JPH07270466A - Failure state discrimination method - Google Patents

Abstract

PURPOSE:To alleviate a burden due to a discrimination analysis and accurately discriminate failure states in discriminating the causes of a grounding failure occurring in a power cable on the basis of a waveform analysis as well as the discrimination analysis. CONSTITUTION:The waveform of earth current flowing in a power cable is arrested with CT and judgement is made as to whether a failure cause is attributable to thunderbolt on the basis of the wave height value of final flashover. Then, a discrimination analysis is undertaken on the basis of the spectrum analysis value of the earth current waveform. Also, failure causes are generally classified on the basis of whether earth current is flowing before the flashover and, then, a discrimination analysis is undertaken on the basis of the spectrum analysis value of the earth current waveform. Thus, a burden due to the discrimination analysis can be alleviated and discrimination accuracy can be improved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a failure appearance determining method for determining the cause of a failure from a ground fault waveform when a ground fault occurs in a power line such as a transmission line or a distribution line.

[0002]

2. Description of the Related Art The main causes of ground faults in power lines are:
Examples include lightning strikes, contact with trees, contact with animals such as birds and snakes, contact with metals such as cranes, adhesion of ice, and lower insulation of insulators. When a ground fault occurs in the power line due to these causes, it is easy to identify the occurrence of the ground fault at the monitoring station, but it is not easy to identify the cause of the fault, and in many cases the worker will not be able to visit the site. I went out to find out the cause of the failure and then did repair work. However, since it is not easy to specify the failure point, it takes a lot of time to find the cause of the failure.

Therefore, the applicant of the present invention has installed a CT on the power line.
The ground fault current waveform was taken in by the method, the power spectrum value obtained by spectrum analysis of the ground fault current waveform was discriminated and analyzed, and a failure cause discrimination method using a neural network was developed. Patent pending. This method was developed based on the knowledge that the ground fault current waveform generated on the power line differs depending on the cause of the failure, and it is possible to estimate the cause of the failure at the monitoring station and then go to the site for repair. This has the effect of significantly reducing the work time.

[0004]

However, in the failure appearance discriminating method of this prior application, the failure appearance is determined by performing the discriminant analysis using the spectral analysis value of the ground fault current waveform and the ground fault current value. However, there is a problem that the final flashover is not clear and the determination accuracy is lowered depending on how to take this point. The present invention has been made to solve the problems of the invention of the prior application and to provide a failure appearance determining method capable of accurately determining the cause of a failure.

[0005]

SUMMARY OF THE INVENTION A first invention made to solve the above problems is a method for determining a cause of a failure from a ground fault current waveform of a power line, in which a peak value of final flashover is used. This method is characterized by roughly classifying whether or not the cause of the failure is a lightning strike, and then performing a discriminant analysis based on the spectrum analysis value of the ground fault current waveform. A second aspect of the present invention is a method of determining a cause of a failure from a ground fault current waveform of a power line, wherein the cause of the fault is roughly classified according to the presence or absence of a ground fault current before the final flashover, and then the ground fault current waveform It is characterized by performing discriminant analysis based on spectrum analysis values. It should be noted that more preferable results can be obtained by implementing the first invention and the second invention in combination as in the embodiments described later.

[0006]

According to the first aspect of the present invention, whether or not the cause of the failure is lightning strike is first identified by the peak value of the final flashover, which is the raw waveform of the ground fault current, and then the spectrum analysis value of the ground fault current waveform is determined. Since the discriminant analysis is performed, the discriminant analysis can be reduced. Also in the second invention, the presence or absence of the ground fault current before the final flashover is discriminated, the causes of the fault are roughly classified, and the spectrum analysis value of the ground fault current waveform is discriminated and analyzed. The range can be reduced as compared with the conventional case, the discriminant analysis can be reduced, and the cause of failure can be discriminated more accurately.

[0007]

The present invention will be described in more detail below with reference to the drawings. FIG. 1 is a flow chart of the present invention. First, a ground fault current waveform of a power line is detected by CT, this ground fault current waveform is stored for analysis, and an analysis portion before and after final flashover is taken out. In the present invention, it is necessary to specify the final flashover, but as shown in FIG. 2, when the ground fault current waveform generated by the ground fault exceeds a predetermined threshold value, the final flashover occurrence point is defined. Define. If this threshold value is set to about 20 A in the actual line, the final flashover can be surely determined. In addition, in order to perform the waveform analysis, the waveform within the predetermined time range is always stored.

