JP3389052B2 - Partial discharge measurement method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring partial discharge of a power cable and its accessories or a gas-insulated line (GIL) and accessories, and more specifically, after the partial discharge charge amount reaches a predetermined value. The present invention relates to a partial discharge measuring method capable of estimating the time until breakdown.

[0002]

2. Description of the Related Art As a method for measuring a partial discharge of a power cable line, there is a method and apparatus for measuring a partial discharge of a power cable as disclosed in, for example, Japanese Patent Application Laid-Open No. H5-256893. The technique described in the above publication is a method of identifying a partial discharge pulse from the relationship between the measured signal and the applied voltage phase signal and the degree of occurrence thereof, and automatically determining the occurrence of partial discharge with high accuracy. This is a method of collectively monitoring all the tracks.

[0003]

In the above-mentioned conventional partial discharge measuring method, it is possible to automatically and accurately determine the occurrence of partial discharge, but it is possible to cause a dielectric breakdown in the power cable in which partial discharge occurs. The time could not be estimated. In the unlikely event that a power cable line suffers a dielectric breakdown, a power failure or momentary power failure may occur, and there is a serious risk of fire or electric shock. Further, considering the recovery work after the stop, it is important to identify the position where the partial discharge occurs before the stop of the power application. If you know the location of the partial discharge, you can replace the power cable with the minimum length, but if you do not know the location,
All the power cables in the measurement section will be replaced,
The time and construction costs associated with the restoration work will be enormous.

Therefore, when a partial discharge is generated from the power cable line while the partial discharge is being measured, it is possible to locate the point where the partial discharge is generated by locating the position and then immediately stop the power application. Most desirable. However,
In the above-mentioned conventional partial discharge measurement method, there is no idea when the power cable that is generating partial discharge will be destroyed,
In some cases, there was a risk of dielectric breakdown during position orientation. When a cable has a dielectric breakdown, as described above, it not only causes a power failure, a momentary power failure, or a serious risk of fire or electric shock, but is also caught between the connection parts before and after the part where the insulation breakdown occurs. The cables in the section have to be replaced, and the repair requires a lot of time and money. For this reason, the insulation breakdown of the cable must be absolutely avoided, and in the past, in some cases, the voltage application was stopped before sufficient position orientation was performed. As described above, in the conventional partial discharge measuring method, the power cable that has generated the partial discharge is
There is a problem that it is not possible to perform sufficient position orientation because it is not known how much time is left until the destruction.

The present invention has been conceived in view of the above problems, and an object thereof is to accurately estimate the time until the power cable is destroyed when a partial discharge occurs in the power cable, and to break the cable. It is to be able to understand how much time is left until.

[0006]

FIG. 1 shows a tree CV cable for power insulation, in which a part of the sheath is removed and a treeing needle is placed over the outer semiconductive layer in order to simulate a projection defect in the outer semiconductive layer. When a needle stick defect part is provided by piercing the inside,
It shows the relationship between the elapsed time of charging after the occurrence of partial discharge and the amount of discharged electric charge. The two threshold values Q1 and Q2 (Q1 <Q2) of the discharge charge amount are set in advance, the time T ₁ when the discharge charge amount value reaches Q1 and the time T ₂ when the discharge charge amount value reaches Q2, and the power cable From the time of destruction T _BD ,
And time T ₂ -T ₁ between the Q1-Q2, the time T _BD -T ₂ to failure from Q2 is calculated. The ratio of the time from Q1 to Q2 and the time from Q2 to breakdown is k, and calculation is performed by the following equation (1). k = (T _BD −T ₂ ) / (T ₂ −T ₁ ) (1) The cumulative distribution function of the destruction probability with respect to k is obtained as F (k) from the destruction test results of a plurality of cables. FIG. 2 shows an example of the cumulative distribution function F (k).

Next, when partial discharge occurs from the power cable to be measured, the time from reaching the measured partial discharge charge amount Q1 to reaching the measured partial discharge charge amount Q2 is measured, and the cumulative distribution function F (k ), The destruction time at a predetermined destruction probability is calculated. For example, the destruction time at a destruction probability of 30% is k in the case of F (k) = 0.3 shown in FIG.
_{It is} a value obtained by multiplying _0.3 by (T ₂ −T ₁ ) and adding T ₂ . Alternatively, the value obtained by multiplying k _0.3 by (T ₂ −T ₁ ) is
Time from T ₂ when Q2 is reached to destruction (T _BD −
T ₂ ).

