Inspection device and inspection method

Abstract

To provide an inspection device which enables abnormality of an inspection object to be inspected by a simple constitution.SOLUTION: The inspection device comprises a vibration sensor, a microphone, an impulse response calculation unit, a removal unit, a frequency conversion unit, an energy calculation unit, and an abnormality determination unit. The vibration sensor detects the vibration of an inspection object which is induced to vibrate. The microphone collects radiated sounds from the inspection object. The impulse response calculation unit calculates an impulse response between the vibration sensor and the microphone on the basis of the vibration and the sound pressure of the radiated sounds. The removal unit removes unnecessary components from the impulse response. The frequency conversion unit converts the impulse response from which unnecessary components has been removed, to a frequency characteristic. The energy calculation unit calculates acoustic energy between the vibration sensor and the microphone on the basis of the frequency characteristic. The abnormality determination unit determines the presence of abnormality of the inspection object on the basis of the acoustic energy.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The embodiments relate to an inspection apparatus and an inspection method. [Background technology]

[0002] Technologies have been proposed for non-destructive inspection of abnormalities such as a drop in axial force or cracks by applying physical vibrations such as acoustic waves to the object to be inspected and analyzing its behavior. With this type of technology, the object to be inspected is vibrated by acoustic excitation using a speaker, a hammering mechanism, or by a human hammering, and the presence and location of abnormalities can be detected by collecting the sound emitted from the object to be inspected. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2009-284097 A [Patent Document 2] Patent No. 5242313 [Patent Document 3] Patent No. 6114587 Summary of the Invention [Problem to be solved by the invention]

[0004] The embodiments provide an inspection device and an inspection method capable of inspecting an inspection object with a simple configuration. [Means for solving the problem]

[0005] The inspection device of the embodiment includes a vibration sensor, a microphone, an impulse response calculation unit, a removal unit, a frequency conversion unit, an energy calculation unit, and an abnormality determination unit. The vibration sensor detects vibration of an object to be inspected that has been excited to vibrate. The microphone is disposed in the vicinity of the object to be inspected and collects radiated sound from the object to be inspected. The impulse response calculation unit calculates an impulse response between the vibration sensor and the microphone based on the vibration detected by the vibration sensor and the sound pressure of the radiated sound collected by the microphone. The removal unit removes unnecessary components from the impulse response. The frequency conversion unit converts the impulse response from which the unnecessary components have been removed into frequency characteristics. The energy calculation unit calculates acoustic energy between the vibration sensor and the microphone based on the frequency characteristics. The abnormality determination unit determines the presence or absence of an abnormality in the object to be inspected based on the acoustic energy. [Brief description of the drawings]

[0006] [Figure 1] FIG. 1 is a block diagram showing an example of an inspection device according to an embodiment. [Diagram 2] FIG. 2 is a diagram showing the impulse response and the frequency characteristic before and after removal of the excitation sound. [Diagram 3] FIG. 3 is a diagram for explaining the operating principle of the inspection device. [Figure 4] FIG. 4 is a diagram for explaining the sound pressure of each of the microphones. [Diagram 5] FIG. 5 is a diagram for explaining different vibration modes. [Figure 6] FIG. 6 is a diagram for explaining the difference between low-frequency resonance and high-frequency resonance. [Figure 7] FIG. 7 is a diagram for explaining the operation principle of the inspection device shown in FIG. [Figure 8] FIG. 8 is a diagram for further explaining the operating principle of the inspection device shown in FIG. [Figure 9] FIG. 9 is a flowchart illustrating an example of a processing procedure of the inspection device according to the embodiment. [Figure 10]FIG. 10 is a diagram showing an example of a graph obtained in an experiment for verifying the effect of the embodiment. [Figure 11] FIG. 11 is a diagram showing another example of a graph obtained in an experiment for verifying the effect of the embodiment. [Figure 12] FIG. 12 is a diagram showing another example of a graph obtained in an experiment for verifying the effect of the embodiment. [Figure 13] FIG. 13 is a diagram for explaining the real-time cross-correlation processing. [Figure 14] FIG. 14 is a waveform diagram showing an example of an impulse response obtained in real time. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Hereinafter, an embodiment will be described with reference to the drawings. (composition) Fig. 1 is a block diagram showing an example of an inspection device according to an embodiment. The inspection device shown in Fig. 1 includes a vibration sound source 101, a speaker 102, a microphone 103, a laser sensor 107, a processing unit 104, a memory 105, and a display 106. This inspection device excites an inspection object O to vibrate by irradiating the inspection object O with a vibration sound, for example, and collects a sound emitted from the inspection object O to inspect the presence or absence of an abnormality in the inspection object O. The abnormality in the inspection object O is, for example, a crack C generated in the inspection object O.

