JPH09171050A - Insulation diagnostic apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an insulation diagnostic apparatus by which various factors required for the insulation diagnosis of an electric apparatus can be set automatically. SOLUTION: Before the insulation abnormality of an electric apparatus is diagnosed, a signal which detects an antenna for noise detection is scanned at an initial frequency point. On the basis of small-level data out of various data, a low-noise-level judgment value is set (S1 to S7). By using the judgment value, a frequency point which is used for an actual insulation diagnosis is set. Then, a signal which detects an antenna for a signal is scanned at the set frequency point. On the basis of small-level data out of various data, a threshold value which is used to judge an insulation abnormality is set. Then, by using the low-level judgment value and the threshold value, the insulation abnormality of the electric apparatus is diagnosed. Since the respective factors contain the gain of an antenna, the environment of a noise and the like, the accuracy of a diagnosis can be increased.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an insulation diagnosis device for detecting whether or not there is an insulation abnormality in a power electric device such as a gas insulation switchgear, a gas circuit breaker, and a cubicle.

[0002]

2. Description of the Related Art FIG. 1 shows a circuit configuration of an insulation diagnostic device. In the electric device 1, when a partial discharge that is a precursor of an arc flash occurs, electromagnetic waves are radiated to the outside due to high-frequency potential vibration when the partial discharge occurs. At the time of insulation diagnosis, the electromagnetic wave radiated from the electric device 1 is detected by the signal antenna 2. The detected signal contains a wide range of frequency components. This signal is scanned by the tuner 7,
The data is C for each of the plurality of preselected frequency points.
It is input to the PU 10. The CPU 10 compares the levels of a plurality of data with a threshold value to diagnose the insulation inside the electric device 1.

Here, the signal actually detected by the signal antenna 2 includes, in addition to an electromagnetic wave generated inside the electric device 1, an external noise of a specific frequency such as a broadcast wave. This noise causes a decrease in detection sensitivity.
Therefore, it has been proposed to measure external noise with the noise detection antenna 3 before diagnosis and automatically set each frequency point to a noise-free frequency such as a broadcast wave.

In the conventional frequency point setting method, first, the diagnostic frequency is evenly divided by the number of diagnostic points, and the initial diagnostic points f ₁ ... F _n (for example, 400 MH).
z, 405, 410, 415, ...) are determined. The signal detected by the noise detecting antenna 3 is scanned by the tuner 7 at each initial diagnosis point. The signal level obtained by this is as shown in FIG. next,
For frequency points that exceed the preset low noise level determination value BGN (marked with ● in the figure), it is assumed that noise is included, and the frequency points are Δ
It is shifted by f (f _n = f _n + Δf), and the signal level at that frequency is measured again. Then, when the signal level becomes equal to or lower than the low noise level determination value BGN, the frequency is set as a diagnostic frequency point.

If the signal level does not fall below the BGN level even after shifting the diagnostic frequency a predetermined number of times, the frequency point is regarded as an invalid point and is not used for subsequent diagnostics. In FIG. 7B, the frequencies not used are marked with a cross. After setting a predetermined number of diagnostic frequency points in this way, the changeover switch 5 of the insulation diagnostic device 4 is switched to the signal antenna 2 side. Here, since the frequency point of the low noise level is set as the diagnostic frequency point as described above, the insulation diagnostic device 4 can detect the electromagnetic wave inside the electric device 1 with high sensitivity, and thus the diagnostic error will be erroneous. There is no.

[0006]

In the above-mentioned conventional insulation diagnosis apparatus, it is difficult to set the low noise level judgment value BGN. As shown in FIG. 8A, when the surrounding noise level is higher than the set value or the gain of the noise detecting antenna used is higher than the average,
The overall signal level becomes high, and even if the diagnostic frequency point automatic setting process is performed, it is not possible to find a frequency point that is lower than the low noise level determination value BGN as shown in FIG. 8B.

