JPH0862274A - Partial discharge measurement method and device - Google Patents

Abstract

(57) [Summary] (Correction) [Purpose] To provide a device that does not miss partial discharge data in any case and can detect device abnormality even when the noise level is low. [Structure] The input is compared with a threshold value V _th in a comparator 2. The threshold setting circuit 1 determines the threshold V _th at any time. For inputs V _j that exceed the threshold V _th , the comparator 2 produces a read signal 3. The threshold value V _th is changed according to the magnitude and frequency of the input signal so that the number of input signals equal to or higher than the threshold value V _{th in} a unit time always falls within a certain range. When the noise level is low, the threshold value V _th is low, and the low-level noise is also measured, so it can be seen that the measuring instrument is not a failure. When the noise is large, the threshold V _th is increased to suppress the noise. Even in this case, the partial discharge signal can be measured without being dropped.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to partial discharge measurement of high voltage power equipment such as power cables. The magnitude and frequency of partial discharge are useful data for knowing the state of electric power equipment. The partial discharge measuring device records the magnitude and the time of occurrence of the partial discharge generated in a certain period.

[0002]

2. Description of the Related Art At present, a method of capturing a partial discharge signal using an electronic device such as a computer is widely used to observe a partial discharge measurement. This makes it possible to perform detailed analysis with less omission of data as compared with the conventional method of visually observing with an oscilloscope or recording data on a chart paper.

FIG. 3 shows a functional block of a conventional data acquisition circuit. The input is a voltage V _{j obtained} by appropriately converting the partial discharge detection signal. j is the number of partial discharge. There is a threshold setting circuit 1 that provides a threshold V _th . The threshold V _th given by this and the input are compared by the comparator 2,
The read signal 3 is generated when the input is larger than the threshold value V _th . At this time, the input magnitude V is read. The result V _j of reading the size is stored in the personal computer 5.

The threshold V _th is provided for the following reason. One is that there is a small amount of noise that needs to be dropped. The other is that the partial discharge that leads to failure or abnormality of the power equipment should have a certain size, and it is not necessary to monitor the small partial discharge. Conventionally, this threshold value V _th is a fixed value.

[0005]

The conventional data capturing method has the following problems because only the partial discharge having a pulse wave height higher than this is captured as data by using the fixed threshold value V _th . In FIG. 3, the threshold value V _th of the comparator 2 is usually set as a constant value that slightly exceeds the average level of background noise. Background noise is constantly changing in magnitude.

With such a fixed threshold value setting, if the background noise is small and there is no partial discharge generated in the target device (cable, power device), there is almost no input pulse. That is, there is no signal. When operating from a remote location, there is no partial discharge at this time,
I don't know if there is no input pulse or the input is not coming because the measuring instrument is out of order.

When the external noise increases more than when the threshold value is set and the background noise level exceeds the threshold value, in the fixed threshold configuration of FIG. 3, noise constantly exceeding the threshold value V _th enters the computer. Computer processing capacity is limited. For example, there is a limitation that up to several pulses can be recorded in a unit time. Due to this limitation, the computer will miss data if there is too much data input. In other words, when the frequency of noise with a high wave height increases, noise and partial discharge cannot be distinguished, so the computer captures all data that is _{equal to} or higher than the threshold value V _th . Some of this data is unnecessary. When the memory is full, new data can no longer be fetched. The partial discharge data generated thereafter will be missed.

The problem of (1) is that since there is no signal, it is not known whether the device, the signal line, or the like has a fault, or whether there is no data. This is a drawback that occurs because the threshold V _th is too high. The problem is that the computer memory overflows due to noise. This is because the threshold value V _th is too low. SUMMARY OF THE INVENTION It is an object of the present invention to solve such a problem, to provide a partial discharge measuring device capable of determining a failure of a measuring device and not missing necessary discharge pulse data.

[0009]

The present invention counts the number H of input data within a constant time S that exceeds the threshold V _th, and sets the threshold V _th to the hour and time so that it falls within a certain range K to M. The number of pieces of data having a wave height exceeding the threshold value V _th is adjusted so as to match the capability of the computer.

