Timed-event-state-based diagnoser for manufacturing systems

Abstract

This paper proposes an adapted diagnoser for manufacturing systems. This diagnoser combines event and state based models to infer the fault's occurrence using event sequences and state conditions characterized by sensor's readings and commands issued by the controller. Furthermore, this diagnoser uses expectation functions to capture the inherent temporal dynamics of the system represented by time delays between correlated events. Copyright © 2006 International Federation for Information Processing.