Fault diagnosis based on grey-box neural network identification model

Zhaohui Cen, Jiaolong Wei, Rui Jiang

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

3 Citations (Scopus)

Abstract

This paper presents a fault diagnosis (FD) scheme for a class of nonlinear dynamic systems using a novel Grey-Box Neural Network Model (GBNNM). In this GBNNM, a composite structure, including MLP (multi-layer perception) NN (Neural Network) and integer term, is proposed to approximate both nonlinearity and dynamics of object system. Its approximation ability is then proved theoretically. And a self-defined exciting strategy is introduced into NN training to improve NN's generalization ability. Unlike previous NN model based fault diagnosis methods, a quantitative residual, which is obtained from system output and its GBNNM model output, can accurately indicates inconsistency caused by fault, so the improved residual is not essential for our scheme. The proposed FD scheme is applied in a high-fidelity Reaction Wheel (RW) in Satellite Attitude Control System (SACS) in our case study. The results of the case study demonstrate the effectiveness and superiority of our FD scheme.

Original languageEnglish
Title of host publicationICCAS 2010 - International Conference on Control, Automation and Systems
Pages249-254
Number of pages6
Publication statusPublished - 1 Dec 2010
Externally publishedYes
EventInternational Conference on Control, Automation and Systems, ICCAS 2010 - Gyeonggi-do, Korea, Republic of
Duration: 27 Oct 201030 Oct 2010

Other

OtherInternational Conference on Control, Automation and Systems, ICCAS 2010
CountryKorea, Republic of
CityGyeonggi-do
Period27/10/1030/10/10

Fingerprint

Failure analysis
Identification (control systems)
Neural networks
Attitude control
Composite structures
Wheels
Dynamical systems
Satellites
Control systems

Keywords

  • Grey-box neural-network model (GBNNM)
  • Model identification and fault diagnosis
  • Nonlinear dynamic systems
  • Reaction wheel

ASJC Scopus subject areas

  • Artificial Intelligence
  • Control and Systems Engineering

Cite this

Cen, Z., Wei, J., & Jiang, R. (2010). Fault diagnosis based on grey-box neural network identification model. In ICCAS 2010 - International Conference on Control, Automation and Systems (pp. 249-254). [5670320]

Fault diagnosis based on grey-box neural network identification model. / Cen, Zhaohui; Wei, Jiaolong; Jiang, Rui.

ICCAS 2010 - International Conference on Control, Automation and Systems. 2010. p. 249-254 5670320.

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

Cen, Z, Wei, J & Jiang, R 2010, Fault diagnosis based on grey-box neural network identification model. in ICCAS 2010 - International Conference on Control, Automation and Systems., 5670320, pp. 249-254, International Conference on Control, Automation and Systems, ICCAS 2010, Gyeonggi-do, Korea, Republic of, 27/10/10.
Cen Z, Wei J, Jiang R. Fault diagnosis based on grey-box neural network identification model. In ICCAS 2010 - International Conference on Control, Automation and Systems. 2010. p. 249-254. 5670320
Cen, Zhaohui ; Wei, Jiaolong ; Jiang, Rui. / Fault diagnosis based on grey-box neural network identification model. ICCAS 2010 - International Conference on Control, Automation and Systems. 2010. pp. 249-254
@inproceedings{38b463455db246c385600c356d285164,
title = "Fault diagnosis based on grey-box neural network identification model",
abstract = "This paper presents a fault diagnosis (FD) scheme for a class of nonlinear dynamic systems using a novel Grey-Box Neural Network Model (GBNNM). In this GBNNM, a composite structure, including MLP (multi-layer perception) NN (Neural Network) and integer term, is proposed to approximate both nonlinearity and dynamics of object system. Its approximation ability is then proved theoretically. And a self-defined exciting strategy is introduced into NN training to improve NN's generalization ability. Unlike previous NN model based fault diagnosis methods, a quantitative residual, which is obtained from system output and its GBNNM model output, can accurately indicates inconsistency caused by fault, so the improved residual is not essential for our scheme. The proposed FD scheme is applied in a high-fidelity Reaction Wheel (RW) in Satellite Attitude Control System (SACS) in our case study. The results of the case study demonstrate the effectiveness and superiority of our FD scheme.",
keywords = "Grey-box neural-network model (GBNNM), Model identification and fault diagnosis, Nonlinear dynamic systems, Reaction wheel",
author = "Zhaohui Cen and Jiaolong Wei and Rui Jiang",
year = "2010",
month = "12",
day = "1",
language = "English",
isbn = "9781424474530",
pages = "249--254",
booktitle = "ICCAS 2010 - International Conference on Control, Automation and Systems",

}

TY - GEN

T1 - Fault diagnosis based on grey-box neural network identification model

AU - Cen, Zhaohui

AU - Wei, Jiaolong

AU - Jiang, Rui

PY - 2010/12/1

Y1 - 2010/12/1

N2 - This paper presents a fault diagnosis (FD) scheme for a class of nonlinear dynamic systems using a novel Grey-Box Neural Network Model (GBNNM). In this GBNNM, a composite structure, including MLP (multi-layer perception) NN (Neural Network) and integer term, is proposed to approximate both nonlinearity and dynamics of object system. Its approximation ability is then proved theoretically. And a self-defined exciting strategy is introduced into NN training to improve NN's generalization ability. Unlike previous NN model based fault diagnosis methods, a quantitative residual, which is obtained from system output and its GBNNM model output, can accurately indicates inconsistency caused by fault, so the improved residual is not essential for our scheme. The proposed FD scheme is applied in a high-fidelity Reaction Wheel (RW) in Satellite Attitude Control System (SACS) in our case study. The results of the case study demonstrate the effectiveness and superiority of our FD scheme.

AB - This paper presents a fault diagnosis (FD) scheme for a class of nonlinear dynamic systems using a novel Grey-Box Neural Network Model (GBNNM). In this GBNNM, a composite structure, including MLP (multi-layer perception) NN (Neural Network) and integer term, is proposed to approximate both nonlinearity and dynamics of object system. Its approximation ability is then proved theoretically. And a self-defined exciting strategy is introduced into NN training to improve NN's generalization ability. Unlike previous NN model based fault diagnosis methods, a quantitative residual, which is obtained from system output and its GBNNM model output, can accurately indicates inconsistency caused by fault, so the improved residual is not essential for our scheme. The proposed FD scheme is applied in a high-fidelity Reaction Wheel (RW) in Satellite Attitude Control System (SACS) in our case study. The results of the case study demonstrate the effectiveness and superiority of our FD scheme.

KW - Grey-box neural-network model (GBNNM)

KW - Model identification and fault diagnosis

KW - Nonlinear dynamic systems

KW - Reaction wheel

UR - http://www.scopus.com/inward/record.url?scp=78751539790&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78751539790&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:78751539790

SN - 9781424474530

SP - 249

EP - 254

BT - ICCAS 2010 - International Conference on Control, Automation and Systems

ER -