[0010]

According to the present invention, the detection signal from the electric power equipment or the like is compared with a certain threshold value V _th and the pulse wave height V _j higher than this is counted. The threshold value _Vth is changed so that the number of pulses H within a certain period of time becomes a predetermined number of pulses (K to M). When partial discharge does not occur and there is little noise, the threshold value V _th is lowered. Then, part of the noise exceeds the threshold value V _th . The number H of pulses that exceed the threshold V _th in the measurement unit time S is set to K <H <M. Even if partial discharge does not occur at all, noise is introduced, so it can be seen that the measuring instrument, signal line, etc. are not defective. It is possible to confirm the normal operation of the measuring instrument by inputting noise. It makes effective use of noise.
This makes it possible to overcome the above problems.

When partial discharge frequently occurs but a large amount of noise is also generated, the threshold V _th is raised because the number of inputs exceeding the threshold V _th increases too much. As a result, the number H of signal inputs exceeding the threshold value V _th is reduced, and K <H <M. Most of the input is cut. However, if the pulse height due to partial discharge is larger than the background noise, all the cut inputs are noise inputs. No partial discharge is missed.

The structure of the present invention will be described with reference to FIG. The input is compared with a threshold value V _th in the comparator 2. The threshold setting circuit 1 determines the threshold V _th at any time. For inputs V _j that exceed the threshold V _th , the comparator 2 produces a read signal 3. At this time, the magnitude of the input V _j is read (peak detection circuit 4) and recorded in the personal computer 5. At this time, it does not matter if the signal generation time T _j and the number j are recorded at the same time. When the background noise changes abruptly, the number of signals exceeding the threshold value V _th may exceed the capacity of the personal computer. Therefore, in this example, data is temporarily stored in the buffer memory 8. The number U of data stored in the buffer memory 8 is much larger than the number F of data that can be input to the personal computer 5 per unit time S.

It is assumed that there is a lot of noise, there are many inputs exceeding the threshold value V _th , and the number H of read data is large. Since this is stored in the buffer memory, the buffer memory temporarily holds a large amount of data. The data number H is counted as 6 pulses. This is compared with the reference pulse numbers K and M. Since M <H, the threshold value V _th is increased. Then, the number H of read data is reduced. Data transfer from the buffer memory to the personal computer continues continuously during this period.

If M <H, the threshold value V _th is further increased. The number H of read data is further reduced. The buffer memory also receives more data than usual and is all stored. However, because it is first-in first-out,
The oldest data is sequentially input to the personal computer 5 and erased from the buffer memory.

When the threshold value V _th is continuously raised, the number H of data to be read eventually falls within a predetermined range (K <H <
M). Here, the correction of the threshold value V _th is stopped. Thereafter, this threshold value V _th is maintained. When the noise decreases and the number H of data read in the unit measurement time S decreases, H <K
Becomes This can be seen by a pulse count 6 and a comparison 7 with the reference pulse count. In this case, the threshold V _th is lowered. Since noise with a low wave height also comes in, the number of input pulses increases and falls within a predetermined range (K to M).

In this way, the comparison is made by the comparator 2, and the threshold value V
The number of inputs having a wave height of _th or more is ultimately kept within a predetermined range K to M. A large amount of data may be accumulated in the buffer memory 8 within the measurement time S, and a small amount of data may be accumulated within the measurement time. Capacity U
Is taken large enough so that it never overflows the buffer memory. Although the amount F of data that can enter the unit time S in the personal computer is fixed, all data can be input to the personal computer because of the buffer memory.

[0017]

[Embodiment] For simplification, the description will proceed assuming the capability of each circuit block. The measurement unit time S is 1 second. The pulse processing capacity F of the computer is assumed to be 100 shots / second. The reference pulse number set at this time is 40 to 80 shots / second (K = 40 shots / second, M = 80 shots / second), and the threshold value setting circuit resets the threshold value every 1 second. Also consider the buffer capacity of the buffer memory as 10000 data (U
= 10,000).

First, consider a steady state in which no partial discharge has occurred and the magnitude of noise has not changed. In this case, noise data within the reference pulse number range (K to M) is read into the computer. Even if you are doing telemetry
You can check the normal operation of the measuring instrument due to noise.

Consider a case where the base noise changes and increases. Temporary computer processing power (F = 10
Pulses exceeding 0 shots / second are input. For example, H = 5
A pulse of 00 shots / second is stored in the buffer memory. Since the buffer memory capacity U = 10000 or less, the capacity is sufficient. Pulse counting is performed to recognize that pulses exceeding the reference pulse number (40 to 80 shots / second) have been detected. The threshold setting circuit 1 raises the threshold V _th .

