Identification of transient power quality disturbances based on FCM

This paper presents a new hierarchical identification method for transient power quality disturbance based on fuzzy C-mean clustering (FCM) algorithm. The method consists of five function modules. By use of the ensemble empirical mode decomposition (EEMD) and singular value decomposition method, we can extract effective feature vectors layer by layer, which are used as the input of FCM. In this way the optimized classified matrix and clustering centers are obtained. According to the Euclidean distance between the unknown-sample records and the known-sample ones, the disturbance type is identified. Simulation results of common kinds of transient disturbances indicate that this method is accurate and has a potential application in engineering.