Propuesta de procedimiento para configurar una red neuronal artificial de Base Radial con aplicaciones en el diagnóstico de fallos

En este articulo se presenta un procedimiento que permite la configuracion de los parametros de una arquitectura de red neuronal artificial de Base Radial para tareas de diagnostico de fallos, luego de establecer un orden logico para la seleccion de los mismos. Dicho procedimiento garantiza la obtencion del numero necesario de neuronas ocultas a partir de fijar el error de diagnostico deseado por los expertos para cada proceso, permitiendo tambien seleccionar el metodo para estimar los anchos de las neuronas ocultas y la ecuacion de distancia para la propagacion del espacio de entradas. El procedimiento se aplica al proceso de prueba “Tanque Reactor Continuamente Agitado” para demostrar su efectividad. Como resultado de los experimentos realizados, se concluye que la eleccion de la funcion de distancia no influye sobre la seleccion del metodo para estimar los anchos. Se comprueba que esta arquitectura, con el entrenamiento adecuado, exhibe buenas propiedades de sensibilidad y robustez para el diagnostico de fallos.

[1]  Khashayar Khorasani,et al.  Dynamic neural network-based fault diagnosis of gas turbine engines , 2014, Neurocomputing.

[2]  Abdessamad Kobi,et al.  Fault diagnosis of industrial systems by conditional Gaussian network including a distance rejection criterion , 2010, Eng. Appl. Artif. Intell..

[3]  Bonifacio Martín del Brío,et al.  Redes neuronales y sistemas borrosos: introducción, teoría y práctica / Bonifacio Martín del Brio, Alfredo Sanz Molina , 2001 .

[4]  Xuefeng Yan,et al.  Fault Diagnosis in Chemical Process Based on Self-organizing Map Integrated with Fisher Discriminant Analysis , 2013 .

[5]  Heikki N. Koivo,et al.  Application of artificial neural networks in process fault diagnosis , 1991, Autom..

[6]  Diane Larlus,et al.  A Supervised Clustering Algorithm for the Initialization of RBF Neural Network Classifiers , 2007, 2007 IEEE 15th Signal Processing and Communications Applications.

[7]  J. B. Gomm,et al.  Fault diagnosis of a nuclear processing plant at different operating points using neural networks , 1997 .

[8]  Azah Mohamed,et al.  An optimal radial basis function neural network for fault location in a distribution network with high penetration of DG units , 2013 .

[9]  Timo Sorsa,et al.  APPLICATION OF ARTIFICIAL NEURAL NETWORKS IN PROCESS FAULT DIAGNOSIS , 1992 .

[10]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part I: Quantitative model-based methods , 2003, Comput. Chem. Eng..

[11]  Timo Sorsa,et al.  Neural networks in process fault diagnosis , 1991, IEEE Trans. Syst. Man Cybern..

[12]  Paramasivan Saratchandran,et al.  Performance evaluation of a sequential minimal radial basis function (RBF) neural network learning algorithm , 1998, IEEE Trans. Neural Networks.

[13]  Vojislav Kecman,et al.  Learning and Soft Computing: Support Vector Machines, Neural Networks, and Fuzzy Logic Models , 2001 .

[14]  Yukihiro Toyoda,et al.  A parameter optimization method for radial basis function type models , 2003, IEEE Trans. Neural Networks.

[15]  Adrian G. Bors,et al.  Minimal Topology for a Radial Basis Functions Neural Network for Pattern Classification , 1994 .

[16]  Peter L. Lee,et al.  An integrated neural network/expert system approach for fault diagnosis , 1993 .

[17]  Francisco Herrera,et al.  A study on the use of statistical tests for experimentation with neural networks: Analysis of parametric test conditions and non-parametric tests , 2007, Expert Syst. Appl..

[18]  Chrissanthi Angeli,et al.  On-Line Fault Detection Techniques for Technical Systems: A Survey , 2004, Int. J. Comput. Sci. Appl..

[19]  Robert M. Pap,et al.  Fault Diagnosis , 1990, Bayesian Networks in Fault Diagnosis.

[20]  Andrew Ball,et al.  A RBF neural network model for cylinder pressure reconstruction in internal combustion engines , 1996 .

[21]  Juan Julián Merelo Guervós,et al.  Evolving RBF neural networks for time-series forecasting with EvRBF , 2004, Inf. Sci..

[22]  Krzysztof Patan,et al.  Artificial Neural Networks for the Modelling and Fault Diagnosis of Technical Processes , 2008 .

[23]  P. J. Dalianis,et al.  Fault diagnosis in complex systems using artificial neural networks , 1994, 1994 Proceedings of IEEE International Conference on Control and Applications.

[24]  Heikki N. Koivo,et al.  Artificial neural networks in fault diagnosis and control , 1994 .

[25]  A. Nandi,et al.  A comparison of artificial neural networks and other statistical methods for rotating machine condition classification , 1996 .

[26]  D. Seborg,et al.  Pattern Matching in Historical Data , 2002 .

[27]  Nicolaos B. Karayiannis,et al.  Reformulated radial basis neural networks trained by gradient descent , 1999, IEEE Trans. Neural Networks.

[28]  Benjamin Naumann,et al.  Learning And Soft Computing Support Vector Machines Neural Networks And Fuzzy Logic Models , 2016 .