JP2002323532A - Partial discharge detecting method for electric appliance - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately detect partial discharge in an electric appliance such as a distributor in the current-carrying state by simple signal processing of a measurement signal of a sensor. SOLUTION: A signal of partial discharge of an electric appliance (switch 1) such as a distributor in the current-carrying state is measured by a sensor (magnetic-field probe 12), and data on a signal level in a predetermined frequency range where the signal of partial discharge among measured signals are distributed is obtained by frequency analysis of the measured signal of the sensor, an average value of data of the set range in the order of the level of this data is acquired, and detection is made based on this average value by determining presence of partial discharge.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting a partial discharge of an electric device, such as a distribution device, which detects a partial discharge of a current-carrying electric device from a measurement result of a signal generated by the discharge.

[0002]

2. Description of the Related Art Conventionally, on switches, transformers, insulators, and other pillars,
In electrical devices such as power distribution devices and electric motors on the road, when insulation deterioration or insulation abnormality occurs, partial discharge occurs, and the partial discharge generates a weak electromagnetic wave signal in a wide band of KHz to GHz.

[0003] Various methods for detecting partial discharge have conventionally been proposed in order to detect the partial discharge by measuring this broadband signal and to monitor and detect the presence or absence of insulation deterioration or insulation abnormality of electric equipment.

[0004] The measurement sensor and the processing method of the measurement signal in these conventional methods will be described below.

[0005] First, as sensors for measuring partial discharge, there are firstly an antenna such as a rod antenna and a loop antenna, and secondly, there are a current transformer (CT) for an instrument and a magnetic field probe.

[0006] When an antenna is used, an electromagnetic wave signal generated by partial discharge is received and measured. When a CT or a magnetic field probe is used, grounding of a cable ground line or the like of an electric device is performed by partial discharge. A current signal of a partial discharge flowing through a charged portion conductor such as a conductor or a cable conductor is measured through a magnetic field generated by the current signal.

[0007] As an example of the measurement of partial discharge using the first measurement sensor, Japanese Patent Publication No. 7-109428 discloses that a large number of partial discharges are provided on the outer periphery of an object to be measured (where discharge of an electric device is likely to occur). It describes that a minute loop antenna is installed to measure an electromagnetic wave of a partial discharge.

[0008] As an example of the measurement of partial discharge using the second measurement sensor, Japanese Patent No. 2956301 discloses a method of measuring a partial discharge current by attaching a through-type CT to a cable ground wire of a road switch. Is described.

Japanese Patent No. 2541739 discloses a method of processing a measurement signal of a measurement sensor, which is disclosed in Japanese Patent Application Laid-Open No. 2541739. Is subjected to a spectrum measurement process (frequency analysis process) to obtain signal level data (spectrum data) of each frequency at 0.4 MHz intervals, and from these data, excluding known band data of communication radio waves such as broadcast radio waves, A processing method is described in which the integrated value of the remaining data is compared with a reference value, and the presence or absence of partial discharge is determined based on the magnitude of the integrated value and detected.

[0010] Also, Japanese Patent No. 2993931 discloses that
Paying attention to the fact that the signal of the partial discharge is generated in synchronization with the system fundamental wave (50 Hz or 60 Hz), the component of the system fundamental wave and the component of twice the frequency are obtained from the measurement signal of the antenna. It describes that a composite value such as a square root of a sum of squares is calculated, and the presence or absence of a partial discharge is determined and detected based on the magnitude of the composite value.

[0011]

In this type of partial discharge detection method for electric equipment, a measurement signal of a sensor such as an antenna, a CT or a magnetic field probe includes a partial discharge signal and background noise of the environment of the electric equipment. From these measurement signals, the signal of the partial discharge is converted to a high S / N ratio.
It is important to improve the detection accuracy of partial discharge as much as possible.

