Sound source estimation device

Abstract

To provide a sound source estimation device that estimates a sound source of noise without making a user feel uncomfortable.SOLUTION: A sound source estimation device that estimates one of a plurality of components mounted on an object as a candidate for a sound source of noise based upon the frequency of sound generated by the object and the rotating speed of a rotary device mounted on the object estimates one of the plurality of components as the candidate for the sound source of the noise when a correlation value correlating with the amount of variation in the rotating speed within a predetermined time is equal to or larger than a threshold, but estimates no candidate for the sound source of the noise or estimates one of the plurality of components as a candidate which may be the candidate for the sound source of the noise with a less possibility when the correlation value is smaller than the threshold.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a sound source estimation device. [Background technology]

[0002] There is a technology that estimates one of multiple parts mounted on an object as a candidate source of abnormal noise based on the frequency of the sound generated by the object and the rotation speed of a rotating device mounted on the object (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2022-100139 Summary of the Invention [Problem to be solved by the invention]

[0004] When the rotation speed of a rotating device does not change, the frequency of the sound generated from a specific part remains constant. Therefore, it is difficult for a user to recognize such a sound as an abnormal noise. If a part that generates such a sound is estimated as a candidate for the source of the abnormal noise, it may cause discomfort to the user.

[0005] Therefore, an object of the present invention is to provide a sound source estimation device that estimates the sound source of an abnormal sound without giving a sense of discomfort to the user. [Means for solving the problem]

[0006] The above object can be achieved by a sound source estimation device that estimates one of a plurality of parts mounted on an object as a candidate for the source of abnormal noise based on the frequency of sound generated by the object and the rotation speed of a rotating device mounted on the object, wherein if a correlation value that correlates with an amount of change in the rotation speed within a predetermined time is equal to or greater than a threshold value, the sound source estimation device estimates one of the plurality of parts as a candidate for the source of abnormal noise, and if the correlation value is less than the threshold value, the sound source estimation device does not estimate one of the plurality of parts as a candidate for the source of abnormal noise, or estimates one of the plurality of parts as having a low possibility of being a candidate for the source of abnormal noise. [Effects of the Invention]

[0007] According to the present invention, it is possible to provide a sound source estimation device that estimates the sound source of an abnormal sound without giving a sense of discomfort to the user. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1A is a diagram showing a vehicle, which is an object for which an abnormal sound is estimated by the sound source estimation device, FIG. 1B is a diagram showing the configuration of the sound source estimation device, and FIG. 1C is a flowchart outlining the sound source estimation method. [Figure 2] FIG. 2 is a flowchart illustrating the extraction process. [Figure 3] FIG. 3 is a timing chart illustrating the change over time in the engine rotation speed and the motor rotation speed. DETAILED DESCRIPTION OF THE INVENTION

[0009] FIG. 1A is a diagram showing a vehicle 200, which is an object for which an abnormal sound is estimated by the sound source estimation device 100. The vehicle 200 is equipped with an engine 210, a motor 220, and a plurality of other components. The engine 210 and the motor 220 each generate a rotational motion. The plurality of components generate sound and vibration in association with the rotational motion of at least one of the engine 210 and the motor 220. The order of the sound generated by each component in association with the rotational motion of the engine 210 or the motor 220 is determined by the specifications of each component, and is an order specific to each component (hereinafter referred to as the component order).

[0010] The sound source estimation device 100 includes a microphone 10 and detects sounds around the sound source estimation device 100. The vehicle 200 has an engine 210 and a motor 220 mounted in the front compartment. When a user of the vehicle 200 senses that an abnormal sound is coming from the vehicle 200, the user of the sound source estimation device 100 uses the sound source estimation device 100 to estimate the source of the abnormal sound. When estimating the source of the abnormal sound, the user opens the hood of the vehicle 200 and brings the sound source estimation device 100 close to the front compartment. This allows sounds around the engine 210 and the motor 220 to be detected by the microphone 10, and the sound source of the abnormal sound to be estimated by the sound source estimation device 100.

[0011] 1B is a diagram showing the configuration of sound source estimation device 100. Sound source estimation device 100 is configured with a microphone 10, a timer 20, a rotation speed acquisition unit 30, a data recording unit 40, a playback unit 50, an input unit 60, a display unit 70, a component estimation unit 80, and a data editing unit 90. The rotation speed acquisition unit 30, the component estimation unit 80, and the data editing unit 90 are realized by, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory).

[0012] The microphone 10 detects ambient sounds and generates sound data. The timer 20 measures a detection time, which is the time during which sound is detected by the microphone 10. The rotation speed acquisition unit 30 acquires the rotation speeds of the engine 210 and the motor 220 and generates rotation speed data for each. The data recording unit 40 is a memory that records the above-mentioned sound data, detection time, rotation speed data, etc. The playback unit 50 is a device such as a speaker that plays back sound data and extracted sound data (described later) as sound. The input unit 60 is a switch or the like that accepts user operations related to control of the sound source estimation device 100. The display unit 70 is a liquid crystal screen or the like that displays information related to control of the sound source estimation device 100 and control options to the user.

