JP2006098270A - AC rotating machine maintenance method - Google Patents

Abstract

A maintenance method for an AC rotating machine capable of performing a minimum necessary rational maintenance management is provided.
The state of the discharge generated from the AC rotating machine is monitored based on the partial discharge voltage for each phase generated in the winding of the AC rotating machine while the AC rotating machine is rotating, and is regarded as abnormal from the monitoring result. Detects the discharge. And the position of the discharge regarded as the detected abnormality is specified by detecting the ultraviolet rays or sound accompanying the partial discharge while the AC rotating machine is rotating, and the position that needs repair is determined.
[Selection] Figure 1

Description

  The present invention relates to an AC rotating machine maintenance method.

Conventionally, AC rotating machines such as AC motors and AC generators have been maintenance methods based on regular maintenance. That is, the AC rotating machine maintenance method that has been performed so far is called TBM (Time Based Maintenance). For example, as shown in FIG. 16, the operation time is managed (step S10), and a certain amount of time is maintained. When the years passed, maintenance of predetermined items was performed (steps S20 to 30). For example, as shown in step S20, in the open inspection, there is an item for disassembling a part of the AC rotating machine and performing the work, and the cost for the work and the power outage for the work are necessary (for example, Patent Documents). 1).
Japanese Patent No. 2518405

  However, when the AC rotating machine is large, for example, extensive work is required to pull out the rotor during open inspection. In particular, in the case of a large-sized AC generator in which the winding is cooled with hydrogen, work such as extraction and re-encapsulation of hydrogen is required, resulting in a large construction cost. During the construction period, the AC rotating machine is forced to stop operating, and the period required for each work such as pulling out and reassembling the rotor is several days for small ones and tens of days for large ones. There is a problem that it takes a long time. Furthermore, if the rotor extraction work is extensive as described above, and it is determined that there is no problem as a result of the inspection, there is a disadvantage that the work itself is wasted.

Therefore, in order to prevent the lack of equipment maintenance due to cost reduction while maintaining reliability, CBM that performs maintenance suitable for the state of the equipment based on the information on the state of the equipment at present. (Condition Based Maintenance) is indispensable. Here, the CBM grasps the state of the device and performs the minimum necessary rational maintenance management based on the information.
The present invention has been made paying attention to such problems, and an object of the present invention is to provide a maintenance method for an AC rotating machine capable of performing a minimum necessary rational maintenance management.

In order to solve the above-mentioned problem, the invention according to claim 1 is a method for maintaining an AC rotating machine, wherein the AC rotating machine monitors the state of discharge generated from the AC rotating machine, and this monitoring is performed. It comprises an abnormal discharge detecting step for detecting a discharge that is regarded as abnormal from the result, and a discharge position specifying step for specifying the position of the discharge that is regarded as abnormal.
The invention according to claim 2 is the maintenance method for the AC rotating machine according to claim 1, wherein the abnormal discharge detecting step is a partial discharge for each phase generated in the winding of the AC rotating machine. It is characterized by monitoring based on the voltage and detecting the discharge considered as the abnormality.
The invention according to claim 3 is the maintenance method for an AC rotating machine according to claim 1 or 2, wherein the discharge position specifying step detects ultraviolet rays or sound accompanying partial discharge to detect partial discharge. It is characterized by specifying the position of.

  According to the present invention, the state of the AC rotating machine can be grasped without an uninterruptible power, and the maintenance time can be appropriately determined based on the information. The work can be started, and the maintenance by the CBM that performs the minimum necessary reasonable maintenance management becomes possible.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings as appropriate.
FIG. 1 is a flowchart for explaining the procedure of a maintenance method for an AC rotating machine according to the present invention.
As shown in the figure, in this AC rotating machine maintenance method, first, the state of discharge (partial discharge) generated from the AC rotating machine is monitored (step S1). Then, it is determined whether or not a partial discharge that is regarded as abnormal requiring maintenance has been detected (step S2). Next, when a partial discharge that is regarded as abnormal is detected in step S2, the position of the discharge that is regarded as abnormal detected is identified to determine a portion that needs repair (steps S3 to S4). And after specifying the position of the discharge regarded as the detected abnormality in step S4, various inspections, repairs, tests, etc. are performed (steps S20 to S30), and the repair status is confirmed (steps). S40) is performed. In addition, since implementation (process S20-process S30) content of various inspection, repair, a test, etc. is the same as that of the conventional maintenance method, detailed description is abbreviate | omitted. Here, the abnormal discharge detection step corresponds to steps S1 to S2 in FIG. The discharge position specifying step corresponds to steps S3 to S4.

Next, the abnormal discharge detection step will be described in detail.
First, the abnormal discharge detection apparatus applied in the abnormal discharge detection step will be described with reference to FIG.
As shown in FIG. 2, the abnormal discharge detection device includes at least coupling circuits 1 </ b> A to 1 </ b> C, a PDA (Partial Discharge Analyzer) device 2, and a personal computer 3.

