JP2008051708A - Partial discharge monitoring apparatus and method of gas-insulated electric apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To acquire a partial discharge monitoring apparatus and a partial discharge monitoring method for quickly and accurately determining a cause of a generated partial discharge. <P>SOLUTION: A plurality of detectors 5 are installed in a metallic vessel 4 of a gas-insulated electric apparatus, and detect partially-discharged electromagnetic waves. When a strength of detection signals from the detectors 5 exceeds a predetermined threshold, a comparator 6 generates a pulse, and a counter 10 counts the number of the pulses at a regular interval. A calculation section 12 determines a location at which the partial discharge is generated from the number (n) of the measured pulses, calculates a virtual threshold level (Q<SB>T</SB>) of the comparator to each detector in the vicinity of the determined location at which the partial discharge is generated by using an electromagnetic transfer function between the detectors, creates a ϕ (a voltage phase)-n (the number of the measured pulses)-Q (a strength of the electromagnetic waves) distribution pattern by using the virtual threshold level (Q<SB>T</SB>), and determines the cause of the generated partial discharge by comparing it with a previously-stored distribution pattern of each cause of the generated partial discharge. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a partial discharge monitoring device and a partial discharge monitoring method for monitoring an insulation abnormality in a gas-insulated electrical device in which a high voltage conductor is insulated and supported in a metal container filled with an insulating gas.

A conventional partial discharge monitoring device of a gas-insulated electric device is composed of a plurality of partial discharge electromagnetic wave detectors installed at a distance in a metal container, and an abnormal position locating device for locating the occurrence position of the partial discharge, Based on the electromagnetic wave attenuation for each circuit component of the gas-insulated power device stored in advance in the storage unit, the mounting position information of each detector, and the electromagnetic waves detected by each detector, the partial discharge occurrence location is determined. The result is displayed (see, for example, Patent Document 1).
Further, as a conventional method for investigating the cause of the insulation abnormality of the power equipment, a three-dimensional structure composed of a commercial frequency voltage phase (φ), a discharge pulse magnitude (q), and a number (n) generated by a partial discharge caused by the insulation abnormality. A method of classifying partial discharges based on a distribution, so-called φ-qn distribution, is used (see, for example, Non-Patent Document 1).
In order to apply this method, a high-speed discharge signal is converted into a damped oscillation wave by a CR integration circuit or an LC tuning circuit, the detected waveform is peak-held, and then it corresponds to the size of the partial discharge by A / D conversion. To obtain a digital value (see, for example, Non-Patent Document 2)

JP-A-3-259756 E. Gulski et. al. , “PD Quantities for Evaluation of PD Measurements in GIS”, Record of the CIGRE WG 15.03 Conference, IWD 46 (1997) (FIGS. 22-28). IEEJ Technical Report No. 593 “Computer Application Technology for Insulation Measurement” IEEJ (1996) (p.14, right 9th to right 15th, FIG. 3.1)

In the above-described conventional method for investigating the cause of the insulation abnormality of power equipment, when detecting a partial discharge generated in a gas-insulated electric device, a radiated electromagnetic wave in the UHF band where S / N is easily obtained is detected. In order to obtain intensity information of a high frequency signal, it is necessary to perform detection, peak hold, and A / D conversion by a circuit having excellent high frequency characteristics, which causes a problem that the circuit becomes complicated.
In addition, in the conventional partial discharge monitoring device, when determining the location of the partial discharge, information on the electromagnetic wave attenuation for each circuit component is used. Since the partial discharge electromagnetic wave of the above has a high frequency and resonates in the metal container, the sensitivity of the detector also changes depending on the positional relationship between the metal container and the detector. There was a problem that the accuracy of locating the discharge occurrence location was not sufficient.

  The present invention has been made to solve the above problems, and provides a partial discharge monitoring apparatus and a partial discharge monitoring method capable of determining the cause of partial discharge at high speed and with high accuracy. It is an object.

A partial discharge monitoring device for a gas insulated electrical device according to the present invention is installed in a metal container of a gas insulated electrical device, connected to a plurality of detectors for detecting partial discharge electromagnetic waves, and the plurality of detectors, A pulse generating means for generating a pulse when a detection signal intensity of the detector exceeds a predetermined threshold; a pulse counting means for counting the number of the pulses every predetermined time; an installation position of each of the detectors; A storage unit for storing an electromagnetic wave transfer function between the detectors and a partial discharge pattern for each partial discharge occurrence created in advance, and a calculation unit, and the calculation unit is counted by the pulse counting means. The partial discharge occurrence location is determined from the number of measured pulses, and an electromagnetic wave transfer function between the detectors stored in the storage unit is used to determine each detector in the vicinity of the determined partial discharge occurrence location. Calculate the virtual threshold level of the pulse generating means (Q _T), since the calculated the virtual threshold level (Q _T), each detector pulse measurement number for the commercial frequency voltage phase (phi) (n), and power frequency A partial discharge pattern showing a distribution of voltage phase (φ) −number of pulse measurements (n) −electromagnetic wave intensity (Q) is created, and the partial discharge pattern and a portion for each partial discharge occurrence cause stored in advance in the storage unit The cause of partial discharge is determined by comparing the discharge pattern.

