JP3492240B2 - Reactor monitoring device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a reactor monitoring device (loose parts monitor) for detecting an abnormality in a nuclear reactor, and more particularly to a loose parts monitor capable of estimating the cause of the abnormality.

[0002]

2. Description of the Related Art FIG. 4 is a block diagram of a conventional loose parts monitor. In FIG. 4, a signal from an accelerometer 1 mounted on the surface of a reactor pressure vessel (not shown) is input to a loose parts monitoring system board 2. The input signal is amplified by the amplifier 3 and then discriminated into a normal signal and an abnormal signal by the impact detection device 4. The identified result is sent to the diagnostic computer 5, and the diagnostic result is displayed on the display device 6. The output signal of the amplifier 3 is input to the data recorder 7 and recorded therein when an abnormal signal is detected.

As an example of the signal processing procedure in the shock detector 4, the signal processing procedure disclosed in Japanese Patent Publication No. 63-41409 will be described with reference to FIG. FIG. 5 (1) identifies whether the input signal to the impact detection device 4 is a noise spike, which is a normal signal, or a signal (metal collision signal) generated during metal collision, which is one of abnormal signals. The principle for doing is shown. In FIG. 5A, the width (that is, the duration) of the pulse PW corresponding to the time when the level of the envelope waveform E of the noise spike S exceeds the threshold value AB, and the envelope of the metal collision signal S ′. The level of the line waveform E'is the threshold A'-
The width (duration) of the pulse PW 'corresponding to a time greater than B'is compared.

That is, the duration of the pulse PW corresponding to the noise spike S is relatively short, and the duration of the pulse width PW 'corresponding to the metal collision signal S'is relatively long. Therefore, it is possible to identify whether the input signal is the noise spike S or the metal collision signal by comparing the pulse width corresponding to the input signal to the impact detection device 4 with a predetermined reference width. Become. If the signal is a metal collision signal, the alarm signal AL is output.

FIG. 5B shows an example of the structure of the impact detection device 4 which realizes the above-mentioned identification principle. As shown in FIG. 5B, the envelope detector ED first detects the envelope waveform of the input signal. And the threshold crossing detector TCD
Compares the level of the envelope waveform with a predetermined threshold value RS, and generates a pulse signal corresponding to a time larger than the threshold value RS. Further, the pulse width discriminator PWD determines whether the input signal is a normal signal or an abnormal signal based on the width of the pulse signal, and outputs an alarm signal AL if the input signal is an abnormal signal.

[0006]

As described above, the conventional loose parts monitor can detect an abnormal signal having a characteristic different from that of the normal signal. However, even if the abnormal signal is detected, what kind of cause It was difficult to automatically estimate whether the collision was due to a metal collision (collision of loose parts) or a signal other than the metal collision.

Therefore, conventionally, in order to estimate the cause of an abnormal signal, signal processing by a skilled person and examination by a plant engineer are required, and it is difficult to quickly respond to the occurrence of an abnormality. It takes a lot of effort to estimate the cause.

Therefore, an object of the present invention is to provide a reactor monitoring device (loose parts monitor) capable of easily estimating the cause of a detected abnormal signal.

[0009]

The structure of a reactor monitoring apparatus of the present invention for achieving the above object is a detector for detecting an abnormal signal generated by an abnormality of a nuclear reactor, and a detector for detecting the abnormal signal. An arithmetic unit that calculates a predetermined characteristic amount of the abnormal signal, a storage unit that stores the predetermined characteristic amount of the abnormal signal that occurred in the past and data of the cause of occurrence corresponding to the characteristic amount, and the characteristic calculated by the arithmetic unit An estimating unit that estimates the cause of the abnormal signal detected by the detecting unit based on a comparison between the amount and the characteristic amount stored in the storage unit.

As described above, the reactor monitoring device (loose parts monitor) estimates the cause of the detected abnormal signal. Therefore, in order to estimate the cause of occurrence of an abnormal signal, it is not necessary to have signal processing by a skilled person or examination by a plant engineer, and it is possible to quickly respond to an abnormality when an abnormality occurs and the cost required for estimating the cause of an abnormality. Can be reduced.

Further, the storage section further stores the feature amount of the abnormal signal that has occurred in the past in another nuclear reactor and the data of the cause of occurrence corresponding to the feature amount.

