Sound diagnosis system, sound diagnosis method, and sound diagnosis program

Abstract

To provide a technology that can identify a component causing an abnormal noise from among multiple components.SOLUTION: A sound diagnosis system 100 includes a sound pressure calculation unit 205 that calculates the sound pressure of each of a plurality of frequencies at each timing from sound information about sounds emitted by a device, a self-correlation calculation unit 206 that calculates a self-correlation function about the change over time of representative values of the sound pressure of the frequencies included in a target frequency band, a peak extraction unit 208 that extracts one or more peak timing of the self-correlation function, a detection unit 210 that detects a cycle of one or more abnormal sounds on the basis of one or more peak timings, and an identification unit 222 that identifies the component causing each of one or more abnormal sounds from among a plurality of components included in the device on the basis of the cycle.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to a sound diagnostic system, and more particularly to a sound diagnostic system using sound pressure. [Background technology]

[0002] When a device such as a home appliance used at home or an image forming device used in an office breaks down, the device may emit an abnormal sound. A serviceman can identify the broken part from the abnormal sound and replace the broken part. However, when a device equipped with multiple motors breaks down, the serviceman may not be able to identify which part of the device is generating the abnormal sound. In this case, the serviceman cannot identify which part should be repaired or replaced. Therefore, there is a need for a sound diagnosis technology that can easily identify the part generating the abnormal sound in the device.

[0003] Regarding sound diagnosis, for example, Japanese Patent Application Laid-Open No. 2017-138151 (Patent Document 1) discloses a "diagnostic device" that supports the task of identifying the cause of abnormal sounds. The diagnostic device identifies the frequency of the abnormal sound in frequency spectrum waveform data. The diagnostic device extracts information on the period and frequency of the abnormal sound by performing a fast Fourier transform on the frequency components of the abnormal sound in the time axis direction. The extracted information is used to identify the cause of the abnormal sound.

[0004] Japanese Patent Laid-Open Publication No. 4-167699 (Patent Document 2) discloses a "sticking noise detection device." This device calculates the autocorrelation function of the time waveform of the sound. This device judges the presence or absence of an abnormality based on the result of comparing the peak value of the autocorrelation function with a predetermined value. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2017-138151 A [Patent Document 2] Japanese Patent Application Publication No. 4-167699 Summary of the Invention [Problem to be solved by the invention]

[0006] An apparatus having a complex mechanism such as an image forming apparatus has a plurality of components that generate drive sounds. The plurality of components may include a component that can generate an abnormal sound having a plurality of frequency components. Alternatively, the plurality of components may include a component that can generate an abnormal sound multiple times during one rotation or an abnormal sound having a certain duration. The apparatus disclosed in Patent Document 1 cannot accurately detect the period of the abnormal sound of such a component. As a result, the apparatus disclosed in Patent Document 1 cannot accurately identify the component causing the abnormal sound.

[0007] The device disclosed in Patent Document 2 merely determines the presence or absence of an abnormality, and does not identify the part causing the abnormal noise.

[0008] An object of one aspect of the present disclosure is to provide a technique capable of identifying a component causing an abnormal noise from among a plurality of components. [Means for solving the problem]

[0009] According to an embodiment, a sound diagnostic system is provided. The sound diagnostic system includes a first calculation unit, a second calculation unit, an extraction unit, a detection unit, and an identification unit. The first calculation unit calculates the sound pressure of each of a plurality of frequencies at each timing from sound information related to a sound emitted by the device. The second calculation unit calculates an autocorrelation function for a time change of a representative value of the sound pressure of a frequency included in a target frequency band among the plurality of frequencies. The extraction unit extracts one or more peak timings of the autocorrelation function. The detection unit detects a period of each of the one or more abnormal sounds based on the one or more peak timings. The identification unit identifies a causative part for each of the one or more abnormal sounds from among a plurality of parts included in the device based on the period.

[0010] In some aspects, the frequency band of interest includes all of the plurality of frequencies or only a portion of the plurality of frequencies.

[0011] In one aspect, the second calculation unit calculates a first autocorrelation function corresponding to a first target frequency band including all of the multiple frequencies as the autocorrelation function, and in response to a fact that the causal component is not identified based on the first autocorrelation function, the second calculation unit calculates second autocorrelation functions corresponding to each of one or more second target frequency bands that are part of the first target frequency band.

[0012] In one aspect, the second calculation unit determines a range of the time shift amount of the autocorrelation function according to a maximum possible rotation period of the multiple components.

[0013] In one aspect, the second calculation unit removes a trend component from a time change of the representative value. The second calculation unit calculates an autocorrelation function of the time change of the representative value from which the trend component has been removed.

[0014] In one aspect, the extraction unit extracts the target time shift amount as one of one or more peak timings in response to a difference between the autocorrelation at the target time shift amount and the autocorrelation before and after the target time shift amount being equal to or greater than a predetermined threshold.

[0015] In one aspect, the extraction unit narrows down one or more peak timings based on peak information indicating the shape of the peak.

[0016] In one aspect, the identifying unit identifies the causative part by using period information that associates each of the multiple parts with a rotation period.

[0017] According to another embodiment, there is provided a sound diagnosis method for an apparatus. The sound diagnosis method comprises calculating a sound pressure of each of a plurality of frequencies at each timing from sound information related to a sound emitted by the apparatus. The sound diagnosis method comprises calculating an autocorrelation function for a time change of a representative value of a sound pressure of a frequency included in a target frequency band among the plurality of frequencies. The sound diagnosis method comprises extracting one or more peak timings of the autocorrelation function. The sound diagnosis method comprises detecting a period of each of one or more abnormal sounds based on the one or more peak timings. The sound diagnosis method comprises identifying a causative part for each of the one or more abnormal sounds from among a plurality of parts included in the apparatus based on the period.

[0018] According to yet another embodiment, there is provided a sound diagnosis program for an apparatus, the sound diagnosis program causing a computer to execute the sound diagnosis method described above. Effect of the Invention

[0019] According to an embodiment, the sound diagnostic system can identify the part causing the abnormal noise from among a plurality of parts. [Brief description of the drawings]

