Title :
Robust fault detection dynamic observer design for LTI systems
Author :
Wei, Xiukun ; Lihua, Liu ; Cai, Guoqiang ; Qin, Yong ; Jia, Limin
Author_Institution :
State Key Lab. of Rail Traffic Control & Safety, Beijing Jiaotong Univ., Beijing, China
Abstract :
Dynamic observer provides more freedom to meet the diversified performance specifications for fault detection purpose in contrast with static observer. However, little attention is paid on its design and only a few papers investigated how to design a fault detection dynamic observer. In this work, we propose an innovative dynamic observer design approach for linear time invariant (LTI) systems subjected to unknown disturbance and possible faults with the aid of the generalized KYP lemma developed recently. The dynamic observer is designed such that the disturbances are attenuated to a specified level to meet the H∞ performance in the concerned finite frequency interval while the sensitivity to the possible faults in its frequency domain is maximized in the H- index sense. The existence conditions of such an observer are given in terms of linear matrix inequalities (LMIs). An iterative algorithm is presented to achieve the solution. The effectiveness of the proposed method is illustrated by a numerical example.
Keywords :
fault tolerance; frequency-domain analysis; linear systems; observers; robust control; sensitivity; H∞ performance; H-index sense; LMI; LTI systems; diversified performance specifications; finite frequency interval; frequency domain; generalized KYP lemma; innovative dynamic observer design approach; iterative algorithm; linear matrix inequality; linear time invariant systems; robust fault detection dynamic observer design; sensitivity; static observer; unknown disturbance; Design methodology; Fault detection; Frequency domain analysis; Iterative algorithms; Laboratories; Linear matrix inequalities; Rails; Railway safety; Robustness; Traffic control; Dynamic Observer; Fault Detection; GKYP Lemma; Threshold Design;
Conference_Titel :
Control and Decision Conference (CCDC), 2010 Chinese
Conference_Location :
Xuzhou
Print_ISBN :
978-1-4244-5181-4
Electronic_ISBN :
978-1-4244-5182-1
DOI :
10.1109/CCDC.2010.5498164