Early warning of slight changes in systems and plants with application to condition based maintenance

Techniques for early warning of slight changes in systems and plants are useful for condition based maintenance. In this paper we present an approach for this problem. This approach is based on the so-called "asymptotic local" approach for change detection previously introduced by the same authors. Its original principle consists in characterizing a system via some identified model, and then to monitor its changes using some data-to-model distance also derived from identification technique. We show here that this method is of much wider applicability : model reduction can be enforced, biased identification procedures can be used and finally one can even get rid of identification and use instead some much simple Monte-Carlo estimation technique prior to change detection. Experiments on AR model are reported and an example from gas turbine industry is briefly discussed