[0020] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a sound diagnostic system to which the technology of the present disclosure can be applied. [Diagram 2] FIG. 2 is a diagram showing an example of a configuration of the sound diagnostic system according to the first embodiment. [Diagram 3] FIG. 4 is a diagram showing an example of a procedure of a diagnostic process executed by the sound diagnostic system according to the first embodiment. [Figure 4] FIG. 11 is a diagram illustrating an example of sound information. [Diagram 5] FIG. 13 is a diagram illustrating an example of a spectrogram. [Figure 6] FIG. 13 is a diagram showing an example of data on total sound pressure in a target frequency band for each timing in a spectrogram. [Figure 7] FIG. 13 is a diagram illustrating an example of an autocorrelation function. [Figure 8]FIG. 11 is a diagram illustrating an example of peak timing information. [Figure 9] FIG. 13 is a diagram showing an example of the prominence of an autocorrelation peak. [Figure 10] FIG. 11 is a diagram illustrating an example of period information. [Figure 11] FIG. 13 is a diagram illustrating an example of result information indicating a causal component identified by the identification unit. [Figure 12] FIG. 4 is a diagram showing an example of a first screen output by an output unit. [Figure 13] FIG. 13 is a diagram showing an example of a second screen output by the output unit. [Figure 14] 13A to 13C are diagrams illustrating an example of processing in a case where a target frequency band is set so as to include all of a plurality of frequencies included in a spectrogram. [Figure 15] 11A and 11B are diagrams illustrating an example of processing in a case where a target frequency band is set so as to include only a portion of a plurality of frequencies included in a spectrogram. [Figure 16] The autocorrelation function calculated for the time change in the total sound pressure of frequencies included in the target frequency band "0 to 861 Hz" is shown. [Figure 17] The autocorrelation function calculated for the time change in the total sound pressure of frequencies included in the target frequency band "861 to 1722 Hz" is shown. [Figure 18] The autocorrelation function calculated for the time change in the total sound pressure of frequencies included in the target frequency band "1722 to 2583 Hz" is shown. [Figure 19] FIG. 11 is a diagram showing an example of a procedure of a diagnostic process executed by the sound diagnostic system according to the second embodiment. [Figure 20] FIG. 11 is a diagram showing an example of period candidate information. [Figure 21] FIG. 2 is a block diagram showing the hardware configuration of a computer functioning as a server or a diagnostic device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] Hereinafter, embodiments of the technical idea according to the present disclosure will be described with reference to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed descriptions thereof will not be repeated. Also, each embodiment and each modification example etc. may be selectively combined as appropriate.

[0022] <A. Application Example> FIG. 1 is a diagram showing an example of the configuration of a stethoscope diagnosis system to which the technology of the present disclosure is applicable. With reference to FIG. 1, an overview of the stethoscope diagnosis system 100 to which the technology of the present disclosure is applicable and problems in stethoscope diagnosis will be described. At the same time, terms used in this specification will be described.

[0023] (a. Configuration of a System to which the Technology of the Present Disclosure is Applicable) The stethoscope diagnosis system 100 includes, as an example, a diagnosis device 120 and a server 110. According to an embodiment, the diagnosis device 120 may have the functions of the server 110. The device 130 is the object of stethoscope diagnosis. The device 130 is any device having a plurality of drive components. The device 130 includes office equipment such as an image forming device, home appliances, devices in a factory, research equipment, etc. The diagnosis device 120 is communicatively connected to the server 110 via a wireless LAN (Local Area Network) terminal such as a WiFi (registered trademark) router or the Internet communication network.

[0024] The diagnosis device 120 collects sound information regarding the sound emitted by the device 130 and displays the analysis result of the sound information. The serviceman carries the diagnosis device 120 in order to perform maintenance management, repair, etc. of the device to be maintained used by the end user. The diagnosis device 120 acquires at least abnormal sounds emitted by the device to be maintained and displays the analysis result of the acquired abnormal sounds. The analysis result of the abnormal sounds indicates the component causing the abnormal sound among the plurality of components included in the device 130.

[0025] According to an embodiment, the diagnostic device 120 transmits sampled sound information to the server 110. Then, the diagnostic device 120 receives an analysis result from the server 110. Furthermore, the diagnostic device 120 displays the analysis result on a display. The analysis result indicates at least the part causing the abnormal sound. The diagnostic device 120 and the server 110 are connected to each other via a wireless or wired network so that they can transmit and receive data to and from each other.

[0026] According to another embodiment, the diagnostic device 120 analyzes the sampled sound information. Then, the diagnostic device 120 displays the analysis result on a display. The analysis result indicates at least the component causing the abnormal sound. In this case, the sound diagnostic system 100 does not need to include the server 110.

[0027] According to another embodiment, the diagnostic device 120 is a tablet, a smartphone, a personal computer, smart glasses, or any other information processing device. The diagnostic device 120 may include a built-in normal microphone or a directional microphone. The diagnostic device 120 may be configured to be connectable to a normal microphone or a directional microphone.

[0028] The server 110 receives sound information from the diagnostic device 120 via a network. The server 110 analyzes the sound information and returns the analysis result to the diagnostic device 120. The analysis result indicates the part causing the abnormal sound. The procedure for analyzing the sound information will be described in detail later. According to an embodiment, the server 110 can be realized as one or more devices, a virtual machine built on a cloud environment, or a combination of these.

[0029] The device 130 is a device to be diagnosed. The device 130 includes a plurality of moving parts. The moving parts are, for example, parts that are driven by a motor. The sounds emitted by the moving parts may have different frequencies.

[0030] (b. Problems that may occur in sound diagnosis) As described above, the device 130 includes a plurality of driving parts. For example, an image forming device used in an office or the like includes a large number of driving parts. Therefore, even if the diagnostic device 120 collects an abnormal sound, it is not easy to identify which driving part is generating the abnormal sound. In order to solve this problem, the sound diagnostic system 100 calculates the sound pressure of each of a plurality of frequencies from sound information related to the sound generated by the device 130. The sound diagnostic system 100 calculates an autocorrelation function for the time change of the representative value of the sound pressure of the frequency included in the target frequency band. The sound diagnostic system 100 extracts one or more peak timings of the autocorrelation function. The sound diagnostic system 100 detects the period of each of the one or more abnormal sounds based on the one or more peak timings. The sound diagnostic system 100 identifies a causative part for each of the one or more abnormal sounds from among the multiple driving parts included in the device 130 based on the detected period.

[0031] (c. Terms Used in This Specification) Next, the terms used to describe the technology of the present disclosure will be explained.

[0032] In this specification, the term "system" includes a configuration consisting of one or more devices. The system may also include a virtual machine or a container built in a cloud environment, or a configuration consisting of at least a part of these.

[0033] In this specification, the term "device" refers to any information processing device such as a personal computer, a workstation, a server device, a tablet, or a smartphone. The device may also be a combination of these.

[0034] According to one embodiment, the sound diagnosis system 100 may be connected to input / output devices such as a display and a keyboard and used by a user. According to another embodiment, the sound diagnosis system 100 may provide various functions to a user as a service or a web application via a network. In this case, the user may use the functions of the sound diagnosis system 100 via a browser or client software installed on his / her terminal.

[0035] In this specification, the "device to be diagnosed" includes a device installed anywhere, such as a home, an office, a factory, a school, or a research laboratory. As an example, the device to be diagnosed is any device, such as a refrigerator, an air conditioner, an image forming device, a manufacturing device or a robot installed in a factory. The device to be diagnosed includes multiple driving parts. In this specification, the device to be diagnosed may also be simply referred to as a device.

