JPH036451B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method of detecting the position of a fluid leak from a change in its frequency spectrum pattern by utilizing the phenomenon of sound generation when a fluid leaks.

[Prior Art] One of the methods for detecting fluid leakage from a fluid distribution system is to detect a leakage signal using an AE (Acoustic Emission) sensor by utilizing the sound generation phenomenon at the time of fluid leakage. For example, in pressure tube reactors, in order to detect coolant leakage due to defects in the inlet pipes leading to each pressure tube, AE sensors are installed on piping, etc., which are structures that need to be monitored, to prevent leakage. A system configuration is adopted that monitors the accompanying sound emissions.

Conventional detection methods detect the sound pressure level of the sound emitted due to leakage, and issue an alarm when the sound pressure level becomes abnormally high. Of course, in fluid circulation systems, etc., there are also environmental noise components such as active sound sources such as circulation pumps and the sound of fluid flowing, so accurate detection cannot be made only by the magnitude of the detected sound pressure level itself. In consideration of these factors, a method is adopted in which the influence of the above-mentioned environmental noise components is subtracted, and the presence or absence of leakage is detected from the magnitude of the subtracted signal level. Such a conventional technique is described in, for example, Japanese Patent Laid-Open No. 146684/1984.

[Problems to be Solved by the Invention] However, in this conventional technology, an abnormality is basically detected at a monitoring point based on the magnitude of the sound pressure level of the sound emitted due to leakage, as described above. Therefore, even if the presence or absence of fluid leakage can be detected, the location of the leakage cannot be specified. The sound pressure level is
This is because it changes depending on the size of the discharge hole that has occurred, the fluid leakage situation, the distance from the leakage location to the monitoring point (AE sensor installation location), etc.

For this reason, the prior art has the disadvantage that if a pipe breaks at a location that cannot be easily observed, the location of the leak cannot be determined unless all pipes are inspected, and it takes time to respond to the accident. Particularly when the structure to be monitored is the cooling system of a nuclear reactor, etc., it is strongly desired to be able to detect the occurrence of a leak and pinpoint the location of the leak at the same time from the perspective of minimizing radiation exposure.

An object of the present invention is to eliminate the drawbacks of the prior art as described above, and to reduce the distance from the mounting point of the AE sensor to the fluid leak location by performing some kind of signal processing on the signal from the AE sensor. It can be easily and quickly determined, and therefore, when applied to the cooling system piping of a nuclear reactor, for example, workers can immediately know the location of a leak without entering the containment vessel. It is an object of the present invention to provide a fluid leak position detection method that can significantly reduce radiation exposure of workers and enable quick and efficient response to accidents.

[Means for Solving the Problems] The present invention, which can achieve the above objects, is characterized in that the frequency spectrum of the sound emitted due to leakage changes depending on the distance from the leakage occurrence position to the AE sensor installation point. By taking advantage of this fact, we have devised a way to quickly and accurately extract position data without using complex devices such as frequency analyzers that require long processing times.

In the present invention, first, for various distances from the AE sensor to the leakage position, a set of data indicating the ratio of the sound pressure effective value of each divided frequency band to the normalized sound pressure effective value of all frequency bands is prepared in advance as reference data. Look for it as a table. Then, the signal from the AE sensor attached to the structure to be monitored is divided into multiple frequency bands using a filter in the same way as above, and the effective sound pressure value in each band is determined, and the normalized sound pressure value in all frequency bands is calculated. A set of data indicating the ratio of the sound pressure effective value of each frequency band divided to the pressure effective value is obtained. By comparing this data set with the reference data table, the distance from the mounting position of the AE sensor to the leakage position is determined and the fluid leakage position is detected.

When high-temperature, high-pressure fluid such as coolant leaks from piping, etc., ultrasonic waves are generated and propagate through structures such as piping. The ultrasonic waves used at this time are usually in a frequency band of several MHz or less. Therefore, in general, 0.01~
The method of the present invention as described above is carried out by specifying in advance about four desired bands in a frequency band of about 1 MHz. For example, 50~100kHz, 100~200kHz,
This is implemented by specifying four bands in advance: 200 to 300 kHz and 300 to 1000 kHz.

