Modeling and simulation of a reactive packed distillation column using delayed neural networks

Giwa A., KARACAN S.

4th International Conference on Chaotic Modeling and Simulation, CHAOS 2011, Agios Nikolaos, Crete, Yunanistan, 31 Mayıs - 03 Haziran 2011, ss.129-136 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Agios Nikolaos, Crete
  • Basıldığı Ülke: Yunanistan
  • Sayfa Sayıları: ss.129-136
  • Anahtar Kelimeler: Correlation coefficient (R), Delayed Neural Network (DNN), MATLAB, Mean squared error (MSE), Nonlinear AutoRegressive (NAR), Nonlinear AutoRegressive with eXogenous inputs (NARX), Nonlinear Input-Output (IO), Reactive packed distillation column
  • Ankara Üniversitesi Adresli: Evet

Özet

© 2019 CHAOS 2011 - 4th Chaotic Modeling and Simulation International Conference, Proceedings. All rights reserved.The complex nature of the reactive packed distillation column owing to the occurrence of both reactions and separations in a single unit brought up the need for the search for a very robust tool of representing the process. In view of this, delayed neural networks are considered as tools that can handle this problem effectively. As such, in this work, Nonlinear AutoRegressive, Nonlinear AutoRegressive with eXogenous inputs and Nonlinear Input-Output models are developed and simulated with the aid of MATLAB R2010b to predict the top and bottom sections temperatures. The predicted results obtained from the Input-Output models were not satisfactory. However, observing the good agreements from the plots as well as the correlation coefficients and the mean squared errors between the predicted results of the NAR and NARX models and the experimental ones showed that these two models can be used to represent the reactive packed distillation column.