JP2022010761A - Method for determining insulation deterioration accompanied by electrical tree progress - Google Patents

Abstract

To provide a method for determining the insulation deterioration of an epoxy mold power device in consideration of the degree of progress of an electric tree.SOLUTION: A method for determining the insulation deterioration accompanied by the electric tree progress includes the steps of: detecting an electric signal related to the partial discharge generated from an epoxy mold power device and obtaining the φ-q distribution which is a relationship between an applied voltage phase angle φ and the discharge electric charge amount q; calculating a dispersion value Var relating to the discharge charge amount q from the result of the φ-q distribution obtained in the step; and determining that the electric tree is progressed in the epoxy mold power device when the dispersion value Var exceeds a predetermined threshold value.SELECTED DRAWING: Figure 10B

Description

In power receiving and transforming equipment such as transformers and high-voltage boards, it is important to predict the life and prevent accidents. For this reason, predictive maintenance has been proposed in which measures are taken in advance, paying attention to precursory phenomena that lead to equipment deterioration and accidents that occur in power receiving and transforming equipment. One of such precursors that occurs in power receiving / transforming equipment is partial discharge that occurs when the insulation function in the power receiving / transforming equipment deteriorates.

If a defect such as a void is present in the insulator of an electric power device, a partial discharge occurs. Specifically, a partial discharge occurs when the voltage applied to the defect exceeds the critical voltage or spark voltage due to partial dielectric breakdown in the defect. If this partial discharge is repeated, the insulator will eventually break down as a whole, and there is a possibility that a power failure or other trouble may occur in power transmission and distribution. On the other hand, it is extremely complicated to stop power transmission and distribution for detecting this partial discharge.

Therefore, in order to detect the partial discharge of the insulator without stopping the power transmission and distribution, that is, in the live-line state, a method of detecting the partial discharge by detecting the ultrasonic wave generated at the time of the partial discharge has been proposed ( For example, see Patent Document 1 below).

Japanese Unexamined Patent Publication No. 2004-335953

By the way, in electric power equipment such as transformers, there are those in which insulating oil is used as an insulating material and those in which a solid insulating material such as an epoxy resin is used. In the case of an electric power device having such a solid insulating material, if the state in which partial discharge occurs continues, the insulating material is eroded in a dendritic shape to generate an electric tree. As a result, the dielectric strength is reduced, and eventually the entire road is destroyed.

Once an electric tree is generated in an electric power device, this electric tree does not disappear and progresses over time. If the electric tree is fully developed, electric power equipment may be destroyed in the near future. If all roads are destroyed in an energized state, a power outage accident due to a ground fault will occur, so all road destruction is an event that should be avoided.

On the other hand, even when an electric tree is generated in an electric power device, depending on the degree of progress, there are some that require urgent replacement work of the electric power device and some that do not require such urgent work. .. The replacement work of electric power equipment involves a power outage work, and the work inevitably incurs costs. Furthermore, since the number of electric power devices is enormous, it is not practical to perform replacement work for all electric power devices. Therefore, it is preferable to preferentially replace the electric power equipment that needs to be replaced. However, at this time, there is no known method for ranking the need for replacement of these electric power devices.

In view of the above problems, it is an object of the present invention to provide a method for determining insulation deterioration of an epoxy molded electric power device in consideration of the degree of progress of an electric tree.

The present invention is a method for determining insulation deterioration due to the progress of an electric tree in an epoxy molded electric power device.
The step (a) of detecting the electric signal related to the partial discharge generated from the epoxy mold power device and obtaining the φ−q distribution which is the relationship between the applied voltage phase angle φ and the discharge charge amount q.
From the result of the φ−q distribution obtained in the step (a), the step (b) of calculating the dispersion value Var related to the discharge charge amount q, and
When the dispersion value Var exceeds at least a predetermined threshold value, it is characterized by having a step (c) of determining that an electric tree is developing in the epoxy mold power device.

