JPH0560827A - Insulation deterioration detector - Google Patents

Abstract

PURPOSE:To obtain an insulation deterioration detector which can specify the location where insulation deterioration occurs even when the insulation deterioration occurs at a section which does not belong to a transmission line. CONSTITUTION:A power transmission line 1 is surrounded with a shield line 2 and, at the same time, the line 2 is earthed through an earthing conductor 3. Then a first annular magnetic sensor 4 is put around the line 1, with a first signal line 5 being led out from the sensor 4, and a second annular magnetic sensor 6 is put around the conductor 3. A second signal 7 is led out from the sensor 6 and the signals lines 5 and 7 are connected to a frequency spectrum analysis circuit 8 so as to detect signals induced in the signal lines 5 and 7 as frequency spectra.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an insulation deterioration detecting device, and more particularly to a device of a type which can always inspect in a live state.

[0002]

2. Description of the Related Art Various devices have been proposed as a device for detecting a state of insulation deterioration of a power transmission line or the like.
We proposed a device that detects which part of the power transmission line is causing insulation deterioration.

This utilizes a corona discharge or a partial discharge generated when the insulation of a power transmission path is deteriorated, and the discharge position can be specified by detecting the direction and intensity of these discharges. The high frequency noise generated by the discharge is detected by a plurality of sensors provided in the power transmission path.

[0004]

However, the above-described device cannot detect when the signal due to partial discharge is small and when it is buried in general continuous noise. That is, the partial discharge signal at the initial stage of insulation deterioration is considerably smaller than the brush noise of the motor and other extraneous noise, and even if only the peak of the waveform is monitored, it is buried in these noises. For this reason, it is difficult to catch signs of the cable or early signs of insulation deterioration.

Further, in the above-mentioned apparatus, it is possible to determine at which position on the power transmission path the insulation deterioration occurs. However, the insulation deterioration part occurs at a portion other than the power transmission path, for example, a load such as a motor. In some cases this wasn't possible until we could pinpoint it.

The present invention has been made in view of the above matters, and has a high detection sensitivity even when the S / N is poor, and specifies the insulation deterioration position even when the insulation deterioration occurs in a portion other than the power transmission path. It is a technical object to provide an insulation deterioration detecting device capable of performing the above.

[0007]

In order to solve the above technical problems, the present invention has the following constitution. That is, the power transmission path is surrounded by a shield wire, the shield wire is grounded by a ground wire, and a first annular magnetic sensor is provided so as to surround the power transmission path, and a first signal line is connected from the first annular magnetic sensor. A second wire which surrounds the ground wire while being led out
An annular magnetic sensor is provided, a second signal line is derived from the second annular magnetic sensor, the first signal line and the second signal line are connected to a frequency spectrum analysis circuit, and signals induced in the respective signal lines are provided. Is configured to be detected as a frequency spectrum.

By providing the frequency spectrum analysis circuit with a Fourier transform unit, the frequency component can be analyzed more strictly, and the presence or absence of insulation deterioration or the position can be determined by a computer by signal processing in the Fourier transform unit. It will be easy to do.

[0009]

The waveforms of the signals induced in the first and second signal lines are the inductance of the magnetic sensor and the cable connected thereto for a pulsed signal having a fast rising edge such as a partial discharge signal due to insulation deterioration. It becomes a decaying waveform with the time constant of the circuit determined by the capacitance and so on. Since this signal is small at the initial stage of insulation deterioration, it is difficult to detect the signal itself due to the presence of other noise.

In the present invention, since the frequency spectrum of the signal induced in each signal line is detected, it can be easily distinguished from other noises by paying attention to the shape of the frequency peak and the frequency spectrum. The characteristics of the spectrum of each signal and noise will be described.
The partial discharge signal due to insulation deterioration has an attenuation waveform having a circuit time constant determined by the inductance of the magnetic sensor and the capacitance of the cable connected to it, so the frequency spectrum becomes a Lorentz type having a peak at the frequency determined by the reciprocal of the time constant.

