JP2877901B2 - Noise determination method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION The present invention relates to a noise discrimination method, and more particularly, to a method of determining a noise criterion by adding a margin A to an average level of each frequency band for noise generated from a product. The present invention relates to a method of judging this in a device for automatically inspecting by providing and comparing.

[Prior art and its problems] When it is required in quality control to judge whether noise generated from products such as motors, bearings, household appliances, etc. satisfies a certain criterion. There are many.
Such a determination is made empirically by the auditor's hearing, but is generally performed.

However, such empirical judgments tend to change due to individual differences between human inspectors, and even if the same person, the judgment results vary depending on the daily situation, and the judgment is made under a stable constant standard. It is difficult to make a decision.

[Means for Solving the Problems] Accordingly, it is an object of the present invention to provide an automatic noise inspection method capable of performing accurate noise determination based on a certain determination criterion.

In order to achieve this object, according to the present invention, a test box for shutting off external noise and inputting only noise generated from a product to be inspected by a microphone is used. Using a meter, frequency analyzer, computer, and CRT to determine the sound pressure level of the noise generated from the product under test, the noise generated from non-defective products is subjected to frequency analysis to determine the average for each frequency band. Setting a criterion by adding a margin value A to the level; and performing frequency analysis on the noise generated from the product to be inspected by the frequency analyzer, and analyzing the analysis value for each frequency band to the criterion of the corresponding band. The computer determines the number of the analysis values exceeding the criterion and the difference of the analysis value exceeding the criterion from the criterion. And determining the defective product when the number of the analysis values exceeding the determination criterion for each band exceeds the set number, obtained by the comparison, and displaying the determination result on the CRT. And a noise discrimination method comprising:

[Operation] When performing a noise test, instead of performing noise discrimination based on overhaul of auditory noise, discrimination is performed for each frequency band by frequency analysis, and the number of analysis values that exceed the determination standard for each band is determined. When the number exceeds the set number, it is determined that the product is defective, and the state of the defective product is displayed on the CRT, so that simple and highly accurate determination can be performed.

Example An example of the present invention will be described below.

FIG. 1 is a schematic block diagram showing an example of a noise discriminating apparatus to which the present invention can be applied. In an appropriate position in an inspection box 10 in which a product G to be inspected is housed, a power supply to the product to be inspected is provided. A power supply connector 11 for operating the device under test and a microphone 12 that is sensitive to noise generated when the product to be tested operates or the like are attached. The inspection box 10 has a structure that shuts off external noise so as to respond only to the noise generated from the product G to be inspected.

The microphone 12 is connected to a sound level meter 14 that generates a signal indicating the sound pressure detected by the microphone 12, and an output signal indicating the sound pressure level from the sound level meter 14 is a frequency analyzer 16
The frequency analysis of the noise level signal is then provided to the computer 18 for signal processing.

Next, the operation of the present invention will be described with reference to the flowchart of FIG.

First, data is collected for N (about 10 to 20) products known to be non-defective in order to set a non-defective reference value (step 100). First, in order to collect data on the first non-defective product, the non-defective product is stored in an inspection box 10, and the non-defective product is connected to the power supply connector 11 and operated. 12
The sound pressure thus collected is then analyzed by the frequency analyzer 16 (step 102).

FIG. 3 is a graph showing an example of the frequency analysis performed by the frequency analyzer 16. In FIG. 3, a 2000HP compressor is used as a product to be inspected.
It shows the case where the same analysis was performed. In FIG. 3, the lower horizontal axis shows the octave band center frequency for octave analysis on 9 logarithmic scales ranging from 31.5 Hz to 8000 Hz by two times, and the upper horizontal axis shows. The 1/3 octave band center frequency for 1/3 octave analysis is also scaled on a logarithmic scale over 27 bands from 25 Hz to 10 KHz by 2 ^1/3 times, and the vertical axis shows each frequency. The band sound pressure level in the band is graduated in dB. In the graph, 1/3 Examples of octave analysis is shown by the line L _1, each over a 27 band
The sound pressure level for each 1/3 octave band was measured and plotted with a black circle. Further, examples of octave analysis is shown by the line L _2, is plotted by white circles sound pressure level is measured for each octave band across 9 band.

Hereinafter, in the present embodiment, the frequency analysis is performed at 25 Hz to 10 KHz of 27.
The description will proceed with the case of 1/3 octave analysis across the bands.

First good and the noise generated from the product is 1/3 frequency analysis by the frequency analyzer 16, the 1/3 frequency analysis data as shown in the line L ₁ which is plotted with black circles in FIG. 3 can be obtained (Step 102), take out the first good product from the test box 10, and then store the second good product in the test box 10,
A similar 1/3 octave analysis test is performed to collect 27 bands of data. When the 1/3 octave analysis data of 27 bands is collected for all N good products in this way (Y in step 104), the collected data is then averaged for each band to obtain an average band level. Is calculated (step 106).

