Neuroidentification of a biotechnological process: Issues and application

Fermentations are highly non-linear time-varying and inaccurately modelled processes. Our work towards the generation of a neural network model of an industrial secondary metabolic fermentation process is presented in this paper. The construction methodology and the various issues arising during the...

Full description

Saved in:
Bibliographic Details
Published in:Neurocomputing (Amsterdam) 1995-09, Vol.9 (1), p.63-79
Main Authors: Tsaptsinos, D., Tang, R., Leigh, J.R.
Format: Article
Language:English
Subjects:
Artifical neural networks
Fermentation process
Multi-layer perceptrons
Pruning
System identification
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Fermentations are highly non-linear time-varying and inaccurately modelled processes. Our work towards the generation of a neural network model of an industrial secondary metabolic fermentation process is presented in this paper. The construction methodology and the various issues arising during the design, training and testing stages are also discussed. Such issues addressed in this paper include the treatment of dynamics and the determination of the number of hidden nodes. The usefulness of the neural network model with regard to the on-line estimation of fundamental measurements is demonstrated and its part in a control system is highlighted.
ISSN:0925-2312
1872-8286
DOI:10.1016/0925-2312(94)00053-U