Wheel section abnormality detection method and abnormality detection device

Abstract

To detect the abnormality of a contact between a back plate and a disk rotor in a wheel part, etc., on the basis of the sound generated in the wheel part and further identify a component.SOLUTION: The present invention acquires the sound generated from an engine with a microphone and generates sound data (step 1), and generates frequency data associated with a vehicle speed on the basis of frequency analysis by an FFT (step 2). The same is compared with a basic characteristic per vehicle speed and an abnormal noise is extracted (steps 3, 4). The same is verified against the eigen frequency of a component near the wheel part and a component that can be a candidate is temporarily identified (step 5). When a regression coefficient α1 that indicates a correlation between sound pressure and vehicle speed is greater than or equal to a threshold, it is determined that the abnormal noise comes from contact between a rotor and the component and an indication is made to that effect (steps 6, 7, 8). When there is no correlation with the vehicle speed, it is determined that the abnormal noise is a disturbance noise.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an abnormality detection technique for detecting an abnormality such as component contact in a wheel portion of a vehicle such as an automobile based on a sound generated in the wheel portion.

Background Art

[0002] For example, a back plate provided to protect a disk rotor of a braking device may come into contact with the disk rotor, resulting in abnormal noise. It is generally difficult for an operator or the like to accurately determine such an abnormality in component contact with a rotating body.

[0003] Patent Document 1 discloses a technique in which an acceleration sensor is installed in a bearing or its peripheral components to detect abnormal vibration in a wheel bearing portion and prevent a serious accident, and an acceleration sensor is installed at a reference site that is a position away from the influence of the vibration of the bearing portion. When in specific driving conditions (for example, straight running at 40 km / h), the signals of each sensor are subjected to FFT processing, and the presence or absence of an abnormality is determined from a comparison between the two, such as the peak level of the frequency spectrum.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the above technique, acceleration sensors are required for both the vibration site and the reference site. When there are a plurality of candidate components that can cause vibration, it is necessary to provide acceleration sensors for all components. Therefore, in reality, it cannot be applied to inspections in which the target component is not specified to be one. Also, when candidate components are close to each other, it becomes difficult to separate each vibration.

Means for Solving the Problems

[0006] The method for detecting an abnormality in a wheel part according to the present invention is as follows: Acquire the sound generated near the wheel part of the target vehicle while Vehicle speed changing it to generate sound data, Generate frequency data associated with Vehicle speed by converting this sound data according to frequency, Extract abnormal sounds by comparing this frequency data with the basic frequency data associated with the corresponding Vehicle speed , Identify candidate parts by collating the frequency characteristics of this abnormal sound with the natural vibration frequencies of the parts, and determine whether the part is abnormal based on the correlation between the sound characteristics of the abnormal sound and Vehicle speed and Present the result including the identified parts.

Advantages of the Invention

[0007] According to the present invention, by focusing on the correlation between the sound characteristics of the abnormal sound and Vehicle speed it is possible to reliably identify whether the abnormal sound is an abnormal sound that should be considered abnormal or external disturbance noise, and it is possible to identify the parts in an abnormal state by collating the frequency characteristics of the abnormal sound with the natural vibration frequencies of the parts. Therefore, it is possible to detect abnormalities for a plurality of parts only by acquiring sound.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Mode for Carrying Out the Invention

[0009] Hereinafter, an embodiment of the present invention will be described. This embodiment of the abnormal detection of the wheel part is executed as an inspection for confirming that there is no contact or interference of parts with respect to the rotating body in the wheel part, for example, in the final inspection process of the automobile production line. Generally, in the final vehicle inspection process, an inspector test-drives the completed vehicle to be inspected on a free roller, and conducts inspections on a number of items such as the engine, meters, brakes, etc. During this test drive on the free roller, an acceleration operation is performed until the vehicle speed reaches an appropriate high speed range (for example, 120 km / h) over an appropriate inspection time (for example, about 15 to 20 seconds), and an abnormality inspection is performed based on the sound generated from the wheel part during that time. Hereinafter, the abnormal detection device of this embodiment will also be referred to as a "wheel contact inspection device", and this wheel contact inspection device is configured as a part of the inspection device in the final vehicle inspection process.

[0010] FIG. 1 shows a functional block diagram of the wheel contact inspection device of the first embodiment. The wheel contact inspection device of the first embodiment includes a sound measurement / acquisition unit 10, a vehicle motion state acquisition unit 70, a frequency conversion unit 20, a abnormal sound determination unit 30, a natural frequency acquisition unit 40, a abnormal sound component identification unit 50, and a display unit 60.

