Aplicación de Redes Neuronales en la Detección de Regímenes Degradados en el Proceso Wedm

E. Portillo, I. Cabanes, M. Marcos, A. Zubizarreta

Resumen

Este artículo presenta los resultados de un estudio comparativo de distintas configuraciones de redes neuronales aplicadas al proceso de corte por electroerosión por hilo (WEDM). El objetivo perseguido es detectar con antelación comportamientos de corte degradado que alertan del riesgo creciente de la rotura de la herramienta empleada en este proceso de mecanizado: el hilo. Cuando esto sucede, disminuye la productividad de manera significativa. Así, partiendo de un trabajo previo en el que se identificaron diferentes tipos de comportamientos degradados, se ha realizado un estudio comparativo contemplando distintos criterios. Entre ellos, destaca la comparación de arquitecturas clásicas de red y, más concretamente, la arquitectura estática Perceptrón Multicapa, y la arquitectura recurrente Elman. La conclusión del trabajo ha sido que la arquitectura Elman constituye la alternativa más adecuada para la detección de la degradación del proceso.

Palabras clave

WEDM; electroerosión; RNA; redes neuronales artificiales; Perceptrón Multicapa; Elman

Texto completo:

PDF

Referencias

Behrens, A. y J. Ginzel (2003). Neuro-Fuzzy Process Control System for Sinking EDM. Journal of Manufacturing processes, Vol. 5, No. 1, pp. 33-39.

Cabanes, I., E. Portillo, M. Marcos, y J.A. Sánchez (2008). Online prevention of wire breakage in wire electro-discharge machining. Robotics and Computer-Integrated Manufacturing, Vol. 24, No. 2, pp. 287-298.

Cabanes, I., E. Portillo, M. Marcos, y J.A. Sánchez (2008). Industrial application for on-line detection of instability and wire breakage in wire EDM). Journal of Materials Processing Technology, Vol. 195, No. 1-3, pp. 101-109.

Elman, J. L. (1990). Finding Structure in Time. Cognitive Science, Vol. 14, pp. 179-211.

Fenggou, C. y Y. Dayong (2004). The study of high efficiency and intelligent optimization system in EDM sinking process. Journal of Materials Processing Technology, Vol. 149, pp. 83–87.

Ho K. H., S. T. Newman, S. Rahimifard y R. D. Allen (2004). State of the art in wire electrical discharge machining (WEDM). International Journal of Machine Tools and Manufacture, Vol. 44, pp. 1247-1259.

Huang, J.T. y Y.S. Liao (2000). A wire edm maintenance and fault-diagnosis expert system integrated with an artificial neural network. Int. J. Prod. Res., Vol. 38, No. 5, pp. 1071- 1082.

Isasi Viñuela, P. y I. M. Galván León (2004). Redes de Neuronas Artificiales. Un enfoque práctico, Ed. Pearson Educación, S.A., Madrid.

Kao, J. Y. y Y. S. Tarng (1997). A neural-network approach for the on-line monitoring of the electrical discharge machining process. Journal of Materials Processing Technology, Vol. 69, No. 1-3, pp. 112-119.

Kunieda, M.; S. Saga, H. Yoshino, T. Ohta, y M. Kobayashi (2001). Control of discharge locations in EDM with locally imposed high electric field. Proceedings of the XIII ISEM, Vol. 2, pp. 485-495.

Lauwers B., J. P. Kruth, P. H. Bleys, B. Van Coppenolle, L. Stevens, R. Derighetti (1999). Wire rupture prevention using on-line pulse localisation in WEDMB. Vdi Berichte, Vol. 1405, pp. 203-213.

Liao, Y.S. y J.C. Woo (1997a). The effects of machining settings on the behaviour of pulse trains in the WEDM process. Journal of Materials Processing Technology, Vol. 71, pp 433-439.

Liao, Y.S., Y.Y. Chu and M.T. Yan (1997b). Study of wire breaking process and monitoring of WEDM. International Journal of Machine Tools and Manufacture, Vol. 37 (4), pp. 555-567.

Liao Y. S, Woo, J. C. Design of a fuzzy control system for the adaptive control of WEDM process. International Journal of Machine Tools and Manufacture. 2000; Vol. 40 (15), p. 2293-2307.

Mediliyegedara, T.K.K.R., A.K.M. De Silva, D.K. Harrison, J.A. McGeough (2004). An intelligent pulse classification system for electro-chemical discharge, machining (ECDM)—a preliminary study- Journal of Materials Processing Technology, Vol. 149, pp. 499-503.

Obara, H.; M. Abe y T. Ohsumi (1999). Control of Wire Breakage during Wire EDM. International Journal of Electrical Machining, No. 4.

Portillo, E., I. Cabanes, M. Marcos, D. Orive, J. A. Sánchez (2007). Design of a virtual instrumentation system for a machining process. IEEE Transactions on Instrumentation and Measurement. Vol. 56, No. 6, 2007, pp. 2616-2622.

Sarle W.S. Neural FAQ (2002) ftp://ftp.sas.com/pub/neural/FAQ.html.

Sarle W.S. (1995) Stopped training and other remedies for overfitting. Proceedings of the 27th Symposium on the Interface of Computing Science and Statistics. 1995. - pp. 352-360.

Shoda K., Y. Kaneko, H. Nishimura, M. Kunieda, M. X. Fan (1995). Development of adaptive control system to prevent EDM wire breakage. EDM technology, Vol. 3, pp. 17–22.

Tarng, Y.S., S.C. Ma, L.K Chung (1995). Determination of optimal cutting parameters in WEDM. International Journal of Machine Tools and Manufacture, Vol. 35, No. 12, pp. 1693-1701.

Tetko I.V., Livingstone D.J. y Luik A.I. (1995) Neural Network Studies. 1. Comparison of overfitting and overtraining. Journal of Chemical Information and Computer Science. Vol. 35. – pp. 826-833.

Tiira, T. (1999). Detecting teleseismic events using artificial neural networks. Computers & Geosciences, Vol. 25, pp. 929-38.

Tsai, K. y P. Wang (2001). Predictions on surface finish in electrical discharge machining based upon neural network models. International Journal of Machine Tools & Manufacture, Vol. 41, pp. 1385–1403.

Valverde, R., Gachet, D. (2007) Identificación de sistemas dinámicos utilizando redes neuronales RBF. RIAI, Vol. 4, No. 2, pp. 32-42.

Wang, G (2004). Two-Phase Reverse Neural Network Approach for Modeling a Complicate Manufacturing Process with Small Sample Size. Neural Information Processing, Vol. 2, No. 1.

Wu, J y M.H. Li (2001). The identification of the servo control state in wire electrical discharge machining process. ISEM XIII, pp. 423-433.

Yan, M.T. y Y.S. Liao (1995). Adaptive control of WEDM process using the fuzzy control strategy. ISEM XI, pp. 343- 352.

Yan, Liao y Chang (2001), On-line Estimation of Workpiece Height by using Neural Networks and Hierarchical adaptive Control of WEDM. J. Advanced Manufacturing Technology, Vol.18, No. 2, pp. 884-89.

Abstract Views

404
Metrics Loading ...

Metrics powered by PLOS ALM




Creative Commons License

Esta revista se publica bajo una Licencia Creative Commons Attribution-NonCommercial-CompartirIgual 4.0 International (CC BY-NC-SA 4.0)

Universitat Politècnica de València     https://doi.org/10.4995/riai

e-ISSN: 1697-7920     ISSN: 1697-7912