Layer-recurrent network in identifying a nonlinear system

Farah Hani Nordin, Farrukh Hafiz Nagi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Layer-Recurrent Network (LRN) is a dynamic neural network and is seen as a promising black box model in identifying a nonlinear system injected with nonlinear input signal. In this paper, LRN will be used to identify a nonlinear, state space 3-axis satellite model. Open loop identification is applied and methodology on nonlinear system identification is presented where the best pair of input and output data is first measured. Using the simulated data, six LRN models are used to identify the satellite dynamics. It is shown that only 200 epochs are needed to train a network to converge to a reasonable mean squared value (mse). LRN output is then compared with the state space model where it shows that LRN model is capable to produce similar results as the state space satellite model without knowing the system's state and prior knowledge of the system.

Original languageEnglish
Title of host publication2008 International Conference on Control, Automation and Systems, ICCAS 2008
Pages387-391
Number of pages5
DOIs
Publication statusPublished - 01 Dec 2008
Event2008 International Conference on Control, Automation and Systems, ICCAS 2008 - Seoul, Korea, Republic of
Duration: 14 Oct 200817 Oct 2008

Publication series

Name2008 International Conference on Control, Automation and Systems, ICCAS 2008

Other

Other2008 International Conference on Control, Automation and Systems, ICCAS 2008
CountryKorea, Republic of
CitySeoul
Period14/10/0817/10/08

Fingerprint

Nonlinear systems
Satellites
Identification (control systems)
Neural networks

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering

Cite this

Nordin, F. H., & Nagi, F. H. (2008). Layer-recurrent network in identifying a nonlinear system. In 2008 International Conference on Control, Automation and Systems, ICCAS 2008 (pp. 387-391). [4694674] (2008 International Conference on Control, Automation and Systems, ICCAS 2008). https://doi.org/10.1109/ICCAS.2008.4694674
Nordin, Farah Hani ; Nagi, Farrukh Hafiz. / Layer-recurrent network in identifying a nonlinear system. 2008 International Conference on Control, Automation and Systems, ICCAS 2008. 2008. pp. 387-391 (2008 International Conference on Control, Automation and Systems, ICCAS 2008).
@inproceedings{074ecedb7deb4c5f90281ccdce7ee5ef,
title = "Layer-recurrent network in identifying a nonlinear system",
abstract = "Layer-Recurrent Network (LRN) is a dynamic neural network and is seen as a promising black box model in identifying a nonlinear system injected with nonlinear input signal. In this paper, LRN will be used to identify a nonlinear, state space 3-axis satellite model. Open loop identification is applied and methodology on nonlinear system identification is presented where the best pair of input and output data is first measured. Using the simulated data, six LRN models are used to identify the satellite dynamics. It is shown that only 200 epochs are needed to train a network to converge to a reasonable mean squared value (mse). LRN output is then compared with the state space model where it shows that LRN model is capable to produce similar results as the state space satellite model without knowing the system's state and prior knowledge of the system.",
author = "Nordin, {Farah Hani} and Nagi, {Farrukh Hafiz}",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICCAS.2008.4694674",
language = "English",
isbn = "9788995003893",
series = "2008 International Conference on Control, Automation and Systems, ICCAS 2008",
pages = "387--391",
booktitle = "2008 International Conference on Control, Automation and Systems, ICCAS 2008",

}

Nordin, FH & Nagi, FH 2008, Layer-recurrent network in identifying a nonlinear system. in 2008 International Conference on Control, Automation and Systems, ICCAS 2008., 4694674, 2008 International Conference on Control, Automation and Systems, ICCAS 2008, pp. 387-391, 2008 International Conference on Control, Automation and Systems, ICCAS 2008, Seoul, Korea, Republic of, 14/10/08. https://doi.org/10.1109/ICCAS.2008.4694674

Layer-recurrent network in identifying a nonlinear system. / Nordin, Farah Hani; Nagi, Farrukh Hafiz.

2008 International Conference on Control, Automation and Systems, ICCAS 2008. 2008. p. 387-391 4694674 (2008 International Conference on Control, Automation and Systems, ICCAS 2008).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Layer-recurrent network in identifying a nonlinear system

AU - Nordin, Farah Hani

AU - Nagi, Farrukh Hafiz

PY - 2008/12/1

Y1 - 2008/12/1

N2 - Layer-Recurrent Network (LRN) is a dynamic neural network and is seen as a promising black box model in identifying a nonlinear system injected with nonlinear input signal. In this paper, LRN will be used to identify a nonlinear, state space 3-axis satellite model. Open loop identification is applied and methodology on nonlinear system identification is presented where the best pair of input and output data is first measured. Using the simulated data, six LRN models are used to identify the satellite dynamics. It is shown that only 200 epochs are needed to train a network to converge to a reasonable mean squared value (mse). LRN output is then compared with the state space model where it shows that LRN model is capable to produce similar results as the state space satellite model without knowing the system's state and prior knowledge of the system.

AB - Layer-Recurrent Network (LRN) is a dynamic neural network and is seen as a promising black box model in identifying a nonlinear system injected with nonlinear input signal. In this paper, LRN will be used to identify a nonlinear, state space 3-axis satellite model. Open loop identification is applied and methodology on nonlinear system identification is presented where the best pair of input and output data is first measured. Using the simulated data, six LRN models are used to identify the satellite dynamics. It is shown that only 200 epochs are needed to train a network to converge to a reasonable mean squared value (mse). LRN output is then compared with the state space model where it shows that LRN model is capable to produce similar results as the state space satellite model without knowing the system's state and prior knowledge of the system.

UR - http://www.scopus.com/inward/record.url?scp=58149087554&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58149087554&partnerID=8YFLogxK

U2 - 10.1109/ICCAS.2008.4694674

DO - 10.1109/ICCAS.2008.4694674

M3 - Conference contribution

SN - 9788995003893

T3 - 2008 International Conference on Control, Automation and Systems, ICCAS 2008

SP - 387

EP - 391

BT - 2008 International Conference on Control, Automation and Systems, ICCAS 2008

ER -

Nordin FH, Nagi FH. Layer-recurrent network in identifying a nonlinear system. In 2008 International Conference on Control, Automation and Systems, ICCAS 2008. 2008. p. 387-391. 4694674. (2008 International Conference on Control, Automation and Systems, ICCAS 2008). https://doi.org/10.1109/ICCAS.2008.4694674