[0008] In addition, there are various methods for exciting vibrations in the inspection object O, such as emitting excitation sound from a speaker, using a machine such as a hammering mechanism, or hammering excitation by a person. In other words, the acoustic excitation signal may be generated by any method. In the embodiment, a method of emitting excitation sound from a speaker will be described as an example.

[0009] 1, a vibration sound source 101 serving as a vibration unit generates an acoustic vibration signal for generating a vibration sound to be irradiated to an inspection object O. The vibration sound may be, for example, a one-point hitting sound. The speaker 102 is disposed to face the inspection object O, and irradiates the inspection object O with a vibration sound in accordance with an acoustic vibration signal input from the vibration sound source 101. The vibration sound causes the inspection object O to vibrate in the direction D as a whole, and a radiation sound is emitted from the inspection object O in association with this vibration.

[0010] The microphone 103 is placed near the inspection object O, such as 2.5 cm from the inspection object O. The microphone 103 collects radiated sound from the inspection object O, converts the collected radiated sound into an electrical signal, and outputs the electrical signal to the processing unit 104.

[0011] The laser sensor 107 is a non-contact vibration sensor such as an LDV (Laser Doppler Vibrometer), and is disposed at an arbitrary distance from the microphone 103. The laser sensor 107 irradiates the inspection object O with laser light, measures the vibration of the inspection object O based on the Doppler shift of the reflected laser light, and generates a vibration measurement signal.

[0012] The processing unit 104 is a processor having a digital signal processor such as a CPU, ASIC, FPGA, or DSP, and performing various processes related to the inspection device. The processing unit 104 may be composed of a single CPU or the like, or may be composed of multiple CPUs or the like. The processing unit 104 operates as an impulse response calculation unit 1041, a removal unit 1042, a frequency conversion unit 1043, an energy calculation unit 1044, and an abnormality determination unit 1045 by executing an acoustic inspection program stored in the memory 105, for example.

[0013] The impulse response calculation unit 1041 samples the acoustic signal collected via the microphone 103 and the vibration measurement signal output from the laser sensor 107 according to a sampling frequency, and performs analog-to-digital (AD) conversion to obtain digital inspection data. Then, the impulse response calculation unit 1041 performs digital calculation processing based on the vibration measurement signal (digital data) of the vibration collected via the laser sensor 107 and the sound pressure (digital data) of the radiated sound collected via the microphone 103, and calculates the impulse response between the microphone 103 and the laser sensor 107. For example, the impulse response calculation unit 1041 calculates the impulse response by adaptive identification processing using a convolution operation.

[0014] The removal unit 1042 removes unnecessary components of the excitation sound N from the impulse response calculated by the impulse response calculation unit 1041, and extracts components of the vibration radiation sound S associated with the vibration of the crack C of the inspection object O.

[0015] Fig. 2 shows the impulse response and frequency characteristics before and after removal of the excitation sound. As shown in Fig. 2, the impulse response collected by the impulse response calculation unit 1041 includes a component of the excitation sound N (dotted line), a component of the vibration radiation sound S (solid line S), and a component of the weak vibration radiation sound A (dashed line). Therefore, the frequency characteristics calculated based on the impulse response also include the frequency characteristics of the excitation sound N, the frequency characteristics of the vibration radiation sound S, and the frequency characteristics of the weak vibration radiation sound A.

[0016] The weak vibration radiation sound A caused by the vibration of the crack C of the inspection object O is weaker than the excitation sound N. In other words, the excitation sound N is necessary for exciting the vibration radiation sound S, but is unnecessary noise for determining an abnormality of the inspection object O. Therefore, the removal unit 1042 removes the components of the excitation sound N that become noise from the impulse response.