Further, from the cumulative distribution function F (k) to F
(K ₀₎ = ₀ the ratio k ₀ determined at the time of, determined ratio upper limit value of the time fracture probability is 0% from k ₀ T _BD0 [= k ₀ (T
₂ -T ₁₎ + T _2] can be estimated. By _{obtaining the} time T _BD0 above, the probability of dielectric breakdown is 0%.
The upper limit value of the time can be known. For example, when locating, the operator can know how much time is left when the probability of dielectric breakdown is 0, and position locating with confidence. You can Furthermore, at the time of measurement after the partial discharge occurs from the power cable to be measured and the discharge charge amount reaches Q2, the measured (T ₂ −T ₁ ) time and T ₂ to measurement time T ₃ k _m calculated in the time between [=
(T ₃ -T ₂₎ / by determining the failure probability from the cumulative distribution function _{F (k) (T 2 -T} 1) ] Based on the destruction probability can be estimated at the measurement time T _3.

The present invention focuses on the above points and solves the above problems as follows. (1) The cumulative distribution function F (k) of the destruction probability with respect to the ratio k is obtained as described above, and when partial discharge occurs from the power cable to be measured, the measured discharge charge amount reaches Q1. From the time T ₂ -T ₁ from reaching to Q2 and the cumulative distribution function F (k), the time from when the discharge charge amount at a predetermined destruction probability reaches Q2 to destruction is estimated. (2) The ratio k ₀ when F (k ₀ ) = 0 is obtained from the cumulative distribution function F (k), and the power cable to be measured is measured.
Measured discharge charge when partial discharge occurs from the
Time T ₂ -T ₁ up but reached Q2 from reaching Q1
It was measured, and the time T ₂ -T _1, from the calculated ratio k _0, fracture probability during subsequent measurements when the partial discharge occurs discharge charge quantity from the power cable of the measurement object has reached Q2 is Estimate time 0. (3) partial discharge is generated from the measurement object of a power cable
Then, after the measured discharge charge amount reaches Q1,
Time T ₂ -T ₁ is measured until the discharge charge amount is Q
During the subsequent time of reaching the second measurement, the measurement object
Calculated as the time T ₂ -T ₁ power cable, the time ratio k _m since the amount discharge charge reaches the Q2 to the measurement time T _3, the
The ratio k _m, determine the failure probability of the measurement time T ₃ from the cumulative distribution function F (k).

According to the inventions of claims 1 to 3,
Since the above (1) to (3) are performed, it is possible to stochastically estimate the breakdown time of the cable in which the partial discharge has occurred, or to estimate the destruction probability, and to perform the position locating work after the partial discharge has occurred. It is possible to do so with a margin, and it is possible to prevent the insulation breakdown of the cable. Moreover, the cable breakage time can be estimated by freely setting the breakage probability in accordance with the purpose and the importance, and flexible estimation is possible.

[0011]

FIG. 3 is a diagram showing an example of a partial discharge test apparatus for a power cable. In this example, a 66 kVC having air terminals 1 at both ends as shown in FIG.
V cable 2 (conductor size 200 sq, insulator thickness 11
mm) is used to simulate a protrusion defect in the outer semiconductive layer, a part of the sheath of the cable 2 is removed, and a needle needle defect in which a treeing needle is inserted into the insulator from above the outer semiconductive layer. Part 4 was formed. An AC voltage is applied to the cable 2 from the charging transformer 3, and the partial discharge pulse generated at that time is detected by the partial discharge detectors 5 and 5'connected to the sheaths on both sides of the needle stick defect portion 4, The output of the detector 5.5 'was input to the partial discharge measuring device 7 via the balancer 6 to measure the partial discharge charge amount. Then, the output of the partial discharge measuring instrument 7 and the applied potential phase of the AC voltage applied to the cable 2 are input to the analysis device 9 and the digital oscilloscope 10 configured by a computer or the like, and the partial discharge signal and the applied voltage phase signal are input. The partial discharge signal was identified based on the relationship between the above, the degree of occurrence, and the amount of partial discharge charge was determined and analyzed.

Using the above test equipment, the applied voltage is 160 k
Ten tests were conducted for each of V and 110 kV. FIG. 4 is a diagram showing an example of the relationship between the maximum discharge charge amount and the partial discharge duration at 110 kV obtained as described above (an example of a test result of four cables is shown in the same figure). As shown in the figure, with each cable, the maximum discharge charge amount increases with time, and eventually dielectric breakdown occurs.