As described in the aforementioned Japanese Patent No. 2541739, the partial discharge signal is obtained by removing the data of the frequency band of a known communication radio wave such as a broadcast radio wave from the frequency spectrum data of the measurement signal of the sensor. In order to improve the S / N, in practice, all frequencies of the sensor measurement signal are used by being allocated to some kind of broadcast or communication, and cannot be detected unless data in all communication radio frequency bands is removed. Therefore, the output of the data in the frequency band of each communication radio wave is large, and only the data in the frequency band of a special radio wave such as a broadcast radio wave output almost all day is removed.

In this case, the other communication radio wave data is
The noise data remains after the removal, and these noise data greatly affect the S / N depending on the detection location and the detection time zone, and the S / N of the partial discharge signal is reduced, thereby lowering the detection accuracy of the partial discharge. There is a problem.

Further, as described in Japanese Patent No. 2993931, when only the system fundamental wave and its twice frequency component of the measurement signal of the sensor are used for detecting the partial discharge, the partial discharge is used. The other generated frequency components are not used for detection, the measurement signal is not effectively used, and the detection accuracy cannot be sufficiently improved depending on the frequency signal distribution of the partial discharge.

In addition, complicated signal processing such as integration calculation of the composite value of the data of the system fundamental wave and its double frequency component is required, and there is a problem that it cannot be detected by simple signal processing.

According to the present invention, a partial discharge signal is extracted with a high S / N from a measurement signal of a sensor, and the signal is extracted by a simple signal processing.
It is an object to accurately and surely detect partial discharge of an electric device.

[0017]

According to a first aspect of the present invention, there is provided a method for detecting a partial discharge of an electric device, the method comprising detecting a partial discharge signal generated in an energized electric device such as a distribution device. Is measured by a sensor, and by analyzing the frequency of the measurement signal of the sensor, data of a signal level in a predetermined frequency range in which the signal of the partial discharge is distributed among the measurement signals is obtained, and a setting range in a level order of the respective data is obtained. Is determined, and the presence or absence of the partial discharge is determined and detected based on the average value.

Therefore, the signal level data of each frequency in the predetermined frequency range in which the frequency component of the partial discharge is distributed is obtained from the measurement signal of the sensor, and the average value of the data in the set range in the order of the levels of these data is obtained. Is required, the influence of the background noise is large, and the data of the part where the detection error is likely to occur is removed.
The average value of the data in the signal level range in which the frequency component of the partial discharge is distributed at a uniform signal level is obtained, and the level of the partial discharge is accurately obtained by a simple calculation.

Since the presence or absence of a partial discharge is determined based on the average value and the like, and the partial discharge is detected with high accuracy, the background noise is eliminated as much as possible by a simple signal processing so that the signal of the partial discharge is reduced. The partial discharge of the electric device is accurately and reliably detected from the obtained signal at a high S / N.

In the method for detecting partial discharge of an electric device according to a second aspect of the present invention, a partial discharge signal generated in an energized electric device such as a distribution device is measured by a sensor, and background noise in the environment of the electric device is measured. Measured by a sensor,
By analyzing the frequency of the measurement signal of the partial discharge of the sensor and the measurement signal of the background noise, signal level data in a predetermined frequency range in which the partial discharge signal is distributed is obtained for each of the measurement signals, and the level order of each of the measurement signals is determined. The average value of the data in the set range is determined, and the presence or absence of partial discharge is determined and detected based on a comparison between the two average values.

Therefore, in this case, each of the measurement signal of the partial discharge of the sensor and the measurement signal of the background noise of the measurement environment is subjected to the same signal processing as the processing of claim 1 to obtain an average value. Based on the comparison of the average values, the influence of the background noise can be more favorably eliminated and the signal of the partial discharge can be obtained with a higher S / N from the difference or the like. Is detected.

In a third aspect of the present invention, a partial discharge signal generated in a current-carrying electrical device such as a distribution device is measured by a sensor, and a frequency analysis of the measurement signal of the sensor is performed. The data of the signal level in a predetermined frequency range in which the partial discharge signal is distributed among the measurement signals is obtained, the data of a predetermined level order is obtained from each data, and the presence or absence of the partial discharge is determined based on the signal level of this data. To detect.