Further, the partial discharge monitoring method for a gas-insulated electrical apparatus according to the present invention provides each of the above-described methods when the output intensity of a plurality of partial-discharge electromagnetic wave detectors installed in a metal container of the gas-insulated electrical apparatus exceeds a predetermined threshold. A first step of generating a pulse by means of pulse generation means connected to the detector, a second step of counting the pulse every predetermined time, and a location where a partial discharge is generated from the number of measured pulses measured for each detector A third step of determining the virtual threshold level (Q _T ) of the pulse generating means for each detector in the vicinity of the determined partial discharge generation location using the electromagnetic wave transfer function between the detectors acquired in advance. From the calculated fourth threshold value, the calculated virtual threshold level (Q _T ), and the pulse measurement number (n) of each detector for the commercial frequency voltage phase (φ), the commercial frequency voltage phase (φ) −pulse The fifth step of creating a partial discharge pattern showing the distribution of measured number (n) -electromagnetic wave intensity (Q), and comparing the created partial discharge pattern with a partial discharge pattern for each partial discharge occurrence cause acquired in advance. Thus, a sixth step of determining the cause of occurrence of partial discharge is provided.

  According to this invention, when the partial discharge electromagnetic wave exceeds a preset threshold value, a pulse is generated by the pulse generating means, and this pulse is counted to determine the location and cause of the partial discharge. Despite being able to simplify the circuit configuration up to the part, the threshold level of multiple comparators is corrected based on the signal propagation transfer function in the gas-insulated electrical device to obtain electromagnetic wave intensity information by secondary partial discharge. Therefore, there is an effect that the cause of the partial discharge can be determined with high speed and high accuracy.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a partial discharge monitoring device for a gas-insulated electric device according to Embodiment 1 of the present invention. In the gas-insulated electrical apparatus 1, the conductor 2 is supported by an insulating support called a spacer 3 in a metal container 4, and an insulating gas 33 is sealed in the metal container inside 4.
In such a gas-insulated electrical apparatus 1, a partial discharge 34 is generated at the time of an insulation abnormality due to mixing of foreign matter into the metal container 4 or insulation deterioration of the spacer 3, and an electromagnetic wave 35 is radiated accordingly. Although the electromagnetic wave 35 is detected by the detector 5 installed on the wall surface of the metal container 4, the electromagnetic wave 35 is attenuated as it propagates inside the metal container, and therefore the electromagnetic wave is generated no matter where the insulation abnormality occurs in the gas-insulated electrical apparatus 1. A plurality of detectors 5 are installed so that 35 can be detected.

The signal intensity of the electromagnetic wave 35 detected by each detector 5 is compared with the trigger level (threshold level) of the comparator (comparator: pulse generating means) 6. When the intensity of the electromagnetic wave 35 is higher, the comparator 6 Outputs pulse voltage. This pulse voltage is once converted into an optical signal by the E / O converter 7, transmitted for a predetermined distance by the optical fiber cable 8, and then returned to the electric signal by the O / E converter 9 again. The electromagnetic wave 35 that has been pulsed and transmitted in this way is counted by a counter (pulse counting means) 10 every predetermined time, and the number of pulse measurements in the arithmetic unit 12 and the phase of the commercial frequency voltage applied to the conductor 2. From the distribution or the like, it is determined whether the electromagnetic wave 3 is an external noise or a partial discharge due to an insulation abnormality, what kind of abnormality is caused if it is a partial discharge, and a partial discharge occurrence location or the like.
The determination of the cause of occurrence of partial discharge is performed by determining the installation position of each detector 5 stored in the storage unit 11, the electromagnetic wave transfer function between the detectors 5, and the partial discharge pattern for each cause of occurrence of partial discharge. This is implemented based on (commercial frequency voltage phase (φ) −number of pulse measurements (or occurrence frequency) (n) −distribution pattern of electromagnetic wave intensity (Q)). The determination result is displayed on the display unit 13.

2 and 3 show examples of partial discharge electromagnetic waves detected by a gas insulated switchgear (GIS). In the figure, the horizontal axis is the commercial frequency voltage phase (φ) applied to the conductor, the vertical axis is the pulse measurement number (n) of the pulsed discharge electromagnetic wave, the depth axis is the electromagnetic wave intensity (Q), and the pulses are counted. This corresponds to the trigger level (Q _T ).
In general, a large number of discharge electromagnetic waves are detected when the trigger level is sufficiently low, but the number of detected discharge electromagnetic waves decreases as the trigger level increases. Further, when the position of the discharge generation source is fixed, for example, in the case of an insulation abnormality such as a metal foreign object adhering to the bottom surface of the metal container 4 or a metal foreign object adhering to the surface of the spacer 3, Each discharge exhibits a certain pattern with respect to the voltage phase (φ). On the other hand, when a metal foreign object is mixed in the metal container and is not fixed, the foreign object is charged by the electric field formed by the conductor 2 and the metal container 4, and may move around the metal container 4 by electrostatic force. . The movement of the foreign matter does not completely follow the voltage change of the conductor 2 and repeats the contact with the metal container 4 regardless of the voltage change. Therefore, the discharge generated at the time of contact becomes independent of the voltage phase (φ). As a result, a disordered pattern may be exhibited with respect to the voltage phase (φ).

