JPH02236447A - Counting means for ae signal in ae measuring method - Google Patents

Abstract

PURPOSE:To always count the frequency in generation of an acoustic emission (AE) signal in a stable condition by detecting the envelope of the AE signal to detect the lowest voltage level of the envelope detection signal. CONSTITUTION:The signal outputted from an envelope detecting circuit 4 and a circuit signal are converted by an A/D converter, and signals are transferred to a personal computer 7 through a communication means. The personal computer 7 calculates the value of a lowest voltage level Vmin of the inputted detection signal and uses a value Vth obtained by Vth=k.Vmin ((k) is a threshold voltage coefficient and is a constant to make the value of the threshold voltage Vth slightly higher than a background noise Vbn of an envelope detection signal 10) as the threshold voltage to calculate the frequency in generation of the taken-in AE detection signal. The frequency in generation is counted by an AE signal counting part 5 in accordance with this method; and when the AE signal is generated more than a set value, a message of the occurrence of abnormality is sent from the personal computer 7 to a display part 8.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a means and apparatus for counting AE signals generated from a subject.

(Prior art) Acoustic emission (hereinafter referred to as A
It is known that a signal (abbreviated as E) is generated. Therefore, an AE sensor is attached to the subject to be inspected, and the AE
A method is used to count the number of signals received by a sensor and determine whether an abnormality has occurred. By the way, the above-mentioned A
As an E signal counting means, for example, Japanese Patent Application Laid-Open No. 1109-1983
There is No. 09. In this embodiment, the AE signal generated due to the wear of the sliding member is input to a comparison circuit, and only the signal exceeding a set threshold voltage is converted into pulses, and the pulses are counted. There is.

Furthermore, the above-mentioned AE counting means may not be able to accurately count the number of AE signals generated. That is, this is a case where AE signals generated from a subject whose background noise (hereinafter abbreviated as BGN) constantly increases or decreases are counted. If the threshold voltage is kept constant, for example, BGN increases and exceeds the threshold voltage. Or, if BGN becomes smaller, BG
The threshold voltage level is higher than that of N, and there are problems such as the inability to detect low-level AE signals.

Therefore, in order to solve the above-mentioned problem,
No. 24 has been proposed. That is, this is a method of automatically changing the threshold voltage level according to the level of BGM generated from the rotating machine. In this proposal, the number of revolutions of the rotating machine is detected and the threshold voltage is automatically changed. However, the method generally used as a means for counting AE signals at present is as follows. A commonly used method is to obtain the average value of the AE signal and add a DC voltage to it to obtain a threshold voltage.

[Problem to be solved by the invention]

Assume that the above-mentioned conventional technique is used to diagnose the sliding part of a rotary compressor used in a room air conditioner as a representative example of a sliding motion device.

There are more than 100 types of rotary compressors used in room air conditioners, and the BGN of the AE signal generated differs depending on the model. Furthermore, even if the compressor is the same, the BGN level will differ depending on the operating conditions (rotation speed, compression ratio of fluorocarbon gas, etc.) during inspection. Therefore, with the method proposed in the above-mentioned Japanese Patent Laid-Open No. 55-110909, it is impossible to accurately detect the AE signal generated by an abnormality in the sliding part. As a possible solution to the above, a method can be considered in which a DC voltage is added to the average value of the AE signal in the embodiment described above, and this is used as the threshold voltage. However, in the case of a rotary compressor, even if it is a normal product, it is always synchronized with rotational movement, and one AE signal (hereinafter abbreviated as stationary AE wave) is generated per rotation. Furthermore, the peak value of the stationary AE wave varies greatly depending on the model or operating conditions, similar to the BGN.

In other words, in the case of a compressor with a low level of steady AE waves, the threshold voltage will be a low level, but if the level of steady AE waves is high, the threshold voltage will naturally be at a high level due to the influence of the steady AE waves. Put it away.