Therefore, in this case, the calculation of the average value of claim 1 is not performed, and data of a predetermined level order is selected and obtained from the data of the measurement signal of the sensor in the predetermined frequency range. Based on the signal level, the partial discharge is accurately and reliably detected by simpler signal processing.

Further, in the method for detecting partial discharge of electric equipment according to the present invention, a partial discharge signal generated in an energized electric equipment such as a distribution equipment is measured by a sensor, and background noise of the environment of the electric equipment is measured. Measured by the sensor, frequency analysis of the sensor partial discharge measurement signal and the background noise measurement signal, for each measurement signal, obtains signal level data in a predetermined frequency range where the partial discharge signal is distributed. Data of a predetermined level order of each signal is obtained, and the presence or absence of partial discharge is determined and detected by comparing the signal levels of the data of the two predetermined level orders.

Therefore, also in this case, the calculation for calculating the average value of the measurement signal of the partial discharge and the measurement signal of the background noise of the second aspect is not performed. From the data in the frequency range, the data of each predetermined level is selected and obtained, and based on the comparison of the signal levels of the data of the two predetermined levels, the partial discharge is detected more accurately than in the case of claim 3. You.

[0026]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG.
This will be described with reference to FIGS. (1) First, an embodiment of the present invention will be described with reference to FIGS.
This will be described with reference to FIG. FIG. 2 shows a configuration for detecting partial discharge of a multi-circuit switch 1 installed on a road, which is an example of a road power distribution device.

The switch 1 shown in FIG. 2 is in a state where a front door (not shown) is open, and the metal outer box 2 is at the ground potential.

Further, a metal inner box 4 having a front door 3 is located in the upper part of the outer box 2, and a plurality of metal lids 5 which are so-called closed or open doors are stored in the inner box 4. A switch contact portion (movable electrode portion) is provided in these lids 5.

Also, cables 6 for each phase of each circuit drawn from the switch contacts of each lid 5 are located in the lower part of the outer box 2, and these cables 6 have a cable conductor at the center. The cable conductor is covered with an outer sheath (cable sheath) 7 of an electrical insulator such as rubber.

The metal parts such as the inner box 4 and the lid 5 are connected to the outer box 2 and all are kept at the ground potential.

When the switch 1 is energized (during operation),
System current of each phase of each circuit flows to each switch contact portion in the closed state via the cable conductor of each cable 6.

When insulation deterioration or insulation abnormality occurs in the switch 1 due to long-term use or the like, for example, a contact between each switch contact portion on the high potential side which is a charged portion and the inner wall of the lid 5 at the ground potential. Partial discharge occurs.

At this time, the one-phase equivalent circuit of the switch 1 is
FIG. 3 shows a schematic view, in which 5 'is a switch contact portion in the cover 5, 8 is a cable conductor of the cable 6, 9 is a cable conductor 8 and a shielding layer 18 for taking a ground potential. The cable capacitor 10 between them is the insulation resistance at the connection location of the outer cover 7, and G is the location where partial discharge occurs.

When a partial discharge occurs, for example, K
A high-frequency current of partial discharge occurs in a wide frequency band of Hz to GHz, and this high-frequency current is supplied to the cable conductor 8 of the charging part and the outer box 2 serving as a grounding conductor, as shown by the arrow in FIG.
If there is a cable ground line 11 and each of the lids 5, the cable also flows through the ground line 11.

There is also a path that flows to the ground via a ground point of another device or the like and the cable capacitor 9, and returns from the ground to the generation location G via the outer box 2.

Further, the high frequency radio waves of the partial discharge distributed in the wide frequency band are radiated into the atmosphere from the generation location.

Therefore, a rod antenna or a loop antenna may be used as a sensor, the sensor may be brought close to the generation point G to receive and measure a high frequency radio wave of a partial discharge, and the partial discharge may be detected using the measurement signal. In this embodiment, the magnetic field probe 12 shown in FIG. 2 is used as a sensor in order to obtain a measurement signal of the partial discharge at a higher S / N than the reception of radio waves by those antennas.

Since the high-frequency current flowing through the cable conductor 8 can be detected at a level higher than the high-frequency current flowing through each ground conductor, in this embodiment, the switch 1 is energized (during operation). Next, the magnetic field probe 12 is brought into contact with or close to the cable 6 to be measured.

