JPH11223654A - Partial discharge identifier - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform easy identification of a discharge source and estimation of insulation risk. SOLUTION: A partial discharge identifier 30 includes a partial discharge detecting sensor 8 to detect partial discharge from a gas insulating equipment, a BPF(band-pass filter) 32A, a wave detector 33A and a P/H(peak hold circuit) 34A to measure the intensity of a first frequency band of a signal from the partial discharge detecting sensor 8, a BPF 32B, a wave detector 33B and a P/H 34B to measure the intensity of a second frequency band higher than the first frequency band of the signal, and a data analyzer 35 to identify a discharge source of partial discharge in accordance with plural correlation statistic parameters of the intensity of two frequency components or a mutual correlation coefficient between two frequency components.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for diagnosing an insulation abnormality in a power device such as a gas-insulated device, and more particularly to a partial discharge identification device for determining occurrence of a partial discharge and identifying a partial discharge power source. is there.

[0002]

2. Description of the Related Art FIG.
This will be described with reference to FIGS. FIG. 11 is a diagram showing a configuration of a conventional partial discharge identification device.

In FIG. 11, 1 is an aerial wire, 2 is a gas bushing, 3 is a gas-insulated bus, 4 is an instrument transformer, 5 is an arrester, 6 is a high-voltage conductor, 6a is an inner surface of the bus tank, and 7 is an insulating spacer. It is.

FIG. 1 is installed at a power station or a substation.
1 has a structure in which a central conductor (high-voltage conductor 6) to which a high voltage is applied is supported by an insulating spacer 7 and covered by a metal tank.

As shown in FIG. 11, an aerial wire 1 is drawn through a gas bushing 2 and is connected to a gas-insulated bus 3. An instrument transformer 4 and an arrester 5 are connected to the gas-insulated bus 3. These busbars and bushings are separated by an insulating spacer 7 which is a dielectric. The high-voltage conductor 6 is electrically insulated from the tank metal conductor by the insulating spacer 7 and is mechanically held.

In such a gas insulated device, it is considered that detection of partial discharge is effective in order to prevent a dielectric breakdown accident caused by a metal foreign substance present in a metal tank or a crack in the insulating spacer 7 beforehand. ing. Normally, partial discharge occurs as a predictive phenomenon before all-road breakdown affecting the power system is reached. By detecting this partial discharge with high sensitivity and taking appropriate measures, a ground fault can be prevented beforehand. In order to prevent a ground fault, it is necessary to accurately determine that partial discharge has occurred, and to identify the form of partial discharge.

In FIG. 11, reference numeral 8 denotes a partial discharge detection sensor for detecting an electromagnetic wave generated by the discharge;
Reference numeral 0 denotes a conventional partial discharge identification device for identifying a partial discharge from a frequency spectrum disclosed in, for example, Japanese Patent Application Laid-Open No. 3-269274. Reference numeral 20 denotes, for example, Technical Report No. 593 of the Institute of Electrical Engineers of Japan. This is another conventional partial discharge identification device for measuring the phase characteristics of discharge generation described on pages 16 to 19 in June.

Further, in FIG. 11, reference numeral 11 denotes a broadband amplifier for amplifying a signal from the partial discharge detection sensor 8, 12 denotes a frequency spectrum measuring device (spectrum analyzer) for measuring a frequency spectrum of a signal from the broadband amplifier 11, Reference numeral 13 denotes a data analysis device such as a computer. The frequency spectrum is analyzed by the data analyzer 13. The method takes advantage of the fact that the frequency spectrum is different depending on the power source.

Further, in FIG. 1, a signal from the partial discharge detection sensor 8 is analyzed by a computer or the like through an amplifier 21, a band pass filter (BPF _H ) 22, a detector 23, and a peak hold circuit (P / H) 24. The data is analyzed by the device 26. In addition, 25 is an application voltage zero cross trigger signal.

The partial discharge identification device 10 shown in FIG.
It is intended to determine the partial discharge power and noise based on the frequency spectrum measured by the spectrum analyzer 12.
When a partial discharge occurs, an electromagnetic wave is emitted. The frequency spectrum of this electromagnetic wave is based on the fact that it has a characteristic shape for the type of partial discharge power source and the type of noise. That is, it is intended to identify the partial discharge power source from the frequency spectrum measured by the spectrum analyzer 12.

