Neural network adaptive control of the penicillin acylase fermentation

Mei-Jywan Syu, C. B. Chang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

A neural network was used as the adaptive controller for the control of pH during a batch fermentation of the penicillin acylase. Recurrent backpropagation network (RBPN) with a transfer function was chosen as the controller model for its better ability in longer-term identification. The model was operated by two phases. During the first phase, it was set as the process model and trained by a fixed set of on-line acquired data. The predicted control action was obtained during the second phase. A moving window with size of 15 for supplying training data was determined and applied for on-line learning.

Original languageEnglish
Title of host publicationIEEE International Conference on Neural Networks - Conference Proceedings
PublisherIEEE
Pages639-644
Number of pages6
Volume2
Publication statusPublished - 1997
EventProceedings of the 1997 IEEE International Conference on Neural Networks. Part 4 (of 4) - Houston, TX, USA
Duration: 1997 Jun 91997 Jun 12

Other

OtherProceedings of the 1997 IEEE International Conference on Neural Networks. Part 4 (of 4)
CityHouston, TX, USA
Period97-06-0997-06-12

Fingerprint

Fermentation
Neural networks
Controllers
Backpropagation
Transfer functions
Identification (control systems)

All Science Journal Classification (ASJC) codes

  • Software
  • Control and Systems Engineering
  • Artificial Intelligence

Cite this

Syu, M-J., & Chang, C. B. (1997). Neural network adaptive control of the penicillin acylase fermentation. In IEEE International Conference on Neural Networks - Conference Proceedings (Vol. 2, pp. 639-644). IEEE.
Syu, Mei-Jywan ; Chang, C. B. / Neural network adaptive control of the penicillin acylase fermentation. IEEE International Conference on Neural Networks - Conference Proceedings. Vol. 2 IEEE, 1997. pp. 639-644
@inproceedings{865492174ad449bc9df8b27b1016cf4b,
title = "Neural network adaptive control of the penicillin acylase fermentation",
abstract = "A neural network was used as the adaptive controller for the control of pH during a batch fermentation of the penicillin acylase. Recurrent backpropagation network (RBPN) with a transfer function was chosen as the controller model for its better ability in longer-term identification. The model was operated by two phases. During the first phase, it was set as the process model and trained by a fixed set of on-line acquired data. The predicted control action was obtained during the second phase. A moving window with size of 15 for supplying training data was determined and applied for on-line learning.",
author = "Mei-Jywan Syu and Chang, {C. B.}",
year = "1997",
language = "English",
volume = "2",
pages = "639--644",
booktitle = "IEEE International Conference on Neural Networks - Conference Proceedings",
publisher = "IEEE",

}

Syu, M-J & Chang, CB 1997, Neural network adaptive control of the penicillin acylase fermentation. in IEEE International Conference on Neural Networks - Conference Proceedings. vol. 2, IEEE, pp. 639-644, Proceedings of the 1997 IEEE International Conference on Neural Networks. Part 4 (of 4), Houston, TX, USA, 97-06-09.

Neural network adaptive control of the penicillin acylase fermentation. / Syu, Mei-Jywan; Chang, C. B.

IEEE International Conference on Neural Networks - Conference Proceedings. Vol. 2 IEEE, 1997. p. 639-644.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Neural network adaptive control of the penicillin acylase fermentation

AU - Syu, Mei-Jywan

AU - Chang, C. B.

PY - 1997

Y1 - 1997

N2 - A neural network was used as the adaptive controller for the control of pH during a batch fermentation of the penicillin acylase. Recurrent backpropagation network (RBPN) with a transfer function was chosen as the controller model for its better ability in longer-term identification. The model was operated by two phases. During the first phase, it was set as the process model and trained by a fixed set of on-line acquired data. The predicted control action was obtained during the second phase. A moving window with size of 15 for supplying training data was determined and applied for on-line learning.

AB - A neural network was used as the adaptive controller for the control of pH during a batch fermentation of the penicillin acylase. Recurrent backpropagation network (RBPN) with a transfer function was chosen as the controller model for its better ability in longer-term identification. The model was operated by two phases. During the first phase, it was set as the process model and trained by a fixed set of on-line acquired data. The predicted control action was obtained during the second phase. A moving window with size of 15 for supplying training data was determined and applied for on-line learning.

UR - http://www.scopus.com/inward/record.url?scp=0030688766&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030688766&partnerID=8YFLogxK

M3 - Conference contribution

VL - 2

SP - 639

EP - 644

BT - IEEE International Conference on Neural Networks - Conference Proceedings

PB - IEEE

ER -

Syu M-J, Chang CB. Neural network adaptive control of the penicillin acylase fermentation. In IEEE International Conference on Neural Networks - Conference Proceedings. Vol. 2. IEEE. 1997. p. 639-644