Residual based fault detection using an output observer

Optimal methods for designing observers used for fault detection have been proposed in Viswanadam and Minto (1988). Since then, many authors have presented different design methods, most of them based on the H/sub /spl infin// technique. In this paper, a procedure for designing an l/sub /spl infin// optimal observer is presented, where the optimal observer is obtained from the solution of a linear programming (LP) problem. As a measure of the sensitivity of the fault detection scheme, the minimum detectable fault and maximum undetectable fault are evaluated.

[1]  J. Pearson,et al.  l^{1} -optimal feedback controllers for MIMO discrete-time systems , 1987 .

[2]  N. Viswanadham,et al.  Robust Observer Design with Application to Fault Detection , 1988, 1988 American Control Conference.

[3]  An L∞ optimal performance approach to robust feedback control , 1989, 1989 American Control Conference.

[4]  P. Frank,et al.  Fault detection via factorization approach , 1990 .

[5]  Paul M. Frank,et al.  Robust observer design via factorization approach , 1990, 29th IEEE Conference on Decision and Control.

[6]  Paul M. Frank,et al.  A frequency domain approach to fault detection of uncertain dynamic systems , 1993, Proceedings of 32nd IEEE Conference on Decision and Control.

[7]  Steven X. Ding,et al.  Frequency domain approach to optimally robust residual generation and evaluation for model-based fault diagnosis , 1994, Autom..

[8]  J. Kantor,et al.  Computing bounds for a simple fault detection scheme , 1994, Proceedings of 1994 American Control Conference - ACC '94.