A new method for power quality disturbance classification based on the proliferating V-detectors algorithm

In this paper, a new method for power quality (PQ) disturbance classification based on immune systems method is proposed. After the S-matrix is obtained using S-transform of the raw data, four normalized, statistical features for PQ disturbances are extracted for classification. The proposed classification method, based on the proliferating V-detectors algorithm, is then applied. Here, the state space is divided into self and non-self regions, with the self region corresponding to all possible inputs. Initially, a set of non-self detectors is generated to cover the non-self region. Then the data for one class are removed separately, and the existing detectors are allowed to proliferate class-detectors into the uncovered region. This process is repeated until each class has its own set of class-detectors. Finally, every new PQ disturbance datum is tested to compute its class, by matching either the class-detectors or non-self detectors. The proposed method was applied to a four-class PQ disturbance classification problem. The simulation results verify the applicability and effectiveness of the proposed classification method.