Neural networks based model predictive control of an industrial polypropylene process

A model predictive control (MPC) strategy based on a feedforward neural network model is proposed for an industrial polypropylene process. To infer product properties on-line, a dynamic process model is developed and a recursive prediction error method is used to update the model parameters when there is a significant model prediction error. To obtain optimal control strategy during grade transitions, a nonlinear MPC controller is developed based on a neural network model, which is trained using the input-output data of the process model. Performance of the nonlinear controller is compared with a conventional PID controller. Application results indicate that the MPC controller can obtain satisfactory performance and consequently results in significant reduction in transition time and product variability.

[1]  Cláudio Augusto Oller do Nascimento,et al.  Use of neural networks for modeling of olefin polymerization in high pressure tubular reactors , 1994 .

[2]  John F. MacGregor,et al.  Nonlinear product property control in industrial gas‐phase polyethylene reactors , 1993 .

[3]  Panos J. Antsaklis,et al.  Neural networks for control systems , 1990, IEEE Trans. Neural Networks.

[4]  Jie Zhang,et al.  Dynamic modelling of an industrial polypropylene reactor and its application in melt index prediction during grade transitions , 2002, Proceedings of the 2002 American Control Conference (IEEE Cat. No.CH37301).

[5]  J. Macgregor,et al.  On‐line inference of polymer properties in an industrial polyethylene reactor , 1991 .

[6]  Jie Zhang,et al.  Developing robust non-linear models through bootstrap aggregated neural networks , 1999, Neurocomputing.

[7]  Masahiro Ohshima,et al.  Quality control of polymer production processes , 2000 .

[8]  Morimasa Ogawa,et al.  Quality inferential control of an industrial high density polyethylene process , 1999 .

[9]  D. C. Psichogios,et al.  Direct and indirect model based control using artificial neural networks , 1991 .

[10]  T. Crowley,et al.  On-line monitoring and control of a batch polymerization reactor , 1996 .

[11]  Babatunde A. Ogunnaike.  On-line modelling and predictive control of an industrial terpolymerization reactor , 1994 .

[12]  Peter J. Gawthrop,et al.  Neural networks for control systems - A survey , 1992, Autom..

[13]  Manfred Morari,et al.  Model predictive control: Theory and practice - A survey , 1989, Autom..

[14]  Carlos E. Garcia,et al.  Internal model control. A unifying review and some new results , 1982 .

[15]  John E. Dennis,et al.  Numerical methods for unconstrained optimization and nonlinear equations , 1983, Prentice Hall series in computational mathematics.