In this embodiment, first, it is discriminated whether or not the cause of the ground fault accident is a lightning strike based on the maximum peak value of the final flashover. That is, as shown in FIG. 3, in the case of lightning, the peak value of the final flashover becomes much larger than the ground fault current value due to the subsequent current, so that, for example, the peak value of more than twice the ground fault current value If it is detected, it should be judged as lightning. On the other hand, in the case of other causes, as shown in FIGS. 4 to 7, the difference between the peak value of the final flashover and the ground fault current value due to the subsequent flow is relatively small. This makes it possible to accurately and easily identify whether or not it is a thunder. If it is determined that there is thunder, no further determination is necessary.

Next, the cause of the accident is roughly classified according to the presence or absence of the ground fault current before the final flashover. 3 to 7 show ground fault current waveforms (raw waveforms). Fig. 3 shows ground fault current waveforms caused by lightning strikes, Fig. 4 shows ground fault current waveforms caused by contact with trees (cherry blossoms), and Fig. 5 shows contact with birds. Ground fault current waveform, Fig. 6
Shows a ground fault current waveform due to ice adhesion, and FIG. 7 shows a ground fault current waveform due to contact of a metal rod. As can be seen from these figures, there is almost no ground fault current before the final flashover in the case of lightning strikes, ice adhesion, contact with metal rods, etc., but in the case of contact with trees or birds, the final flashover occurs. There is a clear ground fault current in front of. When these trees and birds are the cause, it is considered that the leakage current gradually flows and the surface of the contact object is carbonized, resulting in final flashover. Also, in the case of insulator fouling, the leakage current gradually flows to reach the final flashover, so that a clear ground fault current occurs before the final flashover as in the case of trees and birds. Therefore, in the present invention, at the stage of the raw waveform before the spectrum analysis of the ground fault current waveform, the failure causes are roughly classified according to the presence or absence of the ground fault current before the final flashover.

On the other hand, the extracted ground fault current waveform is subjected to spectrum analysis as in the prior invention. Since the ground-fault current waveform may include a DC component and the maximum amplitude is different, it is necessary to normalize the waveform before performing spectrum analysis. Therefore, as shown in the flow of FIG. 1, the direct current component is cut off, the ground fault current value is divided by the maximum amplitude, and any ground fault current waveform is normalized to have the same maximum amplitude. Next, the normalized ground fault current waveform is spectrally analyzed. The result is as shown in Fig. 8 and below. Although the peak of the spectrum intensity appears in the commercial frequency part, various peaks also appear in the frequency part (harmonics) of integral multiples, and the pattern has a ground fault. It depends on the cause of the accident.

For example, in the case of the tree shown in FIG. 8, a peak with a large spectral intensity appears only in the commercial frequency portion, and almost no peak appears in other portions. Figure 9
In the case of the bird shown in (1), a slight peak is generated even at the frequency three times the commercial frequency. In addition, in the case of the snake shown in Fig. 10, in addition to the large peak of the commercial frequency part, 1000 Hz
A large number of complex peaks occur in the area up to.

As described above, when the ground fault current waveform is spectrally analyzed, a common power spectrum value can be obtained if the cause of the ground fault is the same. Therefore, this is analyzed using the multiple discriminant analysis method and the neural network method. By doing so, the cause is estimated. The multiple discriminant analysis process is a method of discriminating by showing numerical values the possibility of a plurality of causes, and this analysis process is repeatedly executed to narrow down to a higher rank and to estimate the cause. As described above, in the present invention, the cause of failure is roughly classified in advance based on the raw waveform, so that the cause of failure can be easily estimated.

In the discriminant analysis system of the present embodiment, the ground fault current waveform is divided into before and after the final flashover, three waves immediately before the flashover and three waves immediately after the flashover are taken out, respectively, Fourier-transformed into power spectra, and then each A group of each cause of failure is created with the power spectrum value at a frequency that is an integer multiple of the commercial frequency indicating the characteristic of the cause of failure and the ground fault current value before, during, and after the flashover of the failure current as representative characteristics. Derive the criteria (discriminant) for dividing. Here, the newly obtained data is applied to the discriminant to perform the analysis process,
From the result, the cause of failure which is listed up (analyzed with high possibility) is narrowed down, and the analysis process is performed again only for the narrowed down cause of failure, and the final judgment result is obtained. Also, in the case of neuro processing, the possibility of each cause is output as a numerical value, and this is a method of performing discrimination.In this system, the power spectrum value and the ground fault current value of the above-mentioned representative characteristics are previously learned as learning data in the input layer, Middle layer, output layer 3
The neural network input layer of the back-propagation model consisting of layers is randomly given to the input layer several thousand times for learning, new data is input to that layer, and the one with the largest value among the output layer failure causes is output as the discrimination result. To do.