[Function] The effective value of the sound pressure of the emitted sound caused by leakage is attenuated as it propagates through piping, etc., and the rate of attenuation is more significant in the higher frequency range. In other words, the detected leakage sound spectrum changes depending on the distance from the leakage position to the mounting position of the AE sensor. Therefore, the total frequency band is divided into multiple bands, the effective sound pressure value is normalized over the entire frequency band, and the ratio of the effective sound pressure value of each divided frequency band is calculated. , if the correlation with the distance from the AE sensor to the leak location is actually measured and determined in advance, it is possible to detect leak occurrence by comparing the ratio of the effective sound pressure value for each frequency band with the above correlation at the time of leak detection. The distance to the location can be detected.

[Example] Hereinafter, the present invention will be explained in more detail based on the drawings. FIG. 1 is a block diagram showing an example of a device suitable for implementing the method of the present invention, and FIG. 2 is an explanatory diagram showing the mounting position of the AE sensor. A large number of AE sensors 1a, 1b, . . . are attached to a structure to be monitored (for example, piping of a fluid circulation system, etc.). In this example, as shown in FIG.
A waveguide rod 13 is connected to the connection portions b, . . . , and AE sensors 1a, 1b, . The main piping 2 has a circulation pump 14
etc. are installed, and they become a noise source for each AE sensor. The AE sensors 1a,..., 1n are
Since the leakage signal that is normally input is a signal in the frequency band of about 0.01 to 1 MHz, any type of device that can convert ultrasonic vibrations in that frequency band into electrical signals can be used, and usually piezoelectric elements are used frequently.

For example, assume that a leakage point 15 occurs in the inlet pipe 12a. The leakage signal detected by the AE sensor 1a is
It is amplified by a preamplifier 2a and a main amplifier 3a. These amplified signals are passed through bandpass filters 4aa,..., 4ak and effective voltmeters 5aa, 4ak, respectively.
..., 5ak, and A/D (analog-digital) converters 6aa, ..., 6ak to an arithmetic unit 7 consisting of a minicomputer or the like. Here, the bandpass filter 4aa passes the detected signal over almost the entire frequency band of the AE signal from 0.01 to 1MHz, and the obtained signal has an effective value of 5.
A/D converter 6aa converts it into an effective voltage value.
After the signal is converted into a digital signal, it is input to the arithmetic unit 7, whereby the normalized effective sound pressure value in the entire frequency band is calculated.

Other bandpass filters 4ab,..., 4ak are:
Each of the above-mentioned total frequency bands is divided into (k-1) different frequency bands, and the effective sound pressure value in each band is an effective value total of 5.
ab, . . . , 5ak, and the resulting voltages are converted into digital signals by A/D converters 6ab, . As mentioned above, since it is usually good to divide the entire frequency band into about 4, the value of k is selected to be about 5. If the number of divisions is too small, the detection accuracy of distance data will decrease, which is undesirable. If the number of divisions is too large, it is expected that the detection accuracy will improve, but the device configuration will become significantly complicated and the cost will increase, which is not very preferable.

Now, the calculator 7 calculates whether there is a difference between the past value and the current value for each frequency band, and if a significant difference occurs in a certain frequency band, it can be determined that leakage has occurred. can.

At the same time, the calculator 7 calculates the ratio of the effective sound pressure values in each frequency band to the normalized effective sound pressure values in all frequency bands calculated as described above. Calculated by. For example, FIGS. 3A, B, and C are graphs showing power spectra of leakage signals when leakage occurs. Here, the vertical axis is normalized as (power)/(maximum value of power). The figure A is
This is the case where the AE sensor is located very close to the leakage position, and Figure B shows the case where the AE sensor is located 5m away from the leakage position, and Figure C shows the case where the AE sensor is located 15m away from the leakage position. As is clear from comparing the figures, the effective value of the sound pressure of the leaked sound is attenuated as it propagates through pipes, etc., and the degree of attenuation is more significant in the higher frequency range. Therefore, the power spectrum changes depending on the distance from the leakage position to the AE sensor.

Therefore, we experimentally varied the distance from the AE sensor to the leakage point, and for each distance, we obtained data showing the ratio of the sound pressure effective value in each frequency band divided to the normalized sound pressure effective value in all frequency bands. A set of data is obtained in advance and stored in the arithmetic unit 7 as a reference data table.

Since the AE sensors 1a,..., 1n are usually installed at separate locations as shown in Fig. 2, leakage occurs from one AE sensor (1a in Fig. 2) to the other AE sensors 1b,..., 1n. It will be far away from point 15.
Therefore, at the same time as the leakage occurs, the AE sensor other than AE sensor 1a
A leakage signal is also detected by the sensor 1b, etc., but as described above, the level is lower than that of the leakage signal from the AE sensor 1a, and the spectrum should be on the lower frequency side.

