JP3263318B2 - Diagnosis method of wall peeling - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for diagnosing a peeling of a wall surface by diagnosing a peeling state of the wall surface by collecting a striking sound generated when the wall surface is hit with a percussion instrument. It is.

[0002]

2. Description of the Related Art Conventionally, as a method for diagnosing a peeling state of a tile wall surface, for example, a percussion method for judging a sound of striking a wall surface with a hammer or the like has been generally performed. However, in the method of hearing and judging the impact sound as described above, there is an individual difference between persons who judge, and it is not possible to accurately determine the peeling state.

FIG. 8 is a block diagram for explaining a conventional tile wall peeling diagnosis apparatus in which the peeling state is reliably determined regardless of individual differences in view of the above points. FIG. FIG. 2B is a configuration diagram of the diagnosis device, FIG. 2B is a configuration diagram of a contact portion of the separation diagnosis device, and FIG. 2C is an explanatory diagram illustrating a diagnosis method of the separation diagnosis device.

In FIG. 8, reference numeral 1 denotes a peeling diagnosis device for diagnosing the peeling state of the wall surface 6 to be measured. A wall contact portion 2 for contacting the wall surface 6 to be measured is provided at the tip of the peeling diagnosis device 1. The wall contact portion 2 is provided with a hitting ball 3 for hitting the wall surface and a sound collecting microphone 4 for collecting a hitting sound. Reference numeral 5 denotes an input switch for hitting the hitting ball 3 against the wall surface 6 to be measured.

[0005] 7a, 7b, 7c collects the impact sound generated when the impact ball 3 hits the wall surface 6 to be measured by the sound collecting microphone 4, and the result is analyzed by an analyzing device (not shown) to bond the wall surface. 3 shows light-emitting diodes of three colors adapted to display intensity; if the adhesive strength is 4 kg / cm ² or more, blue light-emitting diodes 7a, 1-4 kg / cm
If it is ² , the yellow light emitting diode 7c, 0 to 1 g / cm ²
In this case, each is indicated by the red light emitting diode 7b.
R> as shown.

Accordingly, the hitting ball 3 hits the wall surface 6 to be measured by applying the switch 5 with the wall surface contact portion 2 provided at the tip of the peeling diagnosis apparatus 1 having the above-mentioned configuration being applied to the wall surface 6 to be measured. That is, at that time, the hitting sound generated is collected by the sound collecting microphone 4, and the hitting sound is analyzed by the analyzer, and the adhesive strength of the measured wall 6 is displayed by the three light emitting diodes 7a to 7c. By performing the partial repair by judging the peeling state, it is possible to reduce the cost and enhance the safety of the reform.

[0007]

However, in the above-described conventional apparatus for diagnosing wall peeling, the wall surface to be measured 6
However, even if the adhesive strength could be identified by the light emitting diodes 7a to 7c of three colors, the degree of peeling such as the depth of peeling could not be detected accurately.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems of the prior art, and an object of the present invention is to provide a method of diagnosing a peeling of a wall surface capable of accurately diagnosing a peeling state such as a peeling depth. Aim.

[0009]

A method of diagnosing separation of a wall surface according to the present invention collects a striking sound when striking a wall surface to be measured, analyzes the waveform of the striking sound, and analyzes the power spectrum density. In a method of diagnosing a peeling of a wall surface based on which a peeling portion of a measured wall surface is determined, as a characteristic amount of a spectral waveform, the horizontal axis is a predetermined frequency range and the upper limit of the vertical axis is a peak point of the spectral waveform. power spectrum waveform and the peak of the lower drawn as a 0
Area was sandwiched between a straight line parallel to the street abscissa point, a first area between the lower limit value of the resonance frequency and the frequency range corresponding to the peak point of the spectrum waveform, the peak of the spectral waveform Extracting a second area between a resonance frequency corresponding to a point and an upper limit value of a frequency range and a first to second area ratio, and determining a peeling portion based on the extracted feature amount; It is characterized by the following.

