A QUANTITATIVE PERFORMANCE INDEX FOR OBSERVER-BASED MONITORING SYSTEMS

Model-based monitoring systems based on state observer theory are attr active for machine monitoring because practical, inexpensive, and reli able sensors can be located remote to the signal(s) of interest. Then, a model of the machine plus an estimation algorithm are utilized to c onvert the output of the remote sensors to signals representing the de sired local behavior. While this type of monitoring system has shown m uch promise in the laboratory, it has not been widely accepted by indu stry because, in practice, these systems often have poor performance w ith respect to accuracy, bandwidth, reliability (false alarms), and ro bustness. In this paper, the limitations of the deterministic state ob server are investigated quantitatively from the machine monitoring vie wpoint. The limitations in the transient and steady-state observer per formance are quantified based on the estimation error bounds, and from these error bounds, performance indices are selected. Then, based on the relationships between the indices, a main index is determined in o rder to represent the overall observer performance. The index is the c ondition number of the observer eigenvectors in L(2) norm. It is shown that observers with small condition numbers are guaranteed to have sm all error bounds. This index can be utilized as a quality condition fo r any linear observer regardless of how it is designed as well as form the basis for an observer design methodology for high performance obs erver-based monitoring systems.