[0011] The sound measurement and acquisition unit 10 includes a microphone that acquires sound generated from the wheel section and converts it into an electrical signal, i.e., sound data, and a recording unit that temporarily stores this sound data. The microphone is placed outside the vehicle so that it can collect the sound from the wheel section of the vehicle running on free rollers. The microphone's directivity and frequency characteristics are selected according to the position relative to the vehicle being measured and the required frequency band. Typically, a microphone with directivity toward the wheel section of the vehicle is used. A microphone array or the like may be used to localize the sound source and obtain sound data from which noise has been removed. It is also possible to use an existing microphone, for example, one installed to collect horn sounds during the finished vehicle inspection process.

[0012] As described above, the wheel contact inspection is performed for, for example, 15 to 20 seconds while increasing the vehicle speed, so sound data having a duration of, for example, 15 to 20 seconds is acquired. Note that, since the vehicle speed changes over time, the sound data includes sounds at various vehicle speeds. In addition, the sound data acquired during the test run also includes engine sounds, etc.

[0013] The vehicle motion state acquisition unit 70 acquires the vehicle motion state, in this embodiment, the vehicle speed, from the signal of the vehicle speed sensor on the vehicle side, the rotation of the free rollers, etc., and outputs it to be associated with the sound data. 。

[0014] The frequency conversion unit 20 generates frequency data by converting the acquired sound data according to frequency using a frequency analysis method such as FFT (Fast Fourier Transform) or wavelet analysis. In particular, since it is necessary to ultimately determine the time change in sound pressure of a specific sound (change with respect to vehicle speed), frequency analysis is performed for each relatively short time domain to generate frequency data for each time.

[0015] Frequency data, for example when represented two-dimensionally, is represented as a frequency spectrum with the horizontal axis being frequency and the vertical axis being sound pressure (or power). Also, by stacking the conversion results for each time in a time series, frequency data may be handled as so-called spectrogram, which is three-dimensional data including time.

[0016] Figure 3 shows an example of the case where sound data obtained by wheel contact inspection is processed by FFT and displayed as a spectrogram. Here, in accordance with the format of a general spectrogram, the horizontal axis represents time and the vertical axis represents frequency. The total time length is about 15 seconds. The frequency includes up to around 10 kHz, for example. And the power (dBA) (or sound pressure or amplitude) of a certain frequency at a certain time point is represented by the brightness and color of each point. Although the attached figure is a black-and-white image, in the actual spectrogram display, for example, high power is in red, medium power is in yellow-green, and low power is in blue, and they are color-coded, and the power is indicated by the change in brightness within each color. What appears in a concentrated form at the bottom of the spectrogram in Figure 3 (indicated by reference sign E) is the engine sound during acceleration. Since the acceleration ends at time t1, the engine sound suddenly decreases.

[0017] Note that the graph shown below the spectrogram display in Figure 3 represents the original sound data with the vertical axis being amplitude.

[0018] The abnormal sound detector 30 extracts abnormal sounds by comparing frequency data associated with vehicle speed with basic frequency data. The abnormal sound detector 30 includes a storage unit in which frequency data for normal vehicles at each vehicle speed is stored in advance as basic frequency data. In one example, the frequency data is a frequency spectrum of sound pressure versus frequency, and the data of the vehicle under test is compared with the basic data to extract differences in sound pressure peaks as abnormal sounds. In other words, sound pressure peaks not included in the basic frequency data are abnormal sounds. The influence of the engine sound during acceleration described above can basically be eliminated by comparing with this basic frequency data associated with vehicle speed. Note that the aforementioned three-dimensional spectrogram may be used instead of the two-dimensional frequency spectrum, and abnormal sounds may be extracted by comparing with the basic spectrogram.

[0019] The natural frequency acquisition unit 40 includes a parts database that compiles the names, part numbers, and natural frequencies of the many parts in the wheel section that are the target of abnormality detection, and outputs this information to the abnormal noise part identification unit 50. The natural frequency of each part is measured using a hammering test or other method on the actual target part, or a value theoretically calculated using CAE. The natural frequency is a value specific to the part, and is uniquely determined by the part's material, shape, fixing method, etc.

[0020] The abnormal noise component identification unit 50 compares the frequency characteristics of the extracted abnormal noise with the natural frequencies of multiple components included in the wheel unit to tentatively identify candidate components, and determines whether the component is abnormal based on the correlation between the sound characteristics of the abnormal noise and the vehicle's motion state. Essentially, a component with a natural frequency (or its overtones) approximately equal to the frequency of the sound pressure peak of the abnormal noise can be a candidate for the source of the abnormal noise. If the abnormal noise is actually caused by vibration of this candidate component (e.g., vibration due to contact with a rotating body such as a disc rotor), the sound characteristics (e.g., sound pressure) of the abnormal noise should change when the motion state of the vehicle, which is the source of the vibration, changes. Therefore, the abnormal noise component identification unit 50 calculates the correlation between the sound pressure (peak sound pressure) of the abnormal noise corresponding to the natural frequency of the component selected as a candidate and vehicle speed, for example, by linear regression, and determines whether there is a correlation between the two based on the regression coefficient.