According to the diligent research of the present inventor, it has been found that when the partial discharge progresses to the extent that the electric tree progresses in the epoxy mold power device, the dispersion value of the discharge charge amount q obtained from the φ−q distribution increases extremely. rice field. That is, according to the above method, by confirming that the value of the dispersion value Var of the discharge charge amount q obtained from the φ−q distribution exceeds a predetermined threshold value, the degree of insulation deterioration of the epoxy molded power device is confirmed. It can be judged that there is a high possibility that

Since the insulating portion of the epoxy molded power device is normally closed, the degree of development of the electric tree generated in the insulating portion cannot usually be visually determined. For this reason, even if it is possible to detect that a partial discharge has occurred, the priority of the replacement work may be lowered because the epoxy mold power equipment needs to be replaced preferentially and has deteriorated to some extent. It is not possible to know whether the degree of deterioration is such that it can be allowed to occur.

On the other hand, according to the above method, when the dispersion value Var is a value exceeding the threshold value, the progress of the electric tree can be detected, so that the deterioration of the epoxy molded power device has progressed to some extent. Can be recognized. Therefore, by using this method, it can be useful for the replacement work plan of the epoxy mold power equipment at each site.

Further, the above method can be executed by detecting the electric signal related to the partial discharge, measuring the φ−q distribution based on the electric signal, and obtaining the dispersion value by calculation from this result. Therefore, the determination work can be performed without a power outage and by simply bringing simple equipment to the site.

As a specific example, the step (a) includes a step of installing an antenna in the vicinity of the epoxy mold power device and a step of receiving a radio wave signal by the antenna, and is based on the strength of the received radio wave signal. This can be a step of obtaining the φ−q distribution.

Further, as another example, the step (a) includes a step of measuring the current flowing through the ground wire of the epoxy mold power device with a current sensor, and the φ−q distribution is measured based on the intensity of the measured current. It can be a step to obtain. As the current sensor, for example, high frequency CT can be used.

The predetermined threshold value may be, for example, a predetermined value of 150 or more and 250 or less. As an example, the predetermined threshold value can be 200.

According to the diligent research of the present inventor, it can be estimated that when the dispersion value Var exceeds 200, the tree length of the electric tree is advanced by about 1.5 mm. Although it varies depending on the value of the electric power used, the thickness of the epoxy mold resin used as an insulating member in the epoxy mold electric power device is about 5 mm to 15 mm. By setting a predetermined threshold value to the above value, it can be detected that the tree length of the electric tree reaches about 10% to 30% or more of the thickness of the epoxy mold resin, so that it is highly necessary to replace the device. It can be detected in advance, and the risk of destruction of all roads can be prevented.

If it is possible to detect whether or not partial discharge has occurred in the epoxy mold power equipment installed at each site, the above method should be applied only to the power equipment in which partial discharge has occurred. It does not matter if it is executed.

According to the present invention, it is possible to certify an epoxy molded electric power device having a particularly high need for replacement.

It is a drawing schematically showing one embodiment of the method of determining the insulation deterioration with the progress of the electric tree of the epoxy mold electric power device. It is a block diagram which shows the aspect of FIG. 1 more schematically. It is a drawing schematically showing one embodiment of the method of determining the insulation deterioration with the progress of the electric tree of the epoxy mold electric power device. It is a block diagram which shows the aspect of FIG. 3 more schematically. It is a drawing which shows the structure of an experimental system schematically. It is a schematic drawing which shows the structure of the test piece for generating a partial discharge. It is a photographed image of a test piece by a macroscope. FIG. 6 is a graph showing a φ−q distribution based on a current signal detected by a partial discharge analyzer at the time of imaging in FIG. 6A. It is a photographed image of a test piece by a macroscope. FIG. 7 is a graph showing a φ−q distribution based on a current signal detected by a partial discharge analyzer at the time of imaging in FIG. 7A. It is a photographed image of a test piece by a macroscope. FIG. 8 is a graph showing a φ−q distribution based on a current signal detected by a partial discharge analyzer at the time of imaging in FIG. 8A. It is a graph which shows the relationship between the dispersion value Var which concerns on the discharge charge amount q obtained from the result of φq distribution, and the electric tree length. It is a graph which shows the φ−q distribution based on the current signal measured by the actual device 5b. It is a graph which shows the φ−q distribution based on the current signal measured by the actual device 5c.

