Control of dissolved oxygen concentration using neural network in a batch bioreactor

Mete T., ÖZKAN G., HAPOĞLU H., Alpbaz M.

COMPUTER APPLICATIONS IN ENGINEERING EDUCATION, cilt.20, sa.4, ss.619-628, 2012 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 20 Sayı: 4
  • Basım Tarihi: 2012
  • Doi Numarası: 10.1002/cae.20430
  • Dergi Adı: COMPUTER APPLICATIONS IN ENGINEERING EDUCATION
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.619-628
  • Anahtar Kelimeler: artificial neural networks, model-based control, NARMA-L2, fermentation bioreactor model, dissolved oxygen concentration control, system identification, PREDICTIVE CONTROL, ADAPTIVE-CONTROL, FERMENTATION, IMPLEMENTATION, TEMPERATURE, REACTOR
  • Ankara Üniversitesi Adresli: Evet

Özet

Artificial neural networks (ANN) have been utilized for many chemical applications because of their ability to learn system features. This paper presents the use of feedforward neural networks for dynamic modelling and dissolved oxygen (DO) control of a batch yeast fermentation. The ARMAX model of this nonlinear process is also presented. Model verification is tested by using experimental data. Different ANNs are trained using the backpropagation learning algorithm. The resulting ANNs are introduced in a Model Predictive Control (MPC) scheme to test the control performance of the structure. At system, output variable that is DO concentration and adjusting variable that is air flow rate are chosen. The robustness of this control structure is studied in the case of setpoint changes. Results obtained with NN-MPC is also presented and compared with those obtained with Nonlinear Auto Regressive Moving Average (NARMA-L2) control strategy. (C) 2010 Wiley Periodicals, Inc. Comput Appl Eng Educ 20: 619628, 2012