Inspection method and inspection device for sound generation device

Abstract

To detect one-side blowing of a double horn in completed vehicle inspection.SOLUTION: A horn inspection device comprises: a sound signal acquisition part 10 which acquires a sound generated by a vehicle to be inspected and generates a sound signal; an object sound extraction part 20 which detects the sound that a horn device 1 generates from the sound signal; a sound signal processing part 30 which includes one-side blowing detection part 31; a vehicle information acquisition part 40 which acquires information related to the vehicle to be inspected; a storage part 50 which stores various data; and a display part 60 which displays inspection results. The one-side blowing detection part 31 analyzes frequency characteristics to determine whether horns A, B are blowing based upon the numbers of peaks appearing in frequency windows A, B corresponding to fundamental frequencies and harmonics of the horns A, B.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an inspection technique for inspecting whether a sound generating device including a plurality of sound generating bodies that each emit sounds of different frequencies is emitting sounds correctly. [Background technology]

[0002] For example, in the final stage of an automobile production line, in the finished vehicle inspection process, inspectors test drive the finished vehicle on free rollers and inspect a number of items, including the engine, meters, brakes, horn, and lights. Generally, during this inspection process, the presence or absence of abnormalities is determined by the inspector's sensory evaluation of various sounds emitted by various parts of the vehicle. For example, in the case of a horn, the inspector sounds the horn and listens to it to confirm that the sound is normal.

[0003] On the other hand, instead of such sensory tests, attempts have been made to detect abnormalities by capturing sounds with a microphone and analyzing the signals.

[0004] Patent Document 1 describes a technology for an automobile horn inspection device that processes the sound of the horn picked up by a microphone using FFT to obtain spectral data, calculates the number of peaks and valleys in the sound pressure level and the maximum peak frequency in this spectral data, and uses membership functions for these detected values ​​to determine whether the horn is pass or fail. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 3136084 Summary of the Invention [Problem to be solved by the invention]

[0006] The test subject may be a sound-generating device that includes multiple sound-generating elements, each emitting a different frequency, and that simultaneously emits sound from these multiple sound-generating elements. For example, a double horn, which has two sound-generating elements with different fundamental frequencies, is known as a type of automobile horn. In such double horns, a malfunction or abnormality known as "single-ended ringing" can occur, in which one of the sound-generating elements does not produce sound. However, in this case of "single-ended ringing," the horn sound itself is sufficiently loud, and the only difference is the sound quality, making it difficult to detect through a sensory inspection by an inspector.

[0007] Furthermore, the technology disclosed in Patent Document 1 cannot handle sounds generated by multiple sound generating bodies such as a double horn. [Means for solving the problem]

[0008] The present invention provides a method for inspecting a sound-generating device including a plurality of sound-generating bodies that emit sounds of different frequencies, the method comprising the steps of: Acquire the sound emitted by the sound generator, The frequency characteristics of this acquired sound are analyzed, Separating the sound components into a plurality of frequency bands for each sound generator corresponding to the fundamental frequency of each sound generator and the frequencies of its overtones; Whether or not each of the plurality of sound generators is emitting sound is determined based on the number of peak frequencies that appear in the frequency band corresponding to each sound generator.

[0009] That is, for each individual sound generator, a frequency band is set corresponding to the fundamental frequency of that sound generator and the frequencies of its overtones. The frequency characteristics obtained by analyzing the sound emitted by the sound generator include the fundamental frequencies of all sound generators and the peaks of their overtones. For example, if a first sound generator with a certain fundamental frequency is correctly emitting sound, peak frequencies will appear in many of the frequency bands set for this fundamental frequency. The frequency band set for a second sound generator with a different fundamental frequency is shifted from the frequency band for the first sound generator, so the peak frequency of the sound from the first sound generator will rarely appear.

[0010] Therefore, whether or not each sound generating element is producing sound is determined based on the number of each peak frequency. [Effects of the Invention]