[0036] In this specification, a "part (driving part)" is a part that is included in the device to be diagnosed. The part is typically a rotational driving part. The part may be a motor, or a combination of a motor and a power transmission part. The part may also be a combination of a motor, a power transmission part, and a part beyond that. Or the part may be a part that is driven by a motor.

[0037] In this specification, the term "sound information" refers to sounds emitted by the device to be diagnosed. The sound information may also be data picked up by the diagnostic device 120 using a microphone, i.e., sampled sound data. The sampled sound data is digital data represented at any resolution.

[0038] As used herein, "sound pressure" refers to the variation in pressure from atmospheric pressure due to sound. Sound pressure is defined as the effective value of the maximum amplitude.

[0039] The "overall sound pressure (overall sound intensity)" in this specification is a representative value of the sound pressure of the frequency components included in the target frequency band at each timing of the spectrogram. The overall sound pressure (representative value of sound pressure) is, for example, a sum or an average value. The spectrogram includes a plurality of frequency components for each timing. For example, assuming that at timing A of the spectrogram, frequency components X, Y, and Z are included in the target frequency band. In this case, the overall sound pressure (overall sound intensity) at timing A is calculated by the formula "overall sound pressure at timing A = sound pressure of frequency component X + sound pressure of frequency component Y + sound pressure of frequency component Z". Or, the overall sound pressure (overall sound intensity) at timing A may be the average value of the sound pressures of frequency components X, Y, and Z.

[0040] The "autocorrelation function" in this specification shows the relationship between the time shift amount τ, the time waveform x(t) of the overall sound pressure (representative value of sound pressure), and the autocorrelation between the waveform x(t) and the waveform x(t+τ) obtained by shifting x(t) by the time shift amount τ. "Autocorrelation" represents the degree of similarity between the waveform x(t) and the waveform x(t+τ). "Autocorrelation" is also referred to as the "autocorrelation coefficient".

[0041] When x(t) is a discrete signal, the autocorrelation function R(τ) is typically represented by the following formula (1). Note that the autocorrelation function R(τ) may be normalized using the mean and variance of x(t) so as to take values between -1 and 1.

[0042]

Equation

[0043] The "peak timing" in this specification is a timing in the graph of autocorrelation that has an autocorrelation higher than a certain level with respect to the autocorrelations of the surrounding timings.

[0044] <B. Embodiment 1> (a. Device Configuration) Fig. 2 is a diagram showing an example of a configuration of the sound diagnostic system according to the first embodiment. A part or all of the configuration shown in Fig. 2 may be realized as hardware. Also, a part of the configuration shown in Fig. 2 may be realized as software. In this case, the sound diagnostic system 100 includes one or more processors for executing each piece of software, one or more memories, and one or more storages.

[0045] The sound diagnostic system 100 includes a microphone 202, a sound collection unit 204, a sound pressure calculation unit 205, an autocorrelation calculation unit 206, a peak extraction unit 208, and a detection unit 210. The sound diagnostic system 100 further includes a storage unit 212, an identification unit 222, and an output unit 224. The storage unit 212 stores period information 214 and a UI (User Interface) program 218.

[0046] According to an aspect, diagnostic device 120 includes microphone 202, sound collection unit 204, and output unit 224. Server 110 includes sound pressure calculation unit 205, autocorrelation calculation unit 206, peak extraction unit 208, detection unit 210, storage unit 212, and identification unit 222.

[0047] According to another aspect, diagnostic device 120 may also have the functions of server 110, in addition to microphone 202, sound collection unit 204, and output unit 224. In this case, diagnostic device 120 includes sound pressure calculation unit 205, autocorrelation calculation unit 206, peak extraction unit 208, detection unit 210, storage unit 212, and identification unit 222.

[0048] The microphone 202 and the sound collector 204 act as a collector that collects sound information about sounds emitted by the device.

[0049] The microphone 202 converts sound into an electrical signal and outputs the electrical signal to the sound collector 204. The device 130 includes multiple sound sources (driving components). Therefore, the microphone 202 picks up a composite sound emitted by the multiple sound sources.

[0050] The sound collection unit 204 converts the electrical signal output from the microphone 202 into a digital signal to generate sound information (sampling result). The sound collection unit 204 converts an analog signal at an arbitrary resolution into sound information that is a digital signal. The sound collection unit 204 outputs the generated sound information.

[0051] The sound pressure calculation unit 205 is an example of a "first calculation unit" in the present disclosure. The sound pressure calculation unit 205 calculates the sound pressure of each of a plurality of frequencies included in each timing of the sound information. More specifically, the sound pressure calculation unit 205 performs a Short-Time Fourier Transform (STFT) on the sound information (sound data). The sound pressure calculation unit 205 obtains a frequency spectrum waveform (spectrogram) by performing the STFT on the sound information. The spectrogram represents the time change of the sound pressure distribution for each frequency.

[0052] The sound pressure calculation unit 205 calculates the total sound pressure (representative value of sound pressure) of the frequencies included in the target frequency band among the multiple frequencies at each timing. The total sound pressure is, for example, the sum or average value of the sound pressures of the frequencies. The target frequency band may include all of the multiple frequencies for which the sound pressures are calculated by the sound pressure calculation unit 205, or may include only a part of the multiple frequencies.

[0053] The autocorrelation calculation unit 206 is an example of a "second calculation unit" in the present disclosure. The autocorrelation calculation unit 206 calculates an autocorrelation function for a time change in the total sound pressure (representative value of sound pressure) of frequencies included in the target frequency band. That is, the autocorrelation calculation unit 206 calculates an autocorrelation function for a time waveform x(t) of the total sound pressure of frequencies included in the target frequency band. The autocorrelation calculation unit 206 calculates the autocorrelation between the time waveform x(t) and a time waveform x(t+τ) obtained by shifting the time waveform x(t) by a time shift amount τ. The autocorrelation calculation unit 206 calculates an autocorrelation function R(τ) that indicates a relationship between the time shift amount τ and the autocorrelation while changing the time shift amount τ within a predetermined range. The details of the autocorrelation calculation unit 206 will be described later with reference to FIG. 7.

[0054] The peak extraction unit 208 extracts one or more peak timings of the autocorrelation function R(τ). More specifically, the peak extraction unit 208 extracts, as the peak timing, a time shift amount τ at which the difference between the autocorrelation at each timing and the adjacent autocorrelation is equal to or greater than a preset value. Alternatively, the peak extraction unit 208 may extract the peak timing based on a prominence value (prominence) of the autocorrelation. Details of the peak extraction unit 208 will be described later with reference to Figs. 8 and 9.

[0055] The detection unit 210 detects the period of each of the one or more abnormal sounds based on the one or more peak timings. For example, the detection unit 210 detects the peak timing with the greatest peak height (prominence value) among the one or more extracted peak timings as the period. Alternatively, when multiple peak timings are extracted, the detection unit 210 may detect the period of the abnormal sound based on the interval between the peak timings. For example, when peak timings of 1.5 sec and 3.0 sec are extracted, the detection unit 210 3.0-1.5=1.5sec may be detected as a period.

