Method and device for inspection display of vehicle

Abstract

To determine whether sound from a specific site is normal in the process of inspecting a finished automobile and specify the generation source if the sound is abnormal.SOLUTION: The inspection display device includes a sound acquisition unit 1, a vibration acquisition unit 10, an abnormal sound determination unit 20, an abnormal sound vibration source estimation unit 30, a data storage unit 40, and a display unit 50. A directional microphone of the sound acquisition unit 1 and a laser doppler vibration meter of the vibration acquisition unit 10 are arranged to point to the specific site. When the acquired sound is determined to be abnormal, vibration data of a corresponding predetermined time length is acquired and the possibility that the specific site is the generation source of the abnormal sound is determined from the correlation between the amplitudes and the consistency degree of the frequencies.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method and apparatus for determining whether a sound produced from a specific part of a vehicle such as an automobile is a normal sound during an inspection process of the vehicle, and, if the sound is abnormal, for identifying the source of the sound and displaying the result. [Background technology]

[0002] For example, in the finished vehicle inspection process, which is the final stage of an automobile production line, an inspector test drives the finished vehicle to be inspected on free rollers and inspects many items such as the engine, meters, brakes, horn, lights, etc. In general, during this inspection process, the presence or absence of abnormalities is determined for various sounds emitted from each part of the vehicle through so-called sensory evaluation based on the inspector's senses. For example, the sensory inspection determines whether sounds generated from the engine or transmission during acceleration, or from the brakes during braking, are normal or not. It is very difficult to identify the source of abnormal sounds through sensory inspection using the inspector's ears.

[0003] Instead of such sensory testing, attempts have been made to detect abnormalities by acquiring sounds or vibrations using a microphone or a sensor and analyzing the signals. For example, Patent Document 1 describes an analysis method in which, in order to identify the source of noise inside the vehicle caused by an electric power steering device that can generate vibrations at multiple locations during steering, a microphone is placed inside the vehicle and vibration sensors made of acceleration sensors are attached to multiple locations assumed to be noise sources of the electric power steering device, and which location of the vibration is the noise source is identified from the correlation between the waveform of the noise collected by the microphone and the detection signal of each vibration sensor. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2012-215412 A Summary of the Invention [Problem to be solved by the invention]

[0005] The technology of Patent Document 1 requires vibration sensors to be placed in advance at multiple candidate locations, and is a technology that can be applied to identify noise sources in so-called experimental vehicles. However, it cannot be applied to sound inspection in the above-mentioned finished vehicle inspection process, where the installation of vibration sensors is not possible. [Means for solving the problem]

[0006] The vehicle inspection and display method according to the present invention comprises the steps of: Sound is picked up by a microphone aimed at a specific part of the vehicle being inspected. In parallel with the measurement of the sound, a non-contact vibration sensor is oriented toward the specific part to obtain the vibration of the specific part, Determine whether the acquired sound is normal or abnormal, Identifying the specific part as the source of the abnormal sound based on the correlation between the sound determined to be the abnormal sound and the corresponding vibration, Display the results. Effect of the Invention

[0007] According to this invention, by acquiring sound and vibration via a microphone and a non-contact vibration sensor facing the same direction from outside the vehicle, it is possible to easily inspect and display whether a sound coming from a specific location is an abnormal noise, and whether this abnormal noise is indeed originating from that specific location. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a functional block diagram of a first embodiment in which the present invention is applied to abnormal sound inspection in a finished vehicle inspection process. [Diagram 2] 4 is a flowchart showing the flow of processing in the first embodiment. [Diagram 3] FIG. 4 is an explanatory diagram showing a display example on a display unit according to the first embodiment. [Figure 4]FIG. 11 is a functional block diagram of a second embodiment. [Diagram 5] FIG. 11 is an explanatory diagram of a sound and vibration measurement set according to a second embodiment. [Figure 6] 10 is a flowchart showing the flow of processing according to a second embodiment. [Figure 7] FIG. 11 is an explanatory diagram showing a display example on a display unit according to a second embodiment. [Figure 8] An explanatory diagram of a sound and vibration measurement set that allows the distance to the vehicle to be changed. [Figure 9] FIG. 13 is an explanatory diagram of another example of a sound and vibration measurement set that allows the distance to the vehicle to be changed. [Figure 10] An explanatory diagram of a sound and vibration measurement set with multiple devices arranged in a planar shape. [Figure 11] FIG. 13 is a functional block diagram of a third embodiment. [Figure 12] 13 is a flowchart showing the flow of processing according to a third embodiment. [Figure 13] FIG. 13 is an explanatory diagram showing a display example on a display unit according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment of the present invention will be described in which the present invention is applied to an abnormal sound inspection in a finished vehicle inspection process for automobiles. In general, in a finished vehicle inspection process, an inspector performs a test run of the finished vehicle to be inspected on a free roller, and inspects many items including the engine, meters, brakes, etc. In the finished vehicle inspection, an inspection is performed to check whether the sound generated from a specific part of the vehicle includes an abnormal sound. As an example, the outer surface of the brake disc or the brake caliper is set as the specific part, and an inspection is performed to check whether the sound generated during braking includes an abnormal sound. Alternatively, the present invention can be applied to other inspections such as an inspection to check whether the sound generated from the engine during acceleration includes an abnormal sound, with an appropriate part of the engine visible from inside a pit under the vehicle as the specific part. In particular, when an abnormal sound is judged to be present, the position of the source of the sound is to be identified. The inspection display device of each embodiment is configured as a part of a finished vehicle inspection device.

