JPH02183155A - Method and device for detecting minute foreign objects in stationary electrical equipment, and collision sound sensor provided in this device - Google Patents

Abstract

PURPOSE:To enable satisfactory discrimination of a sound as emitted when fine foreign matters collide one another from extraneous noises by detecting a collision sound within an equipment to determine a characteristic of the collision sound. CONSTITUTION:Sounds as emitted when a conductive foreign matter 3i existing in a bus container 1 for a gas insulation switch gear and a foreign matter 30 existing outside the container 7 collide separately on a walls of the container 1 are detected with sound sensors 11a and 11b and sound sensors 11c and 11d set on an outer surface of the container 1. Sound signals detected with the sensors 11a-11d are amplified separately with amplifiers 12a-12d. Signals after the amplification of outputs of the sensors 11a and 11b and outputs of the sensors 11c and 11d are inputted into foreign matter identification index computing device 15 for computing a foreign matter identification index. The signals after the amplification of the outputs of the sensor 11a and 11b are inputted into a sound source distribution computing device 14 through a position evaluator 13. A fine foreign matter discriminator 16 identifies a collision sound due to the foreign matter 3i within the container 1 or a sound due to another cause from a foreign matter identification index as output of the computing device 15 and a sound source distribution as output of the computing device 14.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides a method for detecting minute foreign matter in stationary electrical equipment, which detects whether or not there is a minute foreign matter that causes insulation deterioration in the stationary electrical machine. and a device, and a collision sound sensor for detecting the collision sound caused by the minute foreign matter hitting the stationary electrical equipment from inside, and in particular, the stationary electrical equipment suitable for accurately distinguishing the collision sound caused by the internal minute foreign matter from external noise. The present invention relates to a method and device for detecting minute foreign objects, and a collision sound sensor included in the device.

[Prior Art] If a partial discharge occurs inside a stationary electrical device, such as an electrical device that handles high voltage, this may trigger a fatal dielectric breakdown accident. Therefore, in the past, internal discharge was detected to prevent dielectric breakdown accidents in advance.

Conventionally, as a method for detecting the presence or absence of internal discharge, a microphone is installed in a stationary electrical device and the sound accompanying the internal discharge is detected. However, when detecting only this sound, it may not be possible to distinguish it from external noise.
In the conventional technology described in Publication No. 421, the generation of an electric pulse caused by internal discharge is detected, and when a sound is detected after a predetermined time has elapsed since the detection of this electric pulse, it is determined that this sound is caused by partial discharge. are doing.

[Problems to be Solved by the Invention] The above-mentioned conventional technology detects the presence or absence of partial discharge, and zero partial discharge, which may not be detected in time to prevent a fatal insulation breakdown accident, is detected, for example, inside the equipment. Microscopic foreign matter on the micron order, such as tiny pieces of equipment components, foreign matter that has peeled off from the equipment surface, or foreign matter that has melted from equipment components due to the previous partial discharge, can cause magnetic fields caused by the electric field inside the equipment or changes in the electric field. This is often caused by the particles jumping around due to the influence of foreign objects. Therefore, by detecting the sound of collision of this minute foreign object hitting equipment, it is possible to check the possibility of partial discharge occurring before partial discharge occurs. becomes possible. However, in this case, unlike the prior art described above, only the collision sound is detected, making it difficult to distinguish this from external noise.

A first object of the present invention is to provide a method and apparatus for detecting minute foreign objects in stationary electrical equipment, which can effectively distinguish the sound of a minute foreign object inside one device colliding with the device from external noise.

A second object of the present invention is to provide a collision sound sensor that can detect the sound of a minute foreign object colliding with a device from inside, while distinguishing it well from external noise.

[Means for Solving Problem 11] The first objective is to detect a collision sound, determine the characteristics of this collision sound, and determine whether or not this characteristic is biased compared to a predetermined characteristic. , achieved.

The second object is achieved by accommodating an element that converts low frequency sound into an electrical signal and an element that converts high frequency sound into an electrical signal in one container.

[Effect] External noise is caused by a variety of miscellaneous causes, and its characteristics vary.On the other hand, the collision noise generated when minute foreign objects bounce around under the influence of the electric and magnetic fields generated inside the device is caused by the electric and magnetic fields. As a result of being subject to regulations, its characteristics become biased. Therefore, it is possible to distinguish between the two based on the difference in the characteristics of the detected sound.

Further, the minute foreign matter inside is a hard object because it is a piece of material that constitutes one device. Therefore.

The collision sound will contain many high frequency components. On the other hand, external noise detected when a soft foreign object such as rain hits a device contains many low frequency components.