[0056] The detection unit 210 may detect the periods of the multiple abnormal sounds from the multiple peak timings. For example, the detection unit 210 detects the shortest peak timing among the multiple peak timings as the period of the first abnormal sound. Next, the detection unit 210 extracts one or more remaining peak timings whose difference from an integer multiple of the period of the first abnormal sound exceeds a threshold value from the multiple peak timings. The detection unit 210 detects the shortest peak timing among the extracted one or more remaining peak timings as the period of the second abnormal sound. By repeating the same process, the detection unit 210 may detect the period of the kth abnormal sound. k is an integer equal to or greater than 3.

[0057] The storage unit 212 stores data necessary for processing by the identification unit 222, and UI data provided to the user. The storage unit 212 stores periodicity information 214. The periodicity information 214 indicates the period of the sound emitted by each driving part. The storage unit 212 stores a UI program 218. The UI program 218 is UI data or a program provided to the user. The UI program 218 includes UI design data, an event program linked to each UI component, and the like.

[0058] The output unit 224 outputs the analysis result of the sound information. The analysis result indicates at least the part causing the abnormal sound. As an example, the output unit 224 displays the screens shown in Figs. 12 and 13 on the display of the diagnostic device 120. Alternatively, the output unit 224 may transmit display data to another smartphone 226, a PC (Personal Computer) 228, or any other device.

[0059] (b. Sound diagnosis procedure) FIG. 3 is a diagram showing an example of the procedure of a diagnostic process executed by the sound diagnostic system according to the first embodiment. An overview of the diagnostic process of the sound diagnostic system 100 will be described with reference to FIG. 3. Details of each process in FIG. 3 will be described later with reference to FIGS. 4 to 13. According to a certain aspect, the processor of the sound diagnostic system 100 may load a program for executing the process in FIG. 3 from a storage into a memory and execute the program. In another aspect, some or all of the process may be realized as a combination of circuit elements configured to execute the process.

[0060] In step S301, the diagnostic device 120 picks up the sound of the device 130 to be diagnosed with a microphone (microphone 202). More specifically, the microphone 202 of the diagnostic device 120 picks up the sound emitted by the device 130 in response to an operation by a user (e.g., a serviceman). The sound collector 204 converts the sound (analog signal) into sampled sound information (digital data). The diagnostic device 120 transmits the sound information to the server 110. In the following steps, the sound information is transmitted to the server 110. In a certain aspect, when the diagnostic device 120 also has the function of the server 110, the diagnostic device 120 does not transmit the sound information to the server 110. As an example, the diagnostic device 120 acquires sound information 400 as shown in FIG. 4.

[0061] In step S302, the server 110 performs STFT on the acquired sound information. As a result, the server 110 calculates a spectrogram that represents a time change in sound pressure distribution for each frequency. As an example, the server 110 acquires a spectrogram 500 as shown in FIG. 5.

[0062] In step S303, server 110 calculates the total sound pressure (representative value of sound pressure) of all of the multiple frequencies included in the spectrogram for each timing. Typically, server 110 calculates the sum of the sound pressures of each frequency for each timing in the spectrogram as the total sound pressure. Alternatively, server 110 calculates the average value of the sound pressures of each frequency for each timing in the spectrogram as the total sound pressure. As an example, server 110 acquires data 600 as shown in Fig. 6. Data 600 indicates the total sound pressure for each timing.

[0063] In step S304, the server 110 calculates an autocorrelation function for the time change of the total sound pressure. As an example, the server 110 calculates an autocorrelation function 700 as shown in Fig. 7. The autocorrelation function 700 indicates the relationship between the time soft quantity τ and the autocorrelation.

[0064] In step S305, the server 110 extracts one or more peak timings of the autocorrelation function. A peak timing is a timing that has a certain or greater autocorrelation compared to the timings before and after it in the autocorrelation graph. In other words, it is a portion that is greatly convex in the autocorrelation graph. As an example, the server 110 obtains peak timing information 800 that indicates a peak timing as shown in FIG. 8.

[0065] In step S306, the server 110 determines whether one or more peak timings have been extracted.

[0066] If one or more peak timings are not extracted (NO in step S306), the process proceeds to step S307. In step S307, the server 110 generates result information indicating that the part causing the abnormal noise cannot be identified.

[0067] If one or more peak timings have been extracted (NO in step S306), the process proceeds to step S308. In step S308, the server 110 detects the period of each of the one or more abnormal sounds based on the one or more peak timings indicated by the peak timing information 800.

[0068] In step S309, the server 110 determines whether a period that is effective for identifying the part causing the abnormal noise has been detected. For example, the server 110 determines that a period whose difference from any of the periods indicated by the period information 214 is less than a predetermined threshold is a valid period.

[0069] If a period effective for identifying the part causing the abnormal noise is not detected (NO in step S309), the process proceeds to step S307.

[0070] If a period effective for identifying the part causing the abnormal noise is detected (YES in step S309), the process proceeds to step S310. In step S310, the server 110 identifies, as the causative part, a driving part that emits a sound having a period whose difference from the detected period is less than a predetermined threshold, based on the period information 214. The server 110 generates result information indicating the identified causative part.

[0071] After step S307 or step S310, the process proceeds to step S311. In step S311, server 110 transmits the result information to diagnostic device 120. Diagnostic device 120 displays the result indicated by the received result information on a display. According to an aspect, diagnostic device 120 may transfer the result information to another device (such as a PC or a smartphone).

[0072] (c. Details of each step) Next, each process in the flow of Fig. 3 will be described in detail with reference to Fig. 4 to Fig. 13. More specifically, the processes in steps S301 to S311 will be described in detail with reference to Fig. 4 to Fig. 13.

[0073] 4 is a diagram showing an example of sound information. Sound information 400 is an example of the sound information acquired in step S301. Sound information 400 represents sound pressure over time.

[0074] FIG. 5 is a diagram showing an example of a spectrogram. Spectrogram 500 is an example of the spectrogram calculated in step S302. The sound diagnosis system 100 calculates the spectrogram 500 by performing STFT on the sound information 400. The sound information 400 includes a plurality of frequency components. That is, the sound information is a collection of a plurality of sounds having different frequencies. By performing STFT on the sound information 400, the spectrogram 500 shows the time change of the sound pressure distribution of each frequency component. In other words, the spectrogram 500 includes information on various sounds emitted by each part of the device 130.

[0075] Fig. 6 is a diagram showing an example of data on total sound pressure in a target frequency band for each timing in a spectrogram. Data 600 shown in Fig. 6 is an example of data calculated in step S303. Data 600 shows a change over time in total sound pressure (representative value of sound pressure) of all frequency components included in spectrogram 500. The total sound pressure is the sum or average value of the sound pressure of all frequency components.

