A small current ground fault detection method based on the combination of vmd and gray correlation degree

Abstract

The invention discloses a small current grounding fault detection method based on the combination of VMD and gray correlation degree, which includes collecting and monitoring the zero-sequence voltage of the distribution network busbar, and collecting the zero-sequence current waveform at the same time; using VMD to decompose each zero-sequence current into low-frequency IMF component and high-frequency IMF component, and sample them to obtain low-frequency sequence and high-frequency sequence; calculate low-frequency average gray correlation degree, high-frequency average gray correlation degree, and judge according to low-frequency average gray correlation degree and high-frequency average gray correlation degree Whether a single-phase-to-earth fault occurs and the type of fault. The method of the invention is not affected by the location of the fault, the magnitude of the closing angle of the fault and the time of the fault, has high detection accuracy, and has certain robustness to noise and good adaptability.

Description

technical field [0001] The invention belongs to the technical field of distribution network, and more specifically relates to a small current grounding fault detection method. Background technique [0002] Single-phase-to-ground fault handling in distribution network is related to personal safety and power supply reliability, and has received extensive attention in recent years. On the one hand, more than 70% of faults and blackouts are caused by medium-voltage distribution networks, and single-phase grounding is the most important fault form of distribution networks, accounting for more than 80% of the total faults. At present, my country's distribution network mainly adopts ungrounded and arc-suppressing coil grounding systems. When a single-phase ground fault occurs, traditionally it mainly relies on manual testing to determine the fault line, and the ground fault handling capability of the distribution network automation system is also limited. It is far behind the short...

AI Technical Summary

Problems solved by technology

However, the selection of wavelet basis functions is too dependent on prior knowledge, so the adaptability of the algorithm is relatively poor, and its adaptability needs to be improved; The input signal can be decomposed, with strong adaptability and simple calculation, but it is prone to modal aliasing during the decomposition process, and the intuitiveness of the decomposition results needs to be improved

In addition, improved EMD algorithms such as CEEMD and EEMD are gradually applied to fault detection. The principle is to improve the performance of EMD algorithms by adding different types of noise. Such algorithms reduce the occurrence of modal aliasing, but there is decomposition The post information is not intuitive; the Prony algorithm realizes the fitting of the input signal by superimposing multiple exponential functions of arbitrary amplitude, phase and attenuation factor, and the decomposition of the unstable signal is accurate and intuitive, but the algorithm stability is vulnerable to high frequency noise; in recent years, with the continuous deepening of research, many scholars have proposed the use of chaotic oscillators and S-transform methods to achieve fault detection

However, most algorithms have the problem of large amount of information after decomposition and are not easy to be used.

Images