[0011] According to the present invention, for a sound generating device including a plurality of sound generating elements, such as a double horn, it is possible to reliably test whether all of the sound generating elements are emitting sound correctly. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a functional block diagram of a first embodiment in which the present invention is applied to the inspection of double horns in the finished automobile inspection process. [Figure 2] FIG. 3 is an explanatory diagram of frequency characteristics of the first embodiment. [Figure 3] 4 is a flowchart showing the flow of processing in the first embodiment. [Figure 4] FIG. 3 is an explanatory diagram showing a display example on the display unit of the first embodiment. [Figure 5] FIG. 10 is a functional block diagram of a second embodiment. [Figure 6] FIG. 10 is an explanatory diagram of frequency characteristics of the second embodiment. [Figure 7] 10 is a flowchart showing the flow of processing in a second embodiment. [Figure 8] FIG. 10 is an explanatory diagram showing a display example on a display unit according to a second embodiment. [Figure 9] FIG. 10 is a functional block diagram of a third embodiment. [Figure 10] FIG. 10 is an explanatory diagram of frequency characteristics of the third embodiment. [Figure 11] 10 is a flowchart showing the flow of processing according to a third embodiment. [Figure 12] FIG. 11 is an explanatory diagram showing a display example on a display unit according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] An embodiment of the present invention applied to the inspection of automobile horns will be described below. This horn inspection is typically performed during the final vehicle inspection process, which is the final stage of an automobile production line. In general, during this process, an inspector test-drives the completed vehicle on a free roller and inspects various components, including the engine, meters, and brakes. During this inspection, the inspector sounds the horn by pressing a horn switch on the steering wheel according to a predetermined inspection sequence. The horn is then inspected for normal operation based on the sound. This section describes the detection of a "single-horn" condition, in which only one horn emits sound, when the horn device being inspected is a double horn consisting of two horns (referred to as "horn A" and "horn B" for the sake of explanation) with different fundamental frequencies.

[0014] The horn inspection device of the embodiment is not an independent device, but is configured as part of an inspection device that performs a series of finished vehicle inspections.

[0015] 1 shows a functional block diagram of a horn inspection device according to the first embodiment. The horn inspection device according to the first embodiment includes a sound signal acquisition unit 10 that acquires sounds emitted from a vehicle under inspection and generates a sound signal, a target sound extraction unit 20 that extracts sounds emitted by the horn device 1 from the sound signal, a sound signal processing unit 30 that includes a one-sided sound detection unit 31 (described later), a vehicle information acquisition unit 40 that acquires information about the vehicle under inspection, a memory unit 50 that stores various data, and a display unit 60 that displays the inspection results.

[0016] The sound signal acquisition unit 10 includes a microphone that acquires sounds generated by 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 can be of any suitable type, such as an omnidirectional microphone installed outside the vehicle, a shotgun-type directional condenser microphone, or an ultra-small MEMS microphone. Sound source localization using a microphone array or the like may be used to obtain sound data from which noise has been removed. The microphone's directivity, sampling frequency, frequency band, sensitivity characteristics, and other factors are appropriately selected according to the measurement target and sound field environment.

[0017] The target sound extraction unit 20 extracts the target sound (i.e., a horn sound) from the continuous sound data acquired by the sound signal acquisition unit 10 in substantially real time, and cuts out sound data containing the target sound. For example, if the test target is a horn, it is determined to be a horn sound based on its characteristic frequency band and sound pressure level. To cite a specific example, it determines that a sound with a frequency of 1 to 3 kHz and a sound pressure level of 75 dBA or higher is a horn sound, and extracts sound data for a period of approximately 2 to 5 seconds before and after the timing of detection of this sound.

[0018] In addition, a signal indicating that the inspector has pressed the horn switch may be obtained via the vehicle information acquisition unit 40, and sound data for the period when the horn switch is on may be extracted based on this signal.

[0019] The sound signal processing unit 30 processes the sound data extracted by the target sound extraction unit 20 using, for example, FFT or the like, to analyze the frequency characteristics. This results in spectrum data with frequency and sound pressure level as parameters. It then determines whether peak frequencies with sufficient sound pressure levels to be considered fundamental frequencies or overtones exist within the frequency windows (i.e., frequency bands) set for each of horns A and B. Based on the relationship between the number of peak frequencies appearing in the frequency window for horn A and the number of peak frequencies appearing in the frequency window for horn B, the partial vibration detection unit 31 determines whether both horns A and B are vibrating, or whether only one of horns A and B is vibrating, and if so, which horn is vibrating. The processing by the sound signal processing unit 30 will be described in more detail below.

[0020] The vehicle information acquisition unit 40 acquires various information about the vehicle to be inspected from a server that serves as a vehicle database or from the vehicle itself. The vehicle information includes information about the vehicle itself, such as the vehicle identification number and production control number, and multiple parameters that indicate the vehicle's operating state, such as vehicle speed and engine RPM, obtained via CAN communication or the like. For example, dynamic time-series vehicle signals, such as a steering signal when the inspector steers the vehicle, an accelerator pedal position signal, and a brake pedal operation signal, are also acquired via CAN communication or the like. In addition, information such as the vehicle model of the vehicle to be inspected, the model of the horn device 1 installed in this vehicle model, and the fundamental frequencies of horns A and B included in the horn device 1 are acquired from the vehicle database via the vehicle information acquisition unit 40.

[0021] The memory unit 50 is a storage means including a hard disk drive or the like for storing sound data acquired via a microphone, vehicle data (such as the vehicle identification number and vehicle specifications), data calculated for horn judgment, information such as threshold values, judgment result data, and calculation procedures. It is configured by a server computer in the factory included in the finished vehicle inspection device or a cloud computer connected via a communication line. Recorded past data is useful for ensuring inspection traceability and can also be used for future deep learning. The stored data may also include various related information such as environmental information (such as inspection date and time, weather, temperature, and inspector name) and equipment information (such as equipment name and equipment operating status).