  The coupling circuits 1 </ b> A to 1 </ b> C are circuits that electrically connect an AC rotating machine (in this example, the AC motor 4) such as a three-phase AC motor or AC generator to be diagnosed and the PDA device 2. For example, it comprises a capacitor C1 and a resistor R1. One end of the capacitor C <b> 1 is connected to a power supply terminal of the AC motor 4, for example, and the other end is connected to the input terminal 5 of the PDA device 2. One end of the resistor R1 is connected to the input terminal 5 of the PDA device 2, and the other end is grounded.

As shown in FIG. 2, the PDA device 2 includes filter circuits 6 </ b> A to 6 </ b> C and a central processing unit (CPU) 7.
The filter circuits 6A to 6C include a low-pass filter (LPF) 8 that extracts a commercial frequency component from the voltage at the power supply terminal of the AC motor 4 supplied via the coupling circuit 1, and a partial discharge voltage component from the voltage. It consists of a high pass filter (HPF) 9 to be taken out. Here, the cut-off frequency of the low-pass filter 8 is, for example, 100 [Hz], and the cut-off frequency of the high-pass filter 9 is, for example, 100 [MHz].

The central processing unit 7 determines the magnitude of the partial discharge voltage for each phase of the commercial frequency of the AC motor 4 extracted by the high-pass filter 9 and the frequency of occurrence of the partial discharge voltage. Here, FIG. 3 shows an example of the partial discharge voltage obtained by the central processing unit 7 and its occurrence frequency.
Further, the central processing unit 7 calculates an Nqn (Normalized Quantity Number) value by the following equation (1) based on the partial discharge voltage and its occurrence frequency (frequency) obtained as shown in FIG. 3, for example.

Here, in the equation (1), Vpdvid is a voltage value obtained by dividing the partial discharge voltage in accordance with the number of frequencies as shown in FIG. Pi is a value of occurrence frequency (frequency) as shown in FIG.
The personal computer 3 uses a pair of data on the Nqn value of the partial discharge generated at the timing when the potential of the iron core becomes larger than the winding of the AC motor 4 and the load (or temperature) of the AC motor 4. It is designed to be stored as a database.

  Further, the personal computer 3 creates a scatter diagram with the load of the AC motor 4 on the X axis and the Nqn value on the Y axis, and calculates the correlation coefficient between the load and the Nqn value. Furthermore, the personal computer 3 determines (diagnose) the presence or absence of loosening of the windings of the AC motor 4 to be diagnosed based on the calculated correlation coefficient as described later.

  By the way, when an AC rotating machine such as an AC electric motor or an AC generator is operated for a long time, the winding loosens due to aging of the wedge or the insulator that fixes the winding to the rotor, and a gap is generated between the insulator and the iron core. . In addition, the winding may be loosened due to production defects. When the size of the gap increases and the potential gradient (V / m) of the gap exceeds the dielectric breakdown voltage of air, a discharge is generated inside the gap. This is generally referred to as partial discharge. When loosening occurs in the winding, the winding moves easily inside the iron core, so the size of the air gap changes depending on the load applied to the winding and the temperature change of the winding. Change appears. Therefore, the inventor gradually changed the load or output of an AC motor or AC generator with known winding looseness (wedge looseness), and the partial discharge magnitude (Nqn value) each time the load or output was changed. ) Was obtained, and experimental results as shown in FIGS. 4 to 7 were obtained.

Here, FIG. 4 shows the case of an AC generator, where the wedge loosening rate is 20.3%. FIG. 5 shows an AC generator in which the wedge looseness is 50%. FIG. 6 shows the case of an AC motor in which the loosening rate of the wedge is 60%. FIG. 7 shows an AC generator in which the wedge looseness rate is 0% (new). Further, when the correlation coefficient R was obtained from the experimental results, the following results were obtained. That, ^R 2 = 0.2831 in the case of FIG. 4, ^R 2 = 0.3995 in the case of FIG. 5, ^R 2 = 0.7989 in the case of FIG. 6, in the case of FIG. 7 with ^R 2 = 0.0108 is there.

As a result of examining the above results, it can be seen that there is a difference in the correlation coefficient between the AC motor or AC generator in which the winding has loosened and the AC motor or AC generator in which no looseness has occurred. That is, in an AC motor or an AC generator in which the winding is significantly loosened, a strong correlation appears between the change in the load or output and the change in the partial discharge (see, for example, FIG. 6). In no AC motor or AC generator, there is no correlation between the change in load or output and the change in partial discharge (see, for example, FIG. 7).
Then, the abnormal discharge detection process in the maintenance method of the AC rotating machine of the present invention is carried out using the abnormal discharge detection device based on the above knowledge.

