JP3922459B2 - Separation and cavity detection method and apparatus by percussion method - Google Patents

Description

  The present invention relates to a sound percussion method that analyzes the difference in response sound obtained by striking the surface of an object to be inspected such as a structure or various composite materials, and detects internal deterioration such as cracks, separation, and cavities by nondestructive inspection The present invention relates to a method and an apparatus for peeling and detecting a cavity.

There are methods such as the percussion method, the radar method, the infrared method, and the ultrasonic method for detecting the internal deterioration of the inspection object by non-destructive inspection, but all have difficulties in reproducibility, reliability, efficiency, application conditions, etc. Yes, full-scale practical use has not been achieved, and further technological development is desired.
Of these methods, the conventional method that relies on human hearing has a problem with detection accuracy and reproducibility, so the characteristics such as the frequency and amplitude of the impact sound are analyzed by a computer to automatically remove and remove the cavity. Methods of detection are being considered. For this purpose, it is necessary to clearly distinguish the difference in sound quality between the part with and without the separation and the cavity.

However, in the case of striking position, separation, cavity size, and various composite materials, the frequency component of the striking sound differs depending on the application amount of the impregnated adhesive resin, etc., and the frequency component of the deteriorated part and the healthy part approximate. It is not easy to clearly identify the difference in sound quality.
For example, in Japanese Patent Laid-Open No. 5-322861, the waveform and frequency spectrum of a hitting sound are analyzed, and in a place where there is separation shown in FIG. 4 (a), a sound having a lower frequency than in a place where there is no separation shown in FIG. The excellence is clear and there are differences, but the criteria for identifying the differences are not clearly stated.
JP-A-5-322861

  The problem to be solved is that it is difficult to discriminate the difference in sound quality of different impact sounds between where there is separation or void and where the present invention clearly identifies the difference in sound quality of the impact sound. An object of the present invention is to provide a method and an apparatus for detecting peeling and cavity by a percussive method.

The present applicant notices that the hitting sound changes to a high sound at a part where there is a separation or a cavity, and as a result of various observations, it is remarkable that a waveform having a considerably high frequency overlaps with the waveform of the first wave in particular. Observed.
For example, the sound of a musical instrument is generated when an object resonates. For this reason, if there is separation, the mass of the portion becomes smaller than that of the portion without separation, and thus it is considered that the hitting sound changes to a high sound.

The sound of a musical instrument is made by synthesizing a plurality of sine waves whose frequencies have an integer ratio relationship.
Of these sine waves, the sine wave with the lowest frequency is the fundamental wave, and the other sine waves are harmonics. The timbre of a musical instrument is determined by how much harmonic component it contains, and it is the frequency of the fundamental wave (fundamental frequency) that represents the pitch (pitch).
Therefore, if the fundamental wave included in the first wave is cut out and its frequency is specified, the pitch (pitch) of the hitting sound can be known, and as a result, the difference in sound quality that is different between the part with and without the separation can be clearly identified. It becomes like this.

  From the above viewpoints, the present invention has the main feature of identifying a part where there is no separation and a part where there is no separation, using the fundamental frequency included in the first wave of the response sound as a criterion.

In the present invention, only the first wave of the response sound is extracted and the fundamental frequency is used as a criterion. Therefore, the separation evaluation and the cavity are performed using a simple identification circuit as compared with the conventional method of analyzing the entire waveform and frequency spectrum. Can be detected.
In addition, since the fundamental frequency of the first wave of the response sound has relatively few fluctuation elements, a stable evaluation result can be obtained.

Embodiments of the present invention will be described below.
FIG. 1 shows a configuration diagram of a separation and cavity detection apparatus embodying the present invention.
The peeling and cavity detecting device is configured by connecting a personal computer 4 to a hitting inspection unit 3 in which the hitting unit 1 and the inspection unit 2 are integrated.
The hammering section 1 is provided with a hammer 5 driven by a solenoid and a control device 6 for controlling the driving of the hammer 5, and the inspection section 2 is provided with a microphone 7 and an analyzer 8 for analyzing sound waves received by the microphone 7.
The analyzer 8 is configured using a DSP (Digital Signal Processing Integrated Circuit) that substitutes analog circuit elements such as coils, capacitors, transistors, and operational amplifiers for digital signal processing.