Here, since the average band level for each band is a decibel value dB, it is obtained as follows. For example, N sound pressure levels in a band with a center frequency of 1000 Hz are represented by S ₁ (d
B), S ₂ (dB), ... Sn (dB) ... S _N (dB)
To explain the case where these sound pressure levels are averaged, Sn = 10 · log (Mn / Mo) (where Mn is the noise energy and Mo is the reference value), so each actual sound level is Mn = Mo × 10 ^{Sn / 10} , and therefore N in the band of center frequency 1000 Hz
The average level of the actual measured sound Smean (1000) Becomes The average level of the other bands is determined in the same manner, and thus the average level of each band over 27 bands Smean (25), Smean (31.5), Smean (40),.
.., Smean (1000),..., Smean (10,000) are calculated (step 106), and then a certain margin value A is added to each average level to determine judgment criteria R ₁ , R ₂ ,. · · R ₂₇ is determined (Suppu 108). The margin value A is set to about 3 to 4 dB in consideration of the environmental state and the situation at that time. That is, for example, the determination reference value in the band of the center frequency 1000 Hz, that is, the determination reference value R ₁₇ (1000) of the _17th band counted from the band of the center frequency 25 Hz is R ₁₇ (1000) = Sm
ean (1000) + A, and a criterion value R in a band having a center frequency of 400 Hz, for example, the 13th band.
₁₃ (400) is represented by R ₁₃ (400) = Smean (400) + A.

If the determination criteria for each band are determined in this way (step 108), then the actual inspection of the product to be inspected is performed.

In order to perform the inspection, the product to be inspected is stored in the inspection box 10 and data is collected from the sound level meter 14 as described above (step 11).
0), 1/3 octave analysis over 27 bands is performed (step 112). When sound pressure level data D ₁ , D ₂ ,... D ₂₇ of 27 bands are determined by 1/3 octave analysis, the data D _{1 to} D ₂₇ are respectively determined for each band determined previously. criterion R ₁ to R ₂₇ and being compared each data to determine whether there within criterion (step 114), if there is data exceeding the criterion, the data exceeds the criterion The number of bands and the difference between the exceeded data and the criterion are determined (step 116), and it is determined whether the product to be inspected is a non-defective product or a defective product based on the determined number and difference of the exceeded bands. Processing is performed (step 118).

In the present embodiment, the non-defective or defective judgment processing is performed as described below.

First, if all of the data D _{1 to} D ₂₇ of the 27 bands are smaller than the corresponding determination criteria R _{1 to} R _{27, the} product to be inspected is, of course, determined to be non-defective.

Further, 27 among bands of data D ₁ to D ₂₇ single, if eight or more data beyond the corresponding criterion is determined as defective.

If any of the 27 data exceeds the corresponding determination criterion between 1 and 7, the non-defective or defective product is determined based on the table in FIG.

In FIG. 4, the abscissa indicates the total number of data exceeding the criterion, and the ordinate indicates the difference of the data exceeding the criterion from the criterion. The values shown are shown. In the figure, black circles indicate defective products, white circles indicate non-defective products, and triangles indicate cases where neither can be said. For example, center frequency 40
Only two data, the data D ₁₃ (400) of the band of 0 Hz and the data D ₁₇ (1000) of the band of the center frequency 1000 Hz,
Each exceeds the corresponding criterion R ₁₃ (400) and R ₁₇ (1000), and the exceeded value is, for example, D ₁₃ (400) −R ₁₃ (400) = 2 dB D ₁₇ (1000) −R ₁₇ (1000) = 3 dB, the horizontal axis in FIG. 4 is 2 and the vertical axis is 2d
Since B + 3dB = 5dB, it becomes 5. Therefore, since the intersection of the horizontal axis 2 and the vertical axis 5 in FIG. 4 is a black circle, the inspected product is determined to be defective.

In this way, when it is determined whether the product is good or defective, the result of the determination is displayed on a display such as a CRT 20 (step 120), or some device is directly controlled by a signal of the result. Is also good.

[Effects of the Invention] As described above, according to the present invention, a margin is added to an average level for each frequency band to set a criterion, and when the number of analysis values exceeding the criterion exceeds the set number, Since the product is judged to be defective and the result of the judgment is displayed on the CRT, the noise test can be performed automatically, which has the effect of facilitating the test. The noise discrimination is not performed by the overhaul, but the discrimination is performed for each frequency band by the frequency analysis, so that there is an effect that highly accurate discrimination is possible.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a noise discriminating apparatus to which a noise discriminating method according to one embodiment of the present invention can be applied. FIG. 2 is a flowchart for explaining the noise discriminating method according to one embodiment of the present invention. 3 and 4 are diagrams for explaining the operation of the present invention. In the figure, 10 is an inspection box, 12 is a microphone, G is a product to be inspected, 14 is a sound level meter, 16 is a frequency analyzer, 18 is a computer, and 20 is a CRT.

Claims (1)

(57) [Claims] 1. A product to be inspected (G) which shuts off external noise.
Using a test box (10) that inputs only the noise generated by the microphone (12), the signal representing the sound pressure from the microphone (12) is converted to a sound level meter (14) and a frequency analyzer (1
6) Using a computer (18) and a CRT (20), to determine the sound pressure level of the noise generated from the product to be inspected, analyze the frequency of the noise generated from non-defective products, Setting a judgment criterion by adding a margin value A to the average level of each product, and analyzing the noise generated from the product to be inspected by the frequency analyzer (1).
6) frequency analysis, comparing the analysis value for each frequency band with the criterion of the corresponding band, the number of the analysis values exceeding the criterion, and the analysis value exceeding the criterion, Determining the difference from the criterion by the computer (18); and determining that the number of the analysis values exceeding the criterion for each band, obtained by the comparison, exceeds a set number. Making a determination and displaying the result of the determination on the CRT (20).