On the other hand, in the partial discharge discriminating apparatus 20 for acquiring the phase characteristic shown in FIG. 11, the peak value of the intensity of a specific partial frequency band of a single partial discharge pulse is held, and the held value is held. To the data analysis device 26
Take in. The peak-held value is taken into the data analyzer 26 together with the phase information, and its shape is analyzed, for example, to identify a partial discharge.

FIG. 12 shows the frequency spectra of air discharge and discharge in GIS (Gas Insulated Switchgear) in order to show the problem of the method of identifying the discharge power source by the frequency spectrum of partial discharge. In the figure, (a) is a partial discharge frequency spectrum in the outside air, and (b) is a GIS.
2 shows a partial discharge frequency spectrum in FIG.

As shown in FIG. 12, in the air discharge, the low frequency component is slightly large, but the frequency at which the spectrum peak occurs is the same for both, and it is not possible to distinguish the discharge source from the form of the spectrum. Sometimes it was difficult. A signal having a large component in a specific frequency band, such as narrow band noise, was easily separated by a frequency spectrum. However, both the external air discharge and the discharge inside the GIS include a broadband spectrum, and there may be no significant difference in the frequency spectrum. In such a case, it is difficult to identify the partial discharge power from the shape of the frequency spectrum. In addition, in order to measure a spectrum, a measurement frequency must be scanned, which causes a problem that the apparatus becomes expensive.

FIG. 13 is a diagram illustrating the problem of the method of identifying the partial discharge power source performed using the phase-maximum charge amount of the discharge performed in the prior art. Maximum discharge charge Q of discharge in air bushing
The phase characteristic of max was shown. The patterns of the maximum discharge charge amount of the discharge in the GIS and the discharge in the air are similar. In the prior art, it was intended to identify whether the discharge was in the GIS tank or in the outside air from the pattern of these data. As can be seen from FIG. 13, the patterns of both phase characteristics are very similar, and in such a case, it is difficult to identify them.

[0015]

As described above, in the conventional partial discharge discriminating apparatus, that is, in the case of a method based on the frequency spectrum or the method based on the phase characteristic of the partial discharge, there are cases where it is difficult to identify the discharge power source. There was a problem.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and has as its object to provide a partial discharge identification device capable of easily identifying a discharge power source and estimating an insulation risk. And

[0017]

According to the present invention, there is provided a partial discharge detecting apparatus according to the present invention, wherein a partial discharge detection sensor for detecting a partial discharge of a power device and a first signal of a signal from the partial discharge detection sensor are provided.
First frequency intensity measuring means for measuring the intensity of the frequency band of the first, second frequency intensity measuring means for measuring the intensity of a second frequency band higher than the first frequency band of the signal, And a data analyzer for identifying the discharge power of the partial discharge based on the statistical parameter of the correlation between the two frequency component intensities from the second frequency intensity measuring means.

Further, in the partial discharge identification device according to the present invention, the first frequency intensity measuring means may output the first frequency intensity of the signal.
And a first peak hold circuit for peak-holding the output of the first band-pass means, and wherein the second frequency intensity measuring means includes And a second peak-hold circuit for peak-holding the output of the second band-pass means.

Further, in the partial discharge identification device according to the present invention, the data analysis device identifies a discharge source of the partial discharge based on a cross-correlation coefficient between two frequency components as the statistical parameter.

Further, in the partial discharge identification device according to the present invention, the data analysis device identifies a discharge source of the partial discharge based on a slope of a regression line of a correlation between two frequency components as the statistical parameter. is there.

Further, in the partial discharge discriminating apparatus according to the present invention, the data analysis apparatus may be configured so that the statistical ratio relates to the intensity ratio of the sum of the intensities of the discharge events in which the intensity ratio between the two frequency components has a specific value. The discharge power of the partial discharge is identified based on the distribution.

Further, in the partial discharge discriminating apparatus according to the present invention, the data analysis apparatus may include a statistical parameter of data between two frequency components for each discharge event within a specific phase range with respect to the applied voltage phase. And the discharge power of the partial discharge is identified by comparing the phases of the statistical parameters.