[0021] If the noise extracted as an abnormal sound does not correlate with the vehicle's motion state, such as vehicle speed, the sound is likely to be disturbance noise and is not treated as an abnormal part. If there is a certain correlation between the two, the part selected as a candidate is ultimately determined to be abnormal (in this embodiment, the part is in contact with the rotating body).

[0022] In the spectrogram shown in Figure 3, an example, a continuous abnormal noise with a constant frequency labeled "BP" can be seen. Although it is not clear in the black and white image of Figure 3, the line representing this abnormal noise changes color to indicate an increase in sound pressure as the vehicle speed increases (time elapses up to t1). Furthermore, because the tires continue to rotate after time t1, it is shown that the abnormal noise continues, in contrast to the sudden decrease in engine sound (E) at time t1.

[0023] The display unit 60 is a display means for displaying the inspection results to relevant parties such as workers in the vicinity of the inspection device, inspectors operating the vehicles, managers, and data scientists who utilize the data. For example, it may include a liquid crystal display or an organic EL display. If any notification sound or audio is involved, it may also include a sound source, amplifier, speaker, etc. for generating and emitting the sound. The images, videos, sounds, audio, etc. displayed here are generated by image generation means and audio / audio generation means included in the display unit 60. Specific examples of what is displayed on the display will be described later.

[0024] FIG. 2 shows, as a flowchart, the flow of processing of the wheel contact inspection apparatus of the above-described first embodiment. First, the microphone of the sound measurement / acquisition unit 10 collects the sound near the wheel portion of the vehicle that is test-running on the free roller and acquires it as sound data (step 1). Next, as the frequency conversion unit 20 described above, the sound data is converted according to frequency using a frequency analysis method such as FFT (Fast Fourier Transform) or wavelet analysis to generate frequency data. Specifically, frequency analysis is performed for each relatively short time domain to generate frequency data at individual times (step 2). Also, as the vehicle motion state, the information on the vehicle speed when the sound of the wheel portion is being acquired is read (step 3).

[0025] Next, the frequency data associated with the vehicle speed (for example, a two-dimensional frequency spectrum) is compared with the basic frequency data corresponding to the same vehicle speed, and the difference between the two, that is, the presence or absence of a sound pressure peak that does not exist in the basic frequency spectrum is determined (step 4). If there is no difference between the two, the process proceeds to step 10, where it is determined to be normal and a display to that effect is made.

[0026] If there is a difference from the basic frequency data, after obtaining information on the natural frequencies of a plurality of components near the wheel part from the component database (step 5), proceed to step 6 to determine whether the sound regarded as abnormal noise is emitted from the components of the wheel part, and also identify which component emitted the abnormal sound. Specifically, after temporarily identifying a plurality of candidate components by comparing the frequency of the sound pressure peak that becomes abnormal noise with the natural frequencies in the component database, using a large number of frequency data associated with the vehicle speed, calculate the correlation between the sound pressure and the vehicle speed for the sound (including harmonics) of the natural frequency of the candidate component by linear regression, and obtain the regression coefficient α1 (when the regression line is y = α1x + β1) (step 6). Then, determine the presence or absence of the correlation between the two based on whether this regression coefficient α1 is equal to or greater than a predetermined threshold value (step 7). That is, regarding abnormal noise caused by contact between components such as contact between the back plate and the disc rotor, attention is paid to the relationship that as the vehicle speed increases, the energy applied to the component increases and the sound pressure increases. If the sound pressure does not change even when the vehicle speed increases, it is determined that it is some kind of external disturbance noise. This calculation of linear regression and determination of the magnitude of the regression coefficient α1 are made for each of the plurality of components. If there is no component for which the regression coefficient α1 is equal to or greater than the threshold value, proceed to step 10, determine that it is normal, and display to that effect.

[0027] Here, for a certain component (for example, component A) having a natural frequency corresponding to abnormal noise, if the regression coefficient α1 is equal to or greater than the threshold value, proceed to step 8, determine that component A is abnormal (that is, the abnormal sound source), and display to that effect.

[0028] Finally, in step 9, determine whether re-inspection is required due to some reason. If re-inspection is necessary, return to step 1, and if not, end the inspection.

[0029] As described above, the wheel contact inspection device of the embodiment described above can detect contact between a component such as a back plate and a rotating body such as a disc rotor by collecting sounds near the wheel section with a microphone and processing the signals. In particular, by utilizing information on the natural frequencies of multiple components near the wheel section and the correlation between the sound pressure of the extracted abnormal noise and the vehicle speed, it is possible to identify the component that is making the abnormal noise due to contact while eliminating the influence of other noises, even in an environment where other noises are present. Therefore, wheel contact inspection can be performed efficiently during test runs on free rollers during the completed vehicle inspection process.