Further, in the case of the embodiment, in addition to reading the result of the multiple discriminant analysis and the result of the neuro processing into the inference system known as the expert system, the information about the date and time such as the season and the time around the site is also taken in to determine the cause of the ground fault. Make a comprehensive judgment. In other words, the result of the multiple discriminant analysis method, which has no learning ability but can derive a discriminant with an appropriate amount of data and carry out the analysis, and the enormous amount of data are required to create the network in the initial stage. The results of the neural network method that has the ability to learn each time new data is obtained are weighted by taking advantage of their respective advantages and which analysis result is to be emphasized according to the data amount of the determined failure cause. To judge. At that time, in order to further improve the probability of positive discrimination, the information on the season and time at the time of occurrence of a ground fault is taken in, the impossibility of the cause of the fault is determined before the determination, and the cause of the discrimination is deleted. Limit the results. This is because, for example, if a confusing power spectrum value occurs in the winter even though the snake is hibernating, it may be mistaken for a ground fault due to the snake, or heavy machinery such as a crane (metal This is to prevent erroneous determination that a ground fault is caused by a stick. In addition to this, in case of a subtle final judgment, the actual results of failure causes at the same time in the past and the cumulative number of each failure cause are taken in from the database that stores the failure result data, and used in the comprehensive judgment. More accurate inference can be made by adding weighting to which discrimination result is selected. In this way, the cause of failure can be determined accurately and in detail.

[0015]

As described above, according to the first invention and the second invention, the spectrum analysis value is discriminated and analyzed after the failure cause is roughly classified from the raw waveform of the ground fault current.
It is possible to reduce the discriminant analysis and accurately determine the cause of failure. Therefore, the use of the present invention has an advantage that the cause of the ground fault can be accurately known at a remote monitoring place or the like, and the countermeasure according to the cause can be immediately taken. Therefore, the present invention greatly contributes to the industry as a failure aspect determination method that solves the conventional problems.

[Brief description of drawings]

FIG. 1 is a flow chart showing an embodiment of the present invention.

FIG. 2 is a waveform diagram of final flashover.

FIG. 3 is a waveform diagram showing a waveform of a ground fault current caused by a lightning strike.

FIG. 4 is a waveform diagram showing a waveform of a ground fault current caused by a tree.

FIG. 5 is a waveform diagram showing a waveform of a ground fault current due to a bird.

FIG. 6 is a waveform diagram showing a waveform of a ground fault current caused by ice.

FIG. 7 is a waveform diagram showing a waveform of a ground fault current generated by a metal rod.

FIG. 8 is a graph showing a spectrum of a ground fault current waveform generated by a tree.

FIG. 9 is a graph showing a spectrum of a ground fault current waveform generated by a bird.

FIG. 10 is a graph showing a spectrum of a ground fault current waveform generated by a snake.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hiroyuki Katsukawa 2-56, Sudamachi, Mizuho-ku, Nagoya, Aichi Prefecture Insulator Nihon Honjo Co., Ltd. (72) Inventor Satoshi Morikawa 2-56, Sudamachi, Mizuho-ku, Nagoya, Aichi Issue: Insulator of Nihonhon Co., Ltd.

Claims (3)

[Claims] 1. A ground fault current waveform in which a fault cause is discriminated from a ground fault current waveform of a power line, by roughly classifying whether or not the fault cause is a lightning strike based on the peak value of the final flashover. A method for discriminating a failure aspect, which is characterized by performing discriminant analysis based on spectral analysis values of. 2. A method of determining a failure cause from a ground fault current waveform of a power line, wherein the cause of failure is roughly classified according to the presence or absence of a ground fault current before final flashover, and then a spectrum analysis of the ground fault current waveform is performed. A method for discriminating a failure aspect, which is characterized by performing discriminant analysis based on values. 3. The method for determining a failure aspect according to claim 1, wherein when the ground fault current exceeds a predetermined threshold value, the point where the final flashover occurs is defined and a discriminant analysis is performed.