Next, the abnormal discharge detection process in the maintenance method of the present invention, which is performed using the abnormal discharge detection apparatus configured as described above, will be described in more detail.
Here, the maintenance method using the abnormal discharge detection device described above can be applied to both an AC motor and an AC generator. However, in the following description, a case where the maintenance method is applied to the AC motor 4 will be described.
In the maintenance method of this embodiment, the loosening rate of the winding of the AC motor is known in order to diagnose the loosening of the winding of the AC motor (the looseness of the wedge) and monitor the change of the partial discharge. For a plurality of AC motors having different rates, a correlation coefficient between the change in load and the Nqn value corresponding thereto is obtained in advance, and a graph or the like in which the looseness of the winding is correlated with the correlation coefficient, for example, It is necessary to prepare in advance as shown in FIG. The preparation work for the preparation will be described below.

First, a plurality of AC motors with known wedge loosening rates used to fix the windings are prepared. Next, one end of each of the coupling circuits 1A to 1C of the abnormal discharge detection device shown in FIG. 2 is connected to each power supply terminal of the AC motor, and then the AC motor is rotated to a predetermined load state. Under this constant load, each part of the abnormal discharge detection device performs the following processing.
The filter circuits 6A to 6C take out components of the commercial frequency from the respective voltages of the respective power supply terminals of the AC motor by the low-pass filter 8, take out the components of the partial discharge voltage from the respective voltages by the high-pass filter 9, and take out these components. Is supplied to the central processing unit 7.

The central processing unit 7 determines the magnitude of the partial discharge voltage for each phase of the commercial frequency of the AC motor extracted by the high-pass filter 9 and the frequency of occurrence thereof over a predetermined time. Further, the central processing unit 7 calculates the Nqn value by the above equation (1) based on the obtained result, and transmits the calculation result to the personal computer 3.
Thereafter, the load of the alternating current machine is gradually changed. In order to change the load, the Nqn value corresponding to the load is obtained as described above, and the obtained Nqn value is notified to the personal computer 3.
Further, the personal computer 3 creates a scatter diagram such as shown in FIG. 6 from the load change data and the corresponding Nqn data, and the correlation coefficient between the load and the Nqn value based on this scatter diagram. Is calculated.

With the above processing, the correlation coefficient between the load and the Nqn value is calculated for the AC motor whose wedge looseness rate is known. For the remaining AC motors with known wedge loosening rates, the correlation coefficient between the load and the Nqn value is calculated in the same manner as the AC motor described above.
In this way, for a plurality of AC motors with known wedge loosening rates, the correlation coefficient between the load and the Npn value is obtained, and the relationship between the correlation coefficient and the wedge loosening rate is represented, for example, in FIG. The result is obtained and stored (registered) in the personal computer 3 in advance.

Next, on the premise of the above-described work, the partial discharge in the AC motor 4 to be diagnosed is monitored while the AC rotary machine is rotating and the partial discharge that is regarded as abnormal is detected by the abnormal discharge detection device of FIG. A detection method will be described.
In this case, when one end of the coupling circuits 1A to 1C of the abnormal discharge detection device of FIG. 2 is connected to each power supply terminal of the AC motor 4, the result is as shown in FIG. Thereafter, the AC motor 4 is rotated to a predetermined load state. Under this constant load, each part of the abnormal discharge detection device performs the following processing.
Each of the filter circuits 6A to 6C extracts a commercial frequency component from each voltage of each power supply terminal of the AC motor 4 by the low-pass filter 8, and extracts a partial discharge voltage component from each voltage by the high-pass filter 9. The components are supplied to the central processing unit 7.

The central processing unit 7 obtains the magnitude of the partial discharge voltage for each phase of the commercial frequency of the AC motor 4 extracted by the high-pass filter 9 and the frequency of occurrence of the partial discharge voltage over a predetermined time. Further, the central processing unit 7 calculates the Nqn value by the above equation (1) based on the obtained result, and transmits the calculation result to the personal computer 3.
Thereafter, the load of the alternating current machine 4 is gradually changed, and each time the load is changed, the Nqn value corresponding to the load is obtained as described above, and the obtained Nqn value is notified to the personal computer 3. The partial discharge is monitored (step S1 above).

The personal computer 3 creates a scatter diagram based on the load change data and the corresponding Nqn data, and calculates a correlation coefficient between the load and the Nqn value based on the created scatter diagram. The personal computer 3 detects, based on the calculated correlation coefficient, a partial discharge that is regarded as abnormal by, for example, determining whether or not the winding is loose with reference to FIG. 8 (step S2). For example, according to FIG. 8, when the calculated correlation coefficient is 0.6, for example, the looseness rate of the wedge is 54%, and the looseness rate of the wedge can be quantitatively evaluated.
In this way, the change in partial discharge can be monitored in an uninterrupted state, and the partial discharge that is regarded as abnormal can be detected from the monitoring result.