With this, the storage unit can accumulate more information (predetermined feature amount of the abnormal signal that has occurred in the past and data of the cause of occurrence corresponding to the feature amount), so that the estimation with higher accuracy is possible. Is possible.

Further, a processing device including a detection unit and a calculation unit is arranged for each of the plurality of plants, and a center device including a storage unit and an estimation unit is arranged apart from the plurality of plants. The center device and the center device are connected by a communication line, respectively, and the feature amount calculated by the calculation unit of the one processing device is transmitted from the one processing device to the center device, and the estimation result by the estimation unit is obtained from the center device. It is transmitted to one processing device.

As described above, since the storage section and the estimation section are shared by using communication, it is not necessary to provide them for each plant, and the cost can be reduced.

The predetermined feature amount is, for example, the primary moment radius r1 of the abnormal signal and the secondary moment radius r.
These are a second- and third-order moment radius r3, an amount α indicating the attenuation characteristic of the waveform of the abnormal signal, an amount CF and RMS indicating the sharpness of the peak of the waveform of the abnormal signal.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below. However, the technical scope of the present invention is not limited to this embodiment.

FIG. 1 is a block diagram of a reactor monitoring device (loose parts monitor) according to an embodiment of the present invention. The configuration of FIG. 1 will be described in comparison with FIG. 4 described above. The loose parts monitor system board 2 includes an A / D converter 8 in place of the data recorder 7 of FIG. It is input to the / D converter 8. Also,
The trigger signal line 9 is connected to the A / D converter 8 from the impact detection device 4. Furthermore, the output of the A / D converter 8 is input to the diagnostic computer 5. Also, a database 1 described later
0 is connected to the diagnostic computer 5.

The operation of the loose parts monitor of FIG. 1 will be described. The output signal from the amplifier 3 is the impact detection device 4
It is also input to the A / D converter 8.
Then, the A / D converter 8 converts the input signal into digital data and holds it. On the other hand, when the signal input to the impact detection device 4 is determined to be an abnormal signal based on the above-described identification principle, the trigger signal is transmitted from the impact detection device 4 to A / A via the trigger signal line 9. It is sent to the D converter 8.

When the A / D converter 8 receives the trigger signal, the A / D converter 8 sends digital data corresponding to the abnormal signal to the diagnostic computer 5. Diagnostic computer 5 is A / D
The following arithmetic expressions (1) to (6) are executed on the digital data of the abnormal signal sent from the converter 8 (that is, the digitized waveform data of the abnormal signal) to obtain six feature values r.
Calculate 1, r2, r3, α, CF, RMS. Feature amount r
1, r2, and r3 are quantities that indicate the characteristics of the signal spectrum, and indicate the first-order moment radius, the second-order moment radius, and the third-order moment radius, respectively. Also, α is
It is a quantity indicating the attenuation characteristic of the waveform. Furthermore, CF and RMS
Is a quantity indicating the sharpness of the peak of the waveform.

[0020]

[Equation 1]

On the other hand, the database 10 stores the feature quantities of abnormal signals that have occurred in the past and the corresponding data of the cause of occurrence. The diagnostic computer 5 reads the past data accumulated in the database 10 after the calculation of the characteristic amount. Then, the degree of coincidence ε _k between the feature amount of the new abnormal signal obtained by the above equation and the feature amounts of all the abnormal signals accumulated in the database is calculated according to the following equation (7).

[0022]

[Equation 2]

In the diagnostic computer 5, the degree of coincidence ε _k obtained according to the equation (7) is minimized (most coincident with the newly obtained feature amount of abnormal signal).
The feature amount of the abnormal signal accumulated in 0 is specified. Or
Among the characteristic amounts of the abnormal signal accumulated in the database, a plurality of characteristic amounts of the abnormal signal may be specified in the ascending order of matching (matching).

Then, the diagnostic computer 5 estimates that the cause of occurrence corresponding to the specified characteristic amount of the abnormal signal is the cause of the new abnormal signal. Then, the estimation result is displayed on the display device 6. Further, when the characteristic quantities of a plurality of abnormal signals are specified, the causes of occurrence corresponding to each are displayed on the display device 6.