Therefore, if high-frequency sound and low-frequency sound can be detected with one sensor, it becomes possible to distinguish the source of collision sound without providing multiple sensors.

[Embodiment] Preferred embodiments of the present invention will be described below with reference to the drawings.6 Figure 1 shows a micro foreign object detection device in a stationary electrical equipment according to an embodiment of the present invention, which is installed below a gas insulated switchgear IC. A high voltage is applied between the cylindrical GIS bus container 1 and the center conductor 2, and an insulating gas is filled between them. ing. Conductive foreign matter 3 existing in the GIS bus bar container 1
1 and the foreign object 3o existing outside the GIs busbar container 1 on the container 1 wall, propagate through the container 1 wall, and
Acoustic sensors lla and llb and acoustic sensors llc and 1. Detected by ld.

Foreign matter 31 inside the GIS bus bar container 1 is removed from the container 1 or the center conductor 2
It is often a piece of or peeled off, so it is a metal that has conductivity or magnetism.

Moreover, the size thereof is quite small, on the order of microns. Therefore, the foreign matter 31 is caused to jump due to the electric field extending radially from the center conductor 2 to the inner surface of the container 1. The direction in which it jumps depends considerably on the direction of the electric field. On the other hand, external foreign matter 3
o is sand, dust, etc. blown by the wind, and air vibrations such as thunder may also be transmitted to the container 1 as external noise.

Acoustic sensors lla and 11b are AE sensors and detect high frequency components of propagating sound. Acoustic sensor llc, li
d is an acceleration sensor that detects low frequency components of propagating sound. Incidentally, only one of the acceleration sensors llc and lid may be used, and the acoustic signals detected by the acoustic sensors lla to lid are sent to the amplifiers 12a to 12, respectively.
amplified by d. The amplified signals of the outputs of the AE sensors lla and llb and the outputs of the acceleration sensors lie and lid are input to a foreign object identification index calculator 15 that calculates a foreign object identification index used for determining the type of sound source. Also, the amplified signals of the outputs of the AE sensors lla and llb are as follows.

It is input to the position evaluator 13. The evaluation result of the sound source position for each acoustic signal generation is input to the sound source distribution calculator 14, and the average position of the sound source and its standard deviation are determined. The minute foreign object discriminator 16 selectively determines whether it is a collision sound caused by the conductive minute foreign object 31 inside the GIS bus container 1 or not based on the foreign object identification index output from the calculator 15 and the sound source distribution output from the calculator 14. The cause of the sound is identified as described below.

Next, the position evaluator 13, which is the main part of the embodiment shown in FIG.
The sound source distribution calculator 14, the foreign object identification index calculator 15, and the minute foreign object discriminator 16 will be explained in detail.

FIG. 2 is a detailed configuration diagram of the position evaluator 13. timer 1
32 includes a counter (not shown), and a comparator 131
The clock signal of the clock generator 133 is counted using the pulse signal from either one of the comparators a and 131b as a count start signal, the counting is ended using the pulse signal from the other comparator as a count stop signal, and the counted value is calculated based on the time difference. It is designed to output to a distance converter 134. Comparator 131a.

131b are connected to the AE sensors lla and llb, respectively, and output a pulse signal to the timer 132 when the signals from each AE sensor lla and flb exceed a set value. Therefore, the clock signal count value of the timer 132 corresponds to the difference in distance between the source of the collision sound and each AE sensor lla, llb. In other words, the time difference/distance converter 134 outputs this distance difference. The clock generator 133 is
It is a pulse generator that is precisely adjusted so that the pulse generation period is a time reference.

The time difference/distance converter 134 has a count value/distance conversion coefficient setting value calculated in advance from the period of the clock signal and the sound speed of the GIS bus container 1, and has a function of multiplying the output of the timer 132. By combining circuits having these functions, it becomes possible to evaluate the position of the sound source with reference to the center position of the installation points of the AE sensors lla and llb.

FIG. 3 is a detailed configuration diagram of the sound source distribution calculator 14 shown in FIG. 1. The sound source distribution calculator 14 has a function of calculating the average value and standard deviation of the sound source positions.

When the input signal (output signal of the position evaluator 13) is input, a bypass filter 141 extracts only the fluctuation in the sound source position, a square calculator 142 calculates the square of the fluctuation,
Integration is performed by a low-pass filter 143, and a square root is determined by a square root calculator 144. On the other hand, the 0-wire configuration in which the average value of the sound source position is obtained by passing the input signal through the low-pass filter 145 can be realized with either an analog circuit or a digital circuit.