[0010] 1 shows a functional block diagram of the inspection display device of the first embodiment. The inspection display device of the first embodiment includes a sound acquisition unit 1, a vibration acquisition unit 10, an abnormal sound determination unit 20, an abnormal sound vibration source estimation unit 30, a data storage unit 40, and a display unit 50.

[0011] The sound acquisition unit 1 includes a microphone that acquires sounds generated from the vehicle under test and converts them into electrical signals, i.e., sound data, and a buffer that temporarily stores the sound data. The microphone has high directivity so that it can collect sounds generated near specific locations, such as the above-mentioned brake discs and brake calipers, and is placed outside the vehicle in a position that points toward the specific location. The frequency characteristics of the microphone are selected according to the measurement target.

[0012] The vibration acquisition unit 10 is composed of a non-contact vibration sensor capable of detecting local vibration, and acquires vibrations in a specific portion, for example, the above-mentioned brake disc or brake caliper, in parallel with the measurement of sound by a microphone. As the non-contact vibration sensor, a laser Doppler vibrometer or a laser displacement meter can be used, and in one embodiment, a laser Doppler vibrometer is used. The laser Doppler vibrometer is arranged adjacent to a directional microphone, and is combined with the microphone so that the directional directions of both are the same. The laser Doppler vibrometer can measure the magnitude (displacement amount, acceleration, etc.) and frequency of vibration of a local target portion by irradiating a laser beam to the target portion. With the laser Doppler vibrometer, the vibration measurement position can be detected as the laser irradiation position by visual inspection or via an appropriate means, so that it is easy to direct the laser beam together with the microphone to the specific portion to be inspected.

[0013] The abnormal sound determination unit 20 cuts out a sound of a predetermined length from the continuous sound acquired by the sound acquisition unit 1, determines whether the sound is normal or abnormal (abnormal), and extracts sound data including the abnormal sound. A sound of a predetermined length can be cut out using as a trigger a vehicle signal (such as a CAN signal) indicating a specific vehicle state, a control signal of an inspection device such as a free roller, a predetermined action of an inspector, or a sound feature amount exceeding a predetermined threshold value, etc. In an example of inspecting the sound when the brakes are applied, the inspector detects that the brake pedal is depressed, and cuts out sound data of a length of several seconds before and after that, and determines whether the sound is normal or abnormal.

[0014] As a trigger for extracting sound data, for example, a condition such as the vehicle speed reaching 100 km / h, the engine speed reaching 3000 rpm, an inspector pressing / releasing a switch of a specific device of the vehicle, etc. Alternatively, a condition such as a sound pressure level of 80 dBA or more at 1 kHz may be set as a threshold value as a sound feature, and when a sound corresponding to this condition is detected, sound data having a duration of about several seconds before and after the detection may be extracted.

[0015] The determination of whether a sound is normal or abnormal may be made by any known method. For example, the normal state of a sound may be learned and quantified using a deep neural network (DNN) or the like, the amount of deviation from that value may be expressed as the degree of abnormality, and the degree of abnormality may be compared with a threshold value to determine whether the sound is normal or abnormal.

[0016] In addition, in order to determine whether a sound is normal or abnormal, abnormal sound quality may be quantified using the characteristics of the abnormal sound, such as the harmonic characteristics of the sound and their frequency fluctuation characteristics, the cepstrum envelope and its slope, and quefrequency characteristics.

[0017] Then, using these sound features, supervised learning can be used, in which a large amount of collected sound data on normal and abnormal sounds is learned using DNNs or other methods to generate a supervised learning model of normal and abnormal sounds and used to determine whether the sound is normal or abnormal, or unsupervised learning can be used, in which only the normal state is learned and the magnitude of the difference from that result is used to determine whether the sound is normal or abnormal.

[0018] Alternatively, a statistical amount such as the MT method may be used to determine whether a sound is normal or abnormal based on an MD value corresponding to the level of significance, or a simple threshold value may be used to determine whether a sound is normal or abnormal.

[0019] When the abnormal sound determination unit 20 determines that certain sound data is an abnormal sound, the abnormal sound vibration source estimation unit 30 determines that a specific part from which vibration was acquired is the source of the abnormal sound based on the correlation between the sound determined to be an abnormal sound and the vibration corresponding to the sound. Vibration data acquired by a laser Doppler vibrometer or the like in parallel with the sound measurement via a microphone is temporarily stored in a buffer included in the vibration acquisition unit 10, and is extracted as vibration data of the same time length as the sound data, for example, and the correlation between the two is determined.