Further, in the partial discharge identification device according to the present invention, the data analysis device may be configured such that a part of a bandwidth of the first frequency band includes a frequency equal to or lower than a lowest cut-off frequency of electromagnetic wave propagation in the power equipment tank. It is a thing.

[0024]

In the present invention, a frequency band on a frequency spectrum to be detected as a partial discharge signal is selected in advance, the intensity of at least two frequency bands for each discharge pulse is measured, and a plurality of discharge pulses are measured. The generation and discharge power of the discharge are identified from statistical parameters of the correlation between the intensities of the two frequencies.

It is a well-known fact that partial discharges occur in pulses at intervals of several tens nsec to several sec or more depending on the discharge power and applied voltage. The inventor of this patent application
It has been found that the frequency spectrum of each pulsed partial discharge has statistical variation. Conventionally,
It was generally known that the intensity of the partial discharge varied, but it was not known that the frequency spectrum of each discharge pulse varied. It has also been found that this statistical variation in the discharge spectrum varies greatly depending on the form of the discharge power source and the gas pressure.

That is, the frequency spectrum of each partial discharge pulse depends on whether it is a discharge in the GIS or a discharge in the outside air, and whether the discharge power is a needle-like metal foreign matter or a void in the spacer. , Has a statistical feature in its variation.

In order to show a typical example, FIG. 14 shows partial discharges when a discharge power source is installed in a GIS filled with SF6 gas at 4 atm and when a discharge power source is installed in the outside air. The variation in the two-frequency correlation was shown.

In FIG. 14, AL is the intensity of the first frequency band, and AH is the intensity of the second frequency band. In other words, the horizontal axis represents the frequency band of the discharge pulse from 200 MHz to 3
Intensity at 00 MHz, vertical axis is 900 MHz to 1000
It is the intensity in MHz.

The length of the needle-like foreign matter used as the discharge power source is 20
mm, and the applied voltage is 300 kV. Comparing the correlation pattern of the discharge in SF6 gas and the correlation pattern of the discharge in the air, it can be seen that the distribution of the discharge in the air is distributed below the gas discharge, that is, on the low frequency side. Recognize. As described above, it is possible to identify whether the detected signal source is a partial discharge in the GIS or a partial discharge in the outside air from the distribution in the two-frequency correlation.

Embodiment 1 First Embodiment A partial discharge identification device according to a first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing a configuration of a partial discharge identification device according to Embodiment 1 of the present invention. In addition,
In the respective drawings, the same reference numerals indicate the same or corresponding parts.

In FIG. 1, reference numeral 8 denotes a partial discharge detection sensor,
30 is a partial discharge identification device, 31 is an amplifier, and 32A is a first
Band pass filter (BP
F _L ) and 32 B are band pass filters (BPF _H ) that pass the second frequency band. 33A and 33
B is a detector, 34A and 34B are peak hold circuits (P / H), and 35 is a data analyzer such as a computer.

Next, the operation of the first embodiment will be described. A certain discharge event occurs, the signal detected by the partial discharge detection sensor 8 is amplified by the amplifier 31, and the frequency of each signal is determined by the band-pass filters 32A and 32B and the detectors 33A and 33B in two frequency bands set in advance. Converted to component strength.

After that, the detected waveform is peak-held by the peak hold circuits 34A and 34B, and the peak-held data is taken into the data analyzer 35. Each peak hold value data is stored as a data set having two values for one discharge event. The data analysis device 35 captures at least several tens of data and performs statistical processing on the data.

FIG. 2 shows a case where a simulated discharge power source is installed in a GIS, a case where a simulated discharge power source is installed in the vicinity of an overhead line outside the GIS, and a case where a mobile phone is used. Shows the correlation between the intensities in two frequency bands in three cases when the sensor is operated near the partial discharge detection sensor.

The frequencies used in the measurement are 200 MHz to 300 MHz in the first frequency band fL and 900 MHz to 1000 MHz in the second frequency band fH. That is, in FIG. 2, AL is the intensity of the first frequency band fL, A
H is the intensity of the second frequency band fH.