[0017] Here, the excitation sound N collected by the microphone 103 includes a component of a direct wave Nd from the speaker 102 and a component of a reflected wave Nr from the inspection object O. The component of the direct wave Nd can be removed by the setting of the microphone 103 or an adaptive identification process. On the other hand, the component of the reflected wave Nr from the inspection object O can be removed by, for example, detecting the maximum peak of the impulse response and removing a component of a predetermined time width including the maximum peak in the impulse response, that is, the number of sampling points equivalent to the time width.

[0018] The removal unit 1042 performs a process of removing components of a predetermined time width including the maximum peak in such an impulse response. As shown in the impulse response in Fig. 2, the reverberation of the vibration radiation sound S is longer than that of the excitation sound N (=Nd+Nr). Therefore, by removing the components of a predetermined time width in the impulse response, the frequency characteristics calculated based on the impulse response output from the removal unit 1042 include only the frequency characteristics of the vibration radiation sound S. Since the frequency characteristics of the vibration radiation sound differ depending on the presence or absence of an abnormality, the presence or absence of an abnormality can be determined based on this difference.

[0019] The frequency conversion unit 1043 converts the impulse response output from the removal unit 1042 into a frequency characteristic. For example, the frequency conversion unit 1043 converts the impulse response into a frequency characteristic using FFT (Fast Fourier Transformation).

[0020] The energy calculation unit 1044 calculates the acoustic energy (average acoustic energy) between the microphone 103 and the laser sensor 107 based on the gain over the entire frequency range in the frequency characteristics output from the frequency conversion unit 1043.

[0021] The abnormality determination unit 1045 determines the presence or absence of an abnormality in the inspection object O and, if an abnormality is present, the degree of progression of the abnormality based on the acoustic energy calculated by the energy calculation unit 1044. For example, the abnormality determination unit 1045 stores an acoustic energy threshold value when an abnormality is present in an inspection object similar to the inspection object O. The abnormality determination unit 1045 determines the presence or absence of an abnormality by comparing the acoustic energy calculated by the energy calculation unit 1044 with the threshold value. Similarly, the abnormality determination unit 1045 stores an acoustic energy threshold value when an abnormality is progressing in an inspection object similar to the inspection object O. The abnormality determination unit 1045 determines the degree of progression of the abnormality by comparing the acoustic energy calculated by the energy calculation unit 1044 with the threshold value.

[0022] The memory 105 is made up of a ROM and a RAM. The ROM stores various programs such as a startup program for the inspection device and an acoustic inspection program executed by the processing unit 104. The RAM can be used as a working memory for various calculations and the like in the processing unit 104.

[0023] The display 106 is a display such as a liquid crystal display or an organic EL display, and displays various images. For example, the display 106 displays the determination result by the abnormality determination unit 1045 regarding the presence or absence of an abnormality and the degree of progression of the abnormality.

[0024] Next, the operation of the inspection device of the embodiment will be described. First, the operating principle of the inspection device will be described. 3 is a diagram for explaining the operating principle of the inspection device. First, the principle when two microphones (a first microphone and a second microphone) are used will be explained. When a structure with a crack is excited, a new resonance similar to a higher-order local vibration occurs around the crack. If it is assumed that the area around the crack vibrates locally in a divided manner due to excitation, the vibration field becomes locally discontinuous and asymmetrical, unlike the mode vibration caused by normal plate resonance. If this vibration field is assumed to be a minute divided vibration, it can be considered as a model in which minute elements (a few mm in size) on the surface of the inspection object O shown in Figure 3 are distributed across multiple two-dimensional flat plates and vibrate with different complex amplitudes.

[0025] When this vibration radiation sound is collected using two microphones, it can be seen that the transfer function G between the microphones is a function of the complex vibration amplitude αi between each element of the divided vibration group and the coefficient βi due to the propagation path ratio related to the distance r from the sound source generating this divided vibration radiation sound, i.e., the position of the divided element to the microphone.