[0022] The display unit 60 is a means for displaying information such as acquired data and inspection results to relevant parties such as the inspector driving the vehicle and the production manager, and is configured, for example, with a liquid crystal display, an organic EL display, various indicator lights, etc. It is not limited to visual devices, and may also be a device that displays audio information, a haptic device that presents tactile information to the hand or arm, etc.

[0023] Next, the detection of one-sided sound from horns A and B by sound signal processing unit 30 will be described. Fig. 2 shows a frequency characteristic diagram (spectrum) obtained by processing the sound from horn device 1, which is a double horn, using FFT (Fast Fourier Transform). Note that Fig. 2 is an explanatory diagram that schematically shows frequency characteristics for the purpose of explanation. The horizontal axis represents frequency, and the vertical axis represents sound pressure level (which may be sound power) equivalent to sound pressure.

[0024] Figure 2(a) shows the frequency characteristics when only horn A, whose fundamental frequency is, for example, 500 Hz, is sounding. In addition to the peak corresponding to the fundamental frequency of 500 Hz, multiple peaks appear corresponding to multiple harmonics, such as the second harmonic, third harmonic, etc.

[0025] Figure 2(b) shows the frequency characteristics when only horn B, whose fundamental frequency is, for example, 400 Hz, is sounding. In addition to the peak corresponding to the fundamental frequency of 400 Hz, multiple peaks appear corresponding to multiple harmonics, such as the second harmonic, third harmonic, etc.

[0026] Figure 2(c) shows the frequency characteristics when both horns A and B are sounding simultaneously (i.e., the normal state of a double horn system). In this case, multiple peaks from horn A and multiple peaks from horn B appear overlapping each other. In response to this, frequency window A, shown as rectangular windows in Figure 2(c), is set to correspond to the fundamental frequency and harmonic frequencies of horn A, respectively, and frequency window B, shown as rectangular windows in Figure 2(c), to correspond to the fundamental frequency and harmonic frequencies of horn B, respectively. Peaks with sufficient sound pressure that appear within each frequency window are counted as peak frequencies. As shown in the figure, the peaks of horn A's fundamental frequency and harmonic frequency appear primarily within frequency window A, and similarly, the peaks of horn B's fundamental frequency and harmonic frequency appear primarily within frequency window B.

[0027] Therefore, if both horn A and horn B are sounding correctly, the number of peak frequencies appearing in frequency window A and the number of peak frequencies appearing in frequency window B will both roughly match the number of harmonics in each (for example, if frequency analysis is performed within the range of 0 to 10 kHz, the number of harmonics included in that range).

[0028] In contrast, if horn A is sounding without horn B being sounded, the number of peak frequencies appearing in frequency window A will nearly match the number of harmonics of horn A, while the number of peak frequencies appearing in frequency window B will be significantly fewer. Conversely, if horn A is not sounding and only horn B is sounding, the number of peak frequencies appearing in frequency window B will nearly match the number of harmonics of horn B, while the number of peak frequencies appearing in frequency window A will be significantly fewer. Note that if the frequency of a certain harmonic of horn A matches or is close to the frequency of a certain harmonic of horn B, peak frequencies will appear in the frequency window corresponding to the silent horn, but the number of such peak frequencies will be small. Furthermore, small peaks may appear in the frequency window corresponding to the silent horn due to the generation of small sounds at frequencies other than harmonics, but this can be ignored by setting an appropriate sound pressure level as the threshold.

[0029] Therefore, in more specific processing, peaks that appear in each frequency window and are equal to or greater than a certain sound pressure level threshold are counted as peak frequencies, and the number of these peak frequencies is compared with an appropriate peak number threshold to determine whether horns A and B are sounding.

[0030] The sound pressure level thresholds are preferably set to appropriate values ​​for each frequency window to correspond to the differences in the fundamental loudness of each harmonic. For example, if the harmonic that produces the maximum sound pressure level of the horn sound is the fifth harmonic, the sound pressure level threshold for the frequency window for the fifth harmonic is set to the highest value, and the sound pressure level thresholds for the frequency windows for the other harmonics are also individually set to correspond to the sound pressure levels of the individual harmonics. Such horn characteristics can be acquired in association with vehicle model information.

[0031] The above-mentioned frequency windows A and B are set based on the fundamental frequencies of horns A and B. The fundamental frequencies are acquired via vehicle information acquisition unit 40. If the fundamental frequency of horn A is 500 Hz and the fundamental frequency of horn B is 400 Hz, in one example, the intermediate frequency of 450 Hz is used as the boundary, and the fundamental frequency window A for horn A is set to 450 to 550 Hz, and the fundamental frequency window B for horn B is set to 350 to 450 Hz, and frequency windows A and B with similar frequency widths are set for each harmonic.