The magnetic field probe 12 comprises a small magnetic field detection sensor formed by housing a coil such as a magnetic field detection search coil in an insulator case 12 ', and is similar to a penetrating type current transformer for an instrument. Instead of a dedicated sensor attached to the cable 6, it is carried to the measurement site together with a detection processing device described later only when necessary, and is used in contact with the outer cover 7 as shown in FIG.

At this time, the magnetic field probe 12 is sufficiently small compared to a rod antenna or the like, has no restrictions on the shape and the like, can easily perform measurement in a narrow place, and can be measured by one probe 12. Therefore, there is an advantage that it is inexpensive as compared with the case where a large number of loop antennas are provided for measurement, and it is possible to work efficiently in a small space.

When the high frequency current of the partial discharge flows through the cable conductor 8, a magnetic field (high frequency magnetic field) corresponding to the high frequency current is generated, and the probe 12 comes into contact with the cable 6 and approaches the cable conductor 8. And the probe 12
The high frequency current of the partial discharge is measured through the measurement of the magnetic field.

In this case, compared to a case where a high frequency radio wave radiated into the atmosphere based on partial discharge is received by an antenna such as a rod antenna or a loop antenna, the influence of noise such as an external radio wave is small. Measurement with high S / N can be performed.

Next, the analog measurement signal of the partial discharge of the magnetic field probe 12 is sent to the high frequency measurement unit 15 of the detection processing device 14 via the connection lead 13.

The measuring unit 15 is composed of, for example, a spectrum analyzer, and performs a frequency analysis process for spectrum measurement on an analog measurement signal of the probe 12.

The predetermined frequency range in which the frequency component of the partial discharge is distributed is set to 0 to 200 MHz, and the signal level data of each frequency in the range of, for example, 0.4 MHz is obtained, and the data of the frequency analysis result at 501 points is obtained. Then, these data are sent to the determination unit 16 of the computer configuration. The judging unit 16 operates as shown in FIG. 1 to detect a partial discharge.

That is, according to the actual measurement, the magnetic field probe 1
When the frequency spectrum data of a predetermined frequency range is input from the high frequency measurement unit 15 with respect to the measurement signal of the partial discharge obtained by bringing the cable 2 closer to the cable conductor 8, the process proceeds from step S _{1 in} FIG. ₁ to step S _2. , The input data are rearranged in ascending order of signal level, and a level distribution is obtained. At this time, the data at the 251st point from the bottom is data of the median (median).

The data at the maximum level and the level near the maximum level (hereinafter referred to as the maximum level) include background noise having a large level such as a broadcast wave, and the level tends to become extremely large due to this noise.

Further, the data of the minimum level and the level in the vicinity thereof (hereinafter referred to as minimum minimum level) have a small signal level of the partial discharge, and the difference from the background noise tends to be small.

Since only the data of the portion where the change in signal level is as flat as possible and the components of the partial discharge are uniformly distributed are extracted, the setting range of the level order excluding the data near the maximum level and the minimum level is limited. Are set in the determination unit 16 as setting sections.

This section setting is performed by the operator based on, for example, the frequency spectrum of the frequency analysis result of the measuring section 15 displayed on the screen of the monitor display section 17 subsequent to the determining section 16 and the rearranged level distribution.

The frequency spectrum of the analysis result displayed on the display unit 17 depends on whether a partial discharge is present (when it occurs) or not (background noise), for example, as shown in FIG.
When the respective data of the spectrum are rearranged in ascending order of the signal level, the data with the highest level is 100% data, and the data with the median is 50% data. The results are as shown by solid lines α ′ and β ′ in FIG.

The solid line α ′ is obtained by rearranging the data of the spectrum of the solid line α with partial discharge in ascending order, and the solid line β ′ is the data of the spectrum of the solid line β without partial discharge (background noise). Are sorted in ascending order of signal level.

