SciELO - Scientific Electronic Library Online

 
vol.13 issue1AsistO: A Qualitative MDP-based Recommender System for Power Plant OperationUsing Machine Learning for Extracting Information from Natural Disaster News Reports author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • Have no similar articlesSimilars in SciELO

Share


Computación y Sistemas

Print version ISSN 1405-5546

Comp. y Sist. vol.13 n.1 México Jul./Sep. 2009

 

Artículos

 

Probabilistic Intelligent Systems for Thermal Power Plants

 

Sistemas Inteligentes Probabilistas para Plantas Termoeléctricas

 

Pablo Héctor Ibargüengoytia, Alberto Reyes and Zenón Flores

 

Instituto de Investigaciones Eléctricas, Av. Reforma 113, Palmira, Cuernavaca, Mor., 62490, México; pibar@iie.org.mx , areyes@iie.org.mx , zfl@iie.org.mx

 

Article received on July 16, 2008
Accepted on April 03, 2009

 

Abstract

Artificial Intelligence applications in large–scale industry, such as thermal power plants, require the ability to manage uncertainty because current applications are large, complex and influenced by unexpected events and their evolution in time. This paper shows some of the efforts developed at the Instituto de Investigaciones Eléctricas (IIE) to assist operators of thermal power plants in the diagnosis and planning tasks using probabilistic intelligent systems. A diagnosis system, a planning system and a decision support system are presented. The diagnosis system is based on qualitative probabilistic networks, and the decision support system uses influence diagrams. The planning system is based on the Markov Decision Processes formalism. These approaches were validated in different power plant applications. Current results have shown that the use of probabilistic techniques can play an important role in the design of intelligent support systems for thermal power plants.

Keywords: power plants, diagnosis, probabilistic reasoning, Bayesian networks, influence diagrams, Markov decision processes.

 

Resumen

Las aplicaciones de Inteligencia Artificial (IA) en industrias de gran escala, como las centrales generadoras termoeléctricas, requieren de la habilidad de manejar incertidumbre ya que estas aplicaciones son complejas e influenciadas por eventos inesperados que evolucionan en el tiempo. Este artículo muestra algunos de los esfuerzos desarrollados en el Instituto de Investigaciones Eléctricas (IIE) para apoyar a los operadores de plantas termoeléctricas en sus tareas de planeación y diagnóstico, usando sistemas inteligentes probabilistas. Se presentan en este artículo un sistema de diagnóstico, un sistema de planificación y un sistema de soporte a las decisiones. El sistema de diagnóstico está basado en redes probabilistas cualitativas y el sistema de diagnóstico en diagramas de influencia. El sistema de planificación está basado en el formalismo de los procesos de decisión de Markov. Estos tres sistemas fueron validados en diferentes aplicaciones dentro de la operación de la planta termoeléctrica. Los resultados obtenidos muestran que las técnicas probabilistas pueden jugar un importante papel en el diseño de sistemas de ayuda en la operación de plantas termoeléctricas.

Palabras clave: Plantas termoeléctricas, diagnóstico, razonamiento probabilista, redes Bayesianas, diagramas de influencia, procesos de decisión de Markov.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

Acknowledgments

Thanks to the anonymous referees for their comments which improved this article. This research is supported by grants from IIE under infrastructure projects 12941 and 11984.

 

References

1. C. Boutilier, T. Dean, and S. Hanks. Decision–theoretic planning: Structural assumptions and computational leverage. Journal of Artificial Intelligence Research, 11:1–94, 1999.        [ Links ]

2. M. Caimi, C. Lanza, and B. Ruiz–Ruiz. An assistant for simulator–based training of plant operator. Marie Curie Fellowships Annals, Vol. 1, 1999.        [ Links ]

3. G.F. Cooper and E. Herskovitz. A Bayesian method for the induction of probabilistic networks from data. Machine Learning, 9(4):309–348, 1992.        [ Links ]

4. R. E. Bellman. Dynamic Programming. Princeton University Press, Princeton, N.J., U.S.A., 1957.        [ Links ]

5. The Elvira Consortium. Elvira: An environment for creating and using probabilistic graphical models. In Proceedings of the First European Workshop on Probabilistic graphical models (PGM'02)", pages 1–11, Cuenca, Spain, 2002.        [ Links ]

6. Z. Flores, P.H. Ibargüengoytia, and E. Morales. On–line diagnosis of gas turbines using probabilistic and qualitative reasoning. In th Intl. Conf.on Intelligent Systems Application to Power Systems, ISAP–2005, Washington, D.C., U.S.A., 2005.        [ Links ]

7. J. Hoey, R. St–Aubin, A. Hu, and C. Boutilier. Spudd: Stochastic planning using decision diagrams. In Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence, UAI–99, pages 279–288, 1999.        [ Links ]

8. C. Lacave, M. Luque, and F. J. Díez. Explanation of Bayesian networks and influence diagrams in Elvira. IEEE Transactions on Systems, Man and CyberneticsPart B: Cybernetics, 37:952–965, 2007. In press.        [ Links ]

9. B. Morales and P.H. Ibargüengoytia. On–line diagnosis using influence diagrams. In L.E. Sucar R. Monroy, G. Arroyo and H. Sossa, editors, Advances in Artificial Intelligence – MICAI 2004, LNAI 2313, pages 546–554, Berlin Heidelberg, 2004. Springer–Verlag.        [ Links ]

10. J. Pearl. Probabilistic reasoning in intelligent systems: networks of plausible inference. Morgan Kaufmann, San Francisco, CA., 1988.        [ Links ]

11. M.L. Puterman. Markov Decision Processes: Discrete stochastic dynamic programming. Wiley, New York, N.Y., U.S.A., 1995.        [ Links ]

12. A. Reyes, L.E. Sucar, and P.H. Ibargüengoytia. Abstraction and refinement for solving markov decision processes. In Proceedings of the European Workshop on Probabilistic Graphical Models (PGM'06)", pages 263–270, Chezch Republic, 2006.        [ Links ]

13. D. Suc and I. Bratko. Qualitative reverse engineering. In Proceedings of the 19th International Conference on Machine Learning, 2000.        [ Links ]

14. Z.A. Vale, C. Ramos, A. Silva, L. Faria, J. Santos, F. Fernandez, C. Rosado, and A. Marques. Socrates an integrated intelligent system for power system control center operator assistance and training. In IASTED International Conference on Artificial Intelligence and Soft Computing, pages 27–30, Cancun, México, 1998.        [ Links ]

15. I.H. Witten and E. Frank. Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann, San Francisco, CA, U.S.A., 2nd edition edition, 2005.        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License