ARPN Journal of Science and Technology Logo

ARPN Journal of Science and Technology >> Volume 7, Issue 1, January 2017

ARPN Journal of Science and Technology


On the Application of Neural Network Predictive Controller For Stirred Tank Reactor

Full Text Pdf Pdf
Author D. I. Lanlege, L.A. Nafiu, U.M.Gana, A.A.Falaye
ISSN 2225-7217
On Pages 301-308
Volume No. 3
Issue No. 3
Issue Date April 01, 2013
Publishing Date April 01, 2013
Keywords Neural Network, proportional Integral Derivative (PID), Predictive Controller, Stirred Tank Reactor and Hybrid Algorithm



Abstract

In this paper, we proposed a Proportional Integral Derivative (PID) Neural Network Algorithm, which is used to model and solve continuous stirred tank mixer (CSTM) problem. This hybrid algorithm is robust and converges fast without being trapped into a local minimal as it is the case with the conventional neural network. We established the characteristics equation governing the dynamics of the continuous stirred tank mixer/ reactor. A controller was formulated tested and found to be consistent. The Proportional Integral Derivative (PID) network was used to simulate typical continuous stirred tank reactor (CSTR) problems, of which predictive accuracy was found to be 96%.


Back
Seperator
    Journal of Computing | Journal of Systems and Software     
Copyrights
2013 ARPN Publishers