An embodiment of a method for determining insulation deterioration due to the progress of an electric tree in an epoxy molded electric power device of the present invention (hereinafter, abbreviated as “this determination method” as appropriate) will be described with reference to the drawings as appropriate. In each drawing, the dimensional ratio in the drawing and the actual dimensional ratio do not always match.

FIG. 1 is a drawing schematically showing an example of an embodiment of this determination method. FIG. 1 shows, as an example, a case where the epoxy mold power device to be diagnosed for deterioration is the transformer 5 in the cubicle 3.

The epoxy mold power device refers to a power device containing an insulator made of an insulating material containing an epoxy mold resin in the insulator material. Examples of such electric power devices include transformers, switches, circuit breakers, bushings and the like.

In the embodiment shown in FIG. 1, the antenna 11, the discharge measuring instrument 12, and the computer 13 are used when executing this determination method. In FIG. 1, the case where the antenna 11 is installed in the vicinity of the transformer 5 in the cubicle 3 is illustrated. The antenna 11 and the discharge measuring instrument 12 are connected by a cable 10. Even when the opening / closing door 4 of the cubicle 3 is closed, the cable 10 can be laid to the outside of the cubicle 3 through the gap of the opening / closing door 4. If there is no gap in the opening / closing door 4, the opening / closing door 4 may not be completely closed, but a gap may be opened so that the cable 10 can pass through.

FIG. 2 is a block diagram illustrating the configuration of FIG. 1 more schematically. The discharge measuring instrument 12 has a waveform analysis unit 12a, a phase analysis unit 12b, and a display unit 12c. The computer 13 has an arithmetic processing unit 13a, an information output unit 13b, and a display unit 13c.

The waveform analysis unit 12a is a calculation means for extracting a signal in a predetermined frequency band from the radio wave signal received by the antenna 11 and detecting the intensity thereof. It is known that when the transformer 5 which is an epoxy mold power device generates a partial discharge, it emits a radio wave signal in the several MHz band to several tens of MHz band (for example, 4 MHz to 40 MHz). Therefore, the waveform analysis unit 12a has a function of detecting the strength of the signal in the frequency band included in the radio wave signal received by the antenna 11.

A known technique can be used as a method for detecting the strength of a signal in the above frequency band included in the radio wave signal from the received radio wave signal. For example, the intensity of the signal in the above frequency band may be extracted by spectrally analyzing the signal, or only the signal in the above frequency band may be extracted by passing through a filter that shields the signal other than the above frequency band. It may be used to detect the strength.

The phase analysis unit 12b is a calculation means for analyzing the voltage waveform of the radio wave signal received by the antenna 11. Specifically, the phase analysis is performed by recognizing the temporal change of the voltage value indicated by the radio wave signal. Specifically, the phase analysis unit 12b detects the intensity of the signal in the several tens of MHz band included in the radio wave signal analyzed by the waveform analysis unit 12a for each phase angle φ.

The display unit 12c is a monitor for visually recognizing the analysis result by the waveform analysis unit 12a and the analysis result by the phase analysis unit 12b. However, it is arbitrary whether or not the discharge measuring instrument 12 includes the display unit 12c.

The arithmetic processing unit 13a derives a discharge charge amount q corresponding to the phase angle φ based on the intensity of the radio wave signal for each phase angle φ analyzed by the phase analysis unit 12b, and has a φ−q distribution (“q−φ”). Also called "distribution"). Further, based on the result of this φ−q distribution, the dispersion value Var related to the discharge charge amount q is calculated.

