On-line Diagnosis of a Technological System : a Fuzzy Pattern Recognition Approach

This paper gives a description of an adaptative on-line diagnosis system for the detection of slow varying changes in dynamical systems. This fault detection system is based on both a learning and a supervisor module. The learning module acts as an adaptatif fuzzy classiier i.e. its parameters are updated on-line. The supervisor module controls the learning periods of the classiier, it makes the adaptation of the classiier parameters on steady states, these parameters are saved without changes as evolutions occur (e.g. slow or abrupt). So, the diagnosis system is able to detect : new operation modes and slow or abrupt varying changes on the plant operation modes.