SciELO - Scientific Electronic Library Online

 
vol.14 issue1Tetrahedral Grid Generators and the Eigenvalue Calculation with Edge ElementsAn Efficient Δ-Causal Distributed Algorithm for Synchronous Cooperative Systems in Unreliable Networks 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.14 n.1 México Jul./Sep. 2010

 

Artículos

 

Multiple Fault Diagnosis in Electrical Power Systems with Dynamic Load Changes Using Probabilistic Neural Networks

 

Diagnóstico de Fallas Múltiples en Sistemas Eléctricos de Potencia con Cambios de Carga Dinámicos Utilizando Redes Neuronales Probabilísticas

 

Juan Pablo Nieto González1, Luis Garza Castañón2 and Rubén Morales Menéndez3

 

1 I.T.E.S.M. Campus Saltillo, Departamento de Mecatrónica Saltillo, Coahuila, México juan.pablo.nieto@itesm.mx

2 I.T.E.S.M. Campus Monterrey, Departamento de Mecatrónica y Automatización, legarza@itesm.mx

3 I.T.E.S.M. Campus Monterrey, Centro de Automatización Industrial Monterrey, Nuevo León, México, rmm@itesm.mx

 

Article received on December 18, 2007
Accepted en February 27, 2009

 

Abstract

Power systems monitoring is particularly challenging due to the presence of dynamic load changes in normal operation mode of network nodes, as well as the presence of both continuous and discrete variables, noisy information and lack or excess of data. This paper proposes a fault diagnosis framework that is able to locate the set of nodes involved in multiple fault events. It detects the faulty nodes, the type of fault in those nodes and the time when it is present. The framework is composed of two phases: In the first phase a probabilistic neural network is trained with the eigenvalues of voltage data collected during normal operation, symmetrical and asymmetrical fault disturbances. The second phase is a sample magnitude comparison used to detect and locate the presence of a fault. A set of simulations are carried out over an electrical power system to show the performance of the proposed framework and a comparison is made against a diagnostic system based on probabilistic logic.

Keywords: Fault Diagnosis, Multiple Faults, Probabilistic Neural Networks, Correlation Matrix, Eigenvalues, Power System, Dynamic Load Changes.

 

Resumen

El monitoreo de sistemas de potencia es particularmente retador debido a la presencia de cambios dinámicos de carga de los nodos de la red en modo de operación normal, así como la presencia de variables continuas y discretas, información con ruido y falta o exceso de datos. Este artículo propone un método de diagnóstico de fallas que es capaz de localizar el conjunto de nodos involucrado en eventos de fallas múltiples. El método detecta los nodos con falla, el tipo de falla y el tiempo en el cual está presente la falla. El método está compuesto de dos fases: En la primera fase una red neuronal probabilística es entrenada con los eigenvalores de los datos de voltaje obtenidos en operación normal así como con fallas simétricas y asimétricas. La segunda fase emplea una comparación entre las muestras para detectar y localizar la presencia de una falla. Se lleva a cabo un conjunto de simulaciones en un sistema eléctrico de potencia para mostrar el desempeño del método propuesto y se realiza una comparación contra un sistema de diagnóstico basado en lógica probabilística.

Palabras clave: Diagnóstico de Fallas, Fallas Múltiples, Redes Neuronales Probabilísticas, Matriz de Correlación, Eigenvalores, Sistemas de Potencia, Cambios Dinámicos de Carga.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

References

1. Barigozzi A., Magni L., & Scattolini R. (2004) A Probabilistic Approach to Fault Diagnosis of Industrial Systems. IEEE Transactions on Control System Technology, 12(6), 201–212.         [ Links ]

2. Bouthiba T. (2005). Fault detection and classification technique in EHV transmission lines based on artificial neural networks. European transactions on electrical power, 15(5), 443–454.         [ Links ]

3. Du D., Lou X. & Wu C. (2005). Dynamic Model of FCCU and its Application in a Hybrid Fault Diagnosis System. International Conference on Control and Automation, Budapest, Hungary, 2, 1014–1017.         [ Links ]

4. Duda R., Hart P., & Stork D. (2001). Pattern Classification. New York: Willey Interscience.         [ Links ]

5. Garza L. (2001). Hybrid Systems Fault Diagnosis with a Probabilistic Logic Reasoning Framework. PhD thesis, Intituto Tecnológico y de Estudios Superiores de Monterrey, Campus Monterrey.         [ Links ]

6. Gentil S., Montmain J. & Combastel C. (2004). Combining FDI and AI Approaches Within Causal Model Based Diagnosis. IEEE Transactions on Systems, Man and Cybernetics. Part B: Cybernetics, 34(5), 2207–2221.         [ Links ]

7. Hartstein R., Rezende K., & Suman A. (2007). Fault Detection in Primary Distribution Systems using Wavelets. International Conference on Power Systems Transients (IPST'07), Lyon, France, 572–580.         [ Links ]