Next, the discharge position specifying step will be described in detail.
By the way, the abnormal discharge detection process by the abnormal discharge detection device is a very effective means for detecting insulation deterioration in an uninterruptible state. After obtaining information that the value exceeded the limit, when trying to check the presence or absence of discharge at the site, in order to confirm where the discharge is occurring and in what aspect In order to do so, it must be confirmed after the power is turned off. That is, in the abnormal discharge detection step, partial discharge that is regarded as abnormal can be detected, but the site and aspect of the discharge cannot be detected. Therefore, in the subsequent discharge position specifying step, the discharge site and its appearance are detected by light (ultraviolet rays) uninterrupted by the discharge position specifying device.

Here, first, the discharge phenomenon and ultraviolet rays will be described.
It is known that the discharge phenomenon involves emission of ultraviolet rays in addition to electromagnetic waves and ultrasonic waves. Discharge is caused by gas atoms in an excited state under the influence of an electric field emitting free electrons and being ionized. At this time, light is emitted in the process in which the ionized atom receives free electrons and returns from the excited state to the ground state.
Now, the wavelength of light emitted when the electron trajectory transitions from n = i to n = j is given by the following equation (2).

Where λ: wavelength of emitted light, m: kinetic mass of electron, e: charge of electron (1.602 × 10 ⁻¹⁹ C), ε ₀ : dielectric constant of vacuum (8.854 × 10 ⁻¹² F) / M), h: Planck's constant (6.624 × 10 ⁻³⁴ J · sec), c: speed of light (3 × 10 ⁸ m / sec)
Here, the inventor of the present application pays attention to the fact that the ionization voltage of each atom is given by the following equation (3), and the emission wavelength of a gas whose ionization voltage is known from the equations (2) and (3) is: I thought that it can be obtained in (4).

  FIG. 9 shows an emission spectrum of a typical gas obtained by the equation (4). From the figure, it can be seen that the wavelength of light generated by the discharge is discretely distributed in the ultraviolet region of 400 nm or less. On the other hand, ultraviolet rays are classified into UV-A from 400 nm to 315 nm, UV-B from 315 nm to 280 nm, and UV-C from 280 nm to 100 nm depending on the wavelength.

  By the way, ultraviolet rays are short-wave electromagnetic waves with a wavelength of 400 nm or less emitted from artificial illumination such as the sun and high-intensity lamps. In particular, in the case of sunlight radiation reaching the ground surface, the emission distribution shown in FIG. Thus, the shorter the wavelength, the smaller the occupation ratio in solar radiation. In particular, the electromagnetic wave in the UV-C region is absorbed by the ozone layer in the atmosphere, and therefore hardly reaches the ground surface. On the other hand, the light emission distribution by the partial discharge is as shown in FIG. 10 and radiates over the entire ultraviolet region. Therefore, by detecting ultraviolet rays in the UV-C region that do not exist in solar radiation that reaches the earth's surface, it is not affected by solar radiation, and even weak discharge light is not affected by sunlight. Ultraviolet rays accompanying light emission from the subject itself due to the occurrence can be detected.

Next, the structure of the discharge position specifying device applied to the discharge position specifying step will be described. FIG. 11 is a schematic configuration diagram of the apparatus.
In the cylindrical case 21, from the object 30 side, the UV filter 22, the UV lens 23, the UV image intensifier 24, the UV lens 23, the UV image intensifier 24, and the UV are arranged so that the optical path L is coaxial. The lens 23 and the video camera 25 are arranged in this order.
Each of the UV image intensifiers 24 is a device for detecting and multiplying a weak ultraviolet image to convert it into a contrasted image, that is, a visualized image. In addition, since a normal CCD camera does not have sensitivity to ultraviolet rays, the discharge phenomenon cannot be imaged.

Each UV image intensifier 24 doubles photoelectrons by processing a light-receiving surface that converts received light into electrons and an infinite number of fine holes (channels) with a diameter of 12 μm on a thin glass plate with a diameter of 25 mm and a thickness of 0.48 mm. MCP (multi-channel plate) and a fiber plate that converts the doubled electrons into light again and transmits them to the subsequent stage. In this embodiment, Ce-Te having sensitivity to ultraviolet rays is selected as the material for the light receiving surface of each UV image intensifier 24. The Ce-Te light-receiving surface is suppressed in sensitivity to light having a wavelength of 320 nm or more and has maximum sensitivity in the vicinity of 250 nm. And MCP made the electron multiplication rate 10 ⁶ times by considering the weak discharge light and making it two steps. Here, since each UV image intensifier 24 has a high amplification factor, it forms an image even with a small amount of light and obstructs discrimination of the discharge light to be originally observed. Therefore, a UV filter 22 that allows only UV-C light to pass therethrough is installed in front of the UV image intensifier 24 to block unnecessary light.

