CN108599104A - Tranformer protection New Algorithm for Inrush Current Identification of Transformer - Google Patents

Abstract

Tranformer protection New Algorithm for Inrush Current Identification of Transformer, for it is quick, intuitively identify excitation surge current.For the difference showed between transformer excitation flow and internal fault current waveform, 4 dimensional analysis are carried out using two kinds of electric currents of wavelet transformation pair, and comparison identification is carried out using two kinds of electric currents of the difference pair of peak value in particular dimensions.Tranformer protection New Algorithm for Inrush Current Identification of Transformer provided by the invention can fast, intuitively realize the discovery to excitation surge current.It is not influenced by the extraneous factors such as interval angle and transformer model；It is criterion simple, intuitive, effective and rapid；Identification is accurate and reliable, and by a large amount of emulation experiments, accuracy rate is up to 100%.

Description

Identification Algorithm of Magnetizing Inrush Current for Transformer Protection

technical field

The invention relates to the technical field of excitation inrush current, in particular to an identification algorithm for excitation inrush current for transformer protection.

Background technique

In recent years, with the country's promotion of the development of smart grid technology and the construction and operation of UHV power grid projects, higher requirements have been placed on the safety and reliability of power transformers in the process of power transmission and transformation in power systems. According to the statistics of State Grid Corporation of China on the protective action rate of relay protection devices in operation in recent years, the correct action rate of transformer protection is relatively low, and even lowers the correct action rate of protective devices in the entire network. This shows that the algorithm criterion in the transformer protection device has certain defects, and the judgment accuracy of the fault needs to be improved. Differential protection based on Kirchhoff's current law (KCL) is always one of the main protections of transformers. The focus of research in differential protection is how to better solve the problem of distinguishing excitation inrush current and internal fault current ^{[2 ]} . Due to the influence of Kirchhoff's current law, when the differential protection is used as the main protection of the transformer, the magnetic circuit of the transformer winding in actual operation will affect the current within the protection range, making the correct rate of transformer protection relatively backward. At present, the principles of discontinuity angle, harmonic restraint and sampling value difference are mainly applied in the field of identifying transformer excitation inrush current. But in actual engineering, there are some problems in the application of these principles. Therefore, it is necessary to explore new protection criteria to make up for the defects.

Contents of the invention

The purpose of the present invention is to provide a transformer protection excitation inrush identification algorithm for fast and intuitive identification of excitation inrush current.

The technical solution adopted by the present invention to solve the technical problem is: transformer protection excitation inrush identification algorithm, which is characterized in that, aiming at the difference between the transformer excitation inrush current and the internal fault current waveform, wavelet transform is used to perform 4 comparisons of the two currents Scale analysis, and use the difference of the peak value in a specific scale to compare and identify the two currents.

The beneficial effect of the invention is that: the transformer protection excitation inrush identification algorithm provided by the invention can quickly and intuitively realize the discovery of the excitation inrush current. It is not affected by external factors such as discontinuity angle and transformer model; the criterion is simple, intuitive, fast and effective; the recognition is accurate and reliable, and the accuracy rate is as high as 100% through a large number of simulation experiments.

Description of drawings

Figure 1 is a simulation diagram of a three-phase transformer;

Figure 2 is a simulated waveform diagram of the transformer no-load closing excitation inrush current;

Fig. 3 is the simulation waveform diagram of the inter-turn short-circuit fault current of transformer A phase;

Detailed ways

The identification method of the present invention will be described in detail below.

When the transformer is switched into no-load or the fault outside the area is removed, and the voltage returns to normal, the core magnetic flux has reached saturation at this time and because the magnetic flux cannot change suddenly, aperiodic transient components will appear. Therefore, an excitation current 6-8 times larger than the rated current of the transformer will be generated, which is the excitation inrush current.

Turn-to-turn short circuit is one of the main forms of transformer internal faults. Recently, a large number of entangled windings are used in large-scale transformer high-voltage windings. The invention performs multi-scale analysis on the signal waveform by means of wavelet transformation, and then obtains new criteria according to the decomposed signal. Fourier transform can transform the signal from time domain to frequency domain analysis. The theory of wavelet transform is based on Fourier transform. At the same time, good time-frequency localization features are added, and a time window that can change with frequency is provided. , which can perform accurate and effective time-frequency processing on the signal.

(1) Simulate the inrush current and internal fault current of the transformer; as shown in Figure 1, use the closing and opening of the three-phase circuit breakers BRK1 and BRK2, close BRK1 at 0.1s to realize the no-load closing of the three-phase transformer , to simulate the excitation inrush current of the transformer, and the simulation time is 0.5s. By connecting the Timed Fault Logic module between the three-phase transformer and BRK2, the occurrence and end of the fault are controlled, and the inter-turn short circuit fault of phase A is simulated to realize the waveform simulation of the internal fault of the three-phase transformer. The fault starts when 0.1s is set and the duration is 0.4s.

(2), analyze the waveform;

There are obvious differences between the excitation inrush current and the internal fault current waveform, as shown in Figure 2 and Figure 3, in which the excitation inrush current is obviously asymmetrical and biased to one side of the time axis, and the waveform contains discontinuous angles. At the same time, it is found that there are a large number of The harmonic components of the harmonic components, in which the proportion of the second harmonic is larger. In a three-phase transformer, under the premise that the closing angle remains unchanged, the waveform of the excitation inrush current of a certain phase will be in a state of symmetrical inrush current, but the other two phases are still biased to one side of the time axis. The internal fault current will gradually become symmetrical about the time axis with the increase of time, the attenuation of the amplitude is not limited by the transformer capacity and the saturation degree of the iron core, there is no discontinuity angle, and the amplitude of the fault current is larger than the excitation inrush current.

Claims (1)

1. Transformer protection excitation inrush identification algorithm, which is characterized in that, aiming at the difference between the transformer excitation inrush current and the internal fault current waveform, the wavelet transform is used to analyze the two currents in 4 scales, and the difference between the peak values in a specific scale is used The value compares and identifies the two currents.