[0032] To detect individual peaks more precisely, narrower frequency windows can be used. For example, to correspond to the frequency width of a horn sound, a frequency range of ±20 Hz can be set, with frequency window A for horn A ranging from 480 to 520 Hz and frequency window B for horn B ranging from 380 to 420 Hz.

[0033] Furthermore, since the frequency width (band width) of the peak sound differs depending on the type of horn (so-called flat type and spiral type), the frequency width may be set in accordance with the characteristics of those types.

[0034] As mentioned above, the fundamental frequency can be obtained from the model information of the horn device 1 included in the vehicle model information, but generally, for automobile horns, a variation of about ±10% is allowed in the fundamental frequency. Therefore, in order to improve the accuracy of peak frequency detection, it is desirable to obtain the actual fundamental frequency from the sound picked up by a microphone and use this measurement result to correct the design fundamental frequency.

[0035] Specifically, by acquiring the frequency characteristics of each horn as shown in Figure 2(c) and reading the frequency at which the peak occurs within the frequency range of this frequency characteristic, the actual fundamental frequency near the fundamental frequency of the design value can be determined. For example, if the detected fundamental frequency is off by +10 Hz from the design value of 400 Hz, the fundamental frequency that serves as the reference for setting the frequency window is corrected to 410 Hz.

[0036] The peak is determined as a point on a frequency characteristic graph such as that shown in FIG. 2 where the curve becomes an upward convex function and the differential value of the curve is zero.

[0037] Incidentally, automobile horns are designed in principle to produce a peak sound with the maximum sound pressure level at some harmonic frequency due to the resonance of vibrations at the fundamental frequency, and to produce a sound at the legally specified sound pressure level (87 to 115 dB).

[0038] FIG. 4 is an explanatory diagram showing an example of a display on the display unit 60 of the first embodiment. This example shows an example of a one-sided sound in which no sound is coming from horn B. On the left side of the screen, to show the measurement results, a frequency characteristic graph 41 showing the sound pressure levels in multiple frequency windows A for horn A and a frequency characteristic graph 42 showing the sound pressure levels in multiple frequency windows B for horn B are arranged one above the other. The sound pressure level threshold described above is added to each graph as "threshold." On the screen display, this sound pressure level threshold is shown as a constant value, but as mentioned above, it actually has a different value for each frequency window.

[0039] Horn A is sounding correctly, and as shown in the figure, peak frequencies above the sound pressure level threshold corresponding to the fundamental frequency and harmonic frequencies appear in frequency window A. Table 43 on the right side of the screen indicates that 18 peak frequencies were counted.

[0040] In frequency window B for silent horn B, small peaks appear as small sounds other than the harmonics contained in the sound of horn A. However, there are few peaks above the sound pressure level threshold, and the table on the right side of the screen shows that four peak frequencies were counted.

[0041] In other words, at 2000 Hz, 4000 Hz, 6000 Hz, and 8000 Hz, which are the same frequencies that are multiples of the fundamental frequencies of Horn A and Horn B, peak frequencies with high sound pressure levels also appear in frequency window B.

[0042] Table 43 on the right side of the screen shows that the threshold for the number of peak frequencies is "10" as "Threshold > 10". Therefore, Horn A is judged to be sounding correctly and a "○" mark is displayed, while Horn B is judged not to be sounding and a "×" mark is displayed. Furthermore, the judgment result is displayed in text as "One-sided ringing: present".

[0043] Such a display allows inspectors and other related parties to easily know that the horn device 1 of the vehicle being inspected is producing one-sided noise, and further allows them to see frequency data, criteria for judgment, and the like.

[0044] Figure 3 is a flowchart showing the processing flow of the horn inspection device of the first embodiment. First, the microphone of the sound signal acquisition unit 10 collects the horn sound from a vehicle test-running on a free roller and acquires it as a sound signal (Step 1). Next, the target sound extraction unit 20 separates and extracts the horn sound (Step 2).

[0045] In step 3, the fundamental frequencies of the horns A and B to be inspected are identified. As mentioned above, the fundamental frequency contained in the vehicle model information is used as the basis, and the fundamental frequency of the design value is further corrected by frequency analysis and peak detection of the extracted horn sound. Frequency windows A and B are set using this fundamental frequency.

[0046] In step 4, the extracted horn sound is subjected to frequency analysis using FFT, and the number of peaks that appear within frequency windows A and B is calculated. Here, as mentioned above, the sound pressure level of each peak is compared with the sound pressure level threshold, and peaks below the sound pressure level threshold are not counted.