The data shown in FIG. 4 was obtained for a plurality of cables, and the cumulative distribution function F (k) was obtained by the analyzer 9 as follows. The first threshold value Q1 shown in FIG. 1 is, for example, 20 pC, and the second threshold value Q2 is 1.
00pC, the discharge charge amount is 20pC to 100pC
The time T _{20 -100pC} which increases to ₁₀₀ ° _C and the time T _100-BD from 100pc to the breakdown are obtained, and k is calculated from the following equation (2). k = T _100-BD / T _20-100pC (2) The first threshold charge amount Q1 is approximately 10 pC to 50 pC, and the second threshold charge amount Q2 is, depending on the detection sensitivity. , Approximately 50 to 200 pC is recommended. However, these threshold charge amounts are Q1 <Q
Any value may be used as long as it has a relationship of 2, and is not limited to these charge amounts.

The ratio k was obtained for a plurality of cables as described above, and a histogram was created with the horizontal axis as k and the vertical axis as the frequency of the ratio k as shown in FIG. 5, for example. Then, the cumulative value of the frequency of the histogram is obtained, and each cumulative value is divided by the total of the frequency to calculate the cumulative distribution function F shown in FIG.
(K) was obtained. FIG. 6 is a graph showing the relationship between the ratio k and the destruction probability obtained as described above, and the graph is a graph arranged by a Weibull plot with the ratio k on the horizontal axis and the destruction probability on the vertical axis. As is clear from the figure, when the ratio k is arranged in a Weibull plot, it can be said that the ratio k and the failure probability are on a substantially straight line and follow the Weibull distribution. The Weibull distribution is a distribution that is relatively well suited to represent the life time x and the corresponding cumulative probability F (x). In FIG. 6, the vertical axis represents Y = log log [1 / {1- F
(X)}] is plotted on a Weibull probability paper with the horizontal axis calibrated by X = log x as x = k and F (x) = F (k), and the plotted data matches the Weibull distribution. If so, it becomes a straight line on the Weibull probability paper.

When the above-mentioned test was conducted for the applied voltage of 160 kV, it was found that the distribution of the ratio k coincided with the Weibull distribution in both cases of 160 kV and 110 kV. Distribution function F
(K) was obtained by the Weibull distribution. FIG. 7 shows the applied voltages of 160 kV and 110 k obtained from the Weibull distribution.
The cumulative distribution function F (k) at V is shown. The distribution of k may be any distribution as long as it fits well, and the cumulative distribution function may be obtained from the distribution that most closely follows the lognormal distribution, the gamma distribution, the exponential distribution, or the like.

Next, a CV cable having a defect in the outer semiconductive layer was subjected to a voltage application test, and the voltage application was stopped based on the estimated breakdown time obtained by the cumulative distribution function, and the insulation breakdown of the cable was performed. It was tested whether it was possible to prevent. The cable used is a 66 kVCV cable (conductor size 200 sq, insulation thickness 11 mm), and the applied voltage is 110.
It was set to kV and 160 kV. The cable used was made by inserting foreign matter into the outer semiconductive layer, and it has been confirmed that there is a defect in the outer conductive protrusion. The test equipment used is Figure 3.
Is similar to that shown in, and the partial discharge charge amount is 20 pC.
Time to reach 100 pC after reaching T
_20pC-100pC was measured, and the estimated destruction time when the destruction probability was 5% and 30% was estimated using the cumulative distribution function shown in FIG.

FIG. 8 is a diagram showing an example of the configuration of the analyzing device used for obtaining the estimated breakdown time. The same figure (a)
As shown in FIG. 5, the partial discharge signal identifying section 20 identifies the partial discharge signal and inputs the maximum discharge charge amount to the first and second comparators 21 and 22. Threshold charge amounts Q1 (20 pC) and Q2 (1 set by the discharge charge amount and the variable resistors R1 and R2
00pC) and the corresponding signal. As described later, the applied potential phase 8 of the AC voltage applied to the cable 2 may be input to the partial discharge signal identification unit 20 to identify the partial discharge signal based on the applied potential phase. Since the first and second comparators 21 and 22 generate outputs when a signal larger than the set signal is input, the partial discharge measurement signal corresponds to the charge amount Q1 as shown in FIG. 8B. When the value is reached (time T ₁ ), the first comparator 21 generates an output, and when the partial discharge measurement signal reaches the value corresponding to the charge amount Q ₂ (time T ₂ ), the second comparator 21 The comparator 22 produces an output.