FIG. 2 is an example of a partial discharge pattern when a metal foreign object is fixed to the bottom surface of the metal container 4. The discharge is distributed around the phases of 90 degrees and 270 degrees, which are the peaks of the applied voltage. This is because a partial discharge occurs when the tip electric field of a foreign substance that is a partial discharge generation source exceeds a certain value.
On the other hand, FIG. 3 shows an example of a partial discharge pattern when a metallic foreign object adheres to the surface of the spacer 3. A discharge with a low electromagnetic wave intensity (A in FIG. 3) is distributed around 45 degrees and 225 degrees of the voltage phase, whereas a discharge with a high electromagnetic wave intensity (B in FIG. 3) is 0 degrees. And a phase close to 180 degrees. This is because the electric charge due to the discharge in the voltage cycle of the previous half cycle remains on the creeping surface of the spacer insulator, and as a result, the electric field around the foreign material is emphasized at the start of the next cycle discharge, and a large discharge is generated. On the other hand, such a charged charge is neutralized by the occurrence of several partial discharges, and then reversely charged to the same polarity to relax the electric field around the foreign matter and generate a small discharge. This is probably because it occurs every half cycle.
The above is an example of a part of the partial discharge pattern, but the cause of the insulation abnormality can be determined by capturing the characteristics of the partial discharge pattern including other patterns.

A signal processing method for performing such determination will be described below.
First, the transfer function between the detectors 5 will be described. In the present embodiment, the storage unit 11 includes an electromagnetic wave transfer function between each detector 5 and the gas-insulated electric device 1 as an electromagnetic wave transfer function between the detectors 5 and the gas-insulated electric device 1. The electromagnetic wave transfer function between the parts corresponding to each detector 5 is stored.
Each of the above transfer functions is obtained as follows.

FIG. 4 shows a state in which a part of the gas-insulated electric apparatus 1 is connected to a measuring instrument for measuring the transfer function of the electromagnetic wave of each part. A plurality of detectors S <b> 1 to S <b> 6 are installed on the wall surface of the metal container 4 of the gas insulated electrical device 1, and detect partial discharge electromagnetic waves generated in the gas insulated electrical device 1. The pulse input point Sp1 is a pulse input point that is temporarily provided because the detector S1 is located at the end, and is set in the opposite direction with respect to the adjacent detector S2. Of the transfer function G (1) between the detector S1 and the gas-insulated electric device 1 and the transfer function in the gas-insulated electric device 1, the pulse input point Sp1 is connected to a portion corresponding to the detector S1 and the detector S2. This is necessary for obtaining the transfer function G (1,2) between the corresponding parts.
The pulse input point Sp1 may be the same as that of the detector, for example, or if a spacer is installed and the metal container is broken, a temporary electromagnetic wave injection device may be provided there. Alternatively, if the bushing is installed and the metal container is disconnected, an additional electromagnetic wave injection device may be provided there. In addition to these, if an electromagnetic wave signal can be injected into the metal container, pulse input with various configurations is possible. The point is applicable.

Now, the pulse generator 14 is connected to the pulse input point Sp1 to inject a pulse voltage (calibration pulse), and the oscilloscope 15 is connected to the detector S1 to detect the propagated electromagnetic wave signal. The ratio of the pulse input voltage and the detection voltage at the detector S1 at this time is the transfer function G (Sp1, S1) from the pulse input point Sp1 to the detector S1, and G (Sp1, S1) is From 4,
G (Sp1, S1) = G (p1) + G (p1,1) + G (1) (1)
Can be expressed as
Similarly, when the pulse generator 14 is connected to the detector S1 and the oscilloscope 15 is connected to the detector S2, the ratio of the pulse voltage injected from the detector S1 to the detection voltage at the detector S2 is the detector S1 to the detector S2. Can be measured as a transfer function G (S1, S2) until G (S1, S2) from FIG.
G (S1, S2) = G (1) + G (1,2) + G (2) (2)
Can be expressed as
Further, when the pulse generator 14 is connected to the pulse input point Sp1, and the oscilloscope 15 is connected to the detector S2, the ratio between the pulse voltage injected from the pulse input point Sp1 and the detection voltage at the detector S2 is detected from the pulse input point Sp1. Can be measured as a transfer function G (Sp1, S2) up to the device S2, and G (Sp1, S2) is obtained from FIG.
G (Sp1, S2) = G (p1) + G (p1,2) + G (2) (3)
Can be expressed as
When equations (1) to (3) are solved simultaneously, G (1) = {G (Sp1, S1) + G (S1, S2) -G (Sp1, S2)} / 2
... (4)
G (1) becomes a known value.