[0030] Next, a wheel contact inspection device according to a second embodiment will be described with reference to Figures 4 and 5. The following mainly describes the differences from the first embodiment. The wheel contact inspection device according to the second embodiment uses the half-width of the waveform in the frequency spectrum as the sound characteristic instead of sound pressure.

[0031] As shown in the block diagram of FIG. 4, the wheel contact inspection device of the second embodiment, like the first embodiment, is configured to include a sound measurement and acquisition unit 10, a vehicle motion state acquisition unit 70, a frequency conversion unit 20, an abnormal sound determination unit 30, a natural frequency acquisition unit 40, an abnormal sound component identification unit 50, and a display unit 60.

[0032] Here, when extracting abnormal sounds, the abnormal sound detector 30 compares full widths at half maximum (FWHM) instead of comparing sound pressures as in the first embodiment. When a peak exists in a predetermined frequency range in the sound pressure-frequency characteristics, the maximum (= peak value) and minimum sound pressure values within that range are extracted, and the intermediate value H between the two (H = (maximum value - minimum value) / 2) is calculated. The full width at half maximum is given as the width of the frequency range in which the sound pressure exceeds this intermediate value H. Such full width at half maximum basically corresponds to the height of the sound pressure at each frequency, and while the peak value of sound pressure varies depending on the shape of the peak, this full width at half maximum can represent the magnitude of energy at each frequency without being affected by the peak shape. Therefore, whereas the frequency data is a frequency spectrum of sound pressure vs. frequency in the first embodiment, the second embodiment uses a frequency spectrum of full width at half maximum vs. frequency, where the vertical axis is the full width at half maximum and the horizontal axis is the frequency, for example.

[0033] That is, a basic half-value width-frequency frequency spectrum associated with the vehicle speed is stored in advance, and a half-value width-frequency frequency spectrum associated with the vehicle speed is generated from the sound data of the vehicle to be inspected, and by comparing the two, a peak component of the half-value width that does not exist in the basic frequency data is extracted as an abnormal sound.

[0034] Further, in the abnormal sound component identification unit 50, instead of evaluating the correlation between the sound pressure and the vehicle speed in the first embodiment described above, the correlation between the half-value width of the half-value width peak frequency extracted as an abnormal sound and the vehicle speed is evaluated. That is, the correlation between the half-value width of the abnormal sound corresponding to the natural frequency of the component selected as a candidate and the vehicle speed is obtained as, for example, linear regression, and the presence or absence of the correlation between the two is determined from the regression coefficient.

[0035] FIG. 5 is a flowchart showing the processing flow of the wheel contact inspection in the second embodiment. After acquiring sound data, generating frequency data, and reading the vehicle speed in steps 1 to 3, in step 4A, an abnormal sound is extracted by comparing the half-value width-frequency frequency spectrum associated with the vehicle speed with the basic half-value width-frequency frequency spectrum also associated with the vehicle speed. If there is no difference between the two, the process proceeds to step 10, where it is determined to be normal and a display to that effect is made.

[0036] If an abnormal noise is detected, the natural frequency information of the component is acquired in step 5, and then the process proceeds to step 6A. The half-width peak frequency of the abnormal noise is compared with the natural frequencies in the component database to identify multiple candidate components. Using a large number of frequency data items associated with vehicle speed, the correlation between the half-width of the natural frequency of the candidate components (including harmonics) and vehicle speed is calculated using linear regression, and the regression coefficient α2 (when the regression line is y = α2x + β2) is obtained (step 6A). The presence or absence of a correlation between the two is determined based on whether this regression coefficient α2 is equal to or greater than a predetermined threshold (step 7A). If no component has a regression coefficient α2 equal to or greater than the threshold, the process proceeds to step 10, where the component is determined to be normal, and a message to that effect is displayed. If the regression coefficient α2 for a component (e.g., component A) with a natural frequency corresponding to the abnormal noise is equal to or greater than the threshold, the process proceeds to step 8, where component A is determined to be abnormal (i.e., the source of the abnormal noise), and a message to that effect is displayed.

[0037] In this way, by using the half-width instead of the sound pressure for the frequency data after frequency analysis, the data is less susceptible to the influence of the waveform of the sound pressure peak, as described above.

[0038] Next, a wheel contact inspection device according to a third embodiment will be described with reference to Figures 6 and 7. The wheel contact inspection device according to the third embodiment uses both sound pressure as in the first embodiment and half-value width of the frequency waveform as in the second embodiment as sound characteristics.

[0039] As shown in the block diagram of FIG. 6, the wheel contact inspection device of the third embodiment, like the first embodiment, is configured to include a sound measurement and acquisition unit 10, a vehicle motion state acquisition unit 70, a frequency conversion unit 20, an abnormal sound determination unit 30, a natural frequency acquisition unit 40, an abnormal sound component identification unit 50, and a display unit 60.

