Dynamic modelling of twin rotor multi system in horizontal motion

This paper investigates the parametric linear approach and utilisation of neural networks (NNs) for modelling of a twin rotor multi system (TRMS) in horizontal motion. Parametric modelling using Auto Regressive Modelling (ARX) model using Recursive Least Squares (RLS) algorithm. On the other hand, n...

Full description

Bibliographic Details
Main Authors: Mat Darus, Intan Zaurah, Lokaman, Zainul Aman
Format: Article
Language:English
English
Published: Faculty of Mechanical Engineering 2010
Subjects:
Online Access:http://eprints.utm.my/36669/1/IntanZaurahMat2010_DynamicModellingofTwinRotorMulti.pdf
http://eprints.utm.my/36669/2/201026211.pdf
_version_ 1796857560325685248
author Mat Darus, Intan Zaurah
Lokaman, Zainul Aman
author_facet Mat Darus, Intan Zaurah
Lokaman, Zainul Aman
author_sort Mat Darus, Intan Zaurah
collection ePrints
description This paper investigates the parametric linear approach and utilisation of neural networks (NNs) for modelling of a twin rotor multi system (TRMS) in horizontal motion. Parametric modelling using Auto Regressive Modelling (ARX) model using Recursive Least Squares (RLS) algorithm. On the other hand, non-parametric modelling, makes use of Multi Layer Perceptron-Neural Network (MLP-NN) technique. All of these techniques will be used to characterize the behaviour of TRMS. Comparative assessment between these two techniques was conducted and the MLP-NN shows better results compared to RLS for modelling the TRMS. Mean Square Error (MSE), One Step Ahead (OSA) prediction and Correlation Tests were used for verification and validation of both models. Both models are found to be within the 95% confident level.
first_indexed 2024-03-05T18:59:36Z
format Article
id utm.eprints-36669
institution Universiti Teknologi Malaysia - ePrints
language English
English
last_indexed 2024-03-05T18:59:36Z
publishDate 2010
publisher Faculty of Mechanical Engineering
record_format dspace
spelling utm.eprints-366692014-03-12T08:10:15Z http://eprints.utm.my/36669/ Dynamic modelling of twin rotor multi system in horizontal motion Mat Darus, Intan Zaurah Lokaman, Zainul Aman TJ Mechanical engineering and machinery This paper investigates the parametric linear approach and utilisation of neural networks (NNs) for modelling of a twin rotor multi system (TRMS) in horizontal motion. Parametric modelling using Auto Regressive Modelling (ARX) model using Recursive Least Squares (RLS) algorithm. On the other hand, non-parametric modelling, makes use of Multi Layer Perceptron-Neural Network (MLP-NN) technique. All of these techniques will be used to characterize the behaviour of TRMS. Comparative assessment between these two techniques was conducted and the MLP-NN shows better results compared to RLS for modelling the TRMS. Mean Square Error (MSE), One Step Ahead (OSA) prediction and Correlation Tests were used for verification and validation of both models. Both models are found to be within the 95% confident level. Faculty of Mechanical Engineering 2010-12 Article PeerReviewed application/pdf en http://eprints.utm.my/36669/1/IntanZaurahMat2010_DynamicModellingofTwinRotorMulti.pdf text/html en http://eprints.utm.my/36669/2/201026211.pdf Mat Darus, Intan Zaurah and Lokaman, Zainul Aman (2010) Dynamic modelling of twin rotor multi system in horizontal motion. Jurnal Mekanikal (31). pp. 17-29. ISSN 0127-3396
spellingShingle TJ Mechanical engineering and machinery
Mat Darus, Intan Zaurah
Lokaman, Zainul Aman
Dynamic modelling of twin rotor multi system in horizontal motion
title Dynamic modelling of twin rotor multi system in horizontal motion
title_full Dynamic modelling of twin rotor multi system in horizontal motion
title_fullStr Dynamic modelling of twin rotor multi system in horizontal motion
title_full_unstemmed Dynamic modelling of twin rotor multi system in horizontal motion
title_short Dynamic modelling of twin rotor multi system in horizontal motion
title_sort dynamic modelling of twin rotor multi system in horizontal motion
topic TJ Mechanical engineering and machinery
url http://eprints.utm.my/36669/1/IntanZaurahMat2010_DynamicModellingofTwinRotorMulti.pdf
http://eprints.utm.my/36669/2/201026211.pdf
work_keys_str_mv AT matdarusintanzaurah dynamicmodellingoftwinrotormultisysteminhorizontalmotion
AT lokamanzainulaman dynamicmodellingoftwinrotormultisysteminhorizontalmotion