JPS6214787B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a determination method for determining the deterioration state of a cable due to water trees, which often occur in polyethylene cables, and particularly to a determination method for determining whether water trees occur generally or locally. Water treeing is caused by moisture penetrating the insulation in cables and abnormal electric field areas such as voids in the insulation and protrusions in the semiconducting layer, and it causes a significant deterioration of the insulation properties of the cable. It is. Therefore, conventionally, as a method of predicting the occurrence of such water trees, a method using tan δ and DC leakage current in combination has been recommended. However, the method using tan δ is a numerical value that indicates the average deterioration of the cable, and is hardly suitable for discovering phenomena such as water trees where deterioration is progressing locally. In other words, if the cable has deteriorated to an extreme degree and water trees occur along the entire length of the cable, it can be detected, but as shown in many cases, water trees occur near joints, cable heads, and rising parts of cables. This is because it often occurs locally in places where drains, solvents, chemicals, etc. enter. In short, it is impossible to detect this type of deterioration using the tan δ method. Next, the method using DC leakage current has the advantage of being able to detect local deterioration, but in the case of polyethylene, the resistivity is high, so the water tree in the radial direction of the cable insulation coating layer is at least 1/2
If it does not extend by 2 or more, it cannot be detected. Furthermore, since the measured values are mediocre, it is difficult to determine the characteristics unless a high voltage is applied. This may lead to a new danger of degrading the insulation performance of the cable during measurement. In this way, conventional methods are still insufficient,
No effective concrete measures have been found. In addition, the problem of water tree occurrence is not only important for manufacturers but also for users such as electric power companies, from the standpoint of preventing accidents and predicting the remaining life of existing cables. There was an urgent need to develop a detection method. Therefore, as a method for solving the above problem, the present inventors have developed a method of charging the cable to be measured, then discharging it, and detecting the presence or absence of water trees from the discharge characteristics during such discharge. Among the methods described above, a method has already been proposed in which the instantaneous discharge current at the initial stage of discharge is bypassed, and the discharge characteristics of the aftereffect current are then measured with a delay of a predetermined time to detect the presence or absence of water trees. However, with the former method, if the cable to be measured is a long cable actually installed, the capacitance C of the cable itself increases, and the amount of charge Q during charging and discharging increases.
This has the drawback that the aftereffect current (current with a time delay) in the water tree generation area is masked and cannot be detected. This drawback cannot be improved even if the voltage at the time of measurement is increased, as it only increases the instantaneous discharge current.On the contrary, if the voltage is increased, the current at the time of measurement becomes unstable, and it may cause damage to the human body. This brings about a new drawback in that it also poses a risk to In addition, increasing the measurement voltage will increase the space charge stored in polyethylene.
There is also the problem that the measurement results become unstable. The latter method is a preferable method as it can eliminate the drawbacks mentioned above, but it is difficult to determine whether water trees occur on average over the entire length of the cable or locally concentrated in a part of the cable. It is not possible to determine whether In particular, considering the fact that water treeing often occurs locally at joints, cable heads, cable drains, and places where solvents, chemicals, etc. can infiltrate, cable deterioration due to water treeing is more likely to occur. A major drawback is that it cannot be determined whether it is local or global. In view of these points, the present invention provides a method for detecting whether the presence or absence of water tree generation is local or general, and thereby solves the above-mentioned drawbacks. Charge the measuring cable,
After that, in the method of detecting cable deterioration due to water tree from the discharge characteristics, after bypassing the instantaneous discharge current at the initial stage of discharge, the cumulative discharge charge Q ₁ is calculated for the first few seconds when the aftereffect current is large. , and the subsequent cumulative discharge charge amount Q ₂ for several tens of seconds with a relatively small discharge current, and Q
This method is characterized by determining the deterioration state of the cable by determining ₁ / _Q2 . Figure 1 shows the discharge characteristics of the aftereffect current after bypassing the instantaneous discharge current at the initial stage of discharge.A is a normal cable, B is a cable that has water trees, but has not deteriorated much. C is a cable in which water trees are occurring on average throughout the cable (deterioration is progressing), D is a cable in which parts of the cable are severely degraded. As can be seen from the graph shown in Figure 1, the discharge current characteristics when the water trees (C) occur on the average over the entire cable are similar to those obtained by translating the characteristics of the normal cable (B). If the water tree (2) is progressing significantly locally, the discharge after a certain period of time (for example, 10 seconds) becomes extremely slow. Therefore, the accumulated discharge charge Q ₁ occurs during the first few seconds when the aftereffect current is large (for example, between 3 seconds and 10 seconds), and then during the subsequent tens of seconds when the discharge current is relatively small (for example, between 10 seconds and 60 seconds). ) integrated discharge charge amount
Q ₂ i.e. Q ₁ =∫ ^t=10 _t=3 I(t)dt and Q ₂ =∫ ^t=60 _t
=10 I(t)
If you measure dt and find the ratio Q ₁ /Q ₂ × 100 (%) of the two, it will be close to 100% in the case of average deterioration, and the larger the local deterioration, the more it will decrease. It is possible to determine whether the cable deterioration caused by this is local or global. Incidentally, in the 6.6KV150mm ² CV cable, normal cable A, cable B with water tree generation throughout, cable C with several meters of deteriorated cable jointed to 100m of normal cable to simulate partial deterioration, and cable C with water tree generation throughout the cable. The cable that caused
Table 1 shows the experimental results for each cable D, in which the deteriorated cable was jointed at a length of 100 m.