  By the way, normally, the filter includes a band pass filter that passes a specific wavelength band and a narrow band pass filter that passes only a specific wavelength (see, for example, FIG. 9). A filter that passes only UV-C ultraviolet rays in the range of 280 nm is preferred. In particular, as a result of laboratory tests with five types of narrow band pass filters in filter selection in this embodiment, as shown in FIG. 12, the narrow band pass filter that passes 249.7 nm has the best S / N ratio. It was suitable for visualization of discharge. In the example of the figure, discharge is generated at the bottom center of the screen, but it can be seen that it is difficult to discriminate between the discharge light and the background sunlight with a filter other than 260BP and 249.7NB7.

Here, in the image that has passed through the 249.7NB7 filter shown in FIG. 12, Ne (neon) emission lines are mainly observed as can be seen with reference to FIG. Neon nitrogen (volume ratio 78.088% in air), oxygen (up 20.949%), argon volume ratio in the atmosphere subsequent to (the same 0.93%) larger gas (the 1.8 × 10 ^{- 3} %), and the volume ratio is the largest among gases having bright lines in the UV-C region. Therefore, observing the bright line of Ne (neon) is preferable in discriminating the discharge light because the amount of emitted light is large. Further, in the present embodiment, a narrow band pass filter that passes one or more emission lines of 249.7 nm, that is, Ne (neon) among UV-C ultraviolet rays existing in a wavelength range of 240 to 280 nm is provided. It is adopted as the UV filter 22. Therefore, by adopting such a narrow band pass filter, the discrimination between the discharge light and the background sunlight can be made more reliable.

Here, each of the UV lenses 23 is a lens having sensitivity to ultraviolet rays (particularly, it is preferable that UV-C is sensitive). The video camera 25 is a recording medium that constitutes an image acquisition unit and continuously acquires and records a visualized ultraviolet image that passes through the UV filter 22 and is visualized.
The plurality of UV lenses 23 and the UV image intensifier 24 constitute a visualization conversion unit. In view of the weak radiation of UV-C ultraviolet rays due to partial discharge, the present embodiment illustrates a case where a plurality of UV lenses 23 and UV image intensifiers 24 are arranged.

  An ultraviolet lamp 26 is attached to the outside of the case 21. The ultraviolet lamp 26 is an ultraviolet irradiation means for irradiating the object 30 with ultraviolet rays. For example, a high-intensity lamp 26 such as a high-intensity UV lamp using quartz glass may be used. In particular, an ultraviolet lamp capable of mainly emitting ultraviolet rays having a wavelength in the UV-C region that can pass through the UV filter 22 is preferable. The irradiation direction of the ultraviolet lamp 26 can be adjusted with respect to the case 21.

An openable / closable shutter 27 is attached in front of the ultraviolet lamp 26, and the shutter 27 is opened and closed by a shutter actuator 28 made of a drive motor, an electromagnet or the like.
Reference numeral 29 denotes a controller, which turns on the ultraviolet lamp 26 when an operation switch (not shown) is turned on, and repeatedly opens and closes the shutter 27 at predetermined time intervals through the shutter actuator 28. . Although the opening / closing speed of the shutter 27 is not particularly limited, for example, the opening / closing speed is set to be repeated at intervals of 1 second.

Next, the discharge position specifying step in the maintenance method of the present invention, which is performed using the discharge position specifying apparatus configured as described above, will be described in more detail.
Now, in a state where the AC motor 4 is rotating, the optical path L such as the UV filter 22 arranged in the case 21 is directed to the AC motor 4, and the radiation direction of the ultraviolet lamp 26 is also directed to the same AC motor 4. Then, an operation switch (not shown) is turned on. Then, ultraviolet rays are irradiated from the ultraviolet lamp 26 toward the AC motor 4, but the irradiation of ultraviolet rays is periodically blocked by the shutter 27 that repeatedly opens and closes. In this state, the video camera 25 is also in the recording state. The operation of the video may be controlled by the controller 29 (step S3).

  By setting in this way, the reflected light from the AC motor 4 due to the irradiation of the ultraviolet rays and the light emitted from the AC motor 4 due to the blocking of the ultraviolet rays enter the UV filter 22 alternately. Then, only the reflected light from the AC motor 4 by the ultraviolet lamp 26 and the ultraviolet light emitted by the AC motor 4 pass through the UV filter 22 alternately, so that the influence of sunlight reaching the ground surface is removed.

  The two types of ultraviolet images that alternately pass through the UV filter 22 are doubled and emphasized by two sets of UV lens 23 and UV image intensifier 24, respectively, and converted into a visible state, and in turn the video camera 25. Are recorded and recorded. FIG. 13 shows an image diagram of the two types of ultraviolet images. (A) is the figure of the object by ultraviolet irradiation, (b) is the figure by light emission of the object by partial discharge. Actually, in the figure of (b), only the portion of the partial discharge portion A is an white image.