Further, the arithmetic processing unit 13a compares the calculated dispersion value Var with a predetermined threshold value, and based on the comparison result, the deterioration of the measurement target device (transformer 5 in the cubicle 3 in the example of FIG. 1) is deteriorated. Create information that indicates the degree. The information output unit 13b and the display unit 13c are means for outputting the calculation result by the calculation processing unit 13a. For example, the information output unit 13b outputs information on the calculation result to another computer, smartphone, or printer by wire or wirelessly. Further, the display unit 13c is composed of a monitor and displays information regarding the calculation result. When the discharge measuring instrument 12 includes the display unit 12c, this information may be output to the display unit 12c.

The calculation result referred to here includes at least information indicating the degree of deterioration of the device to be measured (in the example of FIG. 1, the transformer 5 in the cubicle 3). As other information, the dispersion value Var related to the discharge charge amount q may be included, or information related to the φ−q distribution may be included. Further, the value of the tree length of the electric tree traveling in the measurement target device estimated from the value of the dispersion value Var may be included.

If the discharge measuring instrument 12 has a function corresponding to the arithmetic processing unit 13a, the computer 13 may not be used when carrying out this determination method.

Hereinafter, an example of the procedure for executing this determination method will be described.

(Step S1)
The antenna 11 is arranged in the vicinity of the epoxy mold power device (here, the transformer 5) to be diagnosed. When the transformer 5 is deteriorated and partial discharge occurs, a predetermined radio wave signal is transmitted from the transformer 5, so that the antenna 11 is arranged at a position where the radio wave signal can be received. Specifically, the arrangement position of the antenna 11 is preferably a region within 3 m from the outer surface of the transformer 5, and more preferably a region within 2 m. At this time, the antenna 11 may be fixed by a magnet or the like.

As shown in FIG. 1, when the target electric power device (transformer 5) is arranged in a housing such as the cubicle 3, the antenna 11 is arranged in the housing and then the opening / closing door is opened. It is preferable to close 4. As a result, the reception sensitivity of the environmental radio waves existing outside the cubicle 3 by the antenna 11 is lowered, so that misdiagnosis due to disturbance is suppressed.

(Step S2)
The antenna 11 and the discharge measuring instrument 12 are connected by a cable 10, and the radio wave signal received by the antenna 11 is output to the discharge measuring instrument 12. The discharge measuring instrument 12 is connected to the computer 13 so that information can be input and output. The discharge measuring instrument 12 and the computer 13 may be connected by wire or wirelessly.

(Step S3)
The waveform analysis unit 12a provided in the discharge measuring instrument 12 extracts a signal in the tens of MHz band from the radio wave signals received by the antenna 11. After that, the phase analysis unit 12b provided in the discharge measuring instrument 12 analyzes the intensity of the extracted signal for each phase angle φ.

(Step S4)
The arithmetic processing unit 13a provided in the computer 13 derives the discharge charge amount q according to the phase angle φ based on the signal strength for each phase angle φ analyzed by the phase analysis unit 12b, and obtains the φ−q distribution. Calculate. The above steps S1 to S4 correspond to the step (a).

(Step S5)
The arithmetic processing unit 13a provided in the computer 13 calculates the dispersion value Var related to the discharge charge amount q based on the result of the φ−q distribution. This step S5 corresponds to the step (b).

(Step S6)
The arithmetic processing unit 13a provided in the computer 13 compares the value of the variance value Var with the threshold value Th. When the dispersion value Var exceeds the threshold value Th, the arithmetic processing unit 13a determines that the electric tree has advanced to the device to be measured and the deterioration has progressed. This step S6 corresponds to the step (c).

As shown in FIG. 3, a current sensor 11a may be used instead of the antenna 11 to measure the grounding current flowing through the grounding wire 10a. High frequency CT can be used as the current sensor 11a. In this case, as shown in FIG. 4, the information regarding the ground current detected by the current sensor 11a is output to the discharge measuring instrument 12. The waveform analysis unit 12a provided in the discharge measuring instrument 12 extracts a signal in the tens of MHz band from the current signals detected by the current sensor 11a. The following is common to the above-described embodiment.

