Local/global fault diagnosis of event-driven controlled systems based on probabilistic inference

This paper presents a new local/global fault diagnosis strategy for the event-driven controlled systems such as the programmable logic controller (PLC). First of all, the controlled plant is decomposed into some subsystems, and the global diagnosis is formulated using the Bayesian network (BN), which represents the causal relationship between the fault and observation in subsystems. Second, the local diagnoser is developed using the conventional timed Markov model (TMM), and the local diagnosis results are used to specify the conditional probability assigned to each arc in the BN. By exploiting the local/global diagnosis architecture, the computational burden for the diagnosis can be drastically reduced. As the result, large scale diagnosis problems in the practical situation can be solved. Finally, the usefulness of the proposed strategy is verified through some experimental results of an automatic transfer line.