Then, in the case of the solid line β ′ in FIG.
Since the signal level of the data having a level higher than the order of approximately 75% increases sharply due to background noise such as a broadcast radio wave or a communication radio wave, the set section is, for example, data having a median value (a 50% order). 25% around
It is set to the data of the section of the order of 75%.

[0054] Based on this interval setting, determination unit 16, in step S ₃ in Fig. 1, he detects the signal of the extracted and partial discharge data only 25% to 75% Rank interval at a high S / N.

[0055] Further, the process proceeds to step S _4, for each data of the extracted interval, obtains the average value of the signal level (partial discharge measurement side average value).

At this time, if a partial discharge has occurred, the data in the extracted section has the frequency component of the high frequency current of the partial discharge uniformly superimposed on the background noise of a substantially constant level. .

Then, the background noise is small,
With respect to the measurement result of the magnetic field probe 12 whose S / N is higher than that in the case of the radio wave reception of the antenna, an average value of only data in a signal level range in which the influence of unnecessary noise is further reduced is obtained. The signal level at the time of occurrence is sufficiently higher than the signal level of the background noise when it is not occurring.

[0058] Therefore, when comparing the partial discharge measurement side average value of obtained in step S ₄ and a suitable threshold value (set value), the average value in the event of a partial discharge is sufficiently larger than the threshold value In addition, the presence or absence of partial discharge can be reliably determined and detected based only on the partial discharge measurement side average value.

However, in this embodiment, in order to detect partial discharge with higher accuracy, the magnetic field probe 12
Is brought into contact with the outer sheath 7 of the cable 6 and the probe 12 is brought close to the cable conductor 8 to measure not only the partial discharge current passing through the conductor 8 but also to perform comparative measurement as shown in FIG. To the outer sheath 7 of the cable 6
L (= several tens of centimeters or more) away from the probe 1
The background noise in the measurement environment of the switch 1 is also measured with the cable 2 away from the cable conductor 8.

That is, since the magnetic field of the high-frequency current flowing through the cable conductor 8 attenuates in inverse proportion to the square of the distance, when the magnetic field probe 12 is separated from the cable 6 by several tens of centimeters or more, the high frequency of the partial discharge is reduced. The measurement level of the current is attenuated by several tens of decibels and becomes extremely small as compared with the case where the current is closely attached to the cable 6.

On the other hand, in the measurement environment, communication radio waves such as broadcast radio waves, radio waves generated from various industrial equipments and the like mainly exist as background noise of space propagation, and the magnetic field probe 12 picks up this noise.

The level of the background noise picked up by the magnetic field probe 12 hardly changes even when the probe 12 is moved closer to or away from the cable conductor 8. The magnetic field probe 12 is separated from the cable 6 by several tens of centimeters or more, and the background noise in the measurement environment is measured in this state.

The measurement signal of the magnetic field probe 12 separated from the cable conductor 8 is also measured by the high-frequency measurement unit 1.
The frequency analysis is performed according to step 5 to obtain data of a predetermined frequency range of the frequency analysis result of, for example, 501 points, and these data are sent to the determination unit 16.

[0064] At this time, the determination unit 16 returns to step S ₁ through step S ₅ from step S ₄ in FIG. 1, the loop of steps S _1, S ₆ data input wait from the high-frequency measuring section 15 state to have, from the measurement unit 15, based on a comparison measurement, the data of a predetermined frequency range of background noise in the measurement signal is input, the process proceeds from step S ₆ to step S _7, the respective data in ascending order of the signal level Sort to find the level distribution.

The spectrum of the background noise is, for example, as shown by the solid line β in FIG. 5A, and each data rearranged in ascending order is shown by the solid line β ′ in FIG. Become like

[0066] In addition, the process proceeds to step S _8, the solid line β '
Also for each data, extracts only data of the setting period of the signal level 25% to 75% Rank, in step S _9, obtaining their mean values (noise measurement side average value).

[0067] When the partial discharge measurement side mean and noise measurement side average value is obtained, the process proceeds to step S ₁₀ through step S _5, you obtain a difference between the average value back from the partial discharge measurement side mean Remove ground noise.

