Combining First Principles Models and Neural Networks for Generic Model Control

  • Janos Abonyi
  • Janos Madar
  • Ferenc Szeifert


Generic Model Control (GMC) is a control algorithm capable of using non-linear process model directly. In GMC, mostly, first-principles models derived from dynamic mass, energy and momentum balances are used. When the process is not perfectly known, the unknown parts of first principles models can be represented by black-box models, e.g. by neural networks. This paper is devoted to the application of such hybrid models in GMC. It is shown that the first principles part of the hybrid model determines the dominant structure of the controller, while the black-box elements are used as state and/or disturbance estimators. The sensitivity approach is used for the identification of the neural network elements of the control-relevant hybrid model. The underlying framework is illustrated by the temperature control of a continuous stirred tank reactor (CSTR) where a neural network is used to model the heat released by an exothermic chemical reaction.


Extended Kalman Filter Continuous Stir Tank Reactor Principle Model Dynamic Matrix Control Underlying Framework 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Lee P.L., Sullivian G. R., “Generic Model Control (GMC)”, Comput. Chem. Engng.,vol.12 no.6, pp.573–580, (1987)Google Scholar
  2. 2.
    Lee P. L., Newell R.B., Generic model control-a case study, Canad. J. Chem. Engng., 67, 478–483,1989CrossRefGoogle Scholar
  3. 3.
    Xie X.Q., Zhou D.H,. Jin Y.H., A Novel Approach to Constrained Generic Model Control based on Quadratic Programming, Industrial & Engineering Chemistry Research, 39, 4, 989–998, 2000CrossRefGoogle Scholar
  4. 4.
    Signal P. D., Lee, P. L., “Robust stability and performance analysis of generic model control (GMC)”, Chem. Eng. Comm.vol.124, pp.57–76, (1993)Google Scholar
  5. 5.
    Dimitris C. P., Ungar L. H., “A hybrid neural network-first-principles approach to process modelling”, AIChE Journal,vol.38 no.10, pp. 1499–1511, (1992)Google Scholar
  6. 6.
    Thompson M. L., Kramer M.A., “Model chemical processes using priori knowledge and neural networks” AIChE Journal,vol.40 no.8, pp. 1328–1340, (1994)Google Scholar
  7. 7.
    Pottman M., Pearson R.K., Block--Oriented NARMAX Models with Output Multiplicities, AIChE Journal, 44, 131–140, 1998CrossRefGoogle Scholar
  8. 8.
    Bequette B. W., Sistu P. B., “Nonlinear predictive control of uncertain processes: application to CSTR”, AIChE Journal,vol.37 no.11, pp. 1711–1723, (1991)Google Scholar
  9. 9.
    Cho W., Lee J., Edgar T. F., “Control system design based on a nonlinear first-order plus time delay model”, J. Proc. Vont., vol.7 no.1, pp. 65–73, (1997)Google Scholar
  10. 10.
    Hernandez E., Arkun Y., “Study of the control-relevant propeties of back-propagation neural network models of nonlinear dynamical systems”, Comput. chem. Engng.,vol.16 no.4, pp. 227–240, (1992)Google Scholar

Copyright information

© Springer-Verlag London 2002

Authors and Affiliations

  • Janos Abonyi
    • 1
  • Janos Madar
    • 1
  • Ferenc Szeifert
    • 1
  1. 1.Department of Process EngineeringUniversity of VeszprémVeszprémHungary

Personalised recommendations