[0020] This correlation is estimated from, for example, the correlation between the magnitude of sound and vibration, the degree of agreement of frequency characteristics, etc. For example, for sound data, the maximum sound pressure level Spre and frequency FSpre of the sound before the occurrence of the sound regarded as an abnormal sound, and the maximum sound pressure level Spost and frequency FSpost when the sound regarded as an abnormal sound occurs are obtained, and the difference Com_s (=Spost-Spre) between the maximum sound pressure levels before and after the occurrence is calculated. Similarly, for vibration data, the maximum vibration level Vpre and frequency FVpre before the occurrence of the sound regarded as an abnormal sound, and the maximum vibration level Vpost and frequency FVpost when the sound regarded as an abnormal sound occurs are obtained, and the difference Com_v (=Vpost-Vpre) between the maximum amplitude levels before and after the occurrence is calculated. Then, whether or not it is a vibration source (i.e., a source of generation) of the abnormal sound is determined based on the correlation α between the difference Com_s between the maximum sound pressure levels before and after the occurrence and the difference Com_v between the maximum amplitude levels, and the degree of agreement Δ between the frequency characteristics of the sound and vibration.

[0021] The correlation α is calculated by the following formula.

[0022] α= (Com_v / Com_s) = (Vpost-Vpre) / (Spost-Spre) ···(1) The degree of coincidence Δ of frequency characteristics is calculated as follows using the frequencies FSpost, FSpre of the maximum sound pressure levels and the frequencies FVpost, FVpre of the maximum vibration levels.

[0023] Δ1=(n * FVpost) / (m * FSpost) ···(2) where n and m are arbitrary integers and are selected so that Δ1 approaches 1.

[0024] Δ2=(n * FVpre) / (m * FSpre) ···(3) where n and m are arbitrary integers and are selected so that Δ2 approaches 1.

[0025] That is, the degree of coincidence Δ is determined taking into account overtones.

[0026] It is also possible to treat only Δ1 as the degree of coincidence Δ and omit Δ2.

[0027] Specific numerical examples of the correlation α and the degree of agreement Δ will be described later.

[0028] The data storage unit 40 is a storage unit consisting of an HDD or the like for storing data acquired by the sound acquisition unit 1 and the vibration acquisition unit 10, and data such as the results of calculations or judgments by the abnormal sound judgment unit 20 and the abnormal sound vibration source estimation unit 30, and is configured by a server computer included in the finished vehicle inspection device, a cloud computer connected via a communication line, etc. The data storage unit 40 stores acquired data related to the inspection of one specific part, various calculated values, judgment results, display contents, and other data as one data set.

[0029] The display unit 50 is a means for displaying acquired data, inspection results, and other information to relevant parties such as an inspector driving the vehicle and a manufacturing manager, and is configured, for example, with a liquid crystal display, an organic EL display, various indicator lights, etc. It is not limited to being a visual device, and may be a device that displays audio information, a haptic device that presents tactile information to the hand or arm, etc.

[0030] FIG. 3 is an explanatory diagram showing a display example on the display unit 50 of the first embodiment. In this example, it is written at the top that the sound acquired from a specific part such as a brake disc is determined to be an abnormal sound, and that the specific part such as a brake disc to which the laser Doppler vibrometer is directed is highly likely to be the vibration source of the abnormal sound, and the acquired or calculated data values ​​are displayed in a tabular format. The "Amplitude" column in the table lists the values ​​of the maximum sound pressure levels Spre, Spost and the maximum vibration levels Vpre, Vpost described above, and the "Fundamental Frequency / Natural Frequency" column lists the values ​​of the frequencies FSpost, FVpost. In the table, "before" corresponds to the subscript "pre", and "after" corresponds to "post". The "Determination Result" row shows the presence or absence of correlation of the amplitude (the value of correlation α) and the degree of agreement Δ of the frequency characteristics (here, the above-mentioned Δ2 is omitted and only Δ1 is used). "Δ=1" means that the frequency characteristics are identical.

[0031] Next, Fig. 2 shows a flow chart of the process flow of the inspection display device of the first embodiment described above. First, in step 1, the microphone of the sound acquisition unit 1 collects sounds from a specific part, such as the brake disc or brake caliper, from a vehicle test-running on a free roller, and a sound of a predetermined length is cut out and acquired as sound data, for example, triggered by an inspector stepping on the brake pedal. Next, an abnormal sound determination process is performed to determine whether the sound included in this sound data is a normal sound or an abnormal sound (step 2). In step 3, it is determined whether the determination result is a normal sound or an abnormal sound, and if it is a normal sound, the process returns to step 1 and the inspection is repeated for the next vehicle or another specific part.

[0032] If it is determined that the sound is an abnormal sound, the process proceeds from step 3 to step 4, where vibration data of a predetermined length of time corresponding to the sound data is acquired from a buffer included in the vibration acquisition unit 10. Then, in step 5, the source of the abnormal sound (vibration source) is identified based on the correlation and degree of agreement between the vibration and the sound. In other words, the possibility that the specific part from which the vibration was acquired is the source of the abnormal sound is determined.

