Robust fault detection filter design for a class of neutral‐type neural networks with time‐varying discrete and unbounded distributed delays. (19th June 2012)

• Record Type: Journal Article
• Authors: Zhang, Xian; Lin, Xue; Wang, Yantao
• Is Part Of: Optimal control applications and methods. Volume 34:Number 5(2013:Sep./Oct.)
• Page Start: 590
• Page End: 607
• Publication Date: 2012-06-19
• Subjects: neutral‐type neural networks -- robust fault detection filter (RFDF) -- unbounded distributed delay -- cone complementarity linearization (CCL); Control theory -- Periodicals; Mathematical optimization -- Periodicals; 629.8312
• DOI: 10.1002/oca.2041
• Languages: English
• ISSNs: 0143-2087

Abstract

SUMMARY: This paper is concerned with the problem of robust fault detection filter design for a class of neutral‐type neural networks with time‐varying discrete and unbounded distributed delays. A Luenberger‐type observer is designed for monitoring fault. By introducing an appropriate Lyapunov–Krasovskii functional and by using Jensen's inequality techniques to deal with its derivative, a new sufficient condition for the existence of robust fault detection filter is proposed in the form of LMIs with nonlinear constraints. To solve the nonlinear problem, a cone complementarity linearization algorithm is proposed. In addition, several numerical examples are provided to illustrate the applicability of the proposed approach. Copyright © 2012 John Wiley & Sons, Ltd.