[0068] Further, the process proceeds to step S _11, compared denoising the partial discharge measurement side average value (the difference value) with a threshold (set value) to determine the presence or absence of a partial discharge, the average value (the difference between Is smaller than or equal to the threshold value, the partial discharge is detected, and the detection result is displayed on the monitor display unit 17.

Therefore, the measurement result of the main measurement in which the magnetic field probe 12 was brought close to the cable conductor 8 and the magnetic field probe 1
2 is obtained from the spectrum data of a predetermined frequency range of the measurement result of the comparative measurement separated from the cable conductor by a simple signal processing for obtaining an average value of each signal level, thereby obtaining a signal of partial discharge at an extremely higher S / N than in the past. Thus, partial discharge can be reliably detected.

The partial discharge can be detected more accurately than before only from the average value of the measurement result of the main measurement. In this embodiment, the measurement result of the main measurement and the measurement result of the comparative measurement are obtained. Since the partial discharge is detected from the comparison of the average values, the influence of the background noise can be further reduced and the partial discharge can be detected very accurately and reliably.

Incidentally, the overall judgment may be made by a combination of the judgment based on the average value of the measurement results of the main measurement and the judgment based on the difference between the average values of the main measurement and the comparative measurement.
In this case, the partial discharge can be detected more accurately and reliably.

In the case of the three-phase AC switch 1, since the high frequency current of the partial discharge flows through the cable conductor 8 of each circuit of the same phase as shown in the equivalent circuit of FIG. One suitable cable 6 may be selected, and the magnetic field probe 12 may be moved closer to or farther from the cable conductor 8 of the cable 6 to detect the partial discharge of each phase. In this case, FIG. As described above, the magnetic field probe 12 may be brought into contact with a portion of the outer skin 7 different from that shown in FIG.

Then, based on the detection of the three-phase partial discharge,
It is possible to know insulation deterioration and insulation abnormality of the switch 1.

In the above-described embodiment, the section to be extracted is a section having a rank of 25% to 75%. However, the set section (predetermined section in the order of level) is not suitable for the situation such as background noise at the site. It may be set appropriately according to, for example, 4
It may be a section of 0% to 60% rank.

Further, in order to remove unnecessary noise of a relatively large level which is constantly generated such as broadcast radio waves and to detect the noise, the predetermined frequency range is set to, for example, 10 to avoid broadcast radio waves.
Two ranges of 70 MHz and 110 to 200 MHz may be used, and three or more ranges may be used as needed.

When the predetermined frequency range is set to a plurality of ranges, the average value is obtained for each range, the partial discharges are individually detected, and the principle of majority decision is determined by comprehensive judgment based on a combination of these detection results. For example, the partial discharge may be detected, and in this case, the detection accuracy is further improved.

(Other Embodiment) Next, another embodiment of the present invention will be described with reference to FIG. In the first embodiment, a level distribution is obtained from each data in a predetermined frequency range obtained from a frequency spectrum of a measurement signal of the magnetic field probe 12, and only data in a set range in a level order is extracted from the data, and based on an average value thereof. However, in this embodiment, in order to further simplify the signal processing, based on each data in the predetermined frequency range of FIG. 7A obtained by the frequency spectrum measurement of the high frequency measurement unit 15,
The determination unit 16 rearranges the signal levels in ascending or descending order as shown in FIG. 3B, and calculates the median data Dm [dB] of the 50% rank in each of the main measurement and the comparative measurement.
m] as data of a predetermined level order.

Note that Dmin in FIG. 6B is data of the lowest level order, and Dmax is data of the highest level order.

The magnetic field probe 12 for both measurements
, The presence / absence of partial discharge is determined and detected from the difference of the median data Dm based on the measurement signal of (i).

In this case, the determination unit 16 obtains the median data Dm of the 50% rank from the rearranged data instead of the processes of steps S ₃ and S ₄ and steps S ₈ and S ₉ in FIG. Perform processing.

[0081] Further, in step S ₅ of FIG. 1, the data Dm in both median is determined whether or not required.