Next, the foreign object identification index calculator 15 will be explained.

The foreign object identification index is an index for determining whether the detected sound is the collision sound of a conductive minute foreign object or noise such as noise, and can be derived by calculation processing of the detected sound signal. . The impact sound of a hard, small foreign object contains more high-frequency components than the impact sound of a soft, large foreign object.
From the frequency distribution of the detected sound, it is possible to estimate to some extent what type of foreign object collision sound it is. The foreign object identification index is a representative value of the frequency distribution of the detected sound, and is defined as the ratio of the amplitude A of the high frequency component of the detected sound to the amplitude AL of the low frequency component. In this embodiment, the sound at one measurement point is measured by two sensors. The high frequency component of the acoustic signal is detected by the AE sensor.
Since the low frequency component is detected by an acceleration sensor, the foreign object identification index R is.

It can be calculated by This foreign object identification index takes a large value because the conductive micro foreign object to be detected is hard and small, and the collision excites many high frequency components, and it takes a small value for the impact sound of a large and soft object such as plastic. There is a tendency to take Hereinafter, the functions and operations of the foreign object identification index calculator 15 will be explained in detail using FIG. 4.

AE sensor lla, llb and acceleration sensor 11c
*The acoustic signal detected by the lid is first detected by the detector 151a.
~151d, respectively. AE sensor lla,
When the detection signal output of the detected acoustic signal of llb exceeds the set value, the output of the corresponding comparator 152, 156 becomes "11". At this time, the output of the corresponding timing circuit 153, 157 becomes "11'' for a certain period of time"1", and the peak values of the AE sensor output detection signal and the acceleration sensor output detection signal in that certain time width are the peak values of the peak holder 154.
a, 154b, 158a, 158b.

peak holders 154a, 154b and 158a,
The output of 158b becomes the input of dividers 155.159, and a foreign object identification index signal is obtained as the output of divider 155.159.

FIG. 5 shows a time chart of the foreign object identification index calculator 15. Since the correct value of the foreign object identification index cannot be obtained unless each peak value is held, the timing circuit 153
．． After the peak value is held at 157, a pulse with a fixed time width is output, and the dividers 155 and 159 are configured to output a foreign object identification index signal, which is the division result, only while this pulse is output. It is.

The micro foreign object discriminator 16 determines whether the detected sound is a collision sound of a conductive micro foreign object inside the GIS bus container 1 or a sound generated by other factors, based on the average value and standard deviation of the sound source position and the value of the foreign object identification index. and generates an alarm when the collision sound of a conductive micro foreign object inside the GIS bus bar container 1 is detected.

Figure 6 shows the difference in position evaluation results depending on the sound source.
The collision sound of the conductive minute foreign matter 31 inside the S busbar container 1 has a stable sound source position because the foreign matter 31 is regulated by the electric field and bounces (Fig. 2(a)). The sound source position ((b) in the same figure) varies. Also, electrical noise is detected by the AE sensor lla.

The sound source position appears to be extremely stable because it enters the measurement line of llb at the same time ((c) in the same figure).

For example, if a tag indicating the name of a part is hung on a string on the outside surface of the GIS bus container 1, and the tag swings in the wind and collides with the container, the length of the string may cause a relatively high noise source as shown in Figure 6(a). There may be cases where the variation in position becomes smaller. Furthermore, if the sound source is located far enough away from the installation interval between the AE sensors lla and llb, there is almost no difference in the arrival time of the sound between the two sensors, so the sound source location evaluation result shown in Figure 6(a) may occur. It's possible. This applies to thunder, etc.

In order to cope with such special cases, this embodiment uses two pieces of information: the sound source distribution and the foreign object identification index.
Collision sounds of minute conductive foreign objects inside the S busbar container 1 are selectively detected. That is, in the case of a sound that is generated very far away, such as thunder, the attenuation of high-frequency components is large because it propagates through the air, and the acoustic sensors lla to lid detect the sound as a sound containing many low-frequency components. Furthermore, tags and other tags installed on the outer surface of the tank are extremely large compared to the conductive micro foreign objects that are the object of detection, and the impact sound has a large low frequency component. Therefore, these sound sources can be identified by using a foreign object identification index.

The configuration and operation of the minute foreign matter discriminator 16 of this embodiment will be explained with reference to FIG. Among the foreign object identification index signals obtained at the two measurement points, the maximum value selection circuit 161 selects the larger value and makes it the output of the maximum value selection circuit 161. The sound is G
During the process of propagation through the IS bus bar container 1, the higher the frequency, the more likely it is to be attenuated, so the larger value of the foreign object identification indices is selected. The output of the comparator 163a becomes "1" when the foreign object identification index exceeds the set value. Comparator 16
The output of 3b becomes #11' when the value of the sound source position standard deviation signal (output signal of the sound source distribution calculator 14) is within the set range.