[0033] Next, in step 6, the results and the like are displayed on the display unit 50 and the data is stored in the data storage unit 40. In step 7, it is determined whether or not to continue the inspection, and if YES, the process returns to step 1 to continue the inspection of the next vehicle or other specific parts.

[0034] Next, a specific example of determining the correlation between sound and vibration will be described using the numerical values ​​shown in Fig. 3. In the example of Fig. 3, for sound, the maximum sound pressure level in the period (e.g., 3 seconds) immediately before it is determined to be an abnormal sound is 50 dBA, and the maximum sound pressure level in the period (e.g., 3 seconds) after it is determined to be an abnormal sound is 80 dBA, and the frequency at that time is 2 kHz. Similarly, for vibration, the maximum vibration levels before and after are 25 dB and 50 dB, respectively, and the frequency is 20 Hz.

[0035] The correlation α for the amplitude is given by the above-mentioned formula (1): α=(50-25) / (80-50)=25 / 30=0.834 In this case, the amplitude of both the sound and the vibration is large when it is judged to be an abnormal sound, so there is a correlation. If the amplitude of the vibration is not large, the value of the correlation α will be small, so there is no correlation. Generally, if the value of α is in the range of 0.5 to 2, there is a correlation.

[0036] The frequency matching degree Δ takes into account the case of harmonics. If n = 1 and m = 100 in the above formula (2), then Δ=(1*2000) / (100*20)=1 Therefore, it is determined that the frequencies of the sound and vibration match each other.

[0037] If we assume that the sound frequency is 2010 Hz, then Δ = 1.005, which is not 1, so we make an initial judgment that the frequencies of the sound and vibration do not match.

[0038] In this case, the degree of deviation ΔΔ, which indicates the degree of deviation between the two frequencies with respect to the vibration frequency (20 Hz in the example of FIG. 3), is further calculated, and the final judgment (secondary judgment) of the degree of match is made based on the magnitude of this degree of deviation ΔΔ. In other words, the degree of deviation ΔΔ is calculated as follows.

[0039] ΔΔ=ABS((1*2010)-(100*20)) / 20=10 / 20=0.5 The threshold value for the deviation ΔΔ in the secondary judgment can be set to, for example, about 0.1 (i.e., 10% error), and if the deviation ΔΔ is equal to or less than the threshold value, it is finally judged that the frequencies of the sound and vibration match.

[0040] For example, if the deviation threshold is 0.1, and the sound frequency (FSpost) is in the range of 1998-2002 (Hz), it is determined that the sound and vibration frequencies match. In the above example, in the case of 2010 Hz, the deviation ΔΔ is 0.5, and therefore it is ultimately determined that the sound and vibration frequencies do not match.

[0041] In the example of FIG. 3, the correlation α is 0.83 and the frequency coincidence Δ is 1, so that a specific part whose vibration was measured using the laser Doppler vibrometer, such as a brake disc or brake caliper, is identified as the source of the abnormal noise.

[0042] As described above, according to the first embodiment, by combining a laser Doppler vibrometer capable of measuring local vibrations without contact with a directional microphone, it is possible to acquire sound and vibration from a specific part, such as a brake disc, in parallel, and thus easily verify that the sound coming from the specific part is an abnormal sound, and that this abnormal sound is indeed originating from the specific part.

[0043] Next, an inspection and display device of a second embodiment will be described with reference to Fig. 4 to Fig. 7. The following mainly describes the differences from the first embodiment. The second embodiment is intended to improve the accuracy of estimating that a specific part is the source of an abnormal sound by utilizing the characteristic that sound attenuates with distance.

[0044] 4 shows a functional block diagram of the inspection display device of the second embodiment. As in the first embodiment, the inspection display device of the second embodiment includes a sound acquisition unit 1, a vibration acquisition unit 10, an abnormal sound determination unit 20, an abnormal sound vibration source estimation unit 30, a data storage unit 40, and a display unit 50, and further includes a distance measurement unit 60 that measures the distance between the microphone of the sound acquisition unit 1 and a specific part to be measured.

[0045] The distance measuring unit 60 desirably uses a non-contact distance measuring device such as a laser range finder, but may be of any type, and may be something like a simple tape measure.

[0046] 5 shows, as an example, a sound / vibration measurement set 100 in which a microphone 101 of the sound acquisition unit 1, a laser Doppler vibrometer 102 of the vibration acquisition unit 10, and a laser distance meter 103 of the distance measurement unit 60 are all mounted on one bracket 110. The distance L from the microphone 101 to a specific part can be changed by moving the sound / vibration measurement set 100 closer to or farther away from the vehicle. For example, measurements can be performed at two points, a distance L1 of 100 cm and a distance L2 of 50 cm.

[0047] Next, Fig. 6 shows a flow chart of the process flow of the inspection display device of the second embodiment. First, in step 1, the microphone of the sound acquisition unit 1 collects sounds from a specific part, such as the brake disc or brake caliper, from a vehicle test-running on a free roller, and a sound of a predetermined length is cut out and acquired as sound data, for example, triggered by an inspector stepping on the brake pedal. Next, an abnormal sound determination process is performed to determine whether the sound included in this sound data is a normal sound or an abnormal sound (step 2). In step 3, it is determined whether the determination result is a normal sound or an abnormal sound, and if it is a normal sound, the process returns to step 1 and the inspection is repeated for the next vehicle or another specific part.

