Active Fault Diagnosis for Hybrid Systems Based on Sensitivity Analysis and EKF

Publication: Research - peer-review › Conference article in Journal

Standard

Active Fault Diagnosis for Hybrid Systems Based on Sensitivity Analysis and EKF. / Gholami, Mehdi ; Schiøler, Henrik ; Bak, Thomas.

In: American Control Conference. Proceedings, 02.07.2011, p. 244-249.

Publication: Research - peer-review › Conference article in Journal

Publication: Research - peer-review › Conference article in Journal

Bibtex

@article{cf00f536799948a5b69c9a275ce321c9,

title = "Active Fault Diagnosis for Hybrid Systems Based on Sensitivity Analysis and EKF",

publisher = "American Automatic Control Council",

author = "Mehdi Gholami and Henrik Schiøler and Thomas Bak",

year = "2011",

pages = "244--249",

journal = "American Control Conference. Proceedings",

issn = "0743-1619",

}

RIS

TY - CONF

T1 - Active Fault Diagnosis for Hybrid Systems Based on Sensitivity Analysis and EKF

A1 - Gholami,Mehdi

A1 - Schiøler,Henrik

A1 - Bak,Thomas

AU - Gholami,Mehdi

AU - Schiøler,Henrik

AU - Bak,Thomas

PB - American Automatic Control Council

PY - 2011/7/2

Y1 - 2011/7/2

N2 - An active fault diagnosis approach for different kinds of faults is proposed. The input of the approach is designed off-line based on sensitivity analysis such that the maximum sensitivity for each individual system parameter is obtained. Using maximum sensitivity, results in a better precision in the estimation of the corresponding parameter. The fault detection and isolation is done by comparing the nominal parameters with those estimated by Extended Kalman Filter (EKF). In study, Gaussian noise is used as the input disturbance as well as the measurement noise for simulation. The method is implemented on the large scale livestock hybrid ventilation model which was obtained during previous research

AB - An active fault diagnosis approach for different kinds of faults is proposed. The input of the approach is designed off-line based on sensitivity analysis such that the maximum sensitivity for each individual system parameter is obtained. Using maximum sensitivity, results in a better precision in the estimation of the corresponding parameter. The fault detection and isolation is done by comparing the nominal parameters with those estimated by Extended Kalman Filter (EKF). In study, Gaussian noise is used as the input disturbance as well as the measurement noise for simulation. The method is implemented on the large scale livestock hybrid ventilation model which was obtained during previous research

KW - Active fault diagnosis, sensitivity analysis, hybrid systems, EKF

UR - http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5991038

JO - American Control Conference. Proceedings

JF - American Control Conference. Proceedings

SN - 0743-1619

SP - 244

EP - 249

ER -