The peeling and cavity detecting device is configured as described above, and the surface of the inspection object 9 is hit with a hammer 5 with a constant force under the control of the control device 6, and the response sound is received by the microphone 7, and the analysis device 8. The frequency characteristics are analyzed with.
The response sound is not a sound wave propagating through the air generated on the striking surface, but a sound wave that passes through the inside of the inspection object 9 and is reflected on the bottom surface is received as a response sound.
The analysis result is transferred to the personal computer 4 and statistically analyzed. Based on the knowledge that the first wave of the response sound changes to a high sound where there is a gap, it is verified whether there is a gap inside the inspection object 9. To do.
In addition, when performing simply, it is also possible to make a determination only with an analyzer without using a personal computer.

The test first hits a plurality of locations where it is confirmed that there are no voids in the inspection object 9 several times, and decomposes the first wave of each response sound into a fundamental wave and a harmonic component to obtain a fundamental frequency. , And the average value is obtained as the reference frequency f0, and at the same time, the standard deviation σ is calculated.
Next, the portion to be inspected of the inspection object 9 is hit several times, the first wave of each response sound is similarly decomposed into the fundamental wave and the harmonic component, the fundamental frequency is specified, and the average value is obtained. The measurement frequency is f1.

  Next, when the sample distribution is close to the normal distribution, the significance level (risk rate) α = 5% test based on Chebyshev's inequality that only about 5% is 1.95σ or more away from the reference frequency f0 When the difference between the measurement frequency f1 and the reference frequency f0 is 1.95σ or less, it is determined that the OK (no gap) is OK, and when the difference is 1.95σ or more, NG (the gap is present) is determined.

  In the batting inspection, the same position of the inspection object 9 is struck with the hammer 5 several times to determine the presence or absence of a gap. However, the hammer 5 is moved at equal intervals and scanned in the vertical and horizontal directions. By hitting several times, it is possible to quantitatively grasp the size and depth of the gap inside the inspection object 9.

  The harder the substance is, the more likely it is to transmit vibration. Since sound is also vibration, the speed of sound transmission is higher in a dense, hard substance than in air. The specific acoustic resistance value (acoustic impedance) representing this is about 4 × 10 2 for air and about 10 7 for metal, and there is an order of magnitude difference. Therefore, sound reflection occurs at the interface having this difference.

  The sound generated when the object 9 is struck over a part with a gap reaches the gap and is reflected. At this time, the sound that reaches the bottom surface and reflects is overlapped with the sound that reflects in the middle gap, but the sound that reaches the bottom surface and returns is lower in frequency because vibration takes longer than the sound reflected in the middle gap ( The wavelength becomes longer).

  Therefore, when the high-frequency wave reflected by the air gap in the middle and the low-frequency wave reflected by the bottom surface overlap, the half-cycle of the first wave, as shown in FIG. When two or more waves arrive at the same time, the vibration at that point will be the sum of the vibrations when each wave arrives separately), so that the wave height of the higher frequency will always increase. For the above reason, it is considered that the first wave of the response sound changes to a higher sound at a place with a gap than at a place without a gap.

FIG. 3 is a flowchart showing the processing contents of the peeling and cavity detecting apparatus embodying the present invention.
First, in step 101, the analog sound wave signal S received by the microphone 7 is A / D converted into a digital sound wave signal S (n).
At this time, the analog sound wave signal S is passed through an LPF (low pass filter) before A / D conversion to remove high frequency components unnecessary for analysis. As the cut-off frequency in this case, the highest frequency observed by hitting the inspection object 9 with the hammer 5 is set.
At the same time, the analog sound wave signal S is amplified to a voltage that meets the specifications of the input terminal of the A / D converter using an amplifier.

In the next step 102, the digital sound wave signal S (n) is multiplied by the rectangular wave window function W (n) to limit the time, and the first wave digital sound wave signal SW (n) = S (n) for one cycle. ) × W (n).
At this time, the timing of applying the window function W (n) in order to cut out the first-wave sound wave generated immediately after the impact is synchronized with the drive pulse of the hammer 5.
Further, in order to set the width to be cut out for one cycle, the observation time of the window function W (n) is set in accordance with the wavelength of the sound wave of the first wave observed by hitting the inspection object 9 with the hammer 5.