[0026] If vibration becomes more likely due to differences in the degree of deterioration (crack depth, etc.), αi increases due to increases in vibration displacement, velocity, and vibration volume velocity qi (= vibration velocity x element area). Here, the volume velocity qi can also be expressed as a ratio to q1, qi = αiq1. Depending on the size and progress of the deterioration, the radiation position of the radiated sound also changes, and therefore βi also changes. Therefore, based on changes in the transfer function G between the two microphones, it is possible to measure the vibration caused by acoustically exciting the deteriorated area.

[0027] The transfer function G (impulse response) between two microphones is expressed by formula (1). The asterisk (*) in formula (1) is a symbol representing a complex conjugate.

[0028]

number

[0029] 4, the distance from the vibration element Ei to the first microphone is r1i, the distance from the vibration element Ei to the second microphone is r2i, the sound pressure of the first microphone is P1, and the sound pressure of the second microphone is P2. If the volume velocity in the vibration element Ei (= vibration velocity vi in ​​the vibration element Ei × area si of the vibration element Ei) is qi, P1 and P2 can be expressed by formula (2).

[0030]

number

[0031] Figure 5 is a diagram for explaining different vibration modes. (A) shows point vibration, (B) shows uniform vibration of the entire surface, and (C) shows local vibration. In a typical vibration field without cracks, the transfer function G between microphones in the case of a point vibration source or when the entire surface vibrates uniformly is shown in equation (3).

[0032]

number

[0033] As shown in equation (3), since the transfer function G between the two microphones does not include the vibration information qi, the radiated sound cannot be detected.

[0034] The transfer function G between the microphones when the microphones are moved away from the target area is shown in equation (4).

[0035]

number

[0036] As shown in equation (4), in this case too, the vibration information qi is not included, so that the radiated sound cannot be detected.

[0037] On the other hand, in the case of the local vibration in Figure 5(C), since it is not a regular vibration mode radiation, it is possible to detect irregular local radiation. In other words, as shown in formula (5), the key point is that there is a difference in the propagation distance ri from each part to the microphone. In other words, in the vibration generated by exciting the crack, the difference in the propagation distance ri from each part vibrating with locally different volume velocity qi (amplitude phase) to the microphone also has an effect. Therefore, it is possible to detect the radiated sound.

[0038]

number

[0039] Figure 6 is a diagram for explaining the difference between low-frequency resonance and high-frequency resonance. The local vibration in Figure 5(C) corresponds to high-frequency resonance. In high-frequency resonance, the amplitude of each element of the divided vibration varies greatly. In addition, high-frequency resonance is prone to phase shift due to the difference in distance from each element of the divided vibration to the microphone, and is prone to sound pressure interference. In either case, the radiated sound is easy to detect using two microphones, and resonance caused by cracks or a decrease in axial force falls into this category.

[0040] Equation (6) represents the difference in distance from each element of the divided vibration to the microphone.

[0041]

number

[0042] Equation (7) shows the coefficients that change in the vibration radiation surface.

[0043]

number

[0044] Equation (8) shows the transfer function G at high frequency resonance.

[0045]

number

[0046] On the other hand, low-frequency resonance is a type of resonance in which the entire structure resonates, and there is little variation between each element of the divided vibration, so sound pressure interference due to incoming sound pressure is unlikely to occur. Therefore, it is difficult to obtain information based on low-frequency resonance, such as deterioration of vibration damping characteristics, by detecting with two microphones.

[0047] Therefore, in this embodiment, instead of using two microphones, measurements are performed separately using microphone 103 that can obtain overall vibration information and laser sensor 107 that can obtain local vibration information, as shown in Figure 1, and the transfer function of the radiated sound pressure from the vibration of the vibration element irradiated with laser to microphone 103 is measured.

[0048] Fig. 7 is a diagram for explaining the operating principle of the inspection device shown in Fig. 1. In Fig. 7, a random signal such as white noise is amplified by an amplifier, and an acoustic excitation sound N is emitted from a speaker 102. The cracked panel (inspection object O) irradiated with this is vibrated and emits a panel vibration radiation sound S. If there is a crack, a weak vibration radiation sound A is also generated. The panel vibration radiation sound S and the weak vibration radiation sound A are collected by a microphone 103 and converted into an electric signal.

