JPH052047A - Apparatus for predictive maintenance - Google Patents

Abstract

(57) [Abstract] [Purpose] A sensor that is easy to install and that can detect over a wide frequency range is realized, and at the same time, by appropriately processing the obtained sensor output signal, it is possible to reliably To provide an excellent predictive maintenance device capable of detecting an abnormality and discriminating the content of the abnormality. [Structure] As a sensor, a piezoelectric composite material sensor 10 formed by forming a piezoelectric composite material into a sheet shape and disposing electrodes on both sides thereof is used. It has a signal processing means 20 for processing an output signal from the piezoelectric composite material sensor 10. The signal processing unit 20 compares the storage unit 21 that stores a fixed signal pattern as a reference signal in advance with the signal pattern stored in the storage unit 21 and the output signal from the piezoelectric composite material sensor 10 to determine whether there is an abnormality. And a determination means 23 for determining.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a predictive maintenance apparatus for detecting the presence / absence of abnormality of high voltage electric equipment, and more particularly to signal processing for determining the presence / absence of abnormality using a piezoelectric composite material sensor.

[0002]

2. Description of the Related Art Various types of high-voltage equipment are used in substations, switchyards, etc. These high-voltage equipment are housed in a sealed metal container. In such a structure, since the internal state of the device cannot be directly grasped from the outside, some means is used to monitor whether or not the internal device state is abnormal.

A predictive maintenance device is a device for monitoring such abnormalities of high-voltage equipment. This predictive maintenance device
It is a system that monitors the state of a device using a plurality of various sensors. Among them, the detection of partial discharge that occurs due to some electrical abnormality can detect the deterioration of insulation inside the device, the electrical connection failure, the presence or absence of the inclusion of metal foreign matter, etc. It is extremely effective in monitoring abnormalities.

By the way, since partial discharge generates not only electric signals but also mechanical vibrations, when detecting partial discharges, the presence or absence of partial discharges is determined by detecting these electric signals and mechanical vibrations. Is being done. Also, among high-voltage equipment, in switching devices such as circuit breakers, disconnectors, and grounding devices, mechanical vibrations are generated during opening and closing operations. You can figure it out. Furthermore, mechanical vibration occurs when noise due to rainfall or a ground fault due to dielectric breakdown occurs.

The above-mentioned various factors, that is, the mechanical vibrations caused by the partial discharge, the switching operation, or the ground fault have a wide frequency range of several tens Hz to several MHz. There is usually a fixed pattern of such changes. Further, conventionally, as a detection means for detecting mechanical vibration and determining the presence or absence of abnormality,
Accelerometers and ultrasonic sensors that use piezoelectric elements are used.

[0006]

However, as described above, when the mechanical vibration is detected by using the acceleration sensor or the ultrasonic sensor, the natural frequency of the sensor itself is determined, so that the frequency that can be detected is specified. There was a problem that it would end up. Therefore, even if you try to detect mechanical vibration in a wide range of frequencies, it cannot be detected sufficiently, and
Since the detectable frequency range is limited in this way, it was not possible to recognize the mechanical vibration pattern as described above. Therefore, although it is possible to detect the presence or absence of mechanical vibration, it is difficult to reliably detect an abnormality and identify the content of the abnormality.

As a method of compensating for such a defect, a method of using a plurality of sensors having different detection frequencies is conceivable. In this case, mounting is complicated and it is necessary to process a plurality of sensor signals. However, it has a drawback that the device becomes large and economically disadvantageous. further,
Since the area required for mounting the sensor is wide, there is a drawback that it is easily affected by external noise, and the number of sensors that can be mounted is naturally limited. Further, even if a large number of sensors are attached, it is necessary to compensate for variations in sensitivity among the sensors, and there is a drawback in that adjustment requires a lot of time and labor.

The present invention has been proposed in order to solve the problems of the prior art as described above, and an object of the present invention is to realize a sensor that is easy to install and can detect over a wide frequency range. At the same time, by appropriately processing the obtained sensor output signal, it is possible to provide an excellent predictive maintenance device capable of surely detecting an abnormality of the high-voltage equipment and discriminating the content of the abnormality.