As a result, even if the AE signals of the same wave height level are generated by sliding part defects of the same size, in the case of a compressor with low steady AE waves, this AE signal can be counted, but the steady AE wave level is In a high-performance compressor, the threshold voltage is high, resulting in a problem in which the AE signal cannot be counted.

This also causes a drop in the detection accuracy of the AE signal generated due to damage to the sliding part.

In order to solve the above-mentioned problems, it is possible to solve them by manually setting the best threshold voltage for each compressor according to its operating conditions and model. Also, as another means,
All BGN levels and steady AE wave levels for compressor models and operating conditions are stored in the computer (approximately 3 million or more data), and each time the computer stores the threshold voltage level that matches the operating conditions of each compressor. A solution is possible by using a method that allows selection.

However, when considering the case of diagnosing the sliding parts of a compressor, the above solution is not realistically possible. As an example, consider diagnosis of sliding parts in a manufacturing line where different types of compressors flow at a tact time of 1 unit/10 seconds. 10 humans
Adjusting the threshold voltage within seconds is too late.

Furthermore, another method of selecting a threshold voltage using a computer is considered to be impractical in view of the amount of data.

An object of the present invention is to provide a means and apparatus that can always count the number of AE signals generated under stable conditions, regardless of the form in which the AE signals are generated from a subject.

[Means to solve the problem]

The above purpose is to perform envelope detection of the AE signal, detect the lowest voltage level of the envelope detection compensation, and count the number of generated AE signals by setting a voltage value several times the lowest voltage level as a threshold voltage. It can be achieved.

[Effect]

The minimum voltage level value of the envelope detection signal mentioned above changes almost in proportion to the fluctuation of the BGM. Further, as described above, the average value level is affected by changes in the peak value level of the steady waveform, whereas the minimum voltage level value is hardly affected by changes in the peak value of the steady waveform.

As described above, the present invention performs envelope detection of the AE signal,
If the lowest voltage level is used as the reference and the threshold voltage, no matter what form of AE signal, A.
The invention is characterized by the ability to count the number of E signals generated.

〔Example〕

Examples of the present invention will be described below with reference to the drawings. In this embodiment, a rotary compressor used in a room air conditioner will be used as an example of a sliding motion device.

FIG. 1 is an overall configuration diagram of an apparatus for counting the number of generated AE signals according to an embodiment of the present invention. An AE sensor 2 is attached to the rotary compressor 1, and the output of the AE sensor 2 is amplified by an amplifier 3, detected by an envelope detection circuit 4, and then input to an AE signal counter 5. On the other hand, the rotation signal of the compressor is also detected by the rotation signal detection sensor 9, and is similarly input to the AE signal counting section 5.

Although the AE signal counting section 5 can be constructed using an analog circuit, a digital system will be described here.

The AE signal counting section 5 is composed of an A/D converter 6 and a personal computer 7. That is, the signal output from the envelope detection circuit 4 and the rotation signal are transmitted to the A/D converter 6.
A/D conversion is performed at the . The signal is transferred to the personal computer 7 via communication means such as GPIB. In the personal computer 7, the lowest voltage level V m l of the input detection signals
Compare the n values and use the following formula to find the value Vth
The number of occurrences of captured AE detection signals is calculated using the threshold voltage as the threshold voltage.

V th = k-V-i- -
(1) k in the above equation is the threshold voltage coefficient, and the second
As shown in the figure, the value of Vih is the background noise of the envelope detection signal 10. This is a constant to set the value to a slightly higher value. Next, the number of occurrences is counted in the AE signal counting section 5 using the above method, and if an AE signal exceeding the set value is generated, a message indicating that an abnormality has occurred is sent from the personal computer 7 to the display section 8. It will be done.

Next, we will take the AE signal generated from the rotary compressor.
The effects of the above invention example will be specifically described in comparison with the conventional example. FIG. 3 shows the envelope detection waveform of the AE signal generated from the rotary compressor. As shown in the figure, in a rotary compressor, a steady AE wave 2 is generated once per rotation.
0 occurs.