On the other hand, continuous wave noise has a sharp spectrum having a peak at the repetition frequency of the continuous wave. With other pulsed noise, the rise of the waveform is slower than that of the partial discharge signal, so the spectrum has a peak at a frequency different from the frequency for the time constant of the circuit, and the spectrum shape is different from the Lorentz type. It becomes something like a different Gaussian type.

These characteristics of the spectrum make it possible to distinguish between partial discharge signals, continuous noise, and pulse noise. On the other hand, the first provided so as to surround the power transmission line
The signals from the annular magnetic sensor are the partial discharge signal generated by the load connected to the power transmission path and the partial discharge signal of the power transmission path. The signal from the second annular magnetic sensor provided so as to surround the ground wire is Since the partial discharge signal is mainly generated in the power transmission path, it is possible to detect the presence or absence of abnormality on the load side by comparing these two signals.

By using the Fourier transform in the frequency spectrum analysis circuit, the frequency component can be analyzed more strictly, and automatic measurement using a computer is also possible. It should be noted that the apparatus of the present invention can always be inspected in a live state without stopping the power supply.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described with reference to FIGS. <Embodiment 1> A power transmission line 1 connects a power source P to a motor M which is a load, and is surrounded by a shield wire 2.
One end of the shielded wire 2 on the power supply side is grounded (GN
The other end of the ground wire 3 which is D) is connected.

A first annular magnetic sensor 4 is provided so as to surround the shield wire 2 and the ground wire 3. The first annular magnetic sensor 4 is made of an amorphous metal containing cobalt as a main component, and the first signal line 5 is wound around the first annular magnetic sensor 4.

On the other hand, the ground wire 3 is provided with a second annular magnetic sensor 6 surrounding the ground wire 3. The second annular magnetic sensor 6 is also made of the same material as the first annular magnetic sensor 4. A second signal line 7 is wound around the second annular magnetic sensor 6.

The first signal line 5 and the second signal line 7 are connected to a frequency spectrum analysis circuit 8. The frequency spectrum analysis circuit 8 measures the intensity, frequency spectrum, and phase of the signal induced on each signal line 5, 7.

Further, the frequency spectrum analysis circuit 8 can be provided with a Fourier transform section 8a, and in this case the input waveform can be decomposed into a simple sine wave, so that post-processing by a computer is easy and measurement is possible. Can be automated.

When the level of the signal obtained from each of the signal lines 5 and 7 is low, a preamplifier A is inserted between each of the signal lines 5 and 7 and the frequency spectrum analysis circuit 8 if necessary. You can

As shown in FIG. 3, bandpass filters F1, F2 and F3 can be inserted between the preamplifier A and the frequency spectrum analysis circuit 8. When this configuration is adopted, the frequency spectrum and phase of each band can be measured.

An operation example will be described below. Each signal line 5, 7
The partial discharge signal obtained from the waveform has a waveform as shown in FIG. 4, and since it is a small signal, it is difficult to detect because it is buried in noise at the early stage of insulation deterioration.

Here, the frequency spectrum of the partial discharge signal induced in each signal line 5, 7 is Lorentz type as shown in FIG. This is determined by the nature of the partial discharge signal at the time of insulation deterioration and the time constant of the magnetic sensor circuit. On the other hand, the other noises have the shapes shown in FIGS.
It is possible to distinguish between the two because the position of the peak or the shape of the spectrum is different.

Next, an example of a partial discharge signal mixed with noise will be given. In FIG. 8, the partial discharge signal is mixed with the continuous wave noise, but it is very difficult to detect the signal only with this.
On the other hand, the spectrum at this time clearly shows the presence of the partial discharge signal as shown in FIG. This is clear in comparison with the noise-only spectrum diagram 10. It should be noted that there is no significant difference between the waveform diagram of FIG. 17 showing only continuous wave noise and the waveform diagram of FIG.

Thus, even under adverse conditions, if attention is paid to the frequency peak and the shape, the distinction from other noises becomes clear.
The above judgment is to judge the insulation deterioration of the power transmission line 1 based on the signals from the respective signal lines 5 and 7, but the signal from the first signal line 5 is generated by the load connected to the power transmission line. The partial discharge signal and the partial discharge signal of the electric power transmission path, and the signal from the second signal line 7 is mainly the partial discharge signal generated in the electric power transmission path. Therefore, by comparing these two signals, the load side The presence or absence of insulation deterioration can be detected.

