Identification of power quality disturbances based on improved TT transform and support vector classifier

Power quality disturbances (PQDs) identification is a crucial procedure in ameliorating the power quality performance for utilities and power consumers. The improved TT transform is proposed in this paper to extract the feature information of 5 typical power quality disturbance signals (e.g. voltage swell, sag, interruption, harmonics, oscillation). Two tuning parameters r and m are introduced in TT transform to improve the energy concentration and resolution as well. In accordance with the feature information, the categories of PQDs are identified intelligently by the support vector classifier (SVC). Ultimately, compared with back propagation (BP) network, the classification accuracy of SVC demonstrates preferable performance, which verifies the proposed method an elegant approach in PQDs identification.