Abnormal sound diagnosis system

Abstract

To provide technology which can appropriately calculate a rate of a period in which a frequency of maximum sound pressure and a generated frequency are substantially matched.SOLUTION: An abnormal sound diagnosis system includes at least one computer. The computer performs processing for acquiring sound spectrum data, processing for acquiring vehicle data, processing for setting a predicted frequency band and a time zone in which presence or absence of abnormal sound is evaluated for the sound spectrum data, processing for calculating a rate of a period in which a frequency of maximum sound pressure and a generated frequency of sound which a device generates are substantially matched in the predicted frequency band and the time zone, which are set, processing for extracting a time zone in which an operation mode in which the generated frequency is included in the predicted frequency band is described from the time zone which is set, when the rate of the period exceeds a first prescribed value, and processing for calculating the rate of the period in which the frequency of maximum sound pressure and the generated frequency of the device are substantially matched in the predicted frequency which is set and the extracted time zone.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The technology disclosed in this specification relates to an abnormal sound diagnosis system.

[0002] Patent Document 1 discloses an abnormal sound diagnosis system for vehicles. This abnormal sound diagnosis system acquires sound spectrum data that describes the frequency spectrum of sounds recorded in a vehicle over a predetermined period of time, and sets, for the sound spectrum data, a predicted frequency band of an abnormal sound and a time period for evaluating the presence or absence of the abnormal sound. The abnormal sound diagnosis system uses the set predicted frequency band and time period to identify candidates for equipment that may be the source of the abnormal sound. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-33848 A Summary of the Invention [Problem to be solved by the invention]

[0004] Among abnormal sound diagnosis systems, there are vehicle diagnosis systems that acquire sound spectrum data and vehicle data that describes the operating modes of equipment mounted on the vehicle over a predetermined period of time. This system calculates the percentage of a period during which the frequency of maximum sound pressure described in the sound spectrum data substantially matches the frequency of sound generated by the equipment in the operating mode described in the vehicle data, within a set predicted frequency band and time period.

[0005] Some of the devices mounted on a vehicle have a generation frequency that changes depending on the operation mode of the device. The operation mode changes depending on the running state of the vehicle. Therefore, the set time period may include a time period in which no abnormal noise occurs. Therefore, the proportion of the period in which the frequency of the maximum sound pressure and the generation frequency substantially match in the set time period is smaller than the proportion of the period in which the frequency of the maximum sound pressure and the generation frequency substantially match in the time period in which the abnormal noise is likely to occur.

[0006] This specification provides a technique that can appropriately calculate the ratio of the period during which the frequency of the maximum sound pressure and the generated frequency substantially match. [Means for solving the problem]

[0007] In a first aspect of the present technology, an abnormal sound diagnosis system for a vehicle includes at least one computer, and the computer performs the following steps: acquires sound spectrum data describing a frequency spectrum of a sound recorded in the vehicle over a predetermined period of time; acquires vehicle data describing an operation mode of an apparatus mounted in the vehicle over the predetermined period of time; sets, for the sound spectrum data, a predicted frequency band of an abnormal sound predicted and a time period for evaluating the presence or absence of the abnormal sound; and, in the set predicted frequency band and time period, calculates a frequency of a maximum sound pressure described in the sound spectrum data and a frequency of an operation mode of an apparatus mounted in the vehicle over the predetermined period of time. the process of calculating a proportion of a period during which the frequency of the sound generated by the device in the operation mode described in the vehicle data substantially coincides with the predicted frequency band; when the proportion of the period exceeds a first predetermined value, extracting from the set time period in the sound spectrum data a time period in which an operation mode in the vehicle data in which the generation frequency of the device is included in the predicted frequency band is described; and calculating, in the set predicted frequency and the extracted time period, a proportion of a period during which the frequency of the maximum sound pressure described in the sound spectrum data substantially coincides with the generation frequency of the device in the operation mode described in the vehicle data.

[0008] According to the above configuration, when the ratio of the period exceeds a first predetermined value, the abnormal sound diagnosis system extracts, from the set period, a time period in which an operation mode in which the occurrence frequency of the equipment is included in the expected frequency band is described. This makes it possible to exclude time periods in which there is a high possibility that abnormal sounds are not occurring. Therefore, it is possible to accurately calculate the ratio of the period in which the frequency of the maximum sound pressure described in the sound spectrum data and the occurrence frequency of the equipment in the operation mode described in the vehicle data substantially match. As a result, by using the ratio of the period, it is possible to appropriately identify candidates for the equipment that is the source of the abnormal sound. Note that machine learning may be used to identify candidates for the equipment that is the source of the abnormal sound. [Brief description of the drawings]

[0009] [Figure 1] FIG. 2 is a block diagram of an abnormal sound diagnosis system 2. [Diagram 2] 13 is a flowchart of a specific process. [Diagram 3] FIG. 4 is a diagram showing an example of sound spectrum data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] (Example) The abnormal sound diagnosis system 2 in FIG. 1 is a system for identifying candidates for devices that may be the source of an abnormal sound occurring in a vehicle (hereinafter, referred to as "abnormal sound candidates"). The abnormal sound diagnosis system 2 includes a diagnostic device 4 and a recording device 6. The diagnostic device 4 includes a monitor 10, an operation unit 12, a communication interface 14, and a control unit 20. Hereinafter, the interface will be referred to as "I / F". The communication interface 14 is an interface for executing communication with other devices (e.g., the recording device 6). The control unit 20 is composed of a computer including a CPU, ROM, RAM, etc.