  Then, when the recorded image is reproduced, in the state where there is no partial discharge, only a dark image and a black-and-white image of the AC motor 4 are displayed alternately, but in the AC motor 4 where the partial discharge occurs, Two types of images, that is, an ultraviolet image in which only the light emission part of the partial discharge is white and a black and white image of the AC motor 4 are displayed alternately. As a result, the presence / absence of partial discharge is detected, and an ultraviolet image in which only the partial discharge light-emitting portion is white and the entire image of the AC motor 4 are alternately displayed, whereby the position of the partial discharge in the AC motor 4 and its By visually recognizing the aspect, it is possible to determine a part that needs repair (step S4). In particular, by adjusting the playback speed etc. and projecting two types of images alternately so that the afterimage of one image and the next projected image appear to completely overlap with each other, partial discharge can be more easily performed. It becomes easy to specify the position. It should be noted that the adjustment of the appearance of the afterimage can be adjusted by adjusting the reproduction speed, for example, by extending the video display time of the ultraviolet image by light emission.

At this time, by acquiring the image of the AC motor 4 as an ultraviolet image, the image of the AC motor 4 itself is naturally suppressed to a darker state. The whitened portion due to light emission is naturally emphasized.
Here, without actively irradiating from the ultraviolet lamp 26, an image of light emission by partial discharge is acquired by the apparatus, and an image of the AC motor 4 is taken by another camera in synchronization with both. It is also conceivable to detect the discharge position by superimposing these images. However, not only another camera that captures the image of the AC motor 4 is required, but also the position for receiving the image is different, so superposition processing is required.

  On the other hand, in this discharge position specifying device, only one camera is required to capture an image, and since both the image of the discharge and the image of the AC motor 4 are acquired at the position where the optical path L is coaxial, the device configuration is simple. Turn into. In addition, since this discharge position specifying device does not directly superimpose two images on the image processing unit, but alternately displays them, for example, specifies the position of partial discharge using the visual afterimage phenomenon. The process of superimposing two images is also unnecessary.

Next, the operation and effect of this AC rotating machine maintenance method will be described.
According to the maintenance method of this embodiment, in the abnormal discharge detection process, the loosening of the winding is judged while the AC motor is rotating, that is, under the operating state, and the partial discharge state is monitored by the loosening of the winding. Therefore, it can be diagnosed at an arbitrary timing whether or not the state of the partial discharge is a discharge regarded as an abnormality requiring maintenance. Furthermore, since the degree of partial discharge (winding looseness) can be grasped quantitatively, it is possible to select an optimal timing for the maintenance according to the state of the motor or the like.

  Then, according to the maintenance method of this embodiment, when a discharge that is considered abnormal in the abnormal discharge detection step is detected, the discharge position specifying step is performed by light (ultraviolet rays) from an AC rotating machine by a discharge position specifying device. Therefore, if an open-type AC rotating machine is used, it is possible to visually identify the location of the discharge that is considered abnormal without stopping the operation, and to prevent the discharge caused by the insulation deterioration. It is possible to perform maintenance measures that are detected at an early stage and prevent the accident from spreading. In the case of a closed AC rotating machine, a maintenance voltage can be applied by applying a test voltage and visually confirming the site where the discharge occurs during open inspection.

  As described above, according to the maintenance method for an AC rotating machine according to the present invention, in the abnormal discharge detection step, the state of the discharge generated from the AC rotating machine in the abnormal discharge detection step is changed to AC. Monitor while the rotating machine is rotating. Then, when a discharge that is regarded as abnormal is detected in the abnormal discharge detection step, the position of the discharge that is regarded as a detected abnormality is specified in the discharge position specifying step while the AC rotating machine is operating or stopped. . Therefore, it is possible to grasp the state of the AC rotating machine without an uninterruptible power and to appropriately determine the maintenance time based on the information, and to determine the place where maintenance is deemed necessary in an uninterruptible state in advance and then start the maintenance work Therefore, it is possible to perform maintenance by performing a minimum necessary rational maintenance management. Therefore, it is possible to rationalize regular inspections based on periodic inspections that have been carried out in TBM, that is, inspections of AC rotating machines, that is, maintenance by CBM, and overall cost reductions such as support for extending the life of equipment. It is possible to provide a maintenance method for an AC rotating machine that can measure

In addition, the maintenance method of the AC rotating machine according to the present invention is not limited to the above embodiment, and various modifications can be made without departing from the gist of the present invention.
For example, in the above-described embodiment, in the abnormal discharge detection step, the magnitude of the partial discharge voltage for each phase of the AC motor 4 and the frequency of occurrence of the partial discharge are obtained over a predetermined time, and based on the obtained results ( 1) The Nqn value is calculated by the equation. Then, the load of the AC motor 4 is gradually changed, and each time the load is changed, the Nqn value corresponding to the load is obtained as described above. Further, a scatter diagram is created based on the load change data and the corresponding Nqn data, a correlation coefficient between the load and the Nqn value is calculated based on the created scatter diagram, and the calculated correlation For example, with reference to FIG. 8, the presence or absence of loosening of the winding is determined by the number.