[0009]

The present invention focuses on a piezoelectric composite material sensor which has been recently developed as a sensor for detecting mechanical vibration. This piezoelectric composite material sensor is a sensor in which piezoelectric fine particles are mixed with a polymer material to form a sheet.
It has a flat frequency characteristic over a wide frequency range of MHz, and has an excellent feature that a conventional piezoelectric sensor that can faithfully detect a mechanical signal pattern is not present. Besides, because it is a sheet shape and rich in elasticity,
It has the advantage that it is extremely easy to attach even to a container having a complicated shape. Therefore, the piezoelectric composite material sensor has a feature that a signal having a wide frequency range such as mechanical vibration can be detected by one sensor with high sensitivity without being affected by external noise.

That is, the predictive maintenance device according to the present invention is
As the sensor, a piezoelectric composite material sensor formed by forming a piezoelectric composite material into a sheet and disposing electrodes on both sides thereof is used, and also has a signal processing means for processing an output signal from the piezoelectric composite material sensor. Moreover, the signal processing means compares the signal pattern stored in the storage means with the storage means for storing a constant signal pattern as a reference signal in advance, and the output signal from the piezoelectric composite material sensor to determine whether or not there is an abnormality. And a determining means for determining.

[0011]

That is, according to the present invention, the sheet-shaped piezoelectric composite material sensor attached to the device can detect a signal over a wide frequency range. Then, the input signal from the piezoelectric composite material sensor is sent to the signal processing means, and the determination means can determine the presence or absence of abnormality. In this case, various signal patterns corresponding to the predictable factors of mechanical vibration are stored in the storage means in advance as reference signals, and these signal patterns are compared with the input signal in the determination means. By this comparison, the abnormality can be surely detected and the content of the abnormality can be identified. For example, the presence / absence of partial discharge, the content of abnormality such as noise or ground fault can be identified, and the operating state of the device can be monitored.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the predictive maintenance device according to the present invention will be described below with reference to FIG. In this case, FIG. 1 shows an embodiment in which the present invention is applied to a gas-insulated bus bar. First, the configuration of this embodiment will be described. Figure 1
In the figure, 1 is a closed container, 2 is a high voltage conductor, 3 is an insulating gas such as SF ₆ gas, 4 is an insulating spacer, 10
Shows a piezoelectric composite sensor according to the present invention. The number of the piezoelectric composite material sensors 10 is n from S1 to Sn, and the piezoelectric composite material sensors 10 are individually connected to the signal processing unit 20 via the corresponding signal lines 11 of L1 to Ln. In this case, in the figure, for convenience, only two sensors S1 and Sn and two signal lines L1 and Ln are shown. The signal processing unit 20 is a signal processing unit that processes an output signal from the piezoelectric composite material sensor 10, and includes a storage unit (storage unit) 21, an input unit 22, and a calculation unit (determination unit).
23 and a display unit 24.

Then, a fixed signal pattern is previously stored in the storage unit 21 as a reference signal, various setting values are input by the input unit 22, and the arithmetic unit 23
The signal pattern stored in the storage unit 21 and the output signal from the piezoelectric composite material sensor 10 are compared and calculated to determine the presence or absence of abnormality. In this case, in the storage unit 21 of the signal processing unit, various predictable factors of mechanical vibration, for example, various signal patterns due to opening / closing operation, partial discharge, noise, ground fault, etc. are stored in advance as reference signals. Keep it. Further, the calculation result (determination result) by the calculation unit 23 is displayed on the display unit 24 as necessary. In FIG. 1, the number of piezoelectric composite material sensors 10 is n, but one piezoelectric composite material sensor 10 will be described below to simplify the description.

Next, the operation of this embodiment will be described.
That is, in the present embodiment having the above configuration, since the piezoelectric composite material sensor 10 has a wide range of frequency characteristics, various mechanical vibrations can be widely detected without being limited to a specific frequency. It is possible. And
Since the obtained mechanical vibration signal has a certain pattern for each cause, various signals for each cause previously stored in the storage unit 21 by the calculation unit 23 of the signal processing unit 20 are used. By performing a comparison calculation with the signal pattern of, it is possible to automatically determine whether or not there is an abnormality and identify the content of the abnormality. Further, it is also possible to display a comparison between the stored signal pattern and the actually detected signal on the display unit 24.

