Neural Network Control of CSTR for Reversible Reaction Using Reverence Model Approach

In this work, non-linear control of CSTR for reversible reaction is carried out using Neural Network as design tool. The Model Reverence approach in used to design ANN controller. The idea is to have a control system that will be able to achieve improvement in the level of conversion and to be able to track set point change and reject load disturbance. We use PID control scheme as benchmark to study the performance of the controller. The comparison shows that ANN controller out perform PID in the extreme range of non-linearity. This paper represents a preliminary effort to design a simplified neutral network control scheme for a class of non-linear process. Future works will involve further investigation of the effectiveness of thin approach for the real industrial chemical process

[1]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[2]  Mohd Azlan Hussain,et al.  Adaptive sliding mode control with neural network based hybrid models , 2004 .

[3]  Rutherford Aris,et al.  An analysis of chemical reactor stability and control—I: The possibility of local control, with perfect or imperfect control mechanisms , 1958 .

[4]  Marco Saerens,et al.  Neural controller based on back-propagation algorithm , 1991 .

[5]  Barry Lennox,et al.  Industrial application of neural networks — an investigation , 2001 .

[6]  Michael Nikolaou,et al.  Control of nonlinear dynamical systems modeled by recurrent neural networks , 1993 .

[7]  Gholamreza Zahedi,et al.  A Neural Network Approach for Prediction of the CuO-ZnO-Al2O3 Catalyst Deactivation , 2005 .

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

[9]  Yang Yinghua,et al.  Neural network pole placement controller for nonlinear systems through linearisation , 1997, Proceedings of the 1997 American Control Conference (Cat. No.97CH36041).

[10]  Dale E. Seborg,et al.  Nonlinear internal model control strategy for neural network models , 1992 .

[11]  Kenneth Levenberg.  A METHOD FOR THE SOLUTION OF CERTAIN NON – LINEAR PROBLEMS IN LEAST SQUARES , 1944 .

[12]  Neal R. Amundson,et al.  An analysis of chemical reactor stability and control—XVI , 1981 .

[13]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[14]  Martin T. Hagan,et al.  Neural network design , 1995 .

[15]  Richard S. Sutton,et al.  Neuronlike adaptive elements that can solve difficult learning control problems , 1983, IEEE Transactions on Systems, Man, and Cybernetics.