However, in the maintenance method for an AC rotating machine according to the present invention, the Nqn value may be replaced with a maximum discharge voltage value (Qmax).
In this case, the magnitude of the partial discharge voltage for each phase of the AC motor 4 and the frequency of occurrence of the partial discharge voltage are respectively obtained, and the maximum value is extracted as Qmax from the obtained partial discharge voltages. . However, since the partial discharge voltage includes noise, considering the influence of the noise, for example, the maximum one of the partial discharge voltages having a frequency of occurrence of 10 or more per second is preferably set to Qmax. .

  Then, the load on the AC motor 4 is gradually changed, and each time this load is changed, Qmax corresponding to the load is obtained. Further, a scatter diagram is created based on the load change data and the corresponding Qmax data, and the correlation coefficient between the load and Qmax is calculated based on the created scatter diagram. Thus, it is determined whether or not there is a discharge regarded as abnormal.

  Further, for example, in the above-described embodiment, in the discharge position specifying device in the discharge position specifying step, the alternately acquired images are reproduced so that they appear to overlap each other by the afterimage phenomenon. You do not have to use it. It may be played back slowly in a frame feed. Since the ultraviolet image by partial discharge can be easily identified, when the ultraviolet image by partial discharge is detected, the approximate position of the light emission is stored (see FIG. 13B), and the alternating current imaged next is displayed. The position of the partial discharge can be specified by looking at the corresponding position in the image of the electric motor 4 (see FIG. 13A). However, it is easier to visually recognize the image after reproducing it so as to overlap using the afterimage phenomenon.

  Moreover, in the said embodiment, although the video camera 25 is illustrated as said image acquisition means in the discharge position specification apparatus, a digital camera etc. may be sufficient. In the case of a digital camera, a still image is captured in synchronization with the opening and closing of the shutter 27, and an ultraviolet image reflected by ultraviolet irradiation and an ultraviolet image emitted by partial discharge (a dark image if there is no light emission). May be obtained alternately.

  Moreover, in the said embodiment, in the discharge position specification apparatus, although the camera was illustrated as an image acquisition means and it demonstrated by the case where the presence or absence of partial discharge was confirmed after imaging | photography, it is not limited to this. For example, the image acquisition means is a display device such as a display, and on the spot, an ultraviolet image by partial discharge and an ultraviolet image by reflection from the AC motor 4 are alternately displayed, and whether or not there is partial discharge on the spot. And you may make it specify the discharge position in case the partial discharge has generate | occur | produced.

In addition, color conversion processing may be performed only on an ultraviolet image by partial discharge by image processing, that is, the white portion may be converted into red or the like to emphasize the partial discharge portion. In this case, since the ultraviolet image resulting from the reflection from the AC motor 4 is a black and white image, the position of the partial discharge can be visually recognized more easily.
Further, the case 21 and the ultraviolet lamp may be separate.

  In the example of the above embodiment, the UV filter 22 employs a narrow band pass filter that passes a bright line of 249.7 nm Ne (neon), but is not limited thereto, and the narrow band pass filter has a wavelength of You may observe by the narrow-band pass filter which passes the bright line of 1 or 2 or more Ne (neon) of another wavelength among UV-C ultraviolet rays which exists in the range of 240-280 nm.

Moreover, in the said embodiment, although the position of discharge is specified by detecting the light (ultraviolet ray) by discharge in a discharge position specification process, it is not limited to this, For example, by detecting the sound by discharge The position of discharge may be specified.
Here, as another example of the discharge position specifying device used in the discharge position specifying step, an example in which the position of the discharge is specified by detecting sound due to discharge will be described in detail below.

FIG. 14 is a block diagram showing a configuration of another example of the discharge position specifying device, and FIG. 15 is a characteristic diagram showing output waveforms at each element of the discharge position specifying device.
In these figures, reference numeral 11 denotes an acoustic detection unit such as an ultrasonic microphone having a strong directivity for detecting the sound caused by the discharge, and the sound receiving unit can swing freely. 12 is a first-stage amplifier, 13 is a high-pass filter with a cutoff frequency of, for example, 10 kHz, 14 is a second-stage amplifier, 15 is a rectifier, 16 is a low-pass filter with a cutoff frequency of, for example, 200 Hz, and 17 is an A / A A D converter, 18 is a central processing unit, and 19 is a display. Reference numeral 20 denotes an acoustic azimuth detecting unit such as a pointer by a laser diode, which is attached in the vicinity of the acoustic detecting unit 11 and visually confirms the location of the discharge sound by indicating the direction of the acoustic due to the discharge with light or the like. Can do. These devices are integrally assembled in dimensions that are easy to carry and are configured so that electrical maintenance personnel can take them to the site for measurement.

