A new method for internal overvoltage identification of distribution network

Abstract

The invention relates to a novel internal overvoltage identification method of a distribution network. The atomic decomposition algorithm optimized based on the imperial colonial competition algorithm is applied to the analysis of overvoltage signals. The optimized algorithm can quickly and effectively extract the internal characteristics of the signal, and is not easily disturbed by noise; based on the atomic decomposition algorithm, three-phase characteristic atoms are constructed Spectrum, transforming the waveform signal containing only time-domain information into high-dimensional features containing time-frequency information, which completely describes the time-frequency characteristics of the overvoltage signal; using the convolutional neural network to directly identify the three-phase characteristic atomic spectrum, It solves the deficiency of high-dimensional feature recognition based on shallow learning recognition algorithm, and avoids the subjectivity and complexity in the process of feature reduction. The method for identifying the overvoltage type of the distribution network of the present invention has higher identification accuracy and stronger adaptability through verification of simulation waveforms and physical models.

Description

technical field [0001] The invention relates to the field of distribution networks, in particular to a novel internal overvoltage identification method of a distribution network. Background technique [0002] Traditional operating experience and research have shown that overvoltage may cause insulation breakdown of equipment and lines, affect the normal operation of the power system, and eventually bring huge economic losses and personal hazards. According to statistics, overvoltage accidents in the distribution network account for about 70% to 80% of the overvoltage accidents in the entire power system. Quickly and accurately identifying the type of overvoltage will help engineers and technicians find out the cause of the accident in a timely manner and provide a basis for proposing The voltage suppression method and the improvement of distribution network insulation coordination provide a basis, which is of great significance for improving the self-healing ability of distr...

AI Technical Summary

Problems solved by technology

This type of shallow classifier has a good recognition effect on low-dimensional features, but it is often difficult to recognize high-dimensional features due to structural limitations.

Images