Further, the present invention is characterized in that the degree of the depth of the peeled portion is determined based on the frequency shift of the first and second area ratios.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiment 1 FIG. Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a configuration diagram showing a wall peeling diagnosis device used in the present invention. Exfoliation diagnostic apparatus 10 of illustrated configuration
Has a configuration basically similar to that of the conventional example shown in FIG. 8, but the sound collecting microphone 4 is attached to the apparatus main body side, and a signal processor 8 is connected to the sound collecting microphone 4, Performs a spectrum analysis of a waveform of the striking sound from a sound collection result of the striking sound when striking the measured wall surface, and extracts a characteristic amount of a spectral waveform based on the power spectral density. The peeled portion is determined based on the information, and the result of the determination is displayed on the display panel 9.

A flow of a series of operations by the peeling diagnosis apparatus 10 will be described with reference to FIG. First, when the switch 5 is turned on by applying the wall contact portion 2 to the wall surface 6 to be measured, the hitting ball 3 hits the wall surface 6 to be measured (step S1). The striking sound is collected by the sound collecting microphone 4 (step S2). Since the sound is an analog waveform, the signal processor 8 digitally converts and records the response waveform so as to obtain a response waveform that is easy to process (step S3). The signal processor 8 analyzes the frequency spectrum of the obtained response waveform (spectrum S4) and recognizes the pattern (step S5).

The pattern recognition by the signal processor 8 is as follows.
As shown in FIG. 3, the characteristic amount of the spectrum waveform is extracted from the power spectral density, and the peeling portion is determined based on the extracted characteristic amount. Here, the area of the spectrum waveform is normalized so as to be 1.0 pu, the lower limit value fmin of the frequency range is 0 Hz, and the upper limit value fmax is the upper limit value of the audible frequency in order to obtain consistency with the human sensory test. 12,000 Hz. The peak value of the spectrum waveform shown in FIG. 3 increases as the frequency of the striking sound becomes sharper. Therefore, the frequency range indicated by hatching in FIG. 3 is the area A between the resonance frequency and the lower limit of the frequency range. The range increases with the area B between the resonance frequency and the upper limit of the frequency range. The area A is smaller and the area B is larger as the main resonance frequency f ₀ is lower.

The signal processor 8 uses the power spectrum density shown in FIG. 3 to calculate three types of characteristic quantities of the area A, the area B and their area ratio (the position of the resonance frequency f ₀ ) as the characteristic quantities of the spectrum waveform. The extracted portion is determined based on the extracted feature amount, and the result is displayed and stored on the display panel 9 (step S6). Further, the signal processor 8 includes:
As shown in FIG. 4, the degree of the depth of the peeled portion is determined based on the frequency shift of the area ratio, and the result is displayed and stored on the display panel 9 (step S6). That is, FIG.
Is a case where the depth of the peeled portion is 15 mm, FIG. 4B is a case where the depth of the peeled portion is 20 mm, FIG. 4C is a case where the depth of the peeled portion is 30 mm, and FIG. It can be seen that the peak of the frequency shifts to the right as the depth of the peeled portion increases. As described above, the reason why the peak point (main resonance frequency) shifts to the right side where the frequency is higher as the depth of the peeled portion increases is that the deeper the peeled portion, the greater the thickness of the peeled portion and the greater the rigidity. This is because the frequency of the impact sound increases.

As a result of examining the extent to which the state of the peeled portion can be determined based on such characteristic amounts, it was found from experimental results that the determination based on three types of characteristic amounts has the highest correct answer rate in fuzzy logic. . That is, FIG. 5 shows three types of feature amounts, that is, area A, area B, and their area ratio A / B as feature amounts. FIG.
FIG. 7 shows a case where two kinds of feature amounts of the area B are taken, and FIG. 7 shows the correct answer rate of “depth and area” when two kinds of feature amounts of the peak point PSD and the peak point are taken as the feature amounts. The correct answer rate for "depth" and the correct answer rate for "whether sound or peeled" are shown. As can be understood from these figures, the three types of feature quantities shown in FIG. It was confirmed that the peeling diagnosis based on the peeling was the most preferable for the diagnosis of the peeled portions having different depths and sizes.