[0011] The recording device 6 records sound data and vehicle data for a predetermined period of time. The sound data is data corresponding to sounds generated from the vehicle. The sound data includes vehicle road noise, engine noise, switching noise of the PCU, etc. The vehicle data includes, for example, vehicle speed, engine speed, motor speed, control pattern, etc. The control pattern is information corresponding to the carrier frequency of the PCU.

[0012] (Specific processing; Figure 2) The identification process executed by the control unit 20 of the diagnostic device 4 will be described with reference to Fig. 2. The control unit 20 starts the process in Fig. 2 when an operation for starting the identification process is received from a user. At the start point of Fig. 2, the control unit 20 has already received the sound data of the sound recorded in the vehicle and the vehicle data from the recording device 6.

[0013] In S10, the control unit 20 acquires sound spectrum data by performing a short-time Fourier transform on the received sound data. The control unit 20 displays the acquired sound spectrum data on the monitor 10. The vertical axis of the sound spectrum data 30 in Fig. 3 indicates the frequency of the sound, and the horizontal axis indicates time. The color of each pixel in the sound spectrum data 30 indicates the sound pressure level (dB). In other words, the sound spectrum data 30 is data that indicates the change over time of the frequency spectrum of the sound.

[0014] In S12 of FIG. 2, the control unit 20 monitors whether a setting operation for setting an expected frequency band and a first evaluation time period is received from the user. The expected frequency band is a frequency band of an expected abnormal sound. The first evaluation time period is a time period during which the presence or absence of the abnormal sound is diagnosed. When the control unit 20 receives a setting operation from the user, the control unit 20 judges YES in S12 and proceeds to S14. As shown in FIG. 3, the user can set a rectangular area 32 surrounded by the expected frequency band and the first evaluation time period by operating the operation unit 12 of the diagnostic device 4. In FIG. 3, the frequency band from frequency F1 to frequency F2 is the expected frequency band, and time T0 to time T6 is the first evaluation time period.

[0015] In S14 of FIG. 2, the control unit 20 calculates the first ratio in the predicted frequency band and the first evaluation time period. Specifically, the control unit 20 calculates the frequency of the maximum sound pressure in the predicted frequency band and the first evaluation time period using the sound spectrum data for each unit time period, and calculates the generation frequency of the sound generated by the device (hereinafter referred to as the "candidate device") that may be the source of the abnormal sound using the vehicle data and vehicle specifications. The vehicle specifications may be information pre-stored in the diagnostic device 4, or may be information input by the user. As an example, when the PCU is the candidate device, the control unit 20 calculates the generation frequency using the carrier frequency (vehicle data), the motor rotation speed (vehicle data), and the number of pole pairs (vehicle specifications). The control unit 20 calculates the ratio of the period during which the calculated frequency of the maximum sound pressure and the calculated generation frequency of the candidate device substantially match as the first ratio. Here, "substantially match" includes not only the case where the frequency of the maximum sound pressure and the generation frequency of the candidate device completely match, but also the case where the difference between the frequency of the maximum sound pressure and the generation frequency of the candidate device is several percent.

[0016] In S16, the control unit 20 determines whether the first ratio calculated in S14 is equal to or greater than a first predetermined value (e.g., 10%). If the first ratio is equal to or greater than the first predetermined value (YES in S16), the control unit 20 proceeds to S20. On the other hand, if the first ratio is less than the first predetermined value (NO in S16), the control unit 20 proceeds to S30.

[0017] In S20, the control unit 20 executes an extraction process to extract a second evaluation time period from the first evaluation time period. The extraction process is a process for extracting a time period in which the generation frequency of the candidate device is likely to be included in the expected frequency band from the first evaluation time period. Specifically, the control unit 20 determines the second evaluation time period by using the control pattern and the expected frequency band included in the vehicle data. As shown in FIG. 3, the control unit 20 stores the control pattern and the abnormal sound frequency band in association with each other. The abnormal sound frequency band is information indicating a frequency band in which an abnormal sound may occur when the vehicle is operating with the control pattern associated with the abnormal sound frequency band. The control unit 20 determines the time period in which the expected frequency band and the abnormal sound frequency band match as the second evaluation time period. In the case of FIG. 3, the frequencies F1 and F2 are 6500 Hz and 8000 Hz, respectively. Therefore, the control unit 20 determines that the expected frequency band matches the abnormal sound frequency band corresponding to the control pattern "3", and determines the time period in which the control pattern is "3" as the second evaluation time period. Specifically, the control unit 20 determines the times T0 to T1, T2 to T3, and T4 to T5 as the second evaluation time period.

