Industrial reactor modelling with artificial neural networks

Abstract

This paper describes the development of artificial neural network based models of a large-scale industrial reaction process. The process operates continuously but is subject to changes in product demand and variations in raw material quality. The models are required for the purpose of process optimisation which poses additional modelling demands as predictions of process constraints are also required. A blend of artificial neural network and standard linear regression models was found to be most appropriate. The models have subsequently been incorporated within an advisory system to provide the operators with advice on process set-points to maximise process profitability within the constraints. Results from process operation serve to demonstrate the predictive ability of the models and serve to highlight important application lessons. An overview of the advisory system concept is included to place the models in context.

Keywords

Artificial neural networks modelling industrial reactor linear regression advisory system

Get full access to this article

View all access options for this article.

References

Gent, C. , Montague, G.A. , Buttress, J. , and Morris, A.J. 1996. 'A neural network based advisory system for reactor optimisation '. In preparation to be submitted to Trans I Chem E.

Hoskins, J.C. , and Himmelblau, D.M. 1988. 'Artificial neural network models of knowledge representation in chemical engineering', Comput Chem Engineering 12(9/10), 881-890.

Hoskins, J.C. , Kaliyur, K.M. , and Himmelblau, D.M. 1991. 'Fault diagnosis in complex chemical plants using artificial neural networks', AIChE Journal , 37(1), 137-141.

Hunt, K.J. , and Sbarbaro, D. 1992. 'Studies in neural network based control'. In: Warwick, K. , Irwin, G. W. , and Hunt, K. J. (Eds) Neural networks for control and systems. Peter Peregrinus

Karim, M.N. , and Rivera, S.L. 1992. 'Application of neural networks in bioprocess state estimation' , Proc Amer Control Conf, WA14, 495-499.

Lakrori, M. , and Cheruy, A. 1987. 'Un nouvelle possibilite pour las conduite des bioprocedes' , Biosciences, 6(5), 154-160

Lightbody, G. , Wu, Q.H. , and Irwin, G.W. 1992. 'Control applications for feedforward networks' In: Warwick, K. , Irwin, G. W. , and Hunt, K. J. (Eds) Neural networks for control and systems, 51-71.

Manchanda, S. , and Tham, M.T. 1991. 'An evaluation of non-linear control strategy', Proc IEE Cont '91, 706-711.

McGreavy, C. , Lu, M.L. , and Wang, X.Z. 1994. 'Characterisation of the behaviour and product distribution in fluid catalytic cracking using neural networks', Chem Eng Sci, 49, 4717-4724.

10.

Montague, G.A. , Morris, A.J. , and Tham, M.T. 1992. 'Enhancing bioprocess operability with generic software sensors'. J Biotechnol, 25, 183-201.

11.

Montague, G.A. , Gent, C. , Morris, A.J. , and Buttress, J. 1995. 'Industrial reactor modelling with artificial neural networks . Submitted to Proc IEE PtD.

12.

Patton, R.J. , Chen, T.M. , and Slew, T.M. 1994. 'Fault diagnosis in non-liner dynamic systems via neural networks', Proc Control '94, Conference Publication No 389, IEE, 1346-1351.

13.

Thilbault, J. , Van Breusegem, V. , and Cheruy, A. 1990. 'On-line prediction of fermentation variables using neural networks', Biotechnol Bioeng, 36, 1041-1048.

14.

Turner, P. , Montague, G.A. , and Morris, A.J. 1994. 'Neural networks in process plant modelling and control ', Computing and Control Engineering Journal, June, 131-134.

15.

Turner, P. , Montague, G.A. , and Morris, A.J. 1996. 'Non-linear and direction dependent process modelling using neural networks', Proc IEE Pt D, 143(1),44-48.

16.

Venkatasubramanian, B. , Vaidyanathan, R. , and Yamamato, Y. 1990. 'Process fault detection and diagnosis using neural networks-1. Steady state processes', Comput Chem Engng, 14(7), 699-712.

17.

Willis, M.J. , Di Massimo, C. , Montague, G.A. , Tham, M.T. , and Morris, A.J. 1992. 'Artificial neural networks in process estimation and control', Automatica, 28(6), 1181-1187.