The generation mechanism of this stationary AE wave will be described below.

FIG. 4 shows the structure of a rotary compressor. Motor 30, crankshaft 31, upper bearing 32, lower bearing 33, cylinder 34. It is composed of main parts such as rollers 35. FIG. 5 shows a cross section taken along line AA' in FIG. Compression chamber (Pa) and suction chamber (Ps)
The vane 36 provided to separate the
), the pressure condition is P
Since a > P S, the tilting motion will occur in the direction shown in the figure. However, as soon as the compression stroke ends and the fluorocarbon gas suction stroke begins, the pressure condition changes to P a <
As shown in the same figure (b), the vane 36
collides with the vane slot 37 at a rapid speed. The energy of this collision is the cause of generating stationary AE waves.

As described above, if an AE signal is generated even in a normal product, the following problem will occur. FIG. 6 shows a method for counting AE signals when the average value voltage is used as a reference and a voltage value several times the average value is used as the threshold voltage Vme. The envelope detection waveform shown in Figure 6 is an example of a waveform that occurs when the sliding parts of a rotary compressor are damaged. The abnormal AE wave 50 is generated due to damage to the other sliding parts of the steady AE wave 20, and the value of Vme is lower than the wave height level of the abnormal AE wave 50 when the steady AE wave level is small in FIG.
It is possible to count E waves 5o.

However, as shown in the same figure (b), when the wave height level of the stationary AE wave 20 becomes large, V naturally increases due to its influence.
The value of me becomes high. As a result, Figure 6(a)
This means that abnormal AE waves 50 having the same wave height level as the abnormal AE waves 50 shown in Figure 1 cannot be counted. In this regard, the AE signal counting method according to the present invention uses the lowest voltage level V wa t n of the envelope detection signal as a reference and determines its threshold voltage vth using the above-mentioned equation (1). As the values of Vth are shown in a) and (b),
The abnormal AE waves 50 can be counted without being affected by the steady AE waves 20.

FIG. 7 is an example of a flowchart when diagnosing the sliding part of a compressor, which is one sliding motion device, using the above-mentioned AE signal counting method. From the envelope detection signal transferred to the personal computer 7 shown in FIG. 1, a signal corresponding to one rotation period is extracted by the rotation signal transferred at the same time. Next, the lowest voltage level V min of the extracted signals is calculated, and the calculation of the above equation (1) is performed to calculate the determined threshold voltage v
The number N of AE signal occurrences is counted by th. Next, the number N of AE occurrences is compared with a set value Kt, and if it exceeds Ks, it is determined that there is an abnormality in the sliding part, and this is communicated to the outside through the display section 8 shown in FIG. Note that although the above embodiment has been explained limited to one rotation period, in particular
There is no need to limit the number of AE signals generated per period, and the accuracy improves as the signal acquisition period increases.

Next, other embodiments will be described. The peak value of the AE signal caused by cracks in the material, scratches on the sliding parts, etc.
The greater the degree of abnormality, the higher it becomes. That is, if the peak value can be detected, it is possible to estimate the degree of abnormality.

In the present invention, by providing several stages of threshold voltages as described above, it is possible to count the number of occurrences of A E (g) and also detect its peak value. This is an example of a flowchart for diagnosing the sliding parts of a rotary compressor, which is one of the devices.Similar to the embodiment shown in FIG. 7, the minimum voltage level V m t n of the envelope detection signal is calculated. , Next, set the threshold voltage coefficient k shown in the above-mentioned equation (1) in several stages, and calculate the number of signals generated in the envelope detection signal at each threshold voltage V th t tVthx, Vtha. Count.

Next, the average value N of the counted number of AE occurrences Nl, NZ, Na
．． A set value Ktt. is determined to determine the degree of abnormality. K▲2
The degree of abnormality in the compressor sliding parts is determined. FIG. 9 is an example of the envelope detection waveform of the AE signal generated from the compressor. N.C. in the abnormal state shown in Fig. 3(a), which is relatively mild in severity. is approximately 1.7, but the abnormality progresses and the figure (
It can be seen that N1 increases to 3 when reaching the dangerous state b).