[0040] Here, when extracting abnormal noise, the abnormal noise detector 30 performs both a comparison with the basic characteristic of sound pressure as in the first embodiment and a comparison with the basic characteristic of full width at half maximum (FWHM) as in the second embodiment.

[0041] Then, in the abnormal noise component identification unit 50, both the evaluation of the correlation between the sound pressure and the vehicle speed in the first embodiment described above and the evaluation of the correlation between the half-value width and the vehicle speed in the second embodiment are performed.

[0042] FIG. 7 is a flowchart showing the flow of the wheel contact inspection process in the third embodiment. After acquiring sound data, generating frequency data, and reading the vehicle speed in steps 1 to 3, information on the natural frequencies of the components around the wheel part is acquired in step 5.

[0043] Next, in step 6, using a large number of frequency data associated with the vehicle speed, the correlation between the sound pressure and the vehicle speed for the sounds (including overtones) of the natural frequencies of a large number of components is calculated by linear regression, and the regression coefficient α1 (when the regression line is y = α1x + β1) is obtained. Also, in step 4, by comparing the frequency spectrum of the sound pressure - frequency associated with the vehicle speed with the frequency spectrum of the basic sound pressure - frequency corresponding to the same vehicle speed, the presence or absence of a difference between the two is determined. If there is a difference between the two, that is, if there is an abnormal noise, for the candidate components having the natural frequency corresponding to the frequency of this abnormal noise, it is determined whether there is a correlation between the sound pressure and the vehicle speed based on whether the above regression coefficient α1 is greater than or equal to a predetermined threshold (step 7).

[0044] If these determinations are YES, then in step 6A, using a large number of frequency data associated with the vehicle speed, the correlation between the half-value width and the vehicle speed for the sounds (including overtones) of the natural frequencies of a large number of components is calculated by linear regression, and the regression coefficient α2 (when the regression line is y = α2x + β2) is obtained. Also, in step 4A, by comparing the frequency spectrum of the half-value width - frequency associated with the vehicle speed with the frequency spectrum of the basic half-value width - frequency corresponding to the same vehicle speed, the presence or absence of a difference between the two is determined. If there is a difference between the two, that is, if there is an abnormal noise, for the candidate components having the natural frequency corresponding to this abnormal noise, it is determined whether there is a correlation between the half-value width and the vehicle speed based on whether the above regression coefficient α2 is greater than or equal to a predetermined threshold (step 7A).

[0045] If the determinations in steps 4, 7, 4A, and 7A for a certain part A are all YES, then the part is determined to be abnormal (i.e., the source of an abnormal sound) and a message to that effect is displayed (step 8). If any of the results are NO, the process proceeds to step 10, where the part is determined to be normal and a message to that effect is displayed.

[0046] In the third embodiment, the vehicle speed change regarding the sound pressure of the abnormal noise and the vehicle speed change regarding the half-value width are AND conditions, so abnormality determination becomes stricter and false detection due to noise and the like is reduced.

[0047] 7, the regression coefficients α1 and α2 are calculated for all parts around the wheel before extracting abnormal noises by comparing with the basic characteristics. As in the first and second embodiments described above, after extracting abnormal noises, the regression coefficients α1 and α2 may be calculated only for provisional candidate parts having natural frequencies corresponding to the abnormal noises.

[0048] Furthermore, in the above example, the processing for the sound pressure is performed first, but the processing for the half-value width may be performed first.

[0049] Next, a wheel contact inspection device according to a fourth embodiment will be described with reference to Figures 8 and 9. The wheel contact inspection device according to the fourth embodiment is designed to more accurately determine whether an abnormal noise is coming from the wheel section, depending on whether the abnormal noise contains a fluctuating sound component (a sound in which the sound of a predetermined high frequency component fluctuates by about 1 to 10 Hz) that corresponds to the rotational speed of the tire.

[0050] For example, if the cause of the abnormal noise is interference between the back plate and the disc rotor, the abnormal noise occurs when part of the back plate comes into contact with the disc rotor with each rotation of the disc rotor (i.e., with each rotation of the tire), and in many cases the components of the abnormal noise contain a fluctuating sound component corresponding to the rotational speed of the tire.For example, if the tire diameter is 64 cm and the vehicle speed is 10 km / h, the rotational frequency of the tire is approximately 1.3 Hz, so the vibration noise of a part with a natural frequency of, for example, 5 kHz will contain a fluctuating sound component of approximately 1.3 Hz.