[0049] Meanwhile, the laser sensor 107 measures the vibrations on the surface of the inspection object O, and a vibration measurement signal is acquired. Then, an impulse response (vibration radiation transfer function) between the microphone 103 and the laser sensor 107 is measured. In this impulse response, the panel acoustic excitation wave (excitation sound N) is removed as an unnecessary component, thereby extracting the buried weak vibration radiation sound A, and further, the presence or absence of weak vibrations generated by cracks can be diagnosed based on the frequency characteristics in the wideband region.

[0050] When an excitation sound is directed toward the object to be inspected O, the object to be inspected O vibrates in the direction D as a whole (Fig. 1). At this time, the parts of the object to be inspected O that do not have any abnormalities, such as crack C, vibrate uniformly in the direction D. On the other hand, as shown in Fig. 3, the part of crack C and its surroundings have a localized, discontinuous, and asymmetric vibration field, unlike the mode vibration due to plate resonance in other parts.

[0051] The vibration generated by such a crack C is represented by a model in which a large number of vibration elements Ei (i=1, 2, ..., N) of a few mm in size are distributed on a two-dimensional flat plate, and each vibration element Ei vibrates with a different complex amplitude, as shown in Fig. 4. The inspection device of the embodiment detects radiated sound based on the vibration of such a model. Therefore, the inspection device of the embodiment detects radiated sound from the inspection object O with the microphone 103 and directly detects the vibration of the inspection object O with the laser sensor 107.

[0052] Fig. 8 is a diagram for further explaining the operating principle of the inspection device shown in Fig. 1. In Fig. 8, if the excitation force is F, the measurement value by the laser sensor is v1, and the microphone sound pressure is Pv, the relationship between each parameter is shown in (9).

[0053]

number

[0054] Equation (10) represents an impulse response G including an acoustic excitation wave (excitation sound N) and a microphone sound pressure PR of the acoustic excitation wave (excitation sound N).

[0055]

number

[0056] Equation (11) represents the panel vibration radiated sound Pv at the position of the microphone 103.

[0057]

number

[0058] Equation (12) shows the impulse response G from which the acoustic excitation wave (excitation sound N) has been removed.

[0059]

number

[0060] In formula (12), if it is assumed that the n-th order resonance damping rate ξ is constant and the damping rate is different between the entire panel and the laser vibration portion (rear dispersion control), formula (13) is obtained.

[0061]

number

[0062] (action) 9 is a flowchart showing an example of a processing procedure of the inspection apparatus according to the embodiment. The processing in FIG. In step S1, the vibration sound source 101 irradiates the inspection object O with a vibration sound.

[0063] In step S21, the microphone 103 collects sound. In step S22, the laser sensor 107 measures the vibration of the surface of the inspection object O.

[0064] In step S 3 , the processing unit 104 calculates an impulse response between the microphone 103 and the laser sensor 107 based on the sound pressure of the acoustic signal collected by the microphone 103 and the vibration measurement signal acquired by the laser sensor 107 . In step S4, the processing unit 104 removes the excitation sound component from the calculated impulse response.

[0065] In step S5, the processing unit 104 converts the impulse response from which the excitation sound component has been removed into a frequency characteristic, for example, by FFT. In step S6, the processing unit 104 calculates the average acoustic energy between the microphone 103 and the laser sensor 107 from the frequency characteristics. In step S7, the processing unit 104 determines the presence or absence of an abnormality in the inspection object O and the extent of the abnormality, for example, by comparing the calculated average acoustic energy with a threshold value. In step S8, the processing unit 104 outputs the determination result on the presence or absence of an abnormality and the degree of its progression to, for example, the display 106 as the abnormality diagnosis result.

[0066] Since the laser sensor 107 measures vibration at one point on the panel, it is not necessarily able to capture vibrations caused by deterioration such as weak cracks that occur locally on the vibrating surface of the panel. To diagnose the presence or absence of abnormalities using only the laser sensor 107, a mechanism for scanning the entire panel with the laser sensor 107 is required. Similarly, when using a contact-type acceleration sensor, it is essential to measure at multiple points in order to diagnose the presence or absence of abnormalities.

[0067] Therefore, in the embodiment, the impulse response between two points, the laser vibration and the microphone sound pressure, is measured. Then, by removing the acoustic excitation wave with the maximum peak from the impulse response, it is possible to reproduce vibration information in a wider area than one point acquired by the laser. In this way, it becomes possible to detect weak vibrations that are difficult to acquire by one point measurement using a laser.