[0048] If it is determined that the sound is an abnormal sound, the process proceeds from step 3 to step 4, and vibration data of a predetermined length of time corresponding to the sound data is acquired from a buffer included in the vibration acquisition unit 10.

[0049] Next, proceed to step 11 and obtain information on the distance L from the microphone 101 to the specific part. Then, in step 12, determine whether there is sound data and vibration data obtained for the same specific part at different distances L. If NO, return to step 1, change the distance L, and repeat the inspection.

[0050] If the answer is YES in step 12, proceed to step 13, and identify the source of the abnormal sound using at least two measurement results with different distances from each other. In other words, determine the possibility that the specific part being measured is the source of the abnormal sound.

[0051] Then, in step 6, the results and the like are displayed on the display unit 50 and the data is stored in the data storage unit 40. In step 7, it is determined whether or not to continue the inspection, and if YES, the process returns to step 1 to continue the inspection of the next vehicle or other specific parts.

[0052] FIG. 7 is an explanatory diagram showing a display example on the display unit 50 of the second embodiment. The display of this example is displayed in addition to the display shown in FIG. 3 showing the results at each distance. In the display example of FIG. 7, it is written at the top that the vibration source of the abnormal sound is the wheel cover part, and the probability that this is the source of the abnormal sound is 83%, and the waveforms of the sound data and vibration data acquired when the distance is 50 cm and the waveforms of the sound data and vibration data acquired when the distance is 100 cm are shown in a table. In addition, the "Result" column of the table displays the probability that the sound and vibration acquired when the distance is 50 cm are the same as the sound and vibration acquired when the distance is 100 cm.

[0053] As is well known, the loudness of a sound is inversely proportional to the square of the distance. Therefore, when distance L is 100 cm, the amplitude of the sound is 1 / 4 of that when distance L is 50 cm. If this relationship holds, the probability that it is the same sound (sound from the same source) is set to 100%. If the amplitude does not change, there is a high possibility that the sound source is incorrect, and the probability that it is the same sound is displayed as a low value.

[0054] On the other hand, since the magnitude of the vibration amplitude does not depend on the distance, the amplitude when the distance L is 50 cm should be equal to that when it is 100 cm. Therefore, confirm that the ratio of the two amplitudes is 1 and there is no change. If the vibration amplitude increases or decreases, the amount of change is displayed as the probability of it being the same vibration. In the example in Figure 7, the change in distance L has caused the vibration amplitude to increase by 0.98 times, so the probability that it is the same vibration is displayed as 98%.

[0055] For each of the cases where the distance L is 50 cm and 100 cm, the source of the abnormal sound is identified by the same process as in the first embodiment. In the second embodiment, the reliability of identifying the source of the abnormal sound can be further increased by comparing the data for the two distances.

[0056] FIG. 8 shows another example of the configuration of the sound / vibration measurement set 100. In this example, a bracket 110 on which a microphone 101, a laser Doppler vibrometer 102, and a laser distance meter 103 are mounted is configured to be movable along a guide rail 111. The guide rail 111 extends along the direction of orientation of the laser Doppler vibrometer 102 and the like. The bracket 110 moves forward and backward by an appropriate method, either electrically or manually. For example, the guide rail 111 may be formed with teeth, and the bracket 110 may be configured to move forward and backward by a motor as a so-called rack and pinion mechanism. Alternatively, the guide rail 111 may be formed in a cylindrical shape having a screw thread, and the bracket 110 may be configured to move forward and backward by the rotation of the guide rail 111 as a so-called screw mechanism.

[0057] With this configuration, inspections at positions with different distances L can be performed continuously and efficiently.

[0058] In the example of FIG. 9, the position of the laser range finder 103 is fixed, and a bracket 110 having a microphone 101 and a laser Doppler vibrometer 102 is configured to move back and forth.

[0059] In addition, a sound / vibration measurement set 100 including a microphone 101 and a laser Doppler vibrometer 102 may be placed in advance at least two locations at different distances L from a specific part, and sound and vibration may be measured simultaneously at the two locations.

[0060] 10 shows a different configuration example of the sound / vibration measurement set. In this example, a plurality of sound / vibration measurement sets 100, each including a microphone 101 and a laser Doppler vibrometer 102, are arranged in a plane. More specifically, a plurality of sound / vibration measurement sets 100 are arranged along a surface facing the surface of a specific part to be measured, and each of the microphones 101 and the laser Doppler vibrometers 102 is directed to a plurality of points constituting the surface of the specific part. The plurality of sound / vibration measurement sets 100 can move forward and backward in unison along a guide rail 111 to change the distance between the specific part and the sound / vibration measurement sets 100. This distance is measured by a single laser distance meter 103.

[0061] In the figure, four sound / vibration measurement sets 100 are shown aligned in a row, but in reality, they are arranged in a 2×2 configuration.