On the contrary, if the surrounding noise level is low or the gain of the noise detecting antenna used is smaller than the average, the point where noise actually exists may be set as the diagnostic frequency point. Become.
In this case, the detection sensitivity of the insulation diagnosis device is reduced. Furthermore, the above-mentioned problems also occur due to changes in the frequency band subject to insulation diagnosis. That is,
When the frequency f is low, the low noise level is high, and the frequency f
The low noise level is generally low when is high.
Therefore, the low noise level determination value BGN must be adjusted depending on the frequency band used for diagnosis, but this was also difficult.

Further, in the conventional insulation diagnosis apparatus, it is difficult to set the abnormality determination level used when performing insulation diagnosis. The gain of the signal antenna for detecting partial discharge varies depending on its size, structure, mounting location, and target supply frequency band. Therefore, it is necessary to set the threshold level according to the gain of the antenna, but it is difficult. An object of the present invention is to automatically set various coefficients necessary for diagnosis in a device for diagnosing insulation of electric equipment.

[0009]

In order to achieve the above object, the present invention provides a noise detecting device for diagnosing insulation of an electric device by detecting an electromagnetic wave generated by a partial discharge generated inside the electric device. A means for obtaining a plurality of data at a plurality of frequency points from the signal detected by the antenna for use, and a low noise level judgment value used for searching for a noise-free frequency point based on the data having a small level among the plurality of data. Provide a means to set automatically.

Since such a low noise level judgment value can be set automatically, it can be easily set at the site where the electric equipment is installed. Further, since the level is set according to the antenna gain, the noise environment, etc., when searching for a diagnostic frequency point described below, it is possible to reliably search for a frequency point that is not affected by noise. Therefore, it is not necessary to perform a test of the antenna gain or the like in the field in advance.

Further, according to the present invention, in an apparatus for diagnosing insulation of an electric device by detecting an electromagnetic wave generated by a partial discharge generated inside the electric device, a plurality of frequency points are detected from a signal detected by a signal antenna. It is possible to provide a means for obtaining a plurality of data in 1) and a means for automatically setting an abnormality determination threshold value used for diagnosing insulation of an electric device based on the data of a low level among the plurality of data.

Since the threshold value for such insulation abnormality can be set automatically, it can be easily set at the site where the electric device 1 is installed. Also, since the threshold value is set according to the antenna gain, noise environment, etc., it is possible to set the optimum level even if the antenna gain changes depending on the installation location of electric equipment, frequency band, etc. The diagnostic accuracy can be improved. Therefore, in advance,
It is no longer necessary to test the antenna gain on site.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows a circuit configuration of an insulation diagnostic device. In the figure, 1 is an electric device. A signal antenna 2 detects a signal radiated from the electric device 1. Reference numeral 3 denotes a noise detection antenna, which is arranged at a place where a signal radiated from the electric device 1 is not detected and where the signal antenna 2 has substantially the same conditions. The output signals of the signal antenna 2 and the noise detection antenna 3 are input to the insulation diagnostic device 4.

In the insulation diagnostic device 4, the signals of the signal antenna 2 and the noise detection antenna 3 are changed over to a switch 5.
Is selected by switching and input to the preamplifier 6. The signal amplified by the preamplifier 6 is scanned by the tuner 7 for each predetermined frequency point. Tuner 7
The frequency points to be scanned by are set by the CPU 10. The output of the tuner 7 is input to the CPU 10 via the detection circuit 8 and the A / D converter 9.

The CPU 10 automatically sets a low noise level determination value based on an input signal, determines a frequency point to be scanned by the tuner 7, automatically sets a threshold value for electrical equipment insulation abnormality determination, and electrical equipment insulation abnormality. Perform automatic diagnosis of. Insulation diagnostic device 4 configured as described above
In (1), before the insulation diagnosis of the electric device 1 is started, the changeover switch 5 is switched to the noise detection antenna 2 side, and the CPU 10 sets the diagnosis frequency point to be scanned by the tuner 7. For this procedure,
This will be described with reference to the flowcharts of FIGS. 2 and 3 and the graph of FIG.

In step S1, the frequency points are first initialized. This initial frequency point is determined by evenly dividing the monitoring frequency band by the number of diagnostic points. For example, when setting 40 diagnosis points in the frequency band of 400 to 800 MHz, 410, 420,
430 ... 800 MHz. In step S2, the tuner 7 scans to obtain data of the received signal level at each set frequency point. FIG. 4A shows this data arranged for each frequency point.