FIG. 2 is a diagram showing an example of the distribution of the characteristic amount according to the nature of the signal detected by the impact detection device 4. 2 (1) to (4) are waveforms when the simulated impact is given to the reactor equipment with the loose parts monitor installed in the plant. FIG. 2A shows the case where the abnormal signal is a low-frequency shock signal. FIG. 2B shows the case where the abnormal signal is a shock signal having a slightly high frequency. FIG. 2C shows the case where the abnormal signal is a high-frequency shock signal. FIG. 2 (4) shows a case where the abnormal signal is electrical spike noise. As is clear from these figures, the distribution of the characteristic amount according to the property of the signal can be obtained, so that the cause of the abnormal signal can be estimated with high accuracy.

FIG. 3 is a block diagram of a reactor monitoring device (loose parts monitor) according to another embodiment of the present invention. The configuration of FIG. 3 will be described in comparison with FIG. 4 described above. Loose parts monitor system board 2 which is a processing device.
Is equipped with an A / D converter 8 instead of the data recorder 7 of FIG. 4, and the output from the amplifier 3 is input to the A / D converter 8. In addition, the trigger signal line 9 is
It is connected to the / D converter 8. Furthermore, the output of the A / D converter 8 is input to the diagnostic computer 5. The diagnostic computer 5 is connected to the remote diagnostic computer 14 in the diagnostic center 13, which is the center device, via the communication device 11, the communication line 12, and the communication device 11 'in the remote diagnostic center 13. There is. Further, a database 10 'is connected to the remote diagnosis computer 14. In addition, the diagnostic center 13
A plurality of loose parts monitor system boards 2 ′ and 2 ″ installed in another nuclear reactor plant are connected via the communication line 12.

The operation of the loose parts monitor shown in FIG. 3 will be described. Similar to the above-described embodiment, the output signal from the amplifier 3 is input to the impact detection device 4 and
It is input to the / D converter 8. And the A / D converter 8
Converts the input signal into digital data and holds it. On the other hand, the signal input to the impact detection device 4 is
If it is determined to be an abnormal signal based on the above-described identification principle in that, a trigger signal is sent from the impact detection device 4 to the A / D converter 8 via the trigger signal line 9.

When the A / D converter 8 receives the trigger signal, it sends digital data corresponding to the abnormal signal to the diagnostic computer 5. Diagnostic computer 5 is A / D
The above equations (1) to (6) are executed on the digital data of the abnormal signal sent from the converter 8 (that is, the digitized waveform data of the abnormal signal) to obtain six feature values r.
Calculate 1, r2, r3, α, CF, RMS. And
The diagnostic computer 5 reports the result to the communication device 11, the communication line 12, and the communication device 11 ′ in the diagnostic center 13.
The data of the obtained characteristic amount is transmitted to the remote diagnosis computer 14 via the, and the remote diagnosis computer 14 stores them in the database 10 ′. Therefore, in the database 10 ', information on abnormal signals detected in the past in all loose parts monitors connected by the communication line 12 (that is, feature amount and corresponding cause data).
Is accumulated.

When the remote diagnosis computer 14 receives the characteristic amount, it reads the past data accumulated in the database 10 '. Then, the degree of coincidence between the received characteristic amount and the characteristic amounts of all the abnormal signals accumulated in the database 10 ′ is calculated according to the above equation (7).

Then, the remote diagnosis computer 14 stores the abnormal signal accumulated in the database 10 in which the degree of coincidence obtained according to the equation (7) is the smallest (which is the closest to the characteristic amount of the newly obtained abnormal signal). Specify the feature amount of. Alternatively, among the feature quantities of the abnormal signal accumulated in the database, a plurality of feature quantities of the abnormal signal may be specified in the order of the smallest matching degree (the best matching).

Then, the remote diagnosis computer 14 estimates that the cause of occurrence corresponding to the specified characteristic amount of the abnormal signal is the cause of the new abnormal signal. Then, the remote diagnosis computer 14 transmits the estimation result of the occurrence cause to the diagnosis computer 5 via the communication device 11 ′, the communication line 12, and the communication device 11. When receiving the estimation result, the diagnostic computer 5 displays it on the display device 6. In addition, when the characteristic quantities of a plurality of abnormal signals are specified, the causes of occurrence corresponding to each are transmitted and displayed on the display device 6.

In this embodiment, the database 1
Since information of abnormal signals detected by loose parts monitors in other nuclear reactor plants is also stored in 0 ', a large amount of information is stored in the database 10',
Since information on anomalies that have occurred in other nuclear reactor plants can be immediately used, more accurate estimation can be performed.