[0062] In this way, by arranging the microphone 101 and the laser Doppler vibrometer 102 in a planar manner to acquire sound and vibration, the reliability of identifying the source of the abnormal sound is increased.

[0063] When the specific portion has an uneven shape, the positions of the multiple sound / vibration measurement sets 100 may be shifted so that the distance between each measurement point is the same. Alternatively, each sound / vibration measurement set 100 may include a laser distance meter 103 to measure the distance to each measurement point.

[0064] Next, an inspection display device of a third embodiment will be described with reference to Figures 11 to 13. In the third embodiment, the same processing as in the first embodiment is performed on sound data deemed to be an abnormal sound and the corresponding vibration data, and the same processing as in the first embodiment is also performed on sound data and vibration data deemed to be normal sounds, and the results of both are compared to prevent errors in determining whether an abnormal sound exists or identifying the source of the abnormal sound.

[0065] 11 shows a functional block diagram of the inspection display device of the third embodiment. As in the second embodiment, the inspection display device of the third embodiment includes a sound acquisition unit 1, a vibration acquisition unit 10, an abnormal sound determination unit 20, an abnormal sound vibration source estimation unit 30, a data storage unit 40, a display unit 50, and a distance measurement unit 60, and further includes a normal sound determination unit 21, a normal sound vibration source estimation unit 31, and an estimation result comparison unit 80.

[0066] The normal sound determination unit 21, like the abnormal sound determination unit 20 described above, cuts out a sound of a predetermined length from the continuous sound acquired by the sound acquisition unit 1, determines whether the sound is normal or abnormal (abnormal) and extracts sound data deemed to be normal. The method of cutting out a sound of a predetermined length and the method of determining whether the sound is normal or abnormal can be similarly applied to those described for the abnormal sound determination unit 20. In an example of inspecting sounds during braking, the inspector detects that the brake pedal is depressed, cuts out sound data of a length of several seconds before and after the detection, and determines whether the sound is normal or abnormal.

[0067] In the third embodiment, at least two inspections are required, one to obtain sound data deemed to be an abnormal sound and one to obtain sound data deemed to be a normal sound. For example, in an example where the sound produced when the brakes are applied is inspected, the inspector will depress the brake pedal at least twice to apply the brakes twice.

[0068] For sound data that is deemed to be normal sound, vibration data of a corresponding duration is also acquired via the vibration acquisition unit 10.

[0069] The normal sound vibration source estimation unit 31, like the above-mentioned abnormal sound vibration source estimation unit 30, estimates the correlation between the sound and the vibration using the sound data regarded as a normal sound and the corresponding vibration data, based on the correlation between the sound and the vibration and the degree of agreement of the frequency characteristics. That is, it identifies the source of the sound regarded as a normal sound. For example, if the sound is a sound generated during braking, the brake disc or the brake caliper is set as a specific part and vibration is measured by the laser Doppler vibrometer 102, and it is determined whether or not this specific part is the vibration source of the sound regarded as a normal sound based on the correlation between the sound and the vibration.

[0070] The estimation result comparison unit 80 compares the abnormal sound and vibration characteristics obtained by the abnormal sound vibration source estimation unit 30 with the normal sound and vibration characteristics obtained by the normal sound vibration source estimation unit 31, and outputs the final estimation result. For example, if there is no substantial difference between the sound and vibration characteristics regarded as an abnormal sound and the sound and vibration characteristics regarded as a normal sound, there is doubt about the determination of an abnormal sound and the identification of its vibration source, and the possibility of it being the abnormal sound vibration source is estimated to be low.

[0071] FIG. 13 is an explanatory diagram showing a display example on the display unit 50 of the third embodiment. In this example, it is written at the top that there is a low possibility that a specific part such as a brake disc to which the laser Doppler vibrometer is directed is the vibration source of the abnormal sound, and the numerical values ​​of the acquired or calculated data are summarized and displayed in a table format. This table is similar to the table of the first embodiment shown in FIG. 3, but the "magnitude of amplitude" and "fundamental frequency / natural frequency" of the sound and vibration are displayed for both abnormal sounds and normal sounds. In the "magnitude of amplitude" column, for example, the amplitude before the occurrence of the brake operating sound and the amplitude during the occurrence of the brake operating sound are represented by the numerical values ​​before and after the arrow. In the "judgment result" row, the correlation of the amplitude (value of correlation α) and the degree of agreement Δ of the frequency are written, as in FIG. 3 of the first embodiment.

[0072] 13, the sound data of the abnormal sound has a maximum sound pressure level of 50 dBA in the period before the braking sound occurs (for example, 3 seconds), a maximum sound pressure level of 80 dBA in the period after the braking sound occurs (for example, 3 seconds), and a frequency of 400 Hz. Similarly, the maximum vibration levels before and after the braking sound occur are 40 dB and 66 dB, respectively, and the frequency is 20 Hz.