[0076] Fig. 7 is a diagram showing an example of an autocorrelation function. An autocorrelation function 700 shown in Fig. 7 is an example of the autocorrelation function calculated in step S304. The autocorrelation function 700 is calculated for the time waveform x(t) represented by the data 600 shown in Fig. 6.

[0077] For example, the time waveform x(t) is expressed as a discrete function according to the time resolution Δt of the spectrogram calculated by the sound pressure calculation unit 205. Δt is, for example, 0.005 sec. In this case, the autocorrelation calculation unit 206 calculates the autocorrelation between the time waveform x(t) and the time waveform x(t+n×Δt). n is a natural number and is also called a lag. The time shift amount τ is expressed as n×Δt. The autocorrelation calculation unit 206 calculates the autocorrelation function 700 by changing the lag n within a predetermined range.

[0078] It is also possible to calculate an autocorrelation function for the sound information 400 shown in Fig. 4. However, the capacity of the data 600 shown in Fig. 6 is smaller than the capacity of the sound information 400 shown in Fig. 4. Therefore, the calculation load when calculating an autocorrelation function for the data 600 shown in Fig. 6 is smaller than the calculation load when calculating an autocorrelation function for the sound information 400 shown in Fig. 4.

[0079] Furthermore, to further reduce the calculation load when calculating the autocorrelation function 700, the range of the time shift amount τ may be determined according to the maximum possible rotation period of the multiple parts included in the device 130. For example, if the maximum period indicated by the period information 214 is 3 seconds, 3 seconds or a time slightly longer than 3 seconds is determined as the upper limit value of the range of the time shift amount τ. The server 110 calculates the autocorrelation function 700 by changing the time shift amount τ in the range from 0 to the upper limit value.

[0080] When a periodic abnormal noise is generated by a driving part, the total sound pressure over time has periodicity. In this case, periodic peaks appear in the autocorrelation function 700 as shown in Fig. 7. Therefore, the peak timing corresponds to the period of the abnormal noise.

[0081] In order to more accurately determine the period of the abnormal sound, the server 110 preferably removes trend components from the time change indicated by the data 600. The trend components represent long-term trends in the data, such as a certain increasing or decreasing trend that occurs over time. For this reason, the server 110 removes the trend components, for example, by applying a high-pass filter to the data 600. The server 110 may calculate an autocorrelation function for the time change in the total sound pressure from which the trend components have been removed.

[0082] Fig. 8 is a diagram showing an example of peak timing information. Peak timing information 800 shown in Fig. 8 is an example of peak timing information indicating one or more peak timings extracted in step S305. The peak timing information 800 indicates the peak timing of the autocorrelation function 700 shown in Fig. 7.

[0083] The server 110 (peak extraction unit 208) sequentially changes the time shift amount τ from 0 to the upper limit value, and determines whether or not the target time shift amount is a peak timing.

[0084] Specifically, the server 110 (peak extraction unit 208) compares the autocorrelation at the target time shift amount with the autocorrelation at the adjacent time shift amount. It is assumed that the autocorrelation at the target time shift amount is higher than the autocorrelation at the adjacent time shift amount by a predetermined threshold or more. In this case, the server 110 (peak extraction unit 208) determines the target time shift amount as a peak timing. The server 110 acquires the autocorrelation at that time as a peak value. In FIG. 8, "x" indicates a peak.

[0085] Server 110 may exclude peaks that exhibit autocorrelation equal to or less than a preset threshold (for example, 0.1). Server 110 extracts the time shift amount τ of the remaining peaks that have not been excluded as the peak timing.

[0086] 9 is a diagram showing an example of the prominence of an autocorrelation peak. The server 110 (peak extraction unit 208) may detect the autocorrelation peak timing based on the prominence value (prominence) of the peak. Prominence is a method of calculating the prominence of each peak based on the relative position of a peak to other peaks.

[0087] A specific method of calculating prominence is as follows. First, the server 110 focuses on a certain peak. Next, the server 110 extends the horizontal line of the peak to the left and right until the horizontal line of the peak intersects with a higher peak or reaches the left and right ends (graph ends) of the signal. Next, the server 110 finds a signal whose autocorrelation is a minimum value on the right and left sides of the section to which the horizontal line is extended. Next, the server 110 sets the larger of the minimum value of the right section and the minimum value of the left section as the reference value. Next, the server 110 sets the difference between the autocorrelation of the peak and the reference value as the prominence value. The prominence value represents the degree of protrusion from the surroundings (the height of the peak relative to the surroundings). The server 110 excludes peaks that are equal to or less than a preset prominence value (for example, 0.1). The server 110 extracts the time shift amount τ of the remaining peaks that have not been excluded as the peak timing.

[0088] Note that the server 110 may limit the length of the horizontal line to the left and right in order to calculate a more accurate prominence value.

[0089] According to one aspect, after detecting the time shift amount of the target as a peak timing, the server 110 may not detect the time shift amount for a certain period from the peak timing as a peak timing, thereby suppressing the detection of consecutive peaks due to noise.

[0090] According to an aspect, the server 110 may narrow down one or more peak timings based on the peak information. The peak information indicates the shape of the peak. The peak information indicates, for example, the height of the peak (the value of autocorrelation or prominence value), the kurtosis of the peak, or the difference or ratio between the autocorrelation of the peak and the maximum value of the autocorrelation. For example, the server 110 excludes from the peak timings a time shift amount having an autocorrelation whose difference from the maximum value of the autocorrelation is equal to or less than a predetermined threshold (for example, half the maximum value). This prevents a time shift amount with a small autocorrelation from being erroneously detected as a peak timing. Alternatively, the server 110 may exclude from the peak timings a time shift amount whose peak height is equal to or less than a predetermined threshold. Alternatively, the server 110 may exclude from the peak timings a time shift amount whose peak kurtosis is equal to or less than a predetermined threshold.

[0091] The server 110 (detection unit 210) detects the period of each of the one or more abnormal sounds based on the one or more peak timings.

[0092] The identifying unit 222 compares the detected period with the period information 214 to identify the component causing the abnormal noise.

[0093] Fig. 10 is a diagram showing an example of the period information. As shown in Fig. 10, the period information 214 associates each of a plurality of driving parts included in the device 130 with a rotation period. According to a certain aspect, the period information 214 may be expressed as a table of a relational database. In addition, the period information 214 may be expressed in any other data format such as JSON (JavaScript (registered trademark) Object Notation).

[0094] 11 is a diagram showing an example of result information indicating a causal part identified by the identifying unit 222. The identifying unit 222 compares the detected period with the period for each part in the period information 214. The identifying unit 222 searches for a part that emits a sound with a rotation period close to the detected period.

[0095] 11 is generated when the detection period is "0.1676 sec" and the rotation period of part X is "0.1674 sec." Since the detection period "0.1676 sec" and the rotation period of part X is "0.1674 sec" are close to each other, the identifying unit 222 selects part X as a candidate for the part causing the abnormal sound with the detection period "0.1676 sec."

