A functional graph approach for alarm filtering and fault recovery for automated production systems

Alarm processing is one of the functions of the monitoring and the supervision of automated productions systems. However, these systems become more and more complex. Actually, this induces that standard alarm processors do not manage efficiently all the alarms that occur. This paper deals with intelligent alarm processing system (IAPS). The goal of this type of system is to reduce the number of alarms that are presented to the operators and to assist them to select corrective actions in case of fault occurrence. In this paper, we propose a prototype developed under PROLOG. Different concepts such as validation, causal inhibition or synthesis alarms characterize this prototype features. The filtering process is independent of the nature of the process (continuous or discrete) and of the way the raw alarms are produced.