learning | en |
dc.subject.keywords | Statistic analysis | en |
dc.subject.keywords | Classification | en |
dc.subject.keywords | Short circuit | en |
dc.subject.keywords | Señales de vibración | es_MX |
dc.subject.keywords | Transformadores eléctricos | es_MX |
dc.subject.keywords | Aprendizaje automático | es_MX |
dc.subject.keywords | Análisis estadístico | es_MX |
dc.subject.keywords | Clasificación | es_MX |
dc.subject.keywords | Cortocircuito | es_MX |
dc.contributor.id | info:eu-repo/dai/mx/cvu/329800 | es_MX |
dc.contributor.role | director | es_MX |
dc.type.version | info:eu-repo/semantics/publishedVersion | es_MX |
dc.contributor.two | ARTURO GARCIA PEREZ | es_MX |
dc.contributor.idtwo | info:eu-repo/dai/mx/cvu/64698 | es_MX |
dc.contributor.roletwo | director | es_MX |
dc.description.abstractEnglish | Electrical transformers are essential elements in industrial processes where the reliability and life span of these machines can be significantly improved by modeling a diagnosis strategy to monitor the working conditions in these devices. The short circuit turns (SCTs) in the windings significantly contribute to damage in electrical transformers; therefore, early detection during the initial stages is vital to prevent more extensive deterioration and schedule the appropriate maintenance. This work proposes a methodology based on the vibration signals analysis in the time and frequency domain to detect and classify a healthy state and several levels of SCTs. For both studies, the proposal consists of several steps. Starting with signal processing of the steady state, decimation, and computing of the Fast Fourier Transform (FFT) in the frequency domain, and the use of the raw steady state signal in the time domain. Followed by the extraction of statistical parameters. Subsequently, two techniques are used to select and reduce the features and keep those with the highest quality to distinguish the damage severity. Finally, three machine learning algorithms are trained and validated to classify the conditions under test. A comparative examination is made with the classification results achieved in each domain after applying the methodology to ensure a feasible and satisfactory assessment of the operating conditions of the transformer. Each step of the procedure was developed using mathematical software, and the results achieved show the effectiveness of this proposal in precisely identifying and classifying the severity of short circuit damage in the windings with the addition of low computational cost and fast processing time. | en |
Appears in Collections: | Maestría en Ingeniería Eléctrica (Instrumentación y Sistemas Digitales) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
DANIEL ALEJANDRO ZAMBRANO ROMÁN_Tesis24.pdf | 6.16 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.