[0016]

As described above, according to the present invention, a striking sound when a striking wall surface is measured is collected, the waveform of the striking sound is analyzed, and the measured sound is measured based on the power spectrum density. In the method of diagnosing the peeling of the wall surface to determine the peeling portion of the wall surface,
The upper limit of the vertical axis is the peak point of the spectrum waveform is the lower limit drawn as a 0 as well as the horizontal axis and the predetermined frequency range
Parallel to the horizontal axis passing through the power spectrum waveform and the peak point
Area was sandwiched between a straight line, the spectral waveform
A first area between the lower limit value of the resonance frequency and the frequency range corresponding to the peak point, the peak point of the spectrum waveform
The second between the corresponding resonance frequency and the upper limit of the frequency range
And the first and second area ratios are extracted, and the peeling portion is determined based on the extracted feature amounts,
It is possible to accurately diagnose the peeled state of the peeled portions having different depths and sizes.

Further, the degree of the depth of the peeled portion can be accurately determined based on the frequency shift of the first and second area ratios.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing a wall peeling diagnosis apparatus used in the present invention.

FIG. 2 is a flowchart illustrating an operation of the peeling diagnosis apparatus of FIG. 1;

FIG. 3 is a waveform chart for explaining a spectrum waveform and a feature amount normalized for pattern recognition.

FIG. 4 is an explanatory diagram for determining the depth of a peeled portion.

FIG. 5 is an explanatory diagram showing a correct answer rate when three types of feature amounts of an area A, an area B, and an area ratio A / B thereof are taken as the feature amounts;

FIG. 6 is an explanatory diagram showing a correct answer rate when two types of feature amounts, area A and area B, are taken as feature amounts;

FIG. 7 is an explanatory diagram showing a correct answer rate when two types of feature amounts, a peak point frequency and a peak point PSD, are taken as feature amounts;

FIG. 8 is an explanatory view of a tile wall peeling diagnosis apparatus according to a conventional example.

[Explanation of symbols]

10 Peeling diagnosis device, 2 wall contact part, 3 impact ball, 4
Sound collecting microphone, 8 signal processor, 9 display board.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References Akira Kawada, Shigeo Hayashi, Diagnosis method of wall painting with neural network, Gi-en, Japan, No. 80, pp. 16-20 (58) Fields surveyed (Int. Cl ^7, DB name) G01N 29/00 -. 29/28 JICST file (JOIS)

Claims (2)

(57) [Claims] 1. A wall surface which collects a striking sound when striking a wall surface to be measured, performs a spectrum analysis of a waveform of the striking sound, and determines a peeling portion of the wall surface to be measured based on its power spectrum density. in the peeling diagnostic method, as a feature of the spectrum waveform, the peak upper limit of the vertical axis the peak point of the spectrum waveform and power spectrum waveform drawn the lower limit set to 0 as well as the horizontal axis and the predetermined frequency range
Area was sandwiched between a straight line parallel to the street abscissa point, a first area between the lower limit value of the resonance frequency and the frequency range corresponding to the peak point of the spectrum waveform, the peak of the spectral waveform Extracting a second area between a resonance frequency corresponding to a point and an upper limit value of a frequency range and a first to second area ratio, and determining a peeling portion based on the extracted feature amount; A method for diagnosing peeling of a wall surface, characterized in that: 2. The method for diagnosing wall separation according to claim 1, wherein the degree of the depth of the separation portion is determined based on a frequency shift of the first and second area ratios. Diagnostic method.