Since the threshold value V _th has risen, V
It is assumed that a pulse of 40 shots / second (H = 40) with _j > V _th is detected. This is well within the processing capacity of the computer (F = 100 shots / second, H <F). After that, in the above example, (100-40) = 60 pulses are reduced from the buffer memory every second. (500-100)
After /60=6.7 seconds, the steady state can be restored. At this time, data is not lost unless the number of pulses temporarily exceeding the processing capacity exceeds 10,000. As can be seen from the above example, the larger the buffer capacity U, the smaller the possibility of missing data. However, since the cost and power consumption will naturally increase, the buffer capacity may be determined in consideration of economic efficiency.

Consider a case where the base noise is reduced.
Temporarily, no signal including noise comes to the computer at all. However, when recognizing that the number H of detected pulses is less than the reference pulse K, the threshold setting circuit 1 lowers the threshold V _th . As a result, after a while, it is possible to return to the steady state (state H = F in which pulses of about the reference pulse number are detected). Next, let us consider a case where, for example, when there is a noise of 70 shots / second, a partial discharge larger than the background noise of 40 shots / second occurs. The number of signals larger than the threshold value V _th is 11
It will be 0 shots / second.

Therefore, transiently, 110-100 =
You will miss a pulse of 10 shots / second. This one
No problem if 0 shots / second is all noise. However, it is possible that a partial discharge is included in the missed signal. However, no buffer memory is needed to allow this omission. The data acquisition circuit according to the present invention includes 11
When the number of pulses of 0 shots / second is recognized, the threshold value setting circuit raises the threshold value V _th to control the number of pulses to 40 to 80 shots / second. As a result, the partial discharge pulse is detected at 40 shots / second and the noise level is detected at 0-40 shots / second. Therefore, no partial discharge pulse is steadily missed. It only reduces the mixing of noise. There is no problem even if noise is dropped.

In this way, the present invention can easily solve the problems of not knowing whether the measuring device or the signal line is broken because there is no signal, and the problem of dropping the partial discharge signal due to too much noise. Can be resolved.
It is a useful invention. FIG. 2 shows a specific configuration. Here, an example in which a digital comparator is used as the comparator 2 is shown. The input waveform is subjected to A / D conversion 10 to be a digital data string and input to the comparator 2.
The comparator 2 opens the gate 11 when data having a size _{equal to} or larger than the set threshold value V _th is input. The peak value V _{j of the} pulse is detected by the peak detection circuit 12. This is immediately stored in the buffer memory 8. These data are read by the computer 5 in fixed numbers. The noise level changes and the number of pulses H to be processed is too small (H <
K) or excessive (M <H), the threshold setting circuit 1 automatically changes the threshold so that the reference processing pulse number (K <H <M) is achieved.

[0024]

INDUSTRIAL APPLICABILITY The present invention is effectively used when a partial discharge signal is taken into a computer and processed when measuring a partial discharge of a cable or a power device to which a high voltage is applied.

[Brief description of drawings]

FIG. 1 is a functional block diagram of a data acquisition circuit of the present invention.

FIG. 2 is a diagram showing an embodiment of the present invention.

FIG. 3 is a functional block diagram of a conventional data acquisition circuit.

[Explanation of symbols]

 1 Threshold Setting Circuit 2 Comparator (Digital Comparator) 3 Read Signal 4 Size Read 5 Personal Computer (Computer) 6 Pulse Number Count 7 Comparison with Reference Pulse Number 8 Buffer Memory 10 A / D Conversion 11 Gate 12-peak detection circuit

Claims (2)

[Claims] 1. To measure a partial discharge of a power device,
Comparing the pulse height threshold V _th the detection signal, a comparator for extracting only the threshold V _th or more signals, a detection circuit for detecting a threshold value V _th is larger than the detection signal, the threshold V _th of the measurement unit time S
A counting device for counting the input frequency H of a larger signal,
The number H of signals is defined as a reference frequency K in the measurement unit time S,
M (K <M), and if H <K, lower the threshold V _th ,
A threshold setting circuit for raising the threshold V _th if M <H,
A partial discharge measuring apparatus characterized in that the threshold V _th is changed so that the input frequency H of a signal having a wave height higher than the threshold V _th is K <H <M. 2. In order to measure a partial discharge of a power device,
The threshold V _th is compared with the wave height of the detection signal, a signal equal to or higher than the threshold V _th is captured, and the threshold V _th within the measurement unit time S is set.
The number H of the above signals is counted, and this is the reference frequency K to M (K
A partial discharge measuring method, characterized in that the threshold V _th is changed so as to fall within <M).