As a result, the output of the AND circuit 164a is such that the frequency distribution of the detected sound contains many high frequency components like the collision sound of a conductive micro foreign object, and the sound source distribution is not as narrow as electrical noise, but also over a wide range of the GIS bus container 1. It becomes "1" when the distance is not wide, such as when a foreign object collides with the object. However, in order to perform such an operation, the comparator 163a must be adjusted in advance through experiments or the like.

It is necessary to determine the setting value of 163b. Comparator 163c has the same function as comparator 163b. The differentiator 162 is for detecting a change in the sound source position from the average sound source position signal from the sound source distribution calculator 14. The comparator 163d becomes 11'' when the output of the differentiator 162 exceeds the set value.
The output of the circuit 164b is such that the sound source distribution is not so wide that a foreign object collides with the outer surface of the GIs busbar vessel 1, but not so narrow in the case of electrical noise, that the sound source position changes, and that the detected sound contains many high-frequency components. If it is included, it becomes 111 It.

That is, by providing this determination section, foreign matter detection becomes possible even in a transient state in which a conductive minute foreign matter inside the GIS busbar container moves. When the internal minute foreign matter 31 moves, when the electric field strength varies in the axial direction of the busbar container 1 shown in FIG. Possibly moving to a higher location. The output of the OR circuit 165 is determined when the detected sound contains many high frequency components and the variation in the sound source position is small, or when the detected sound contains many high frequency components and the variation in the sound source position is relatively small.
In FIG. 7, the comparator 163b and the comparator 163c become "1" when the sound source position moves.
have the same function, but since the comparator 163c assumes that the sound source moves, the comparator 163c
The setting value is larger than that of 3b.

As described above, in detecting minute foreign objects inside the GIS bus container 1, information on the sound source distribution is used to determine whether the sound source is inside or outside the GIS bus container 1. It becomes possible to distinguish between the sound of a conductive foreign object colliding with the inner surface and the sound of a conductive foreign object colliding with the inner surface.

The various circuits and parts used in this example are as follows.

All can be achieved using existing products or a combination of them. However, it goes without saying that the functions of these various circuits can be replaced by computer software processing.
In this embodiment, two acoustic sensors are used to measure the foreign object identification index, but the foreign object identification index can also be measured using a broadband AE sensor. Further, the type of sound source can be discriminated by comparing the frequency distribution itself without using a foreign object identification index. The sound source position estimation can also be realized by other means such as the correlation method.

Effects specific to this embodiment include the following.

(1) By using a method for determining the type of sound source based on the frequency distribution of the detected sound, malfunctions of the micro foreign object detector can be further reduced, which has the effect of improving the reliability of the micro foreign object detector.

(2) By representing the sound source distribution with the standard deviation and the frequency distribution with the foreign object identification index, pattern matching processing is no longer necessary, which simplifies signal processing and has the effect of simplifying the micro foreign object detector. There is.

A second embodiment of the present invention will be described with reference to FIG.

FIG. 8(a) is a configuration diagram of a minute foreign matter detection device that detects foreign matter inside the GIS bus bar container 1. FIG. In this example, G
The distortion of the GIS busbar vessel 1 is used to identify whether the collision sound is inside the IS' busbar vessel 1 or from outside. FIG. 8(b) shows that when a minute foreign object 31 collides with the container 1 from within the GIS bus container 1, the AE sensor II
It is a figure which showed the amplitude change of the collision vibration detected by a. On the other hand, FIG. 8(c) shows that when a minute foreign object 3o collides with the container 1 from outside the GIS bus container 1, the AE sensor lla
It is a figure showing the amplitude change of the collision vibration detected by. The collision vibration is attenuated as it propagates through the container 1. However, to which side of the container 1 does the initial amplitude swing?

That is, whether the initial phase is positive or negative depends on whether the minute foreign matter collides with the container 1 from inside or outside. The positive and negative of this initial phase (of course, AE
The positive and negative values are reversed depending on whether the sensor is provided on the outer surface or the inner surface of the container 1. ) is preserved regardless of vibration propagation. In this embodiment, by detecting whether the initial phase is positive or negative, it is possible to identify whether the minute foreign matter colliding with the GIS busbar container 1 is internal or external.