[0018] 2, the control unit 20 calculates a second ratio in the predicted frequency band and the second evaluation time period by using the frequency of the maximum sound pressure calculated in S14 and the frequency generated by the candidate device calculated in S14. The second ratio is the ratio of the period during which the frequency of the maximum sound pressure and the frequency generated by the candidate device substantially match, during the second evaluation time period.

[0019] In S24, the control unit 20 determines whether the second ratio is equal to or greater than a second predetermined value (50%). The second predetermined value is greater than the first predetermined value. If the second ratio is equal to or greater than the second predetermined value (YES in S24), the control unit 20 proceeds to S26. On the other hand, if the second ratio is less than the second predetermined value (NO in S24), the control unit 20 proceeds to S30.

[0020] In S26, the control unit 20 outputs the candidate device as an abnormal sound candidate. As one example, the control unit 20 displays a message indicating that the candidate device is an abnormal sound candidate on the monitor 10. When S26 ends, the control unit 20 ends the processing in FIG. 2.

[0021] In S30, the control unit 20 does not output the candidate device as an abnormal sound candidate. When S30 ends, the control unit 20 ends the processing in FIG.

[0022] As described above, the abnormal sound diagnosis system 2 for a vehicle includes the control unit 20 (one example of "at least one computer"). The control unit 20 acquires sound spectrum data describing the frequency spectrum of sounds recorded in the vehicle over a predetermined period of time (S10 in FIG. 2), and acquires vehicle data describing the operation mode of the devices mounted on the vehicle over a predetermined period of time. The control unit 20 sets, for the sound spectrum data, an expected frequency band of the expected abnormal sound and a first evaluation time period in which the presence or absence of the abnormal sound is evaluated (S12), and calculates the proportion of a period during which the frequency of the maximum sound pressure described in the sound spectrum data substantially matches the generation frequency of the sound generated by the candidate device in the control pattern described in the vehicle data, within the expected frequency band and the first evaluation time period (S14). When the first ratio exceeds a first predetermined value (YES in S16), the control unit 20 extracts from the first evaluation time period in the sound spectrum data a time period in which a control pattern in the vehicle data in which the generation frequency of the candidate device falls within the predicted frequency band (S20), and calculates a second ratio in which the frequency of the maximum sound pressure described in the sound spectrum data substantially matches the generation frequency of the candidate device in the control pattern described in the vehicle data in the predicted frequency and the second evaluation time period (S22).

[0023] According to the above configuration, when the first ratio exceeds the first predetermined value, the abnormal sound diagnosis system 2 extracts from the first evaluation time period a time period in which a control pattern is described in which the generation frequency of a candidate device falls within the expected frequency band. This makes it possible to exclude time periods in which there is a high possibility that an abnormal sound is not occurring (times T1 to T2, T3 to T4, and T5 to T6 in FIG. 3). Therefore, it is possible to accurately calculate the ratio of the period in which the frequency of the maximum sound pressure described in the sound spectrum data and the generation frequency of the candidate device substantially match (i.e., the second ratio). As a result, it is possible to appropriately identify abnormal sound candidates. [Explanation of symbols]

[0024] 2: abnormal sound diagnosis system, 4: diagnosis device, 6: recording device, 10: monitor, 12: operation unit, 14: communication interface, 20: control unit, 30: sound spectrum data, 32: rectangular area

Claims

[Claim 1] An abnormal sound diagnosis system for a vehicle, comprising: At least one computer, the computer comprising: obtaining sound spectrum data describing a frequency spectrum of sounds recorded in the vehicle over a predetermined period of time; acquiring vehicle data describing operational modes of equipment installed in the vehicle over the predetermined period of time; A process of setting an expected frequency band of an abnormal sound and a time period for evaluating the presence or absence of the abnormal sound for the sound spectrum data; A process of calculating a ratio of a period during which the frequency of the maximum sound pressure described in the sound spectrum data substantially coincides with the frequency of the sound generated by the device in the operation mode described in the vehicle data, in the set predicted frequency band and time period; A process of extracting, when the ratio of the period exceeds a first predetermined value, a time period in which an operation mode in which the generated frequency of the device falls within the predicted frequency band in the vehicle data is described from the set time period in the sound spectrum data; A process of calculating a ratio of a period during which the frequency of maximum sound pressure described in the sound spectrum data substantially coincides with the generated frequency of the device in the operation mode described in the vehicle data, for the set predicted frequency and the extracted time period; This abnormal noise diagnosis system performs the following:

Images