Further analysis reveals that the waveforms of the signals induced in the first and second signal lines are different from those of the magnetic sensor for a pulsed signal having a fast rising edge such as a partial discharge signal due to insulation deterioration. The decay waveform has a time constant of the circuit determined by the inductance and the capacitance of the cable connected to it (Fig. 4). Since this signal is small at the initial stage of insulation deterioration, it is difficult to detect the signal itself due to the presence of other noise.

In the present invention, since the frequency spectrum of the signal induced on each signal line is detected, it can be easily distinguished from other noises by paying attention to the frequency peak and the shape of the frequency spectrum. The characteristics of the spectrum of each signal and noise will be described.
The partial discharge signal due to insulation deterioration has an attenuation waveform with the time constant of the circuit determined by the inductance of the magnetic sensor and the capacitance of the cable connected to it, so the frequency spectrum is Lorentz type with a peak at the frequency determined by the reciprocal of the time constant ( Figure 5).

Further, the continuous wave noise has a sharp spectrum having a peak at the repetition frequency of the continuous wave (FIG. 6,
(Fig. 10). With other pulsed noise, the rise of the waveform is slower than that of the partial discharge signal, so the spectrum has a peak at a frequency different from the frequency for the time constant of the circuit, and the spectrum shape is different from the Lorentz type. It becomes something like a different Gaussian type (Fig. 7,
(Fig. 11).

These characteristics of the spectrum make it possible to distinguish between partial discharge signals, continuous noise, and pulse noise. On the other hand, the first provided so as to surround the power transmission line
The signals from the annular magnetic sensor are the partial discharge signal generated by the load connected to the power transmission path and the partial discharge signal of the power transmission path. The signal from the second annular magnetic sensor provided so as to surround the ground wire is Since the partial discharge signal is mainly generated in the power transmission path, it is possible to detect the presence or absence of abnormality on the load side by comparing these two signals.

As described above, if the frequency peak is focused even under adverse conditions, the distinction from other noises becomes clear and the S / N ratio is greatly improved. Further, even when the peaks of the frequencies of other noises and partial discharges are at the same position, the shape of the frequency spectrum shows a unique form, so that the determination is easy.

The above judgment is to judge the insulation deterioration of the power transmission line 1 based on the signals from the respective signal lines 5 and 7, but the signal of the second signal line 7 is mainly connected to the power transmission line. Since it is a partial discharge component generated in a load, the presence or absence of insulation deterioration on the load side can be detected by detecting only this component.

Next, the operation of the frequency analysis device shown in FIG. 1 will be supplementarily described. As described above, the partial discharge generated by the insulation deterioration has a signal centered on the frequency f0 determined by the circuit. On the other hand, for example, considering frequencies f1 and fh vertically separated by Δf, the result is as shown in FIG.

At this time, the setting of Δf can be set as shown in FIG. 13 by appropriately adjusting the partial discharge signal and the pulse noise. Here, the solid line is the partial discharge signal and the broken line is the pulse noise. It is assumed that Ml, M0, and Mh are obtained by monitoring the levels of the signals fl, f0, and fh at this time. On the other hand, Mstdl, Mstd0, Mstdh obtained for the standard partial discharge signal
Use to calculate each ratio as below.

[0033]

[Equation 1]

The values of Ratio (l) and Ratio (h) are naturally close to 1 for partial discharge, and are considerably smaller than 1 for pulse noise. This makes it possible to actually distinguish the partial discharge signal from the pulse noise. Further, when the frequencies are slightly deviated as shown in FIG. 14, Ratio (l) is smaller than 1 and Ratio (h) is larger than 1, so that they can be distinguished. When there is no noise having a peak, both Ratio (l) and Ratio (h) are larger than 1, so that they can be distinguished.