In FIG. 8(,), sound is detected by the AE sensor lla installed on the outer surface of the GIS busbar container 1, and the amplifier 1
Amplify by 2a. The AE sensor lla outputs a positive voltage (the phase corresponds to positive) at the compressive force.

A material having a characteristic of generating a negative voltage (the phase corresponds to the negative voltage) with tension is used. The output of the comparator 201 becomes "1" when the input signal exceeds the set value, while the output of the comparator 202 becomes "'1"' when the input signal is below the set value. For this reason, the AE sensor lla
When the comparator 201 detects the compressive force, the output of the comparator 201 becomes “
1'', and when the tension is detected, the output of the comparator 202 becomes 11171.One-shot circuit 203a
, 203b has a function of holding the output at "1" for a certain period of time after the comparator output becomes "1" and has a configuration of 0 or more, so that the foreign object 31 collides inside the GIS bus bar container 1. In this case, the output of the one-shot circuit 203a becomes "1";
The output of b becomes "1". This is because when the foreign object 31 collides with the inner surface of the GIS bus container 1, the GIS bus container 1 is distorted outward, a compressive force is applied to the AE sensor lla, and the foreign object 3o hits the outer surface. When the collision occurs, the GIS busbar container 1 is distorted inward A
This is because tension acts on the E sensor lla. After compressive force and tension work, tension and
Compressive force acts, and positive and negative voltages appear alternately in the output of the AE sensor lla, but the one-shot circuits 203a and 2
03b does not respond at this time, and its output is determined depending on whether the first force applied is compression or tension.With the configuration described above, it is possible to determine whether the collision sound occurred inside the GIS busbar container 1 or outside. Recognize. Furthermore, if a method using a foreign object identification index similar to that of the first embodiment is added to this embodiment, it goes without saying that the noise removal ability can be further improved.

The unique effects of this embodiment are that the device configuration is simple;
One advantage is that the micro foreign object detection device can be simplified.

A third embodiment of the present invention will be described with reference to FIG.

FIG. 9 is a block diagram of a micro foreign object detection device that detects foreign objects in the GIS busbar section. Differences in temporal fluctuations in the rate of sound generation are used.

In FIG. 9, sound is detected by an AE sensor lla installed on the outer surface of the GIS busbar container 1, and amplified by an amplifier 12a. The output of the comparator 211 is It 17 when the input signal (amplifier 12a output) exceeds the set value.
It becomes 1. The sound occurrence rate calculator 212 includes a comparator 2
11 outputs a voltage corresponding to the frequency at which the output becomes 111", that is, the sound generation rate per unit time. Effective value calculator 2
13 outputs a voltage according to the variation range of the output voltage of the arithmetic unit 212, that is, the variation range of the sound generation rate. The comparator 214 outputs '1' when the output voltage of the effective value calculator 213 is below the set value.

In other words, in this embodiment, when the fluctuation range of the sound generation rate is small, it is set to 111''.

The number of collisions per unit time of conductive micro foreign objects 31 inside the GIS bus container 1 that are blown away by Coulomb force is relatively constant, whereas the number of foreign objects that fly up due to wind etc. and collide with the outer surface of the GIS bus container 1 is relatively constant. The number of collisions per hour varies widely depending on wind speed fluctuations, etc. Therefore, when the fluctuation range of the sound generation rate is small, the output of the comparator 214 is 1", it is natural that the noise removal ability will be further improved if a method using a foreign object identification index is added thereto, as in the first embodiment.

The unique effects of this embodiment are that the device configuration is simplified;
The reason for this is that it can simplify the micro foreign object detection device.

A fourth embodiment of the present invention will be described with reference to FIG.

FIG. 10 is a block diagram of a micro foreign object detection device that detects foreign objects in the GIS bus bar. Judgment is made based on the degree of correlation between incidence and wind speed.

The measurement system in FIG. 10 is the same as that in FIG. 9 up to the sound occurrence rate calculator 212. A correlation calculator 222 calculates the correlation between the wind speed signal output from the anemometer 221 and the sound occurrence rate signal. The correlation strength determiner 223 searches for the maximum value of the correlation strength of the calculation result of the correlation calculator 222, and sets its output to '1'' when the maximum value is less than a set value.

In other words, if there is a correlation between wind speed and sound generation rate, the sound is GI
If the sound is generated outside the S bus vessel 1 and is not related to the wind speed, it is determined that the sound is occurring inside the GIS bus vessel 1.6 In this embodiment, as in the first embodiment, this If a method using a foreign object identification index is added to the method, the noise removal ability will be further improved.

