Detecting faults in a hydraulic system using neural network and observer approaches

D.N. Shields, S. Du, E. Gaura

Research output: Contribution to journalArticle

Abstract

A robust fault detection observer (RFDO), which includes a robust residual, is designed for a general nonlinear system with polynomial-type nonlinearities. Two main theoretical results are given for the existence, stability and the detectability of the RFDO. The approach is applied to a three-tank hydraulic system where two fault scenarios are considered. Three different degree observers, with corresponding residuals, are assessed. This observer approach is then compared to the performance of a neural network approach under the same fault conditions.
Original languageUndefined
Pages (from-to)85-106
Number of pages22
JournalSystems Science
Volume28
Issue number1
Publication statusPublished - 2002

Keywords

  • Asymptotic stability
  • Control system synthesis
  • Differential equations
  • Failure analysis
  • Hydraulic equipment
  • Neural networks
  • Nonlinear control systems
  • Polynomials
  • System stability
  • Polynomial-type nonlinearities
  • Robust fault detection observer
  • Hydraulic system
  • Robustness (control systems)

Cite this

Detecting faults in a hydraulic system using neural network and observer approaches. / Shields, D.N.; Du, S.; Gaura, E.

In: Systems Science, Vol. 28, No. 1, 2002, p. 85-106.

Research output: Contribution to journalArticle

@article{8a4ccd93ac0a484f8ee3a0201040850c,
title = "Detecting faults in a hydraulic system using neural network and observer approaches",
abstract = "A robust fault detection observer (RFDO), which includes a robust residual, is designed for a general nonlinear system with polynomial-type nonlinearities. Two main theoretical results are given for the existence, stability and the detectability of the RFDO. The approach is applied to a three-tank hydraulic system where two fault scenarios are considered. Three different degree observers, with corresponding residuals, are assessed. This observer approach is then compared to the performance of a neural network approach under the same fault conditions.",
keywords = "Asymptotic stability, Control system synthesis, Differential equations, Failure analysis, Hydraulic equipment, Neural networks, Nonlinear control systems, Polynomials, System stability, Polynomial-type nonlinearities, Robust fault detection observer, Hydraulic system, Robustness (control systems)",
author = "D.N. Shields and S. Du and E. Gaura",
year = "2002",
language = "Undefined",
volume = "28",
pages = "85--106",
journal = "Systems Science",
issn = "0137-1223",
publisher = "Tech Univ Wroclaw, Warsaw, Poland",
number = "1",

}

TY - JOUR

T1 - Detecting faults in a hydraulic system using neural network and observer approaches

AU - Shields, D.N.

AU - Du, S.

AU - Gaura, E.

PY - 2002

Y1 - 2002

N2 - A robust fault detection observer (RFDO), which includes a robust residual, is designed for a general nonlinear system with polynomial-type nonlinearities. Two main theoretical results are given for the existence, stability and the detectability of the RFDO. The approach is applied to a three-tank hydraulic system where two fault scenarios are considered. Three different degree observers, with corresponding residuals, are assessed. This observer approach is then compared to the performance of a neural network approach under the same fault conditions.

AB - A robust fault detection observer (RFDO), which includes a robust residual, is designed for a general nonlinear system with polynomial-type nonlinearities. Two main theoretical results are given for the existence, stability and the detectability of the RFDO. The approach is applied to a three-tank hydraulic system where two fault scenarios are considered. Three different degree observers, with corresponding residuals, are assessed. This observer approach is then compared to the performance of a neural network approach under the same fault conditions.

KW - Asymptotic stability

KW - Control system synthesis

KW - Differential equations

KW - Failure analysis

KW - Hydraulic equipment

KW - Neural networks

KW - Nonlinear control systems

KW - Polynomials

KW - System stability

KW - Polynomial-type nonlinearities

KW - Robust fault detection observer

KW - Hydraulic system

KW - Robustness (control systems)

M3 - Article

VL - 28

SP - 85

EP - 106

JO - Systems Science

JF - Systems Science

SN - 0137-1223

IS - 1

ER -