Fault diagnosis model based on Petri net with fuzzy colors

The problem addressed in this paper concerns fault diagnosis based on incomplete and/or uncertain information. The aim of this research is to explore all information which can be supplied by operators or exploitation persons in order to make more efficient fault diagnosis process. A Petri net with fuzzy colors (PNCF) tool has been developed to model functional structure of equipment and to reproduce propagation process of failure. Fuzzy colors are used to characterize information and processed by PNFC rules to identify the type of fault and to determine the sequence of tests which should be applied to locate faulty component. The practice and the usefulness of PNFC in the fault diagnosis problem will be discussed using an example. (C) 1999 Elsevier Science Ltd. All rights reserved.