[0096] According to one aspect, the identifying unit 222 may select a plurality of parts as candidates for the part causing the abnormal noise of the detected cycle. For example, as shown in Fig. 11, when the detection cycle and the rotation cycles of the parts X, Y, and Z are close to each other, the identifying unit 222 may select the parts X, Y, and Z as candidates for the part causing the abnormal noise of the detection cycle.

[0097] As an example, the identifying unit 222 can determine a part p that is a candidate for the part causing the abnormal noise of the detected period, using the following expressions (2) to (5).

[0098]

number

[0099] FIG. 12 is a diagram showing an example of a first screen output by an output unit. According to a certain aspect, the first screen 1200 is displayed on the display of the diagnostic device 120. According to another aspect, the output unit 224 transmits the data of the first screen 1200 to other devices (such as the smartphone 226, the PC 228, etc.). The other devices display the data of the first screen 1200 on the display. The first screen 1200 is generated by the UI program 218.

[0100] The first screen 1200 is a screen before the execution of auscultation diagnosis. The first screen 1200 includes parameter settings necessary for analysis, recording, and execution buttons for analysis, etc. A user (such as a service technician) can obtain an analysis result by selecting a sound file (collected sound information) on the first screen 1200 and pressing the analysis button. When the analysis button is pressed, the first screen 1200 transitions to the second screen 1300.

[0101] FIG. 13 is a diagram showing an example of a second screen output by an output unit. According to a certain aspect, the second screen 1300 is displayed on the display of the diagnostic device 120. According to another aspect, the output unit 224 transmits the data of the second screen 1300 to other devices (such as the smartphone 226, the PC 228, etc.). The other devices display the data of the second screen 1300 on the display. The second screen 1300 is generated by the UI program 218.

[0102] The second screen 1300 is a screen showing the analysis result of a sound file (collected sound information). The second screen 1300 displays the period of the abnormal sound detected based on the peak timing of the autocorrelation function and the part name that emits a sound with a period close to the period of the abnormal sound. When there are multiple parts that emit a sound with a period close to the period of the detected abnormal sound, the second screen 1300 may display multiple part names.

[0103] <C. Embodiment 2> As described above, the target frequency band for which the total sound pressure is calculated may include all of the multiple frequencies included in the spectrogram 500. Alternatively, the target frequency band for which the total sound pressure is calculated may include only a portion of the multiple frequencies included in the spectrogram 500.

[0104] Fig. 14 is a diagram showing a processing example in the case where a target frequency band is set so as to include all of the multiple frequencies included in a spectrogram. Fig. 15 is a diagram showing a processing example in the case where a target frequency band is set so as to include only a portion of the multiple frequencies included in a spectrogram. In Figs. 14 and 15, the upper part shows a spectrogram. The middle part shows the time change (time waveform) of the total sound pressure. The lower part shows an autocorrelation function.

[0105] For example, an image forming apparatus includes a plurality of rollers. An abnormal sound such as a rubbing sound generated by the rollers has only some frequency components and has a small sound pressure. Therefore, when the total sound pressure is calculated for all frequency components, the abnormal sound having only some frequency components is buried in the environmental sound having other frequency components. As a result, as shown in FIG. 14, when the total sound pressure is calculated for all frequency components included in the spectrogram, the peak of the autocorrelation function is not detected. In contrast, as shown in FIG. 15, when the total sound pressure is calculated for some frequency components, the peak of the autocorrelation function is detected.

[0106] Fig. 16 shows an autocorrelation function calculated for the time change in the total sound pressure of frequencies included in the target frequency band "0 to 861 Hz". Fig. 17 shows an autocorrelation function calculated for the time change in the total sound pressure of frequencies included in the target frequency band "861 to 1722 Hz". Fig. 18 shows an autocorrelation function calculated for the time change in the total sound pressure of frequencies included in the target frequency band "1722 to 2583 Hz".

[0107] As shown in Fig. 16 to Fig. 18, the three autocorrelation functions corresponding to the three frequency bands are different from one another. Therefore, the sound diagnostic system according to the second embodiment identifies the period of each of one or more abnormal sounds by using a plurality of frequency bands. The sound diagnostic system according to the second embodiment has the same device configuration as the sound diagnostic system 100 shown in Fig. 2. Therefore, a description of the device configuration of the sound diagnostic system according to the second embodiment will be omitted.

[0108] The sound diagnostic system according to the second embodiment predefines a plurality of frequency bands. The plurality of frequency bands include a first frequency band and one or more second frequency bands. The first frequency band includes all of the plurality of frequencies included in the spectrogram. Each of the one or more second frequency bands includes only a portion of the plurality of frequencies included in the spectrogram. When the one or more second frequency bands include a plurality of second frequency bands, the plurality of second frequency bands are different from each other. Each of the one or more second frequency bands is a part of the first frequency band.

[0109] The one or more second frequency bands include, for example, the frequency bands "0 to 861 Hz", "861 to 1722 Hz", and "1722 to 2583 Hz".

[0110] Note that a second frequency band may partially overlap with another second frequency band. For example, the one or more second frequency bands may include the frequency bands "0 to 861 Hz", "761 to 1722 Hz", and "1622 to 2583 Hz".

[0111] Fig. 19 is a diagram showing an example of the procedure of a diagnostic process executed by the sound diagnostic system according to the second embodiment. The procedure shown in Fig. 19 differs from the procedure shown in Fig. 3 in that it includes steps S1901 to S1908 instead of step S307. Therefore, steps S1901 to S1908 will be described below. Steps S1901 to S1908 are performed if NO is determined in step S306 or step S309. Note that in step S303, a first frequency band is set as a target frequency band. The first frequency band set as the target frequency band corresponds to the "first target frequency band" of the present disclosure.

[0112] In step S1901, the server 110 (the autocorrelation calculation unit 206) selects one unselected second frequency band from among one or more second frequency bands as a target frequency band. The second frequency band selected as the target frequency band corresponds to the "second target frequency band" of the present disclosure.

[0113] In step S1902, the server 110 calculates the total sound pressure (representative value of sound pressure) of the frequencies included in the target frequency band for each timing.

[0114] In step S1903, the server 110 calculates an autocorrelation function for the time change of the total sound pressure.

[0115] In step S1904, the server 110 (peak extractor 208) extracts one or more peak timings of the autocorrelation function.

[0116] In step S1905, the server 110 (detection unit 210) detects the period of each of the one or more abnormal sounds based on the one or more peak timings.

[0117] In step S1906, server 110 registers the detected cycle and the height of the peak at the peak timing corresponding to the cycle (hereinafter referred to as "peak height") in cycle candidate information shown in Fig. 20. The cycle candidate information is stored in storage unit 212 shown in Fig. 2.

[0118] If no peak timing is extracted in step S1904, steps S1905 and S1906 are omitted.