The timer 23 measures the time T ₂ -T ₁ (= T _20-100pC ) from the time when the first comparator 21 generates an output to the time when the second comparator 22 generates an output, and measures it. The time T _{20-100 pC} is sent to the computer 24. The computer 24 uses the cumulative distribution function F created as described above.
From (k), the ratio k when the destruction probability is 5% and 30% is obtained, and the obtained ratio k is multiplied by the time T _20-100pC to obtain the time T
By _{obtaining 100-BD} , the maximum charge amount is Q2 (= 10
Estimate the time from reaching 0 pC) until the cable breaks. Then, the estimated time from when the partial discharge charge amount reaches 100 pC until the destruction probability is reached is output to the display 25.

First, assuming that the probability of destruction is 5%, 110 kV
It was decided to carry out the test in. Since k at this time is k _0.05 = 1.3 as shown in FIG. 7, the discharge charge amount is 20
1. Time T _20pC-100pC increasing from pC to 100pC
Calculate 3 times as estimated failure time. Then, at the time when the estimated breakdown time calculated after the discharge charge amount reached 100 pC, the charging was stopped. The results of the above experiment are shown in Table 1. After conducting the test 5 times, there was no cable that was broken before the stop of the electric power application, and it was possible to stop the electric power supply in the undestructed state.

[0020]

[Table 1]

Next, assuming that the destruction probability is 30%, 160 kV
The same test was carried out. Since k _0.3 at this time is k _0.3 = 0.72 as shown in FIG. 7, 0.72 times the time T _20pC-100pC at which the discharge charge amount increases from 20pC to 100pC is calculated as the estimated breakdown time. . Then, at the time when the estimated breakdown time calculated after the discharge charge amount reached 100 pC, the charging was stopped. The results of the above experiment are shown in Table 2.

[0022]

[Table 2]

As shown in Table 2, No. Only the sample No. 2 suffered cable breakdown before the power was stopped. The breakdown time was 21 seconds after the discharge charge amount reached 100 pC. As described above, in this embodiment, the breakdown time of the CV cable in which the partial discharge has occurred can be estimated stochastically, and the insulation breakdown of the cable can be prevented in advance. Further, since an appropriate destruction probability can be freely set according to the purpose and its importance, there is flexibility in estimating the destruction time.

In this embodiment, the estimation was made based on the maximum partial discharge charge amount, but the average charge amount per unit time may be used, or, as described above, the partial discharge identifying section 10 may be used. The input potential phase 8 may be used to limit the generated phase angle and target partial discharge pulses generated in a certain phase range. In addition, in the present embodiment, in order to obtain the cumulative distribution function, data obtained by an experiment in which a needle is stuck in an insulator with a tree needle to form a defect is used. Is known to cause insulation breakdown, and if it is important to prevent the insulation breakdown of the cable, it is considered to be a safe side and appropriate. Of course, a cumulative distribution function obtained from other defect data may be used.

Further, it is most desirable to obtain the cumulative distribution function for each insulation thickness of the cable and the applied voltage, but if the insulation thickness and the voltage range are allowable, it is possible to use it for the test conducted on other cables. Is. For example, when the applied voltage is very high, the influence of the insulation thickness on the time from the occurrence of discharge to the breakdown becomes small, so the same cumulative distribution function is approximately used for cables with various insulation thicknesses. can do.

[0026]

As described above, according to the present invention, it is possible to stochastically estimate the breakdown time of a CV cable in which partial discharge has occurred, and it is possible to prevent the insulation breakdown of the cable. Further, since an appropriate destruction probability can be freely set according to the purpose and its importance, there is flexibility in estimating the destruction time.

[Brief description of drawings]

FIG. 1 is a diagram showing a relationship between a charge elapsed time and a discharge charge amount in a cable having a defective portion.

FIG. 2 is a diagram showing an example of a cumulative distribution function F (k).

FIG. 3 is a diagram showing a configuration of a test apparatus according to an embodiment of the present invention.

FIG. 4 is a diagram showing an example of the relationship between the maximum discharge charge amount at 110 kV and the partial discharge duration.

FIG. 5 is a histogram showing the ratio k obtained from the discharge charge amount and the partial discharge duration and its frequency.

