Réseaux de neurones récurrents à fonctions de base radiales: RRFR Application au pronostic

This paper introduces a Recurrent Radial Basis Function network (RRBF) for non- linear system prognosis. The training process is divided in two stages. First, the parameters of the RRBF are determined by the unsupervised k-means algorithm. The ineffectiveness of this algorithm is improved by the FuzzyMinMax technique. In the second stage, a multivariable linear regression supervised learning technique is used to determine the weights of the connections between the hidden and output layer. We test the RRBF on the Box and Jenkins furnace database. This application shows that the RRBF is able to predict the evolution of a non-linear system. The performances of the RRBF are compared with those of the TDRBF. The RRBF gives better results for long run predictions. The FuzzyMinMax technique makes the K-means more stable. MOTS-CLES : Maintenance preventive, Surveillance, Pronostic, Reseaux de neurones temporels, RFR - Reseaux de neurones a fonctions de base radiales.

[1]  H. Demmou,et al.  Temporal sequence learning with neural networks for process fault detection , 1993, Proceedings of IEEE Systems Man and Cybernetics Conference - SMC.

[2]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[3]  N. Hernández-Gress.  Système de diagnostic par réseaux de neurones et statistiques : application à la détection d'hypovigilance du conducteur automobile , 1998 .

[4]  Béatrice Devauchelle-Gach.  Diagnostic mécanique des fatigues sur les structures soumises à des vibrations en ambiance de travail , 1991 .

[5]  Sung Yang Bang,et al.  An Efficient Method to Construct a Radial Basis Function Neural Network Classifier , 1997, Neural Networks.

[6]  Claire Cussenot.  Surveillance et diagnostic de la chaine de depollution d'une automobile , 1996 .

[7]  Stephen Jose Hanson,et al.  A Neural Network Autoassociator for Induction Motor Failure Prediction , 1995, NIPS.

[8]  Giovanni Soda,et al.  Unified Integration of Explicit Knowledge and Learning by Example in Recurrent Networks , 1995, IEEE Trans. Knowl. Data Eng..

[9]  Joydeep Ghosh,et al.  Knowledge enhancement and reuse with radial basis function networks , 2002, Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290).

[10]  J. Wade Davis,et al.  Statistical Pattern Recognition , 2003, Technometrics.

[11]  Noureddine Zerhouni,et al.  The RRBF. Dynamic representation of time in radial basis function network , 2001, ETFA.

[12]  Hervé Poulard.  Statistiques et réseaux de neurones pour un système de diagnostic : application au diagnostic de pannes automobiles. (Statistic and neural networks for a diagnosis system: Application to automotive failure detection) , 1996 .

[13]  Lei Xu,et al.  RBF nets, mixture experts, and Bayesian Ying-Yang learning , 1998, Neurocomputing.

[14]  Hamid Demmou,et al.  Using Self-Recurrent Neurons for Fault Detection and Diagnosis , 1995 .

[15]  Michael I. Jordan.  Serial Order: A Parallel Distributed Processing Approach , 1997 .

[16]  Abdoul Karim Armand Toguyeni.  Surveillance et diagnostic en ligne dans les ateliers flexibles de l'industrie manufacturière , 1992 .

[17]  N. Zerhouni,et al.  From the spherical to an elliptic form of the dynamic RBF neural network influence field , 2002, Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290).

[18]  Venkat Venkatasubramanian,et al.  Challenges in the industrial applications of fault diagnostic systems , 2000 .

[19]  Sun-Yuan Kung,et al.  Estimation of elliptical basis function parameters by the EM algorithm with application to speaker verification , 2000, IEEE Trans. Neural Networks Learn. Syst..

[20]  O. Daniel,et al.  Les réseaux de Pétri stochastiques pour l'évaluation des attributs de la sureté de fonctionnement des systèmes manufacturiers , 1995 .

[21]  Mohamad T. Musavi,et al.  On the training of radial basis function classifiers , 1992, Neural Networks.

[22]  Geoffrey E. Hinton,et al.  Phoneme recognition using time-delay neural networks , 1989, IEEE Trans. Acoust. Speech Signal Process..

[23]  Terrence J. Sejnowski,et al.  NETtalk: a parallel network that learns to read aloud , 1988 .

[24]  Marios M. Polycarpou,et al.  Neural-network-based robust fault diagnosis in robotic systems , 1997, IEEE Trans. Neural Networks.

[25]  Tomaso A. Poggio,et al.  Extensions of a Theory of Networks for Approximation and Learning , 1990, NIPS.

[26]  Raghunathan Rengaswamy,et al.  A syntactic pattern-recognition approach for process monitoring and fault diagnosis , 1995 .

[27]  References , 1971 .

[28]  Alain Grumbach,et al.  A Kohonen Map for Temporal Sequences , 1996 .

[29]  P. E. Keller,et al.  Three neural network based, sensor systems for environmental monitoring , 1994, Proceedings of ELECTRO '94.

[30]  Joydeep Ghosh,et al.  A neural network based hybrid system for detection, characterization, and classification of short-duration oceanic signals , 1992 .

[31]  ’. aboratoired,et al.  APPLICATION OF THE DYNAMIC RBF NETWORK IN A MONITORING PROBLEM OF THE PRODUCTION SYSTEMS , 2002 .

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

[33]  James M. Hutchinson,et al.  A radial basis function approach to financial time series analysis , 1993 .

[34]  Jonathan S. Maltz,et al.  NEURAL NETWORKS FOR PNEUMATIC ACTUATOR FAULT DETECTION , 1999 .

[35]  Racoceanu Daniel,et al.  APPLICATION OF THE DYNAMIC RBF NETWORK IN A MONITORING PROBLEM OF THE PRODUCTION SYSTEMS , 2002 .

[36]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[37]  Michael R. Berthold,et al.  A time delay radial basis function network for phoneme recognition , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[38]  P. Weber.  Diagnostic de procédé par l'analyse des estimations paramétriques de modèles de représentation à temps discret , 1999 .

[39]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[40]  P. Werbos,et al.  Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences , 1974 .

[41]  Noureddine Zerhouni,et al.  Modular modeling and analysis of a distributed production system with distant specialised maintenance , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[42]  Christophe Combastel.  Méthodes d'aide à la décision pour la détection et la localisation de défauts dans les entraînements électriques , 2000 .

[43]  Ronald J. Williams,et al.  A Learning Algorithm for Continually Running Fully Recurrent Neural Networks , 1989, Neural Computation.

[44]  Z. Ryad,et al.  The RRBF. Dynamic representation of time in radial basis function network , 2001, ETFA 2001. 8th International Conference on Emerging Technologies and Factory Automation. Proceedings (Cat. No.01TH8597).

[45]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[46]  M. J. Hudak.  RCE classifiers: theory and practice , 1992 .

[47]  C. Micchelli.  Interpolation of scattered data: Distance matrices and conditionally positive definite functions , 1986 .

[48]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[49]  Michael R. Berthold,et al.  Boosting the Performance of RBF Networks with Dynamic Decay Adjustment , 1994, NIPS.

[50]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[51]  M. Basseville,et al.  Surveillance et diagnostic de systèmes dynamiques: approches complémentaires du traitement de signal et de l'intelligence artificielle , 1996 .

[52]  Marios M. Polycarpou,et al.  Neural network based fault detection in robotic manipulators , 1998, IEEE Trans. Robotics Autom..

[53]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[54]  Padhraic Smith,et al.  Detecting novel fault conditions with hidden Markov models and neural networks , 1994 .

[55]  M.H. Hassoun,et al.  Fundamentals of Artificial Neural Networks , 1996, Proceedings of the IEEE.