Rough set model for vibration fault diagnosis of steam turbine-generator set

Extraction of simple and effective decision rules for fault diagnosis is one of the most important issues needed to be addressed in mechanical fault diagnosis, because available information is often inconsistent and redundant. This paper presents a fault diagnosis model for vibration fault diagnosis of a steam turbine-generator set based on rough set theory. From original fault data containing inconsistent and redundant information, the discretized fault symptom attributes are reduced using the genetic algorithm. Then, a set of maximal generalized decision rules with certainty factor and coverage factor are generated by using a proposed value reduction algorithm, and therefore a decision rules base for fault diagnosis is established. When the proposed model is applied to fault diagnosis of a new case, the discretized fault symptom attributes are first matched with the rules in the decision rules base. Then the returned diagnostic decision rules are evaluated by using the proposed algorithm, and finally the diagnosis is made. The general scheme of the proposed model for vibration fault diagnosis of steam turbine-generator set is established in the paper.