  In the next step 103, the digital sound wave signal SW (n) is subjected to FFT (Fast Fourier Transform) processing, and the digital sound wave signal SW (n) continuously distributed on the time axis is decomposed into a fundamental wave and a harmonic component. To a line spectrum X (n) distributed discretely on the frequency axis, thereby identifying the fundamental frequency of the first wave.

  In the next step 104, the fundamental frequency of the first wave is transferred to the personal computer 4 to perform the above-described verification, and the verification result is displayed as OK (no gap) or NG (with gap).

Hereinafter, experimental results of the peeling and cavity detection apparatus embodying the present invention will be described.
In the experiment, a drill hole was drilled in the center of a plurality of test pieces to provide a gap, and the top and four corners of the center hole were hit with a hammer 5 four times.
Each test piece was unified into a square of 100 × 100 mm although the material and thickness were different. The experiment was performed by measuring the wavelength of the fundamental wave, not the fundamental frequency of the first wave of the response sound.
The experimental results for each test piece are shown below.

Test piece 1: A drill hole with a diameter of 16 mm and a depth of 8 mm was formed in the center of an aluminum alloy (5052) having a thickness of 15 mm.
The response sound measurement results are as follows.
Center: 55.25 μs
Corner 1: 67.88μs
Corner 2: 66.13 μs
Corner 3: 60.13 μs
Corner 4: 62.25 μs
From the above measurement results, when the average value of the wavelengths at the four corners (reference wavelength λ ₀ ) and the standard deviation σ are obtained, λ ₀ = 64.09 and σ = 3.065 are obtained.
Next, when the difference between the reference wavelength λ ₀ and the average value of the central wavelength (measurement wavelength λ ₁ ) is obtained,
λ ₀ −λ ₁ = 64.09−55.25 = 8.84.
As described above, since λ ₀ −λ ₁ = 8.84> 1.95σ = 5.98, the determination result is NG, and it was found that the central machining hole can be detected.

Test piece 2: A drill hole with a diameter of 16 mm and a depth of 13 mm was formed in the center of a resin (PP) having a thickness of 20 mm.
The response sound measurement results are as follows.
Center: 83.25 μs
Corner 1: 90.50 μs
Corner 2: 91.50 μs
Corner 3: 88.63μs
Corner 4: 87.50 μs
From the above measurement results, when obtaining the reference wavelength λ0 and the standard deviation σ,
λ ₀ = 89.53 and σ = 1.562.
Next, when the difference between the reference wavelength λ0 and the measurement wavelength λ1 is obtained,
λ ₀ −λ ₁ = 89.53−83.25 = 6.28.
As described above, since λ ₀ −λ ₁ = 6.28> 1.95σ = 3.05, the determination result is similarly NG, and it was found that the center machining hole can be detected.

Test piece 3: A drill hole with a diameter of 16 mm and a depth of 13 mm was formed in the center of a resin (POM) having a thickness of 20 mm.
The response sound measurement results are as follows.
Center: 78.00 μs
Corner 1: 89.75 μs
Corner 2: 92.13 μs
Corner 3: 98.00 μs
Corner 4: 95.38 μs
From the above measurement results, when obtaining the reference wavelength λ ₀ and the standard deviation σ,
λ ₀ = 93.81 and σ = 3.136.
Next, when the difference between the reference wavelength λ ₀ and the measurement wavelength λ ₁ is obtained,
λ ₀ -λ ₁ = 93.81-78.00 = 15.81.
As described above, since λ ₀ −λ ₁ = 15.81> 1.95 and σ = 6.12, the determination result is similarly NG, and it has been found that the center machining hole can be detected.

Test piece 4: A drill hole with a diameter of 16 mm and a depth of 13 mm was formed in the center of a resin (acrylic) having a thickness of 20 mm.
The response sound measurement results are as follows.
Center: 78.00 μs
Corner 1: 83.25 μs
Corner 2: 84.38 μs
Corner 3: 80.50 μs
Corner 4: 80.88 μs
From the above measurement results, when obtaining the reference wavelength λ ₀ and the standard deviation σ,
λ ₀ = 82.25 and σ = 1.618.
Next, when the difference between the reference wavelength λ ₀ and the measurement wavelength λ ₁ is obtained,
λ ₀ −λ ₁ = 82.25−78.00 = 4.25.
As described above, since λ ₀ −λ ₁ = 4.25> 1.95 and σ = 3.16, the determination result is similarly NG, and it has been found that the center machining hole can be detected.

