Competitive learning for self organizing maps used in classification of partial discharges

Autores/as

  • Ruben Jaramillo-Vacio Universidad Autónoma de Aguascalientes. México
  • Carlos Alberto Ochoa Ortiz Zezzatti Laboratorio de Pruebas a Equipos y Materiales, Comisión Federal de Electricidad. Reforma 164, col. Juárez, México, DF, México 2 Centro de Innovación Aplicada en Tecnologías Competitivas
  • Julio César Ponce Gallegos Universidad Autónoma de Ciudad Juárez. Omega 201, fracc. Industrial Delta, CP 37545, León, Guanajuato, México.

DOI:

https://doi.org/10.30973/progmat/2013.5.2/2

Palabras clave:

aprendizaje competitivo, mapas autoorganizados, descargas parciales, métricas de calidad, diagnóstico

Resumen

In this paper different competitive learning algorithms for self-organizing maps (SOM) are experimentally examined. The characterization of the results obtained is presented in terms of quality of SOM. The competitive learning algorithms evaluated through SOM are winner-takes-all, frequency sensitive competitive learning, and rival penalized competitive learning. Case study: their performance in the classification of partial discharges on power cables.

Biografía del autor/a

Ruben Jaramillo-Vacio, Universidad Autónoma de Aguascalientes. México

Received its BSc in Electromechanical Engineering from ITESI (2002), Master in Electrical Engineering from San Luis Potosi University (2005), and Master in Management Engineering from La Salle University (2010). He has joint CIATEC (Conacyt research center) in 2009 to carry out his PhD research in industrial engineering and manufacturing. Since 2005 he is test engineer in CFE-LAPEM in dielectric test, partial discharge diagnosis at power cables. His research interest includes partial discharge diagnosis using intelligence artificial tools.

Carlos Alberto Ochoa Ortiz Zezzatti , Laboratorio de Pruebas a Equipos y Materiales, Comisión Federal de Electricidad. Reforma 164, col. Juárez, México, DF, México 2 Centro de Innovación Aplicada en Tecnologías Competitivas

(BSc 1994; Eng. Master, 2000; PhD, 2004; Postdoctoral researcher, 2006; Industrial postdoctoral research, 2009). He has written three books and eleven chapters in books related to AI. He has supervised ten PhD theses, 21 Master theses and 32 Bachelor theses. He participated in the organization of conferences such as HAIS’07, HAIS’08, ENC’06, ENC’07, ENC’08, MICAI’09, MICAI’10 and MICAI’11. His research interests include evolutionary computation, natural processing language and social data mining.

Julio César Ponce Gallegos , Universidad Autónoma de Ciudad Juárez. Omega 201, fracc. Industrial Delta, CP 37545, León, Guanajuato, México.

Received its BSc degree in computer system engineering from the Universidad Autónoma de Aguascalientes (UAA) in 2003, its MSc degree in computer sciences from the UAA in 2007, and its PhD in evolutionary computation from UAA. He is currently professor in UAA. His research interests include evolutionary computation and data mining

Citas

International Electrotechnical Comission. (2000). IEC 60270 Ed. 2. High-voltage test techniques - Partial discharge measurements. Geneva: IEC, 2000:15-16.

Pollard D. (1982). Quantization and the method of k-means. IEEE Trans. Information Theory 28 (2), 199-205. https://doi.org/10.1109/TIT.1982.1056481

Bezdek J. C. (1981). Pattern recognition with fuzzy objective function algorithms. New York: Plenum Press. https://doi.org/10.1137/1025116

Yang M. S. (1993). A survey of fuzzy clustering. Math Comput Model 18 (11), 1-16. https://doi.org/10.1016/0895-7177(93)90202-A

Höppner F., Klawonn F., Kruse R., Runkler T. Fuzzy. (1999). Cluster Analysis: Methods for Classification Data Analysis and Image Recognition. New York: Wiley.

Wang X., Lin H., Lu J., Yahagi T. (2002). Combining recurrent neural networks with self-organizing maps for channel equalization. IEEE Trans. on Communications. 2002, E85-B (10), 2227-2235.

Xu L, Krzyzak A, Oja E. (1993). Rival penalized competitive learning for clustering analysis, RBF net, and curve detection. IEEE Trans Neural Netw, 4 (4), 636-649. https://doi.org/10.1109/72.238318

Mazroua A. (1993). PD pattern recognition with neural netwoks using the multilayer perception technique. IEEE Transactions on Electrical Insulation, 28, 1082-1089. https://doi.org/10.1109/14.249382

Krivda A. (1995). Automated Recognition of Partial Discharge. IEEE Transactions on Dielectrics and Electrical Insulation, 28, 796-821. https://doi.org/10.1109/94.469976

Kim J, Choi W, Oh S, Park K, Grzybowski S. (2008) Partial Discharge Pattern Recognition Using Fuzzy-Neural Networks (FNNs) Algorithm. IEEE International Power Modulators and High Voltage Conference. Proceedings of 2008, 272-275. https://doi.org/10.1109/IPMC.2008.4743634

Descargas

Publicado

06-11-2013

Cómo citar

Jaramillo-Vacio, R. ., Ochoa Ortiz Zezzatti , C. A., & Ponce Gallegos , J. C. (2013). Competitive learning for self organizing maps used in classification of partial discharges. Programación matemática Y Software, 5(2), 6–12. https://doi.org/10.30973/progmat/2013.5.2/2

Número

Vol. 5 Núm. 2 (2013): Diciembre de 2013

Sección

Artículos

Licencia

Derechos de autor 2013 Ruben Jaramillo-Vacio, Carlos Alberto Ochoa Ortiz Zezzatti , Julio César Ponce Gallegos

Creative Commons License

Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.

Usted es libre de:

Compartir — compartir y redistribuir el material publicado en cualquier medio o formato.
Adaptar — combinar, transformar y construir sobre el material para cualquier propósito, incluso comercialmente.

Bajo las siguientes condiciones:

Atribución — Debe otorgar el crédito correspondiente, proporcionar un enlace a la licencia e indicar si se realizaron cambios. Puede hacerlo de cualquier manera razonable, pero de ninguna manera que sugiera que el licenciador lo respalda a usted o a su uso.
  Sin restricciones adicionales: no puede aplicar términos legales o medidas tecnológicas que restrinjan legalmente a otros a hacer cualquier cosa que permita la licencia.

Artículos más leídos del mismo autor/a