Next, the operation of the discharge position specifying device configured as another example will be described.
In another example of the discharge position specifying device, in the discharge position specifying step in the maintenance method of the AC rotating machine, first, the acoustic direction detection unit 20 such as a laser pointer is used to point the direction of the generation of the discharge sound. After confirming, the sound detection unit 11 is directed to the AC rotating machine in the direction, and a discharge sound having a waveform as shown in FIG. The discharge sound collected by the acoustic detection unit 11 is first amplified to a waveform as shown in FIG. 10B by the first-stage amplifier 12 and passes through the high-pass filter 13 to remove noise. . After the noise is removed, it is amplified again by the second stage amplifier 14. The signal at this time is shown in FIG. Here, the function of noise removal in the high-pass filter 13 will be described. Usually, the sound generated by the discharge exists over a wide frequency band at a high frequency of 10 kHz or more, but the environmental sound of general substation equipment is mostly 10 kHz. Because of the following frequency components, only the discharge sound can be easily extracted by the high-pass filter 13.

  Then, the signal after noise removal is rectified by the rectifier 15 into a waveform as shown in FIG. 10 (d), and then envelope detection is performed by the low-pass filter 16, and the intensity component of the discharge sound as shown in FIG. 10 (e). To extract. Further, the analog signal of the intensity component of the discharge sound is converted into a digital signal by the A / D converter 17 and led to the central processing unit 18. Here, the analog signal after the envelope detection using the A / D converter 17 is converted into a digital signal. The signal before the envelope detection becomes a wide-range signal up to the ultrasonic region (about 40 kHz). On the other hand, the signal after envelope detection is several times the frequency of the power supply voltage (about 200 Hz), the sampling rate can be set low, and the load on the central processing unit 18 can be reduced. Furthermore, the central processing unit 18 can display a spectrum of the frequency component of the signal by FFT (Fast Fourier Transform) processing, and display a signal waveform that allows visual confirmation of the presence or absence of signal phase transition. .

It is a flowchart explaining the procedure of the maintenance method of the alternating current rotating machine which concerns on this invention. It is a block diagram which shows an example of a structure of the abnormal discharge detection apparatus applied to embodiment of this invention method. It is a figure which shows an example of the partial discharge voltage detected and its generation frequency (frequency). It is a figure which shows the relationship between the output and Nqn in the AC generator in which the loosening rate of a wedge is known. It is a figure which shows the relationship between the output and Nqn in the AC generator in which the loosening rate of a wedge is known. FIG. 5 is a diagram showing the relationship between the load and Nqn in an AC motor with a known wedge loosening rate. It is a figure which shows the relationship between the output and Nqn in a new AC generator. It is a figure which shows an example of the correlation coefficient of a load-Nqn value, and the loosening rate of a wedge. It is explanatory drawing which shows the emission spectrum of typical gas. It is a figure explaining light emission distribution. 1 is a schematic configuration diagram of a discharge position specifying device applied to an embodiment of the method of the present invention. It is a figure explaining the result of the laboratory test for filter selection. It is a figure explaining the image of two types of ultraviolet images. It is a block diagram which shows the other structure of a discharge position specification apparatus. It is a characteristic view which shows the output waveform in each element of the discharge position specification apparatus by another structural example. It is a flowchart explaining the procedure of the maintenance method of the conventional alternating current rotating machine.

Explanation of symbols

1A to 1C coupling circuit 2 PDA device 3 personal computer 4 AC motor 6A to 6C filter circuit 7 central processing unit 11 acoustic detection unit 12 first stage amplifier 13 high pass filter 14 second stage amplifier 15 rectifier 16 low pass Filter 17 A / D converter 18 Central processing unit 19 Display 20 Acoustic direction detection unit 21 Case 22 UV filter 23 UV lens 24 Image intensifier 25 Video camera 26 Ultraviolet lamp 27 Shutter 28 Actuator 29 Controller 30 Target L Optical path

Claims (3)

  The state of the discharge generated from the AC rotating machine is monitored while the AC rotating machine is rotating, and an abnormal discharge detecting step for detecting discharge regarded as abnormal from the monitoring result, and the position of the discharge regarded as abnormal are specified. An AC rotating machine maintenance method comprising: a discharge position specifying step.   2. The AC according to claim 1, wherein the abnormal discharge detection step monitors based on a partial discharge voltage for each phase generated in a winding of the AC rotating machine and detects a discharge regarded as the abnormal. Rotating machine maintenance method.   3. The maintenance method for an AC rotating machine according to claim 1, wherein the discharge position specifying step specifies the position of the partial discharge by detecting ultraviolet rays or sound accompanying the partial discharge.