FIG. 3 shows one of the statistical parameters.
The difference of the cross-correlation coefficient shown in the following equation 1 depending on the discharge power is shown.

Ρ = μ _xy / σ _x σ _y Equation 1

Here, σ _x and σ _y are the variances of x and y, respectively. The covariance μ _xy is defined by the following equation 2. The integration range is from 0 (zero) to ∞ (infinity).

[0039] _{μ xy = ∫∫ (x-m} x) (y-m y) f (x, y) dxdy ··· formula 2

X represents the intensity AH of the measured discharge in the first frequency band, y represents the intensity AL of the second frequency band as a variable, and f (x, y) represents the discharge of (x, y). The probability density function that an event occurs. m _x, m _y is x, the average value of y.

Statistically, the correlation coefficient indicates the degree of association between the event x and the event y. It approaches 1 if the two events have a linear correlation, and approaches 0 if they vary.

As can be seen from FIG. 3, the value of the correlation coefficient differs depending on the discharge during the GIS, the discharge in the outside air, and the noise of the mobile phone, and it is possible to identify the signal source from the correlation coefficient. I understand.

FIG. 4 shows the applied voltage dependence of the correlation coefficient in the case of discharge in GIS. It can be seen that the correlation coefficient decreases as the applied voltage increases. This means that the variation in the frequency spectrum for each discharge pulse increases as the applied voltage increases.

That is, the partial discharge identification device according to the first embodiment measures the intensity of at least two frequency bands for each partial discharge pulse of the gas-insulated equipment, and determines the intensity of the two frequency components of the plurality of partial discharge pulses. , That is, the occurrence of a partial discharge and the discharge power are identified from the cross-correlation coefficient between two frequency components.
Note that, as shown in FIG.
The bandpass filter 32A and the detector 33A
And a down converter 41A and a down converter 4A instead of the bandpass filter 32B and the detector 33B.
1B may be used. Downconverters 41A and 4
1B is a general circuit for extracting a signal of a specific frequency component from the input signal. In the partial discharge identification device according to the first embodiment, it is necessary to set the first and second frequencies to be detected to be the optimum frequencies according to the device that detects the partial discharge, the noise environment of the device installation location, and the like. In the down converter, a frequency to be extracted is generally defined by an oscillation frequency of an internal oscillator. Since the oscillation frequency can be controlled by the voltage, the down converter can easily change the frequency to be extracted. In order to make the frequency variable in the configuration example using the bandpass filter, it was necessary to change the element parts. On the other hand, in the configuration example using the down converter, the frequency band can be changed only by changing the voltage for controlling the internal oscillation frequency, and the frequency can be easily changed.

Embodiment 2 Second Embodiment A partial discharge identification device according to a second embodiment of the present invention will be described with reference to FIG. FIG. 5 is a diagram showing a slope of a regression line due to a difference between signal sources in a two-frequency correlation by the partial discharge identification device according to the second embodiment of the present invention. The hardware configuration is the same as in the first embodiment.

In the second embodiment, a method will be described in which the slope of the regression line is used as a parameter for occurrence of partial discharge and identification of discharge power. In the second embodiment, a regression line of the two-frequency correlation data is derived, and a signal detected based on whether the slope value of the line is within a predetermined range is a partial discharge or a partial discharge power source. It is characterized by determining the form.

As an example, FIG. 5 shows the slopes of regression lines calculated for three types of discharge power sources of the data shown in FIG. Since the mobile phone noise has almost no low frequency components, the slope of the regression line is very large. Since the frequency spectrum of the discharge in the GIS has a wide band, the discharge has a predetermined intensity in both frequency bands, and the slope of the regression line in the two-frequency correlation has a specific value. On the other hand, in the case of a discharge in the outside air, since the high-frequency component is slightly smaller than the discharge in the GIS, the gradient is smaller than the gradient of the discharge in the GIS. As described above, it can be determined from the slope of the regression line whether the signal is noise or what kind of partial discharge power source it is.

That is, the partial discharge discriminating apparatus according to the second embodiment uses the slope or intercept of the regression line of the correlation between two frequency components as a statistical parameter.