Similarly, when the measurement range is expanded to the detector S3,
G (S2, S3) = G (2) + G (2,3) + G (3) (4)
G (S1, S3) = G (1) + G (1,3) + G (3) (5)
And solving the equations (2), (4), and (5) simultaneously, G (2) = {G (S1, S2) + G (S2, S3) -G (S1, S3)} / 2
... (6)
G (2) is a known value.
Furthermore, from the equations (2), (4) and (6),
G (1,2) = G (S1, S2) -G (1) -G (2) (7)
Thus, G (1,2) is also a known value.

As described above, the pulse voltage is input to the predetermined detector S (i), the detection signal of the detector S (j) next to the detector S (i) and the detector S ( The transfer function between the gas-insulated electric device and the gas-insulated electric device by performing the procedure of measuring the detection signal of k) for all the detectors or all the detectors included in the predetermined area. Can be obtained.
In addition, for the detector S (m) whose installation location is located at the end of the gas-insulated electrical device, it can be added to any location on the opposite side of the detector S (n) adjacent to the detector S (m). The signal input point S (pm) is set, and the detection signal of the detector S (m) adjacent to S (pm) and the detection signal of the detector S (n) adjacent to S (pm) are measured. Perform the procedure on all detectors located at the ends of all gas-insulated electrical devices or detectors located at the ends of all gas-insulated electrical devices included in a given area. It is possible to determine the transfer function between the device and the transfer function in the gas-insulated electrical device.

  In the above method, the pulse generator and oscilloscope were used to calculate the pulse voltage change as a transfer function. However, the input pulse and detection pulse waveforms were also fast Fourier transformed and replaced with frequency components. It is also possible to obtain a transfer function as an intensity change for each frequency. It is also possible to obtain a transfer function as a characteristic including frequency and phase from the response when the frequency of the sine wave is swept using a network analyzer instead of the pulse generator and the oscilloscope.

The importance of obtaining the transfer function between the detector and the gas-insulated electric device and the transfer function between the parts corresponding to each detector in the gas-insulated electric device will be described next. .
The partial discharge pulse generated in the insulating gas is a high-speed pulse with a rise of 1 ns or less, whereby an electromagnetic wave having a frequency exceeding 1 GHz is radiated. On the other hand, in the case of an airborne corona that occurs outside the gas-insulated electrical device, which causes noise when detecting partial discharge electromagnetic waves, the frequency of the emitted electromagnetic waves is lower than that of the discharge in the insulating gas, and gas insulation from the outside Since it is filtered when propagating into the metal casing of the electric device, the main frequency tends to be 500 MHz or less. For these reasons, partial discharge monitoring of gas-insulated electrical devices often detects electromagnetic waves in the frequency band of several hundreds of MHz to several hundreds of MHz in order to increase sensitivity and improve S / N.

On the other hand, electromagnetic waves propagate in the metal casing while being repeatedly reflected, and resonate at a frequency determined by the structure and dimensions of the metal casing and the conductor. For example, in a gas-insulated bus having a coaxial structure, which is one of the gas-insulated electrical devices, depending on the size, in general, in a region where the frequency exceeds several hundreds of MHz, in addition to the basic TEM mode, a TE mode that is a resonance mode, TM A mode is generated, and an intensity distribution of an electric field or a magnetic field is generated in the axial direction, the circumferential direction, and the radial direction in the metal casing. In other words, the detector output is not simply determined by the distance between the electromagnetic wave source and the detector, the structure on the propagation path, etc., but also depends on the position on the metal housing where the detector is installed and the frequency to be detected. To do.
Therefore, in order to accurately determine the magnitude of the discharge from the intensity of the electromagnetic wave, the transfer function between the detector and the gas-insulated electric device and each detector in the gas-insulated electric device are supported by the method described above. It is important to measure the transfer function between the parts to be measured.

Next, how to determine the cause of the partial discharge from the detected electromagnetic wave will be described. FIG. 5 shows a partial discharge pattern created when the calculation unit 12 makes a determination. FIG. 5 shows a case where a partial discharge is generated in the vicinity of the detector S1 and an electromagnetic wave is detected by the detectors S1 to S4.
First, the voltage phase of the commercial frequency applied to the conductor is divided into a predetermined number of windows, and the number of pulses generated when the electromagnetic wave exceeds the trigger level Vtrig for each window is detected and detected. A characteristic pattern of voltage phase (φ) for each device−number of pulse measurements (n) is created. At this time, it is determined that the detector S1 having the largest number of pulse measurements is closest to the discharge source.

Next, a transfer function relating to the detectors S1 to S4 stored in the storage unit 11 is called, and the trigger level of each detector is corrected. That is, the electromagnetic wave attenuated by the transfer function G (1) between the gas-insulated electrical apparatus 1 and the detector S1 is detected in the detector S1, and the electromagnetic wave having an intensity equal to or higher than the trigger level Vtrig is pulsed. Therefore, substantially Vtrig + G (1) as the trigger level (virtual trigger level)
The electromagnetic waves having the above intensity are counted.
The detector S2 detects electromagnetic waves attenuated by the transfer function G (1,2) in the gas-insulated electrical apparatus 1 and the transfer function G (2) between the gas-insulated electrical apparatus 1 and the detector S2. Since the electromagnetic wave having the intensity equal to or higher than the trigger level Vtrig is pulsed, the trigger level (virtual trigger level) is substantially Vtrig + G (1,2) + G (2).
The electromagnetic waves having the above intensity are counted.
The output to the detectors S3 and S4 can be considered similarly,
The trigger level (virtual trigger level) for the detector S3 is Vtrig + G (1,2) + G (2,3) + G (3)
The trigger level (virtual trigger level) for the detector S4 is Vtrig + G (1,2) + G (2,3) + G (3,4) + G (4)
I believe that.

