System identification using neural networks /
This thesis studies the modelling of nonlinear dynaical systems using neural networks employing the system identification methodology. The most commonlyused learning algorithm for training neural network is the backpropagation algorithm. A computer program was modified and the backpropagation algoti...
Main Author: | |
---|---|
Format: | |
Published: |
Sekudai : UTM,
1993
|
Subjects: |
_version_ | 1826373302279471104 |
---|---|
author | 383392 They, Hong San |
author_facet | 383392 They, Hong San |
author_sort | 383392 They, Hong San |
collection | OCEAN |
description | This thesis studies the modelling of nonlinear dynaical systems using neural networks employing the system identification methodology. The most commonlyused learning algorithm for training neural network is the backpropagation algorithm. A computer program was modified and the backpropagation algotithm was used to train the multilayered perception networks. Some nonlinear dynamical examples will be trained with the backpropagation algorithm. Effect of varying with learning rates and thresholds, network complexity and some new metrics of performance were introduced. |
first_indexed | 2024-03-04T17:00:57Z |
format | |
id | KOHA-OAI-TEST:90783 |
institution | Universiti Teknologi Malaysia - OCEAN |
last_indexed | 2024-03-04T17:00:57Z |
publishDate | 1993 |
publisher | Sekudai : UTM, |
record_format | dspace |
spelling | KOHA-OAI-TEST:907832020-12-19T17:01:02ZSystem identification using neural networks / 383392 They, Hong San Sekudai : UTM,1993This thesis studies the modelling of nonlinear dynaical systems using neural networks employing the system identification methodology. The most commonlyused learning algorithm for training neural network is the backpropagation algorithm. A computer program was modified and the backpropagation algotithm was used to train the multilayered perception networks. Some nonlinear dynamical examples will be trained with the backpropagation algorithm. Effect of varying with learning rates and thresholds, network complexity and some new metrics of performance were introduced.Project paper (Bachelor of Mechanical Engineering) - Universiti Teknologi Malaysia, 1993This thesis studies the modelling of nonlinear dynaical systems using neural networks employing the system identification methodology. The most commonlyused learning algorithm for training neural network is the backpropagation algorithm. A computer program was modified and the backpropagation algotithm was used to train the multilayered perception networks. Some nonlinear dynamical examples will be trained with the backpropagation algorithm. Effect of varying with learning rates and thresholds, network complexity and some new metrics of performance were introduced.20PRZSLSystem identificationSystem analysisArtificial neural networks |
spellingShingle | System identification System analysis Artificial neural networks 383392 They, Hong San System identification using neural networks / |
title | System identification using neural networks / |
title_full | System identification using neural networks / |
title_fullStr | System identification using neural networks / |
title_full_unstemmed | System identification using neural networks / |
title_short | System identification using neural networks / |
title_sort | system identification using neural networks |
topic | System identification System analysis Artificial neural networks |
work_keys_str_mv | AT 383392theyhongsan systemidentificationusingneuralnetworks |