FIG. 6 is a diagram showing a relationship between a ratio k and destruction probability, which are arranged by a Weibull plot.

FIG. 7: Cumulative distribution function F obtained by Weibull distribution
It is a figure which shows an example of (k).

FIG. 8 is a diagram showing an example of a configuration of an analysis device used for obtaining an estimated breakdown time.

[Explanation of symbols]

1 aerial terminal 2 CV cable 3 voltage transformer 4 Needle-stick defect 5,5 'partial discharge detector 6 Balancer 7 Partial discharge measuring instrument 8 section potential phase 9 Analyzer 10 digital oscilloscope 20 Partial discharge signal identification section 21,22 comparator 23 timer 24 computer 25 display

Claims (3)

(57) [Claims] 1. A method for measuring a partial discharge of a power cable, which comprises two thresholds Q1 and Q2 (Q1 <Q1) of the amount of discharged charges in advance.
Q2) is set, and for a plurality of power cables, the time T ₁ when the value of the discharge charge amount reaches Q1, the time T ₁ when the value of the discharge charge amount reaches Q2.
₂ and the time T _{BD at} which the power cable is destroyed,
The time T ₂ −T ₁ from the discharge charge amount reaching Q1 to the Q2 and the time T _BD −T ₂ from the discharge charge amount reaching the Q2 to the breakdown are calculated, and the time T ₂ −T is calculated. The cumulative distribution function F (k) of the destruction probability with respect to the ratio k of ₁ and the time T _BD -T ₂ is obtained, and when partial discharge occurs from the power cable to be measured,
From the time T ₂ -T ₁ from when the measured discharge charge amount reaches Q1 to Q2 and the cumulative distribution function F (k), the discharge charge amount at a predetermined breakdown probability reaches Q2 and the breakdown occurs. A method of measuring partial discharge, characterized by estimating the time to reach. 2. A method for measuring a partial discharge of a power cable, which comprises two thresholds Q1 and Q2 (Q1 <Q1) of discharged electric charges in advance.
Q2) is set, and for a plurality of power cables, the time T ₁ when the value of the discharge charge amount reaches Q1, the time T ₁ when the value of the discharge charge amount reaches Q2.
₂ and the time T _{BD at} which the power cable is destroyed,
The time T ₂ −T ₁ from the discharge charge amount reaching Q1 to the Q2 and the time T _BD −T ₂ from the discharge charge amount reaching the Q2 to the breakdown are calculated, and the time T ₂ −T is calculated. The cumulative distribution function F (k) of the failure probabilities for the ratio k of ₁ and the time T _BD −T ₂ is obtained, and the failure probability is 0 from the cumulative distribution function F (k) of the failure probability.
The ratio k ₀ when F (k ₀ ) = 0 is obtained, and when partial discharge occurs from the power cable to be measured,
The measured discharge charge amount reaches Q2 after reaching Q1.
Time measures the time T ₂ -T ₁ up, and the time T ₂ -T _1, from the calculated ratio k _0, the partial discharge occurs discharge charge quantity from the power cable of the measurement object has reached Q2 Destruction probability is 0 during subsequent measurements
A method for measuring partial discharge, characterized by estimating the time of the discharge. 3. A method for measuring a partial discharge of a power cable, which comprises two threshold values Q1 and Q2 (Q1 <Q1) of discharge charges in advance.
Q2) is set, and for a plurality of power cables, the time T ₁ when the value of the discharge charge amount reaches Q1, the time T ₁ when the value of the discharge charge amount reaches Q2.
₂ and the time T _{BD at} which the power cable is destroyed,
The time T ₂ −T ₁ from the discharge charge amount reaching Q1 to the Q2 and the time T _BD −T ₂ from the discharge charge amount reaching the Q2 to the breakdown are calculated, and the time T ₂ −T is calculated. ₁ and to previously obtain a cumulative distribution function F (k) of the fracture probability with respect to a ratio k of time T _BD -T _2, the partial discharge from the measurement target power cable occurs
The measured discharge charge reaches Q1 and then Q2
Time T ₂ -T ₁ is measured, and at the time of measurement after the discharge charge amount reaches Q 2,
And time T ₂ -T ₁ of the measurement target power cable, determine the time of the ratio k _m since the amount discharge charge reaches the Q2 to the measurement time T _3, the ratio k _m and the cumulative distribution function F ( A partial discharge measuring method, characterized in that the destruction probability at the measurement time T ₃ is obtained from k).