As mentioned above, the lowest voltage level V of the envelope detection signal
m t n as a reference, A at multiple threshold voltages
By counting the number of E signals generated, it is possible to determine the degree of abnormality in the sliding parts of the rotary compressor.

In the above-mentioned embodiment, an example was described in which the threshold voltage was set in three stages, but if a higher threshold voltage is set, it is possible to judge the degree of abnormality with even higher accuracy. It is. In addition, in the above embodiment, as a method of calculating N1, A. Although the average value of the number of E signal occurrences was used, the total value may also be considered. Furthermore, as in the embodiment described in FIG. 7, it is not necessary to limit the AE signal to one period of rotation.
For convenience of explanation, all of the above-mentioned embodiments are described using a rotary compressor as an example, but the invention is not limited to rotary compressors only, and an AE signal is always generated even in a normal product. In the case of an object to be examined, the present invention is highly effective in counting the number of AE signals generated.

The invention described below further improves the accuracy of counting the number of AE signals generated in the embodiment described above. For convenience of explanation, a rotary compressor, which is one type of sliding motion device, will be taken as an example.

In sliding motion devices such as rotary compressors, acoustic signals are always generated as a result of sliding motion. This is caused by the minute contact between sliding parts and the noise of fluids such as fluorocarbon gas.
AE signals generated due to abnormalities are called noise. For example, as shown in FIG. 10, when these sliding noises 60 are large, it may be difficult to distinguish them from abnormal AE waves caused by damage to the sliding parts.

When counting the number of AE signals generated in the above state, in the case of the waveform shown in the figure, the number is 7 per cycle, which results in an error in which the product is determined to be abnormal, regardless of whether it is a normal product.

The present invention has been devised to eliminate the above-mentioned errors. As a result of experiments, it was found that when looking at the generation phase of the abnormal AE wave 50 for one rotation, it always occurs at the same position, whereas the position of the sliding noise 60 is random. That is, the abnormal AE wave is generated at the location of the defect, and the location of the sliding noise changes depending on the operating state each time.

This embodiment was invented in consideration of the characteristics of the above-mentioned sliding noise and abnormal AE waves. FIG. 11 is a specific flowchart for carrying out the invention. As shown in the figure, first, the rotation signal is used as a trigger signal, and the envelope detection waveform of the AE signal is subjected to averaging processing. FIG. 12 shows an average processing waveform 70 after performing the above processing. As shown in the figure, random sliding noise disappears. If the abnormal AE wave 50 is included, it can naturally be extracted clearly as shown by the broken line in the figure.

Next, a signal for one period is extracted from the average processing waveform, and the lowest voltage level V m is obtained from the next extracted signal.
s n is calculated. Furthermore, by executing the calculation of equation (1) above, and using the obtained threshold voltage ■, , , AE
The number N of signal occurrences is counted. Next, by comparing the set value Ki and the above-mentioned N, a determination is made as to whether it is normal or abnormal.

Note that after the calculation process of the lowest voltage level V m s n shown in FIG. 11, as shown in FIG. By counting, the AE signals can be counted with higher precision than the embodiment shown in FIG. 8, and the degree of abnormality can also be determined.

Further, in the above-mentioned embodiment, the explanation was given for one period of rotation, but the counting accuracy of the AE signal improves as the period increases, so it is not particularly necessary to limit it to one period. Furthermore, although the rotation signal was used as the 1.rigger signal used in the averaging process in the above embodiment, any signal may be used as long as it is synchronized or synchronized with the driving of the subject.

As mentioned above, the envelope detection signal is subjected to averaging processing,
Calculate the lowest voltage level V m l n of the waveform,
If the AE signals are counted based on this, it becomes possible to detect only the AE 4M signal generated due to an abnormality with high precision without being affected by other noises.

