Noise analysis method

Abstract

To provide a noise analysis method capable of accurately analyzing reproduction conditions of a vehicle noise.SOLUTION: A noise analysis method selects data within a frequency and time range specified by a user from data indicating a time change of the frequency of a sound generated from a vehicle, extracts a sound having the highest sound pressure as a noise from the data within the frequency and time range specified by the user, and presents reproduction conditions corresponding to the frequency of the sound having the highest sound pressure to the user based on the correspondence relationship between the frequency of the noise and the reproduction conditions stored in advance.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an abnormal sound analysis method. [Background technology]

[0002] In a vehicle, various abnormal noises are generated from various parts and devices such as gears, pumps, etc. Regarding abnormal noises of a vehicle, for example, Patent Document 1 describes a method for outputting reproduction conditions of abnormal noises that match the inquiry information input by a user. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-032535 ​​A Summary of the Invention [Problem to be solved by the invention]

[0004] However, since users do not necessarily input interview information based on their exact memory of abnormal sounds, there is a problem in that the accuracy of the reproduction conditions of abnormal sounds is low.

[0005] Therefore, the present invention has been made in consideration of the above problems, and has an object to provide an abnormal sound analysis method that can analyze the conditions under which abnormal sounds in a vehicle reoccur with high accuracy. [Means for solving the problem]

[0006] The abnormal sound analysis method of the present invention is a method which selects data within a frequency and time range specified by a user from data indicating changes over time in the frequency of sounds generated from a vehicle, extracts the sound with the highest sound pressure from the data within the frequency and time range specified by the user as an abnormal sound, and presents to the user the reproduction conditions corresponding to the frequency of the sound with the highest sound pressure based on a pre-stored correspondence relationship between the frequency of the abnormal sound and reproduction conditions. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a configuration diagram showing an example of an abnormal sound analysis system. [Diagram 2] FIG. 2 is a diagram showing an example of calculating a vehicle speed as a reproduction condition from sound data. [Diagram 3] FIG. 3 is a flowchart showing an example of an abnormal sound analysis method. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] (Configuration of abnormal sound analysis system) 1 is a configuration diagram showing an example of an abnormal sound analysis system 9. The abnormal sound analysis system 9 includes a terminal 1 and an abnormal sound analysis server 2. The terminal 1 and the abnormal sound analysis server 2 communicate with each other via a communication network such as the Internet.

[0009] The terminal 1 is an example of an abnormal sound analysis device. Examples of the terminal 1 include, but are not limited to, a smartphone and a personal computer.

[0010] The terminal 1 has a CPU (Central Processing Unit) 10, a memory 11 such as a ROM (Read Only Memory) or a RAM (Random Access Memory), a communication interface (communication IF) 12, a touch panel 13, a microphone device 14, and a non-volatile storage memory 15. The CPU 10 reads a program from the memory 11 at the time of startup, and forms various software functions according to the program. The communication IF 12 processes communication between the CPU 10 and the abnormal sound analysis server 2.

[0011] The abnormal sound analysis server 2 has a CPU 20, a memory 21 such as a ROM or RAM, a communication IF 22, and a HDD (Hard Disk Drive) 23. The CPU 20 reads a program from the memory 21 at startup and forms various software functions according to the program. The communication IF 22 processes communication between the CPU 20 and the terminal 1.

[0012] In the terminal 1, the microphone device 14 collects sounds generated from the vehicle 8. The CPU 10 analyzes the sounds collected by the microphone device 14 using STFT (Short-Time Fourier Transform). As a result, the CPU 10 generates spectrogram sound data 150 that indicates the relationship between time, frequency, and sound pressure. The sound data 150 is an example of data that indicates the change over time in the frequency of sounds generated from the vehicle 8.