[0013] The part estimation unit 80 estimates candidate sound sources of abnormal noise based on orders calculated from analysis sound data and analysis rotation speed data, which will be described later. The part estimation unit 80 calculates the order of a sound estimated to be an abnormal noise by executing processing, which will be described later, on the analysis sound data and analysis rotation speed data. The part estimation unit 80 acquires part order information for each of a plurality of parts that emit sound in conjunction with the rotation of the engine 210 and the motor 220. The part order of each part is determined by the number of teeth, for example, if the part is a gear. The part order information for each part is stored in advance in the data recording unit 40. The part estimation unit 80 acquires the part order information for each part from the data recording unit 40. The part estimation unit 80 compares the part order information for each part with the calculated order of the abnormal noise, thereby estimating candidate parts that are the sound sources of the abnormal noise.

[0014] The data editing unit 90 edits the sound data and rotation speed data recorded in the data recording unit 40 to generate analysis sound data and analysis rotation speed data. The data editing unit 90 performs frequency analysis on the sound data. The sound data detected by the microphone 10 represents changes in sound pressure over time. The data editing unit 90 performs a fast Fourier transform (FFT) on this sound data to generate frequency sound data that represents changes in the frequency spectrum over time. The data editing unit 90 performs extraction processing to extract any frequency sound data, such as frequency sound data in a specific time range or frequency sound data in a specific frequency band, from the generated frequency sound data, and generates extracted sound data.

[0015] 1C is a flowchart outlining the sound source estimation method. The sound source estimation device 100 acquires sound data (step S1), and acquires rotation speed data corresponding to the acquired sound data (step S2). Next, the sound source estimation device 100 performs frequency analysis on the sound data to generate frequency sound data (step S3), and performs extraction processing on the frequency sound data and rotation speed data (step S4). Next, the sound source estimation device 100 estimates parts that are candidate sound sources of abnormal noise by performing analysis based on the extracted frequency sound data and rotation speed data (step S5), and outputs information about the parts that are candidate sound sources (step S6).

[0016] In step S1, the user brings the microphone 10 of the sound source estimation device 100 close to the running engine 210 and motor 220 and performs an operation to start sound detection on the input unit 60. This starts sound detection by the microphone 10, and detects sounds around the running engine 210 and motor 220 for a predetermined time. The detected sound is recorded in the data recording unit 40 as sound data. Furthermore, the detection time measured by the timer 20 is linked to the sound data and recorded in the data recording unit 40.

[0017] In step S2, the rotation speed acquisition unit 30 acquires the rotation speeds of the engine 210 and the motor 220, for example, from an ECU (Electronic Control Unit) of the vehicle 200. The rotation speed acquisition unit 30 acquires the rotation speeds of the engine 210 and the motor 220 at times corresponding to the detection times recorded in the data recording unit 40, and generates rotation speed data for each. The generated rotation speed data is linked to the sound data and recorded in the data recording unit 40.

[0018] In step S3, the data editing unit 90 performs FFT on the input sound data to generate frequency sound data. In step S4, the data editing unit 90 executes an extraction process, which will be described later, and outputs the frequency sound data and the rotation speed data to the part estimation unit 80 as analysis sound data and analysis rotation speed data, respectively.

[0019] In step S5, the part estimation unit 80 calculates the order of the sound estimated to be an abnormal noise based on the input analysis sound data and analysis rotation speed data. Specifically, the maximum sound frequency, which is the frequency indicating the greatest sound pressure, is extracted for each predetermined time period to generate maximum sound frequency data. This is because the maximum sound frequency is likely to be the frequency of the abnormal noise felt by the user. The part estimation unit 80 calculates the order of the maximum sound frequency data based on the maximum sound frequency data and the analysis rotation speed. Specifically, linear approximation is performed using the least squares method or the like on multiple sets of values ​​of the maximum sound frequency data and the analysis rotation speed data for each predetermined time period. The slope of the straight line obtained by linearly approximating these values ​​is calculated as the order. The part estimation unit 80 estimates parts that are candidate sources of the abnormal noise based on the calculated order. Specifically, the part orders of parts mounted on the vehicle 200 obtained from the data recording unit 40 are compared with the calculated order, and the part having the part order closest to the calculated order is estimated to be a candidate source of the abnormal noise.

[0020] In step S6, the part estimation unit 80 displays information about the part estimated to be a candidate for the source of the abnormal noise on the display unit 70. Information displayed on the display unit 70 includes, for example, the part's name, shape, mounting position, inspection method, repair method, replacement method, etc. The user can inspect, repair, replace, etc., the part estimated to be a candidate for the source of the abnormal noise as necessary.