[0047] In step 5, the number of peak frequencies that appeared in frequency window A and the number of peak frequencies that appeared in frequency window B are each compared with a threshold value (for example, 10) to determine whether either is below the threshold value. If either is below the threshold value, it is determined in step 6 that "single chirp is present," and if both are above the threshold value, it is determined in step 7 that "single chirp is not present." Then, the process proceeds to step 8, where a screen such as that shown in FIG. 4 is displayed on the display unit 60, and data including the determination result is stored in the storage unit 50.

[0048] Next, a horn inspection device according to a second embodiment will be described with reference to FIGS. 5 to 8. The following mainly describes the differences from the first embodiment. In addition to determining whether the double horn is oscillating as described in the first embodiment, the horn inspection device according to the second embodiment determines whether the horn sounds of horns A and B contain abnormal sounds and displays the results. In other words, in the case of a horn, if sounds other than harmonics of the fundamental frequency are contained, the sound quality deteriorates and is perceived by humans as a muddy sound or a buzzing sound. The second embodiment determines whether such abnormal sounds are present.

[0049] 5 shows a functional block diagram of a horn inspection device according to the second embodiment. Similar to the first embodiment, the horn inspection device according to the second embodiment includes a sound signal acquisition unit 10, a target sound extraction unit 20, a sound signal processing unit 30 including a one-sided sound detection unit 31, a vehicle information acquisition unit 40, a storage unit 50, and a display unit 60. In the second embodiment, the sound signal processing unit 30 includes an abnormal sound determination unit 32.

[0050] In addition to the frequency bands corresponding to horn A and horn B, i.e., frequency windows A and B, abnormal sound detection unit 32 sets a plurality of frequency bands for abnormal sound detection (referred to as frequency window C) whose center frequencies do not overlap with those of frequency windows A and B, and determines whether the sounds emitted by horns A and B contain abnormal sounds based on the number of peak frequencies that appear in this frequency band for abnormal sound detection (frequency window C).

[0051] In one embodiment, the frequency window C is set to a frequency band that does not overlap with the frequency window A and the frequency window B (that is, the remaining portion) in the frequency characteristics that are the spectrum.

[0052] Figure 6 is an explanatory diagram of frequency characteristics similar to Figure 2, where (a) shows the frequency characteristics when only horn A, whose fundamental frequency is, for example, 500 Hz, is sounding, and (b) shows the frequency characteristics when only horn B, whose fundamental frequency is, for example, 400 Hz, is sounding.

[0053] Figure 6(c) shows the frequency characteristics when both horn A and horn B are sounding simultaneously, and similar to Figure 2, frequency windows A and B are indicated by rectangular frames. Although not shown by a frame, the frequency band between frequency window A and frequency window B is set as frequency window C.

[0054] As explained in the first embodiment, the presence of a single ringing sound is determined by counting the number of peak frequencies with sufficient sound pressure levels that appear in frequency window A and frequency window B. In the second embodiment, the number of peak frequencies with sufficient sound pressure levels that appear in frequency window C is counted to determine whether an abnormal sound is present, and if this number is equal to or greater than a predetermined threshold, it is determined that an abnormal sound is present. Note that instead of comparing the number of peak frequencies in frequency window C with a threshold, the occurrence rate relative to the number of frequency windows C may be compared with a threshold. An appropriate sound pressure level threshold is also set for peaks in frequency window C, and peaks below this sound pressure level threshold are not counted. Note that the sound pressure level threshold for frequency window C is set to a relatively low sound pressure level, unlike the sound pressure level thresholds for frequency windows A and B.

[0055] By processing using such a frequency window C, it is possible to determine whether or not an abnormal sound outside the harmonic range is included.

[0056] FIG. 8 is an explanatory diagram showing an example of a display on the display unit 60 of the second embodiment. This example shows an example in which horn B is not producing sound, resulting in a one-sided ringing sound, and the sound from horn A, which is ringing, contains an abnormal sound. To show the measurement results, a frequency response graph 81 showing the sound pressure levels in multiple frequency windows A for horn A and a frequency response graph 82 showing the sound pressure levels in multiple frequency windows B for horn B are arranged vertically on the left side of the screen. The sound pressure level threshold described above is indicated by a dashed line on each graph. While the sound pressure level threshold is shown as a constant value on the screen, as mentioned above, it actually varies for each frequency window.

[0057] Table 84 at the bottom right of the screen has the same content as Table 43 in the display example of the first embodiment in Figure 4. In this display example, as explained in the first embodiment, it has been determined that horn B is not sounding, that is, one-sided sounding.

[0058] In the second embodiment, a frequency characteristic graph 83 showing the sound pressure level in frequency window C is displayed above graphs 81 and 82. A sound pressure level threshold indicated by a dashed line is added to this graph. As described above, the number of frequency peaks equal to or greater than this sound pressure level threshold is counted, and table 85 in the upper right corner of the screen indicates that the number of frequency peaks is 12. Table 85 also indicates that the threshold for the number of frequency peaks is "10" as "Threshold < 10". Therefore, it is determined that the horn sound contains an abnormal sound, and an "x" mark is displayed. Furthermore, the determination result is displayed in text as "Abnormal sound: present."