In step S3, the respective data are rearranged in order of increasing level. In step S4, the n _b -th data level from the smallest level is set to the initial BGN level =
BGN ₀ . The value of n _b is, for example,
It can be an appropriate value of about 10% to 20% of the total number of data. In addition, the value of BGN ₀ is an average value such as a simple mean or a square mean of 0 to 10% of data, a -3σ point of a normal distribution is taken, or a span between the minimum value and the maximum value is taken. Can be set to a predetermined ratio.

At step S5, the margin ΔP is added to this value.
By adding the _BGN, low noise level judgment value BGNL (= BG
N ₀ + ΔP _BGN ). In step S6, it is determined whether the low noise level determination value BGNL is smaller than the upper limit value BGNL (MAX). In addition, this upper limit value BGNL
(MAX) is a value necessary for determining whether or not the abnormal oscillation occurs when the amplifier abnormally oscillates due to an abnormal input or the like and the output level becomes an excessive level over the entire frequency. Here, the determination value BGNL is the upper limit value BG
If it is smaller than NL (MAX), the amplifier is operating normally, and its value is directly adopted as the low noise level determination value BGN. In addition, the determination value BGNL is the upper limit value BGNL.
If it is larger than (MAX), the process proceeds to step S7, and the upper limit value BGNL (MAX) is the low noise level determination value BGN.
It is adopted as L.

Since such a method of setting the low noise level determination value can be automatically performed, it can be easily performed at the site where the electric device 1 is installed. Also,
Since the level is set according to the antenna gain, the noise environment, etc., it is not necessary to perform an antenna gain test or the like beforehand on site. Therefore, when searching for a diagnostic frequency point described below, it is possible to reliably search for a frequency point that is not affected by noise.

When the low noise level determination value BGNL is determined as described above, the process proceeds to step S11 and subsequent steps in FIG. 3 to set the diagnostic frequency point. In addition,
The processing from step S11 onward is an example of what is also executed in the conventional insulation diagnosis device.
Of course, other methods can be used. Step S
At 11 and 12, counters m and n are set to 1, respectively.

In step S13, it is determined whether or not the level P _{n of the nth} data is smaller than the low noise level determination value BGNL. Here, if it is smaller (◯ in FIG. 4A), the n-th frequency point is set as an effective point in step S14. If it is larger (● in FIG. 4A), the frequency fn is set to fn + Δf in step S15.
shift to n.

By passing through step S16 to step S17, the processing of the above steps S13 to 15 is repeated N times. This N times is the maximum value of n of f _n , and means that the above process is repeated for all the frequency points in FIG. 4, and all frequency points f _{1 to} f
_When the process for _n is completed, the process proceeds to step S18.

The processing of steps S13 to S16 is repeated by passing through steps S18 to S19. As a result, with respect to the frequency point for which P _n > BGNL is determined in step S13 in the previous process, the determination in step S13 is performed with the frequency shifted by Δfn. By repeating this, frequency points at which the data level becomes equal to or lower than the noise level are set.

When the value of the counter m is 5 in step S18 or the number n _{e which} is determined to be a valid frequency point in step S14 is 30, this process ends. Here, m = 5 means that shifting the frequency by Δf was repeated 5 times. In this way, when the data level does not become lower than the noise level even after being repeated 5 times, the frequency point of the data is invalidated and is not used for the diagnosis of insulation abnormality. Also, n _e = 3
In the case of 0, it means that the process ends when 30 valid frequency points are obtained before the process is repeated 5 times. The numerical value of 5 or 30 can be arbitrarily changed.

As described above, the low noise level judgment value BGN
Is automatically set and the diagnostic frequency points f _{1 to} f _n are automatically set, the signal level at the diagnostic frequency points is, as shown in FIG. 4B, a low noise level determination value BGN or less. Become. Next, the changeover switch 5 is switched to the signal antenna 2 side, and the threshold value for determining the presence or absence of insulation abnormality of the electric device 1 is automatically set in the procedure described below. At this time, the electric device 1
May be running or stopped. The procedure will be described with reference to the flowchart of FIG. 5 and the graph of FIG.