As a signal level monitor, a bandpass filter may be combined and assembled by hardware, or FFT may be used for measurement. If hardware is used, the speed will be faster, while if FFT is used, it is easy to change the frequency setting. This action can be realized by the circuit of FIG. Further, in this example, the analysis is performed based on the levels at three different frequencies, but the accuracy can be improved by increasing the frequency points.

As described above, if attention is paid to the frequency peak even under bad conditions, the distinction from other noise becomes clear.
The / N ratio is greatly improved. Further, even when the peaks of the frequencies of other noises and partial discharges are at the same position, the shape of the frequency spectrum shows a unique form, so that the determination is easy.

The above judgment is to judge the insulation deterioration of the power transmission line 1 based on the signals from the respective signal lines 5 and 7, but the signal of the second signal line 7 is mainly connected to the power transmission line. Since it is the partial discharge component generated in the load, the presence or absence of insulation deterioration on the load side can be detected by detecting only this component.

<Embodiment 2> FIG. 2 is an example applied to a power transmission line 1 in a three-phase AC. The same parts as those described in the conventional example are designated by the same reference numerals, and the description thereof will be omitted. In the case of this embodiment, since the external noise appears equally in each phase, by taking the difference between the phases, the external noise of the in-phase component can be canceled and the S / N ratio can be further improved.

[0039]

According to the present invention, since the external noise and the insulation deterioration discharge are discriminated by the frequency spectrum, the detection sensitivity is high even when the S / N is poor, and the insulation deterioration is transmitted by the power transmission. It is possible to specify the location of insulation deterioration even when it occurs in a part other than the road.

[Brief description of drawings]

FIG. 1 is a circuit diagram showing a first embodiment of the present invention.

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

FIG. 3 is a waveform diagram for explaining the operation of the embodiment of the present invention.

FIG. 4 is a waveform diagram of a partial discharge signal for explaining the operation of the embodiment of the present invention.

FIG. 5 is a spectrum diagram of a partial discharge signal for explaining the operation of the embodiment of the present invention.

FIG. 6 is a spectrum diagram of continuous wave noise for explaining the operation of the embodiment of the present invention.

FIG. 7 is a spectrum diagram of pulse noise for explaining the operation of the embodiment of the present invention.

FIG. 8 is a waveform diagram of a partial discharge signal mixed with continuous wave noise for explaining the operation of the embodiment of the present invention.

FIG. 9 is a spectrum diagram of a partial discharge signal mixed with continuous wave noise for explaining the operation of the embodiment of the present invention.

FIG. 10 is a view for explaining the operation of the embodiment of the present invention.
Spectrum diagram of continuous wave noise

FIG. 11 is a view for explaining the operation of the embodiment of the present invention.
Waveform diagram of pulse noise

FIG. 12 is a view for explaining the operation of the embodiment of the present invention.
Partial discharge signal spectrum diagram and sampling frequency

FIG. 13 is a view for explaining the operation of the embodiment of the present invention,
Spectrum diagram of overlapping partial discharge signals and pulse noise

FIG. 14 is a view for explaining the operation of the embodiment of the present invention.
Spectrum diagram of partial discharge signal and pulse noise with shifted frequency

FIG. 15 is a view for explaining the operation of the embodiment of the present invention.
Waveform diagram of continuous wave noise only

[Explanation of symbols]

 1 power transmission path, 2 shield wire, 3 ground wire, 4 1st annular magnetic sensor, 5 1st signal line, 6 2nd annular magnetic sensor, 7 2nd signal line, 8 frequency spectrum analysis circuit.

Claims (2)

[Claims] 1. A power transmission path is surrounded by a shield wire, the shield wire is grounded by a ground wire, and a first annular magnetic sensor is provided so as to surround the power transmission path. While deriving a signal line, a second annular magnetic sensor surrounding the ground line is provided, a second signal line is derived from the second annular magnetic sensor, and the first signal line and the second signal line are frequency spectrum Connected to the analysis circuit,
An insulation deterioration detecting device, characterized in that it is configured to detect a signal induced in each signal line as a frequency spectrum. 2. The insulation deterioration detecting device according to claim 1, wherein the frequency spectrum analysis circuit includes a Fourier transform unit.