[0059] Such a display allows inspectors and other relevant personnel to easily know whether the horn device 1 of the vehicle being inspected is producing an abnormal noise as well as whether it is producing an uneven sound.

[0060] Figure 7 is a flowchart showing the processing flow of the horn inspection device of the second embodiment described above. The processing from step 1 to step 7 is the same as that of the first embodiment shown in Figure 3, and the presence or absence of one-sided ringing is determined based on the number of peak frequencies that appear in frequency windows A and B. In step 4, in addition to counting the number of peak frequencies that appear in frequency windows A and B, the number of peak frequencies that appear in frequency window C is also counted. Peaks below the sound pressure level threshold are not counted.

[0061] In step 9 following step 7, the number of peak frequencies that appear in frequency window C is compared with a threshold value (for example, 10) to determine whether it is equal to or greater than the threshold value. If it is equal to or greater than the threshold value, it is determined in step 10 that "abnormal noise is present," and if it is less than the threshold value, it is determined that "abnormal noise is not present" in step 11. Then, the process proceeds to step 12, where a screen such as that shown in FIG. 8 is displayed on display unit 60, and data including the determination result is stored in storage unit 50.

[0062] Next, a horn inspection device according to a third embodiment will be described with reference to Figures 9 to 12. In addition to determining whether a double horn is oscillating as described in the first embodiment and whether an abnormal sound is present as described in the second embodiment, the horn inspection device according to the third embodiment determines whether the abnormal sound comes from horn A or B, if any, and displays the determination.

[0063] For example, if the horn is not properly secured to the vehicle body or if the horn is deformed for some reason, secondary peaks that cause abnormal noise may occur at frequencies adjacent to the fundamental frequency and its overtones. In other words, for example, a sound of the fundamental frequency may be generated simultaneously with a sound of a slightly different frequency, which a person may perceive as an abnormal noise. Based on these characteristics, the third embodiment determines which of horns A and B is causing the abnormal noise, for example, when both horns A and B are sounding.

[0064] 9 shows a functional block diagram of a horn inspection device according to the third embodiment. Similar to the second embodiment, the horn inspection device according to the third embodiment includes a sound signal acquisition unit 10, a target sound extraction unit 20, a sound signal processing unit 30 including a one-sided sound detection unit 31 and an abnormal sound determination unit 32, a vehicle information acquisition unit 40, a storage unit 50, and a display unit 60. In the third embodiment, the sound signal processing unit 30 includes an abnormal sound component identification unit 33.

[0065] The abnormal noise component identification unit 33 determines the number of peak frequencies (for convenience, referred to as peak frequencies H) that appear in frequency window A corresponding to horn A and have a sound pressure level equal to or greater than a relatively high first sound pressure level threshold H, and the number of peak frequencies (for convenience, referred to as peak frequencies L) that have a sound pressure level equal to or greater than a relatively low second sound pressure level threshold L, and also determines the number of peak frequencies (peak frequencies H) that appear in frequency window B corresponding to horn B and have a sound pressure level equal to or greater than the relatively high first sound pressure level threshold H, and the number of peak frequencies (peak frequencies L) that have a sound pressure level equal to or greater than a relatively low second sound pressure level threshold L, and compares these in frequency window A and frequency window B to identify abnormal noise components.

[0066] The number of peak frequencies L includes the number of peak frequencies H. That is, peak frequencies H equal to or higher than the first sound pressure level threshold H are also counted as peak frequencies L.

[0067] In one embodiment, the number of peak frequencies L related to frequency window A (for convenience, referred to as the window A peak frequency number) is compared with the number of peak frequencies L related to frequency window B (for convenience, referred to as the window B peak frequency number), and assuming that an abnormal noise is occurring, if the window A peak frequency number and the window B peak frequency number are not similar and the window A peak frequency number is relatively large, horn B is identified as the abnormal sound source, and if the window B peak frequency number is relatively large, horn A is identified as the abnormal sound source.If the window A peak frequency number and the window B peak frequency number are similar, both horn A and horn B are identified as abnormal sound sources.

[0068] The first sound pressure level threshold H can be set to the same value as the sound pressure level threshold for detecting single-tone vibration in the first embodiment, and the second sound pressure level threshold L is set to a lower value. It is also desirable to set the second sound pressure level threshold L to a value appropriate for each harmonic.