[Table] From Table 1 of the experimental results, the ratio between the two is close to 100% for normal cable A, and
In the case of cables where water trees have occurred throughout the entire area, the percentage is in the 70% range, and in the case of cables with locally deteriorated parts of C and D, the percentage is in the 50% range. It can be confirmed that the judgment is possible. Figure 2 shows an example of a detection circuit for carrying out the detection method of the present invention, in which 1 indicates the cable to be measured, 1a its conductor, 1b its polyethylene insulation layer, and 1c its shielding layer. be. S ₁ and S ₂ are changeover switches, e.g. vacuum switch 2
is a DC power source, 3a and 3b are discharge charge amount direct reading devices that sample the discharge charge amount in minute time, integrate it, and display it digitally. 3a is Q ₁ , 3b
displays Q ₂ . Note that the changeover switch _S2 is operated via a timer 4. First, the movable contact a of the changeover switch _S1
is inserted into the fixed contact b side, and the cable insulation layer 1b is charged by the DC power supply 2. A charging time of about 5 minutes is sufficient for the aftereffect current to be approximately saturated. Strictly speaking, it takes a very long time, but in the case of this measurement, this level of current sensitivity is sufficient for measurement. After charging, next, move the movable contact a of the changeover switch _S1 to the fixed contact c side to form a discharge circuit and discharge. At this time, the movable contact d of the changeover switch _S2 , which is interlocked via the timer 4, is on the side of the fixed contact e. Therefore, the initial instantaneous discharge current is discharged to the ground side. If you discharge in this way for 3 seconds,
When the timer 4 is activated, the movable contact d of the switch S ₂ separates from the fixed contact e and enters the fixed contact f side, and the accumulated discharge charge amount for the first few seconds (3 seconds to 10 seconds) where the aftereffect current is large. Q ₁ is the discharge charge direct reading device 3a
Then, the timer 4 is activated, and the movable contact d of the switch _S2 moves away from the fixed contact f and enters the fixed contact g side, and the discharge current is relatively small for several tens of seconds (10 seconds to 60 seconds). The integrated discharge charge amount Q ₂ is measured by the discharge charge amount direct reading device 3b. Note that a suitable voltage for measurement is 100 to 500V. As is clear from the above, according to the present invention, it is possible to determine whether the deterioration of the cable due to the occurrence of water trees is local or general. Also,
At this time, there is no need to make the measurement voltage very high, so there is no danger and there is no possibility that the current during measurement will become unstable.

[Brief explanation of the drawing]

FIG. 1 is a discharge current characteristic diagram, and FIG. 2 is a circuit diagram showing an example of a detection circuit for implementing the present invention. A, B, C, D...Discharge characteristic curve of each cable,
DESCRIPTION OF SYMBOLS 1... Cable to be measured, 1a... Conductor, 1b... Insulating layer, 1c... Shielding layer, 2... DC power supply, 3a, 3b...
Discharge charge amount direct reading device, 4...timer, _S1 , _S2 ...changeover switch.

Claims (1)

[Claims] 1 In the method of charging the cable under test, then discharging it, and detecting cable deterioration due to water tree from the discharge characteristics, after bypassing the instantaneous discharge current at the initial stage of discharge, the initial few seconds when the aftereffect current is large The deterioration state of the cable can be determined by measuring the cumulative amount of discharged charge Q ₁ during the following period and the cumulative amount of discharged charge Q ₂ for several tens of seconds when the discharge current is relatively small, and determining Q ₁ /Q ₂ . A method for determining cable deterioration by water tree, characterized by: