Premium
The design of experiments, training and implementation of nonlinear controllers based on neural networks
Author(s) -
Dayal Bhupinder S.,
Taylor Paul A.,
Macgregor John F.
Publication year - 1994
Publication title -
the canadian journal of chemical engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.404
H-Index - 67
eISSN - 1939-019X
pISSN - 0008-4034
DOI - 10.1002/cjce.5450720618
Subject(s) - artificial neural network , nonlinear system , control theory (sociology) , model predictive control , feed forward , internal model , control engineering , computer science , controller (irrigation) , feedforward neural network , nonlinear control , control (management) , engineering , artificial intelligence , agronomy , physics , quantum mechanics , biology
In the area of nonlinear predictive control, several control schemes using artificial neural networks have been proposed. In this work, the issues relating to the information contents of the data used to train the neural network components of these nonlinear predictive control schemes are considered. This raises questions about the design of experiments. A class of feedback‐feedforward nonlinear controller based on the model predictive structure (also known as Internal Model Control, IMC, structure) is investigated. The implementation and performance of these neural network based controllers, together with comparisons to other nonlinear and linear controllers, are illustrated on two nonlinear continuous‐stirred‐tank‐reactor simulations.