Embodiment 3 Third Embodiment A partial discharge identification device according to a third embodiment of the present invention will be described with reference to FIG. FIG. 6 is a diagram illustrating the angle dependence of the total discharge due to the difference in the signal source in the two-frequency correlation by the partial discharge identification device according to the third embodiment of the present invention. The hardware configuration is the same as in the first embodiment.

In the third embodiment, the characteristics of the discharge power source are to be extracted from the angular distribution of the total discharge intensity.

As shown in FIG. 6A, the angle formed by the intensity vector (AL, AH) of the discharge pulse having a two-frequency correlation is θ
Then, the sum of the intensities of the discharge events existing in θ and θ + dθ is defined by the following equation 3. The range of Σ is i to
k.

[0052] f (θ) = Σ√ {( A L, i) 2 + (A H, i) 2} ··· Equation 3

Here, k is the total number of discharge events within the angles θ and θ + dθ. θ means the intensity ratio of two frequencies.

FIG. 6B shows f (θ) as a function of θ. The characteristics of the discharge in the GIS and the discharge in the air are different, indicating that the discharge power source can be identified. The physical meaning of this function is the distribution of the frequency spectrum.

That is, the partial discharge discriminating apparatus according to the third embodiment uses, as a statistical parameter, a distribution relating to the intensity ratio of the sum of the intensities of the discharge events in which the intensity ratio between two frequency components has a specific value. It is.

Embodiment 4 Embodiment 4 A partial discharge identification device according to Embodiment 4 of the present invention will be described with reference to FIGS. FIG. 7 is a diagram showing data in two-frequency correlation separated by a partial discharge identification device according to Embodiment 4 of the present invention into a positive phase and a negative phase of a power supply phase. FIG. 8 is a diagram illustrating the applied voltage dependence of the negative phase correlation coefficient and the positive phase correlation coefficient by the partial discharge identification device according to the fourth embodiment of the present invention. The hardware configuration is the same as in the first embodiment.

In the fourth embodiment, the correlation of the two frequency bands is plotted by separating the positive-phase discharge and the negative-phase discharge. As shown in FIG. 7, regarding the discharge in the GIS, the discharge in the positive phase is distributed on the lower frequency side than the distribution of the discharge in the negative phase. Also, the negative phase discharge has less variation. It is understood that the occurrence of discharge in the GIS can be identified by using such a characteristic statistical distribution in the GIS as a criterion.

FIG. 8 is a graph in which the applied voltage dependence of the correlation coefficient is plotted separately for a positive phase and a negative phase. As for the correlation coefficient, only the positive-phase discharge decreases as the applied voltage increases. As described above, by performing the statistical analysis by separating the phases, the characteristics of the discharge in the GIS are further emphasized, and the accuracy of the identification of the discharge power source is improved.

That is, the partial discharge discriminating apparatus according to the fourth embodiment uses a statistical parameter of data between two frequency components for each discharge event within a specific phase range with respect to the applied voltage phase. The partial discharge power source is identified by comparing the phases of the statistical parameters.

Embodiment 5 FIG. Embodiment 5 A partial discharge identification device according to Embodiment 5 of the present invention will be described with reference to FIGS. 9 and 10. FIG. The hardware configuration is the same as in the first embodiment.

FIG. 9 is a correlation diagram of two frequencies in a case where a component below the lowest cut-off frequency determined by the shape of the GIS tank is included as the lower frequency band of the two frequency bands for obtaining the correlation.

By using a frequency lower than the lowest cut-off frequency, it is possible to know whether the discharge power is on the tank side or on the center conductor side.

In general, the minimum cutoff frequency is given by the following equation (4), where a is the outer diameter of the GIS center conductor and b is the inner diameter of the GIS tank. Here, c is the speed of light.

Fc = 2c / π (a + b) Equation 4

For example, the center conductor diameter of GIS is 130 m
m, when the tank inner diameter is 550 mm, the minimum cutoff frequency is 280 MHz.

Therefore, as a correlation between two frequencies, the first frequency band is set to include components equal to or lower than the cutoff frequency.
MHz to 300 MHz is selected, and 900 MHz to 1000 MHz is selected as the second frequency. When the frequency is selected in this way, it is possible to effectively discriminate between a case where the foreign matter is on the tank bottom surface and a case where the foreign matter is on the center conductor side.