The simple average value of each data in the predetermined frequency range obtained from the frequency spectrum, the average value of the 25% to 75% section of the one form, and the median value of this form are partially When the signal level [dBm] and the S / N [dB] at the time of the occurrence of the partial discharge and the occurrence of the partial discharge were determined, the results shown in Table 1 below were obtained. Was done.

[0083]

[Table 1]

As is clear from these results, the use of the median value data Dm is significantly more remarkable than the case where the simple average value is used, as in the case of using the average value of the signal levels of the data in the set section. / N is improved, and partial discharge with high accuracy can be detected.

It is to be noted that only the main measurement is performed to obtain only the median data Dm, and the signal level of this data Dm is compared with the signal level of the determined threshold value. The occurrence of the partial discharge may be determined and detected. In this case, the comparison measurement and the processing of the measurement signal are unnecessary, and the detection of the partial discharge with high accuracy can be performed by simpler signal processing.

In addition, depending on the noise condition at the site, etc.,
Instead of the median value data Dm, data before and after, for example, data of a level rank of 40% and data of a level rank of 60% are obtained, and the presence or absence of partial discharge is determined from the magnitude of the signal level or the difference between the signal levels. It may be detected.

By the way, in both of the above embodiments,
The magnetic field probe 12 is brought into contact with or close to the cable 6 to measure the current of the partial discharge flowing through the cable conductor 8, but the magnetic field is applied to various ground conductors of the switch 1 such as the lid 5 and the cable ground line 11. The probe 12 may be contacted to measure the partial discharge current flowing through the conductor.

In the above-described embodiments, the predetermined frequency range in which the partial discharge signal is distributed is 0 to 200 MHz.
Not limited to, may be set appropriately according to the measurement environment and the like, at this time, a plurality of predetermined frequency ranges are set, and the average or median data is obtained for each frequency range to determine the presence or absence of partial discharge. The presence or absence of the final partial discharge may be detected in accordance with, for example, the principle of majority voting by detecting and comprehensively determining the detection results.

Next, the sensor for magnetic measurement of the present invention comprises:
The present invention is not limited to the magnetic field probe 12, and may be, for example, a penetrating CT attached to the cable ground line 11, or may be an antenna such as a rod antenna or a loop antenna.

When the sensor is an antenna, only the measurement signal of the partial discharge received by bringing the antenna close to the location G of the partial discharge, or the measurement signal and the background noise of the background noise received away from the location G of the antenna. With respect to the measurement signal, the high-frequency measurement unit 15 and the determination unit 16 may perform the detection processing of the partial discharge in the same manner as in the above-described two embodiments. By obtaining the average value of the data in the set range in the level order of these data or the signal level of the data having the predetermined level order, it is possible to detect the partial discharge with the same high accuracy as in the above-described two modes. .

The present invention can be applied not only to the detection of partial discharge of road-side distribution equipment such as the switch 1 but also to the detection of partial discharge of electric motors and pole-side distribution equipment. it can.

[0092]

The present invention has the following effects. First, in the case of claim 1, a sensor (magnetic field probe 1
For the measurement signal of 2), the signal level data of each frequency in the predetermined frequency range in which the frequency component of the partial discharge is distributed was obtained, and furthermore, the average value of the data in the level order of these data was obtained. The data of the part where the influence of the ground noise is large and the detection error easily occurs is removed, and the average value of the data in the range where the frequency component of the partial discharge is distributed at a uniform signal level is obtained. It can be obtained with high precision from the calculation of the value.

Then, the presence or absence of a partial discharge can be determined based on the average value and the like, and the partial discharge can be detected with high precision. A signal is obtained with a high S / N, and a partial discharge of an electric device (switch 1) can be accurately and reliably detected from the obtained signal.

In the case of the second aspect, each of the measurement signal of the partial discharge of the sensor and the measurement signal of the background noise of the measurement environment is subjected to the same signal processing as in the first aspect, and an average value is obtained. Based on the comparison of the average values, the influence of the background noise can be more favorably eliminated from the difference or the like, and the signal of the partial discharge can be obtained with a higher S / N. Can be detected with higher accuracy.

