Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm
The problem of constructing an adequate and parsimonious neural network topology for modeling non-linear dynamic system is studied and investigated. Neural networks have been shown to perform function approximation and represent dynamic systems. The network structures are usually guessed or selected...
Main Authors: | , , , , |
---|---|
Format: | Article |
Published: |
Springer
2012
|
Subjects: |
_version_ | 1796857206857007104 |
---|---|
author | Loghmanian, Sayed Mohammad Reza Jamaluddin, Hishamuddin Ahmad, Robiah Yusof, Rubiyah Khalid, Marzuki |
author_facet | Loghmanian, Sayed Mohammad Reza Jamaluddin, Hishamuddin Ahmad, Robiah Yusof, Rubiyah Khalid, Marzuki |
author_sort | Loghmanian, Sayed Mohammad Reza |
collection | ePrints |
description | The problem of constructing an adequate and parsimonious neural network topology for modeling non-linear dynamic system is studied and investigated. Neural networks have been shown to perform function approximation and represent dynamic systems. The network structures are usually guessed or selected in accordance with the designer's prior knowledge. However, the multiplicity of the model parameters makes it troublesome to get an optimum structure. In this paper, an alternative algorithm based on a multi-objective optimization algorithm is proposed. The developed neural network model should fulfil two criteria or objectives namely good predictive accuracy and minimum model structure. The result shows that the proposed algorithm is able to identify simulated examples correctly, and identifies the adequate model for real process data based on a set of solutions called the Pareto optimal set, from which the best network can be selected. |
first_indexed | 2024-03-05T18:54:18Z |
format | Article |
id | utm.eprints-33553 |
institution | Universiti Teknologi Malaysia - ePrints |
last_indexed | 2024-03-05T18:54:18Z |
publishDate | 2012 |
publisher | Springer |
record_format | dspace |
spelling | utm.eprints-335532018-11-30T06:37:35Z http://eprints.utm.my/33553/ Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm Loghmanian, Sayed Mohammad Reza Jamaluddin, Hishamuddin Ahmad, Robiah Yusof, Rubiyah Khalid, Marzuki TJ Mechanical engineering and machinery The problem of constructing an adequate and parsimonious neural network topology for modeling non-linear dynamic system is studied and investigated. Neural networks have been shown to perform function approximation and represent dynamic systems. The network structures are usually guessed or selected in accordance with the designer's prior knowledge. However, the multiplicity of the model parameters makes it troublesome to get an optimum structure. In this paper, an alternative algorithm based on a multi-objective optimization algorithm is proposed. The developed neural network model should fulfil two criteria or objectives namely good predictive accuracy and minimum model structure. The result shows that the proposed algorithm is able to identify simulated examples correctly, and identifies the adequate model for real process data based on a set of solutions called the Pareto optimal set, from which the best network can be selected. Springer 2012-09 Article PeerReviewed Loghmanian, Sayed Mohammad Reza and Jamaluddin, Hishamuddin and Ahmad, Robiah and Yusof, Rubiyah and Khalid, Marzuki (2012) Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm. Neural Computing & Applications, 21 (6). pp. 1281-1295. ISSN 0941-0643 (Print); 1433-3058 (Electronic) http://dx.doi.org/10.1007/s00521-011-0560-3 DOI:10.1007/s00521-011-0560-3 |
spellingShingle | TJ Mechanical engineering and machinery Loghmanian, Sayed Mohammad Reza Jamaluddin, Hishamuddin Ahmad, Robiah Yusof, Rubiyah Khalid, Marzuki Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm |
title | Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm |
title_full | Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm |
title_fullStr | Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm |
title_full_unstemmed | Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm |
title_short | Structure optimization of neural network for dynamic system modeling using multi-objective genetic algorithm |
title_sort | structure optimization of neural network for dynamic system modeling using multi objective genetic algorithm |
topic | TJ Mechanical engineering and machinery |
work_keys_str_mv | AT loghmaniansayedmohammadreza structureoptimizationofneuralnetworkfordynamicsystemmodelingusingmultiobjectivegeneticalgorithm AT jamaluddinhishamuddin structureoptimizationofneuralnetworkfordynamicsystemmodelingusingmultiobjectivegeneticalgorithm AT ahmadrobiah structureoptimizationofneuralnetworkfordynamicsystemmodelingusingmultiobjectivegeneticalgorithm AT yusofrubiyah structureoptimizationofneuralnetworkfordynamicsystemmodelingusingmultiobjectivegeneticalgorithm AT khalidmarzuki structureoptimizationofneuralnetworkfordynamicsystemmodelingusingmultiobjectivegeneticalgorithm |