As a result of correcting the trigger level for each detector based on the transfer function in this way, the outputs of the detectors S1 to S4 detected the electromagnetic waves generated at the partial discharge occurrence positions simultaneously at a plurality of predetermined virtual trigger levels. You can think of it as the same thing.
Therefore, these are handled in an integrated manner, and the above-described characteristic pattern of voltage phase (φ) −number of pulse measurements (n) for each detector is changed according to the value (Q _T ) of the virtual trigger level for each detector. When arranged in the axis (Q) direction orthogonal to the phase (φ) axis and the number of pulse measurements (n) axis, as shown in FIG. 5, the partial discharge pattern (commercial frequency voltage phase (φ) −number of pulse measurements (n) − Electromagnetic wave intensity (Q) distribution pattern).

  Next, partial discharge patterns (commercial frequency voltage phase (φ) −number of pulse measurements (or occurrence frequency) (n) −electromagnetic wave intensity (Q) distribution pattern) for each cause of partial discharge stored in the storage unit 11) Call and compare and select the ones with similar characteristics.

  As described above with reference to FIGS. 2 and 3, examples of partial discharge sources and electromagnetic wave distributions have a close relationship between the type of partial discharge source and the partial discharge pattern. By storing it in the storage unit 11, it can be used to determine the cause of partial discharge.

In the above description, it is determined that the detector S1 having the largest number of pulse measurements is closest to the discharge source, and the virtual trigger level is calculated for each detector. explain in detail.
As an example,
G (1) + G (1,2) <G (2)
Or
G (2) -G (1)> G (1,2)
In such a case, even if there is a partial discharge generation source in the vicinity of the detector S2, the number of pulse measurements detected by the detector S1 is larger, and an erroneous determination that the discharge source is in the vicinity of the detector S1. Will be made. As described above, variations in the values of transfer functions G (i), G (j),... Between the plurality of detectors Si, Sj,. If the value is larger than the values of transfer functions G (i, j), G (j, k),... In the electric device, the position of the discharge source may be erroneously determined.

  Therefore, in the detector used for determining the discharge occurrence position, variation in the transfer function between the detector and the gas-insulated electric device is the transfer function in the gas-insulated electric device in the area where the detector is installed. It is preferable to select one smaller than the value and determine that the vicinity of the detector having the largest number of pulse measurements detected therein is the discharge generation source.

When the detectors are divided into a plurality of groups depending on the size of the transfer function, the discharge generation position is determined for each group, and a plurality of candidates are given as the discharge generation positions. Next, the correction of the trigger level of each detector is performed on each discharge generation position candidate. Since the number of pulse measurements decreases as the trigger level increases, when the descending order of the number of pulse measurements does not match the ascending order of the corrected trigger level (virtual trigger level), the discharge generation position is another position. Further, if there is a contradiction when it is assumed that the discharge generation position is in the vicinity of one of the detectors, it is possible that the discharge generation position is somewhere between the detectors. In this case, the trigger level is corrected by assigning a half of the transfer function in the gas-insulated electric apparatus, assuming that the position is simply between the two detectors. If a plurality of candidates remain even when the discharge generation position is selected and narrowed down as described above, the cause of discharge generation is determined for each position. When a plurality of determination results are obtained as the cause of occurrence of discharge, each result is displayed as possible.
By doing as described above, the discharge occurrence position can be obtained with high accuracy.

FIG. 6 is a flowchart showing a determination method of the partial discharge monitoring apparatus according to the first embodiment of the present invention, and shows the partial discharge monitoring method when the detectors are divided into a plurality of groups according to the size of the transfer function.
In FIG. 6, an electromagnetic wave signal is acquired in each detector (step S101), the trigger level set to a predetermined value by the comparator is compared with the peak value of the acquired signal (step S102), and the signal is If it is below the trigger level, there is no abnormality display (step S117), but if the trigger level is exceeded, the comparator generates a pulse for each electromagnetic wave signal (step S103). The pulse is counted by a counter (step S104), the number of pulses measured per unit time (n) is compared with a predetermined value (step S105), and if it is less than the set number, there is no abnormality display (step S117). If the number exceeds the set number, a determination operation is performed by the calculation unit.

The calculation unit calls up a database related to the transfer function stored in the storage unit (step S106), and transfers the transfer functions G (i), G (j),... Between each detector and the gas-insulated electric device. The value is compared with the values of transfer functions G (i, j), G (j, k),... In the gas-insulated electrical device, and a detector used for determining the discharge occurrence position is selected and transmitted. A plurality of detectors are grouped according to the values of the functions G (i), G (j),... (Step S107).
Next, it is temporarily set that the vicinity of the detector having the largest number of pulse measurements for each group is the discharge generation position (step S108), and the detector having the largest number of pulse measurements for each group and the detectors on both sides thereof. Is temporarily set to be a discharge generation position (step S109).