FIG. 9 shows the correlation when the needle-shaped metal is installed on the center conductor side and the tank bottom side in the GIS. Also,
In FIG. 10, the frequency band on the low frequency side is set to 40 for comparison.
The case where the frequency is set to 0 MHz to 500 MHz is shown.

The difference in correlation between the case where the discharge power source is installed on the center conductor side and the case where the discharge power source is installed on the tank side is the first difference shown in FIG.
It can be understood that the frequency band is larger when the frequency band is below the cutoff frequency.

The effect of the fifth embodiment is that the difference in the position of the discharge power source in the radial direction of the tank can be easily detected in a two-frequency correlation pattern.

That is, in the partial discharge discriminating apparatus according to the fifth embodiment, of the two frequencies, a part of the bandwidth on the low frequency side includes frequencies lower than the lowest cutoff frequency of electromagnetic wave propagation in the GIS tank.

In each of the embodiments described above, the gas insulated equipment has been described. In each of these embodiments, it is necessary to select an optimum frequency band. However, it is necessary to select a cable, a transformer, a generator and the like. The present invention can also be applied as a partial discharge identification device in another device in which discharge may occur.

[0072]

As described above, the partial discharge identification device according to the present invention includes a partial discharge detection sensor for detecting a partial discharge of a power device, and an intensity of a signal from the partial discharge detection sensor in a first frequency band. Frequency intensity measuring means for measuring the intensity of the signal, second frequency intensity measuring means for measuring the intensity of a second frequency band higher than the first frequency band of the signal, and the first and second frequencies A data analyzer for identifying the discharge power of the partial discharge based on the statistical parameter of the correlation of the intensity of the two frequency components from the intensity measurement means, so that the discharge power can be easily identified, and the insulation risk can be reduced. This has the effect that it can be estimated.

Further, as described above, in the partial discharge identification device according to the present invention, the first frequency intensity measuring means includes a first band pass means for passing a first frequency band of the signal; A first peak hold circuit for peak-holding an output of the first band-pass means, wherein the second frequency intensity measuring means passes the second frequency band of the signal. And a second peak hold circuit for peak-holding the output of the second band-pass means, so that it is possible to easily identify the discharge power source and to estimate the insulation risk. .

Further, in the partial discharge identification device according to the present invention, as described above, the data analysis device identifies the discharge source of the partial discharge based on the cross-correlation coefficient between two frequency components as the statistical parameter. Therefore, it is possible to easily identify the discharge power source and to estimate the insulation risk.

Further, as described above, in the partial discharge discriminating apparatus according to the present invention, the data analysis apparatus may be configured such that the data analysis apparatus uses the discharge power of partial discharge based on the slope of the regression line of the correlation between the two frequency components as the statistical parameter. Therefore, the discharge power source can be easily identified, and the degree of insulation risk can be estimated.

Further, as described above, in the partial discharge discriminating apparatus according to the present invention, the data analysis apparatus determines that the intensity ratio between the two frequency components has a specific value as the statistical parameter. Since the discharge power source of the partial discharge is identified based on the distribution related to the intensity ratio of the sum, it is possible to easily identify the discharge power source and to estimate the insulation risk.

Further, as described above, in the partial discharge discriminating apparatus according to the present invention, the data analyzing apparatus may be configured such that the data analyzing apparatus sets the frequency between two frequency components for each discharge event within a specific phase range with respect to the applied voltage phase. Using the statistical parameters of the data, the discharge power of the partial discharge is identified by comparing the phases of the statistical parameters, so that the discharge power can be easily identified, and the effect that the insulation risk can be estimated can be obtained. Play.

Further, as described above, in the partial discharge identification device according to the present invention, the data analysis device may be configured such that a part of the bandwidth of the first frequency band is a minimum cut-off frequency of electromagnetic wave propagation in the power equipment tank. Since the following frequencies are included, it is possible to easily identify the discharge power source and to estimate the insulation risk.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a partial discharge identification device according to Embodiment 1 of the present invention.

FIG. 2 is a diagram showing two-frequency correlation data of various signal sources according to the first embodiment of the present invention.

