High resistance ground fault detection method based on flexible DC distribution network

Abstract

The invention discloses a method for detecting a high-resistance grounding fault in a flexible direct current distribution network. First, the characteristic mode IMF is extracted by using a complementary set empirical mode decomposition algorithm for the transient zero-mode current 1 components, and the IMF 1 Perform the first-order difference operation to obtain the abrupt singular point, calculate the cumulative slope sum near the singular point, and compare the slope sum value with the start-up threshold to distinguish the fault state from the normal state; then, use the Prony algorithm to analyze the IMF 1 Parameter identification of components to obtain IMF 1 The characteristic frequency component and the DC component in the component, and then calculate the energy ratio of the characteristic frequency component and the DC component, and further distinguish different states through the difference in the value of the energy ratio. Compared with the existing methods, the present invention can adapt to accurate feature extraction under strong noise environment, and the feature extraction process has self-adaptability, convenient application, high detection accuracy, can keenly determine the operating state of the distribution network system, and simultaneously improves the calculation speed.

Description

technical field [0001] The invention belongs to the technical field of power grid fault detection, and in particular relates to a high-resistance grounding fault detection method of a flexible DC distribution network. Background technique [0002] Recently, the number of DC loads is increasing day by day. When using the traditional AC power distribution system to realize the consumption of distributed energy and the power supply of DC loads, a large number of power electronic converter equipment is required, and the investment cost is greatly increased. On the contrary, if the consumption of distributed energy and the power supply of DC loads are carried out through the DC power distribution system, a large amount of converter equipment can be saved and investment costs can be reduced. In addition to the advantages in terms of economy, compared with the traditional AC power distribution system, the advantages of the DC power distribution system are also reflected in: improvi...

AI Technical Summary

Problems solved by technology

[0006] 1) High-resistance fault feature extraction: The fixed basis function leads to insufficient feature characterization; the feature extraction process is not adaptive. Although the EMD algorithm is adaptive, it is prone to problems such as mode aliasing and endpoint effects; unclear meaning

[0007] 2) High-impedance fault detection criterion: the start-up criterion of the existing HIF detection method is not sensitive, and the distribution network system is misjudged to be in a normal state; the start-up criterion and the discrimination criterion are not constructed separately, resulting in low HIF detection accuracy ; Computationally complex, and most criterion constructions have no physical meaning

Images