A unique effect of this embodiment is that it is possible to eliminate noise sources that are directly caused by wind and are currently not expected, thereby improving the reliability of the micro foreign object detection device.

A fifth embodiment of the present invention will be explained with reference to FIG.

FIG. 11 is a configuration diagram of a micro foreign object detection device that detects foreign objects in a GIS busbar section. In this example, the collision sound is
The frequency distribution of the detected sound level is used to identify whether the sound has occurred inside or outside the busbar vessel 1.

In FIG. 11, sound is detected by the AE sensor lla,
The signal is amplified by the amplifier 12a. Sound level distribution calculator 2
31, the number of occurrences of each level of sound is counted.

The frequency distribution determiner 232 searches whether the occurrence frequency tends to decrease as the sound level increases, and -
If there is a decreasing trend, the output is set to “1” @
The number of foreign objects flying around inside the GIS bus bar container 1 due to Coulomb force is small.

Since these are often collision sounds of the same foreign object, there is a certain tendency in the frequency distribution of the collision sound levels.

On the other hand, since the foreign objects that collide with the outer surface of the GIS busbar container 1 vary in size, material, etc., the frequency distribution of the collision sound level also varies. Figure 12 (
Figures a) and (b) are examples of measuring the frequency distribution of internal and external impact sound levels. Since sounds below the threshold are not counted, sounds below the threshold are not shown. From this figure, it can be seen that the configuration shown in FIG. 11 makes it possible to identify whether the collision sound is generated inside or outside the GIS busbar container. In this embodiment as well, as in the first embodiment, if a method using a foreign object identification index is added to this, the noise removal ability is further improved.

A unique advantage of this embodiment is that information regarding the mass of the colliding object can be obtained from the collision sound level distribution, and a function for estimating the amount of foreign matter can be added to the minute foreign matter detection device.

A sixth embodiment of the present invention will be described with reference to FIG.

FIG. 13 is a configuration diagram of a micro foreign object detection device that detects foreign objects in a GIS bus bar section. In this example, the collision sound is
In order to identify whether the sound is generated inside or outside the bus bar container 1, it is determined whether the sound source is at the bottom of the GIS container or elsewhere by evaluating the position of the sound source.

If there is a sound source other than the bottom of the GIS container, all GIS
It is determined that the sound is coming from outside the bus bar container 1.

In FIG. 13, three AE sensors lla, flb, and lle are installed on the outer surface of the GIS busbar container 1, and a six-position evaluator 241 amplifies each sensor output with amplifiers 12a, 12b, and 12e. Calculated from the difference in arrival time of sound from the sensor. Collision position determiner 2
42, whether the sound source position is at the bottom of the GIS bus container 1 or elsewhere is calculated from the coordinate values obtained from the sound source position evaluation result, and when the sound source position is at the bottom of the GIS bus container 1, the output is '1''. ', and it is determined that there is a foreign object 31 inside.The conductive minute foreign object inside the GIS bus container 1 repeatedly moves up and down between the center conductor 2 and the container 1 due to Coulomb force and gravity, so the collision position is the GIS bus container. It will be at the bottom of 1.

On the other hand, foreign objects carried by the wind etc. collide with the GIS busbar container 1 at an angle determined by the direction of the wind and the resultant force of gravity.

The probability that the collision location is other than the lower part of the GIS busbar container 1 is extremely high. In this embodiment as well, as in the first embodiment, if a method using a foreign object identification index is added to this, the noise removal ability is further improved.

An advantage unique to this embodiment is that the location of the sound source, that is, the location where the conductive foreign object is present, can be determined, making it easier to formulate a repair plan after detection.

In each of the embodiments described above, various determinations are made using dedicated individual circuits, but as described in the first embodiment,
Since the functions of these dedicated circuits can be performed by a computer, it is of course possible to construct a minute foreign matter detection device by using a normal computer. Further, in each of the above embodiments, a GIS bus bar container was explained as an example, but it goes without saying that the micro foreign object detection device of the present invention can be applied to other stationary electrical equipment, such as a high voltage transformer. Furthermore, it goes without saying that the present invention can be applied not only to high-voltage stationary electrical equipment, but also to cases where minute foreign objects inside are bounced around due to internal magnetic fields.

In the micro foreign object detection device explained in Fig. 1, the AE sensor l
The sensor la and the acceleration (AC) sensor lie are placed close to each other, and the AE sensor llb and the acceleration sensor lid are placed close to each other, and these are connected to the sensor 11a.