Next, in the case of claim 3, the data of the predetermined level order is selected from the data in the predetermined frequency range of the measurement signal of the sensor without performing the calculation for obtaining the average value of claim 1. , Based on the signal level of this data,
Since the presence or absence of the partial discharge is determined and detected, the partial discharge of the electric device can be accurately and reliably detected by simpler signal processing.

Also, in the case of the fourth aspect, the same operation as in the third aspect is performed for both the measurement signals without performing an operation for obtaining an average value of the partial discharge measurement signal and the background noise measurement signal of the second aspect. Then, data of a predetermined level order is selected and obtained from the data of each predetermined frequency range, and the presence or absence of partial discharge is determined and detected based on a comparison between the two data, so that extremely simple signal processing is performed. In this case, the partial discharge of the electric device can be detected with higher accuracy than in the case of

[Brief description of the drawings]

FIG. 1 is a partial flowchart of signal processing according to an embodiment of the present invention.

FIG. 2 is a configuration explanatory diagram of one embodiment of the present invention.

FIG. 3 is an explanatory diagram of a current of the partial discharge in FIG. 1;

FIGS. 4A and 4B are explanatory diagrams of a measurement state of the magnetic field probe of FIG. 1;

5A is a frequency spectrum of the measurement result of FIG. 1, and FIG. 5B is a spectrum obtained by rearranging the spectrum of FIG. 1A in the order of signal level.

FIG. 6 is an explanatory diagram of another measurement state of the magnetic field probe of FIG. 1;

7A is a frequency spectrum according to another embodiment of the present invention, and FIG. 7B is a spectrum obtained by rearranging the spectrum of FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Multi-circuit switch 5 Lid 6 Cable 8 Cable conductor 11 Cable ground wire 12 Magnetic field probe 13 Detection processing device

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Noriyasu Goto 47 Numezu Takaune-cho, Ukyo-ku, Kyoto-shi Inside Nissin Electric Co., Ltd. (72) Inventor Toru Yoshikawa 47-Takaune-cho Umezu, Ukyo-ku, Kyoto-shi F-term in Nissin Electric Co., Ltd. (reference) 2G015 AA06 BA08 CA01

Claims (4)

[Claims] 1. A sensor detects a partial discharge signal generated in a current-carrying electric device such as a power distribution device, and a frequency distribution of the measurement signal of the sensor causes a distribution of the partial discharge signal in the measurement signal to be distributed. Determining a signal level data in a predetermined frequency range to obtain, an average value of the data in a set range in the order of the level of each of the data, based on the average value, determine the presence or absence of the partial discharge and detect. Partial discharge detection method for electrical equipment. 2. A sensor measures a partial discharge signal generated in a current-carrying electric device such as a distribution device, and measures a background noise of an environment of the electric device by the sensor, and measures a partial discharge of the sensor. By analyzing the frequency of the signal and the measurement signal of the background noise, for each of the two measurement signals, signal level data in a predetermined frequency range in which the partial discharge signal is distributed is determined, and the level order of each of the two measurement signals is set. A partial discharge detection method for electrical equipment, comprising: determining an average value of data in a range; determining whether or not the partial discharge exists by comparing the two average values; 3. A sensor detects a partial discharge signal generated in an energized electrical device such as a power distribution device, and a frequency analysis of the sensor measurement signal distributes the partial discharge signal among the measurement signals. Determining data of a signal level within a predetermined frequency range, obtaining data of a predetermined level order from each of the data, determining and detecting the presence or absence of the partial discharge based on the data of the predetermined level order. Partial discharge detection method for electrical equipment. 4. A sensor measures a partial discharge signal generated in a current-carrying electric device such as a power distribution device, and measures background noise in an environment of the electric device by the sensor, and measures partial discharge of the sensor. A signal level data of a predetermined frequency range in which the partial discharge signal is distributed is obtained for each of the two measurement signals by frequency analysis of the measurement signal of the signal and the background noise, and a predetermined level order of each of the two measurement signals is obtained. And detecting the presence or absence of the partial discharge based on a comparison between the signal levels of the data of the two predetermined levels and detecting the partial discharge.