The leakage signal from the AE sensor 1a is frequency-divided by bandpass filters 4aa,..., 4ak,
Effective value meter 5aa,..., 5ak, A/D converter 6aa,
. . , 6ak, the signal enters a computing unit 7 consisting of a minicomputer or the like, and a set of data indicating the ratio of the effective sound pressure value of each divided frequency band to the normalized effective sound pressure value of the entire frequency band is determined. The same applies to the AE sensors 1b, . . . , 1n. In this way, the computing unit 7 usually first selects the AE sensor exhibiting the spectrum on the highest frequency side from among the leakage signals from the respective AE sensors. In the case of FIG. 2, the AE sensor 1a is selected. In the computing unit 7,
A set of detection data based on the leakage signal from the AE sensor 1a is compared with an already stored reference data table, and the most similar set of reference data is found. This can be done, for example, by selecting the one that minimizes the moment of deviation between the two (the method of least squares). Then, the distance corresponding to that set of reference data is determined by that AE sensor (1a in this case).
This is the distance from the leak point 15. In the case of a simple piping structure as shown in FIG. 2, the leak location can be immediately determined from this distance.

In the above, we looked at the signal from the AE sensor closest to the leakage point, but the distance to the leakage point can be determined by similarly comparing leakage signals from other AE sensors with the reference data table. Therefore, if the transmission path of ultrasonic vibrations due to leakage is complex, the distances from multiple AE sensors can be determined by data processing using the leakage signals of these multiple AE sensors, and the position where they match can be detected as a leakage signal. It can be specified as a location. Since standardized effective values are used as position detection data, accurate detection can be performed without being affected by the size of the leakage hole or the sound pressure level itself due to the distance to the leakage location. be.

Although the configuration and operation of only the AE sensor 1a and the circuits connected to it have been explained here, in reality there are many other AE sensors 1b,
..., 1n are attached, and as shown in the figure, a similar circuit is connected to each AE sensor, and each AE sensor generates a leak and detects the leak position through the same operation. It turns out.

However, the maximum value of the power spectrum decreases as a function of distance. Therefore, the presence or absence of leakage can be detected by averaging the level of the power spectrum, but if the distance is 20 meters or more from the leakage position, the level approaches the noise level, and it becomes difficult to even detect the presence or absence of leakage. However, even in such a state, the spectrum of the detected signal will certainly change, so according to the method of the present invention, it is possible to determine whether there is leakage or not.
It is possible to determine the distance from the AE sensor installation position to the leakage position.

Although a preferred embodiment of the present invention has been described in detail above, it goes without saying that the present invention is not limited to this configuration, and that various modifications are possible. For example, in terms of equipment, if a multiplexer or the like is inserted between the main amplifier and the bandpass filter to switch the signals from each AE sensor in a time-sharing manner, then the bandpass filter, rms meter, and It is also possible to significantly reduce the number of A/D converters installed.

[Effects of the Invention] Since the present invention is a fluid leak position detection method configured as described above, the device configuration is relatively simple, whereas the conventional technology could only detect the presence or absence of a leak. However, since the location of fluid leaks can be detected quickly and accurately, it does not take time to respond to accidents.As a result, it is possible to prevent serious accidents from occurring in the structures that should be monitored, and improve the efficiency of system usage. It can also have excellent effects, such as increasing the safety of workers.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of a device suitable for carrying out the present invention, FIG. 2 is an explanatory diagram showing an example of the installation situation of the AE sensor, and FIGS. 3A, B,
C is a diagram showing an example of a leakage sound spectrum. 1a,...,1n...AE sensor, 4aa,...,4
nk……Band pass filter, 5aa,…, 5nk…
Effective value meter, 6aa,...,6nk...A/D converter,
7... Arithmetic unit, 10... Main piping, 11... Manifold, 12a, 12b... Inlet pipe, 13... Wave guide rod.

Claims (1)

[Claims] 1 In advance, for various distances from the AE sensor to the leak location, obtain a set of data as a reference data table showing the ratio of the sound pressure effective value of each divided frequency band to the normalized sound pressure effective value of all frequency bands. Then, the signal from the AE sensor attached to the structure to be monitored is divided into multiple frequency bands using a filter in the same manner as above, and the effective sound pressure value in each band is determined, and the signal is normalized in all frequency bands. By obtaining a set of data indicating the ratio of the effective sound pressure value of each frequency band divided into the effective sound pressure value obtained in the AE sensor and comparing this data set with the reference data table, A fluid leak position detection method characterized by determining the distance from an installation position to a leak position.