[0069] Figure 10 is an explanatory diagram of frequency characteristics similar to Figure 2, where (a) shows the frequency characteristics when only horn A, which has a fundamental frequency of, for example, 500 Hz, is sounding, and (b) shows the frequency characteristics when only horn B, which has a fundamental frequency of, for example, 400 Hz, is sounding. In this example, horn A emits abnormal noise, and abnormal noise peaks with relatively small sound pressure levels can be seen adjacent to the peaks of the fundamental frequency and harmonic frequencies.

[0070] In Figure 10, frequency window A is drawn superimposed on the frequency characteristics of horn A, and frequency window B is drawn superimposed on the frequency characteristics of horn B. Furthermore, the first sound pressure level threshold H and the second sound pressure level threshold L are drawn superimposed on the respective frequency characteristics, assuming that they are constant levels.

[0071] As can be seen from the relationship between the peaks in the frequency characteristics of horn A and frequency window A, peaks of the original fundamental frequency and harmonic frequencies appear in frequency window A, and their sound pressure levels are equal to or higher than sound pressure level threshold H. In contrast, relatively small sound pressure level peaks occurring adjacent to the fundamental frequency and harmonic frequencies that indicate the abnormal noise emitted by horn A are not detected in frequency window A. However, because frequency window B is in a different frequency band from frequency window A, at least some of the sound pressure level peaks that indicate the abnormal noise appear in frequency window B. Therefore, if these peaks have sound pressure levels higher than second sound pressure level threshold L, they are counted as peak frequencies L for frequency window B, and the number of window B peak frequencies exceeds the number of peaks of the original fundamental frequency and harmonic frequencies of horn B (which are counted as peak frequencies H for frequency window B). Peaks lower than second sound pressure level threshold L have little effect as actual abnormal noise and can be ignored.

[0072] The same is true when horn B contains abnormal noise; at least some of the relatively small sound pressure level peaks that result in abnormal noise occurring near the fundamental frequency or harmonic frequencies of horn B appear in frequency window A, not frequency window B.

[0073] Using this method, abnormal noise component identification unit 33 can identify the horn in a double horn that is the source of the abnormal noise. Note that this component identification is based on the premise that it has been determined that the horn sound from horn device 1 as a whole contains an abnormal noise, as in the second embodiment. Furthermore, if the horn sound is one-sided and contains an abnormal noise, then naturally the ringing horn is the source of the abnormal noise.

[0074] 12 is an explanatory diagram showing an example of a display on the display unit 60 of the third embodiment. This example shows a case where sound is coming from both horns A and B, and the sound from horn A contains an abnormal noise. Table 84 at the bottom right of the screen and table 85 at the top right are the same as tables 84 and 85 of the second embodiment shown in FIG. 8, and display the presence or absence of intermittent ringing and abnormal noise together with the number of peak frequencies in each frequency window A, B, and C and their threshold values.

[0075] In the third example, information related to identifying the abnormal noise component and the judgment results are displayed on the left side of the screen as table 121. This table 121 is pre-populated with three rows: "Both horn A and horn B make abnormal noise," "Only horn A makes abnormal noise," and "Only horn B makes abnormal noise." The corresponding row is indicated by a circle or an cross in the right column. The corresponding row, "Only horn A makes abnormal noise," is displayed in red. The center column lists the number of peak frequencies in window A and window B, along with the number of peak frequencies H, as a "judgment method." In the table, "window B threshold level L = 38" indicates that the number of peak frequencies in window B is 38, and "window A threshold level L = 20" indicates that the number of peak frequencies in window A is 20. "window B threshold level H" and "window A threshold level H" indicate the number of peak frequencies H in frequency window B and frequency window A, respectively.

[0076] In the example of Fig. 12, since the number of peak frequencies in window B is large, horn A is determined to be the foreign sound source. Any specific method may be used to determine whether the foreign sound source is present from the number of peaks, and for example, it can be determined based on whether the number of peak frequencies in window B is larger or smaller than the allowable range (for example, ±20%) of the number of peak frequencies in window A. Alternatively, a comparison of the number with an appropriate threshold value may be made.

[0077] Note that because the fundamental frequencies of Horn A and Horn B are different, if frequency analysis is performed within the range of, for example, 0 to 10 kHz, the number of harmonics included within that range will differ to begin with. Therefore, when evaluating the number of peak frequencies in Window A and the number of peak frequencies in Window B, it is desirable to take into account the difference in the number of fundamental harmonics.

[0078] Figure 11 is a flowchart showing the processing flow of the horn inspection device of the third embodiment described above. The processing from step 1 to step 11 is the same as that of the second embodiment shown in Figure 7, and the presence or absence of one-sided ringing is determined based on the number of peak frequencies that appear in frequency windows A and B, and the presence or absence of abnormal noise is determined based on the number of peak frequencies that appear in frequency window C. If it is determined that there is no abnormal noise, there is no need to identify the source of the abnormal noise, and the process proceeds from step 11 to step 18, where the determination results and the like are displayed and the data are saved.

