Disturbance Waveform Detection Technology Based on Probability Density of Remaining Data

With the increasing application of disturbance waveform data, a general disturbance detection technology faced with various transient waveforms has practical significance for simplifying the detection process of specific disturbances and improving the accuracy of the specific disturbance detection. To this end, a detection algorithm for transient disturbance waveforms based on the probability density of remaining data is proposed in this paper. Firstly, the method of obtaining the remaining waveform data and its probability density is presented. Then, the probability distribution law of noise in power waveform data is studied based on the field measured data. Finally, according to the probability density difference of the remaining waveform data in adjacent periods, a transient waveform detection algorithm based on Wasserstein distance between probability densities is established. The threshold is determined by the box plot. A large amount of field measured data verifies the effectiveness of the proposed disturbance detection algorithm.