[0119] In step S1907, the server 110 determines whether or not all of the one or more second frequency bands have been selected as the target frequency bands.

[0120] If none of the one or more second frequency bands have been selected as the target frequency band (NO in step S1907), the process returns to step S1901.

[0121] If all of the one or more second frequency bands have been selected as the target frequency bands (YES in step S1907), the process proceeds to step S1908. In step S1908, server 110 (identifier 222) selects the period corresponding to the largest peak height from the period candidate information.

[0122] After step S1908, the process proceeds to step S310. In step S310 after step S1908, the server 110 (identification unit 222) identifies the causing part. Specifically, the server 110 identifies a driving part that emits a sound whose period is less than a predetermined threshold value when compared to the period selected in step S1908 as the causing part.

[0123] FIG. 20 is a diagram showing an example of period candidate information. Period candidate information 2000 shown in FIG. 20 indicates, for each of a plurality of second frequency bands, the detected period and the peak height of the peak timing corresponding to that period. The peak height is represented, for example, by an autocorrelation value or a prominence value. Period candidate information 2000 indicates "0.756 sec" as the period corresponding to the largest peak height. Therefore, in step S310, server 110 identifies a part that emits a sound with a period close to the period "0.756 sec" as the part causing the abnormal sound.

[0124] In this way, server 110 according to the second embodiment first calculates a first autocorrelation function corresponding to a frequency band including all of the multiple frequencies included in the spectrogram. Thereafter, server 110 performs steps S1901 to S1908 in response to the fact that a causal component is not identified based on the first autocorrelation function. That is, server 110 calculates a second autocorrelation function corresponding to each of one or more second frequency bands that are part of the first frequency band. Then, server 110 identifies a causal component based on the second autocorrelation function.

[0125] <D.バリエーション> In the present specification, a configuration in which diagnostic device 120 includes microphone 202 has been described, but the application of the technology of the present disclosure is not limited to this. According to a certain aspect, device 130 to be diagnosed may include a microphone. In this case, device 130 transmits sound information to diagnostic device 120 or server 110.

[0126] The sound information may be analog sound data. In this case, the detection unit 210 may have an encoding function for the analog sound data. The sound information may be digital sound data. As an example, the sound collection unit 204 may encode the sound acquired by the microphone 202. The sound information including the multiple frequencies emitted by the device 130 may include a sampling result (sampling information) of the recorded sound. The predetermined period for extracting the information regarding the multiple sound pressures may be a period of any length preset in the sound diagnosis system 100. For example, the predetermined period may include a period of any length, such as several milliseconds, several seconds, or several minutes. The information regarding the sound pressure may include information such as the total sound pressure of the sound of each frequency, the relative sound pressure of the sound of each frequency, and the peak timing included in the total sound pressure.

[0127] In one aspect, the apparatus 130 may be an image forming apparatus including a plurality of driving sources, in which case the plurality of driving sources may drive any of the fixing belt, the paper discharge roller, the paper feed roller, and the fixing gear.

[0128] In this specification, timing may include not only a momentary point in a time series but also any length of time (period). As an example, the sound diagnostic system 100 may analyze sound information divided into 1 μsec intervals. In this case, a certain timing means a certain 1 μsec period. Thus, timing may be any period such as 1 μsec, 1 millisecond, etc. The sound pressure within a certain period (e.g., a 1 μsec period) may be, for example, the average or median value of the sound pressure within that period.

[0129] All or part of the sound pressure calculation section 205, the autocorrelation calculation section 206, the peak extraction section 208, the detection section 210, the storage section 212, the identification section 222, and the output section 224 may be realized by a computer.

[0130] Fig. 21 is a block diagram showing the hardware configuration of a computer that functions as a server or a diagnostic device. A computer 2100 shown in Fig. 21 includes, as main components, a processor 1, a RAM (Random Access Memory) 2, a storage device 3, a communication interface 4, an input device 5, and a display 6. Each component is connected to each other by a data bus.

[0131] The processor 1 includes, for example, a CPU (Central Processing Unit) and an MPU (Micro Processing Unit), and executes a program 7 stored in a storage device 3 .

[0132] The RAM 2 volatilely stores data generated by the execution of the program 7 by the processor 1, data input via the input device 5, data received from the outside via the communication interface 4, and the like.

[0133] The storage device 3 stores data in a non-volatile manner and is realized by a hard disk device, a solid state drive (SSD) device, or the like.

[0134] The input device 5 includes, for example, a touch panel, a keyboard, a mouse, etc. The display 6 is, for example, a liquid crystal display or an organic EL (Electro Luminescence) display.

[0135] The processing in the computer 2100 is realized by the cooperation of each piece of hardware and a program 7 executed by the processor 1. The program 7 is stored in the storage device 3 in advance. Alternatively, the program 7 may be stored in a CD-ROM (Compact Disc-Read Only Memory) or other recording medium and distributed. Alternatively, the program 7 may be provided as a downloadable application program by an information provider connected to the Internet. Such a program 7 is read from the recording medium by an optical disk drive (not shown) or other reading device, or downloaded via the communication interface 4, and then temporarily stored in the storage device 3.

[0136] Each component constituting the computer 2100 shown in Fig. 21 is a general one. Therefore, one of the essential parts of the technical idea according to the present disclosure can be said to be the program 7 stored in the RAM 2, the storage device 3, a CD-ROM or other recording medium, or the program 7 downloadable via a network. The recording medium may include a non-transitory (non-volatile) computer-readable data recording medium. Note that the operation of each piece of hardware of the computer 2100 is well known, and therefore detailed description will not be repeated.

[0137] Note that the recording medium is not limited to CD-ROM, FD (Flexible Disk), and hard disk, and may be a medium that fixedly carries a program, such as magnetic tape, cassette tape, optical disk (MO (Magnetic Optical Disc) / MD (Mini Disc) / DVD (Digital Versatile Disc)), IC (Integrated Circuit) card (including memory card), optical card, mask ROM, EPROM (Electronically Programmable Read-Only Memory), EEPROM (Electronically Erasable Programmable Read-Only Memory), semiconductor memory such as flash ROM, etc.

[0138] Program 7 includes not only programs directly executable by processor 1, but also programs in source program form, compressed programs, encrypted programs, etc.