[0013] The touch panel 13 receives touch input from the user and displays an image output by the CPU 10. The CPU 10 outputs the sound data 150 to the touch panel 13 as a spectrogram and displays it on the screen, and receives designation of the range of the spectrogram of the sound data 150 by the user's touch input to the touch panel 13. The CPU 10 selects data of the frequency and time range designated by the user from the sound data 150. This enables the CPU 10 to identify the period and frequency band in which the user recognized an abnormal sound.

[0014] The CPU 10 extracts the sound with the highest sound pressure as an abnormal sound from the data within the frequency and time range specified by the user. Therefore, the CPU 10 can identify the sound that is estimated to be most unpleasant to the user as an abnormal sound without relying on the user's memory. The CPU 10 accesses the abnormal sound analysis server 2 via the communication network 90 to estimate the reproduction conditions from the frequency of the sound with the highest sound pressure.

[0015] The abnormal sound analysis server 2 has a vehicle DB 240 and an abnormal sound DB 241 pre-stored in the HDD 23. The CPU 20 of the abnormal sound analysis server 2 provides the vehicle DB 240 and the abnormal sound DB 241 via the communication network 90 in response to a request from the CPU 10 of the terminal 1.

[0016] The abnormal sound DB 241 indicates the correspondence between the frequency of an abnormal sound and the reproduction conditions. The type of abnormal sound, the frequency f, and the reproduction conditions are registered in advance in the abnormal sound DB 241. For example, high-order gear noise, engine chain noise, fuel rumbling, and pump operation noise are registered as the types of abnormal sounds.

[0017] V(f)=f×3.6 / 1000×r / n / k (1) R(f) = f × 60 / m (2)

[0018] The high-order gear noise is generated according to the speed of the vehicle 8, and the engine chain noise is generated according to the engine speed of the vehicle 8. The vehicle speed is calculated from the frequency f of the abnormal sound by the above formula (1), and the rotation speed is calculated from the frequency f of the abnormal sound by the above formula (2). Formulas (1) and (2) are registered in the abnormal sound DB 241 as reproduction conditions. The tire outer diameter r, number of gear teeth n, and constant k of the vehicle 8 in formula (1), and the number of chain sprocket teeth m of the vehicle 8 in formula (2) are registered in the vehicle DB 240.

[0019] The CPU 10 acquires the tire outer diameter r, the number of gear teeth n, the number of chain sprocket teeth m, and the constant k of the vehicle 8 from the vehicle DB 240, and acquires the formulas (1) and (2) from the abnormal sound DB 241. As a result, the CPU 10 calculates the vehicle speed V(f) for reproducing the high-order gear noise according to the frequency of the sound with the highest sound pressure in the sound data 150, and calculates the rotation speed R(f) for reproducing the engine chain noise.

[0020] The fuel rumbling noise occurs when the frequency f of the abnormal noise is, for example, 2000 to 2400 (Hz). As a reproduction condition of the fuel rumbling noise, a fuel pressure of the low pressure side fuel injection device of the vehicle 8 of, for example, 200 to 700 (kPa) is registered.

[0021] The pump operating noise occurs when the frequency f of the abnormal noise is, for example, 5000 to 8000 (Hz). As a reproduction condition of the pump operating noise, direct fuel injection is registered.

[0022] The CPU 10 obtains the frequency f of the fuel gurgle and pump operation sound from the abnormal sound DB 241, compares it with the frequency of the sound with the highest sound pressure in the sound data 150, determines whether the abnormal sound recognized by the user corresponds to either the fuel gurgle or the pump operation sound, and obtains the judgment conditions for the corresponding abnormal sound.

[0023] In this way, the CPU 10 outputs to the touch panel 13 the reproduction conditions of the corresponding abnormal sound from among the vehicle speed V(f) of the reproduction conditions of high-order gear noise, the rotation speed R(f) of the reproduction conditions of engine chain noise, the fuel rumbling sound, and the pump operation sound, starting from the frequency of the sound with the highest sound pressure in the sound data 150. In this way, the reproduction conditions of the abnormal sound are presented to the user on the screen of the touch panel 13.

