Identification and predictive control of the petroleum catalytic cracking process
DOI:
https://doi.org/10.31416/rsdv.v1i1.209Keywords:
Artificial Neural Networks, Algorithm DMC, Complex ProcessesAbstract
Artificial Neural Networks (ANNs) constitute a tool that has recently become used for a great number of successful applications. Its capacity with respect to empirical modeling of complex processes has been stimulating its application in the engineering field. This paper treats the control of the petroleum fluid catalytic cracking process based
on neural networks using the multilayer kind. This process is known by its strong non-linearity and interactions between its variables. For this control strategy, a network is used as internal model for the controller. The conventional model of Lee & Kugelman (1973) was used to obtain data. The stationary behavior was analyzed by bifurcation graphs for coke concentration in the spent and regenerated catalyst. From graphs it was determinate the stationary states expected for coke concentration in the catalyst inside the operation region. The riser temperature control based on ANNs was controlled as well as the dynamic matrix control (DMC). For the tuned parameters, the control based on ANNs was more efficient and conservative.
References
ALGHAZZAWI A., LENNOX B. - Model predictive control monitoring using multivariate statistics. Journal of Process Control 19, 314–327, 2009.
ARAÚJO A. S., M. G. F. Rodriguez. Desenvolvimento de Catalisadores Zeolíticos Destinados
à Reforma do Metano.Tese de Doutorado. Área de Concentração: Operações e Processos, Centro de Ciências e Tecnologia, Universidade Federal de Campina Grande. 2006.
ARMOR J. N. - A history of industrial catalysis. Catalysis Today, 2010.
DARAOUI N., DUFOUR P., HAMMOURI H., HOTTOT A. - Model predictive control during the primary drying stage of lyophilisation. Control Engineering Practice. 18, 483–494, 2010.
DE SOUZA JR. M. B. - Redes Neuronais Multicamadas Aplicadas a Modelagem e Controle de Processos Químicos. Tese de Doutorado. COPPE/PEQ/UFRJ. Rio de Janeiro, 1993.
DOEDEL E. J., FAIRGRIEVE T. F., SANDSTEDE B., CHAMPNEYS A. R., KUZNETSOV Y. A., WANG X. – Continuation and Bifurcation Software for Ordinary Differential Equations. Manual AUTO97. March 29, 1998.
FIEDLER FERRARA N. & PRADO C. P. C. – Caos - Uma Introdução. Editora Edgard Blircher Ltda. S. Paulo – SP. 1995.
FILETI A. M. F. – Controle Avançado de Processos: Adaptativo, Preditivo e Fuzzy. Modelagem Matemática via Redes Neurais. Aplicações em Processos Químicos. Material especialmente elaborado para curso ministrado na Escola Politécnica da Universidade Federal da Bahia. Julho, 1998.
LEE W. & KUGELMAN A. M. – Number of steady-state operating points and local stability of open-loop fluid catalytic cracker. Ind. Eng. Chem. Pocess Des. Dev., 12, 197-204, 1973.
LEONARD J. & KRAMER, M. A., Improvement of the Backpropagation Algorithm for Training Neural Networks, Computers chem. Engng., 14, N° 3, 337-341,1990.
NARENDRA, K. S. AND PARTHASARATHY, K. "Identification and Control of Dynamical Systems Using Neural Networks," IEEE Trans. Neural Networks, 1(1), 4-27, 1999.
RUMELHART D. E. & McCLELLAND J. L. – Parallel Distributed Processing: Explorations in the Microstructure of Cognition. 1, Foundations the Mit Press. Cambridge. 1986.
SCATTOLINI R. - Architectures for distributed and hierarchical Model Predictive Control – A review. Journal of Process Control.19, 723– 731, 2009.
STEWART B. T., VENKAT A. N., RAWLINGS, J. B., WRIGHT, S. J., PANNOCCHIA G. -Cooperative distributed model predictive control. Systems & Control Letters 59, 460_469, 2010.
SU H., MCAVOY T. J., WERBOS P. – Long- term Predictions of Chemical Processes Using Recorrent Neural Neworks: a Parallel Training Approach. Ind. Eng. Chem. Res., 31, 1338-1352, 1992.
