Abstract
In this paper we develop neural network-based models to estimate the top and bottom product composition in a pilot plant distillation column. We study and compare the performance of several recurrent neural network architectures, namely, the Multilayer Neural Network in Parallel-configuration (MLNP), the Jordan Sequential Neural Network (JSNN), the Elman Recurrent Neural Network (ELNN), the Diagonal Recurrent Neural Network DRNN and the State Predictor Neural Network (SPNN). The models obtained can produce multi-stepahead predictions, and therefore can be considered an alternative for on-line composition analyzers. We find that the JSNN-based model gives worse results with respect to the other models, when used in the identification of the distillation process under consideration. This may be due to structural limitations of networks of the JSNN type.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
J. L. Elman. Finding structure in time. Cognitive Science, 14:179–211, 1990.
I. Humphrey and S. Salazar. Simulación estática y dinámica e identificación con redes neurales recurrentes de una columna de destilación binaria metanol-agua. Undergraduate thesis. Universidad Simón Bolívar. Caracas, 1996.
M. I. Jordan. Attractor dynamics and parallelism in a connectionist sequential machine. In Proceedings of the Eighth Annual Conference of the Cognitive Science Society, pages 531–546. Hillsdale: Erlbaum, 1986.
C. C Ku and K. Lee. Diagonal recurrent neural networks for dynamic systems control. IEEE Transactions on Neural Networks, 6(1):144–155, January 1995.
A. U. Levin. Neural Networks in Dynamical Systems: a System Theoretic Approach. PhD thesis, Yale University, 1992.
W. Luyben, editor. Practical Distillation Control. Van Nostrand Reinhold, 1992.
E. Margaglio. Modélisation et commande de processus par des méthodes neuronales et floues: applications aux colonnes de distillation. PhD thesis, Institut National des Sciences Appliquées (INSA) de Rennes, May 1999.
E. Margaglio, R. Lamanna, and P.-Y Glorennec. An experimental comparison of recurrent neural networks in nonlinear identification. In Proceedings of 2nd IMACS International Multiconference on Computing Engineering in Systems Applications, volume 2, pages 329–334, April 1998.
K. Narendra and K. Parthasarathy. Identification and control of dynamical systems using neural networks. IEEE Transactions on Neural Networks, 1(1):4–27, March 1990.
O. Nerrand, P. Roussel-Ragot, L. Personnaz, and G. Dreyfus. Neural networks and nonlinear adaptive filtering: Unifying concepts and new algorithms. Neural Computation, 5(4):165–199, 1993.
I. Rivals. Modélisation et commande de processus par réseaux de neurones: Applications au pilotage d’un véhicule autonome. PhD thesis, Université de Paris 6, January 1995.
H. T. Siegelmann and E. D. Sontag. Some recent results on computing with “neu-ral nets”. In Proceedings of the 30st IEEE Conference on Decision and Control, pages 1476–1481, December 1992.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Margaglio, E., Lamanna, R., Glorennec, PY. (2000). Analysis and Comparison of Recurrent Neural Networks for the Identification of a Pilot Plant Distillation Column. In: Monard, M.C., Sichman, J.S. (eds) Advances in Artificial Intelligence. IBERAMIA SBIA 2000 2000. Lecture Notes in Computer Science(), vol 1952. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44399-1_46
Download citation
DOI: https://doi.org/10.1007/3-540-44399-1_46
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41276-2
Online ISBN: 978-3-540-44399-5
eBook Packages: Springer Book Archive