[0024] In this way, the terminal 1 does not need to ask the user, but extracts the sound with the maximum sound pressure estimated to be an abnormal sound from the sound data 150, and presents the reproduction conditions corresponding to that frequency to the user. Therefore, it is possible to analyze the reproduction conditions of abnormal sounds of the vehicle 8 with high accuracy, without relying on the user's memory of abnormal sounds.

[0025] (Example of reproducible conditions) Fig. 2 is a diagram showing an example of calculating vehicle speed as a reproduction condition from sound data 150. Fig. 2 shows, from the top of the page, sound data 150, time changes in the frequency of the sound with maximum sound pressure, and time changes in vehicle speed.

[0026] The CPU 10 displays the sound data 150 as a spectrogram on the touch panel 13. In the spectrogram, the horizontal axis indicates time, the vertical axis indicates frequency, and the shade of color indicates sound pressure. In the spectrogram, the lighter the color of the area, the higher the sound pressure. The user specifies a rectangular area A to be analyzed on the touch panel 13. As an example, the area A is assumed to be in the frequency range of 5000 to 8000 (Hz) and the time range of Ta to Tb (seconds). The CPU 10 selects the area A of the sound data 150, and extracts the sound of the abnormal sound area S with the maximum sound pressure, that is, the frequency component, from the area A.

[0027] The CPU 10 samples the frequency of the abnormal sound region S, and calculates the vehicle speed for each sampled frequency using formula (1). As an example, the CPU 10 estimates that the vehicle speed is decelerated from 40 (km / h) to 28 (km / h) from time Ta to time Tb as the reproduction condition of the high-order gear noise. In addition, since the frequency of the abnormal sound is 5000 to 8000 (Hz) in this example, the CPU 10 estimates that the reproduction condition of the pump operating sound is that the fuel injection is direct injection.

[0028] (Method of analyzing abnormal noise) 3 is a flowchart showing an example of an abnormal sound analysis method. The terminal 1 executes the abnormal sound analysis method when the user inputs an instruction to analyze the abnormal sound from the touch panel 13, for example.

[0029] First, the CPU 10 collects sounds from the vehicle 8 using the microphone device 14 (step St1). Next, the CPU 10 generates sound data 150 by analyzing the collected sounds using STFT (step St2) and displays the sound data 150 as a spectrogram on the touch panel 13 (step St3). Next, the CPU 10 accepts designation of the time and frequency ranges of the sound data 150 from the user via the touch panel 13 (step St4).

[0030] Next, the CPU 10 extracts the sound with the maximum sound pressure as an abnormal sound from the range specified by the user in the sound data 150 (step St5), and detects its frequency (step St6). Next, the CPU 10 accesses the vehicle DB 240 and the abnormal sound DB 241 of the abnormal sound analysis server 2 (step St7). Next, the CPU 10 estimates the reproduction conditions of the abnormal sound from the frequency of the sound with the maximum sound pressure based on the vehicle DB 240 and the abnormal sound DB 241, and displays them on the touch panel 13 (step St8). In this manner, the abnormal sound analysis method is executed.

[0031] The above-described embodiment is a preferred example of the present invention, but the present invention is not limited to this embodiment and can be modified in various ways without departing from the scope of the present invention. [Explanation of symbols]

[0032] 1 terminal, 2 abnormal sound analysis servers, 8 vehicles, 10,20 CPUs, 150 sound data

Claims

[Claim 1] Selecting data within a frequency and time range designated by a user from data showing time changes in frequency of sounds generated from a vehicle; extracting, as an abnormal noise, a sound having the highest sound pressure from the data within the frequency and time range designated by the user; presenting to the user the reproduction condition corresponding to the frequency of the sound having the highest sound pressure based on a correspondence relationship between the frequency of the abnormal sound and the reproduction condition stored in advance; Abnormal noise analysis method.

Images