Homology recognition of multi voltage sag events based on perceptual hash sequence

Inhalt:
At present, in the power grid, a single voltage sag event will be recorded by different monitoring terminals. Due to the needs of voltage sag location, propagation law and regional power grid sag evaluation, it is necessary to merge the voltage sag events in a certain period of time. Therefore, a multi voltage sag event homology recognition method based on perceptual hash sequence similarity is proposed in this paper. Firstly, the sampling rate of the recorded data is unified, and the number and time of abrupt changes are detected by S-transform; secondly, according to the recording situation and sudden change time, the sag data segment is extracted, and all possible waveforms of the reference data are obtained by using the transformer transfer matrix; then, the gram angle field is used to convert the one-dimensional data of voltage sag into two-dimensional images, and the perceptual hash algorithm is used to convert it into hash sequences. The similarity between sequences is compared by Euclidean distance to realize the homologous identification of voltage sag events. Finally, the simulation data generated by IEEE30 bus system are used to verify the accuracy and practicability of this method.