Neural network approach to identify batch cell growth

Mei-Jywan Syu, George T. Tsao

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

2 Citations (Scopus)

Abstract

A saturation-type transfer function with a backpropagation neural network (BPNN) was proposed for solving the modeling problem of batch cell growth system. Batch chemical processes are usually influenced by their initial conditions. For batch cell cultures, the initial state strongly governs the growth pattern during the timecourse. In modeling a chemical system, we are always interested in how to model the outcome of the system related to some affecting factors. In a batch system, some of the initial conditions are certainly important affecting factors. Trying to model the cell growth with information concerning only the initial conditions is not yet possible from a kinetic approach. The difficulty comes from numerical analysis and insufficient knowledge regarding certain growth parameters as they vary with time. Accordingly, neural network methodology with the concept developed earlier was proposed to solve this problem. The feasibility and capability of the neural network to model the pattern of batch cell growth by providing initial conditions only is tested in this study. A 2-3-8 BPNN with initial glucose and cell concentrations as the two inputs, cell densities measured at eight each hours as the eight outputs was thus constructed. The simulation and prediction results of this BPNN are presented to demonstrate the performance and applicability of this newly discovered transfer function. Sensitivity analysis of the initial factors from this neural network model (NNM) is also discussed. The optimization of the initial conditions for this system is also performed.

Original languageEnglish
Title of host publication1993 IEEE International Conference on Neural Networks
PublisherPubl by IEEE
Pages1742-1747
Number of pages6
ISBN (Print)0780312007
Publication statusPublished - 1993 Jan 1
Event1993 IEEE International Conference on Neural Networks - San Francisco, California, USA
Duration: 1993 Mar 281993 Apr 1

Publication series

Name1993 IEEE International Conference on Neural Networks

Other

Other1993 IEEE International Conference on Neural Networks
CitySan Francisco, California, USA
Period93-03-2893-04-01

Fingerprint

Cell growth
Neural networks
Backpropagation
Batch cell culture
Transfer functions
Sensitivity analysis
Glucose
Numerical analysis
Kinetics

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Control and Systems Engineering
  • Software
  • Artificial Intelligence

Cite this

Syu, M-J., & Tsao, G. T. (1993). Neural network approach to identify batch cell growth. In 1993 IEEE International Conference on Neural Networks (pp. 1742-1747). (1993 IEEE International Conference on Neural Networks). Publ by IEEE.
Syu, Mei-Jywan ; Tsao, George T. / Neural network approach to identify batch cell growth. 1993 IEEE International Conference on Neural Networks. Publ by IEEE, 1993. pp. 1742-1747 (1993 IEEE International Conference on Neural Networks).
@inproceedings{309a20b40f33485bb196b12bc1c5b4be,
title = "Neural network approach to identify batch cell growth",
abstract = "A saturation-type transfer function with a backpropagation neural network (BPNN) was proposed for solving the modeling problem of batch cell growth system. Batch chemical processes are usually influenced by their initial conditions. For batch cell cultures, the initial state strongly governs the growth pattern during the timecourse. In modeling a chemical system, we are always interested in how to model the outcome of the system related to some affecting factors. In a batch system, some of the initial conditions are certainly important affecting factors. Trying to model the cell growth with information concerning only the initial conditions is not yet possible from a kinetic approach. The difficulty comes from numerical analysis and insufficient knowledge regarding certain growth parameters as they vary with time. Accordingly, neural network methodology with the concept developed earlier was proposed to solve this problem. The feasibility and capability of the neural network to model the pattern of batch cell growth by providing initial conditions only is tested in this study. A 2-3-8 BPNN with initial glucose and cell concentrations as the two inputs, cell densities measured at eight each hours as the eight outputs was thus constructed. The simulation and prediction results of this BPNN are presented to demonstrate the performance and applicability of this newly discovered transfer function. Sensitivity analysis of the initial factors from this neural network model (NNM) is also discussed. The optimization of the initial conditions for this system is also performed.",
author = "Mei-Jywan Syu and Tsao, {George T.}",
year = "1993",
month = "1",
day = "1",
language = "English",
isbn = "0780312007",
series = "1993 IEEE International Conference on Neural Networks",
publisher = "Publ by IEEE",
pages = "1742--1747",
booktitle = "1993 IEEE International Conference on Neural Networks",