[0079] If it is determined that abnormal noise is present, the process proceeds from step 10 to step 13, where it is determined whether the number of peak frequencies in window A and the number of peak frequencies in window B are similar to each other. If they are similar, it is determined in step 14 that both horn A and horn B are producing abnormal noise.

[0080] If the two are not similar, the process proceeds to step 15, where it is determined whether the number of window A peak frequencies is equal to or greater than a predetermined threshold. The threshold is set appropriately, taking into consideration the number of fundamental frequencies and harmonic frequencies included in the frequency range for which frequency analysis was performed. If the number of window A peak frequencies is equal to or greater than the threshold, then in step 16, horn B is identified as the abnormal sound source. If the number of window A peak frequencies is less than the threshold, then in step 17, horn A is identified as the abnormal sound source. Then, in step 18, a screen such as the one shown in FIG. 12 is displayed on display unit 60, and data including the determination result is stored in memory unit 50.

[0081] The above describes an embodiment in which the present invention is applied to the inspection of double horns in the finished vehicle inspection process, but the present invention is not limited to the above embodiment and can be applied in a variety of ways. For example, the present invention can be widely applied to the inspection of sound-generating devices that emit two sounds simultaneously, such as warning sound generators, in addition to automobile horn devices, and can also be applied to inspections in repair shops, not limited to finished vehicle inspection processes. Furthermore, sound-generating devices are not limited to those that emit simple sounds, such as horn devices. [Explanation of symbols]

[0082] 1...Horn device 10...Sound signal acquisition unit 20...Target sound extraction section 30...Sound signal processing section 31...One-sided sound detection unit 40...Vehicle information acquisition unit 50...Storage section 60...Display section

Claims

1. A method for inspecting a sound-generating device including a plurality of sound-generating bodies that each emit sounds of different frequencies, comprising: Acquire the sound emitted by the sound generator, The frequency characteristics of this acquired sound are analyzed, Separating the sound components into a plurality of frequency bands for each sound generator corresponding to the fundamental frequency of each sound generator and the frequencies of its overtones; determining whether each of the plurality of sound producing bodies is emitting sound based on the number of peak frequencies that appear in the frequency bands corresponding to each sound producing body; A method for inspecting a sound-producing device.

2. The sound generating device is a horn device including a plurality of horns provided in a vehicle. A method for inspecting the sound-producing device according to claim 1.

3. Identifying the vehicle model of the vehicle to be inspected and setting the frequency band using the fundamental frequency of each horn that is predetermined according to the vehicle model. A method for inspecting a sound-producing device according to claim 2.

4. The actual peak frequency of the fundamental frequency is calculated from the acquired sound. Using this actual peak frequency, a fundamental frequency that is predetermined depending on the vehicle model is corrected to set the frequency band. The method for inspecting the sound-producing device according to claim 3.

5. In addition to the frequency bands corresponding to the individual sound-producing bodies, a plurality of frequency bands for detecting abnormal noise are set, the center frequencies of which do not overlap with the frequency bands described above. Further, based on the number of peak frequencies that appear in this frequency band for detecting abnormal sounds, it is determined whether or not the sounds emitted by the multiple sound-producing bodies contain abnormal sounds. A method for inspecting the sound-producing device according to claim 1.

6. Identifying which sound generator is emitting the abnormal sound based on the number of peak frequencies having a sound pressure equal to or greater than a relatively high first sound pressure threshold and the number of peak frequencies having a sound pressure equal to or greater than a relatively low second sound pressure threshold that appear in multiple frequency bands for each sound generator.

6. A method for inspecting a sound-producing device according to claim 5.

7. Used to inspect the horn device in the finished vehicle inspection process, The inspection results are displayed together with related information on an output device in the finished vehicle inspection process, and are also stored in a storage device. A method for inspecting a sound-producing device according to claim 2.

8. Counting the peaks having a sound pressure equal to or greater than a predetermined sound pressure threshold as peak frequencies among the peaks included in each frequency band. A method for inspecting the sound-producing device according to claim 1.

9. The sound pressure threshold is set to a different value for each frequency band depending on the magnitude of the fundamental sound pressure of each harmonic. The method for inspecting a sound-producing device according to claim 8.

10. An inspection device for a sound-generating device including a plurality of sound-generating bodies that each emit sounds of different frequencies, a sound signal acquisition unit that acquires a sound emitted from the sound generating device and generates a sound signal; a sound signal processing unit that processes the sound signal; Equipped with The sound signal processing unit includes: Analyzing the frequency characteristics of the acquired sound, separating the sound components into a plurality of frequency bands provided for each sound generator corresponding to the fundamental frequency of each sound generator and the frequencies of its overtones, and determining whether each of the plurality of sound generators is emitting sound based on the number of peak frequencies that appear in the frequency band corresponding to each sound generator. Testing equipment for sound generation devices.

Images