[0139] <E. Postscript> As described above, this embodiment includes the following disclosures. [Configuration 1] A sound diagnosis system, a first calculation unit that calculates the sound pressure of each of a plurality of frequencies at each timing from sound information regarding the sound emitted by the device; a second calculation unit that calculates an autocorrelation function for the time change of the representative value of the sound pressure of the frequencies included in the target frequency band among the plurality of frequencies; an extraction unit that extracts one or more peak timings of the autocorrelation function; a detection unit that detects the period of each of one or more abnormal sounds based on the one or more peak timings; a sound diagnosis system, comprising a specification unit that specifies a cause component for each of the one or more abnormal sounds from among a plurality of components included in the device based on the period. [Configuration 2] 2. The sound diagnostic system according to claim 1, wherein the target frequency band includes all of the plurality of frequencies or only a portion of the plurality of frequencies. [Configuration 3] The second calculation unit is Calculating a first autocorrelation function corresponding to a first target frequency band including all of the plurality of frequencies as the autocorrelation function; the sound diagnostic system according to configuration 1, wherein, in response to the causative component not being identified based on the first autocorrelation function, a second autocorrelation function corresponding to each of one or more second target frequency bands that are part of the first target frequency band is calculated. [Configuration 4] 4. The sound diagnosis system according to any one of configurations 1 to 3, wherein the second calculation unit determines a range of a time shift amount of the autocorrelation function according to a maximum possible rotation period of the plurality of components. [Configuration 5] The second calculation unit is removing trend components from the time variation of the representative value; 5. The sound diagnosis system according to any one of configurations 1 to 4, wherein the autocorrelation function is calculated for a time change of the representative value from which the trend component has been removed. [Configuration 6] The sound diagnostic system according to any one of configurations 1 to 5, wherein the extraction unit extracts the time shift amount of the target as one of the one or more peak timings in response to a difference between an autocorrelation at the time shift amount of the target and an autocorrelation before and after the time shift amount of the target being equal to or greater than a predetermined threshold. [Configuration 7] 7. The sound diagnosis system according to any one of configurations 1 to 6, wherein the extraction unit narrows down the one or more peak timings based on peak information indicating a shape of a peak. [Configuration 8] 8. The sound diagnosis system according to any one of configurations 1 to 7, wherein the identifying unit identifies the causative part by using period information in which each of the plurality of parts is associated with a rotation period. [Configuration 9] A sound diagnostic method, comprising: Calculating a sound pressure of each of a plurality of frequencies at each timing from sound information related to a sound emitted by the device; Calculating an autocorrelation function of a time change in the representative value of the sound pressure of a frequency included in a target frequency band among the plurality of frequencies; extracting one or more peak timings of the autocorrelation function; Detecting a period of each of the one or more abnormal sounds based on the one or more peak timings; and identifying a causative part for each of the one or more abnormal sounds from among a plurality of parts included in the device based on the period. [Configuration 10] A sound diagnosis program for causing a computer to execute a sound diagnosis method, The sound diagnosis method includes: Calculating a sound pressure of each of a plurality of frequencies at each timing from sound information related to a sound emitted by the device; Calculating an autocorrelation function of a time change in the representative value of the sound pressure of a frequency included in a target frequency band among the plurality of frequencies; extracting one or more peak timings of the autocorrelation function; Detecting a period of each of the one or more abnormal sounds based on the one or more peak timings; and identifying a causative part for each of the one or more abnormal sounds from among a plurality of parts included in the device based on the period.

[0140] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is indicated by the claims, not by the above description, and is intended to include all modifications within the meaning and scope of the claims. In addition, the disclosure contents described in the embodiments and each modified example are intended to be implemented, as far as possible, either alone or in combination. [Explanation of symbols]

[0141] 1 processor, 2 RAM, 3 storage device, 4 communication interface, 5 input device, 6 display, 7 program, 100 sound diagnosis system, 110 server, 120 diagnosis device, 130 device, 202 microphone, 204 sound collection unit, 205 sound pressure calculation unit, 206 autocorrelation calculation unit, 208 peak extraction unit, 210 detection unit, 212 storage unit, 214 period information, 218 UI program, 222 identification unit, 224 output unit, 226 smartphone, 400 sound information, 500 spectrogram, 600 data, 800 peak timing information, 1100 result information, 1200 first screen, 1300 second screen, 2000 period candidate information, 2100 computer.

Claims

1. 1. A sound diagnostic system, comprising: a first calculator that calculates a sound pressure of each of a plurality of frequencies at each timing from sound information related to a sound emitted by the device; a second calculation unit that calculates an autocorrelation function of a time change in the representative value of the sound pressure of a frequency included in a target frequency band among the plurality of frequencies; an extractor for extracting one or more peak timings of the autocorrelation function; a detection unit that detects a period of each of the one or more abnormal sounds based on the one or more peak timings; an identification unit that identifies a causative part for each of the one or more abnormal sounds from among a plurality of parts included in the device based on the period.

2. The sound diagnostic system according to claim 1 , wherein the target frequency band includes all of the plurality of frequencies or only a portion of the plurality of frequencies.

3. The second calculation unit is Calculating a first autocorrelation function corresponding to a first target frequency band including all of the plurality of frequencies as the autocorrelation function; 2. The sound diagnostic system according to claim 1, further comprising: a second autocorrelation function that calculates, in response to the causative component not being identified based on the first autocorrelation function, a second autocorrelation function that corresponds to each of one or more second target frequency bands that are part of the first target frequency band.

4. The sound diagnosis system according to claim 1 , wherein the second calculator determines a range of a time shift amount of the autocorrelation function in accordance with a maximum possible rotation period of the plurality of components.

5. The second calculation unit is removing trend components from the time variation of the representative value; The sound diagnosis system according to claim 1 , wherein the autocorrelation function is calculated for a time change of the representative value from which the trend component has been removed.

6. 4. The sound diagnostic system according to claim 1, wherein the extraction unit extracts the time shift amount of the object as one of the one or more peak timings in response to a difference between an autocorrelation at the time shift amount of the object and an autocorrelation before and after the time shift amount of the object being equal to or greater than a predetermined threshold.

7. The sound diagnosis system according to claim 1 , wherein the extraction unit narrows down the one or more peak timings based on peak information indicating a shape of a peak.

8. The sound diagnosis system according to claim 1 , wherein the identifying unit identifies the causative part by using period information in which each of the plurality of parts is associated with a rotation period.

9. A sound diagnostic method, comprising: Calculating a sound pressure of each of a plurality of frequencies at each timing from sound information related to a sound emitted by the device; Calculating an autocorrelation function of a time change in the representative value of the sound pressure of a frequency included in a target frequency band among the plurality of frequencies; extracting one or more peak timings of the autocorrelation function; Detecting a period of each of the one or more abnormal sounds based on the one or more peak timings; and identifying a causative part for each of the one or more abnormal sounds from among a plurality of parts included in the device based on the period.

10. A sound diagnosis program for causing a computer to execute a sound diagnosis method, The sound diagnosis method includes: Calculating a sound pressure of each of a plurality of frequencies at each timing from sound information related to a sound emitted by the device; Calculating an autocorrelation function of a time change in the representative value of the sound pressure of a frequency included in a target frequency band among the plurality of frequencies; extracting one or more peak timings of the autocorrelation function; Detecting a period of each of the one or more abnormal sounds based on the one or more peak timings; and identifying a causative part for each of the one or more abnormal sounds from among a plurality of parts included in the device based on the period.

Images