FIG. 3 is a diagram showing a discharge power dependency of a correlation coefficient according to the first embodiment of the present invention.

FIG. 4 is a diagram showing an applied voltage dependency of a correlation coefficient according to the first embodiment of the present invention.

FIG. 5 is a diagram showing data indicating that a discharge power source can be identified from a slope of a regression line of a two-frequency correlation according to the second embodiment of the present invention.

FIG. 6 is a diagram showing an angle dependence of a discharge sum in two-frequency correlation according to Embodiment 3 of the present invention.

FIG. 7 is a diagram showing data obtained by separating data in a two-frequency correlation into a positive phase and a negative phase of a power supply phase according to Embodiment 4 of the present invention.

FIG. 8 is a diagram showing the applied voltage dependence of each of the negative phase and positive phase correlation coefficients according to Embodiment 4 of the present invention.

FIG. 9 is a diagram showing two-frequency correlation data when 200 to 300 MHz is selected as the low frequency band fL according to the fifth embodiment of the present invention.

FIG. 10 is a diagram illustrating two-frequency correlation data when 300 to 400 MHz is selected as a low-frequency band fL according to Embodiment 5 of the present invention.

FIG. 11 is a diagram showing a configuration of a gas insulation device and a conventional partial discharge identification device.

FIG. 12 is a comparison diagram of partial discharge frequency spectra in air and gas.

FIG. 13 is a diagram showing phase characteristics of the maximum discharge charge amount of partial discharges in air and gas.

FIG. 14 is a diagram showing data of two-frequency correlation between air discharge during GIS and air discharge outside GIS.

FIG. 15 is a block diagram showing another configuration of the partial discharge identification device according to Embodiment 1 of the present invention.

[Explanation of symbols]

Reference Signs List 8 partial discharge detection sensor, 30, 40 partial discharge identification device, 31 amplifier, 32A bandpass filter (BPF _L ) that passes first frequency band, 32B bandpass filter (BPF _H ) that passes second frequency band , 3
3A, 33B detector, 34A, 34B Peak hold circuit (P / H), 35 Data analyzer, 41A, 41
B Down converter.

Continued on the front page (72) Inventor Keiichi Ito 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsubishi Electric Corporation

Claims (7)

[Claims] 1. A partial discharge detection sensor that detects a partial discharge of a power device, a first frequency intensity measurement unit that measures an intensity of a signal from the partial discharge detection sensor in a first frequency band, A second frequency intensity measuring unit for measuring the intensity of a second frequency band higher than the first frequency band; and a statistical parameter of a correlation between two frequency component intensities from the first and second frequency intensity measuring units. And a data analysis device for identifying a discharge power source of the partial discharge based on the data. 2. The first frequency intensity measuring means includes a first band-pass means for passing a first frequency band of the signal, and a first band-pass means for peak-holding an output of the first band-pass means. A peak hold circuit, wherein the second frequency intensity measuring means includes a second band-pass means for passing a second frequency band of the signal;
And a second peak hold circuit for peak-holding the output of said band-pass means. 3. The partial discharge identification device according to claim 2, wherein the data analysis device identifies a discharge source of the partial discharge based on a cross-correlation coefficient between two frequency components as the statistical parameter. . 4. The partial discharge according to claim 2, wherein the data analysis device identifies a discharge source of the partial discharge based on a slope of a regression line of a correlation between two frequency components as the statistical parameter. Identification device. 5. The data analysis device according to claim 1, wherein the discharge power of the partial discharge is based on a distribution relating to an intensity ratio of a sum of intensities of discharge events in which an intensity ratio between two frequency components is a specific value as the statistical parameter. 3. The partial discharge identification device according to claim 2, wherein the identification is performed. 6. The data analysis apparatus according to claim 1, wherein each of the discharge events in a certain specific phase range with respect to the applied voltage phase is performed.
Using statistical parameters of data between two frequency components,
3. The partial discharge identification device according to claim 2, wherein a discharge power of the partial discharge is identified by comparing the phases of the statistical parameters. 7. The data analysis device according to claim 2, wherein a part of the bandwidth of the first frequency band includes a frequency equal to or lower than a lowest cut-off frequency of electromagnetic wave propagation in the power equipment tank. Partial discharge identification device.