Subsequently, for each discharge occurrence position temporarily set in step S107 and step S109, the trigger level of the comparator corresponding to each detector in the vicinity of the discharge occurrence position is used using the transfer function called in step S106. corrected, to calculate the virtual trigger level (Q _T) (step S110).
Next, for each discharge occurrence position, it is compared whether the calculated ascending order of the virtual trigger level matches the descending order of the number of pulse measurements measured by the corresponding counter, and the temporarily set discharge occurrence Among the positions, a matching temporary setting position is selected (step S111), and it is determined that the matching temporary setting position is a discharge generation position (step S112). Here, if the virtual trigger level is compared with the ascending order of the virtual trigger level and the descending order of the number of pulse measurements, redundancy is provided, and if a plurality of positions are selected while giving priority, erroneous determination can be prevented.

Subsequently, a characteristic pattern of voltage phase (φ) −number of pulse measurements (n) is created for each detector, and obtained in step S110 for each detector based on the selected discharge generation position. A virtual trigger level (Q _T ) is assigned (step S113).
Further, the calculation unit integrates the characteristic pattern of voltage phase (φ) for each detector−number of pulse measurements (n) and the virtual trigger level (Q _T ) assigned to each detector to obtain a voltage phase ( A distribution pattern of (φ) −number of pulse measurements (n) −electromagnetic wave intensity (Q) is created (step S114), and a discharge pattern (voltage phase (φ) −pulse for each cause of partial discharge stored in the storage unit) A database of the number of measurements (n) -electromagnetic wave intensity (Q) distribution pattern) is called (step S115), compared, and similar characteristics are selected to determine the cause of occurrence (step S114).
Finally, the determination result is displayed (step S116).

Hereinafter, the determination procedure of the discharge occurrence position in steps S111 and S112 will be described more specifically with reference to FIGS.
FIG. 7 schematically shows the positions of the detectors S1 to S7 of the partial discharge monitoring device, the number of detected pulses, and the transfer function of each part. For simplification of explanation, here, the transfer function (attenuation rate) between the detector and the gas-insulated electric device is 10 [dB] in the detector group 1 (S1, S2, S5, S6) (circle mark). In the detector group 2 (S3, S4, S7) (Δ mark), 18 [dB] is set, and the transfer functions in the gas-insulated electrical apparatus are all 6 [dB] between adjacent detectors. Assume that S1 to S7 are arranged in series.
The number of pulse measurements detected for each of the detectors S1 to S7 is S5>S4>S6>S2>S3>S1> S7 in order from the largest. The two groups divided by the transfer function between the detector and the gas-insulated electrical device, each of which has the largest number of pulse measurements, the vicinity of the detectors S4 and S5, and the intermediate position S3 between these adjacent detectors. −4, S4-5, and S5-6 are temporarily set to the discharge generation positions.
FIG. 8 shows the calculated transfer functions from these five temporarily set positions to each detector. For the intermediate position between the detectors, the transfer function up to the positions of the detectors on both sides is calculated as 3 [dB] which is half of 6 [dB]. The value obtained by adding the threshold level of the comparator to these transfer functions is the correction value of the trigger level (virtual trigger level), but since the threshold values of each comparator are generally the same, the virtual trigger level is now The threshold value of the comparator is omitted for comparing the magnitudes of.
Based on the results of FIG. 8, the descending order of the number of pulse measurements and the ascending order of the virtual trigger level at each temporarily set position are summarized in FIG. 9 for each detector. In the figure, the combinations surrounded by dotted lines are cases where the calculated values of the virtual trigger level are the same, so the order may be switched.
From FIG. 9, the temporary setting position where the ascending order of the virtual trigger level most closely matches the descending order of the number of pulses is S4, followed by S4-5 and S5.
Therefore, it is determined that S4 (determination order 1) is most likely as the discharge occurrence position, and that S4-5 and S5 (determination order 2) are also possible.

As described above, according to the configuration of the first embodiment of the present invention, the detected high frequency electromagnetic wave signal is pulsed for each signal and then calculated and determined, so the circuit configuration from the detector to the calculation unit is simple. In addition, the processing load on the calculation unit is small and the determination can be made at high speed.
In addition, in order to obtain second-order electromagnetic wave intensity information by correcting the threshold levels of multiple comparators based on the signal propagation transfer function in the gas-insulated electrical apparatus, the cause and location of the partial discharge are determined. There is an effect that can be determined with high accuracy.
In addition, since the electromagnetic wave intensity derived from the discharge source itself can be obtained by correcting the waveform distortion and attenuation accompanying propagation / resonance of the electromagnetic wave, it is possible to accurately obtain the voltage phase-number of pulse measurements-electromagnetic wave intensity characteristics of the discharged electromagnetic wave. It is possible to determine the cause of occurrence of discharge.