[0021] Step S4 will be described in detail. FIG. 2 is a flowchart illustrating the extraction process. The data editing unit 90 displays frequency sound data on the display unit 70 (step S41) and waits for an operation to be input by the user. The user selects a desired frequency band or a desired time range from the frequency sound data (step S42). The data editing unit 90 extracts any frequency sound data selected by the user from the sound data to generate extracted sound data. Next, the data editing unit 90 generates extracted rotation speed data by extracting the time range selected by the user from the rotation speed data (step S43). Next, the sound source estimation device 100 plays the extracted sound data on the playback unit 50 (step S44). This allows the user to confirm the extracted sound.

[0022] Next, the sound source estimation device 100 displays options such as "Sounds other than abnormal sounds are included" and "Sounds other than abnormal sounds are not included" on the display unit 70 (step S45). The user determines whether or not the sound confirmed in step S44 includes noise other than abnormal sounds (step S46), and selects one of the options using the input unit 60. If "Sounds other than abnormal sounds are included" is selected (step S47), the sound source estimation device 100 displays the frequency sound data again (step S41) and returns to a state of waiting for an operation to be input by the user.

[0023] If "No sounds other than abnormal noises are included" is selected (step S48), the data editing unit 90 sorts out extracted rotation speed data in which the rotation speed has changed within the selected time range, and extracted rotation speed data in which the rotation speed has not changed within the selected time range (step S49-1). For example, if the amount of rotation speed within the selected time range is equal to or greater than a threshold, the data is sorted as extracted rotation speed data in which the rotation speed has changed. If the amount of change in rotation speed within the selected time range is less than the threshold, the data is sorted as extracted rotation speed data in which the rotation speed has not changed.

[0024] FIG. 3 is a timing chart illustrating the time variations of engine speed and motor speed. Within the selected time range from time t1 to t2, the engine speed remains almost constant, while the motor speed gradually increases. Therefore, the amount of change in engine speed is less than the threshold, and the data is selected as extracted rotation speed data indicating no change in rotation speed. The amount of change in motor speed is equal to or greater than the threshold, and the data is selected as extracted rotation speed data indicating a change in rotation speed. Here, the amount of change is, for example, the value obtained by integrating the rotation speed with respect to the average value of the rotation speed within the selected time range. Instead of the amount of change in rotation speed, for example, the length of the rotation speed trajectory within the selected time range may be used. In other words, any value other than the amount of change described above may be used as long as it correlates with the amount of change in rotation speed. Potential sources of abnormal noise caused by engine speed include, for example, an engine chain and a balance shaft. Potential sources of abnormal noise caused by motor speed include, for example, a counter gear.

[0025] Next, the data editing unit 90 sets the extracted rotation speed data in which the rotation speed has changed and the corresponding extracted sound data as analysis rotation speed data and analysis sound data, respectively (step S49-2). After that, step S5 is executed.

[0026] In this way, the candidate sound sources of the abnormal noise are estimated based on extracted rotation speed data in which the rotation speed is changing, and the candidate sound sources of the abnormal noise are not estimated based on extracted rotation speed data in which the rotation speed is not changing. For example, in the example of Fig. 3, the parts estimated based on the motor rotation speed are displayed as the candidate sound sources of the abnormal noise on the display unit 70. This prevents parts that emit unchanged sounds that are difficult for the user to notice from being estimated as the candidate sound sources of the abnormal noise, and avoids giving the user an unpleasant feeling.

[0027] In the above embodiment, the candidate sound sources of the abnormal noise are not estimated based on extracted rotation speed data in which the rotation speed does not change, but this is not limiting. For example, the candidate sound sources of the abnormal noise may be estimated based on extracted rotation speed data in which the rotation speed does not change, and these candidate sound sources may be displayed on the display unit 70 as being unlikely to be the source of the abnormal noise.

[0028] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the specific embodiments, and various modifications and variations are possible within the scope of the gist of the present invention as defined in the claims. [Explanation of symbols]

[0029] 100 Sound source estimation device 200 Vehicle 210 Engine 220 Motor

Claims

[Claim 1] A sound source estimation device that estimates one of a plurality of parts mounted on an object as a candidate for an abnormal noise source based on the frequency of a sound generated from the object and the rotation speed of a rotation device mounted on the object, a sound source estimation device that estimates one of the plurality of parts as a candidate for the source of the abnormal noise when a correlation value that correlates with the amount of change in the rotation speed within a predetermined time is equal to or greater than a threshold value, and that does not estimate one of the plurality of parts as a candidate for the source of the abnormal noise when the correlation value is less than the threshold value, or estimates one of the plurality of parts as having a low possibility of being a candidate for the source of the abnormal noise.

Images