It is installed on the outer surface of the busbar container 1 at a predetermined distance in front of the busbar container 11c. If the object to which each sensor is installed is small, the number of sensors to be installed can be small, but if the object is large, the number of sensors will increase accordingly, and installation will take time. Further, it is preferable that the AE sensor and the AC sensor are arranged close to each other and detect acoustic waves at the same location. Therefore, care must be taken when installing the sensor. Therefore, in the present invention, as shown in FIG. 14, one sensor 4 simultaneously detects the low frequency component (corresponding to the AC sensor output) and the high frequency component (corresponding to the AE sensor output) of the acoustic wave. ,
After amplifying the detection signal with preamplifier 5.

In this embodiment, a signal equivalent to the AC sensor output is extracted by a filter 6 that passes a signal from 1kHz to 50kHz, and a signal corresponding to the output of the AC sensor is extracted.
Filter 7 that passes signals from 00kHz to 300k)Iz
Signals corresponding to the AE sensor output are extracted, and these signals are used as the sensor signals in each of the embodiments described above.

As the above-mentioned sensor 4, for example, a piezoelectric element shown in FIG. 15(a) is used. The piezoelectric element shown in FIG. 15(a) is a thin semicircular piezoelectric element 4a that converts high frequency components of acoustic waves into corresponding electrical signals.

It is constructed by combining thick semicircular piezoelectric elements 4b into a single disk shape, which converts low frequency components of acoustic waves into corresponding electrical signals. And 1 which does not show these
The detection signal is taken out from the terminal connected to each surface of the piezoelectric elements 4a and 4b and inputted to the preamplifier 5. It goes without saying that the detection signals of the piezoelectric elements 4a and 4b may be extracted and used separately. FIG. 15(b) is a diagram showing another configuration of the sensor 4 explained in FIG. 14. In this example, 1
Using a disc-shaped piezoelectric element 4C, a detection signal corresponding to the high frequency component of the acoustic wave is extracted from the thickness direction, and a detection signal corresponding to the low frequency component of the acoustic wave is extracted from the diameter direction. . By using such a sensor,
Even if the stationary electrical equipment to be detected is large, the sensor can be easily and quickly attached.

[Effects of the Invention] According to the method and device for detecting foreign objects in stationary electrical equipment of the present invention, collision sounds caused by foreign objects in stationary electrical equipment can be discriminated and detected from external noise, so partial discharges in stationary electrical equipment can be detected in advance. It becomes possible to take measures to avoid this. Further, according to the collision sound sensor of the present invention, since signals corresponding to the AE sensor output and the AC sensor output can be extracted with one sensor, the sensor can be installed easily and quickly.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a micro foreign object detection device according to a first embodiment of the present invention applied to a GIS busbar container, FIG. 2 is a detailed block diagram of the position evaluator shown in FIG. 1, and FIG. FIG. 4 is a detailed configuration diagram of the sound source distribution calculator shown in FIG. 1, FIG. 5 is a time chart explaining the operation of the foreign object identification index calculator, and FIG. Figure (a),
(b) and (c) are graphs showing the sound source location evaluation results based on GIS internal foreign matter, external foreign matter, and electrical noise, respectively. Figure 7 is a detailed configuration diagram of the minute foreign matter discriminator shown in Figure 1. Figure 8 (a)
) is a configuration diagram of a micro foreign object detection device according to a second embodiment of the present invention, FIG. FIG. 10 is a configuration diagram of a minute foreign matter detection device according to a fourth embodiment of the present invention, and FIG. 11 is a configuration diagram of a minute foreign matter detection device according to a fifth embodiment of the present invention. Figure 12(a),
(b) is a graph showing the frequency distribution of collision sounds inside and outside the GIS,
FIG. 13 is a configuration diagram of a minute foreign matter detection device according to the sixth embodiment of the present invention, and FIG. 14 is the present invention (invention according to claim 13)
FIG. 15(a) is a diagram showing a configuration for separating high-frequency components and low-frequency components of a detection signal when using a sensor according to the above.
(b) is a configuration diagram of a sensor according to an embodiment of the present invention. 1... OIS bus bar container, 2... Center conductor, 31...
・Internal minute foreign matter, 3o... External minute foreign matter, 4... Sensor, 4 a + 4 b + 4 c... Piezoelectric element, 5... Preamplifier, 6°7... Filter, lla
, llb, 1ie-AE sensor, 11c. lid... AC sensor, 12a-12d... amplifier, 13... position evaluator, 14... sound source distribution calculator,
15... Foreign object identification index calculator, 16... Minute foreign object discriminator.