As a more specific example, the operation when the metallic foreign matter is mixed in the container 1 of FIG. 1 will be described.
The foreign metal particles mixed in the container 1 move around the bottom of the container 1 by the AC voltage applied to the high-voltage conductor 2. In this state, when the metallic foreign matter collides with the bottom surface of the container 1, mechanical vibration and partial discharge are generated. This signal is detected by the piezoelectric composite material sensor 10, but the mechanical signal becomes a random signal with respect to the period of the voltage. On the other hand, when a partial discharge occurs inside the insulating spacer 4, a partial discharge synchronized with the voltage occurs. In this case, the presence / absence of a foreign metal substance or the occurrence of a partial discharge can be identified based on whether or not there is a signal component synchronized with the voltage.

As described above, according to the preventive maintenance device of this embodiment, since the sheet-shaped piezoelectric composite material sensor 10 is used, a mechanical signal can be faithfully detected over a wide frequency range. By learning and memorizing, it becomes possible to identify the operating status of the device, the presence or absence of abnormality, and the cause of corona occurrence. Therefore, it is possible to obtain much more accurate device internal information as compared with the related art.

By the way, in the above description, the method of signal processing and calculation was not described, but as such a system, for example, a system using a neural network can be adopted. The use of a neural network has the feature that signal input, learning, and determination can be completely automated. As a result, more accurate pattern recognition can be performed, and more reliable operation monitoring and abnormality diagnosis can be performed. The configuration of the signal processing means is
The present invention is not limited to this, and it is possible to freely select. For example, the determination means for determining the presence / absence of abnormality does not necessarily have to perform comparison by arithmetic processing, and comparison and determination by other methods are possible.

It is needless to say that the number of piezoelectric composite material sensors used in the present invention can be selected as appropriate, and the larger the number of sensors used, the more information can be collected. Further, regarding the mounting position of the piezoelectric composite material sensor according to the present invention, it is needless to say that the piezoelectric composite material sensor is installed around the insulating spacer or in the vicinity of a portion where abnormality may occur such as a bus bar. It can be attached to a place where mechanical operation needs to be monitored, such as a circuit breaker, a circuit breaker, or a grounding switch. In this case, the same effect can be obtained.

[0019]

As described above, according to the present invention, the piezoelectric composite material sensor which is easy to install and can detect over a wide frequency range is used, and the obtained sensor output signal is used. Since the signal processing means can appropriately process, it is possible to provide an excellent predictive maintenance device capable of surely detecting the abnormality of the high-voltage equipment and discriminating the content of the abnormality.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram showing an embodiment in which a predictive maintenance device according to the present invention is applied to a gas insulated busbar.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Container 2 ... High-voltage conductor 3 ... Insulating gas 4 ... Insulation spacer 10 ... Piezoelectric composite material sensor 11 ... Signal line 20 ... Signal processing part (signal processing means) 21 ... Storage part (storage means) 22 ... Input part 23 ... Calculation unit (determination means) 24 ... Display unit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hiroshi Murase 2-1, Ukishimacho, Kawasaki-ku, Kawasaki-shi, Kanagawa Stock company Toshiba Hamakawasaki factory

Claims (1)

Claim: What is claimed is: 1. A predictive maintenance device, wherein a sensor is attached to a high-voltage device, and the presence or absence of an abnormality is determined by the signal of the sensor. A piezoelectric composite material sensor in which electrodes are arranged on the electrode, and the device has a signal processing means for processing an output signal from the piezoelectric composite material sensor, and the signal processing means is fixed. Storage means for pre-storing the signal pattern as a reference signal and a determination means for comparing the signal pattern stored in the storage means with the output signal from the piezoelectric composite material sensor to determine the presence or absence of abnormality. A featured predictive maintenance device.