Test piece 5: Drill holes with diameters of 20, 30, and 10 mm were formed at three central positions of a resin (CARBON / EPOXY) having a thickness of 25 mm.
The response sound measurement results are as follows.
Center: 48.25 μs
Corner 1: 98.75 μs
Corner 2: 98.75 μs
Corner 3: 86.63 μs
Corner 4: 85.88 μs
From the above measurement results, when obtaining the reference wavelength λ ₀ and the standard deviation σ,
λ ₀ = 92.50 and σ = 6.253.
Next, when the difference between the reference wavelength λ ₀ and the measurement wavelength λ ₁ is obtained,
λ ₀ −λ ₁ = 92.50−48.25 = 44.25.
As described above, since λ ₀ −λ ₁ = 44.25> 1.95 and σ = 12.19, the determination result is similarly NG, and it has been found that the center machining hole can be detected.

It is a block diagram of the peeling and cavity detection apparatus which implemented this invention. It is explanatory drawing where the wave height of the higher frequency becomes high in the half cycle of the first wave. It is a flowchart of the processing content of the peeling and cavity detection apparatus which implemented this invention. It is a figure which shows the waveform and frequency spectrum of a hit sound.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Blowing part 2 Inspection part 3 Blowing inspection unit 4 Personal computer 5 Hammer 6 Control apparatus 7 Microphone 8 Analyzer

Claims (8)

In a method of detecting peeling and cavities inside the inspection object by analyzing the difference in response sound obtained by hitting the surface of the inspection object,
A separation and cavity detection method using a sounding method, wherein a part having separation or a cavity is identified from a part having or without a cavity based on a fundamental frequency included in the first wave of the response sound. In a method of detecting peeling and cavities inside the inspection object by analyzing the difference in response sound obtained by hitting the surface of the inspection object,
A cutting step of cutting the first sound wave signal by multiplying the response sound by a window function;
A specifying step of decomposing the first wave sound wave signal into a fundamental wave and a harmonic component by FFT processing to specify a fundamental frequency of the first wave;
A determination step of determining the presence or absence of delamination and cavities by comparing the fundamental frequency of the first wave with a reference frequency set in advance as a measurement frequency;
A peeling and cavity detection method by a sound-striking method, characterized by comprising:   3. The peeling and cavity detection method according to claim 2, wherein the timing of applying the window function is synchronized with a driving pulse of a hammer that strikes the surface of the inspection object.   3. The peeling and cavity detection method according to claim 2, wherein the observation time of the window function is one cycle of the sound wave signal of the first wave.   The striking frequency according to claim 2, wherein the measurement frequency in the determination step is an average value of a fundamental frequency of a first wave of a response sound obtained by hitting a portion to be inspected of the inspection object a plurality of times. Sound separation and cavity detection method.   The reference frequency in the determination step is an average value of the fundamental frequency of the first wave of the response sound obtained by hitting a location where it is confirmed that there is no separation or cavity inside the inspection object. 3. A peeling and cavity detection method by a percussion method according to claim 2.   The determination in the determination step is performed by multiplying the standard deviation of the fundamental frequency of the first wave of the response sound obtained by hitting a location where it is confirmed that there is no separation or cavity inside the inspection object by a predetermined coefficient. 3. The peeling and cavity detection method according to claim 2, wherein the difference between the measured value and the difference between the measurement frequency and the reference frequency is compared. In a device that detects the separation and cavities inside the inspection object by analyzing the difference in response sound obtained by striking the surface of the inspection object,
Cutting means for cutting out the first wave sound wave signal by multiplying the response sound by a window function;
Identifying means for decomposing the first wave acoustic signal into a fundamental wave and a harmonic component by FET processing to identify a fundamental frequency of the first wave;
A determination means for determining the presence or absence of separation and a cavity by comparing the fundamental frequency of the first wave with a reference frequency set in advance as a measurement frequency;
An apparatus for detecting peeling and cavity by a sounding method, characterized by comprising:

Images