In addition, according to such a configuration, since the pulsed detection signal can be easily converted into an electric signal, the signal can be transmitted from the detector to the counter by eliminating the influence of noise as much as possible. It becomes possible to do.
In the above embodiment, the comparator and the counter are connected by the optical fiber cable. However, the same effect can be obtained by connecting the counter and the arithmetic unit by the optical fiber cable.

In addition, when acquiring the transfer function, it is necessary to measure the propagation characteristics of the pulse between the detectors one after the other with the two adjacent detectors, which is necessary to determine the cause of the partial discharge. The transfer function of each part can be easily obtained.
For the transfer function related to the detector located at the end, the transfer function is obtained by providing the pulse input point on the opposite side of the adjacent detector, so the transfer function related to the detector at the end can be easily obtained. Obtainable.

Embodiment 2. FIG.
FIG. 10 is a block diagram showing a partial discharge monitoring device for a high voltage switchboard according to Embodiment 2 of the present invention. The high-voltage switchboard 36 accommodates a conductor 38 as a high-voltage charging unit, a circuit breaker (not shown), and the like in a metal casing 37 and is supported by an insulator (not shown). In some cases, the conductor 38 and the insulator are placed in a metal container (not shown) and sealed with an insulating gas. However, in the case of a high voltage switchboard with a relatively low voltage class, the air is simply sealed without sealing the insulating gas. You may have been exposed to. In such a high voltage switchboard 36, when the insulator is deteriorated, a partial discharge 39 is generated, and an electromagnetic wave 40 is radiated accordingly.
The electromagnetic wave is detected by the detector 27 installed in the metal casing 37 or a metal container (not shown), but the electromagnetic wave 40 is attenuated as it propagates inside the metal casing 37, so an insulation abnormality occurs anywhere in the high voltage switchboard 36. Even so, a plurality of detectors 27 are installed so that the electromagnetic wave 40 can be detected. The signal intensity of the electromagnetic wave 40 detected by the detector 27 is compared with the trigger level of the comparator 28. When the intensity of the electromagnetic wave 40 is higher, the comparator 28 outputs a pulse voltage. The pulsed electromagnetic wave 40 is counted by the counter 29, and the signal transfer function between the high-voltage switchboard 36 and each detector 27 stored in the storage unit 30 and the calculation unit 31 in the calculation unit 31, as in the first embodiment. The trigger level of each comparator 28 is corrected using a transfer function when a signal propagates through the high-voltage switchboard 36, and the distribution of the number of measurements with respect to the corrected virtual trigger level and the phase of the commercial frequency voltage applied to the conductor 38. Are used to obtain a pattern of voltage phase-number of pulse measurements (or occurrence frequency) -electromagnetic wave intensity, and by comparing with a plurality of patterns stored in advance, whether the detected electromagnetic wave 40 is external noise or due to insulation abnormality It is determined whether the discharge is a partial discharge or what kind of abnormality is caused by the partial discharge. The determination result is displayed on the display unit 32.

  With such a configuration, even in the partial discharge monitoring device of the high-voltage switchboard, there is an effect that the cause of the partial discharge can be determined with high speed and high accuracy with a simple circuit configuration.

It is a block diagram which shows the partial discharge monitoring apparatus of the gas insulated electrical apparatus by Embodiment 1 of this invention. It is a figure which shows an example of the partial discharge pattern concerning Embodiment 1 of this invention. It is a figure which shows the other example of the partial discharge pattern concerning Embodiment 1 of this invention. It is a figure explaining the measuring method of the electromagnetic wave transfer function concerning Embodiment 1 of this invention. It is a figure which shows the example of a measurement of the partial discharge pattern concerning Embodiment 1 of this invention. It is a flowchart which shows the determination method of the partial discharge monitoring apparatus by Embodiment 1 of this invention. It is a figure explaining the discharge generation position determination method of the partial discharge monitoring apparatus by Embodiment 1 of this invention. It is a figure explaining the discharge generation position determination method of the partial discharge monitoring apparatus by Embodiment 1 of this invention. It is a figure explaining the discharge generation position determination method of the partial discharge monitoring apparatus by Embodiment 1 of this invention. It is a block diagram which shows the partial discharge monitoring apparatus of the high voltage switchboard by Embodiment 2 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Gas insulation electric device, 2,38 conductor, 3 spacer, 4 metal container, 5,27 detector, 6,28 comparator, 7 E / O converter, 8 optical fiber cable, 9 O / E converter, 10 , 29 Counter, 11, 30 Storage unit, 12, 31 Calculation unit, 13, 32 Display unit, 14 Pulse generator, 15 Oscilloscope, 33 Insulating gas, 34, 39 Partial discharge, 35, 40 Electromagnetic wave, 36 High voltage switchboard, 37 Metal enclosure.