In step S21, the diagnostic frequency points f1 to fn are first set in the tuner 7. In step S22, the tuner 7 scans to obtain received signal level data at each diagnostic frequency point. FIG. 6 shows this data arranged for each frequency.

In step S23, each data is rearranged in order of increasing level. In step S24, the level of the n _s- th data from the smallest level is set to the initial threshold value AJ.
Let L ₀ . The value of n _s is determined almost in the same manner as n _b described above. At step S25, the margin ΔPj is added to this value to obtain the final threshold value AJL = AJL.
₀ + Pj.

At step S26, the threshold value AJL is set to the upper limit value A.
It is determined whether or not it is smaller than JL (MAX). In addition,
This upper limit value AJL (MAX) is the above-mentioned upper limit value BGNL.
It has the same meaning as (MAZ). Here, if the threshold value AJL is smaller than the upper limit value AJL (MAX), that value is directly adopted as the threshold value AJL. If the threshold value AJL is larger than the upper limit value AJL (MAX), the process proceeds to step S7, and the upper limit value AJL (MAX) is adopted as the low noise level determination value AJL.

Since the method of setting the threshold value used for determining the insulation abnormality can be automatically performed, the electric device 1
Can be easily done at the site where is installed. Therefore, it is not necessary to perform a test of the antenna gain or the like in the field in advance. In addition, even if the antenna gain changes depending on the installation location and frequency band of electrical equipment, the threshold level is set according to the antenna gain and noise environment, etc., so that the optimum level can be set. The accuracy can be increased.

When the diagnosis frequency point and the abnormality determination threshold value are set as described above, the actual insulation diagnosis is executed. The tuner 7 scans the diagnostic frequency point while the changeover switch 5 is switched to the signal antenna 2 side. Thereby, the electromagnetic wave radiated by the electric device 1 is detected for each diagnostic frequency point. CPU10
Calculates the average level increase value and compares it with a threshold value to determine the level, and based on the result, diagnoses whether or not an insulation abnormality has occurred inside the electric device 1. Since this specific method is known, only a brief description is given here.

[0031]

According to the present invention, in an apparatus for diagnosing insulation of electric equipment, various coefficients required for diagnosis can be automatically set.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of an insulation diagnostic device according to the present invention and a conventional one.

FIG. 2 is a flowchart for explaining a method of setting a low noise level determination value according to the present invention.

FIG. 3 is a flowchart for explaining a frequency point setting method according to the present invention.

FIG. 4 is a graph for explaining a method of setting a low noise level determination value BGN according to the present invention.

FIG. 5 is a flowchart for explaining a method of setting a threshold value for insulation abnormality determination according to the present invention.

FIG. 6 is a graph for explaining a method of setting a threshold value for insulation abnormality determination according to the present invention.

FIG. 7 is a graph (No. 1) for explaining the low noise level determination value BGN in the conventional insulation diagnostic device.

FIG. 8 is a graph (No. 2) for explaining the low noise level determination value BGN in the conventional insulation diagnostic device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Electrical equipment 2 ... Signal antenna 3 ... Noise detection antenna 4 ... Insulation diagnostic device 5 ... Changeover switch 6 ... Preamplifier 7 ... Tuner 8 ... Detection circuit 9 ... A / D converter 10 ... CPU

Claims (2)

[Claims] 1. An apparatus for diagnosing insulation of an electric device by detecting an electromagnetic wave generated by a partial discharge generated inside an electric device, wherein a plurality of frequency points at a plurality of frequency points are detected from a signal detected by a noise detecting antenna. A means for obtaining data, and means for automatically setting a low noise level judgment value used to search for a noise-free frequency point based on data having a low level among the plurality of data. Insulation diagnostic device. 2. An apparatus for diagnosing insulation of an electric device by detecting an electromagnetic wave generated by a partial discharge generated inside the electric device, wherein a plurality of data at a plurality of frequency points are detected from a signal detected by a signal antenna. And a means for automatically setting an abnormality determination threshold value used for diagnosing insulation of the electric device based on data of a low level among the plurality of data. Insulation diagnostic device.