In view of the large computational quantity involved in the S-transform, a new classification of voltage sags, which is based on single fundamental frequency S-transform, has been proposed for its improvement. First, the acquisition of module vector is achieved through S-transform of single fundamental frequency, followed by the extraction of such characteristic parameters as mutation point numbers, sag magnitude and sag amplitude standard deviation from this vector. Finally, a classification is made of voltage sag signals based on the characteristic parameters input in the decision tree. Simulation results show that the proposed method is effective for the classification of voltage sags.