Unlock instant, AI-driven research and patent intelligence for your innovation.

Power distribution network grounding fault line selection method

A technology of ground fault and line selection method, which is applied to the fault location, detects faults according to conductor types, and measures electricity, etc., which can solve the problems of inconspicuous transient process, rapid removal of unfavorable faults, and inability to select correct lines.

Pending Publication Date: 2021-09-10
SHENYANG INST OF ENG
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This will cause unnecessary short-term power outages on sound lines. At the same time, after selecting faulty lines, the workload of manual line inspection to find fault points is also very heavy, which is not conducive to rapid fault removal, increases power outage time, and reduces power supply reliability.
[0003] When the grounding occurs when the phase voltage and phase are zero, the transient process will be very inconspicuous, especially when the fault grounding resistance is high, the transient fault characteristics can hardly be seen, which will cause incorrect line selection
However, the steady-state line method often ignores the transient information with rich fault characteristics, resulting in low line selection accuracy.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Power distribution network grounding fault line selection method
  • Power distribution network grounding fault line selection method
  • Power distribution network grounding fault line selection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0126] The present invention selects a real ground fault waveform information of a substation admission as a source of sample data, and takes 50 distribution networks that have 4 lines, a total of 200 zero-order current signals, a total of 200 sample data. The respective fault degree value is calculated by the fault degree function, and the results are shown in Table 1.

[0127] Table 1 Sample data sheet

[0128]

[0129] Import 200 sample data into MATLAB for model verification, 160 sets of data as training samples, 40 sets of data as test samples. BP neural network model and GA-LM improved BP line model training test results, respectively Figure 4 , Figure 5 Indicated. In the figure, the abscissa represents the first group to the 40th group of test samples, each 4 sets of samples, corresponding to 4 outlines of the distribution network; the ordinate is the output of the GA-LM improved BP algorithm, indicating the line The degree of failure.

[0130] The output results of the 4...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a power distribution network grounding fault line selection method. According to the method, a fifth harmonic method, an active component method and a transient energy method are adopted to extract fifth harmonic components, active components and transient energy from zero-sequence current signals to serve as fault features, fault measures of the zero-sequence current signals are defined, then the fault measures serve as input of a model, and according to different judgment capacities of fault steady-state and transient information, and by means of the powerful macroscopic search capability and the good global optimization performance of the genetic algorithm, during training, the genetic algorithm is firstly used for searching the weight of the neural network, the search range is narrowed down, and then the BP network improved by the LM is combined, and then a four-layer neural network of an improved LM-BP algorithm model based on a genetic algorithm is formed to carry out accurate solution so as to achieve the purpose of global search, and local minimum can be avoided, so that the line selection model achieves relatively high precision so as to improve the stability and accuracy of line selection.

Description

Technical field [0001] The present invention belongs to the technical field of distribution network fault discriminant, and,,,,,,,, Background technique [0002] The most common fault form in the distribution system is single-phase ground fault. After a single phase phase failure, the neutral point non-effective grounding system occurs, since only a high impedance current loop can only be constituted between the fault point and the neutral point, so that the grounding point current of the system is small, and the signal is difficult to capture. It is difficult to have a fault line to be discovered. However, when the small current grounding system is single-phase, the system's line voltage is still three-in-line, in order to avoid the sudden disconnection of production, there is no need to trip immediately, the power sector stipulates that the system remains for 1-2 hours. At this time, the sound phase voltage rises to normal. Double, if long-termband fault operation, the system ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G01R31/08G01R31/52G01R31/58
CPCG01R31/086G01R31/088G01R31/52G01R31/58
Inventor 高庆忠齐建明马艳娟赵琰王健李昱材姜河王东来林盛罗金鸣宋世巍
Owner SHENYANG INST OF ENG