8. He Q. Peter, Wang J., & Joe Qin S. (2004). A New Fault Diagnosis Method Using Fault Directions in Fisher Discriminant Analysis. TWMCC Texas–Winsconsin Modeling and Control Consortium. Technical report number TWMCC–2004–05. Austin, USA: The University of Texas at Austin. Department of Chemical Engineering.         [ Links ]

9. Johnson R. & Wichern D. (2002). Applied Multivariate Statistical Analysis. New Jersery: Prentice Hall.         [ Links ]

10. Liang J., & Wang N. (2003). Fault Detection and Isolation Based on PCA: An Industrial Reheating Furnance Case Study. International Conference on Systems, Man and Cybernetics, Hangzhou, China, 2, 1193–1198.         [ Links ]

11. Liang W., Wang B., Ma L., Zhang J. & Gao J. (2005). Fault Diagnosis Approach Based on the Integration of Qualitative Model and Quantitative Knowledge of Signed Directed Graph. Fourth International Conference on Machine Learning and Cybernetics, Guangzhou, China 4, 2251–2256.         [ Links ]

12. Nieto J.P., Garza L.E. & Morales R. (2007). Fault Detection Combining PCA, Control Charts and Statistic Operation Limits. Research on Computing Science, 27, 267–278.         [ Links ]

13. Nieto J.P., Garza L.E. & Morales R. (2007). Multiple Fault Diagnosis in Electrical Power Systems with Probabilistic Neural Networks. 6th Mexican International Conference on Artificial Intelligence. Aguascalientes, Mexico, 71–81 .         [ Links ]

14. Nieto J.P., Garza L.E., Garza M. & Morales R. (2008). Fault Diagnosis of Industrial Systems with Bayesian Networks and Neural Networks. 7th Mexican International Conference on Artificial Intelligence. Mexico, 998–1008.         [ Links ]

15. Nieto J.P., Garza L.E., Rabhi A., & El Hajjaji A. (2008). Fault Diagnosis of a Vehicle with soft Computing Methods. 7th Mexican International Conference on Artificial Intelligence. Mexico, 492–502.         [ Links ]

16. Nieto J.P., Garza L.E., Rabhi A., & El Hajjaji A. (2008). Fault Detection and Diagnosis of a Vehicle Combining Multidimensional Scaling, Percentiles Range and PCA. 9th International conference on Sciences and Techniques of Automatic control & computer engineering. Tunis, Tunisia, 1–11.         [ Links ]

17. Nieto J.P., Garza L.E., Rabhi A., El Hajjaji A. & Morales R.(2009). Vehicle Fault Detection and Diagnosis combining AANN and ANFIS. 7th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes. Barcelona, Spain. 1079–1084.         [ Links ]

18. Nieto J.P. & Garza L.E. (in press). Fault Detection with Statistical and Soft Computing Methods. In Lea M. Simon (Ed), Fault Detection: Theory, Methods and Systems. New York: Nova Science Publishers, Inc.         [ Links ]

19. Ren H. & Mi Z. (2006). Power System Fault Diagnosis Modeling Techniques based on Encoded Petri Nets. IEEE Power Engineering Society General Meeting. Montreal, Canada, 6, 1111–1118.         [ Links ]

20. Shi W., Yan H., & Ma K. (2005). A New Method of Early Fault Diagnosis Based on Machine Learning. Fourth International Conference on Machine Learning and Cybernetics. Guangzhou, China, 6, 3271–3276.         [ Links ]

21. Venkatasubramanian V., Rengaswamy R., Yin K. & Kavuri S. (2003). A review of process fault detection and diagnosis Part I, Part II and Part III. Computers and Chemical Engineering. 27, 293311.         [ Links ]

22. Wang W., Bai X., Zhao W., Ding J. & Fang Z. (2005). Hybrid Power System Model and the Method for Fault Diagnosis. IEEE/PES Transmission and Distribution Conference & Exhibition: Asia and Pacific. Dalian, China, 1–5.         [ Links ]

23. Xu L. & Chow M. (2005). Power Distribution Systems Fault Cause Identification Using Logistic Regression and Artificial Neural Network. 13th International Conference on Intelligent Systems Application to Power Systems. North Carolina, USA, 163–168.         [ Links ] Yongli Z., Limin H. & Jinling L. (2006) . Bayesian Networks–Based Approach for Power Systems fault Diagnosis. IEEE Transactions on power Delivery, 21(2), 634–639.         [ Links ]

24. Zanzouri N. & Tagina M. (2002). A Comparative Study of Hybrid System Monitoring Based on Bond Graph and Petri Net Modeling. International Conference on Systems, Man and Cybernetics. Tunis, Tunisia, vol(4), 1–6.         [ Links ]

25. Zhang D., Dai S., Zheng Y., Zhang R. & Mu P. (2000). Researches and Applications of a Hybrid Fault Diagnosis System. 3r world Congress on Intelligent Control and Automation. Hefei, China, 215– 219.         [ Links ]

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