Classification for voltage sags based on generalized S-transform

The classification and recognition of voltage sags are the important part of power quality management program. This paper proposes a new classification method of voltage sags based on generalized S-transform. It firstly determines the adjustment factor λ of the Gaussian window according to the main frequency of voltage sags, and the modulus matrixes are regarded as the standard templates; through comparing the similarities between generalized S-transform modulus matrixes of test signals and the standard templates, the classification process can be finished. In the process of similarity comparison, the concept of normalization and the idea of calculating similarities at different frequencies are proposed to enlarge the differences among different templates. This proposed method realizes the disturbance classification with simple calculation of the similarities among different sags; no extra classification processes are needed. Simulation and the analysis of field data show that the proposed method is easy to achieve the purpose of classification and has high accuracy and noise robustness.