[0068] (effect) 10 is a diagram showing an example of a graph obtained in an experiment to verify the effect of the embodiment. For the verification experiment, an abnormal weak vibration model was created that cannot be measured by the laser sensor 107 but can be measured by the microphone 103. This model was created by creating a simulated signal (weak vibration radiation sound) and adding it only to the actually measured microphone signal.

[0069] Using the above model, a vibration measurement signal was obtained by the laser sensor 107, a sound pressure signal was obtained by the microphone 103, and a mutual impulse response was calculated. By removing the maximum peak component from this impulse response, the micro-vibrations buried in the acoustic excitation were reproduced. Furthermore, when a simulated signal was added, the signal band of the micro-vibrations increased, and when the vibration was stopped (simulated normal), the signal band of the micro-vibrations disappeared. This enabled us to confirm the operating principle of the measurement device of the embodiment.

[0070] 11 and 12 are diagrams showing other examples of graphs obtained in an experiment to verify the effect of the embodiment. The ratio between the attenuation rate of the vibration field in one-point measurement by the laser sensor 107 and the attenuation rate of the vibration field generated by the vibration radiation sound of the entire object collected as microphone sound pressure is reflected in the two-point impulse response of the microphone sound pressure to the laser vibration. This was verified by sticking tape on the back surface of the inspection object O and comparing the difference in vibration attenuation with that without the tape.

[0071] Figure 11 shows the waveforms after removing the excitation wave from the impulse response between two microphones, comparing the case without tape and the case with tape (three pieces). The impulse response G is expressed by equation (14).

[0072]

number

[0073] The graph in Figure 11 shows almost the same resonant frequency characteristics in both cases, making it difficult to find any differences. Therefore, the vibration radiation wave cannot be clearly captured.

[0074] FIG. 12 shows a comparison of waveforms obtained after removing the excitation wave from the two-point impulse response of the vibration measured by the laser sensor and the sound pressure collected by the microphone in the embodiment, between the case without tape and the case with tape.

[0075] If the damping ratio in the case without tape is γ1 and the impulse response is G1, then equation (15) holds.

[0076]

number

[0077] If the attenuation ratio in the case with tape is γ2 and the impulse response is G2, then equation (16) holds.

[0078]

number

[0079] 12 clearly shows the decay of the vibration radiation wave, and this confirmed the effectiveness of the method of the embodiment for analyzing the impulse response between two points, the laser sensor 107 and the microphone. The impulse response G when the damping ratio due to the presence or absence of tape is directly reflected in the waveform is expressed by Equation (17).

[0080]

number

[0081] As described above, according to the embodiment, in an inspection device for inspecting an object for abnormality, vibration information obtained by measuring the vibration of the object for inspection with one laser sensor and vibration information by radiated sound collected from the object for inspection with one microphone are acquired. As a result, the presence or absence of a crack and its progress appear as a change in the transfer function between the laser sensor and the microphone, and therefore the presence or absence of a crack and the degree of its progress can be determined with a simple configuration that uses only the laser sensor and the microphone.

[0082] By exciting weak vibrations more strongly with acoustic excitation, measuring the spatial transfer function (impulse response) with two microphones, and removing the excitation wave with the maximum peak component that appears there, it is possible to extract weak vibrations buried in the excitation wave. With this method, when the entire target surface is vibrating in a resonance mode, such as a flat plate resonance vibration, it is difficult to detect changes in attenuation of the vibration itself.

[0083] On the other hand, there is a method that uses a laser to directly measure local changes in vibrating parts due to abnormalities such as peeling. However, laser vibration can only obtain vibration information at one point on the surface of a structure. Microphone sound pressure has the advantage of being able to record all sounds emitted from vibrating parts of the entire object at one point, but when each sound is acquired separately, noise such as disturbance effects is also added. In particular, when detecting weak vibration emission sounds, it is difficult to expect high accuracy unless noise is removed.

