A fault detection and isolation method for complex industrial systems

A method for the fault detection in complex industrial systems is presented. The plant devices, the sensors, the actuators, and diagnostic tests are described as discrete-event systems. A formal composition rule of these models is given and an inference procedure for the identification and isolation of the faults is discussed. Temporal information is also considered to speed up the fault isolation procedure and, in some cases, to uniquely identify faults with the same static fault signature.