Internal Overvoltage Identification Method of Distribution Network Based on Block Time Spectrum and Deep Learning Algorithm

Abstract

The invention relates to a power distribution network internal overvoltage identification method based on a partitioned time-frequency spectrum and a deep learning algorithm. Firstly, an overvoltage signal is obtained; then, the overvoltage signal is decomposed; then, band-pass filtering is conducted on components according to a set frequency band to construct a time-frequency matrix; then, a partitioned time-frequency spectrogram is obtained; finally, the deep learning algorithm is adopted for classification and identification. The method adopts the partitioned time-frequency spectrogram as an input of the deep learning algorithm, various overvoltage characteristics can be independently learned, and classification and identification of internal overvoltage of a power distribution network are achieved.

Description

technical field [0001] The invention relates to the field of distribution networks, in particular to an internal overvoltage identification method of a distribution network based on block time spectrum and deep learning algorithm. Background technique [0002] The safe and reliable operation of the power grid is an important guarantee for the scientific development of the national economy and people's livelihood. With the rapid development of smart grid in my country, the requirements for power supply reliability and power quality of distribution network are getting higher and higher. Operation experience shows that overvoltage is one of the important factors affecting the safe operation of distribution network. The internal overvoltage of the distribution network lasts for a long time, which poses a great threat to the safe and reliable operation of the distribution network. Therefore, it is very necessary to detect the overvoltage in the distribution network in time and ...

AI Technical Summary

Problems solved by technology

Due to the diversity of overvoltage types, it is often necessary to seek multiple feature quantities to characterize the characteristic mode of the overvoltage signal to achieve the purpose of identification, and the classification algorithms applied to the classification and identification of distribution network overvoltage are mainly based on relatively mature methods. machine learning algorithms that do not have the ability to learn by themselves

Images