[inspection]
Hereinafter, the point that the tree length of the electric tree generated in the device to be measured can be estimated from the dispersion value Var related to the discharge charge amount q will be described based on the experimental results.

(Verification 1)
FIG. 5A is a schematic drawing showing the configuration of the experimental system. A container 32 containing a plate-shaped test piece 33 was installed on a test table 31 placed in the test space 30. The test piece 33 was housed in the container 32 with insulating oil filled in the container 32 so that creeping discharge did not occur in the test piece 33.

The container 32 is made of transparent acrylic. The test piece 33 inside the container 32 was observed and photographed by a macroscope 38 (Z16APO manufactured by Leica, zoom range 0.57 to 9.2 times) installed on the outside of the container 32. The test piece 33 was installed on the insulating sheet 41.

A transformer 37 to which a power supply Vcc was connected was installed in the test space 30. A high voltage of 14 to 17 kV was applied to the test piece 33 at 60 Hz via the transformer 37. A ground wire 33a is connected to the test piece 33, and the current flowing through the ground wire 33a is detected by a high frequency CT39 (I-125-1HF manufactured by PRODYN, 120 kHz to 600 MHz).

The current signal detected by the high frequency CT39 is detected by the partial discharge analyzer 34 (DAC-PD-9 manufactured by Soken Electric Co., Ltd., 40 kHz to 40 MHz). The phase of the current signal detected by the partial discharge analyzer 34 is detected by the coupling capacitor 35 connected to the same node as the test piece 33 and the phase synchronizer 36 connected to the coupling capacitor 35. That is, the discharge measuring instrument 12 is simulated by the partial discharge analyzer 34, the coupling capacitor 35, and the phase synchronizer 36.

FIG. 5B is a schematic drawing showing the configuration of the test piece 33 for generating a partial discharge. A small piece 51 of epoxy resin (bisphenol A type, without filler) to which an AC voltage of 14 to 17 kV was applied via a needle electrode 53 was used as a test piece 33. A void of about 0.5 mm is provided at the tip of the needle electrode 53, and an aluminum electrode is arranged on the ground side of the small piece 51 via the conductive paste 54 in the plane direction with respect to the test piece 33. It is configured so that the voltage is applied uniformly to the aluminum. The test piece 33 used had dimensions of 30 mm × 20 mm × 5 mm in length × width × depth. The distance from the tip of the needle electrode 53 to the grounded surface (bottom surface) of the test piece 33 was set to about 3 mm.

6A-8B are photographs and graphs showing the results of experiments performed using the above-mentioned experimental system. A high voltage was continuously applied to the test piece 33 to intentionally generate a partial discharge, and the state of the change over time of the test piece 33 was photographed by the macroscope 38. 6A, 7A, and 8A are images taken by the macroscope 38. Time has passed in the order of FIGS. 6A, 7A, and 8A, and a situation in which partial discharge is proceeding in this order is simulated. Specifically, the state of FIG. 6A corresponds to the state immediately after the start of the experiment, the state of FIG. 7A corresponds to the state 15 minutes after the start of the experiment, and the state of FIG. 8A corresponds to the state 70 minutes after the start of the experiment. Corresponds to the later state.

In this experimental system, the partial discharge is set to progress at a high speed in the test piece 33, but in the actual target device (transformer 5), the partial discharge progresses extremely slowly. In an actual transformer 5, as shown in FIGS. 6A, 7A, and 8A, it generally takes several years to several decades for the electric tree to progress.

6B, 7B, and 8B are graphs showing the φ−q distribution based on the current signal detected by the partial discharge analyzer 34 at the time when the photographs of FIGS. 6A, 7A, and 8A are taken, respectively. Is.