[0073] The correlation α for the amplitude is given by the above-mentioned formula (1): α=(66-40) / (80-50)=26 / 30=0.866 In this case, the amplitude of both the sound and the vibration is large when it is judged to be an abnormal sound, so there is a correlation. If the amplitude of the vibration is not large, the value of the correlation α will be small, so there is no correlation. Generally, if the value of α is in the range of 0.5 to 2, there is a correlation.

[0074] The frequency matching degree Δ takes into account the case of harmonics. In the above formula (2), if n = 1 and m = 20, Δ=(1*400) / (20*20)=1 Therefore, it is determined that the frequencies of the sound and vibration match each other.

[0075] If the sound frequency is 401 Hz, then Δ=1.0025, which is not 1, and so a primary determination is made that the sound and vibration frequencies do not match. In this case, as described above, the deviation ΔΔ is further calculated and a final determination of the degree of match (secondary determination) is made. In this example, the deviation ΔΔ is within a 10% error, so they match.

[0076] In the example of FIG. 13, the correlation α is 0.866 and the frequency coincidence Δ is 1, so that a specific portion whose vibration is measured via the laser Doppler vibrometer, such as a brake disc or a brake caliper, is tentatively identified as the source of the abnormal noise.

[0077] Similar processing is performed on sounds deemed to be normal sounds and their corresponding amplitudes. In the example of Fig. 13, for the sound data deemed to be normal sounds, the maximum sound pressure level in the period before the occurrence of the braking sound (e.g., 3 seconds) is 49 dBA, and the maximum sound pressure level in the period after the occurrence of the braking sound (e.g., 3 seconds) is 79 dBA, and the frequency at that time is 401 Hz. Similarly, for vibration, the maximum vibration levels before and after the occurrence are 40 dB and 65 dB, respectively, and the frequency is 20 Hz.

[0078] Based on these values, for sounds deemed normal, the correlation α for amplitude is 0.833, which means there is a correlation. The frequency match Δ is strictly Δ=1.0025, but as mentioned above, the deviation ΔΔ is small, so the frequencies are considered to match.

[0079] Then, by comparing the characteristics of the abnormal sound as exemplified in the table of FIG. 13 with the characteristics of the normal sound, it is determined whether the two sounds are substantially the same. For example, similar to the 10% error method described above, if the difference between the amplitude of the abnormal sound and the amplitude of the normal sound is 10% or less, the two sounds are deemed to have the same characteristics. When comparing the amplitudes, it is desirable to match the frequency characteristics of the two sounds and then compare the magnitude of the amplitudes. Regarding the frequency characteristics, it is also determined whether the normal sound and the abnormal sound have the same frequency characteristics based on the difference in frequency, the difference in the appearance of the peaks in the overall frequency characteristics, etc. If the sound regarded as an abnormal sound and the sound regarded as a normal sound are substantially the same sound, there is a possibility that the determination that the sound was an abnormal sound was erroneous.

[0080] Similarly, for vibrations corresponding to sounds deemed to be abnormal sounds and vibrations corresponding to sounds deemed to be normal sounds, it is determined whether they are substantially the same vibration, i.e., whether they are from the same vibration source, based on the amplitude and frequency characteristics. For example, if the vibrations corresponding to sounds deemed to be abnormal sounds and vibrations corresponding to sounds deemed to be normal sounds are clearly different, there is a high probability that the determination that the sound is abnormal is correct.

[0081] As a final estimation result based on these points of view, the example in FIG. 13 displays that there is a low possibility that the specific part that was the subject of the inspection, such as the brake disc, is the vibration source of the abnormal noise.

[0082] Next, Fig. 12 shows a flow chart of the process flow of the inspection display device of the third embodiment. First, in step 21, the microphone of the sound acquisition unit 1 collects sounds from a specific part, such as the brake disc or brake caliper, of a vehicle test-running on a free roller, and a sound of a predetermined length is extracted and acquired as sound data, triggered, for example, by an inspector stepping on the brake pedal. Next, it is determined whether the sound contained in this sound data is normal or abnormal (step 22). If it is determined to be an abnormal sound, the process proceeds to step 23, and vibration data of a predetermined length corresponding to the sound data is acquired from a buffer included in the vibration acquisition unit 10.

[0083] Next, the process proceeds to step 24, where information on the distance from the microphone to the specific part is acquired. Then, in step 25, it is determined whether the distance / position to the specific part is appropriate. If it is determined that the distance and position of the microphone / laser Doppler vibrometer in this inspection are not appropriate in terms of the sound amplitude and frequency characteristics, the process returns to step 21 and acquires sound again. If the distance and position are appropriate, the process proceeds to step 29 and onwards. Although not shown in the flowchart, the source of abnormal sound vibration is estimated based on the sound data regarded as abnormal sound and the corresponding vibration data, in the same way as in the first embodiment.

[0084] On the other hand, if the sound is deemed to be normal in step 22, the process proceeds from step 22 to step 26, where vibration data of a predetermined length corresponding to the sound data is acquired from the buffer included in the vibration acquisition unit 10. Then, in step 27, information on the distance from the microphone to the specific part is acquired. Furthermore, in step 28, it is determined whether the distance / position to the specific part is appropriate. If it is determined that the distance and position of the microphone / laser Doppler vibrometer in this inspection are not appropriate in terms of the sound amplitude and frequency characteristics, the process returns to step 21 to acquire the sound again. If the distance and position are appropriate, the process proceeds to step 29 and onward. Although not shown in the flowchart, the normal sound vibration source is estimated based on the sound data deemed to be normal in this way and the corresponding vibration data.