Claims (1)

[Scope of Claims] 1. Distinguishing the sound of a minute foreign object bouncing under the regulation of an electric or magnetic field generated inside a stationary electric device from external noise, and detecting the presence of a minute foreign object inside the stationary electric device. A method for detecting a minute foreign object in a stationary electrical device, which is characterized in that the detected collision sound is distinguished as to whether it is a collision sound caused by an internal minute foreign object or an external noise based on the characteristics of the detected collision sound. Method for detecting minute foreign objects inside. 2. A collision sound sensor that detects the sound of a minute foreign object colliding with the stationary electric device and the sound of a minute foreign object outside the stationary electric device bouncing under the regulation of an electric or magnetic field generated inside the stationary electric device; A stationary electrical appliance characterized by comprising means for determining the collision sound characteristics from the detection signal of the collision sound sensor, and means for determining the presence or absence of minute foreign matter inside the stationary electrical equipment from the collision sound characteristics determined by the means. Device for detecting minute foreign objects inside equipment. 3. In claim 2, the distribution of sound source positions is determined as the collision sound characteristics determined from the detection signal of the collision sound sensor, and when the distribution of the sound source positions varies from a predetermined distribution, it is determined that the collision sound is not caused by a minute foreign object inside. A device for detecting minute foreign objects in stationary electrical equipment. 4. The micro foreign object detection device in a stationary electrical equipment according to claim 3, wherein the standard deviation value is used as the distribution value of the sound source position used for determination. 5. In claim 2, the collision sound characteristic obtained from the detection signal of the collision sound sensor is obtained by determining the positive or negative of the first phase of the detection signal, and when viewed from the mounting position of the collision sound sensor, the phase is determined based on the collision of an internal minute foreign object. 1. A micro foreign object detection device in a stationary electric device, characterized in that it determines whether or not it is caused by. 6. In claim 2, the magnitude of the temporal fluctuation of the collision sound generation rate is determined as the collision sound characteristic determined from the detection signal of the collision sound sensor, and when the temporal fluctuation is smaller than a predetermined value, it is determined that there is a minute foreign object inside. A device for detecting minute foreign objects in stationary electrical equipment. 7. In claim 2, the intensity distribution of the detection signal is determined as the collision sound characteristic determined from the detection signal of the collision sound sensor, and if the intensity distribution has less variation than a predetermined distribution, it is determined that there is a minute foreign object inside. A device for detecting minute foreign objects in stationary electrical equipment. 8. In a stationary electrical equipment according to claim 2, the source position of the collision sound is determined from the detection signal of the collision sound sensor, and it is determined whether the foreign object collided at the sound source position is internal. Micro foreign object detection device. 9. A collision sound sensor for detecting the collision sound of a minute foreign object bouncing around the stationary electric device under the regulation of an electric or magnetic field generated inside the stationary electric device and the collision sound caused by a minute foreign object outside the stationary electric device; a wind speed detection device that detects the wind speed of an environment in which a stationary electrical device is installed; a means for determining the collision sound characteristics from a detection signal of the collision sound sensor; and a method for detecting the collision sound characteristics determined by the means by the wind speed detection device. and means for determining whether or not the collision sound detected by the collision sound sensor is caused by an internal minute foreign matter based on whether or not it depends on the wind speed. 10. A collision sound sensor that detects the collision sound of a minute foreign object against the stationary electric device that bounces under the regulation of an electric or magnetic field generated inside the stationary electric device, and the collision sound caused by a minute foreign object outside the stationary electric device; means for determining the type of the sound source from the frequency distribution of the detection signal of the collision sound sensor; means for determining the source position of the collision sound from the detection signal of the collision sound sensor; and the collision sound sensor based on the sound source position and the type of the sound source. A device for detecting minute foreign objects in stationary electrical equipment, comprising means for determining whether a collision sound detected by the device is caused by a minute foreign object inside. 11. Claim 10, further comprising means for determining a change in the source position of the collision sound, where the type of the source of the collision sound is a minute foreign object that is considered to exist inside and the degree of change in the sound source position is below a predetermined value. 1. A device for detecting minute foreign matter in a stationary electrical device, characterized in that it determines that a minute foreign matter is present inside a stationary electrical device. 12. The device for detecting minute foreign matter in a stationary electrical device according to any one of claims 2 to 11, wherein the stationary electrical device is a gas-insulated device. 13. A collision sound sensor used in the device for detecting minute foreign objects in stationary electrical equipment according to any one of claims 2 to 12, wherein an element for converting low-frequency sound into an electrical signal and a high-frequency sound are contained in one container. A collision sound sensor characterized by housing an element that converts sound into an electric signal.