Vibrant fault diagnosis for hydroelectric generator units with a new combination of rough sets and support vector machine

► A classifier based on a novel combination of rough sets and SVM is proposed. ► The basic ideas of rough sets are used to describe the classification results of SVM. ► Samples belonging to lower approximation are considered as precise classification. ► The classification reliabilities for samples belonging to boundary region are given. ► The proposed method is successfully applied in diagnosing the faults of a HGU.