[0085] In step 29, as described above, it is determined whether the characteristics of the sound / vibration regarded as normal sound and the characteristics of the sound / vibration regarded as abnormal sound are substantially identical. If they are different, in step 30, various information including that the probability that it is the source of abnormal sound / vibration is high is displayed and the data is saved. If they are substantially identical, in step 31, various information including that the probability that it is the source of abnormal sound / vibration is low is displayed and the data is saved. In step 32, it is determined whether to continue the inspection, and if YES, the process returns to step 21 and the inspection of the next vehicle or other specific parts is continued.

[0086] In this way, in the third embodiment, by comparing the characteristics of sounds regarded as normal sounds and the corresponding vibrations, it is possible to prevent erroneous estimation of the source of abnormal sounds.

[0087] Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment and can be applied in various ways. For example, in addition to the above-mentioned finished vehicle inspection process, the present invention can be widely used in inspection and maintenance at maintenance factories. [Explanation of symbols]

[0088] 1...Sound acquisition section 10...Vibration acquisition section 20…Abnormal sound detection section 21...Normal sound determination section 30…Abnormal noise and vibration source estimation unit 31…Normal sound vibration source estimation unit 40…Data storage section 50...Display section 60…Distance measuring section 80…Estimation result comparison section 100…Sound and vibration measurement set 101…Microphone 102...Laser Doppler vibrometer 103...Laser rangefinder 111...Guide rail

Claims

1. Sound is picked up by a microphone aimed at a specific part of the vehicle being inspected. In parallel with the measurement of the sound, a non-contact vibration sensor is oriented toward the specific part to obtain the vibration of the specific part, Determine whether the acquired sound is normal or abnormal, Identifying the specific part as the source of the abnormal sound based on the correlation between the sound determined to be the abnormal sound and the corresponding vibration, Display the results, How to show vehicle inspection.

2. A laser Doppler vibrometer or a laser displacement meter is used as the non-contact vibration sensor. The vehicle inspection and indication method according to claim 1.

3. Extracting a sound of a predetermined length and determining whether it is a normal sound or an abnormal sound. The vehicle inspection and indication method according to claim 1.

4. determining a correlation between the sound determined to be an abnormal sound and at least one of the amplitude and frequency of the vibration corresponding to the abnormal sound; The vehicle inspection and indication method according to claim 1.

5. A sound of a predetermined length is extracted using a vehicle signal indicating a specific vehicle state, a control signal of an inspection equipment, a predetermined action of an inspector, or a sound feature exceeding a predetermined threshold as a trigger. The vehicle inspection and indication method according to claim 3.

6. Acquiring sound and vibration at at least two positions at different distances from the specific part; Identifying the specific part as the source of the abnormal noise based on the correlation between the sound and the vibration at each position, Calculate the probability that the sounds obtained at different distances are the same sound based on the differences in sound characteristics at different distances. The vehicle inspection and indication method according to claim 1.

7. A microphone and a non-contact vibration sensor are disposed at a plurality of positions each having a different distance from the specific part, Acquiring sound and vibration in parallel at multiple locations at different distances. The vehicle inspection and indication method according to claim 6.

8. a microphone and a non-contact vibration sensor are supported on the guide rail so as to be movable along the guide rail; Acquiring sound and vibrations sequentially at multiple positions at different distances. The vehicle inspection and indication method according to claim 6.

9. measuring the distance from the specific part to the microphone and the non-contact vibration sensor using a distance sensor; This distance is displayed together with the results shown above. The vehicle inspection and indication method according to claim 6.

10. A plurality of microphones and a non-contact vibration sensor are arranged in a plane facing the specific portion, and the microphones and the non-contact vibration sensor are directed to a plurality of points constituting the plane to acquire sounds and vibrations. The vehicle inspection and indication method according to claim 1.

11. A normal sound is acquired, and a normal sound vibration corresponding to the normal sound is acquired. When the correlation between the sound determined to be an abnormal sound and the vibration corresponding thereto is similar to the correlation between the normal sound and the normal sound vibration, the probability that the specific part is the source of the abnormal sound is reduced. The vehicle inspection and indication method according to claim 1.

12. a sound acquisition unit that acquires sound using a microphone directed to a specific portion of the vehicle to be inspected; a vibration acquisition unit that acquires vibration of the specific part using a non-contact vibration sensor directed to the specific part; an abnormal sound determination unit that determines whether the acquired sound is a normal sound or an abnormal sound; an abnormal sound vibration source estimating unit that identifies the specific portion as the source of the abnormal sound based on a correlation between the sound determined to be an abnormal sound and the vibration corresponding to the abnormal sound; A display unit for displaying the results; A vehicle inspection and display device comprising:

Images