Enhanced discrete event model for system identification with the aim of fault detection

Abstract

In this paper, we present a new model for discrete-event system identification that is suitable for fault detection, called Deterministic Automaton with Outputs and Conditional Transitions (DAOCT). The model is computed from observed fault-free paths, and represents the fault-free system behavior. In practice, a trade-off between size and accuracy of the identified automaton has to be found. In order to obtain compact models, loops are introduced in the model, which implies that sequences that are not observed can be generated by the model leading to an exceeding language. This exceeding language is associated with possible non-detectable faults, and must be reduced in order to use the model for fault detection. We show, in this paper, that the exceeding language generated by the DAOCT is smaller than the exceeding language generated by other models proposed in the literature, reducing, therefore, the number of possible non-detectable faults. We also show that if the identified DAOCT does not have cyclic paths, then the exceeding language is empty, and the model represents all and only all observed fault-free sequences generated by the system. A practical example is used to illustrate the results of the paper.