According to FIGS. 6A, 7A, and 8A, it is confirmed that the electric tree is growing as the generation of the partial discharge continues. In the state of FIG. 6A (hereinafter referred to as “state M1”), the electric tree has not yet been generated. In the state of FIG. 7A (hereinafter referred to as “state M2”), the electric tree is slightly confirmed (Tr2 in the figure). In the state of FIG. 8A (hereinafter referred to as “state M3”), it is confirmed that the growth of the electric tree is considerably progressing (Tr3 in the figure).

Then, according to FIGS. 6B, 7B and 8B, it can be seen that the value of the discharge charge amount q according to the phase angle φ varies with the growth of the electric tree.

Table 1 below shows the correspondence relationship with the tree length by calculating the dispersion value Var related to the discharge charge amount q from the result of the φ−q distribution. FIG. 9 is a graph of Table 1. The reference numerals M1, M2, and M3 shown in Table 1 and FIG. 9 correspond to the above-mentioned states M1, state M2, and state M3, respectively.

According to Table 1 and FIG. 9, it can be seen that the value of the dispersion value Var related to the discharge charge amount q increases as the tree length of the electric tree grows. In particular, in view of the captured images of FIGS. 6A, 7A, and 8A, it can be seen that when the dispersion value Var exceeds 200, partial discharge progresses and the electric tree begins to grow. As described above, since the distance (thickness related to the voltage application direction) from the tip of the needle electrode 53 to the bottom surface of the test piece 33 is about 3 mm, when the dispersion value Var exceeds 200, the tree length becomes almost the thickness. It is estimated that it will reach about 50% of the total.
Further, when the dispersion value Var reaches 1000, the electric tree has advanced to the extent that the tree length of the electric tree reaches a length exceeding 80% of the thickness of the test piece 33, and the state is close to the destruction of the entire road. I understand.

When an electric tree is generated in the test piece 33, partial discharge is likely to occur from the tip of the electric tree. On the other hand, partial discharge also occurs at the root or in the middle of the electric tree other than the tip. As a result, it is presumed that the distance between the high voltage side and the ground side changes, and the strength of the discharge varies. As a result, the φ−q distribution varies, and the variance value increases.

Further, as the electric tree progresses, a remarkable outlier is confirmed on the φ−q distribution (reference numeral A1 region in FIG. 8B). It is probable that this is because a minute electric tree reached the ground side before the dielectric breakdown, resulting in an extremely large discharge. Even if the tip of a part of the electric tree reaches the ground side, dielectric breakdown does not occur immediately. That is, in the state of FIG. 8B, it is considered that the state is stopped before reaching the dielectric breakdown.

As described above, when the electric tree progresses considerably, an extremely large discharge occurs, so that it can be seen that the value of the dispersion value Var becomes remarkably large.

The dispersion value Var is a value indicating the degree of variation in the discharge charge amount q in the “φ−q distribution”, which is the distribution state of the discharge charge amount q according to the phase angle φ. The amount of discharge charge q according to the phase angle φ depends on the height of the voltage applied to the transformer 5 in the field and the thickness (related to the voltage application direction) of the epoxy mold resin built in the transformer 5 in the field. Although its size changes, it has little effect on the tendency of variation. That is, it is possible to grasp the degree of growth of the electric tree actually generated in the transformer 5 in the field by the magnitude of the dispersion value Var.

(Verification 2)
The degree of deterioration of the transformer 5 (actual device) installed at the actual site was verified by using the above-mentioned determination method. The verification was performed on transformers 5 installed at three different sites. Hereinafter, for convenience, the respective transformers 5 will be referred to as "actual device 5a", "actual device 5b", and "actual device 5c". In each case, the φ−q distribution was measured based on the result of measuring the ground current flowing through the ground wire 10a by high frequency CT, and the dispersion value Var was calculated from this measurement result. As described above, the φ−q distribution may be measured based on the radio wave signal measured by the antenna 11.

In the actual device 5a, the ground current due to the discharge could not be measured by the high frequency CT. Therefore, the value of the discharge charge amount q was not displayed on the φ−q distribution, and the dispersion value Var = 0. It is presumed that the actual device 5a did not generate a partial discharge in the first place.

