Distributed monitoring based on chronicles recognition for a milk manufacturing unit

Abstract EN

Developments presented in this paper are devoted to the distributed monitoring of discrete event systems and deals with the problem of the distributed detection of a failure symptom in manufacturing job-shops with time constraints.

In such systems, operation times are included between a minimum and a maximum value.

P-time Petri nets are used for modeling.

The monitoring system is distributed on several monitoring sites and the detection function is distributed on these different sites.

The aim of the detection is to recognise in a distributed way some specific evolutions of the monitored process.

These evolutions are represented by chronicles.

A chronicle is a set of patterns of observable events temporally constrained.

Within the framework of the monitoring, the set of chronicles gives the signatures of the system failures and provides the knowledge necessary for the diagnosis.

Finally, to demonstrate the effectiveness and accuracy of the distributed monitoring approach, an application to a milk production unit is outlined.

The results show that a distributed strategy improves the prevention of constraints violation by performing an early diagnosis.