〔Effect of the invention〕

According to the present invention, in a test object in which an AE signal (stationary AE signal) is always generated even in a normal product, counting the AE signal generated due to an abnormality is not affected by the stationary AE signal. Abnormal AE signals can be counted with high precision without any problems.

Therefore, it has an extremely large industrial effect, such as making it possible to diagnose abnormalities in the above-mentioned specimens, which was difficult in the past.

[Brief explanation of drawings]

FIG. 1 is a diagram showing one embodiment of the present invention, FIGS. 2 to 6 are explanatory diagrams for explaining the embodiment shown in FIG. 1, and FIG.
The figure is a flowchart for explaining the operation of FIG. 1, FIG. 8 is a flowchart for explaining another embodiment of the present invention, and FIG. 9 is an explanation for explaining the embodiment shown in FIG. 8. 11 are flowcharts for explaining another embodiment of the present invention, and FIGS. 10 and 12 are explanatory diagrams for explaining the embodiment shown in FIG. 1... Rotary compressor, 2... AE sensor, 3... Amplification section,
4... Envelope detection circuit, 5... AE signal counting section, 6
...A/D converter, 7...PC, 8...
Display unit, 9... Rotation signal detection sensor, 10... Envelope detection signal, 20... Steady AE wave, 50... Abnormal A
E wave, 60... Sliding noise, 70... Addition average processing waveform. Diagram! Ward 3 Figure 4 Figure 5 Cause +L:Ll Force f, Published 11th Takahiro! (The E control sheet is printed 10 and the evening N iteIF! Fig. 6 fbl E U AE wave size Fig. 8 Fig. 7 Fig. 9 α) congenital disorder

Claims (1)

[Claims] 1. In a device for counting AE signals generated from a subject, the AE signal generated from the subject is envelope-detected, the lowest voltage level in the envelope detected signal is detected, and the lowest voltage level is detected. AE signal counting means in an AE measurement method, characterized in that a value obtained by superimposing a voltage set on a voltage level is used as a threshold voltage, and the number of occurrences of the AE signal is counted. 2. In claim 1, the threshold voltage is provided in several stages, and the average value of the number of AE signal occurrences counted at each threshold voltage is taken as the number of AE signal occurrences. A
AE signal counting means in the E measurement method. 3. An AE signal counting means in an AE measurement method, characterized in that the average value in claim 2 is taken as a total value. 4. In a device that counts AE signals generated from a subject, the AE signal generated from the subject is envelope-detected, and a signal that is synchronized or synchronized with the movement of the subject is used as a trigger signal to calculate the envelope detection signal. is averaged, the lowest voltage level in the waveform after the averaging process is detected, and the value obtained by superimposing the set voltage on the lowest voltage level is set as the threshold voltage, and the number of occurrences of the AE signal is counted. AE signal counting means in an AE measurement method, characterized in that: 5. The threshold voltage as set forth in claim 4 is provided in several stages, and the average value of the number of AE signal occurrences counted at each threshold voltage is taken as the number of AE signal occurrences. A
AE signal counting means in the E measurement method. 6. An AE signal counting means in an AE measurement method, characterized in that the average value in claim 5 is taken as a total value. 7. The number of AE signals generated is determined from claims 1 to 6.
A method of counting using a term counting method. 8. An abnormal state of the sliding part is diagnosed by comparing the number of AE signal occurrences counted by the AE signal counting means in claim 1 or 4 with a reference value. Abnormality diagnosis device for sliding motion equipment. 9. By comparing the number of occurrences of AE signals counted by the AE signal counting means in Claims 2, 3, 5, and 6 with a plurality of reference values, An abnormality diagnosing device for a sliding motion device, characterized by diagnosing the degree of abnormality in a sliding motion device. 10. A method for diagnosing an abnormality in a sliding part of a sliding motion device according to claim 8 or 9.