Further, the function of estimating the cause of occurrence of the diagnostic computer and the database storing the information of the abnormal signal are shared by using communication, so that each plant is
It is not necessary to provide these, and the cost can be reduced.

There is no limit to the number of loose parts monitor system boards 2 connected to the diagnostic center 13. The communication line 12 may be a public line or a dedicated line, and includes both wired and wireless cases.

Further, the database 10 in the embodiment shown in FIG. 3 may store information on abnormal signals generated in other nuclear reactor plants. In this case, the database 10
Unlike in the case of Fig. 5, it is not possible to obtain information on abnormal signals in other reactor plants in real time by communication, but by updating the database regularly, information on abnormal signals in other reactor plants can be obtained. Can be accumulated.

[0036]

As described above, the reactor monitoring apparatus (loose parts monitor) of the present invention has a detecting unit for detecting an abnormal signal generated by an abnormality of the reactor, and a predetermined characteristic amount of the abnormal signal detected by the detecting unit. A storage unit that stores a predetermined feature amount of an abnormal signal that has occurred in the past and data of a cause of occurrence corresponding to the feature amount, and a feature amount that has been calculated by the calculation unit and the storage unit. And an estimation unit that estimates the cause of the abnormal signal detected by the detection unit based on the comparison with the generated feature amount.
In this way, the reactor monitoring device (loose parts monitor)
Estimates the cause of the detected abnormal signal. Therefore, in order to estimate the cause of the abnormal signal generation, signal processing by a skilled person or examination by a plant engineer becomes unnecessary,
When an abnormality occurs, it is possible to quickly respond to the abnormality, and it is possible to reduce the cost required for estimating the cause of the abnormality.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a reactor monitoring device (loose parts monitor) according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a distribution of feature quantities according to the nature of a signal detected by an impact detection device 4.

FIG. 3 is a block configuration diagram of a reactor monitoring device (loose parts monitor) according to another embodiment of the present invention.

FIG. 4 Conventional reactor monitoring device (loose parts monitor)
It is a block configuration diagram of.

FIG. 5 is a diagram for explaining an example of a signal processing procedure in the conventional impact detection device 4.

[Explanation of symbols]

1 accelerometer 2 Loose parts monitoring system board 3 amplifier 4 Impact detection device 5 Diagnostic computer 6 Display device 7 Data recorder 8 A / D converter 9 Trigger signal line 10 database 11 Communication device 12 communication lines 13 Diagnostic center 14 Remote diagnosis computer

─────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A 5-126980 (JP, A) JP-A 6-331507 (JP, A) JP-A 11-39023 (JP, A) JP-A 2- 159526 (JP, A) JP 56-16821 (JP, A) JP 60-104208 (JP, A) JP 60-183591 (JP, A) JP 2000-19286 (JP, A) 58) Fields investigated (Int.Cl. ⁷ , DB name) G21C 17/04

Claims (4)

(57) [Claims] 1. A reactor monitoring device for monitoring a reactor provided in a plant, comprising: a detector for detecting an abnormality signal generated by an abnormality of the reactor; and a predetermined abnormality signal detected by the detector. A calculation unit that calculates a feature amount, a predetermined feature amount of an abnormal signal that has occurred in the past, a storage unit that stores data of a cause of occurrence corresponding to the feature amount, and a feature amount that is calculated by the calculation unit. An estimation unit that estimates the cause of the abnormal signal detected by the detection unit based on comparison with the characteristic amount stored in the storage unit , wherein the predetermined characteristic amount is a first moment of the abnormal signal. radius
A reactor monitoring device including r1 . 2. The storage unit according to claim 1, further comprising a feature amount of an abnormal signal that has occurred in another reactor in the past and data of a cause of occurrence corresponding to the feature amount. Reactor monitoring device. 3. The processing device according to claim 2, wherein the processing device including the detection unit and the calculation unit is arranged for each of a plurality of plants, and the center device including the storage unit and the estimation unit includes a plurality of plants. And each processing device and the center device are connected by a communication line, and the characteristic amount calculated by the calculation unit of the one processing device is transmitted from the one processing device to the center device. The reactor monitoring device, wherein the estimation result by the estimating unit is transmitted from the center device to one processing device. 4. The method according to claim 1, wherein the predetermined feature amount is further a secondary motion of the abnormal signal.
And a third moment radius r3 .