[0051] As shown in the block diagram of FIG. 8 , the wheel contact inspection device of the fourth embodiment, like the first embodiment, includes a sound measurement and acquisition unit 10, a vehicle motion state acquisition unit 70, a frequency conversion unit 20, an abnormal sound determination unit 30, a natural frequency acquisition unit 40, an abnormal sound component identification unit 50, and a display unit 60. The device further includes a time-varying frequency calculation unit 80, which determines the fluctuating sound components of the acquired sound data and calculates whether a fluctuating frequency equivalent to the wheel rotation frequency is included. If a fluctuating component equivalent to the wheel rotation frequency is not included, it is determined that there is no contact between the wheel and a rotating body. The fluctuating component equivalent to the wheel rotation frequency is, for example, approximately 1 to 10 Hz. However, this may be set to a narrower range taking into account the vehicle motion state, i.e., vehicle speed, or may be set to a constant range (e.g., 1 to 10 Hz) regardless of vehicle speed.

[0052] FIG. 9 is a flowchart showing the processing flow of wheel contact inspection in the fourth embodiment. In steps 1 to 3, sound data is acquired, frequency data is generated, and vehicle speed is read, and then in step 5, information on the natural frequency of parts around the wheel is acquired.

[0053] Next, in step 11, the fluctuating sound component contained in the sound data, that is, the time-varying frequency, is calculated based on the frequency data, and in step 12, it is determined whether or not this time-varying frequency is substantially the same as the rotation frequency of the wheel.

[0054] The time-varying frequency can be calculated using a known fluctuating sound analysis algorithm. For example, analysis frames are set in minute time units, and fluctuating sound can be analyzed by using multiple band-pass filters for extracting fluctuating components to find the peaks and valleys of the waveform after passing through the band-pass filters for each analysis frame.

[0055] If it is determined in step 12 that the time-varying frequency is different from the rotational frequency of the wheel, proceed to step 10, determine it as normal, and display to that effect. If the time-varying frequency is substantially the same as the rotational frequency of the wheel, proceed to step 6 and subsequent steps.

[0056] Thereafter, it is the same as each of the above-described embodiments. In step 6, using a number of frequency data associated with the vehicle speed, the correlation between the sound pressure and the vehicle speed for the sound (including overtones) of the natural frequencies of a number of components is calculated by linear regression, and the regression coefficient α1 (when the regression line is y = α1x + β1) is obtained. Also, in step 4, by comparing the frequency spectrum of the sound pressure-frequency associated with the vehicle speed with the frequency spectrum of the basic sound pressure-frequency corresponding to the same vehicle speed, the presence or absence of a difference between the two is discriminated. If there is a difference between the two, that is, if there is an abnormal sound, for the candidate components having the natural frequency corresponding to the frequency of this abnormal sound, it is determined whether there is a correlation between the sound pressure and the vehicle speed based on whether the regression coefficient α1 is equal to or greater than a predetermined threshold value (step 7).

[0057] If the determinations in steps 4 and 7 for a certain component A are both YES, it is determined that component A is abnormal (that is, an abnormal sound source), and a display to that effect is made (step 8). If either is NO, proceed to step 10, determine it as normal, and display to that effect.

[0058] Note that in the process flow shown in the flowchart of FIG. 9, the regression coefficient α1 is calculated for all components around the wheel part before extracting the abnormal sound by comparison with the basic characteristics. Similar to the first and second embodiments described above, after extracting the abnormal sound, the regression coefficient α1 may be calculated only for the provisional candidate components having the natural frequency corresponding to the abnormal sound.

[0059] Also, in the example of the above flowchart, only the sound pressure is targeted, but both the sound pressure and the half-value width may be used as in the third embodiment.

[0060] In this fourth embodiment, the presence of a fluctuating sound component corresponding to the rotation frequency of the wheel is a weighting condition, so abnormality determination is stricter than in the first and second embodiments, and false detection due to noise, etc. is reduced.

[0061] Next, a wheel contact inspection device according to a fifth embodiment will be described with reference to Figures 10 to 13. The wheel contact inspection device according to the fifth embodiment is designed to perform more detailed abnormality inspection by determining whether the correlation between the sound pressure of the natural frequency sound of a part and the vehicle motion state (for example, vehicle speed) is simply linear or nonlinear (for example, a quadratic function).

[0062] As mentioned above, when abnormal noise occurs due to contact between a rotating body and a component, for example, as vehicle speed increases, the vibration energy acting on the component increases, and therefore the sound pressure or half-width tends to increase with vehicle speed. However, the correlation between the two may not necessarily be linear depending on the material of the component. Figure 13 shows the relationship between sound pressure or half-width and vehicle speed. The characteristics of one component, represented by line L1, are linear. On the other hand, the characteristics of another component, represented by line L2, are nonlinear. More specifically, in the process of calculating the correlation coefficient, regression equations are calculated using both linear and nonlinear approximations, and the linearity or nonlinearity is determined by comparing the magnitudes of the correlation coefficients. In the example of Figure 13, for example, the correlation coefficient of line L1 is "0.9714" and the correlation coefficient of line L2 is "0.9978." Comparing the magnitudes of the two, the correlation coefficient is determined to be nonlinear. Note that a nonlinear relationship with vehicle speed may also be linear with other variables, such as vehicle acceleration.

