Wavelet shrinkage data processing for neural networks in bioprocess modeling

BH Chen, John Woodley

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

The modeling of biological systems has now become an essential prerequisite for effective bioprocess design, optimization and analysis. The difficulties present in using conventional techniques to model such a complex system make the application of artificial neural networks (ANNs) to these problems particularly attractive because of their capability for nonlinear mapping and lack of necessity for detailed mechanistic knowledge. However, building a reliable ANN model requires sufficient training data, which may be difficult when data are collected from litre-scale experiments. In this work, a bioconversion (with only limited experimental data) was firstly modeled by a radial basis function (RBF) neural network. Although the model provided a very low variance between experiment and simulation, it tended to result in oscillatory behaviour, which clearly does not reflect the accurate profile of the reaction. In order to overcome this drawback, wavelet shrinkage with biorthogonal filters was used to generate a reconstructed function using the RBF model as a base. The synthesis of N-acetyl-D-neuraminic acid by the enzymatic condensation of pyruvate with N-acetyl-D-mannosamine was used as a case study to show the effectiveness of the approach. The effects of alternative filters and border distortion are also discussed. (C) 2002 Elsevier Science Ltd. All rights reserved.
Original languageEnglish
JournalComputers & Chemical Engineering
Volume26
Issue number11
Pages (from-to)1611-1620
ISSN0098-1354
DOIs
Publication statusPublished - 2002
Externally publishedYes

Cite this

@article{069c3a42a6be4c4a81a5311bfbff48c6,
title = "Wavelet shrinkage data processing for neural networks in bioprocess modeling",
abstract = "The modeling of biological systems has now become an essential prerequisite for effective bioprocess design, optimization and analysis. The difficulties present in using conventional techniques to model such a complex system make the application of artificial neural networks (ANNs) to these problems particularly attractive because of their capability for nonlinear mapping and lack of necessity for detailed mechanistic knowledge. However, building a reliable ANN model requires sufficient training data, which may be difficult when data are collected from litre-scale experiments. In this work, a bioconversion (with only limited experimental data) was firstly modeled by a radial basis function (RBF) neural network. Although the model provided a very low variance between experiment and simulation, it tended to result in oscillatory behaviour, which clearly does not reflect the accurate profile of the reaction. In order to overcome this drawback, wavelet shrinkage with biorthogonal filters was used to generate a reconstructed function using the RBF model as a base. The synthesis of N-acetyl-D-neuraminic acid by the enzymatic condensation of pyruvate with N-acetyl-D-mannosamine was used as a case study to show the effectiveness of the approach. The effects of alternative filters and border distortion are also discussed. (C) 2002 Elsevier Science Ltd. All rights reserved.",
author = "BH Chen and John Woodley",
year = "2002",
doi = "10.1016/S0098-1354(02)00146-1",
language = "English",
volume = "26",
pages = "1611--1620",
journal = "Computers & Chemical Engineering",
issn = "0098-1354",
publisher = "Elsevier",
number = "11",

}

Wavelet shrinkage data processing for neural networks in bioprocess modeling. / Chen, BH; Woodley, John.

In: Computers & Chemical Engineering, Vol. 26, No. 11, 2002, p. 1611-1620.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Wavelet shrinkage data processing for neural networks in bioprocess modeling

AU - Chen, BH

AU - Woodley, John

PY - 2002

Y1 - 2002

N2 - The modeling of biological systems has now become an essential prerequisite for effective bioprocess design, optimization and analysis. The difficulties present in using conventional techniques to model such a complex system make the application of artificial neural networks (ANNs) to these problems particularly attractive because of their capability for nonlinear mapping and lack of necessity for detailed mechanistic knowledge. However, building a reliable ANN model requires sufficient training data, which may be difficult when data are collected from litre-scale experiments. In this work, a bioconversion (with only limited experimental data) was firstly modeled by a radial basis function (RBF) neural network. Although the model provided a very low variance between experiment and simulation, it tended to result in oscillatory behaviour, which clearly does not reflect the accurate profile of the reaction. In order to overcome this drawback, wavelet shrinkage with biorthogonal filters was used to generate a reconstructed function using the RBF model as a base. The synthesis of N-acetyl-D-neuraminic acid by the enzymatic condensation of pyruvate with N-acetyl-D-mannosamine was used as a case study to show the effectiveness of the approach. The effects of alternative filters and border distortion are also discussed. (C) 2002 Elsevier Science Ltd. All rights reserved.

AB - The modeling of biological systems has now become an essential prerequisite for effective bioprocess design, optimization and analysis. The difficulties present in using conventional techniques to model such a complex system make the application of artificial neural networks (ANNs) to these problems particularly attractive because of their capability for nonlinear mapping and lack of necessity for detailed mechanistic knowledge. However, building a reliable ANN model requires sufficient training data, which may be difficult when data are collected from litre-scale experiments. In this work, a bioconversion (with only limited experimental data) was firstly modeled by a radial basis function (RBF) neural network. Although the model provided a very low variance between experiment and simulation, it tended to result in oscillatory behaviour, which clearly does not reflect the accurate profile of the reaction. In order to overcome this drawback, wavelet shrinkage with biorthogonal filters was used to generate a reconstructed function using the RBF model as a base. The synthesis of N-acetyl-D-neuraminic acid by the enzymatic condensation of pyruvate with N-acetyl-D-mannosamine was used as a case study to show the effectiveness of the approach. The effects of alternative filters and border distortion are also discussed. (C) 2002 Elsevier Science Ltd. All rights reserved.

U2 - 10.1016/S0098-1354(02)00146-1

DO - 10.1016/S0098-1354(02)00146-1

M3 - Journal article

VL - 26

SP - 1611

EP - 1620

JO - Computers & Chemical Engineering

JF - Computers & Chemical Engineering

SN - 0098-1354

IS - 11

ER -