FIG. 10A is a result showing the φ−q distribution measured in the actual device 5b. The variance value Var = 12 calculated from the measured values obtained in FIG. 10A. From this result and Table 1 above, it is estimated that the electric tree length generated in the actual device 5b is less than 0.5 mm. Since it is known from the design drawing of the actual device 5b that the thickness of the built-in epoxy mold resin is about 10 mm, the deterioration of the actual device 5b has not progressed to that extent, and the priority of the replacement work is given. Turns out to be low.

FIG. 10B is a result showing the φ−q distribution measured in the actual device 5c. The variance value Var = 229 calculated from the actually measured values obtained in FIG. 10B. From this result and Table 1 above, it is estimated that the electric tree length generated in the actual device 5b is about 1.5 mm. Since it is known from the design drawing of the actual device 5c that the thickness of the built-in epoxy mold resin is about 10 mm, this actual device 5c has a length of 10% or more and 20% or less with respect to the thickness. It can be seen that the electric tree is progressing. From this result, it can be seen that the priority of the replacement work is higher than that of the actual device 5b.

The thickness of the epoxy mold resin provided on the transformer 5 is set by the voltage class of the transformer 5, but is generally 60 mm or less at the thickest, and most of them are about 5 mm to 15 mm. If an electric tree of 10% or more with respect to the resin thickness is developed, there is a risk of destruction of all roads in the near future. Therefore, it is possible to detect whether or not the equipment needs to be replaced according to the value of the dispersion value Var of the discharge charge amount q on the φ−q distribution.

The threshold value Th to be compared with the dispersion value Var may be appropriately adjusted according to the thickness of the epoxy mold resin provided in the transformer 5 installed at the site. Further, the arithmetic processing unit 13a quantifies the degree of necessity of replacement of the transformer 5 according to the magnitude of the difference value between the dispersion value Var and the threshold value Th, and displays the display unit 13c or the like via the information output unit 13b. It may be output to.

3: Cubicle 4: Open / close door 5: Transformer 10: Cable 10a: Ground wire 11: Antenna 11a: Current sensor 12: Discharge measuring instrument 12a: Waveform analysis unit 12b: Phase analysis unit 12c: Display unit 13: Computer 13a: Calculation Processing unit 13b: Information output unit 13c: Display unit 30: Test space 31: Test table 32: Container 33: Test piece 33a: Ground wire 34: Partial discharge analyzer 35: Coupling capacitor 36: Phase synchronizer 37: Transformer 38 : Macroscope 41: Insulation sheet 51: Small piece 53: Needle electrode 54: Conductive paste

Claims (4)

The step (a) of detecting the electric signal related to the partial discharge generated from the epoxy mold power device and obtaining the φ−q distribution which is the relationship between the applied voltage phase angle φ and the discharge charge amount q.
From the result of the φ−q distribution obtained in the step (a), the step (b) of calculating the dispersion value Var related to the discharge charge amount q, and
When the dispersion value Var exceeds a predetermined threshold value, it is characterized by having a step (c) of determining that an electric tree is developing in the epoxy mold electric power device. A method for determining insulation deterioration due to the above. The method for determining insulation deterioration with the progress of an electric tree according to claim 1, wherein the predetermined threshold value is 200. The step (a) includes a step of installing an antenna in the vicinity of the epoxy mold power device and a step of receiving a radio wave signal by the antenna, and the φ−q distribution is based on the strength of the received radio wave signal. The method for determining insulation deterioration due to the progress of an electric tree according to claim 1 or 2, wherein the step is a step of obtaining the above. The step (a) includes a step of measuring the current flowing through the ground wire of the epoxy mold electric power device with a current sensor, and is characterized in that the step of obtaining the φ−q distribution based on the intensity of the measured current. The method for determining insulation deterioration due to the progress of an electric current according to claim 1 or 2.

Images