Discrete-time controllers for temperature control of a polystyrene polymerization reactor


Hapoglu H., Ozkan G., Altuntas S.

JOURNAL OF THE INDIAN CHEMICAL SOCIETY, cilt.96, sa.9, ss.1227-1232, 2019 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 96 Sayı: 9
  • Basım Tarihi: 2019
  • Dergi Adı: JOURNAL OF THE INDIAN CHEMICAL SOCIETY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.1227-1232
  • Anahtar Kelimeler: Polymerization, polystyrene, discrete-time model, zero-order hold, discrete-time controllers, PREDICTIVE CONTROL, NEURAL-NETWORK, BATCH REACTOR, DESIGN
  • Ankara Üniversitesi Adresli: Evet

Özet

For temperature control, styrene free radical polymerization in a 1.1 I batch reactor was presented with the zero-order hold equivalent discrete time models in the present work. Two discrete-time controllers were applied to the reactor. Firstly, a discrete-time controller with three tuning parameters was designed, and its parameters were determined by checking control performance and closed loop pole-placement in z-plane. The sixth order closed loop transfer function between reactor temperature and set point was obtained. The best tuning parameters were determined as -0.1, -0.9 and -0.1. Effect of tuning parameters to stability was investigated. Secondly, the discrete-time proportional-integral-derivative (PID) controller (2DOF) was employed. Simulink control design was utilized for tuning the six parameters of the controller as -10.75, -1.09, -19.26, -0.689, 0.1686, 4.123e-0.5 respectively. These controller performances were compared to each other for a step decrease in set point. It is noted that a very close matching was achieved and the controllers have made the closed loop system desirable and stable with the best tuned parameters determined. In order to obtain good quality 52000 g/mole polystyrene, these two different discrete-time controllers were applied theoretically for tracking previously determined optimal temperature profile. The previously published experimental data were used for validation. Cooling water flow rate was manipulated to obtain better control performance with minimum response time and reduced production cost. In the set point tracking cases studied, the best tuned controllers acted to produce non-oscillatory closed loop behaviour without offset throughout the length of the batch run.