[0063] As shown in the block section of FIG. 10, the wheel contact inspection apparatus of the fifth embodiment, similar to the first embodiment, includes a sound measurement / acquisition unit 10, a vehicle motion state acquisition unit 70, a frequency conversion unit 20, a abnormal sound determination unit 30, a natural frequency acquisition unit 40, an abnormal sound component identification unit 50, and a display unit 60. And further, it is provided with a correlation order calculation unit 90 for determining whether the characteristic of the frequency component extracted as an abnormal sound with respect to the vehicle speed is linear or non-linear. This correlation order calculation unit 90 obtains an approximate curve of the correlation and determines whether it is linear or non-linear from its order. In this embodiment, it is determined whether the characteristic of the component is linear or non-linear from the material included in the component database of the natural frequency acquisition unit 40, but information on whether each component is linear or non-linear may be included in the component database.

[0064] FIG. 11 is a flowchart showing the processing flow of the wheel contact inspection of the fifth embodiment. After performing sound data acquisition, frequency data generation, and vehicle speed reading in steps 1 to 3, information on the natural frequencies of the components around the wheel part is acquired in step 5.

[0065] Next, in step 21, based on the frequency data corresponding to the vehicle speed, regression equations are calculated by both linear approximation / non-linear approximation (quadratic function), and their respective correlation coefficients are calculated. Then, in step 22, it is determined whether the correlation is linear by comparing the magnitudes of the two correlation coefficients.

[0066] If it is determined to be linear in step 22, it proceeds to step 23, and the candidate components are narrowed down to the group of components made of material a. If it is determined to be non-linear, it proceeds to step 24, and the candidate components are narrowed down to the group of components made of material b. In other words, the components of material a are components having linear characteristics, and the components of material b are components having non-linear characteristics.

[0067] Thereafter, it is the same as each of the above-described embodiments. In step 6, using a number of frequency data associated with the vehicle speed, for the group of parts made of material a or material b, the correlation between the sound pressure of the natural vibration frequency sound (including overtones) and the vehicle speed is calculated by linear regression, and the regression coefficient α1 (when the regression line is y = α1x + β1) is obtained. Further, in step 4, by comparing the frequency spectrum of the sound pressure-frequency associated with the vehicle speed with the frequency spectrum of the basic sound pressure-frequency corresponding to the same vehicle speed, the presence or absence of a difference between the two is discriminated. If there is a difference between the two, that is, if there is an abnormal sound, for the candidate parts having a natural vibration frequency corresponding to the frequency of this abnormal sound, it is determined whether there is a correlation between the sound pressure and the vehicle speed based on whether the above regression coefficient α1 is equal to or greater than a predetermined threshold value (step 7).

[0068] If the determinations in steps 4 and 7 for a certain part A are both YES, it is determined that part A is abnormal (that is, an abnormal sound source), and a display to that effect is made (step 8). If either is NO, proceed to step 10, determine that it is normal, and make a display to that effect.

[0069] According to this embodiment, since the candidate parts are narrowed down depending on whether the correlation is linear or non-linear, it becomes easier to identify the abnormal parts.

[0070] FIG. 12 is a flowchart showing another example of the processing flow of the wheel contact inspection of the fifth embodiment. In this example, in steps 21 and 22, the same as the flowchart of FIG. 11, it is discriminated whether the correlation is linear. However, if it is discriminated in step 22 that it is not linear, proceed to step 25 and display "correlated with acceleration". In particular, no narrowing down of candidate parts is performed.

[0071] Next, an example of the display shown on the display of the display unit 60 is shown in FIG. 14. In this display example, in order to visually represent the abnormal noise due to contact with the rotating body, the spectrogram display described above is arranged in the right part of the screen (the part indicated by reference sign A), and the original sound data is shown as a two-dimensional graph below the spectrogram display. As described above, the spectrogram is actually a color display. And in this display example, linear portions of different colors due to abnormal noise appear around 5 kHz during an inspection period of about 15 seconds (not clear in the black-and-white image of the attached drawing).

[0072] Also, at the upper part on the left side of the screen, as indicated by reference sign B, the presence or absence of contact with the rotating body is displayed with a relatively large label. In the illustrated example, since contact is detected, it is displayed in red as "Contact present". If no contact is detected, it will be displayed as "Contact absent". Further, when there is contact, below part B, as part C, the part name of the specified part (in one example, BP: back plate), as part D, the frequency of the abnormal noise (in one example, 5 kHz), as part E, the sound pressure thereof (in one example, -30 dB), and as part F, the half-value width thereof (in one example, 100 Hz) are respectively displayed. Inspectors and the like can make a final determination of the abnormality, that is, part contact, by comparing these displayed items with the spectrogram.