[0084] Therefore, in the embodiment, the excitation unit (excitation sound source 101) applies an excitation signal to the structure to vibrate the entire target part. Then, the acoustic excitation wave is removed from the initial maximum peak of the impulse response calculated from the vibration of one point on the surface of the inspection object obtained by the vibration sensor (laser sensor 107) and the microphone sound pressure. This makes it possible to extract the weak vibration radiation sound that was buried, and to judge the change in vibration characteristics caused by deterioration of the structure (reduction in axial force, cracks, deterioration of vibration control, etc.) from the frequency characteristics, thereby diagnosing abnormalities in the structure. In this way, it is possible to supplement useful vibration information that could not be detected by the two-microphone method.

[0085] Moreover, while the two-microphone system requires strict setting of the installation positions of the two microphones, the positional relationship between the microphone 103 and the laser sensor 107 in this embodiment is not sensitive, which further simplifies the work procedure.

[0086] In other words, the placement of the two microphones is determined by the distance from the position of each divided element point of the weak vibration to the microphone and the volume velocity of each point. Even if the scale of the weak vibration is constant, the results will vary depending on this and the distance to the reference microphone. In an actual site, the scale and position of the weak vibration generation are unknown, so it is difficult to derive the optimal microphone spacing.

[0087] In contrast, in the embodiment, the object can be inspected from data collected separately by the microphone and the laser sensor, so there is no need to consider the positions of the microphone and the laser sensor. This eliminates the need for trial and error regarding the installation positions of the microphone and the laser sensor and the distance between them, greatly simplifying the work procedure. To increase the S / N ratio, it is advantageous to place the microphone as close as possible, and simply moving the microphone closer to the object simplifies the system design.

[0088] Furthermore, microphones tend to have lower measurement resolution in the vicinity of weak vibrations than laser sensors. On the other hand, even if weak vibrations occur far from the microphone's installation position, the laser sensor in front of the microphone can detect the weak vibrations with a high S / N ratio. This information can be used to detect the presence or absence of weak vibrations. In other words, by using a vibration sensor, it is possible to detect weak vibrations far from the microphone, and higher measurement resolution can be expected than with methods using multiple microphones.

[0089] <About real-time processing> Here, a process for calculating an impulse response in real time will be described. For example, an inspection vehicle is known that irradiates a detection sound toward a road from a traveling vehicle and collects acoustic information including a reflected sound with a microphone to inspect the road. In this type of technical field, a technology that can calculate an impulse response in real time is highly demanded.

[0090] Well-known methods for calculating an impulse response from two signals are adaptive filter processing and cross-correlation processing. Of these, cross-correlation processing is advantageous for real-time processing.

[0091] FIG. 13 is a diagram for explaining real-time cross-correlation processing. For example, consider collecting acoustic signals using two microphones. The acoustic signals sampled at each point of AD by the microphone of channel 1 (ch1) and the microphone of channel 2 (ch2) are arranged in time series (n, n-1, ...) and stored in a memory (not shown). Then, from the data of each channel, positive correlation data between the channels based on the current time n is obtained, and cross-correlation in time series is calculated sequentially. By such processing, a cross-correlation output that changes from moment to moment with the movement of the microphone is obtained, and the impulse response between the two signals can be calculated in real time.

[0092] FIG. 14 is a waveform diagram showing an example of an impulse response obtained in real time. The impulse response includes a waveform of an excitation sound indicated by a dotted line and a waveform of a vibration-radiated sound indicated by a dashed line. The waveform of the excitation sound includes a first peak and a second peak. The level of the first peak is a function of the distance between the microphone of channel 1 and the microphone of channel 2. The appearance position of the second peak is a function of the distance from the starting point of the first peak between the object to be inspected and a reference microphone (for example, the one closer to the object to be inspected: channel 1).

[0093] Therefore, in real-time processing, the area from the start of the first peak to the start of the second peak is treated as the processing target (detection window) as shown in the dotted rectangular area in Fig. 14. Then, as described in the embodiment, the removal of the unnecessary component, which is the first peak, is applied to this section, and frequency analysis is performed. This type of processing makes it possible to detect the presence or absence of an abnormality by capturing faint sounds in real time.

[0094] Of course, the cross-correlation processing can be similarly applied even when the system is stopped. In other words, according to the embodiment, by using the cross-correlation processing, it is possible to provide an acoustic inspection technique that can be applied to both batch processing and real-time processing.