}

Syu, M-J & Tsao, GT 1993, Neural network approach to identify batch cell growth. in 1993 IEEE International Conference on Neural Networks. 1993 IEEE International Conference on Neural Networks, Publ by IEEE, pp. 1742-1747, 1993 IEEE International Conference on Neural Networks, San Francisco, California, USA, 93-03-28.

Neural network approach to identify batch cell growth. / Syu, Mei-Jywan; Tsao, George T.

1993 IEEE International Conference on Neural Networks. Publ by IEEE, 1993. p. 1742-1747 (1993 IEEE International Conference on Neural Networks).

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

TY - GEN

T1 - Neural network approach to identify batch cell growth

AU - Syu, Mei-Jywan

AU - Tsao, George T.

PY - 1993/1/1

Y1 - 1993/1/1

N2 - A saturation-type transfer function with a backpropagation neural network (BPNN) was proposed for solving the modeling problem of batch cell growth system. Batch chemical processes are usually influenced by their initial conditions. For batch cell cultures, the initial state strongly governs the growth pattern during the timecourse. In modeling a chemical system, we are always interested in how to model the outcome of the system related to some affecting factors. In a batch system, some of the initial conditions are certainly important affecting factors. Trying to model the cell growth with information concerning only the initial conditions is not yet possible from a kinetic approach. The difficulty comes from numerical analysis and insufficient knowledge regarding certain growth parameters as they vary with time. Accordingly, neural network methodology with the concept developed earlier was proposed to solve this problem. The feasibility and capability of the neural network to model the pattern of batch cell growth by providing initial conditions only is tested in this study. A 2-3-8 BPNN with initial glucose and cell concentrations as the two inputs, cell densities measured at eight each hours as the eight outputs was thus constructed. The simulation and prediction results of this BPNN are presented to demonstrate the performance and applicability of this newly discovered transfer function. Sensitivity analysis of the initial factors from this neural network model (NNM) is also discussed. The optimization of the initial conditions for this system is also performed.

AB - A saturation-type transfer function with a backpropagation neural network (BPNN) was proposed for solving the modeling problem of batch cell growth system. Batch chemical processes are usually influenced by their initial conditions. For batch cell cultures, the initial state strongly governs the growth pattern during the timecourse. In modeling a chemical system, we are always interested in how to model the outcome of the system related to some affecting factors. In a batch system, some of the initial conditions are certainly important affecting factors. Trying to model the cell growth with information concerning only the initial conditions is not yet possible from a kinetic approach. The difficulty comes from numerical analysis and insufficient knowledge regarding certain growth parameters as they vary with time. Accordingly, neural network methodology with the concept developed earlier was proposed to solve this problem. The feasibility and capability of the neural network to model the pattern of batch cell growth by providing initial conditions only is tested in this study. A 2-3-8 BPNN with initial glucose and cell concentrations as the two inputs, cell densities measured at eight each hours as the eight outputs was thus constructed. The simulation and prediction results of this BPNN are presented to demonstrate the performance and applicability of this newly discovered transfer function. Sensitivity analysis of the initial factors from this neural network model (NNM) is also discussed. The optimization of the initial conditions for this system is also performed.

UR - http://www.scopus.com/inward/record.url?scp=84943266643&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943266643&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84943266643

SN - 0780312007

T3 - 1993 IEEE International Conference on Neural Networks

SP - 1742

EP - 1747

BT - 1993 IEEE International Conference on Neural Networks

PB - Publ by IEEE

ER -

Syu M-J, Tsao GT. Neural network approach to identify batch cell growth. In 1993 IEEE International Conference on Neural Networks. Publ by IEEE. 1993. p. 1742-1747. (1993 IEEE International Conference on Neural Networks).