[0073] The above display is an example. For example, the inspection result may be simply displayed with characters such as OK / NG, or displayed with ○× of graphics. It is desirable to prevent the amount of information on the screen from becoming excessively large so that it can be understood in a short time.

[0074] Also, in the above display example, a spectrogram is used to visually represent the frequency of the abnormal noise and the like, but other displays for visualizing sounds, such as a mel spectrogram or a cepstrum, may be used.

[0075] Next, as another embodiment of the anomaly detection method of the present invention, a spectrogram display may be treated as image data and processed using a pre-created anomaly determination model. That is, as described above, a spectrogram display corresponding to sound data can be generated using frequency data obtained by frequency analysis of sound data. However, an anomaly determination model may be generated in advance by machine learning (or deep learning) using image data of a large number of spectrogram displays, including inspection examples with and without contact between various components such as backplates. By processing the spectrogram display of the sound of the vehicle being inspected as image data using the anomaly determination model, it is possible to determine the presence or absence of abnormal noise and identify the component causing the abnormal noise based on characteristic sound pressure changes, etc., in the image. As described above, in the completed vehicle inspection process where the vehicle speed is increased, changes over time correspond, in simple terms, to changes in vehicle speed.

[0076] While the present invention has been described above as an embodiment in which it is applied to a contact inspection of a back plate or the like in a wheel section during the inspection process of a completed vehicle, the present invention is not limited to this and can be applied in a variety of ways. For example, the presence or absence of contact in a wheel section can be easily inspected during inspection and maintenance at a dealership or repair shop for a commercial vehicle. It is also possible to inspect the presence or absence of contact between parts while the vehicle is actually traveling on the road.

[0077] The information presentation unit is not limited to the above-mentioned display unit that displays images, but may also be one that presents information only by voice. [Explanation of symbols]

[0078] 10...Sound measurement and acquisition section 20...Frequency conversion section 30...Abnormal noise detection section 40...Natural frequency acquisition section 50... Abnormal noise part identification section 60...Display section 70...Vehicle movement information acquisition unit

Claims

1. Acquire the sound generated near the wheel part of the target vehicle while changing the vehicle speed to generate sound data, Generate frequency data associated with the vehicle speed by converting this sound data according to the frequency, Extract abnormal sounds by comparing this frequency data with the basic frequency data associated with the corresponding vehicle speed, Match the frequency characteristics of this abnormal sound with the natural vibration frequency of the parts to identify candidate parts, and determine whether the parts are abnormal based on the correlation between the sound characteristics of the abnormal sound and the vehicle speed, Present the result including the identified parts, Abnormal detection method for the wheel part.

2. The abnormal detection method for the wheel part according to Claim 1, wherein the sound characteristic is sound pressure.

3. The abnormal detection method for the wheel part according to Claim 1, wherein the sound characteristic is the half-value width of the waveform in the frequency spectrum.

4. Determine whether it is abnormal based on the regression coefficient of the linear regression of two variables, namely the sound pressure as the sound characteristic of the abnormal sound or the half-value width of the waveform in the frequency spectrum, and the vehicle speed, in the abnormal detection method for the wheel part according to Claim 1.

5. Extract the fluctuating sound component with a relatively long period included in the sound data, Determine it as abnormal under the condition that the frequency of this fluctuating sound component corresponds to the rotational frequency of the wheel, in the abnormal detection method for the wheel part according to any one of Claims 1 to 4.

6. Determine whether the correlation between two variables, namely the sound pressure as the sound characteristic of the abnormal sound or the half-value width of the waveform in the frequency spectrum, and the vehicle speed, is linear or non-linear, Narrow down the candidate parts according to whether it is linear or non-linear, in the abnormal detection method for the wheel part according to Claim 1.

7. Generate a spectrogram display corresponding to the sound data using the frequency data, Generate an abnormal sound determination model by machine learning a large number of image data of spectrogram displays in advance, Process the spectrogram display of the sound data to be inspected with the abnormal sound determination model to determine the abnormal sound and identify the parts in the abnormal state, in the abnormal detection method for the wheel part according to Claim 1.

8. A sound acquisition unit that acquires the sound generated near the wheel part of the target vehicle while changing the vehicle speed to generate sound data, A frequency data generation unit that generates frequency data associated with the vehicle speed by converting this sound data according to the frequency, An abnormal sound determination unit that extracts abnormal sounds by comparing this frequency data with the basic frequency data associated with the corresponding vehicle speed, A component identification unit that identifies candidate components by comparing the frequency characteristics of the abnormal noise with the natural vibration frequencies of the components, and determines whether the components are abnormal based on the correlation between the sound characteristics of the abnormal noise and the vehicle speed; An information presentation unit that presents information on the results including the identified components; An abnormal detection device for a wheel unit comprising the above.

Images