Claims (9)

When installed in a metal container of a gas-insulated electrical device and connected to a plurality of detectors that detect partial discharge electromagnetic waves and the plurality of detectors, respectively, and when the detection signal intensity of the detector exceeds a predetermined threshold value Pulse generating means for generating pulses, pulse counting means for counting the number of pulses every predetermined time, installation positions of the detectors, electromagnetic wave transfer functions between the detectors, and parts created in advance A storage unit that stores a partial discharge pattern for each cause of discharge generation, and a calculation unit, wherein the calculation unit determines a partial discharge occurrence location from the number of pulse measurements counted by each pulse counting unit, and using electromagnetic wave transmission function between the storage unit in the stored above each detector, to calculate the virtual threshold level of the pulse generating means for each detector are determined the partial discharge generation location near (Q _T), Starring Was the virtual threshold level (Q _T), since the commercial frequency voltage phase (phi) pulse measurement number of each detector for (n), commercial frequency voltage phase (phi) - pulse measurement number (n) - intensity of electromagnetic waves ( Q) A partial discharge pattern showing a distribution is created, and the cause of partial discharge is determined by comparing the partial discharge pattern with a partial discharge pattern for each partial discharge occurrence cause stored in advance in the storage unit. A partial discharge monitoring device for a gas insulated electric device. The storage unit includes an electromagnetic wave transfer function between each detector and the gas-insulated electric device as an electromagnetic wave transfer function between the detectors, and an electromagnetic wave transmission between parts corresponding to the respective detectors in the gas-insulated electric device. The partial discharge monitoring device for a gas-insulated electric apparatus according to claim 1, wherein the function is stored. The arithmetic unit has a variation in the value of the electromagnetic wave transfer function between each detector and the gas-insulated electric device that is larger than the value of the electromagnetic wave transfer function between the parts corresponding to the respective detectors in the gas-insulated electric device. 3. A gas-insulated electrical apparatus according to claim 2, wherein a detector having a small variation is selected, and the vicinity of the detector having the largest number of pulse measurements among the selected detectors is set as a partial discharge occurrence location. Partial discharge monitoring device. The calculation unit divides multiple detectors into multiple groups according to the value of the electromagnetic wave transfer function between each detector and the gas-insulated electrical device, and near the detector with the largest number of pulse measurements for each group Is used as a temporary partial discharge occurrence location, and the pulse generation means for each detector near each temporary partial discharge occurrence location for each group using the electromagnetic wave transfer function between the detectors stored in the storage unit. 3. A virtual threshold level is calculated, and a temporary partial discharge occurrence location of a group in which the descending order of the number of pulse measurements for each detector and the ascending order of the virtual threshold level are closest is specified as the partial discharge occurrence location. A partial discharge monitoring device for the gas-insulated electric device according to claim. The partial discharge monitoring device for a gas-insulated electrical apparatus according to any one of claims 1 to 4, wherein the pulse generating means and the pulse counting means are connected by an optical fiber. 6. The partial discharge monitoring device for a gas-insulated electrical apparatus according to claim 1, wherein the pulse counting means and the calculation unit are connected by an optical fiber. A first step of generating a pulse by a pulse generating means connected to each detector when the output intensity of a plurality of partial discharge electromagnetic wave detectors installed in a metal container of a gas insulated electric device exceeds a predetermined threshold A second step of counting the pulses every predetermined time, a third step of determining a partial discharge occurrence location from the number of measured pulses measured for the detectors, and between the detectors acquired in advance. The fourth step of calculating the virtual threshold level (Q _T ) of the pulse generating means for each detector in the vicinity of the determined partial discharge generation location using the electromagnetic wave transfer function, the calculated virtual threshold level (Q _T ) and , Partial discharge pattern showing distribution of commercial frequency voltage phase (φ) −number of pulse measurements (n) −electromagnetic wave intensity (Q) from the number of pulse measurements (n) of each detector with respect to the commercial frequency voltage phase (φ) And a sixth step of determining the cause of partial discharge by comparing the created partial discharge pattern with a partial discharge pattern for each partial discharge occurrence cause acquired in advance. A partial discharge monitoring method for a gas-insulated electric device, characterized in that Of the plurality of partial discharge electromagnetic wave detectors installed in the metal container of the gas-insulated electrical apparatus, a sine wave having a calibration pulse or frequency swept is input to a predetermined detector S (i), and the detector S ( i) The procedure of measuring the detection signal of the next adjacent detector S (j) and the detection signal of the two adjacent detectors S (k) is included in all detectors or in a predetermined area. As an electromagnetic wave transfer function between the detectors, an electromagnetic wave transfer function between the detectors and the gas-insulated electric device, and the detectors in the gas-insulated electric device. The partial discharge monitoring method for a gas-insulated electric apparatus according to claim 7, wherein an electromagnetic wave transfer function between corresponding parts is acquired. Among the plurality of partial discharge electromagnetic wave detectors installed in the metal container of the gas insulated electrical device, the detector S (m) whose installation location is located at the end of the gas insulated electrical device is the detector S (m). A signal input point S (pm) is set at an arbitrary location on the opposite side to the adjacent detector S (n), and the detection signal of the detector S (m) next to S (pm) and two The procedure of measuring the detection signal of the adjacent detector S (n) is located at the detector located at the end of all gas insulated electrical devices or at the end of all gas insulated electrical devices included in a given area. 9. The partial discharge monitoring method for a gas-insulated electric device according to claim 8, wherein an electromagnetic wave transfer function relating to the detector whose installation location is located at the end of the gas-insulated electric device is obtained by performing the detection on the detector.

Images