[0095] Furthermore, real-time processing can be performed by applying cross-correlation processing to impulse responses obtained from information collected separately by a vibration sensor and a microphone, not limited to impulse responses obtained from two microphones. In other words, the microphone and the vibration sensor can be regarded as a first sensor that outputs a first signal and a second sensor that outputs a second signal, respectively.

[0096] Therefore, the processing target is the range from the start of the first peak to the start of the second peak from the impulse response obtained by the cross-correlation between the first signal and the second signal. The first peak is found by the distance between the first sensor and the second sensor, and the start of the second peak is found by the time width according to the distance between the object to be inspected and the reference sensor. Then, by performing frequency analysis on the impulse response waveform from which the first peak has been removed, real-time non-destructive inspection can be realized regardless of the type of sensor.

[0097] It should be noted that the present invention is not limited to the above-mentioned embodiment. For example, instead of the vibration sound source 101 and the speaker 102 in Fig. 1, a striking mechanism that physically strikes the inspection object O may be used. Also, instead of the laser sensor 107, a contact sensor that is attached to the inspection object O and collects the vibration of the inspection object O may be used. For example, a small hammer, an acceleration sensor, a small microphone, and a small terminal may be packaged together to simulate a hand-carry measurement.

[0098] Although an embodiment of the present invention has been described, this embodiment is presented as an example and is not intended to limit the scope of the invention. This embodiment can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. This embodiment and its modifications are included in the scope of the invention and its equivalents described in the claims, as well as in the scope and gist of the invention. [Explanation of symbols]

[0099] 101... vibration sound source, 102... speaker, 103... microphone, 104... processing unit, 105... memory, 106... display, 107... laser sensor, 1041... impulse response calculation unit, 1042... removal unit, 1043... frequency conversion unit, 1044... energy calculation unit, 1045... abnormality determination unit.

Claims

1. a vibration sensor that detects vibration of an object to be inspected that has been excited to vibrate; A microphone arranged near the object to be inspected to collect radiated sound from the object to be inspected; an impulse response calculation unit that calculates an impulse response between the vibration sensor and the microphone based on the vibration detected by the vibration sensor and the sound pressure of the radiated sound collected by the microphone; A removal unit that removes unnecessary components from the impulse response; a frequency conversion unit that converts the impulse response from which the unnecessary components have been removed into a frequency characteristic; an energy calculation unit that calculates acoustic energy between the vibration sensor and the microphone based on the frequency characteristics; and an abnormality determination unit that determines whether or not the object to be inspected has an abnormality based on the acoustic energy.

2. The inspection device according to claim 1 , wherein the impulse response calculation section calculates the impulse response based on the detected vibration.

3. The inspection device according to claim 1 , wherein the removal unit removes a component of the impulse response that corresponds to a time width according to a distance between a vibration surface of the inspection object and the microphone.

4. The inspection device according to claim 1 , further comprising a vibration unit for exciting vibrations in the object to be inspected.

5. The inspection device according to claim 4 , wherein the vibration unit is a sound source that irradiates a vibration sound toward the object to be inspected.

6. The inspection device according to claim 4 , wherein the vibration unit is a striking mechanism that strikes the object to be inspected.

7. 2. The inspection device according to claim 1, wherein the vibration sensor is a non-contact sensor that measures vibration of the inspection object based on a Doppler shift of laser light irradiated onto the inspection object.

8. The inspection device according to claim 1 , wherein the vibration sensor is a contact sensor attached to the inspection object to collect vibrations of the inspection object.

9. The vibration unit excites vibration in the object to be inspected; A vibration sensor detects vibration of the inspection object; A microphone disposed near the test object collects a sound radiated from the test object; A processor calculates an impulse response between the vibration sensor and the microphone based on the vibration detected by the vibration sensor and the sound pressure of the radiated sound collected by the microphone; the processor removing unwanted components from the impulse response; The processor converts the impulse response from which the unnecessary components have been removed into a frequency characteristic; the processor calculating acoustic energy between the vibration sensor and the microphone based on the frequency characteristics; and determining, by the processor, the presence or absence of an abnormality in the object to be inspected based on the acoustic energy.

Images